Introduction
Judicial review for statutory compliance has an uneasy relationship with the modern administrative state. This paper focuses on the UK and New Zealand, where the general position is that questions of legal interpretation fall for the courts to determine authoritatively.Footnote 1 Regulators must correctly understand and apply the law that governs their decision-making. However, while this basic principle is fundamental to public law conceptions of legality, its application can have soft edges. This reflects the reality that regulators necessarily exercise a degree of discretion where their decision-making engages complicated factual circumstances or an evolving policy position. Accordingly, courts have sometimes preferred a more deferential approach, leaving space for regulators to determine legal questions.Footnote 2 On this view, questions of legal interpretation become mixed questions of law and technocratic policy, and the courts should not focus myopically on the former lest it compromise the proper scope of discretion afforded to the regulator in respect of the latter.
In comparative administrative law, the US Chevron decision has long been taken as representative of this more deferential approach.Footnote 3 In Chevron, the Supreme Court ruled that if the legislative framework is ambiguous, then the courts will defer to the regulator’s own reasonable interpretation of its governing statute. This approach is fundamentally at odds with the starting position at common law and has long been acknowledged to be at odds with basic principle even in the US.Footnote 4 Nonetheless, the Chevron doctrine persisted and became highly influential in no small part because it supplied a compelling rationale for judicial deference: ambiguous legislative drafting should be treated as an implicit legislative delegation to a regulator, who can bring a degree of specialist, technocratic expertise to bear on mixed issues of interpretation and application. This underlying rationale is more sympathetic to the role of discretion in the modern administrative state. Building off of that rationale is an associated pragmatic justification, as Chevron deference allows regulators to apply their legislative mandate in a flexible way that is responsive to changing circumstances and improved technical understandings of the impacts of particular regulatory provisions. In this way, Chevron and its underlying reasoning has provided broad, implicit support across different jurisdictions for courts to depart from insisting on strict adherence to legality and defer to the regulator’s own interpretation of its governing statute.
That is, until now. In June 2024, the Supreme Court released its judgment in Loper Bright,Footnote 5 with a majority overruling Chevron and ostensibly aligning the US approach with the UK position. This paper reviews Loper Bright for a common law audience and argues that it represents an opportunity to reappraise the merits of excessive deference to regulators. In doing so it contributes to a growing comparative interest in US administrative law generally and deference in particular.Footnote 6 Section 1 sets out the principled common law approach to review of regulators’ legal interpretations and suggests that there are good reasons to be sceptical of regulators’ interpretations of their own governing statutes.Footnote 7 Section 2 then examines Chevron as a potential answer to this scepticism, before Loper Bright and the end of Chevron deference is discussed in Section 3.
Section 4 critically examines two example cases of excessive regulatory deference from the UK and New Zealand, which now appear highly anomalous in light of Loper Bright. Those cases are South Yorkshire Transport,Footnote 8 and Unison Networks. Footnote 9 These cases have been selected for several reasons. First, they are extreme examples of deference, and so demonstrate just how far deference can and has been pushed even under an assumption of judicial authority over interpretation. Indeed, and secondly, these cases are Anglo-Commonwealth analogues of Chevron in that they apply deference presumptively and so depart from the courts’ standard role as authoritative legal interpreters. Perhaps the extraordinary deference in these cases can be explained by the fact that they both deal with economic regulation, which is generally acknowledged to be outside the knowledge and experience of generalist courts.Footnote 10 A third reason to focus on these cases, therefore, is their subject matter. Examining examples where some deference to expertise is reasonably expected is useful to demonstrate that the limits of responsible deference are often much closer than they appear.
A final reason is jurisdictional. Neither the UK nor New Zealand has engaged directly with the Chevron doctrine, meaning that questions about the scope of deference to regulators is still very much ‘at large’. Contrast Australia, which maintains a staunch constitutional commitment to judicial independence and so has explicitly considered and rejected Chevron,Footnote 11 and Canada, which has embraced an alternative form of deference through default application of a reasonableness standard of review.Footnote 12 But even in Canada, analogous cases to South Yorkshire Transport and Unison Networks have received more robust judicial scrutiny than UK and New Zealand law seems to allow. For this reason some aspects of the Canadian experience are also briefly discussed in Section 4, to contextualise the anomalous nature of the example cases, but the focus remains squarely on considering the position in the UK and New Zealand where the advent of Loper Bright may have meaningful consequences.
Section 5 concludes with some observations about appropriately balancing legality and discretion. In brief, the argument presented here is that we should more readily insist on the judicial authority to review regulatory decision‑making and end Chevron-style deference. But to clarify, this line of argument should not be taken as an effort to place public law in opposition to the regulatory state. Rather, the motivation is to facilitate legitimate, responsible and rational decision-making in line with the best traditions of both public law and modern approaches to regulation.
1. The general approach at common law
Public law in New Zealand and the UK exhibit many similarities, which extend to administrative law.Footnote 13 When the English common law was transplanted during colonisation, New Zealand inherited the long-standing common law tradition of controlling administrative discretion. New Zealand has also internalised many of the developments of modern English administrative law and so for the purposes of expositing general principle, the two jurisdictions can be taken together.
In both jurisdictions, the basic principle is that questions of law, including statutory interpretation, are for the courts to determine authoritatively. As Lord Diplock explained in the GCHQ case, decided contemporaneously with Chevron:Footnote 14
The decision-maker must understand correctly the law that regulates his [or her] decision-making power and must give effect to it. Whether he [or she] has or not is par excellence a justiciable question to be decided, in the event of dispute, by those persons, the judges, by whom the judicial power of the state is exercisable.
Of course, legalistic control of administrative power is not the only goal of the common law. In practice, the courts can take on a role that is more facilitative rather than restrictive of public power. This tendency has been captured in Carol Harlow and Richard Rawlings’ ‘green light’ theory of administrative law.Footnote 15 The origins of this green light approach have been located in the early progressive, functionalist scholarship of Ivor Jennings, Harold Laski and William Robson, and the Canadian John Willis. These scholars saw value in public power being put to public purposes, a goal that could be frustrated if legalistic control was applied dogmatically. However, Harlow and Rawlings emphasise that while green light theories consider that overly legalistic approaches are undesirable, they maintain a role for law to avoid the ‘unrestricted or arbitrary’ use of state power.Footnote 16 The trick lies in finding the right balance.
We see administrative law seeking to strike the appropriate balance in a number of ways in contemporary doctrine. There is the classic distinction between law and policy, for example, where the courts will be more reluctant to intervene if a substantive policy position is implicated.Footnote 17 Where the merits of the decision are in question, the law will usually keep out of the way. Something like this distinction is also embedded in Wednesbury unreasonableness,Footnote 18 and Edwards v Bairstow factual error,Footnote 19 and so a facilitative approach can be tied to particular grounds of review. There are also situations where the subject matter of the decision is simply inapt for judicial determination. High policy is less amenable to review on legality grounds because the legal content is minimal, and the courts intervene only if the decision involves ‘bad faith, improper motive or manifest absurdity’.Footnote 20 Further, political bodies are usually more appropriately placed to take decisions on those matters because they are democratically accountable.Footnote 21 In all, there are a range of circumstances where the courts directly or indirectly adopt a deferential stance as an outcome of ordinary public law reasoning.
Of course, not all exercises of public power fall into such a neat taxonomy. Many regulatory decisions turn on mixed questions of fact and law, meaning any legal interpretation cannot be sensibly undertaken in isolation from its practical and policy consequences. Indeed, the line between fact and law itself may be taken as relatively fluid.Footnote 22 It is in these situations where the tension between strict maintenance of legality and facilitative approaches to administrative law is most acute. One approach, which aligns with the principle that the courts should closely control for legality, is for the consequences of different interpretations to be considered and addressed by the court in reaching an authoritative view on the question of interpretation. In respect of statutory interpretation questions, for example, it would be unreasonable to assume that Parliament had intended an absurd or cumbersome outcome from the regulator’s perspective, and so the practical consequences themselves will count strongly against any such interpretation. This aligns the regulatory and judicial functions in service of the same overall goal of coherent, legitimate and effective regulation.
Regulators in the modern state are increasingly specialised, and there is a residual concern that the bespoke expertise of the specialist regulator is beyond the capacity of court to sensibly judge. In such situations, there is a view that decisions rendered in the public interest by regulatory officials should not be second-guessed by the courts.Footnote 23 Rather than insisting on legality, courts ought to defer to the regulator’s own view on questions of law, including its interpretation of its governing statute. This approach effectively breaks down the distinction between law and regulatory policy or, put another way, the distinction between the interpretation and application of law. This is the view that ‘an interpretation of the law is a rule for applying the law’ and so interpretation and application must be taken together to constitute the same (regulatory) activity.Footnote 24 Courts insisting on authoritative judicial interpretation of legislation are therefore interfering with the application of regulatory policy, and so are straying beyond their proper role. Instead of the law shaping the legitimate boundaries of regulatory discretion, matters of public policy and the public interest are conflated with any legal standards that apply. As Trevor Allan, hardly a sceptic of common law standards of legality, explains:Footnote 25
When the modern welfare‑regulatory state confers extensive discretionary powers on public agencies, enabling them to perform wide-ranging and perhaps loosely defined public functions, the clear-cut distinction between law and administration, or law and public policy, dissolves. … The executive is no longer a mere conduit for the implementation of programmes authorised by the legislature; it has become an independent source of policy formation and governance, reflecting its own views of the public interest.
In such circumstances, ‘judicial intervention on behalf of common law values must not usurp the exercise of that legitimate discretion’.Footnote 26 Courts should defer to regulators, even in respect of questions of law that would otherwise naturally fall within the courts’ remit and expertise.
There are several reasons to be sceptical about this deferential approach. First, it unattractively places the courts in opposition to regulators and the regulatory state more broadly. Rather than viewing courts as partners in a shared enterprise who can bring ethical scrutiny, an outside perspective and a perceived legitimacy, regulators’ decisions are lionised and courts are instructed to keep out of their way. The best modern ‘green light’ approaches to administrative law do not treat the courts as a barrier to good regulation, but as a crucial part of a system that delivers good administration.Footnote 27 Secondly, justifications for this deferential approach are often question‑begging. Allan, for example, is concerned that courts do not ‘usurp the exercise of … legitimate discretion’, but who determines whether any individual exercise of discretion is in fact legitimate? There is often very limited political accountability for regulatory decisions given the formal independence of many regulatory agencies, and if the courts are also excluded from that task then the matter can only be one for those agencies to determine for themselves. There are obvious rule of law concerns here, particularly in respect of jurisdictional questions, or challenges interrogating proper purpose. The nemo iudex in causa sua principle ought to apply in such cases,Footnote 28 which suggests that an important role for the courts remains.
Thirdly, the differing institutional competencies as between regulators and the courts is usually drawn in terms of specialist expertise, but application of expertise is too limited a description of the regulator’s function. Proponents of deference seem to imagine that the regulator’s interpretation is tightly confined and utterly technocratic – aimed at discovering the correct answer to a narrow question based on the application of experience and specialist knowledge. But this is not how regulators operate in practice. As Allan notes in the quoted passage above, a regulator is an ‘independent source of policy formation and governance’, which makes decisions ‘reflecting its own views of the public interest’. This describes the exercise of a public law power, where considered judgement is being deployed in complex circumstances, rather than the technocratic resolution of a question relating to specialised knowledge. Whenever public law power is being exercised in this way, it is wholly appropriate that it is subject to review for compliance with basic public law principles such as standards of legality. For example, exercising broad judgement implies that such as jurisdictional matters and statutory purpose are at large, and ought to be addressed by the courts using an appropriate degree of scrutiny. Understanding the regulatory function in terms of judgement also suggests that expertise is not the sole or even dominant consideration. There is also room for substantive public law values to be recognised, such as human rights standards. So while expert knowledge may insulate itself from judicial scrutiny, considered judgement must openly demonstrate any claim it makes to legitimacy. Regulators have no special claim to any unique capability to determine questions of the legitimate exercise of public power, and the courts are the appropriate body, on review, to ensure these standards of legitimacy are met.
These are pivotal challenges at the level of principle to the deferential approach to reviewing regulator’s interpretations. The general approach among proponents of deference is to overlook these challenges on pragmatic grounds. Regulation is justified on the basis of instrumental rationality rather than legality, and so the exercise of regulatory discretion engages ‘the language of efficiency, effectiveness, and economy’ rather than fairness, reasonableness and lawfulness.Footnote 29 But the long-term and wide-spread impacts of regulatory decision-making are often difficult to demonstrate, and claims based on expertise do not fare well when outcomes manifestly go awry. The authority claims of regulators to exercise meaningful discretion in these circumstances can appear somewhat threadbare in the absence of the legitimising effects of the law, even if law is not ultimately controlling. A legal rationale for deference is therefore a critical component of any credible argument that regulators’ own interpretations should be conclusive.
2. Chevron deference in the United States
It is here that the comparative influence of Chevron becomes relevant as the standard-bearer of deference to regulators’ interpretations.Footnote 30 The facts of the case are well‑known: the Environmental Protection Agency was required by the Clean Air Act Amendments 1977 to thoroughly review the creation of any major ‘stationary source’ of air pollution. While the Agency initially interpretated ‘stationary source’ to mean any significant addition to an existing plant or factory, from 1981 it adopted a new interpretation that ‘source’ referred to only an entire plant or factory. When this new interpretation was challenged on review, the US Supreme Court deferred to the Agency’s new interpretation. While the courts would insist on authoritatively determining the question if congressional intent pointed towards a clear meaning, in other cases the legislation might be ambiguous or silent. In those cases, the reviewing court was to leave room for the administering agency’s reasonable policy choices implicit in its own interpretation and ‘may not substitute its own construction of a statutory provision’.Footnote 31 In Chevron, ‘stationary source’ was found to be an ambiguous term not admitting of a single correct interpretation, and so the Agency’s reasonable interpretation of that term was not disturbed on review.
Chevron is a challenging precedent for those steeped in the Anglo-Commonwealth tradition of administrative law. In holding that the regulator’s interpretation is authoritative as a matter of law, it presents the opposite starting position to that which applies in the UK and New Zealand, which holds that ‘errors in construing primary or secondary legislation made by inferior tribunals … however specialised and prestigious they may be, are subject to correction by judicial review’.Footnote 32 The case also represented a significant change in the American legal position,Footnote 33 which immediately before Chevron was that ‘the courts are the final authorities on issues of statutory construction’.Footnote 34 The underlying rationale used to justify such a dramatic departure from international and historical precedent is therefore of considerable interest.
The key rationale in the case rests on implied statutory delegation. If Congress had explicitly delegated interpretative authority to the administering agency, then an agency interpretation would be controlling unless it was ‘arbitrary, capricious, or manifestly contrary to statute’.Footnote 35 The Court reasoned that an implicit delegation by Congress through legislative silence or ambiguity should be afforded the same standard of deference. On its own, however, this rationale is less than satisfactory. Why would it be assumed that Congress has delegated interpretative authority to a regulator over the judiciary when it is the judiciary who has the expertise and constitutional authority to determine questions of law? The answer lies in a second rationale, which implicates the regulatory agency’s claims to institutional competency. Where questions of legal interpretation involve reconciling conflicting policy objectives, judges do not have the relevant expertise,Footnote 36 and courts are not sufficiently accountable in political terms.Footnote 37 The Chevron doctrine therefore turns on judgement ‘about comparative competency [between the regulatory agency and reviewing court], undertaken in light of the regulatory structure and applicable constitutional considerations’.Footnote 38 In practice, this can lead to strong deference in a range of circumstances. The Chevron doctrine has led to deference being applied in the face of a regulator’s inconsistent interpretations,Footnote 39 in cases involving regulators overruling previous judicial interpretations,Footnote 40 and in cases determining the regulator’s own statutory jurisdiction.Footnote 41 In each of these cases, insistence on legality would arguably require a much less deferential approach.Footnote 42
Chevron stands in contrast to the common law position outlined in Section 1 above. From a comparative perspective, we can ask why such a different approach has been taken. Several possible reasons present themselves. The US has a particular need to strike a balance in the ‘constant competition for control’ over regulatory institutions as between Congress and the President, which may require a less legalistic interpretation approach.Footnote 43 Additionally, the relative difficulty of making and remaking US legislation compared to many Commonwealth systems may necessitate interpretative flexibility as a key means of updating regulatory frameworks.Footnote 44 Parliamentary systems with a closer alignment between legislative and executive branches may also be more responsive than in the US towards updating authoritative judicial interpretations of regulatory law where needed.Footnote 45 A third possible explanation is the passage of the Administrative Procedure Act 1946 (APA) in the US, which systemised regulatory consultation and resulted in a change of judicial emphasis away from legality in favour of matters of process and reasoning.Footnote 46 Oren Tamir suggests that more robust, legalistic review might be needed to compensate for a lack of embedded notice and comment procedures.Footnote 47 A fourth explanation relates to differences in the legislative drafting process. The UK and New Zealand have centralised legislative drafting models, reflecting both an exclusive conception of legal authority,Footnote 48 and a separation between legislative and policy functions.Footnote 49 The US has a more fragmented process of congressional drafting, which draws on multiple sources of expertise, including regulatory agencies themselves.Footnote 50 This indicates a unique systemic willingness in the US to allow federal regulators to undertake legal functions that is not present to the same degree in other jurisdictions.
These important jurisdictional differences should not be overlooked, and they make direct comparison between Chevron and other jurisdictions difficult. But it is beyond the scope of this paper to determine the precise balance of reasons for Chevron’s idiosyncratic approach. The key point is that outside of the US, Chevron has been taken to represent an important alternative method of resolving the tension between judicial review’s traditional insistence on legality with the facilitative, ‘green light’ model of administrative law. Even in jurisdictions like the UK and New Zealand, where legality is the starting point, there is still a need to balance judicial control with regulatory discretion in workable and legitimate ways. The underlying rationale in Chevron regarding implied statutory delegation and relative institutional competence presents judicial deference as a viable and useful addition to the judicial toolbox on review. This is the relevance of Chevron in common law jurisdictions: proof of the concept that deference has a doctrinal and theoretical legitimacy beyond raw considerations of strict legality review.
3. Loper Bright and the end of Chevron deference
But it is now unclear whether this proof of concept still holds. In June 2024, a majority of the Supreme Court overruled Chevron in Loper Bright Enterprises v Raimondo. The facts of Loper Bright concerned fishing vessels operating in US territorial waters that carry observers responsible for reporting to fishery management councils. In the case, the New England Fishery Management Council updated its rules so that herring fishing vessels operating in the Atlantic would (in certain circumstances) be required to meet the costs of carrying these observers. Loper Bright and other affected fishing businesses challenged the authority of the Council to make the new rule under the APA. In the lower courts, the governing statute was characterised as being ambiguous on the specific point, and so Chevron deference was found to apply.Footnote 51
Chief Justice Roberts delivered the opinion of the Court, presenting a judgment that resonates strongly with the common law approach to review based on legality. It starts with the proposition that the Constitution prescribes that legal interpretation is the responsibility of the courts. While regulatory interpretations may be afforded respect, and so inform any decision that a court may make,Footnote 52 those regulatory interpretations must not be allowed to supersede the interpretative authority of the judiciary. The Court reasoned that the APA reflects this axiomatic understanding of the judicial role. Indeed, the APA expressly requires the courts to ‘interpret constitutional and statutory provisions’.Footnote 53 Ultimately, ‘[t]he deference that Chevron requires of courts reviewing agency action cannot be squared with the APA’.Footnote 54
Chief Justice Roberts’ opinion also rejects the proposition that a statutory ambiguity should be treated as a presumptive delegation by Congress to the relevant agency.Footnote 55 Many statutory ambiguities are likely to be unintentional and, contra Chevron, the courts are usually well placed to adjudicate such ambiguities. Resolving ambiguity is the courts’ central task, ‘and a court is not somehow relieved of its obligation to independently interpret the statute’ where ‘the ambiguity is not a delegation to anybody’.Footnote 56 Chevron posited a broad rule of deference that could be insensitive to the degree of comparative expertise that the regulator actually has or deploys in respect of a given decision. The majority opinion therefore rejected the view that technical expertise changes the ordinary balance between legality and interpretative discretion because courts are expected to address statutory interpretation questions regardless of the subject matter.
These points echo the general approach of the courts in common law jurisdictions when scrutiny for legality is applied. The Court is, in effect, making a judgement call that the supposed benefits of agency decision-making are not sufficient to override these basic principles. The Supreme Court has not applied Chevron deference for a number of years, suggesting that its influence over the coherence of regulatory decision-making is over-stated. This is in material part a result of the fact that Chevron deference has failed to offer a consistent approach to the judicial role. The ‘gap’ or ‘ambiguity’ that stands as a prerequisite for deference only exists where the courts’ interpretative task has ended. But this is not an objective standard, and in the end turns on individual value judgements about the appropriate role of the judiciary in each individual case. As Eric Ip has observed, any finding of legislative ambiguity, indeterminacy or silence is ‘usually an interpretative conclusion rather than a premise justifying interpretation’.Footnote 57
The rhetoric throughout the Loper Bright majority opinion is strong. It expressly rejects Chevron on a principled basis that defends the courts’ ordinary role as interpreters of legislation. But, as must be the case when adjudicating the workings of the administrative state, there are concessions towards a more balanced approach. Take the following passage:Footnote 58
When the best reading of a statute is that it delegates discretionary authority to an agency, the role of the reviewing court under the APA is, as always, to independently interpret the statute and effectuate the will of Congress subject to constitutional limits. … By doing so, a court upholds the traditional conception of the judicial function that the APA adopts.
If the regulator’s interpretation can take effect through interpretative rather than presumptive delegation, how significant is this change in approach really? It is perhaps too early to tell definitively, but that has not prevented some speculation. Adrian Vermeule has suggested that the substantive law has changed very little, with a different style of reasoning being employed but ultimately similar results being reached.Footnote 59 This speculative view is no doubt helped along by the minority’s dissenting opinion which emphasises at length the risk that courts will be drawn into resolving technical questions that are beyond their institutional capability. Vermeule’s approach suggests that this risk is unlikely to arise in practice. Courts may have more cause to examine the legal and factual detail of the decision before them on review – the difference between what Vermeule characterises as a presumptive ‘Chevron wholesale’ approach and a problem-by-problem ‘Chevron retail’ approach. But regulators’ interpretations can still be applied where necessary; Chevron deference is ultimately ‘reframed but not eliminated’.Footnote 60
Vermeule’s views are a timely reminder that one case does not signal the end of the administrative state. But the move from presumptive rule to interpretative inquiry is a genuine shift in approach that merits considered reflection. If nothing else, it means that reasons for deferring to regulators’ interpretations will be legal in character rather than premised on the language of efficacy and technocratic rationality that characterises modern administration. So, for example, we might well see more of the practice that Elizabeth Fisher and others have documented of regulators developing justiciable ‘yardsticks’ for decision-making, which contribute to the robustness of internal processes and which the courts can apply on review.Footnote 61 Developing and utilising yardsticks in this way involves the application of subject matter expertise in a context of legality, and while not simple to achieve it is the kind of responsible decision-making to which specialist regulatory agencies should aspire. This is why the end of Chevron deference is important, as it brings opportunities for discipline and a methodological coherence to review of statutory interpretation that will be familiar to public lawyers on both sides of the Atlantic.
In any case, presumptive Chevron deference has been abandoned, and a return to legal principle has been reasserted in the US. It is therefore apposite to reappraise approaches that call for deference in common law jurisdictions. Pragmatic and presumptive deference is now largely without purchase, no longer secured by the Chevron foothold. If examples of deference to regulatory interpretations of statutes cannot be justified on their own terms, then my argument is that the deferential approaches should be discarded in favour of re‑centring a public law understanding of legality.
4. Excessive regulatory deference in common law jurisdictions
In what follows, I examine two examples from common law jurisdictions where Chevron-style deference has been applied despite the basic principle that legal questions are fundamentally judicial matters. I am critical of these cases – I consider that on detailed examination they can be shown to be wrongly decided because they are overly deferential. The decisions are ideological rather than interpretative, perhaps even more so than standard Chevron deference would usually permit. This is, in my view, problematic.
To be clear, my argument is not that courts in the UK or New Zealand routinely defer to expert regulators. While my instinct is that courts regularly do defer to economic regulators, demonstrating this requires extensive engagement with various doctrinal manifestations of deference across multiple cases, which is beyond the scope of the present analysis. The more limited claim that I do defend is that there is more potential for Chevron‑style deference than is usually appreciated. In providing examples of where this potential is realised, I mean to question the commitment to common law legality and the interpretative authority of the courts. Both cases result in highly deferential outcomes, but until now there has been little meaningful critique of those results. To show precisely how far deference has been applied over and above adherence to common law principle, this section concludes with a brief consideration of the Canadian experience with correctness review. Even in jurisdictions like Canada that endorse deference at the level of principle, the UK and New Zealand examples appear to be anomalous.
(a) Regulatory deference in the United Kingdom
The first case I consider is the House of Lords decision in South Yorkshire Transport,Footnote 62 which involved a challenge to a merger under the Fair Trading Act 1973 (UK). Reform in the mid‑1980s deregulated bus transportation services, resulting in a significant period of consolidation through mergers and acquisitions. This consolidation activity drew political ire, which motivated the Secretary for Transport to refer a particular acquisition by South Yorkshire Transport to the Monopolies and Merger Commission (MMC) for assessment. Before accepting the reference, the MMC was required to satisfy a jurisdictional pre-condition that the acquisition affected ‘a substantial part of the United Kingdom’.Footnote 63 The MMC found that the condition was satisfied based on its own established interpretation that ‘substantial’ meant ‘something real or important as distinct from something merely nominal’.Footnote 64 In applying this interpretation, the MMC found that although the geographic area affected by the acquisition was minor in terms of both area and population, it had a social and political significance disproportionate to its size.
When the MMC ultimately recommended divestiture of the acquired assets, South Yorkshire Transport appealed on the basis that the jurisdictional pre-condition had not in fact been satisfied. Even on a substantive competition assessment, an acquisition affecting less than 5% of the national market could not realistically be considered ‘substantial’, and so the MMC must have been acting beyond its jurisdiction. Further, the MMC was an economic regulator, and so reference to ‘social’ and ‘political’ values outside of economics was inappropriate. In accordance with basic legal principle, South Yorkshire Transport argued, saying so was definitively a matter for the courts.
The House of Lords disagreed with South Yorkshire Transport line of argument, finding that whether the pre-condition had been satisfied was a matter for the MMC to determine. The judgment manifests an evident concern to interpret the statutory language in an enabling manner,Footnote 65 such that it would ‘not unduly fetter the judgment of the commission’.Footnote 66 As a result, the House of Lords explicitly rejected the South Yorkshire Transport argument that questions of jurisdiction are ‘hard-edged’ in the sense that either ‘the commission had jurisdiction or it had not’. The judgment seems to accept, albeit opaquely, that as a general matter the respondent’s position is entirely correct under English law. However, there are exceptions to this general position because:Footnote 67
… the criterion so established may itself be so imprecise that different decision-makers, each acting rationally, might reach differing conclusions when applying it to the facts of a given case. In such a case, the court is entitled to substitute its own opinion for that of the person to whom the decision has been entrusted only if the decision is so aberrant that it cannot be classed as rational.
That more exceptional position is the one that the House of Lords applied. The statutory term ‘substantial’ was vague and broad enough to call for regulatory judgement rather than precise determination.Footnote 68 As such, there were no legitimate grounds for the courts to interfere.
South Yorkshire Transport is often viewed as the UK analogue of Chevron, and it can be easy to see why.Footnote 69 It treats statutory ambiguity as a signal that determination of a legal question is not for the courts, and defers on a point of legal interpretation in respect of which the courts would usually be expected to take an authoritative view. The scope and effect of deference in this case is therefore exceptionally broad without the benefit of close judicial scrutiny.
But for all that, the doctrinal influence of South Yorkshire Transport is less than clear. It certainly represents the high point of judicial deference in a regulatory context, but the case itself is not widely or deeply analysed in the UK scholarship. A key reason for this lack of detailed attention may be due to the establishment of a specialist tribunal as an avenue to challenge regulatory decision-making in the specific economic regulatory context that the decision is primarily concerned with. The Competition Appeal Tribunal has statutory jurisdiction to hear both review and appeal cases, with the latter jurisdiction allowing it to inquire into the substantive merits of a regulatory decision. It hears matters determined by the UK’s competition agency, the Competition and Markets Authority, as well as sector-specific economic regulators such as Ofwat, Ofcom and Ofgem. It is notable, however, that even within the context of specialist review, including in respect of the merits of a regulatory decision, the instinct towards deference is a strong one. Despoina Mantzari’s research into economic regulation suggests that where the Tribunal’s decision is itself appealed to a generalist court, that generalist court tends to prefer that deference be afforded to the initial decision-maker.Footnote 70 Despite the way economic regulation is often structured through administrative law frameworks and processes as an exercise of public power, courts appear deeply inclined to view it as a specialist activity that sits outside of the usual framework for analysis.
Another reason is that the question of deference is not always broached directly. The error of law standard serves as an example here, with matters of law sometimes being treated as factual matters for pragmatic reasons rather than reasons of principle. Perhaps, despite the matter giving rise to a legal question, the courts determine there is no meaningful legal content.Footnote 71 In such a case, there is no legal content for the court to investigate, and the original decision-maker’s view is likely to stand. This is different from the type of issue raised by Chevron deference, although there is some overlap. Introducing fluidity into the distinction between law and fact on pragmatic grounds increases the ‘grey area’ where a specialist body’s expertise may legitimately ‘shape and direct the development of law and practice in that field’.Footnote 72 In this sense it results in a degree of deference, but only after explicit consideration of the legal content of the relevant issues and the appropriate role of a reviewing court. Thankfully, many examples of ‘deference’ in practice are the result of reasoning about law, not conclusory assertions about non-legal expertise.
Despite these factors, the case remains relevant for two reasons. The first reason is that it demonstrates the deferential potential of the UK common law even as it purports to be based on standards of legality rather than regulatory empowerment and facilitation. It is an extreme example, but as a precedent it has not been overruled and could be applied in new contexts. The second reason South Yorkshire Transport remains relevant is that it can be taken as an exemplar by those who champion judicial deference. In arguing that the UK position should be more openly influenced by the deferential approach that applies in Canada in particular, Paul Daly argues that the reasoning in South Yorkshire Transport should be applied more broadly by English courts.Footnote 73 Daly’s view is that ‘[t]he exception developed by the House of Lord in South Yorkshire Transport should be extended to cover administrative interpretations of law’ so that ‘all statutory terms can, in principle, be treated as evaluative’ leaving room for regulatory interpretation to displace judicial scrutiny.Footnote 74 For these reasons, whether South Yorkshire Transport should be accepted as good law is a matter that still stands to be addressed.
Once a detailed examination of the reasoning in the case is undertaken it becomes clear that a deferential approach to legal interpretation was wholly inappropriate. An ordinary interpretation of the MMC’s preferred meaning of ‘substantial’ as ‘something real or important as distinct from something merely nominal’ would have resulted in the application of a de minimis legal test rather than the exercise of expert policy judgement. And, if a legal test was being applied then this would more readily require close judicial assessment as to the appropriate legal standard. The Lords undertook convoluted linguistic analysis to suggest that the MMC had not really applied a de minimis test in order to set up the finding that the courts could not intervene because policy judgement was needed.Footnote 75 But a de minimis test is precisely what the MMC had developed and applied in this instance and, in respect of its previous decision-making, was an approach that by this time was well understood and expected by the transportation industry.Footnote 76 The term ‘substantial’ was therefore not one that had a ‘protean nature’,Footnote 77 nor is its nature ‘evaluative’ rather than legal in the ordinary sense.Footnote 78 The felt need to deliver a judgment that was enabling and facilitative of the MMC’s policy discretion seems to have led the Lords to overlook the possibility that the jurisdictional question could, and perhaps should, have been addressed in legal terms.
It is also difficult to understand precisely what regulatory policy work the MMC’s interpretation is advancing. That interpretation, and the ultimate decision to accept the reference, is manifestly not justified in terms of well-functioning economic markets, which would be expected if regulatory expertise was actually being deployed in a merger context. Instead, vague references to social and political significance are relied on to extend the MMC’s jurisdiction in a manner that may not at all be consistent with regulating merger activity on economic grounds. Here we might acknowledge that a more insightful US comparison than Chevron might be Brown Shoe Co. Footnote 79 Infamously, Brown Shoe ruled unlawful a merger in shoe manufacturing and retail markets with an extremely marginal increase in market concentration because there was a chance that it would adversely affect small business owners. The case has been widely criticised, and that criticism led to more economically coherent approaches being adopted in subsequent cases. It is not difficult to imagine that the jurisdictional threshold represented by ‘substantial’ was intended to avoid a UK equivalent of Brown Shoe, with economic rather than social and political significance being the key consideration. The Lords’ deferential approach left this question not only unanswered, but unasked.
Some commentators have distinguished South Yorkshire Transport from Chevron on the basis that the former case reserves the authoritative determination of interpretative questions for the courts.Footnote 80 The Lords state that the courts retain the authority to determine the ‘criterion for judgment’, which some have taken to mean that it remains for the courts to provide an authoritative view on the scheme of the statute.Footnote 81 But where the ‘criterion for judgment’ remains vague, various applications are consistent with that scheme. This is where policy judgement enters, those who insist on a deferential approach may argue, and the courts should defer to any interpretation that is within the relevant lawful boundaries. This is a sensible rule of administrative law in principle, affording room for a well-placed regulator to craft a substantive policy response within the confines of broad legal boundaries. But again, in the actual case any substantive policy content that might justify deference appears to be absent. The actual criterion is determined to be ‘worth consideration for the purposes of the Act’,Footnote 82 which results in a rather circular chain of logic where a jurisdictional pre-condition is satisfied by reference to a regulator’s view of its own jurisdiction. This contributes nothing to the legal structure of the regime, but instead conflates the regulator’s policy judgement with the legal issue of jurisdiction. With respect, this is an awkward result.
The awkwardness becomes ever more apparent in the difficulties textbook writers have had in assimilating South Yorkshire Transport into the doctrinal canon. One approach is to claim that the case follows from the Anisminic and Page line of cases that no regulatory authority can determine its own jurisdiction.Footnote 83 This approach appears to place emphasis on the general rule expressed in the case without engaging with the fact that the case itself explicitly stands as an exception to that general rule, and so is descriptively unsatisfactory. A second approach is to rationalise away the jurisdictional nature of the pre-condition.Footnote 84 This is to downplay the legal significance of the precise words used, and therefore open more space for legitimate regulatory interpretation. The statutory scheme itself in South Yorkshire Transport seems to count against this approach, as the only function of ‘substantial’ would seem to lie in defining the MMC’s jurisdiction. A final approach is that the decision is simply treated as sui generis. Paul Craig, for instance, summarises the relevant legal position on questions of law as being that the courts will review all errors of law subject to three qualifications. Those qualifications are that the error must be relevant, the institutional nature of decision‑maker may be relevant, and finally that ‘[t]he court will not necessarily substitute its judgment for that of the agency in circumstances like those in South Yorkshire Transport’.Footnote 85 The decision is effectively confined to its own facts. Such bespoke treatment underscores how the case challenges rather than conforms with administrative law doctrine.
In sum, the justification for judicial deference on the basic legal question in South Yorkshire Transport is not obvious. Indeed, close legal scrutiny might have addressed matters of regulatory substance as well as appealing to the instinct to control legality. Perhaps, in the end, the MMC’s decision to accept the reference could have been explained and justified as the correct one from a legal standpoint. But equally, and in my view more in line with principle, the decision could have been made that the reference area was simply not ‘substantial’ for the purposes of the legislation.Footnote 86 The Lords’ overly deferential approach means that we simply cannot know for sure.
(b) Regulatory deference in New Zealand
The New Zealand equivalent of South Yorkshire Transport is Unison Networks. Footnote 87 Prior to 2008, electricity distribution businesses in New Zealand were subject to a ‘thresholds’ based regime of rate control. This approach called for the setting of crude thresholds as a screening device to determine whether directive rate control based on actual cost components and revenue requirements was warranted. If a business breached the threshold set by the regulator, the Commerce Commission, an intensive investigation would be conducted into whether administrative control of that business’s pricing was warranted.
When setting the thresholds, the Commission’s approach was to adopt a policy of administrative convenience rather than economic principle. Rather than setting threshold prices based on an estimate of reasonably incurred costs or future investment needs, the Commission simply required that all electricity distribution businesses maintain their current rates.Footnote 88 As current rates were determined by a monopoly service provider without regard to the drivers of rate control regulation (investment and efficiency incentives, restricting excessive profitability, consumer interests in quality and availability), it manifestly did not further or achieve any legislative policy purpose. An electricity distribution business challenged the Commission’s decision, arguing that the thresholds could not function as an effective screening device, and so failed to address the stated policy purpose ‘to promote the efficient operation of markets directly related to electricity distribution’.Footnote 89 The issue for the Supreme Court was whether the thresholds were set lawfully, with the applicant arguing improper purpose.
The Court’s analysis invoked the need to maintain standards of legality on review, making it quite clear that public decision-makers such as the Commission ‘must exercise their statutory powers in accordance with the statutes which confer them’.Footnote 90 The Court continued:Footnote 91
A statutory power is subject to limits even if it is conferred in unqualified terms. Parliament must have intended that a broadly framed discretion should always be exercised to promote the policy of the Act. These are ascertained from reading the Act as whole. The exercise of the power will be invalid if the decision-maker ‘so uses his discretion as to thwart or run counter to the policy and objects of the Act’. A power granted for a particular purpose must be used for that purpose …
Given the seemingly arbitrary nature of the Commission’s decision, it might be expected that this statement of basic principle was a precursor to judicial intervention. However, in an abrupt volte-face, the Court went on to heavily qualify this statement of principle so that it had virtually no application in the case before it:Footnote 92
[I]n this case, a public body, with expertise in the subject matter, is given broadly expressed power that is designed to achieve economic objectives which are themselves expansively expressed. In such instances Parliament generally contemplates that wide policy considerations will be taken into account in the exercise of the expert body’s powers. The courts in those circumstances are unlikely to intervene unless the body exercising the power has acted in bad faith, has materially misapplied the law, or has exercised the power in a way which cannot rationally be regarded as coming within the statutory purpose.
When reviewing the statutory scheme and language, the Court concluded that the operative provisions of the Act do not ‘explicitly stipulate any necessary attributes of thresholds other than to say that thresholds can be expressed in quantitative or qualitative terms’.Footnote 93 Instead, the Act contemplates that ‘the Commission itself will fashion thresholds that in its judgment are appropriate’ given that it is the ‘body with expertise in the field’.Footnote 94 As the applicant’s challenge did not amount to a provable allegation of bad faith, material unlawfulness or irrationality, the application for review failed. In short, the Supreme Court rested its decision not to intervene on the grounds of deference to a regulatory body with economic policy expertise.
The Unison Networks case has had an astounding influence on the development of administrative law. It has been cited numerous times for the proposition that regulatory decisions should be afforded a high measure of judicial discretion.Footnote 95 In respect of economic regulation in particular, the overall result has been that regulatory decisions are effectively unreviewable.Footnote 96 This is particularly concerning given that the rationale for deference based on Unison Networks is so threadbare in both doctrinal and theoretical terms. Marcelo Rodriguez Ferrere and Andrew Geddis have speculated that part of the influence of the case stems from the fact that it is a rare example of the Supreme Court providing in-principle guidance on administrative law.Footnote 97 This makes it an obvious place for the lower courts to reach to when they attempt to balance the competing drivers of legality and deference. But none of this speaks to the credibility of the decision, and whether it deserves such a following.
It should also be recognised that while Unison Networks is the high point in terms of judicial deference to regulators, it was itself the product of a trend towards increasingly deferential approaches. The door was opened by recognition that administrative practice could be influential in determining questions of statutory interpretation,Footnote 98 although the question of the weight to be afforded to that practice in the calculus of the particular decision was largely left unanswered. There is also academic commentary suggesting the actual approach in New Zealand in respect of certain grounds of review was not all that different from the Chevron line of cases.Footnote 99
New Zealand’s leading public law scholar has labelled Unison Networks administrative law ‘orthodoxy’.Footnote 100 This reading is based on abstracting away from the details of the specific decision in order to focus on the broad statement of underlying principles articulated in the case. From this abstract point of view, Unison Networks can be, and is, taken as the New Zealand analogue of Padfield. Footnote 101 But this reading of the case elides rather than addresses the fact that the actual decision in the case clashes with this broad statement of abstract principle. The actual result in Unison Networks was considerable deference to the regulatory agency on a point of law. From that perspective, the effective non-reviewability of delegated decision-making on all but the most serious grounds stands as a distinct motif of Unison Networks. And, with respect, that motif appears to fall into conflict with the laudable principle commonly associated with Padfield in a way that requires explanation and justification rather than perfunctory acceptance. As already noted, the basic proposition in New Zealand is that the courts have exclusive authority to determine questions of law.Footnote 102 Deference in specific cases is therefore an approach that requires explication, although neither the Court itself nor supporters of its decision explain how the deferential result is to be reconciled with judicial control of administrative discretion. The claim of ‘orthodoxy’ is simply too strong a description when the detail of the case is examined closely.
Unison Networks more explicitly invokes themes of legislative delegation and regulatory expertise than South Yorkshire Transport. Like Chevron, we can take the two justifications together, understanding that implicit delegation is an acceptable reading of the governing legislation because of the fact of regulatory expertise, which the courts do not have. Indeed, although Chevron is not mentioned in the judgment it seems to be an inspiration for the Court. This is perhaps most obvious in the residual grounds for judicial intervention, where irrationality, bad faith, and material misapplication of the law align neatly with Chevron’s reference to ‘arbitrary, capricious, or manifestly contrary to statute’.Footnote 103 But like South Yorkshire Transport we can question what expertise is actually being utilised in the Commission’s decision-making. The decision to roll over existing pricing into the new regulatory regime was completely arbitrary from the standpoint of micro-economic policy. The Commission may be administering a regulatory scheme, but it is not utilising specialist expertise to do so. Acknowledging this point invites us to reassess whether it can be so readily assumed that the statutory scheme contains an implicit delegation in favour of the Commission and away from the courts.
Any detailed assessment of this kind is likely to engage with what Paul Daly has termed the ‘practical justifications’ for deference evident in the governing legislation.Footnote 104 On proper examination of the statutory framework, these practical justifications reveal the reasons underlying the decision to delegate important decision-making powers to a body other than the legislature or courts, and so inform the scope of any delegation (and therefore the extent of deference), as well as the nature of any legal and procedural requirements that validate delegated decision-making. This is a highly contextual approach to justifying deference, but it maintains a degree of structure and rigour given that it is based in the statutory framework that empowers administrative decision-making. Such an approach could supply an institutional rationale for deference where the relative competencies of the regulator and the courts are taken into account. But in Unison Networks the Commission’s superior expertise relative to that of the Court is simply assumed, without any assessment of the nature and limits of deference implied by the statutory scheme, the context of the individual regulatory decision or the potential role of the courts in validating the regulatory action. Indeed, recourse in the Court’s reasoning to broad-based legislative intention seems to only obscure the fact that the Court is not willing to engage at all with the underlying economic policy context or the nature and limits of the Commission’s expertise. Without more detailed justification, the decision to defer to the Commission on this point of law seems out place.
(c) Regulatory deference in Canada
In the introduction I noted that Canada has embraced deference, but in an alternative way to Chevron. This is not the place for a detailed assessment of the Canadian position, but brief reference to some example cases is useful to contextualise South Yorkshire Transport and Unison Networks.
In Canada, deference in the form of a ‘reasonableness’ standard of review has been the dominant standard of review for some time, although in some circumstances a more exacting ‘correctness’ standard of review is still employed. The current position, set out in Vavilov,Footnote 105 clarifies that reasonableness is the default position, although Vavilov has also brought Canadian administrative law back closer to a Diceyan vision of review based on legislative intention.Footnote 106 It is not always clear, however, how cases analogous to South Yorkshire Transport and Unison Networks might be treated under this bifurcated framework. Are statutory appeals on questions of law subject to a correctness standard, for example? In Vavilov’s companion case of Bell Canada v Canada (Attorney-General) this proved decisive,Footnote 107 with the Court finding that the Canadian Radio‑television and Telecommunications Commission had exceeded the limits of its jurisdiction after determining the issue for itself, without deference. Interestingly, the Court also seemed to put weight on the fact that the appeal raised ‘legal questions that go directly to the limits of the CRTC’s statutory grant of power’.Footnote 108 This last statement is difficult to square with Vavilov, and I think it reveals the difficulty in principle with subjecting questions of jurisdiction or statutory purpose to heightened deference on review. If South Yorkshire Transport or Unison Networks were approached in the same way as Bell Canada, judicial scrutiny would almost certainly have been more meaningful and exacting.
Perhaps the closest Canadian analogy to South Yorkshire Transport and Unison Networks in terms of the underlying facts is the (now rather dated) decision of the Supreme Court in ATCO Gas and Pipelines. Footnote 109 In that case, the question involved the authority of a public decision‑maker to mandate the distribution of proceeds by a public utility. Despite the deferential approach that usually characterises Canadian administrative law jurisprudence, the Court found that it could comfortably and appropriately determine the question itself. The justification for the decision turned in part on the fact that the decision-maker had ‘no greater expertise’ than the Court because the relevant statutory provisions had ‘no technical aspect’ to them. Those provisions did invoke vague standards such as ‘public interest’ where policy expertise might have been expected to inform any decision made. The respective courts in South Yorkshire Transport and Unison Networks would almost certainly have demurred in respect of such an open-ended standard. But the Supreme Court of Canada proceeded with confidence that the matter ultimately involved a legal question where its own expertise was both relevant and sufficient. This seems to me to be a more appropriate means of addressing questions of law than the approach taken in South Yorkshire Transport and Unison Networks. Legal questions with legal consequences are treated as matters for the reviewing court.
Paul Daly has argued that the decision in ATCO Gas and Pipelines should be construed narrowly. Relying on the Court’s finding that the relevant provisions ‘have no technical aspect’, Daly suggests that ‘[b]y negative implication, if reviewing the interpretation of narrow, technical concepts, a court should follow accord curial deference, even if the concepts were jurisdictional’.Footnote 110 But this does not necessarily follow. It may seem counter-intuitive, but the more narrow and technical a question, the more it is likely to admit of a single correct answer which the courts can determine. In those situations, many of the standard arguments for according deference based on complexity, polycentricity and public interest policy considerations fall away or simply do not apply to a determinative extent. Regulators and agencies can more directly demonstrate the correctness of any particular approach rather than seeking to rely on a margin of appreciation for it exercising judgement, and this can be factored into the legal analysis undertaken by the court.
The key point for present purposes is that the types of judicial decision represented by South Yorkshire Transport and Unison Networks seem to go beyond that available in even the most deferential jurisdiction in the Anglo-Commonwealth canon. With the overruling of Chevron by Loper Bright, any remaining justification for those cases has disappeared.
5. A call to end Chevron-style deference
My argument has marshalled together a number of reasons to reject Chevron-style deference in the UK and New Zealand. It is time to bring the separate threads together. Chevron deference is at odds with basic common law principle, which holds that questions of law are for the courts to determine authoritatively. The main alternative to this view, the Chevron case itself, has now been categorically overruled by the US Supreme Court. This overruling by Loper Bright brings US law closer to the general legal position in the UK, while casting serious doubts on outlier cases where more extreme deference has been applied. It also directly questions the sustainability of the standard approach in New Zealand. Presumptive approaches to deference based on implied statutory delegation and assumed regulatory expertise have been shown to be ineffective and have been rejected. To the extent that Anglo-Commonwealth administrative law retains scope to accommodate presumptive deference to regulators, it should be reformulated. Confirming that it is for the judiciary to determine questions of regulatory law will avoid arguably incorrect results like those on show in South Yorkshire Transport and Unison Networks. This will make the law more consistent and coherent overall. It will also contribute materially to the legitimacy of regulatory decision-making.
There is, unfortunately, a risk that the argument presented here is treated as taking one side in the common ‘duality … evident in much of the literature’ that views administration in either a sceptical or optimistic light.Footnote 111 Calling for an end to excessive deference in respect of regulator’s interpretations of statutory provisions might superficially be seen to fall on the sceptical side of the debate, perhaps representing the product of an ideological commitment to law rather than politics and policy, liberalism rather than progressivism, and courts rather than regulators. This reading of my argument misconstrues it. All modern legal systems need to accommodate a balance between the legitimacy of independent review processes – including for legality – and regulatory discretion and flexibility. My argument is that a doctrine of presumptive deference does little to promote this balance. The law of judicial review engages in striking this balance in every instance through consideration of the justiciability of claims, articulation of grounds of review, and the application of remedial discretion. But all these features of modern judicial review doctrine, unlike naked deference, turn on the application of public law analysis to reach reasoned and coherent decisions.Footnote 112 Regulator discretion should be the result of the application of public law doctrine, not a product of the denial of the relevance of that doctrine. The Loper Bright case and the end of the Chevron doctrine presents an opportunity to reappraise the notion of deference to expert regulators in those common law jurisdictions where it has sometimes been pushed too far. Both the law and administration of regulation will be better for it.
Open access
