• Specific Year
    Any

Groves, Matthew --- "Substantive Legitimate Expectations in Australian Administrative Law" [2008] MelbULawRw 15; (2008) 32(2) Melbourne University Law Review 470

SUBSTANTIVE LEGITIMATE EXPECTATIONS IN AUSTRALIAN ADMINISTRATIVE LAW

MATTHEW GROVES[∗]

[Judicial review of administrative action has traditionally had a procedural focus. This means that courts examine the procedure by which a decision is made, rather than the decision itself. A denial of natural justice is no exception to review a person dissatisfied with an administrative decision has long been able to complain about the fairness of the decision-making process but not the fairness of the decision itself. English law has recently developed a doctrine of ‘substantive unfairness’ by which an expectation about the outcome of a decision-making process can be protected by the courts in a strong sense. The strength of the protection given under this new doctrine seems to blur the distinction between process and outcomes, which leads judicial review in a radical new direction. This article explains the English doctrine of substantive unfairness and considers whether it can and should be adopted in Australia.]

CONTENTS

I INTRODUCTION

Governments and their agents may create expectations regarding the manner in which administrative powers will be exercised. Expectations of this nature can be generated in many different ways, such as by the issue of policies or procedures to guide the exercise of discretionary powers. Expectations regarding the future exercise of administrative powers may also be generated by public statements or representations, perhaps even promises, or by adoption and regular application of a certain practice. But just as expectations about the exercise of administrative powers may be created, they may also be disappointed. They may be disappointed when a governmental agency has acted in breach of a promise or undertaking made to a particular person or to a class of persons. They may also be disappointed when a government agency has not applied current policy or guidelines in determining a particular case, and without good reason. In such a case, the complaint may be that the policy has been applied inconsistently, perhaps in a way which reflects improper discrimination. In other cases, an existing policy may be changed and a new one applied to the disadvantage of people who stood to benefit from the earlier policy and who may even have conducted their affairs in reliance upon it.

Courts in England and some other jurisdictions have recently accepted that there can be circumstances in which government agencies should be required to fulfil the legitimate expectations they have created.[1] This approach endows an expectation with a substantive quality because it enables the expectation to determine or strongly influence the outcome of, rather than simply the procedures for, administrative decision-making. Australian courts, in contrast, have generally taken the view that expectations about the exercise of administrative powers may only give rise to procedural rights.[2] On this view, an expectation about the exercise of an administrative power might, at best, oblige a decision-maker who intends to act contrary to that expectation to notify affected people and provide them with an opportunity to argue against that course. But the law in Australia imposes no restraints upon a decision-maker beyond these procedural requirements.

This article examines the different approaches governing legitimate expectations in England and Australia. It traces the development of the English approach by which courts can now require governments and their agencies to honour expectations they have created. The article also considers whether it is open to Australian courts to adopt a similar approach without violating fundamental constitutional principles. It will be argued that the increasing role of the Australian Constitution as a source of guiding principle in Australian judicial review, and the associated conceptions of the separation of powers and the limitations on judicial power that flow from the Constitution, preclude any judicial enforcement of substantive legitimate expectations in Australia. But first, attention must be given to some preliminary questions. What precisely is a ‘substantive legitimate expectation’? How may it arise? And how does it differ from the more traditional ‘procedural legitimate expectation’?

II THE CONCEPT OF LEGITIMATE EXPECTATIONS

The scope of the duty to observe the requirements of procedural fairness is now extremely wide. It is well-settled that the duty extends to virtually every exercise of a statutory power which might have an adverse effect on an individual unless there is a very clear legislative indication to the contrary.[3]

Therefore, in almost all cases the important question now is not whether the requirements of procedural fairness apply but what they require in a particular instance. But that was not always the case. During the evolution of procedural fairness, or natural justice as the doctrine was commonly called in this earlier period, many cases focused on the ‘threshold question’ of whether the doctrine applied. The answer to this preliminary question often depended on whether the courts could identify a particular reason or circumstance why natural justice ought to apply. The doctrine of legitimate expectation contributed to the expansion of the duty to observe the requirements of natural justice by extending the duty beyond the relatively narrow range of rights and interests to which natural justice had traditionally applied.[4]

The legitimate expectation doctrine was invoked in a range of cases, the common theme of which was the principle that when administrative officials had created or induced a belief in a person about the possible exercise of their powers, any change affecting this belief should be conditioned by the rules of natural justice. The earliest cases involved people who held a licence, permit or visa which entitled them to enjoy a particular benefit.[5] In these cases, it was held that the grant of the licence, permit or visa created an expectation in the grantee that they would enjoy that benefit for its expected duration and that the benefit would not be ended prematurely unless the person was granted the right to argue against that course. The legitimate expectation doctrine provided important procedural benefits in these cases, namely, the right to be notified of, and to be heard in opposition to, the revocation of an existing benefit.

As the legitimate expectation doctrine gained acceptance, it was invoked in a wider range of cases, which can be conveniently summarised into four categories.[6]

The first was cases in which a person had relied upon a policy or norm of general application but was then subjected to a different policy or norm. The second category, which was a slight variation on the first, included cases in which a policy or norm of general application existed and continued but was not applied to the case at hand. A third category arose when an individual received a promise or representation which was not honoured due to a subsequent change to a policy or norm of general application. A fourth category, which was a variation on the third, arose when an individual received a promise or representation which was subsequently dishonoured, not because there had been a general change in policy, but rather because the decision-maker had changed its mind in that instance.

The legitimate expectation doctrine in these various manifestations was criticised as serving little purpose. More particularly, it was said to be a procedural device that added ‘little, if anything, to the concept of a right.’[7] But proponents of the legitimate expectation doctrine suggested that it enabled natural justice to extend beyond ‘enforceable legal rights’ to ‘expectations’ of various sorts.[8] That possibility provided an important bridge by which the rules of natural justice could venture into new territory.

Despite the growing body of cases in which the legitimate expectation doctrine was invoked and an increasing acceptance of the doctrine’s role in the evolution of the duty to observe the rules of natural justice, key questions about the doctrine remained. One difficulty arose from the frequently mentioned requirement that an expectation should be reasonable.[9] This requirement provided an apparently objective quality to the concept and, therefore, was thought to provide a useful limit by precluding the recognition of expectations that were somehow unrealistic or inappropriate. However, the logically related issue to any requirement of reasonableness, which has been a longstanding source of uncertainty, was whether an expectation ought to be assessed in subjective or objective terms. The requirement that an expectation be reasonable poses the question of ‘reasonable according to whom?’

A second difficulty was whether a person who raised a legitimate expectation needed to also prove reliance upon it.[10] The point that underpinned any requirement of reliance was the extent to which the legitimate expectation doctrine, and administrative law more generally, should be influenced by considerations of estoppel. The possible influence of estoppel shrouded important related questions. To what extent is it appropriate to use private law concepts in the law that relates to the exercise of public powers? Can private law concepts be used in public law with any theoretical coherence when there is longstanding authority that crucial aspects of private law, particularly the right to damages, do not extend to public law?[11]

The final difficulty was the extent to which the legitimate expectation doctrine might extend to determining actual outcomes in administrative decision-making, as opposed to procedural requirements. An expectation of this last kind — a substantive legitimate expectation — is based on a promise or representation about an actual advantage or benefit. They can be distinguished from all forms of the traditional legitimate expectation (which are procedural legitimate expectations) because the latter are confined to the procedure to be followed before a decision is made. A majority of the Hong Kong Court of Final Appeal defined the substantive legitimate expectation doctrine in the following terms:

The doctrine recognizes that, in the absence of any overriding reason of law or policy excluding its operation, situations may arise in which persons may have a legitimate expectation of a substantive outcome or benefit, in which event failing to honour the expectation may, in particular circumstances, result in such unfairness to individuals as to amount to an abuse of power justifying intervention by the court.[12]

The substantive legitimate expectation doctrine commonly arises in two scenarios. The first is when a person who enjoys a benefit or advantage argues that they expect that the benefit or advantage will continue. In this instance, the substantive legitimate expectation can effectively preclude a decision-maker from exercising a discretionary power to revoke the benefit or advantage because revocation is only permitted in very limited circumstances. The other scenario is when a person does not yet enjoy a benefit or advantage but argues that they rightfully expect that it will be granted. In this instance, the substantive legitimate expectation can effectively force decision-makers to grant the benefit or advantage because the court can require decision-makers to take account of both the substantive legitimate expectation and the circumstances upon which it is based. The important quality in each form of substantive legitimate expectation is that it leads a court very close to determining the outcome of administrative decision-making, rather than only its procedure. This move from procedure to substance is a radical one that takes judicial review of administrative action well beyond its traditional boundaries. The next Part of this article explains how this radical step occurred in England and the subsequent refinements which have been made to that doctrine.

III SUBSTANTIVE LEGITIMATE EXPECTATIONS IN ENGLAND

The decision in R v North and East Devon Health Authority; Ex parte Coughlan (‘Coughlan’)[13] marked the decisive English acceptance of substantive legitimate expectations, or substantive unfairness as it is known in England. However, key elements of the doctrine were developed 15 years earlier in the House of Lords decision in Re Preston (‘Preston’).[14] Preston alleged that he had reached an agreement with tax authorities by which he would pay an amount of tax and withdraw his outstanding claims, and the tax authorities would cease investigating him. The claim failed because Preston could not prove the existence of any agreement or undertaking, but the House of Lords made it clear that if an agreement or undertaking had been proven Preston could have sought judicial review of its breach on the ground of ‘unfairness’. Their Lordships rested this conclusion on a curious blend of public law fairness and private law estoppel. They concluded that the tax authorities were obliged by statute to act ‘fairly’. This obligation would, in some cases, prevent the tax authorities from acting in a manner that could amount to a breach of contract, or the breach of a representation that would give rise to an estoppel, if the tax authority making the representation were a private firm rather than a public authority.[15] If the tax authorities acted unfairly according to these principles, which were guided to an uncertain extent by estoppel, the resulting decision would amount to an abuse of their statutory powers.[16]

Several comments can be made about the Preston case. First, the House of Lords placed no reliance upon the requirements of natural justice, which suggests that their Lordships conceived ‘fairness’ as something quite distinct from natural justice. Once ‘fairness’ is separated from natural justice, it is a small conceptual step to accept that it could form an independent ground of judicial review as appears to have later happened with substantive unfairness.[17] Secondly, fairness in this sense has a strong connection with private law, notably estoppel. The reasoning of the House of Lords suggests that fairness, in a form enforceable against a public official, would arise in situations similar to those of equitable estoppel.[18] Thirdly, the House of Lords did not suggest that this new form of fairness could enable or require a public official to act beyond or contrary to the limits of their statutory powers. Accordingly, fairness could not be used to enforce an ultra vires agreement or undertaking. Even if an undertaking or agreement was within power, it would not become automatically enforceable. Their Lordships made it clear that the tax authorities could not simply make a binding promise regarding the exercise of their powers (in the form of failing to pursue a claim). Such a promise or undertaking would normally conflict with the basic duty of tax authorities to collect revenue,[19] though the Lords made it clear that there could be special circumstances in which it would be unjust or unfair for the tax authorities to enforce this basic duty.[20] Their Lordships suggested that the decisive factor was whether enforcement of a liability by tax authorities would breach an undertaking or agreement.[21] On this view, the tax authorities could resile from an undertaking or representation if new evidence arose or the circumstances of the case changed significantly, but outside of those instances any attempt to resile from an undertaking could amount to an abuse of power.

The suggestion that unfairness could amount to an abuse of power foreshadowed the rise of substantive unfairness as a separate ground of judicial review, but it also raised several questions. Perhaps the most important was why or how unfair behaviour by a public official could amount to an ‘abuse of power’ sufficient to constitute a separate ground of review and attract judicial relief. Professor Paul Craig argues that the content and circumstances of the representations of the public official are crucial: unfairness can be transformed into an abuse of power that is an error of law if the representations create expectations that are normatively justified and lead to reliance on the part of the person affected.[22]

Unfairness of this kind differs from the traditional procedural legitimate expectation in two ways. First, it requires a subjective belief on the part of the person affected, though the requirement of normative justification overlays an objective element. Secondly, the requirement of reliance, or detrimental reliance to use the language of estoppel, suggests that a representation alone is not enough — the representation must have had an effect on the mind or behaviour of the person affected. However, later English cases illustrate that neither of these considerations have proven essential.

IV COUGHLAN: THE ACCEPTANCE OF UNFAIRNESS IN ITS OWN RIGHT

On one view, Preston did not represent a radical development in English law because it drew together the threads of many earlier English decisions which had invoked either estoppel or other arguments to conclude that fairness could, in some cases, require either procedural protection of a strict standard or something more.[23] Further such cases arose after Preston and some even relied on the connection established in Preston between unfairness and abuse of power, but none elaborated on that connection in any significant way.[24] The point was decisively revisited by the Court of Appeal of England and Wales in Coughlan,[25] where it was held that there can be situations in which expectations generated by promises or representations made by public authorities must be fulfilled. The substantive legitimate expectation doctrine was recognised in the form foreshadowed in Preston — substantive unfairness as a form of abuse of power — which marked a significant departure from the procedural legitimate expectation doctrine.

The circumstances of the Coughlan case were as follows. In 1971, Ms Coughlan was badly injured in a car accident. She was hospitalised in New Court Hospital from 1971 to 1993, when she and other residents of the hospital were persuaded by the health authority to move to Mardon House. Ms Coughlan was told that Mardon House would be her ‘home for life’,[26] but in 1998 the health authority decided to close Mardon House and relocate its residents. The health authority had regard to the undertaking it had given to Ms Coughlan and the others but concluded that better services could be provided to all concerned in other institutions. Ms Coughlan sought judicial review of this decision.

The Court of Appeal held that, having regard to the undertaking given in 1993 and Ms Coughlan’s reliance upon it, the decision to close Mardon House was unfair and thus an abuse of power. The Court of Appeal reached this conclusion by use of a threefold approach to the promises, representations and legitimate expectations that could arise from government action. This taxonomy was not used to describe or explain the character of each expectation, but rather to distinguish the different questions that each sort of expectation might pose for the court. The first category was expectations for which the government would only be required to ‘bear in mind its previous policy or other representation, giving it the weight it thinks right, but no more, before deciding whether to change course.’[27] The court would apply the Wednesbury standard of unreasonableness to these expectations,[28] and would only overturn a decision if satisfied that it was entirely irrational or unreasonable. The second category of expectations was those in which a government’s ‘promise or practice induces a legitimate expectation of, for example, being consulted before a particular decision is taken.’[29] In these instances, the court would require consultation with a person affected in accordance with the expectation, after which the expectation could be disregarded if there were appropriate reasons to do so and if that decision was within power.

These categories are not controversial. In the first category, the court is required to apply the various grounds of review that could be encompassed under the rubric of rationality. In the second category, the same principles would apply but with the added requirement to observe the requirements of procedural fairness as determined by the circumstances of the case at hand.[30]

The standard of review would usually be stricter in the second category because the requirements of procedural fairness would inevitably dictate closer attention to the circumstances and expectations of the person affected. But both categories provide little more than a convenient label for the traditional procedural legitimate expectation, and point to the different ways in which a court can examine the process of decision-making. Both categories also presume that the court will apply conventional grounds of review in any application for judicial review.

But the third category of expectation identified by the Court of Appeal was quite different. According to the Court, the operation of this third form of expectation can be described thus:

Where the court considers that a lawful promise or practice has induced a legitimate expectation of a benefit which is substantive, not simply procedural … the court will in a proper case decide whether to frustrate the expectation is so unfair that to take a new and different course will amount to an abuse of power. Here, once the legitimacy of the expectation is established, the court will have the task of weighing the requirements of fairness against any overriding interest relied upon for the change of policy.[31]

The Court of Appeal left no doubt that an expectation falling within this last category could be recognised in judicial review.[32] More particularly, when a public official had created an expectation of a substantive benefit and then acted contrary to that expectation, the court could find that conduct to be an abuse of power and, therefore, unlawful. In a superficial sense, this reasoning breaks no new ground in judicial review because it suggests that the court simply determines the validity of an administrative decision by reference to a ground of judicial review (abuse of power). The significance lay in the way that this ground was applied. The Court of Appeal accepted that the lawfulness of any attempt to renege on a promise, or change the policy upon which an expectation was based, would depend on whether the court was satisfied that there was an ‘overriding’ interest or reason to do so.[33] The Court made clear that this balancing of individual and wider public interests, which would determine whether the public could override the personal, would take account of the fairness of any outcome.[34] According to this view, attention is directed to an issue previously beyond the scope of judicial review: the fairness or merits of the ultimate decision.

A key criticism of this approach is that the Court of Appeal provided no guidance on how or when an exercise of power may become an ‘abuse’. The Court of Appeal simply asserted that it was the role of the Court to determine whether conduct amounting to an abuse of power existed and ‘for the court to say whether the consequent frustration of the individual’s expectation is so unfair as to be a misuse of … power.’[35] Within this conception of the Court’s role in detecting an abuse of power, the only clear touchstone appears to be that the decision is one that the Court does not think should stand. On this view, review on the ground of substantive unfairness amounting to an abuse of power contains no discernible legal principle.

The reasoning of the Court of Appeal might also be criticised for usurping or infringing upon the role of the executive by drawing a court too close to the merits of administrative decision-making. The Court of Appeal was clearly mindful of this issue when it accepted that governments could, in some circumstances, change or resile from statements or policies, though it maintained that any such action would be subject to review for abuse of power. According to the Court of Appeal, the freedom granted to the executive within the broad limits of abuse of power ‘recognises the primacy of the public authority both in administration and in policy development but it insists, where these functions come into tension, upon the adjudicative role of the court to ensure fairness to the individual.’[36]

The balance that this passage appears to strike is arguably an illusion. The freedom that the Court of Appeal seemed willing to grant to executive action was significantly undercut by its emphatic assertion that it was for a court, not the executive government, to determine whether conduct by the executive had given rise to an abuse of power and whether there was a sufficient countervailing public interest to allow the decision to stand. The assumption by the Court of Appeal of the role of balancing or assessing questions of public interest in administrative decision-making is apt to lead the judiciary deep into the territory of the executive arm of government.

Mark Aronson, Bruce Dyer and Matthew Groves identify a common thread between the apparent absence of principle in Coughlan and the potential of the case to draw courts towards merits review. They conclude that Coughlan ‘maximised judicial discretion at the cost of legal certainty.’[37]

They suggest that the problems arising from Coughlan and its progeny

are partly semantic, but largely much more profound. … The vast bulk of judicial review applicants want substantive outcomes, not procedural outcomes, and the courts have traditionally refused them this. That is the province of merits review.[38]

But the theoretical divide between judicial and merits review should not obscure the practical effect judicial review may have. Judicial review clearly has the potential to affect the ultimate or substantive outcome of administrative decision-making.[39] The balancing exercise adopted in Coughlan was radical because it drew the Court directly towards the final stage of decision-making.

A final point worth noting about Coughlan was the obvious tension that arose when the Court of Appeal attempted to simultaneously recognise the controversial nature of its reasoning while also leaving the way open for the expansion and refinement of the substantive legitimate expectation doctrine. The Court of Appeal sought to minimise the possible controversy of its reasoning by suggesting that the recognition of substantive legitimate expectations was not a large doctrinal step because the wider concept of abuse of power in which it was based had become well-settled with cases such as Preston.[40] At the same time, however, the Court of Appeal acknowledged that the evolution of the legitimate expectation doctrine, whether substantive or procedural, might help to clarify the very concept of abuse of power from which it was drawn. The Court of Appeal explained that the

[l]egitimate expectation may play different parts in different aspects of public law. The limits to its role have yet to be finally determined by the courts. Its application is still being developed on a case by case basis. Even where it reflects procedural expectations, for example, concerning consultation, it may be affected by an overriding public interest. It may operate as an aspect of good administration, qualifying the intrinsic rationality of policy choices. And without injury to the Wednesbury doctrine it may furnish a proper basis for the application of the now established concept of abuse of power.[41]

This passage invites several comments. First, the Court envisaged that the substantive legitimate expectation doctrine could expand but hesitated to predict the possible direction of that growth. It could even be argued that the suggestion by the Court that the substantive legitimate expectation doctrine might expand on a case-by-case basis anticipated that at least some of that expansion would occur on a pragmatic rather than principled basis. A second and logically related question is exactly which doctrine might evolve — the substantive legitimate expectation doctrine or the abuse of power doctrine? The passage quoted suggests that the former might provide coherence to the latter. But how can the substantive legitimate expectation doctrine be drawn out of the abuse of power doctrine and then advanced as a basis for the very doctrine from which it was drawn? This reasoning is apt to weaken the coherence of both doctrines because it suggests that the uncertain basis of each doctrine can be overcome by reference to the other. That approach arguably conceals more than it reveals about the ultimate foundation of both the doctrine of substantive legitimate expectation and the doctrine of abuse of power. A further issue arises from the Court’s mention of good administration and the possibility that this might limit the policy choices available to administrative officials. This reference seemed to hint at the role that European law might play in the development of English judicial review. European public law has long accepted that a range of principles of good administration may have legal consequences. Many of these principles, such as the right to confidentiality in dealings with government or consistency in administrative decision-making, embody normative values in relation to the quality of government and its actions.[42] The legal dimension granted to these principles enables European law to take account of matters that have traditionally been beyond the scope of judicial review in England because they concern the merits of the decision-making process or the decision itself. However, some subsequent English cases revealed that English courts were prepared to introduce considerations of good administration as a purely domestic principle.

V THE POST-COUGHLAN ADJUSTMENT AND ENTRENCHMENT OF THE DOCTRINE IN ENGLAND

The Court of Appeal recently observed that the substantive legitimate expectation doctrine had been ‘developed and refined’[43] since Coughlan, but that that case continued to provide the ‘benchmark’ explanation of the concept.[44] That observation is correct in the sense that, while the doctrine has been invoked in many subsequent cases, few have expanded significantly upon the reasoning used in Coughlan. Those cases that have examined the reasoning of Coughlan have essentially refined rather than questioned the doctrinal basis of the substantive legitimate expectation doctrine. Some of those refinements have sought to replace Coughlan’s original foundation of fairness (which logically flows from its emphasis on the avoidance of unfairness) with wider norms of governance. These tentative steps are novel because the values of judicial review are traditionally assumed rather than explained.[45]

The widespread acceptance of the substantive legitimate expectation doctrine has also precipitated developments in related areas of judicial review, which signal what might happen if Coughlan is adopted in Australia. Parts V(A)–(E) of this article examine the most important refinements of Coughlan and the apparent consequences of Coughlan for other principles of judicial review.

A The Refinement of Coughlan

The first important refinement of Coughlan came just a month later in R v Secretary of State for Education and Employment; Ex parte Begbie (‘Begbie’).[46] Mrs Begbie sought judicial review of decisions affecting the assistance provided for her disabled child’s education. Her complaint was based on promises allegedly made by an opposition party which were altered when the party assumed government. The application could have been dismissed on the simple basis that any expectation created by an opposition party was not one induced by a ‘public’ agency and therefore beyond the boundaries of any possible legitimate expectation,[47] or that the complex legislative changes made upon the change of government precluded Mrs Begbie’s claim.[48] However, Laws LJ made his first attempt to provide a more coherent foundation for the substantive legitimate expectation doctrine.[49] His Lordship suggested that ‘[a]buse of power has become, or is fast becoming, the root concept which governs and conditions our general principles of public law.’[50] According to this view, the controversial point of Coughlan was not whether the substantive legitimate expectation doctrine ought to be accepted, but how the concept should be articulated within the wider rubric of abuse of power. Laws LJ explained:

The difficulty, and at once therefore the challenge, in translating this root concept or first principle into hard clear law is to be found in this question, to which the court addressed itself in the Coughlan case: where a breach of a legitimate expectation is established, how may the breach be justified to this court? In the first of the three categories given in Ex parte Coughlan, the test is limited to the Wednesbury principle. But in the third (where there is a legitimate expectation of a substantive benefit) the court must decide ‘whether to frustrate the expectation is so unfair that to take a new and different course will amount to an abuse of power.’ … However the first category may also involve deprivation of a substantive benefit. What marks the true difference between the two?[51]

Laws LJ also queried the adoption in Coughlan of different approaches to the review of each category. Coughlan held that expectations in the first category were amenable to review for irrationality/unreasonableness, while the substantive legitimate expectation category was amenable to review by way of a consideration of fairness in a particular case.[52] Laws LJ reasoned that these principles of review, like the expectations to which they were applied, possessed overlapping qualities. He explained:

Fairness and reasonableness (and their contraries) are objective concepts; otherwise there would be no public law, or if there were it would be palm tree justice. But each is a spectrum, not a single point, and they shade into one another.[53]

At this point, it should be noted that Coughlan’s first category covered expectations involving policies or promises of wide or general application. Change to policies or promises of this type usually occurs at the macro-political level by reference to complex political, social and economic considerations.[54] The cases cited in Coughlan as examples of its third category of legitimate expectation, the substantive variety, involved specific promises to one or only a few people.[55] The focus of a decision here is inevitably at the micro level. Laws LJ doubted that these distinctions could be ‘hermetically sealed’.[56] His Lordship was clearly correct. Policies and procedures of general application can, when applied to individual cases, cause hardship that appears no different from that of the substantive expectation category. Similarly, cases in which the disappointment of a substantive legitimate expectation strongly affects only one person can raise issues relevant to the macro-political level. Laws LJ suggested that the overlapping qualities of these decisions warranted review not by different grounds but by a differing intensity of review. His Lordship explained:

The more the decision challenged lies in what may inelegantly be called the macro-political field, the less intrusive will be the court’s supervision. More than this: in that field, true abuse of power is less likely to be found, since within it changes of policy, fuelled by broad conceptions of the public interest, may more readily be accepted as taking precedence over the interests of groups which enjoyed expectations generated by an earlier policy.[57]

This approach brings into much sharper focus the role of the court in any decision to recognise a substantive legitimate expectation. It does not provide an explanation of the meaning of fairness (or reasonableness) but rather an indication of when and why the court might intervene.

In R (Bibi) v Newham London Borough Council (‘Bibi’), a differently constituted Court of Appeal subsequently conceded that the obvious tension in the competing values of preserving the freedom of decision-makers, on the one hand, and ensuring fairness to the holders of legitimate expectations on the other, could leave courts with an invidious choice of ‘which good we attain and which we forego [sic].’[58] The Court reasoned:

There are administrative and democratic gains in preserving for the authority the possibility in the future of coming to different conclusions as to the allocation of resources from those to which it is currently wedded. On the other hand there is value in holding authorities to promises which they have made, thus upholding responsible public administration and allowing people to plan their lives sensibly. The task for the law in this area is to establish who makes the choice of priorities and what principles are to be followed.[59]

This approach retains the balancing exercise of Coughlan, by which the competing demands of individual fairness and the needs or justifications offered by the decision-maker can be weighed, but acknowledges more openly that the crucial question of which arm of government should perform that balancing exercise is as important as the principles by which the balancing exercise is performed. It is important to note that in England this difficulty is not confined to judicial review. Some commentators have argued that the extent to which the Human Rights Act 1998 (UK) c 42 will affect the willingness of courts to intrude upon executive decision-making will often depend on the ‘relative institutional competence’ of the courts and the executive government to determine the issue at hand.[60] Although the role of courts in human rights applications is clearly different from that in judicial review cases,[61] the use in each area of a sliding scale of review, coupled with the balancing methodology of proportionality, could hasten the move to conceptual unity in the principles governing judicial review and human rights law.

Laws LJ argued for a further shift of focus in R (Nadarajah) v Secretary of State for the Home Department (‘Nadarajah’).[62] There his Lordship expressed dissatisfaction with the generalised (and unsuccessful) attempt by the applicants to found a legitimate expectation on the failure to honour a promise and reliance upon factors similar to those of previous cases. His Lordship’s quibble was not simply about the uncritical use of precedent, but also about the failure of many applicants to identify the reasoning that might underpin their claim to a substantive legitimate expectation. His Lordship complained that

‘[p]rinciple is not … supplied by the call to arms of abuse of power. Abuse of power … is a useful name, for it catches the moral impetus of the rule of law. … But it goes no distance to tell you, case by case, what is lawful and what is not.’[63]

Laws LJ returned to the general principle that had evolved in the legitimate expectation cases, namely, that a public authority that made a promise or followed a practice representing how it would act was required to follow that promise or practice unless there was good reason to do otherwise.[64] His Lordship accepted that this legal obligation to honour promises was ‘grounded in fairness,’[65] but now suggested that its roots ultimately lay in a much deeper principle. Laws LJ explained that principle as follows:

I would … express it rather more broadly as a requirement of good administration, by which public bodies ought to deal straightforwardly and consistently with the public. In my judgment this is a legal standard which, although not found in terms in the European Convention on Human Rights, takes its place alongside such rights as fair trial, and no punishment without law. That being so there is every reason to articulate the limits of this requirement — to describe what may count as good reason to depart from it — as we have come to articulate the limits of other constitutional principles overtly found in the European Convention.[66]

His Lordship continued:

Accordingly a public body’s promise or practice as to future conduct may only be denied, and thus the standard I have expressed may only be departed from, in circumstances where to do so is the public body’s legal duty, or is otherwise, to use a now familiar vocabulary, a proportionate response (of which the court is the judge, or the last judge) having regard to a legitimate aim pursued by the public body in the public interest. The principle that good administration requires public authorities to be held to their promises would be undermined if the law did not insist that any failure or refusal to comply is objectively justified as a proportionate measure in the circumstances.[67]

In later Parts of this article, it will be explained that principles of good government as an underlying or unifying concept for judicial review of administrative action have been rejected in many Australian cases, but it is easy to see why Laws LJ raised it. The invocation and formulation of rules of ‘good administration’ or ‘good government’ might begin to answer the criticisms of the substantive legitimate expectation doctrine as formulated in Coughlan, namely, that the concept contains no explanation of when and why an exercise of power becomes an abuse of power. Ideas of good and bad governance are relatively easy to articulate, particularly by reference to the facts of most legitimate expectation cases. If those ideas are applied through a test of proportionality, they contain some gauge which can indicate when and why a court will intervene. At the same time, however, the invocation of notions of good administration raises new problems. An immediate one is whether those notions contain any real legal standards or simply provide a convenient cloak for judges to impose personal rather than legal principles. A similar criticism can be made of Laws LJ’s attempt to explain the requirement of good administration as a legal standard of a fundamental nature which silently underpins written constitutional documents. Any recourse to principles of this nature is also open to the criticism, made frequently in other areas where courts invoke ‘fundamental’ or ‘unwritten’ legal standards, that such principles are little more than a smokescreen for an erratic and subjective assortment of judicial ideas.[68]

B ‘Conspicuous Unfairness’ A Separate Head of Review or a Sign of Abuse of Power?

Several recent English cases have invoked the concept of ‘conspicuous unfairness’ as an explanation of why or how the disappointment of a legitimate expectation may become an abuse of power. Although this term might seem to echo Coughlan, it was first used several years earlier in a tax case, where it was suggested that any action contrary to a legitimate expectation might be an abuse of power if the decision-maker acted ‘with conspicuous unfairness.’[69] The House of Lords adopted this terminology in a post-Coughlan case and appeared to accept the principle that a public authority which had acted contrary to a representation could abuse its power if it acted with conspicuous unfairness.[70] However, their Lordships did not examine the idea in detail. That was done instead by the Court of Appeal of England and Wales in Secretary of State for the Home Department v The Queen (Rashid) (‘Rashid’).[71]

Rashid was an Iraqi Kurd whose claim for asylum was rejected because the decision-maker was unaware of the policy governing Kurds.[72] This and other failures led the Court to find that the United Kingdom Home Office had acted with ‘flagrant and prolonged incompetence’.[73] Pill LJ explained that this incompetence meant that Rashid’s claim was not a typical one of legitimate expectation but was instead one of

unfairness amounting to an abuse of power, of which legitimate expectation is only one application. The abuse is based on an expectation that a general policy for dealing with asylum applications will be applied and will be applied uniformly.[74]

The serious errors of administration which prevented the proper application of the policy had, according to Pill LJ, caused conspicuous unfairness to Rashid. His Lordship noted that the Home Office had not claimed any form of public interest to justify this unfairness and concluded that ‘the degree of unfairness was such as to amount to an abuse of power requiring the intervention of the court. The persistence of the conduct, and lack of explanation for it, contribute to that conclusion.’[75]

A significant obstacle remained. By the time the errors were uncovered and placed before the Court, the policy had changed. The type of visa that Rashid could have been granted in consequence of the policy of the time was no longer available. Could the Court hold the Home Office to a procedure that no longer existed? The Court of Appeal accepted that the change of policy precluded the grant of refugee status to Rashid but declared that the unfairness caused by the lost opportunity could be addressed by a declaration that Rashid was, by exercise of other powers available to the Home Secretary, entitled to a grant of indefinite leave to remain in England.[76] This remedy was not precisely the same as a grant of refugee status, but the Court concluded that it offered sufficiently similar benefits to be an appropriate remedy.

The decisive issue in Rashid appeared to be the degree of unfairness. If the unfairness was ‘extreme’, or capable of attracting similar descriptors, an abuse of power could be found. But this version of unfairness leaves the court with little more than impressionistic guidance. Collins J recognised as much when he subsequently explained Rashid in the following terms:

The court has to decide whether the unfairness is such that it goes beyond that which should attract no relief other than that afforded by a right of appeal. I recognise that it is not possible to define where the line should be drawn with any precision. Inevitably, the circumstances of an individual case will be the deciding factor. It is only if the court is persuaded that the unfairness is so bad that abuse of power is an appropriate label that it will find in a claimant’s favour.[77]

This reasoning provides no clear criteria by which a court might determine the point at which unfairness had become sufficiently ‘bad’ to warrant intervention. The lack of obvious principle in the notion of conspicuous unfairness has caused some disquiet. One commentator concluded that Rashid represented a substantial step beyond Coughlan by signalling that a court might intervene ‘simply where something has gone badly wrong, even if the court cannot quite put its finger on it.’[78] That interpretation of Rashid was rejected in Secretary of State for the Home Department v The Queen (S),[79] where the Court of Appeal attempted to place Rashid on a more principled footing. In that case, the Court of Appeal was again faced with serious administrative failings but rejected a submission that Rashid had recognised the existence of a broad judicial power to correct such problems. The Court instead focused on the relevance of past illegality to any subsequent decision of the Home Secretary. On this view, the Home Secretary could not simply disclaim the consequences of serious administrative errors merely because the policy that would have been applied but for those errors no longer existed. The Home Secretary was instead required to take account of the illegality that past errors had caused. Carnwath LJ explained:

The issue is not so much whether the unfairness is obvious or conspicuous, but whether it amounts to illegality which on reconsideration the Department has the power to correct. If it has such a power, and there are no countervailing considerations, it should do so. Following Rashid the court has power to order reconsideration on the proper basis.[80]

According to this view, past errors and the unfairness they had caused could not be consigned to history with the outdated policy. They remained relevant to any future decision of the Home Secretary because the circumstances of any new decision ‘might include the present need to remedy injustice caused by past illegality.’[81]

The approach in Secretary of State for the Home Department v The Queen (S) confirmed that Rashid does not, as was initially thought, invest courts with some sort of freestanding power to cure unfairness. Rather, it illustrated that when a decision-maker had without good reason failed to apply a policy or procedure that was lawful, current and relevant, the court could declare that the errors and consequential unfairness should be considered in the exercise of other powers available to provide an outcome similar to that which the applicant might originally have been entitled to. Rashid might also be taken to suggest that there may be a legitimate expectation that policies and procedures will, in the normal course, be applied in cases to which they apply. A longstanding failure to do so could be subsequently cured in the way that happened in Rashid, while the failure to apply a policy or procedure in a particular instance could be attacked on the ground that the failure either constituted a failure to take into account a relevant consideration (the policy or procedure) or amounted to substantive unfairness (the selective non-application of the policy leading to a disadvantageous decision).

C The Doctrinal Break between Estoppel and Public Law

Many early legitimate expectation cases drew openly from the law of estoppel, largely because the cases frequently raised issues of representation, reliance and fairness, all of which feature heavily in estoppel. However, in R v East Sussex County Council; Ex parte Reprotech (Pebsham) Ltd (‘Reprotech’)[82] the House of Lords signalled that any connection that had existed between the two areas was at an end. Lord Hoffmann, with whom the other Law Lords agreed, conceded that there was ‘an analogy between a private law estoppel and the public law concept of a legitimate expectation created by a public authority, the denial of which may amount to an abuse of power’.[83] However, his Lordship continued,

it is no more than an analogy because remedies against public authorities also have to take into account the interests of the general public which the authority exists to promote. Public law can also take into account the hierarchy of individual rights which exist under the Human Rights Act 1998, so that, for example, the individual’s right to a home is accorded a high degree of protection … while ordinary property rights are in general far more limited by considerations of public interest …[84]

Lord Hoffmann concluded that ‘public law has already absorbed whatever is useful from the moral values which underlie the private law concept of estoppel and the time has come for it to stand upon its own two feet.’[85]

Several points can be made about this decisive separation of public and private law. First, longstanding differences had existed between the legitimate expectation doctrine and estoppel. The clear weight of authority suggested that a legitimate expectation could exist even if the person who claimed its benefit could not prove reliance upon the representation from which the expectation arose or, if there was reliance, could not prove any consequential detriment.[86] If the legitimate expectation doctrine could be found in cases that lacked key elements of, or any close parallel with, the circumstances in which estoppel might arise, there seemed little reason to maintain the fiction of a continued parallel between the two areas. Secondly, estoppel is very much directed to a relatively narrow consideration of the issues raised between two parties within which it is often difficult to raise the wider issues of public interest that are present in many public law proceedings. Another reason to formally renounce any continued link between the two areas lies in the different remedial focus of each. Estoppel has always reserved the right to award damages as a remedy in cases where specific performance is not possible.[87] If public law had continued to veer towards estoppel, it would inevitably have had to confront the difficult question of whether public law ought to expand to encompass the remedies available in estoppel (such as the right to damages) or, if that possibility was rejected, how public law could draw upon doctrinal but not remedial principles of estoppel.[88]

The severance of any doctrinal connection between the two areas forecloses all such problems.

D The Separation of Powers

The UK has no formal constitutional instrument that mandates a separation of powers in the binding and overriding manner of Australia’s federal constitutional structure, but observance of the separation of powers is a longstanding part of English law.[89] The decision in Coughlan might be seen as having involved judicial review of an administrative decision, on its merits, and having considered the decision’s fairness in light of a legitimate expectation created by an undertaking given by an administrative agency concerning the future exercise of a discretionary power reposed in it. On this view, the reasoning in Coughlan took the Court beyond its traditional role and into the terrain of the executive government. In their commentary on Coughlan, Paul Craig and Søren Schønberg did not consider the case to be at odds with the ‘classic separation of powers doctrine’.[90] That doctrine, they wrote:

tells us that it is not for the courts to substitute their choice as to how discretion ought to have been exercised for that of the administrative authority. They should not intervene, reassess the matter afresh and decide, for example, that funds ought to be allocated in one way rather than another.[91]

It is true that in Coughlan the English Court of Appeal did not substitute its judgement in relation to the future of Mardon House for that of the health authority.[92] But the practical effect of Coughlan was that Mardon House could not be closed and its residents relocated unless the health authority could provide more persuasive reasons to do so. While such an outcome does not, strictly speaking, lead the courts to make administrative decisions, it clearly restricts the freedom of the executive government in administrative decision-making. It also expands the influence of the courts on the administrative process. A significant weakness in the analysis of Craig and Schønberg and many other academic supporters of substantive legitimate expectations is the failure to acknowledge that the doctrine clearly narrows the freedom of the executive government and, more importantly, the effect that this may have on the relationship between the judicial and executive arms of government.[93]

Craig seemed to anticipate this argument in his treatise on administrative law, where he explains that the nature and purpose of administrative law must be understood by reference to the type of democratic society that its participants desire. This functional explanation of administrative law accepts that all doctrine is inevitably fashioned by equal measures of law and politics. The courts are necessarily dynamic and will ‘decide what particular constraints to impose on administrative action, and more generally on the overall purpose of judicial review.’[94]

According to this view, it is entirely appropriate for the courts to both devise novel principles of review, such as the substantive legitimate expectation doctrine, and openly search for underlying norms for those principles. The judicial dynamism that Craig endorses would surprise most Australian observers, but it is consistent with the suggestion of many other English commentators that the enactment of the Human Rights Act 1998 (UK) c 42 has caused fundamental changes to the constitutional structure of England and the relationship between the courts and government.[95]

E Observations on Coughlan and Its Progeny

Although the English law on substantive legitimate expectations is clearly still evolving, several propositions can be distilled from the law to date. Perhaps the most important is that the precise doctrinal basis of the substantive legitimate expectation doctrine remains uncertain. The Court of Appeal of England and Wales acknowledged this problem in Bibi when it cautioned that:

The case law is replete with words such as ‘legitimate’ and ‘fair’, ‘abuse of power’ and ‘inconsistent with good administration’. When reading the judgments care needs to be taken to distinguish analytical tools from conclusions which encapsulate value judgments but do not give any indication of the route to those conclusions.[96]

The Court was right to draw attention to the difference between the possible analytic concepts by which the substantive legitimate expectation cases might be coherently decided, and the conclusions and value judgements which are frequently invoked in the cases. At the same time, however, the value-laden terms and conclusory phrases which continually arise in this area highlight an obvious connection between judicial values and the legal outcomes that those values lead to. Judicial perceptions of how and why governments ought to behave provide a significant part of the foundation of the substantive legitimate expectation doctrine as it is unfolding in English law. English courts clearly believe that governments should, as a general rule, be held to their word in administrative decision-making.[97] Although this rule is subject to various exceptions, it places importance on the ability of people to assume that public officials will act according to their previously stated intentions unless there is a strong reason to do otherwise. The best explanation for this rule seems to be that offered by Laws LJ in Nadarajah, namely, that it is a requirement or principle of good administration.[98] Such a requirement can certainly provide a more complete explanation for the substantive legitimate expectation doctrine than the concept of abuse of power, partly because it may explain why disappointment of an expectation is thought to be wrong, but also because it acknowledges openly the normative element that must surely underpin the legal recognition of a substantive legitimate expectation.

However, there is less clarity about the extent to which the substantive legitimate expectation doctrine might require or enable a court to step beyond its traditional role in judicial review and veer towards issues that are, according to a traditional conception of the separation of powers, allocated to other arms of government. Courts that uphold a substantive legitimate expectation do not implement that expectation or direct administrative officials to do so.[99] It could be argued, therefore, that the substantive legitimate expectation doctrine does not necessarily take a court beyond its traditional role in judicial review, which is to declare and apply the law. But that conclusion presumes a level of certainty in the law. One continued difficulty with the English cases is that many provide little, if any, detail as to when and why an exercise of power might become an abuse of power. This lack of clarity leaves many of the cases open to the criticism that they lack any certain principle and, therefore, cannot have any educative effect on decision-makers.[100]

When the uncertain foundation of the substantive legitimate expectation doctrine is coupled with the balancing exercise adopted in Coughlan, by which courts weigh the various reasons for and against honouring the subject matter of an expectation, it can be argued that the substantive legitimate expectation doctrine requires courts to apply principles that are legal in name only. The application of such uncertain principles may not be consistent with the judicial role in the traditional conception of the separation of powers. The same point can be made when the substantive legitimate expectation doctrine is conceived as one example of the wider articulation of constitutional norms in English public law. John McMillan has explained this wider trend in English public law, and the role that it envisages for the judiciary, in the following terms:

there are rights — variously thought of as fundamental rights, human rights, or common law assumptions — that inhere in the constitutional structure. It is therefore said to be part of the judicial role to identify, articulate and safeguard those values as constitutional or legal rules. Notions of ‘fairness’, ‘proportionality’ and ‘equality’ quickly emerge as legally enforceable conditions on the exercise of executive power.[101]

According to this view, the English judiciary now believes that it can and should articulate and enforce certain values in the form of constitutional norms. This development has attracted many critics.[102] The many arguments that can be made for and against this direction in English public law cannot be canvassed in any detail here, but there is one point of particular relevance to the substantive legitimate expectation doctrine.[103]

One of the many issues that judges who seek to construct fundamental or underlying principles fail to explain is how these ultimately normative principles may be regarded as legal in character. Jeremy Waldron has asked rhetorically:

how are we supposed to tell whether a given norm practiced and prevalent among the powerful in a society governed by law is actually one of its laws, part of its legal system, as opposed to a moral principle that powerful people happen to accept?[104]

Whatever position one takes on the direction of the English cases, it is clear that the increasing willingness of English judges to expound values and norms as principles of administrative law presents a significant obstacle to the Australian adoption of the substantive legitimate expectation doctrine because the doctrine would take Australian courts beyond what most observers regard as the accepted limits of federal judicial power. Parts VI(D)–(G) of this article explain those constitutional objections to an Australian adoption of the doctrine, though it is useful to first explain other Australian obstacles to it.

VI THE AUSTRALIAN RECEPTION OF THE SUBSTANTIVE LEGITIMATE EXPECTATION DOCTRINE

Any discussion of the Australian reception of Coughlan must distinguish between the form and substance of the grounds of judicial review. ‘Abuse of power’ is formally recognised as a ground of judicial review in some statutory schemes of judicial review in Australia,[105] but there are clear reasons why this statutory ground does not represent an adoption of either the substantive legitimate expectation doctrine or the wider notion of abuse of power articulated in English law. The ground was first enacted in Australia well before Coughlan, and Part VI(B) of this article make clear that, at that time, the substantive legitimate expectation doctrine was not accepted as part of Australian law.[106] The small number of cases in which this ground has been pleaded does not encourage the view that this statutory ground is now taken to encompass English developments, or that the ground provides a vehicle for significant innovations in the law of judicial review.[107] The same reasoning applies to the related statutory ground of review which proscribes an exercise of power that is ‘otherwise contrary to law.’[108] This ground was included in the Administrative Decisions (Judicial Review) Act 1977 (Cth) (‘ADJR Act’) at the suggestion of Sir William Wade and has been reproduced in the subsequent state and territory schemes modelled on that Act. Wade thought that this ground could assist ‘future developments’ in the law of judicial review,[109] but his prediction has not been vindicated. The statutory mechanisms of judicial review that were enacted in several Australian jurisdictions have successfully overcome many of the technical problems of judicial review at common law, such as the absence of any right to reasons for decisions, but they have proved less successful in stimulating any evolution of the grounds of review.[110]

The statutory recognition of apparently expansive grounds of judicial review such as ‘abuse of power’ or ‘otherwise contrary to law’ could in theory provide convenient vehicles for invoking the substantive legitimate expectation doctrine. However, it is likely that any acceptance that the doctrine falls within the scope of these grounds would only occur if it gained acceptance at common law. The cases to date suggest that any such acceptance is unlikely. While Coughlan has been cited in several Australian decisions, the substantive legitimate expectation doctrine has not been accepted as part of Australian law.[111] That position is due largely to the strong doubts expressed about the doctrine by several members of the High Court in Re Minister for Immigration and Multicultural and Indigenous Affairs; Ex parte Lam (‘Lam’).[112] However, closer analysis reveals that doubts about the doctrine have a longer heritage. Parts VI(A)–(C) of this article explain those doubts and Part VI(D) discusses the constitutional objections raised in Lam.

A The Procedural Conception of the Legitimate Expectation Doctrine

As the procedural legitimate expectation doctrine gained acceptance, there also arose a question of whether a doctrine with respect to expectations of a substantive nature might be accepted. In these instances, even the strongest proponents of the legitimate expectation doctrine made it clear that that doctrine did not compel the grant of the ‘substance of the expectation’ but instead ‘the observance of procedural fairness before the substance of the expectation is denied’.[113] This conception of the legitimate expectation doctrine did not simply confine it to procedural issues, but it also allowed the denial of an expectation so long as procedural entitlements were observed. The case of Attorney-General (NSW) v Quin (‘Quin’)[114] illustrated how flimsy procedural protection might be. Quin involved the reorganisation of a state court by legislation to replace existing courts of petty sessions with a local court. The legislation automatically deprived magistrates of the old court of their office, but the government signalled that they would, on application, be appointed to the new court. However, there was a screening process which some former magistrates did not survive. They challenged the decisions not to reappoint them on the ground that they were denied procedural fairness, namely, the right to answer contentions that they were not fit to be reappointed. The New South Wales Court of Appeal ordered that the applications be reheard.[115] Before that was done, the government announced a new policy which required that future appointments be considered on a competitive basis, without preference to magistrates from the old court. Mr Quin, a former magistrate whose application for appointment was pending, argued that the new policy should not apply to him because he had a legitimate expectation that his application would be determined under the old policy.

The High Court did not question the power of the executive to alter its policy regarding the exercise of its statutory power to appoint judicial officers to the local court. Nor did the Court question the legitimacy of either the old or the new policy. The critical question the Court had to decide was whether, in its consideration of Mr Quin’s application, the executive was bound to apply the old policy on account of the legitimate expectation that the old policy may have created. A majority of the Court (Mason CJ, Brennan and Dawson JJ) concluded that the executive was not so bound. They invoked legal principles which prevent public authorities from fettering the exercise of discretionary powers in the future,[116] and reasoned that the government was not bound to adhere to its initial policy. Accordingly, there could be no legitimate expectation that this policy would be applied to Mr Quin.[117]

Mason CJ identified the substantive quality of Mr Quin’s claim, even if it was framed in purely procedural terms: in Mr Quin’s claim that natural justice required that his application be considered according to the first policy, there was an assumption that he would be appointed. The Chief Justice acknowledged that ‘a legitimate expectation may take the form of an expectation of a substantive right, privilege or benefit or of a procedural right, advantage or opportunity’, but cautioned that it is ‘helpful to avoid confusion between the content of the expectation and the resulting right to procedural fairness.’[118] This reasoning is consistent with the dissenting judgments of Deane and Toohey JJ, who each held that Mr Quin was entitled to have his application determined according to the old policy, but accepted that the government could decline to appoint him if it thought he was unfit or unsuitable.[119]

One common theme in the judgments in Quin was that all members of the Court declined to either direct or assume the exercise of the relevant discretionary power. A majority of the Court was also prepared to grant considerable latitude to the executive government to promulgate and revise policies to guide the exercise of discretionary powers, even if that might cause great unfairness to people affected by the policy. Although the dissenting judges held that the requirements of procedural fairness might limit the freedom of the executive government to make decisions affecting people by way of the change of policy, their Honours also accepted that the ultimate decision lay with the executive and that the executive, rather than the courts, should weigh the competing factors of each case. This reasoning is consistent with other cases in which the courts have accepted that administrative decision-making could involve unfair circumstances or lead to outcomes that might be perceived as unfair, but that such problems cannot be resolved by the courts through the requirements of procedural fairness.[120] According to this view, judicial review is directed to issues of procedure rather than substance.

However, that distinction was clearly blurred in the later case of Minister for Immigration and Ethnic Affairs v Teoh (‘Teoh’).[121] The important aspect of Teoh for present purposes was that the legitimate expectation expounded by the majority of the High Court was, in theory at least, entirely procedural in character. A majority of the High Court held that ratification by the executive government of an international treaty constituted a positive statement to the Australian people that the federal executive would act in conformity with the treaty. That ‘positive statement’ provided the foundation for a legitimate expectation that an administrative decision-maker would either take into account the treaty in the exercise of discretionary powers or, if this was not to be the case, provide a person affected by the exercise of these powers with notice of the possibility that the treaty would not be taken into account and the opportunity to argue against that intended course. Mason CJ and Deane J stressed the procedural nature of any expectation that a decision-maker would act in accordance with the treaty. Their Honours explained:

The existence of a legitimate expectation that a decision-maker will act in a particular way does not necessarily compel him or her to act in that way. That is the difference between a legitimate expectation and a binding rule of law. To regard a legitimate expectation as requiring the decision-maker to act in a particular way is tantamount to treating it as a rule of law.[122]

This reasoning is similar to that adopted by the majority of the High Court in Quin, in the sense that it clearly anticipates that the executive may disappoint a legitimate expectation so long as a person who may be affected by this is given notice and an opportunity to argue against that course. McHugh J dissented on the issue of whether ratification of a treaty could itself provide the foundation for a legitimate expectation.[123] McHugh J also adhered to his Honour’s longstanding objection to the possibility that the legitimate expectation doctrine could provide substantive protection of any kind.[124] McHugh J did not explain in detail why his Honour believed that the legitimate expectation doctrine could not extend to provide substantive protection, but this position was a natural extension of his Honour’s strong criticism of the reasoning of the majority. McHugh J essentially doubted whether the legitimate expectation doctrine could or should be invoked to provide procedural protection as frequently as was suggested by the majority. If the strength of the concept within its existing territory was uncertain, as his Honour believed, it could hardly extend to new territory such as substantive protection.

The concerns of McHugh J gained traction when the High Court revisited the legitimate expectation doctrine in Lam,[125] where several members of the High Court expressed doubts about the doctrine.[126] McHugh and Gummow JJ reasoned that the legitimate expectation doctrine had provided a useful procedural device to assist the evolution of the scope of the duty to observe the requirements of procedural fairness, but the modern growth of that duty meant that the concept was now of ‘limited utility’.[127] Hayne and Callinan JJ also questioned the continued utility of the procedural version of the legitimate expectation doctrine, but suggested that it could remain useful if it was refined and applied cautiously in future cases.[128] These various doubts about the legitimate expectation doctrine make it clear that the concept has almost certainly entered its twilight and, therefore, will struggle to retain its traditional procedural role in Australian administrative law, let alone expand in the way that would be required if it were to provide a more substantive form of protection.

In Part VI(G), it will be argued that the constitutional jurisprudence of the High Court of Australia is dominated by a doctrine that may conveniently be described as ‘Dixonian legalism’, or ‘formalism’ as the approach is often currently described.[129] The formalist or legalist approach is one that emphasises rules with a relatively narrow focus rather than principles of general application. It is also characterised by an adherence to traditional principles of statutory interpretation rather than the development of new ones. The formalist approach places relatively little emphasis on the role of international instruments in the development of domestic legal principles.[130] The relative merit of formalism is the subject of strong disagreement which is beyond the scope of this article,[131] but it would be fair to suggest that the reasoning in Lam represents a return to the formalist approach as opposed to the more dynamic approach adopted in Teoh. The important point is that Lam is illustrative of, and consistent with, an interpretive approach that holds sway in the High Court more generally.[132]

B The Rejection of Estoppel in Public Law

Although English courts have recently severed any doctrinal connection between estoppel and public law, no such connection was ever accepted in Australian law. Australian courts have long held that principles of estoppel cannot and should not apply to government agencies in relation to the exercise of powers which are peculiarly governmental.[133]

This principle was established in Minister for Immigration and Ethnic Affairs v Kurtovic (‘Kurtovic’).[134] This case resembled that of Coughlan in that the alleged unfairness was administrative action claimed to be in breach of a prior representation to a particular person. Kurtovic was a non-citizen serving a lengthy sentence of imprisonment. The Minister for Immigration, Local Government and Ethnic Affairs ordered that Kurtovic be deported,[135] but Kurtovic successfully appealed this decision to the Administrative Appeals Tribunal (‘AAT’). The Minister accepted this decision but wrote to Kurtovic warning that any further conviction which rendered Kurtovic liable to deportation would lead to the reconsideration of his deportation. A few years later, the state parole authorities wrote to the Minister expressing concern that Kurtovic might reoffend upon release. The Minister made a new decision, based largely on the original convictions, to deport Kurtovic.

The Full Court of the Federal Court of Australia held that the power conferred on the Minister was exercisable from time to time, whether or not there had been any change in circumstances.[136] The Full Court also emphatically rejected an argument that the Minister was estopped from making the second order by reason of the letter which had advised Kurtovic of the Minister’s acceptance of the AAT’s decision. That letter, the Court considered, did not contain a clear and unambiguous promise that, in the absence of new circumstances, no further order would be made for the deportation of Kurtovic based on the 1983 convictions. However, even if the letter could have been construed as containing such a promise, the promise could not be regarded as binding inasmuch as it would have been an impermissible fetter on a statutory discretion conferred for the public benefit.[137] Gummow J, with whom Neaves and Ryan JJ agreed on this issue, reviewed the law governing estoppel and public officials and concluded that the distinct character of public powers, particularly their application of the rule against fettering those powers, provided strong reasons to reject any application of estoppel to the exercise of public powers. His Honour reasoned that:

in a case of discretion, there is a duty under the statute to exercise a free and unhindered discretion and an estoppel cannot be raised (any more than a contract might be relied upon) to prevent or hinder the exercise of the discretion; the point is that the legislature intends the discretion to be exercised on the basis of a proper understanding of what is required by the statute, and that the repository of the discretion is not to be held to a decision which mistakes or forecloses that understanding.[138]

The reasoning of Gummow J has been cited with approval a sufficient number of times that it is now well-settled that estoppel does not form part of Australian administrative law.[139]

It is important to note that Gummow J also flatly rejected an attempt to marshal the same facts in an argument that the decision was substantively unfair.[140] This submission essentially argued that the Minister’s position had changed, which was unfair in a more substantive sense which, in turn, was ‘contrary to law’ within the meaning of s 5(1)(j) of the ADJR Act. It was under that head that the Full Court considered whether administrative decisions are judicially reviewable when it is claimed that they are substantively unfair. The unfairness was argued to be the inconsistency between the former and current treatment of Kurtovic, namely the apparent change between the decision in 1985 to revoke the deportation order and the second deportation order made in 1988. Reference was made to observations of Lord Denning MR in Laker Airways Ltd v Department of Trade, where his Lordship suggested that

the Crown cannot be estopped from exercising its powers, whether given in a statute or by common law, when it is doing so in the proper exercise of its duty to act for the public good, even though this may work some injustice or unfairness to a private individual …[141]

But Lord Denning MR added that the Crown could

be estopped when it is not properly exercising its powers, but is misusing them; and it does misuse them if it exercises them in circumstances which work injustice or unfairness to the individual without any countervailing benefit for the public …[142]

Gummow J, with whom the other members of the Full Court agreed on this point, identified ‘two fatal objections’[143] to Lord Denning MR’s suggested approach by which courts determined the substantive fairness of administrative decisions ‘by some process of “judicial balancing” between public and private interests.’[144] Those objections were:

First, the question of where the balance lies between competing public and private interests in the exercise of a statutory discretion goes to the merits of the case, and is thus one for the decision-maker, not the courts, to resolve. Secondly, a conclusion that a representation or decision is ultra vires ordinarily will preclude its effectiveness. An ultra vires representation is not a mere factor in favour of which the scales of judicial balancing might be allowed to swing, but peremptorily forecloses such deliberation.[145]

There are two reasons why this rejection of estoppel might also foreclose use of the substantive legitimate expectation doctrine. First, it suggests that any form of ‘balancing’ or weighing of competing interests in judicial review, whether in the form of the test adopted in Coughlan or in the weighing of different factors as might be required in a plea of estoppel, is firmly identified as merits review. Accordingly, the balancing exercise lies beyond the scope of judicial review.[146] Secondly, the Court’s reference to ultra vires representations makes clear that the effect of any representation will be judged according to the ultra vires doctrine of lawfulness rather than any wider principles of fairness. That approach clearly counts against the adoption of the expanded concept of fairness that was accepted in Coughlan.

Mason CJ appeared to take a contrary position in Quin when his Honour suggested that there existed a possible exception to the principle that estoppel was not available against public authorities in respect of the exercise of their discretionary powers. Mason CJ explained that, despite his Honour’s acceptance of this general principle, his Honour did

not deny the availability of estoppel against the Executive, arising from conduct amounting to a representation, when holding the Executive to its representation does not significantly hinder the exercise of the relevant discretion in the public interest. And, as the public interest necessarily comprehends an element of justice to the individual, one cannot exclude the possibility that the courts might in some situations grant relief on the basis that a refusal to hold the Executive to a representation by means of estoppel will occasion greater harm to the public interest by causing grave injustice to the individual who acted on the representation than any detriment to that interest that will arise from holding the Executive to its representation and thus narrowing the exercise of the discretion …[147]

In my view, there are two obvious reasons why this narrow application of estoppel to administrative law is unlikely to find favour. First, the reasoning of Mason CJ would require a court to balance the wider public interest against any possible injustice caused to an individual. That process is strikingly similar to the balancing exercise adopted in Coughlan and would almost certainly encounter all of the objections which could be made against that case, particularly the constitutional ones explained below. Secondly, while it is easy to accept the suggestion of Mason CJ that a truly exceptional case of unfairness can arise, it is difficult to accept that estoppel could offer a solution that the existing grounds of judicial review could not. If, for example, a decision-maker intended to resile from a promise or undertaking but failed to inform the person affected and invite comment on this proposed course of action, the decision could be set aside for a denial of procedural fairness. If such an opportunity was given but the decision-maker ignored the arguments provided by the person affected, or made a decision that was at odds with those arguments and could not be justified, the decision could be set aside on other existing grounds such as the failure to take account of a relevant consideration or unreasonableness. If the decision could be justified, or was one of a number of possible conclusions that appeared reasonably open on the evidence before the decision-maker, the ‘grave injustice’ that concerned Mason CJ would almost certainly not exist.

C Unfairness in the Form of Inconsistent Treatment

Although estoppel has been rejected in Australian administrative law, some cases have suggested that the different treatment of similarly placed people without good reason may be unlawful. This possibility raises issues not unlike estoppel by suggesting that the treatment of one member of a class of people in a particular way creates a basis for similarly placed people to believe that they will be treated in the same way. But cases that have considered this issue have made it clear that any possible principle against inconsistent treatment will not provide a bridge to the adoption of a concept of substantive unfairness. It is useful to distinguish the two types of ‘inconsistency’ that are traditionally alleged. The first is when the same person is treated differently at different points in time.[148] In these cases, the person concerned usually argues that a promise made or policy adopted at one point in time should be honoured at a later time. Claims of this nature will usually fail for the reasons given in Kurtovic and Quin, namely, that the official should not be estopped or fettered from reconsidering the representation or changing the policy. The second form of inconsistency involves the different treatment of different people whose circumstances are similar. In Quin, Dawson J accepted that the requirements of fairness could not dictate the adoption or change of a policy by an administrative official, but his Honour conceded that ‘[i]t may well be different when a particular decision involves, not a change in policy brought about by the normal processes of government decision making, but merely the selective application of an existing policy in an individual case.’[149]

The Full Court of the Federal Court reached a similar conclusion in Bellinz v Commissioner of Taxation when it accepted that fairness would require tax officials to exercise their discretionary powers ‘in a way that does not discriminate [amongst] taxpayers.’[150] The Full Court also suggested that the same principle might support judicial review ‘in matters of administration or procedure where a decision-maker acts unfairly by discriminating [amongst] different categories of persons.’[151] This reasoning is consistent with several other decisions in which the courts have hinted that unequal or inconsistent treatment might, in some circumstances, provide some basis for the grant of relief.[152]

However, there are also cases to the contrary. The Federal Court appears to favour a different position in relation to decisions of the Refugee Review Tribunal (‘RRT’). There have been many cases in which applicants for refugee status have complained that different members of the RRT have reached different conclusions about the same country information in cases that seem to have no material difference. The Federal Court has rejected the suggestion that there may be legal error in the differing conclusions reached in such cases.[153] It could be argued that these cases do not support a general proposition that there is no requirement or principle of equality in administrative decision-making but instead support the more limited principle that refugee decision-making is sufficiently complex and unique that the courts are reluctant to apply a principle of equality in that area.[154]

While the precise standing of a rule against unequal treatment might not be settled, it is clear that the acceptance in some cases that the inconsistent treatment of similarly placed people might be amenable to judicial review has not led to the adoption of limited forms of estoppel or substantive unfairness. The reason may be pragmatic. Any exercise of a discretionary power that applies differing standards or policies to similarly placed people without good reason could be set aside on the ground of: improper purpose (depending on the reason for inconsistent treatment); relevant consideration (depending on what issues led to inconsistent treatment, or what policies or standards were disregarded); considerations of natural justice (depending on whether the person affected was informed of the intended inconsistent treatment and perhaps also given a chance to argue against that course); or unreasonableness (depending on whether the ultimate decision was entirely at odds with the evidence before the decision-maker). The possibility that inconsistent treatment between similar people can be considered under existing grounds of review considerably lessens the scope for a principle of ‘conspicuous unfairness’ as expounded in the English case of Rashid[155]

or the possibility that cases of this nature might provide a foothold for the substantive legitimate expectation doctrine.

D Constitutional Objections to the Substantive Legitimate Expectation Doctrine

The possible effect of the Australian Constitution upon judicial review was relatively neglected until the late 1990s with the advent of challenges to provisions that sought to limit or exclude judicial review of migration decisions.[156] But the potential constitutional obstacles to the substantive legitimate expectation doctrine were signalled earlier by Brennan J in Quin.[157] Although Quin was not commenced under provisions of the Australian Constitution, the principles expounded by Brennan J were fashioned by close reference to the separation of powers doctrine embodied in the Constitution and the attendant limits that the doctrine places on judicial power. Brennan J proceeded from the principle of Marbury v Madison, where the Supreme Court of the United States claimed that it was ‘the province and duty’ of the judicial branch to declare the law.[158] His Honour reasoned that this principle both defined and confined judicial power in equal measure. More particularly, it provided a barrier to the courts from assuming power over the merits of administrative action. Brennan J explained:

The duty and jurisdiction of the court to review administrative action do not go beyond the declaration and enforcing of the law which determines the limits and governs the exercise of the repository’s power. If, in doing so, the court avoids administrative injustice or error, so be it; but the court has no jurisdiction simply to cure administrative injustice or error. The merits of administrative action, to the extent that they can be distinguished from legality, are for the repository of the relevant power and, subject to political control, for the repository alone.[159]

This conception of judicial power and the consequential limits it places upon the proper scope of judicial review led Brennan J to identify several issues relevant to the substantive legitimate expectation doctrine. An important one was that the scope of judicial review should be directed to the ‘protection of individual interests but in terms of the extent of power and the legality of its exercise.’[160] That approach does not lend itself easily to the rights-based focus of the substantive legitimate expectation doctrine. Brennan J also acknowledged that the judicial role envisaged by Marbury v Madison left the court to determine the law, but did not itself provide guidance on what the law might be in any particular case.[161] However, Brennan J reasoned that the courts should be mindful that ‘the judicature is but one of the three coordinate branches of government and that the authority of the judicature is not derived from a superior capacity to balance the interests of the community against the interests of an individual.’[162] This reasoning suggests that the judicial balancing role devised in Coughlan, by which various factors relevant to the circumstances of a particular person are measured against a more general public interest, is not one that Australian courts should embrace.

The central propositions offered by Brennan J were adopted by a majority of the High Court in City of Enfield v Development Assessment Commission (‘Enfield’).[163] In that case, the Court held that the American principle that grants considerable deference to administrators in the adjudication of jurisdictional facts was incompatible with the limited role that Australia’s constitutional arrangements impose upon the functions of the executive.[164] The Court reasoned that administrators could not determine authoritatively (or non-authoritatively, subject to great deference from the courts) legal questions such as jurisdictional facts; such issues were clearly the constitutional province of the courts. However, the High Court stressed that corresponding restrictions applied to the power of the courts to undertake judicial review of administrative action, and that the judicial function could not extend to issues that formed part of the merits of a decision.[165] This conception of the constitutional limits on the role of the courts did not bode well for the substantive legitimate expectation doctrine because that doctrine could easily be characterised as one that drew the courts closer to the merits of a decision than the Australian Constitution, or the High Court, might allow.

That point came to the fore in Lam,[166] when several members of the High Court doubted whether Australia’s constitutional arrangements could allow the adoption of the substantive legitimate expectation doctrine. The basis of the claimed legitimate expectation in Lam was relatively simple. Lam was a non-citizen who had been convicted of criminal offences. While Lam was in prison, he was notified by migration authorities that they were considering whether to revoke his visa on character grounds. Lam was invited to respond and did so with a lengthy letter explaining his circumstances, particularly the position of his two children from a previous relationship. The authorities again wrote to Lam asking for the contact details of the person caring for his children. Lam provided the requested information, but the authorities did not contact the carer.[167] The Minister reviewed a lengthy brief on Lam’s case and decided to cancel his visa.

Lam claimed a denial of procedural fairness, essentially arguing that the second letter from the authorities created a legitimate expectation that they would follow a certain procedure (that no decision would be made before the carer of his children was contacted).[168] Any possible legitimate expectation would clearly have been procedural in nature, so counsel for Lam did not directly rely upon Coughlan,[169] but four members of the High Court took the opportunity to express strong doubts about the constitutional viability of the substantive legitimate expectation doctrine.[170] The strength of those doubts suggests that there are significant constitutional obstacles to the substantive legitimate expectation doctrine in Australia.

Gleeson CJ addressed the substantive legitimate expectation doctrine at a general level, suggesting that the reasoning in Coughlan involved ‘large questions as to the relations between the executive and judicial branches of government.’[171] The Chief Justice concluded that the jurisdiction vested in the High Court by s 75(v) of the Australian Constitution ‘does not exist for the purpose of enabling the judicial branch of government to impose upon the executive branch its ideas of good administration.’[172] McHugh and Gummow JJ, with whom Callinan J agreed on this issue,[173] reached a similar conclusion, but their Honours conceded that the normative values offered in recent English cases on abuse of power bore some similarity to the ‘values concerned in general terms with abuse of power by the executive and legislative branches of government’[174] in Australian constitutional law. However, their Honours cautioned that ‘it would be going much further to give those values an immediate normative operation in applying the Constitution.’[175] This reasoning suggests that the current Australian conception of the separation of powers precludes judges from giving effect to the normative values that have been favoured in recent English cases, such as the notion of good administration or the concept of abuse of power. There is one obvious consequence of this formalist approach to constitutional interpretation, which is that the suggestion by judges that the reach of judicial power under the Constitution does not extend to enable judges to devise or impose norms on administrative action does not prevent the creation or enforcement of those values; it only precludes judicial involvement in the exercise of them. The reasoning of the majority in Lam implies that this outcome is one that is necessarily dictated by the separation of powers doctrine.

McHugh and Gummow JJ also emphasised the differences between the constitutional structures of Australia and England, which dictated that Australian developments in judicial review required particular attention to s 75(v) of the Constitution.[176] Their Honours explained that:

Considerations of the nature and scope of judicial review, whether by this Court under s 75 of the Constitution or otherwise, inevitably involves attention to the text and structure of the document in which s 75 appears. An aspect of the rule of law under the Constitution is that the role or function of Ch III courts does not extend to the performance of the legislative function of translating policy into statutory form or the executive function of administration.[177]

This reasoning foreshadows an important obstacle to the substantive legitimate expectation doctrine and suggests that their Honours see the balancing act required by Coughlan to be within the province of the executive and therefore beyond that of the courts. The approach of McHugh and Gummow JJ illustrates the suggestion of Sir Anthony Mason that the Australian Constitution was fashioned to operate ‘as a delineation of government powers rather than as a charter of citizen’s [sic] rights.’[178] It also precludes the related English proposition that governments should normally be able to be held to their word in administrative decision-making. If notions of good administration and the like can be removed from judicial consideration, an important potential justification for the substantive legitimate expectation doctrine is removed.

McHugh and Gummow JJ also drew attention to the central role that the distinction between jurisdictional and non-jurisdictional error plays in Australian law. Their Honours reasoned that the distinction informed the principles governing s 75(v) of the Constitution which, despite the many problems arising from the jurisdictional error doctrine, provided a touchstone to distinguish those administrative actions which were authorised by law from those which were not.[179] Jurisdictional error may be easily criticised for its imprecision,[180] a problem which is exacerbated because the doctrine embraces different or overlapping forms of error.[181] But the reliance of McHugh and Gummow JJ on jurisdictional error must be understood in light of the crucial role that the doctrine has played in other aspects of the operation of the High Court’s jurisdiction under s 75(v) of the Constitution, particularly the operation of and limits upon privative clauses.[182] The High Court has essentially held that decisions infected with jurisdictional error cannot be regarded as ‘authorised’ by the statute under which they were supposedly made and, therefore, cannot logically be protected by any privative clause contained in the same statute.[183] This approach presumes limitations upon the different arms of government that are relevant to the substantive legitimate expectation doctrine. The courts have claimed exclusive jurisdiction to determine questions of law, such as the application of jurisdictional error. The reservation of that jurisdiction to the High Court has provided a crucial means by which the Court has struck down privative clauses that seek to place administrative decisions beyond effective judicial scrutiny. However, just as this constitutional doctrine reserves certain issues to the sole province of the judicial arm of government, it must concede other issues to the executive arm. The invocation of jurisdictional error by McHugh and Gummow JJ as the principle that guides the reach of s 75(v) of the Constitution reinforces the point that the considerable body of doctrine that the Court has developed to define and defend its role has necessary limits. It is most unlikely that those limits could extend to encompass the substantive legitimate expectation doctrine without a major reformation of wider doctrines.

Any such change would almost certainly need to extend beyond the particular constitutional doctrines that lie in the path of the substantive legitimate expectation doctrine to the wider foundations of the Australian conception of the nature and scope of the judicial role. Mason alluded to this when he suggested that doctrines by which the High Court has expounded its jurisdiction under s 75(v) of the Constitution are underpinned by another more potent element of the separation of powers doctrine, namely, ‘the limited Australian conception of [the] content of judicial power.’[184]

He explained that ‘[t]his conception owes much to the influence of Sir Owen Dixon and his determination that the courts should be insulated from controversial issues which involve policy and which would bring the courts into controversy.’[185] That argument also suggests that the constitutional principles invoked in recent cases, such as jurisdictional error, which tend against the expansion of the judicial function to encompass the balancing exercise of Coughlan, express a longstanding judicial reluctance to engage in issues apt to be identified as ones of policy rather than law.

E Can a Remedy for Serious Administrative Injustice Bypass Constitutional Objections to the Substantive Legitimate Expectation Doctrine?

Although the constitutional obstacles to the adoption of the substantive legitimate expectation doctrine might seem insurmountable, Kirby J has suggested the adoption of a new remedy in judicial review that might bypass many of the obstacles that lie in the path of the doctrine. Kirby J raised this possibility in Re Minister for Immigration and Multicultural Affairs; Ex parte Applicant S20/2002, when his Honour suggested that a remedy for ‘serious administrative injustice’[186] might provide some sort of safety net or fall back in judicial review for cases where the specific grounds of review somehow did not apply.[187] Several comments can be made about this interesting possibility. The first is that this proposed remedy is remarkably similar to the English approach where the court has issued relief by reason of the degree of unfairness, particularly the notion of ‘conspicuous unfairness’ which was explained above.[188]

Kirby J’s suggested remedy is therefore open to many of the criticisms that have been levelled at the English equivalent, namely, that it lacks a coherent legal principle and that it simply provides a cloak for the imposition of subjective judicial impressions rather than legal doctrine. Secondly, the suggestion of Kirby J can only be understood in light of his Honour’s longstanding criticisms of the jurisdictional error doctrine.[189] A remedy to correct serious administrative injustice might be more likely to find favour if the jurisdictional error doctrine were discarded from Australian law, mainly because it would remove a doctrine that currently serves to demarcate the boundaries of judicial review in Australia. The main obstacle to that possibility, which was explained above in Part VI(D), is that the doctrine of jurisdictional error is so deeply embedded in the jurisprudence of the High Court that it is almost impossible that it could be discarded or radically altered without major revisions to the jurisprudence that governs the High Court’s jurisdiction under s 75(v) of the Constitution.

A further comment that can be made about Kirby J’s suggested remedy to correct serious administrative injustice is related to his Honour’s suggestion that the remedy would be available in an extreme case where ‘what has occurred does not truly answer to the description of the legal process that the parliament has laid down.’[190] His Honour asserted that the issue of a remedy in such extreme cases would fall within the scope of judicial power as it is currently understood in Australian constitutional doctrine because it was part of the judicial function to ‘uphold the rule of law … [and to ensure] minimum standards of decision making’.[191] In my view, this reasoning invokes the elements of judicial power in only a superficial or rhetorical sense because the only clear basis upon which the supposed illegality of an administrative decision can be judged is that of the merits or fairness of the decision. The emphasis of Kirby J on the ‘standards of decision making’ contains no guiding legal principle, which any new remedy must surely provide if it is to fall within the accepted scope of judicial power.

F What of the Constitutional Position at the State Level?

The separation of powers doctrine found to be implicit in Chapter III of the Australian Constitution does not apply to the states, though the High Court has held that Chapter III prevents state Parliaments from enacting legislation which invests in state courts functions which are incompatible with their exercise of federal judicial power.[192] Although this doctrine has been successfully invoked only once,[193] it is clearly not the only means by which federal constitutional doctrine can spill over into the principles applicable to state constitutional affairs. Another means is by way of s 73 of the Constitution, which establishes the High Court of Australia as the ultimate Australian court of appeal in both federal and state matters. In the exercise of its appellate jurisdiction, the High Court has promoted development of common law which is uniform throughout the Commonwealth of Australia.[194]

Insofar as principles of judicial review are ones of common law, the High Court has not been disposed to differentiate between the principles to be applied according to whether review is undertaken in exercise of a purely federal jurisdiction, or in exercise of a state jurisdiction which is exercisable by the High Court on appeal pursuant to s 73 of the Constitution. Thus, judicial appreciations of the implications of the separation of powers doctrine, cemented in the Constitution, may affect judicial responses to cases in which the validity of administrative action is challenged, regardless of whether the action was taken by a federal or state agency.

This point is illustrated by Quin,[195] which came before the High Court on appeal from a state court exercising a state supervisory jurisdiction. In theory, the case could have been decided without reference to the separation of powers doctrine, but observations made by Brennan J clearly suggest that the separation of powers doctrine has a bearing on the extent of the supervisory jurisdiction invested in state courts by state legislation. Brennan J stated that:

If it be right to say that the court’s jurisdiction in judicial review goes no further than declaring and enforcing the law prescribing the limits and governing the exercise of power, the next question immediately arises: what is the law? And that question, of course, must be answered by the court itself. In giving its answer, the court needs to remember that the judicature is but one of the three co-ordinate branches of government and that the authority of the judicature is not derived from a superior capacity to balance the interests of the community against the interests of an individual. The repository of administrative power must often balance the interests of the public at large and the interests of minority groups or individuals. The courts are not equipped to evaluate the policy considerations which properly bear on such decisions, nor is the adversary system ideally suited to the doing of administrative justice: interests which are not represented must often be considered.[196]

This reasoning does not anticipate any significant difference between the constitutional limitations that attend state courts and those applicable to courts exercising federal jurisdiction. It also suggests that developments at the state level which might be thought to encourage the adoption of the substantive legitimate expectation doctrine or other novel developments in judicial review are unlikely to do so unless those developments conform to federal constitutional requirements. An obvious example is the Charter of Human Rights and Responsibilities Act 2006 (Vic) (‘Victorian Charter of Rights’), which draws heavily from the Human Rights Act 1998 (UK) c 42.[197] The latter Act is widely acknowledged to have stimulated many changes to English judicial review, particularly the adoption of a more intense standard of review when decisions affect fundamental rights. It is most unlikely that the Victorian Charter of Rights would have a similar effect by reason of the wider constraints imposed by the Australian Constitution.

G Reflections on Australian Objections to the Substantive Legitimate Expectation Doctrine

Australian law has long conceived the legitimate expectation doctrine to be procedural rather than substantive in character. This conception of the doctrine necessarily limited its growth in the years before the Coughlan decision because the legitimate expectation doctrine was left with little, if any, role to play in the modern expansion of the requirements of procedural fairness. The emphatic rejection of estoppel in Australian administrative law highlights a separate but logically related problem for the substantive legitimate expectation doctrine, namely, that there is no coherent body of doctrine in public law which supports the view that considerations of fairness can or should be invoked to ensure that administrative officials may sometimes be held to their word. The absence of estoppel in Australian administrative law removed an important doctrinal stepping stone for a wider, more substantive, notion of fairness. It has also enabled the rule against fettering to retain a level of influence that it has lost in English law.[198]

These obstacles to the substantive legitimate expectation doctrine could be characterised as an illustration of the unprincipled or bottom-up nature of Australian judicial review.[199] According to this approach, judicial review has evolved in Australia without reference to a coherent or unifying principle (which would constitute a top-down doctrine), but instead developed on a case-by-case basis (as occurs with a bottom-up doctrine). At the same time, Australia largely bypassed the intense debate in England about the possible foundations of judicial review, which focused on the rival theories drawn from either the notion of ultra vires or from the common law. In contrast, the statutory intent or ultra vires theory argued that judicial review ultimately rested on parliamentary authority, according to which the role of the courts was to police the limits of a power impliedly intended by Parliament.[200] The common law theory argued that much of the authority for and grounds of judicial review were based ultimately in the common law.[201] The statutory intent or ultra vires theory offered parliamentary authority (in the form of legislation) as the ultimate foundation for judicial review.[202] By the time most commentators settled on one of the compromise theories that drew from each rival theory,[203] English courts had largely discarded references to ‘ultra vires’ in favour of the nomenclature of either ‘the rule of law’ or ‘abuse of power’. Accordingly, this shift has involved doctrines that can easily accommodate the substantive legitimate expectation doctrine.

Early Australian cases on the legitimate expectation doctrine were sometimes explained as embodying differing theories of judicial review not unlike the two main English theories.[204] The tendency to categorise Australian decisions accordingly has continued for more recent decisions on the legitimate expectation doctrine,[205] but it is increasingly clear that the Australian Constitution provides the main point of reference for judicial review principles.[206]

This focus on constitutional issues directs attention to the organising principles by which the High Court articulates the separation of powers doctrine as it affects judicial review, and this presents a far more significant and longstanding obstacle to the substantive legitimate expectation doctrine.

At this point, it is useful to recall the comment of Mason, noted above, that the Australian Constitution was designed as ‘a delineation of government powers rather than as a charter of citizen’s [sic] rights.’[207]

At first glance, this argument illuminates why the Australian constitutional framework might be less receptive to changes within English law which are naturally more attuned to individual rights, such as the substantive legitimate expectation doctrine. However, Mason’s comment draws attention to a much deeper undercurrent of Australian constitutional law. In my view, it may be argued that the institutional focus of the Australian Constitution provided a natural breeding ground for the ‘strict and complete legalism’ favoured by Sir Owen Dixon.[208] This legalism, in turn, provided the foundation for the relatively limited Australian conception to which Mason also refers. It is that limited conception of the judicial function which continues to present the single most significant obstacle to any Australian adoption of the substantive legitimate expectation doctrine because many judges and commentators reject the doctrine on the simple basis that it is incompatible with the Australian doctrine of the separation of powers.[209]

It could be argued that the role of the Australian Constitution can be overstated because, as Selway explains, ‘[i]t provides the ultimate justification for judicial review and sets its parameters, but does not explain the detail of its operation.’[210]

Is it possible that the ‘detail of operation’ of the Australian Constitution can provide room for the adoption of the substantive legitimate expectation doctrine? The approach of Brennan J in Quin suggests not, because it makes emphatically clear that the role of the courts is limited and that these limits simply preclude a doctrine such as that of the substantive legitimate expectation. Stephen Gageler has characterised the approach of Brennan J in Quin as

top down reasoning at the highest level. From the constitutional conception of the nature of judicial power, there is derived a single principle which then informs both the scope and content of judicial review. That single principle is the duty of the court to declare and enforce the law.[211]

According to this view, the adoption of a doctrine such as that of the substantive legitimate expectation is not possible without a wholesale revision to fundamental constitutional doctrines. Any wholesale revision could carry grave risks to the scope of judicial review as it is currently known because the separation of powers doctrine has provided a useful tool by which the High Court has been able to fortify its constitutionally entrenched role to secure an entrenched minimum constitutional guarantee of judicial review and resist successive privative clauses. If the High Court were to countenance a revision to constitutional doctrine that was sufficiently far-reaching to enable the adoption of the substantive legitimate expectation doctrine, it might weaken the very doctrine upon which important aspects of its jurisdiction are anchored.

At this point, it is important to note that the Australian Constitution is not the only avenue of judicial review. The main statutory avenue of judicial review at the federal level, which is the ADJR Act,[212] presents further obstacles to innovations in judicial review. There is widespread agreement that the ADJR Act introduced many important procedural reforms to judicial review,[213] but in recent years the effect of the Act on the substantive law of judicial review has come under criticism. Kirby J has complained that the ADJR Act has ‘retarded’ the evolution of judicial review in Australia because its codification of the grounds of review has ‘somewhat arrested’[214] any further development of existing or new grounds.[215]

There are many cases which illustrate the limited effect of the statutory codification of the grounds of review. In Kioa v West (‘Kioa’), for example, the High Court considered whether the obligation to observe the requirements of natural justice arose from the common law or the statute that conferred the relevant statutory power.[216] The High Court divided sharply on this point, but all members of the Court accepted that the ADJR Act offered no real assistance to determining the source or scope of natural justice.[217] This reasoning suggests that the ADJR Act provided a snapshot of the grounds of review at a point in time but left open the important questions of whether and how that picture might change over time. Some commentators thought that the architects of the ADJR Act had sought to accommodate possible developments in the principles of judicial review with the inclusion of two relatively open-ended grounds that enable review of a decision that was ‘otherwise contrary to law’ or was an ‘exercise of a power in a way that constitutes abuse of the power.’[218] However, both grounds have failed to gain traction and have been ignored by applicants and the courts to such an extent that they may be regarded as dead letters.[219]

Professor Aronson has suggested that the failure of the ADJR Act to stimulate any significant innovation in judicial review in Australia is unsurprising in light of the Act’s emphasis on procedural rather than substantive reform. He drew support from the grounds of review in the ADJR Act, which he argued

say nothing about the rule of law, the separation of powers, fundamental rights and freedoms, principles of good government or (if it be different) good administration, transparency of government, fairness, participation, accountability, consistency of administrative standards, rationality, legality, impartiality, political neutrality or legitimate expectations. Nor does ADJR mention the Thatcher era’s over-arching goals of efficiency, effectiveness and economy … ADJR’s grounds are totally silent on the relatively recent discovery of universal human rights to autonomy, dignity, respect, status and security. Nowhere does ADJR commit to liberal democratic principles, pluralism, or civic republicanism.[220]

On this view, a key problem with the ADJR Act was its focus on correcting the perceived existing problems of judicial review at the expense of any attention to the wider purpose of those changes or the direction in which they might lead judicial review. Could amendments to the ADJR Act overcome these apparent defects and stimulate changes such as the adoption of a concept of substantive unfairness? Aronson questions whether the introduction of a ‘general principles’ clause to the ADJR Act is desirable, largely because it would be difficult to devise a workable statement of ‘over-arching principles’. He was equally concerned that any statement of general principles to guide the operation of the ADJR Act would struggle to strike the right balance (particularly in light of the many and potentially competing principles that could be invoked in support of judicial review) and that there was always the danger that an unduly cautious statement of principles could narrow rather than expand the potential for change.[221] Aronson also questions whether it would be appropriate to leave the task of devising general guiding normative principles for judicial review to the judiciary.[222]

Three key points can be extracted from Aronson’s analysis for the purposes of this article. First, the apparent rejection by Australian law of new developments such as the substantive legitimate expectation doctrine can only be fully understood by reference to the wider framework of judicial review. Although Lam suggests that the principle is incompatible with Australia’s constitutional arrangements, it is important to note that the ADJR Act provides equally infertile terrain for the principle. More generally, it has not encouraged judicial innovation which might have helped pave the way for the acceptance of new grounds of judicial review. Secondly, it is doubtful whether reforms to the ADJR Act may cure the fundamentally procedural nature of the Act. The open-ended grounds of review in the Act that might have encouraged the adoption of new grounds or principles of review have not taken root, and it is difficult to imagine that the judicial approach to those grounds might now suddenly change almost 30 years after their commencement. Any attempt to introduce further new grounds might founder as previous efforts have done. Thirdly, any reform to include a statement of general or guiding principle to the Act might create at least as many problems as it solves. If a guiding statement was introduced to assist the interpretation of the ADJR Act in a manner that enabled or even encouraged the adoption of new principles, such as the substantive legitimate expectation doctrine, this would, if successful, require a fundamental reconsideration of the Australian approach to judicial review. A change of this nature could have such far-reaching consequences that one might ask whether it is appropriate that it be limited to a single Act. One might also ask what effect such a statutory statement might have on the common law approach to judicial review and, more importantly, how the judges might respond to any legislative attempt to ‘guide’ them. Would the judges perceive such guidance to be dictation? If so, how would the courts react? The need for caution would be so strong in the drafting of any amendments to the ADJR Act that it is unlikely that any legislative attempt to reform the ADJR Act along these lines could accommodate these concerns.

The difficulties associated with any reform beg the question of whether more far-reaching reform might be a more appropriate vehicle to stimulate change to the substantive law of judicial review. One such change would be constitutional reform. That option could be considered fanciful in light of the notorious reluctance of the Australian people to vote in favour of constitutional reform, but it is useful to note a neglected model of constitutional reform from another part of the Commonwealth: the Constitution of the Republic of South Africa provides that ‘[e]veryone has the right to administrative action that is lawful, reasonable and procedurally fair.’[223] A related provision of the South African Constitution obliges the federal government to enact legislation to give effect to this right, which has resulted in the enactment of the Promotion of Administrative Justice Act 2000 (RSA).[224] The Constitutional Court of South Africa has confirmed that the constitutional right to administrative justice and the associated legislation that seeks to give effect to that right provide a unifying normative principle to South African administrative law, namely, the promotion of an accountable, plural democracy in which administrative decision-making meets the values and goals expressed in constitutional and legislative documents.[225] The South African model confirms that normative principles can be introduced into a constitutional framework and that the legislature can provide general guidance to the judiciary in a form that does not restrict or infringe upon the judicial role. It must be accepted, however, that the prospect of such radical constitutional change is highly unlikely in Australia.

A more likely reform would be the introduction of a bill or charter of rights along the lines of that introduced in England or New Zealand.[226] At present, it appears that any such reform at the federal level would take the form of legislation rather than a constitutional amendment.[227] Would a bill or charter of rights provide a path by which the substantive legitimate expectation doctrine could be received into Australian administrative law? That question raises many complex issues, which can only be touched upon in this article. There are several reasons why such a bill or charter would not necessarily affect the basic obstacles that lie in the path of the substantive legitimate expectation doctrine. First, there is a growing body of English opinion which suggests that the Human Rights Act 1998 (UK) c 42 has effected a fundamental shift in the relationship between the courts and the legislature and that this shift has greatly influenced English developments in judicial review.[228]

Although the nature and extent of this apparent shift is the subject of considerable debate, it must be accepted that a legislative rather than constitutional bill or charter of rights in Australia could not alter the constitutional relationship between the different arms of government. More particularly, it could not remove the perceived constitutional obstacles to the adoption of the substantive legitimate expectation doctrine into Australian law. Secondly, it is important to note that the many questions about the relationship between the courts and legislature in judicial review, which are articulated in Australia within the doctrine of the separation of powers, have arisen in England since the enactment of the Human Rights Act 1998 (UK) c 42 as part of the debate about the doctrine of proportionality.[229] A key question about proportionality also lies at the heart of Australia’s separation of powers doctrine, namely, to what extent should the courts review administrative decisions?[230] The important point for present purposes is that it cannot be assumed that the adoption of a legislative bill or charter of rights would remove either the constitutional objections to the substantive legitimate expectation doctrine or the underlying issues about the demarcation of the judicial and executive roles.[231]

A bill or charter of rights could provoke one important change. If the law of England is any guide, it could invigorate public law and pave the way for a more dynamic approach to judicial review which could, in turn, pave the way for the adoption of grounds such as the substantive legitimate expectation doctrine. This possibility is not intended to discount the many constitutional objections that might be made to a bill or charter of rights that is introduced by way of legislation rather than constitutional amendment.[232] It is simply to suggest that the legislative bill or charter of rights could provide one means by which the judicial formalism explained above could be overcome. Whether that is desirable is an entirely different question.[233]

VII CONCLUDING OBSERVATIONS

There is a clear difference between a judicial decision that influences administrative decision-making and one which directs it. Coughlan was a novel case in part because it blurred that distinction, but the Court of Appeal of England and Wales did not go to the length of reviewing the decision of the health authority on its merits, as if it had been empowered by statute to hear and determine an appeal against that decision on the merits. Rather, the Court of Appeal found that the health authority had not presented good reasons why it had acted as it did and disappointed what was considered to be Ms Coughlan’s legitimate expectation. Although the Court did not assume the power to make or remake the decision on Ms Coughlan’s living arrangements, the Court’s balancing of the factors for and against the decision of the health authority edged towards a review of the decision on the merits.

But should a new ground of review be rejected simply because it might draw a court closer to the merits of a case? It has long been recognised that some grounds of judicial review which relate to the legality of administrative actions grant the courts considerable latitude in determining what are to be regarded as legal limitations on powers conferred on administrative agencies of government. Unreasonableness in the Wednesbury sense, for example, often involves judicial consideration of the substance of administrative decisions, and, to some extent, appraisal of them according to standards which courts regard as ones to which administrative agencies should conform in the exercise of their functions.[234]

Chief Justice Murray Gleeson has explained that this aspect of the ground of unreasonableness does not necessarily mean that a clear distinction between principles of legality and merits cannot be drawn. The difference between the two, his Honour explained extra-judicially, ‘is not always clear-cut; but neither is the difference between night and day. Twilight does not invalidate the distinction between night and day’.[235] A similar point must surely apply to all grounds of judicial review to the extent that a consideration of the legality of decision-making requires a court to consider the context within which decisions are made. The differing extent to which different grounds of review might require the context of a decision to be considered does not of itself, according to the remarks of Chief Justice Gleeson, obscure the basic distinction between review and appeal.

The separation of powers doctrine enshrined in Chapter III of the Australian Constitution does not, it may be argued, prohibit courts exercising a federal supervisory jurisdiction from adjudging federal administrative action invalid on the ground that relevant legitimate expectations had not been taken into account by the authors of that action. According to this view, a substantive legitimate expectation may be treated as a relevant consideration to which a decision-maker must have regard, as might any unfairness that resulted from the denial of that expectation. However, it is doubtful that constitutional doctrines would permit an Australian court to go further because the separation of powers doctrine may be interpreted as having placed constraints on what courts exercising federal jurisdiction may properly do when the validity of administrative decisions is challenged on the ground that, in the circumstances of the individual case, the decision-maker was bound to exercise a discretionary power in a way which fulfilled the complainant’s legitimate expectation. In cases of this description, it may be argued that the court would be exceeding its judicial powers were it to adjudge the validity of the administrative action by assessing countervailing public and private interests. For this reason, the central judicial task devised in Coughlan is beyond the reach of Australian courts. The issue is ultimately one of boundaries, or rather whether a ground blurs the merits/legalities distinction sufficiently that it may be argued to breach that distinction.

Abuse of power is a recognised ground for judicial review of administrative action under the ADJR Act. The High Court of Australia has not yet had occasion to consider the extent to which application of that ground of review may be constrained by the separation of powers doctrine. While the Court may accept that abuse of power is a ground on which federal administrative action is judicially reviewable, it would, no doubt, insist that the ground not be employed as a subterfuge for merits review. It would likely take the view that the separation of powers doctrine precludes judicial review of federal administrative action to the extent manifested in the case of Coughlan. On this view, the doctrine established in Coughlan could not be fostered in Australia by indirect means.


[∗] BA, LLB (Hons), PhD (Monash); Senior Lecturer, Faculty of Law, Monash University. I am grateful to Bruce Dyer for his helpful comments.

[1] The leading English case is R v North and East Devon Health Authority; Ex parte Coughlan [2001] QB 213 (‘Coughlan’). The case and subsequent developments are discussed in P P Craig, Administrative Law (5th ed, 2003) 646–80; Iain Steele, ‘Substantive Legitimate Expectations: Striking the Right Balance?’ (2005) 121 Law Quarterly Review 300; Mark Elliott, ‘Legitimate Expectations and the Search for Principle: Reflections on Abdi & Nadarajah[2006] Judicial Review 281. For an Australian perspective, see Cameron Stewart, ‘Substantive Unfairness: A New Species of Abuse of Power?’ (2000) 28 Federal Law Review 617; Cameron Stewart, ‘The Doctrine of Substantive Unfairness and the Review of Substantive Legitimate Expectations’ in Matthew Groves and H P Lee (eds), Australian Administrative Law: Fundamentals, Principles and Doctrines (2007) 280.

[2] Mark Aronson, Bruce Dyer and Matthew Groves, Judicial Review of Administrative Action (3rd ed, 2004) 395–400; Ian Holloway, Natural Justice and the High Court of Australia: A Study in Common Law Constitutionalism (2002) 154–95.

[3] See, eg, Haoucher v Minister of State for Immigration and Ethnic Affairs [1990] HCA 22; (1990) 169 CLR 648, 652 (Deane J) (‘Haoucher’); Annetts v McCann (1990) 170 CLR 596, 598 (Mason CJ, Deane and McHugh JJ); Ainsworth v Criminal Justice Commission [1992] HCA 10; (1992) 175 CLR 564, 577 (Mason CJ, Dawson, Toohey and Gaudron JJ); Minister of State for Immigration and Ethnic Affairs v Teoh [1995] HCA 20; (1995) 183 CLR 273, 311 (McHugh J) (‘Teoh’). On the possible limitation or exclusion of the duty to observe the requirements of procedural fairness, see generally Aronson, Dyer and Groves, above n 2, 432–8.

[4] The role of the legitimate expectation doctrine during this period is explained in: Graeme Johnson, ‘Natural Justice and Legitimate Expectations in Australia’ [1985] FedLawRw 2; (1985) 15 Federal Law Review 39; Mark Aronson and Nicola Franklin, Review of Administrative Action (1987) 111–18; Pamela Tate, ‘The Coherence of “Legitimate Expectations” and the Foundations of Natural Justice’ [1988] MonashULawRw 2; (1988) 14 Monash University Law Review 15. The corresponding period in English law has been described as one in which the scope of the legitimate expectation doctrine was dominated by ‘the “procedure only” school of thought’: R (Nadarajah) v Secretary of State for the Home Department [2005] EWCA Civ 1363 (Unreported, Laws and Thomas LJJ and Nelson J, 22 November 2005) [49] (Laws LJ) (‘Nadarajah’).

[5] See, eg, Schmidt v Secretary of State for Home Affairs [1969] 2 Ch 149 (holder of entry permit had a legitimate expectation that they would be able to enter and remain in the country according to the terms of the permit — that expectation could not be defeated without granting the permit holder a chance to put his views against a possible adverse decision); Heatley v Tasmanian Racing and Gaming Commission [1977] HCA 39; (1977) 137 CLR 487 (person who paid entry fee to a racing meeting had a legitimate expectation that they would be able to remain for the whole of the meeting in accordance with the terms of entry); FAI Insurances Ltd v Winneke [1982] HCA 26; (1982) 151 CLR 342 (applicant, who was a preferred provider of insurance to a government scheme, had a legitimate expectation that it would be given notice of, and a hearing to argue against, a possible decision not to renew its status as a preferred provider); Council of Civil Service Unions v Minister for the Civil Service [1985] AC 374 (a union that had been regularly consulted about workplace changes had a legitimate expectation that consultation would, in the absence of exceptional circumstances, continue).

[6] This taxonomy is taken from Craig, Administrative Law, above n 1, 641. See also Secretary of State for the Home Department v The Queen (Rashid) [2005] EWCA Civ 744 (Unreported, Pill, May and Dyson LJJ, 16 June 2005) [45] (Dyson LJ) (‘Rashid’).

[7] Salemi v MacKellar [No 2] [1977] HCA 26; (1977) 137 CLR 396, 404 (Barwick CJ) (‘Salemi’). This view was rejected by the Privy Council in A-G (HK) v Ng Yuen Shiu [1983] 2 AC 629, 636 (Lords Fraser, Scarman, Bridge and Brandon and Sir John Megaw) (‘Ng Yuen Shiu’).

[8] Kioa v West [1985] HCA 81; (1985) 159 CLR 550, 583 (Mason J) (‘Kioa’). This facilitative role of the legitimate expectation doctrine was conceded even by McHugh J, who was a longstanding critic of the concept: see Haoucher [1990] HCA 22; (1990) 169 CLR 648, 680–1.

[9] The requirement was invoked in Heatley v Tasmanian Racing and Gaming Commission [1977] HCA 39; (1977) 137 CLR 487, 508 (Aickin J); Ng Yuen Shiu [1983] 2 AC 629, 636 (Lords Fraser, Scarman, Bridge and Brandon and Sir John Megaw); Haoucher [1990] HCA 22; (1990) 169 CLR 648, 659 (Dawson J) (describing the expectation of the appellant as ‘reasonable enough to be described as legitimate’); Teoh [1995] HCA 20; (1995) 183 CLR 273, 291 (Mason CJ and Deane J), 314 (McHugh J). Variants of ‘reasonable’ were also often used: see, eg, Salemi [1977] HCA 26; (1977) 137 CLR 396, 439 (Stephen J) (suggesting that an expectation must be ‘well-founded’).

[10] Australian law on this point remains unsettled. Some judges have held that the person claiming the benefit of a legitimate expectation need not prove knowledge of or reliance upon the facts supporting the expectation: see, eg, Kioa [1985] HCA 81; (1985) 159 CLR 550, 618 (Brennan J); Haoucher [1990] HCA 22; (1990) 169 CLR 648, 670 (Toohey J); Teoh [1995] HCA 20; (1995) 183 CLR 273, 291 (Mason CJ and Deane J), 301 (Toohey J). Others have suggested that reliance, or at least subjective knowledge of the facts supporting the legitimate expectation, is either essential or highly desirable for a claimant of a legitimate expectation: see, eg, Teoh [1995] HCA 20; (1995) 183 CLR 273, 313–14 (McHugh J); Re Minister for Immigration and Multicultural and Indigenous Affairs; Ex parte Lam (2003) 214 CLR 1, 47 (Callinan J) (‘Lam’). In a more recent case concerning a denial of procedural fairness which did not directly invoke the legitimate expectation doctrine, a majority of the High Court suggested that proof of detriment or prejudice on the part of a person affected was required in some but not all cases: Applicant NAFF of 2002 v Minister for Immigration and Multicultural and Indigenous Affairs [2004] HCA 62; (2004) 221 CLR 1, 12 (McHugh, Gummow, Callinan and Heydon JJ).

[11] See below Part V(C).

[12] Tung v Director of Immigration [2002] 1 HKLRD 561, 600 (Li CJ, Chan and Ribeiro PJJ and Mason NPJ). The circumstances of that case and its use of the legitimate expectation doctrine are explained in Teresa Martin, ‘Hong Kong Right of Abode: Ng Siu Tung & Others v Director of Immigration — Constitutional and Human Rights at the Mercy of China’ (2004) 5 San Diego International Law Journal 465. See also Benny Y T Tai and Kevin K F Yam, ‘The Advent of Substantive Legitimate Expectations in Hong Kong: Two Competing Visions’ [2002] Public Law 688.

[13] [2001] QB 213.

[14] [1984] UKHL 5; [1985] AC 835.

[15] Ibid 852 (Lord Scarman), 866–7 (Lord Templeman).

[16] Ibid 866 (Lord Templeman).

[17] This possibility was flagged before Preston by commentators concerned about the uncritical expansion of the obligation to observe the rules of natural justice. They argued that if ‘fairness’ was conceived in anything other than a procedural sense, the courts would be easily tempted into imprecise and open-ended grounds of review that would greatly and unjustifiably expand the scope of judicial review: see, eg, David J Mullan, ‘Natural Justice and Fairness — Substantive as Well as Procedural Standards for the Review of Administrative Decision-Making?’ (1982) 27 McGill Law Journal 250.

[18] Preston [1984] UKHL 5; [1985] AC 835, 866–7, where Lord Templeman (with whom the other Lords agreed) accepted that conduct that a taxpayer could complain about on the grounds of unfairness could include conduct that might, but did not necessarily have to, amount to a breach of contract or breach of a representation to which estoppel might apply.

[19] Ibid 864 (Lord Templeman).

[20] Ibid 852 (Lord Scarman), 866–7 (Lord Templeman).

[21] Ibid.

[22] Craig, Administrative Law, above n 1, 648–9.

[23] This view was mainly supported by Lord Denning MR: see, eg, HTV Ltd v Price Commission [1976] ICR 170, 185–6; Laker Airways Ltd v Department of Trade [1976] EWCA Civ 10; [1977] QB 643, 707. See also R v Secretary of State for the Home Department; Ex parte Khan [1984] EWCA Civ 8; [1985] 1 All ER 40. There was also early academic support for either the adoption of substantive legitimate expectations or a very rigorous protection of procedural legitimate expectations: see C F Forsyth, ‘The Provenance and Protection of Legitimate Expectations’ (1988) 47 Cambridge Law Journal 238; P P Craig, ‘Representations by Public Bodies’ (1977) 93 Law Quarterly Review 398; P P Craig, ‘Substantive Legitimate Expectations in Domestic and Community Law’ (1996) 55 Cambridge Law Journal 289.

[24] See, eg, R v Ministry of Agriculture Fisheries and Food; Ex parte Hamble (Offshore) Fisheries Ltd [1995] 2 All ER 714; R v Gaming Board for Great Britain; Ex parte Kingsley [1996] COD 241; R v Inland Revenue Commissioners; Ex parte Unilever plc [1996] STC 681 (‘Unilever’). During this time, the idea of substantive legitimate expectations was sometimes doubted or openly disapproved: see, eg, R v Secretary of State for Transport; Ex parte Richmond upon Thames London Borough Council [1994] 1 All ER 577, 596 (Laws J); R v Secretary of State for the Home Department; Ex parte Hargreaves [1997] 1 All ER 397, 412 (Hirst LJ).

[25] [2001] QB 213.

[26] Ibid 214.

[27] Ibid 241 (Lord Woolf MR for the Court).

[28] That standard comes from Associated Provincial Picture Houses Ltd v Wednesbury Corporation [1947] EWCA Civ 1; [1948] 1 KB 223. It requires that a court only find a decision unlawful if it is satisfied that no reasonable decision-maker could have reached that decision. This ground is reproduced in many Australian statutory schemes for judicial review: see, eg, Administrative Decisions (Judicial Review) Act 1977 (Cth) s 5(2)(g) (‘ADJR Act’); Administrative Decisions (Judicial Review) Act 1989 (ACT) s 5(2)(g); Judicial Review Act 1991 (Qld) s 23(g); Judicial Review Act 2000 (Tas) s 20(g).

[29] Coughlan [2001] QB 213, 242 (Lord Woolf MR for the Court).

[30] Although English administrative law generates fewer cases about the content of natural justice than Australian law, the essential requirements of procedural fairness in such instances in England are very similar to those in Australia: see, eg, Sir William Wade and Christopher Forsyth, Administrative Law (9th ed, 2004) 496–522. On the equivalent Australian law, see Aronson, Dyer and Groves, above n 2, 468–562.

[31] Coughlan [2001] QB 213, 242 (Lord Woolf MR for the Court) (emphasis in original).

[32] Ibid (where the Court referred to ‘an enforceable expectation of a substantive benefit’).

[33] Ibid 243.

[34] Ibid 246. See also the Court’s suggestion that the ground of unreasonableness can also touch ‘the intrinsic quality of the decision’: at 243.

[35] Ibid 251.

[36] Ibid 246.

[37] Aronson, Dyer and Groves, above n 2, 355.

[38] Ibid 357–8.

[39] A point established by empirical research which has found that matters remitted to decision-makers after a successful application for judicial review often reach a different outcome when reconsidered in light of the reasons for decision provided by the courts: Robin Creyke and John McMillan, ‘Executive Perceptions of Administrative Law — An Empirical Study’ (2002) 9 Australian Journal of Administrative Law 163.

[40] Coughlan [2001] QB 213, 246–7 (Lord Woolf MR for the Court). The Court referred to the concept of abuse of power as ‘established’: at 247. The Court also explained that ‘the doctrine of legitimate expectation has emerged as a distinct application of the concept of abuse of power in relation to substantive as well as procedural benefits’: at 246. The Court acknowledged, however, that the role of courts in the substantive legitimate expectation doctrine was ‘still controversial’: at 242.

[41] Ibid 247.

[42] See generally Theodore Fortsakis, ‘Principles Governing Good Administration’ (2005) 11 European Public Law 207; Eva Nieto-Garrido and Isaac Martín Delgado, European Administrative Law in the Constitutional Treaty (2007). For an English perspective on these values, see Lord Millett, ‘The Right to Good Administration in European Law’ [2002] Public Law 309.

[43] R (Bancoult) v Secretary of State for Foreign and Commonwealth Affairs [No 2] [2007] EWCA Civ 498; [2008] QB 365, 407 (Sedley LJ).

[44] Ibid. See also at 415 (Waller LJ), 420 (Clarke MR). Coughlan has been cited with apparent approval several times by the House of Lords: see, eg, R v Secretary of State for the Home Department; Ex parte Hindley [2000] UKHL 21; [2001] 1 AC 410, 421, where Lord Hobhouse described the reasoning in Coughlan as ‘valuable’. Coughlan has been mentioned several other times by the House of Lords either with tacit approval or without adverse comment: see, eg, R v Ministry of Defence; Ex parte Walker [2000] UKHL 22; [2000] 2 All ER 917, 924 (Lord Slynn); R v East Sussex County Council; Ex parte Reprotech (Pebsham) Ltd [2002] UKHL 8; [2002] 4 All ER 58, 66 (Lord Hoffmann) (‘Reprotech’); R (Mullen) v Secretary of State for the Home Department [2004] EWCA Civ 243; [2005] 1 AC 1, 48 (Lord Steyn); YL v Birmingham City Council [2007] UKHL 27; [2008] 1 AC 95, 139–40 (Lord Mance).

[45] The absence of open discussion of values in judicial review and the values that might support judicial review, and perhaps administrative law more generally, are considered in Aronson, Dyer and Groves, above n 2, 1–8.

[46] [1999] EWCA Civ 2100; [2000] 1 WLR 1115.

[47] See ibid 1125 (Gibson LJ), 1134 (Sedley LJ), where their Lordships were mindful of this point.

[48] Ibid 1132, where Sedley LJ conceded that the legislative changes provided the Minister a discretion to devise transitional arrangements, including those offered to Mrs Begbie’s daughter.

[49] Sir John Laws had rehearsed a similar argument extra-judicially many years earlier: Sir John Laws, ‘Public Law and Employment Law: Abuse of Power’ [1997] Public Law 455.

[50] Begbie [1999] EWCA Civ 2100; [2000] 1 WLR 1115, 1129.

[51] Ibid 1129–30 (citations omitted).

[52] Coughlan [2001] QB 213, 241–2 (Lord Woolf MR for the Court).

[53] Begbie [1999] EWCA Civ 2100; [2000] 1 WLR 1115, 1130.

[54] The examples of this category cited in Coughlan were: a change of parole policy which affected many prisoners, with the reasons for the policy change being for general administrative and political reasons and not in response to the case of any particular prisoner (in Re Findlay [1985] AC 318); and alteration of a policy affecting the entitlement of prisoners to home leave (in R v Secretary of State for the Home Department; Ex parte Hargreaves [1997] 1 All ER 397).

[55] Examples of this category discussed in Coughlan were largely revenue cases in which quite specific representations had been made to taxpayers, including: Preston [1984] UKHL 5; [1985] AC 835; Unilever [1996] STC 681.

[56] Begbie [1999] EWCA Civ 2100; [2000] 1 WLR 1115, 1130.

[57] Ibid 1131. The House of Lords has made similar remarks in many recent cases: see, eg, R (ProLife Alliance) v British Broadcasting Corporation [2003] UKHL 23; [2004] 1 AC 185, 240, where Lord Hoffmann noted that the various features of the courts and legislature made each institution better suited to resolving different types of disputes.

[58] [2001] EWCA Civ 607; [2002] 1 WLR 237, 245 (Schiemann LJ for the Court).

[59] Ibid 245.

[60] Lord Steyn, ‘Dynamic Interpretation amidst an Orgy of Statutes’ [2004] European Human Rights Law Review 245, 256. See also Jeffrey Jowell, ‘Judicial Deference and Human Rights: A Question of Competence’ in Paul Craig and Richard Rawlings (eds), Law and Administration in Europe: Essays in Honour of Carol Harlow (2003) 67, 71–3.

[61] See Miss Behavin’ Ltd v Belfast City Council [2007] 3 All ER 1007, 1017–18, where Baroness Hale explained that in human rights cases the court is concerned with whether a claimant’s human rights have been infringed. In judicial review cases the court is concerned with whether an applicant’s human rights were taken into account by a decision-maker.

[62] [2005] EWCA Civ 1363 (Unreported, Laws and Thomas LJJ and Nelson J, 22 November 2005).

[63] Ibid [67].

[64] Ibid [68].

[65] Ibid.

[66] Ibid.

[67] Ibid.

[68] For an insightful analysis of such judicial innovation in the context of American constitutional law, see David Crump, ‘How Do the Courts Really Discover Unenumerated Fundamental Rights? Cataloguing the Methods of Judicial Alchemy’ (1996) 19 Harvard Journal of Law and Public Policy 795.

[69] Unilever [1996] STC 681, 695 (Brown LJ).

[70] R v Secretary of State for the Home Department; Ex parte Zeqiri [2002] UKHL 3 (Unreported, Lords Slynn, Mackay, Hoffmann, Millett and Rodger, 24 January 2002) [44] (Lord Hoffmann). This case turned largely on its facts. The House of Lords held that the conduct pleaded as an alleged representation was not as clear as the applicant asserted, so the further claim that the actions contrary to this representation caused conspicuous unfairness failed. The Lords did not refer to Coughlan.

[71] [2005] EWCA Civ 744 (Unreported, Pill, May and Dyson LJJ, 16 June 2005).

[72] At this time, Kurdish claimants comprised a very large portion of refugee applicants and the policy governing Kurds and other Iraqi refugees had changed several times to reflect political changes in the countries from which applicants came. The decision-maker’s ignorance of the policy was, in the circumstances, astonishing. The wider circumstances of the policy changes and the ignorance of the decision-makers of the relevant policies in Rashid are explained in R (A) v Secretary of State for the Home Department [2006] EWHC 526 (Admin) (Unreported, Collins J, 22 March 2006) [1]–[9], [20]–[27].

[73] Rashid [2005] EWCA Civ 744 (Unreported, Pill, May and Dyson LJJ, 16 June 2005) [53] (Dyson LJ).

[74] Ibid [34].

[75] Ibid [36].

[76] This is available under the Immigration Act 1971 (UK) c 77, ss 3–4.

[77] R (A) v Secretary of State for the Home Department [2006] EWHC 526 (Admin) (Unreported, Collins J, 22 March 2006) [34].

[78] Mark Elliott, ‘Legitimate Expectation, Consistency and Abuse of Power: The Rashid Case’ [2005] Judicial Review 281, 285.

[79] [2007] EWCA Civ 546 (Unreported, Carnwath and Moore-Bick LJJ and Lightman J, 19 June 2007).

[80] Ibid [54]. Moore-Bick LJ made similar remarks: at [69].

[81] Ibid [47] (Carnwath LJ).

[82] [2002] UKHL 8; [2002] 4 All ER 58.

[83] Ibid 66.

[84] Ibid.

[85] Ibid.

[86] See, eg, Begbie [1999] EWCA Civ 2100; [2000] 1 WLR 1115, 1124 (Gibson LJ). Gibson LJ notes here, however, that ‘it would be wrong to understate the significance of reliance in this area of the law. It is very much the exception, rather than the rule, that detrimental reliance will not be present when the court finds unfairness in the defeating of a legitimate expectation.’

[87] See R P Meagher, J D Heydon and M J Leeming, Meagher, Gummow and Lehane’s Equity: Doctrines and Remedies (4th ed, 2002) 831–55.

[88] Sedley LJ hinted that English public law could develop a right to damages for some forms of unlawful administrative action in F & I Services Ltd v Customs and Excise Commissioners [2001] STC 939, 959. The idea has not found favour in the cases but was examined in Law Commission, UK, Monetary Remedies in Public Law: A Discussion Paper (2004). The proposals of the Law Commission are analysed in Roderick Bagshaw, ‘Monetary Remedies in Public Law — Misdiagnosis and Misprescription’ (2006) 26 Legal Studies 4. See also Steele, above n 1, 322–7, where Steele suggests that a carefully crafted award of compensation, made by either courts or decision-makers, would not be as radical a step in English law as is widely thought, but concedes that many doctrinal and practical problems would obstruct such a development. Australian law on this issue is less settled. It is clear that no action in damages will lie for unlawful administrative decisions, but tortious liability can sometimes attach to an action done in consequence of an unlawful administrative decision: see, eg, Cowell v Corrective Services Commission of New South Wales (1988) 13 NSWLR 714. The extent to which this possibility relies on a relative theory of nullity in administrative action is considered in Mark Aronson, ‘Nullity’ in Matthew Groves (ed), Law and Government in Australia (2005) 139. The state of the law is unclear following Ruddock v Taylor (2005) 222 CLR 612. In that case, the High Court denied an action in damages brought by people wrongfully detained under s 189(1) of the Migration Act 1958 (Cth), which authorises the detention of people who are, or are reasonably suspected of being, unlawful non-citizens. The power to detain people who are ‘reasonably suspected’ of being unlawful non-citizens essentially protected the Minister, who had made an honest mistake. Although the case turned on this issue, the divergent views expressed by the justices of the Court suggest that there is uncertainty about the extent to which the High Court would allow an action in tort brought by a blameless person who complained about honest but unlawful administrative action.

[89] However, it is widely acknowledged that much of the detail of the doctrine remains either unsettled or is constantly evolving in England: see, eg, Adam Tomkins, ‘The Struggle to Delimit Executive Power in Britain’ in Paul Craig and Adam Tomkins (eds), The Executive and Public Law: Power and Accountability in Comparative Perspective (2006) 16. A recent example is the Constitutional Reform Act 2005 (UK) c 4, which enhanced judicial independence in several ways as part of an attempt to articulate the wider constitutional structure within which the judiciary operates. The Act clarifies many important constitutional principles that for centuries had been assumed rather than explained: see generally Diana Woodhouse, ‘The Constitutional Reform Act 2005 — Defending Judicial Independence the English Way’ (2007) 5 International Journal of Constitutional Law 153.

[90] Paul Craig and Søren Schønberg, ‘Substantive Legitimate Expectations after Coughlan[2000] Public Law 684, 694.

[91] Ibid (emphasis in original).

[92] However, it should be noted that English courts are now empowered to substitute their decisions for those of administrative officials in judicial review proceedings: Supreme Court Act 1981 (UK) c 54, ss 31(5)–(5B), as inserted by Tribunals, Courts and Enforcement Act 2007 (UK) c 15, s 141. These provisions and the equivalent rule of court they replace are discussed in Alexander Horne, ‘The Substitutionary Remedy under CPR 54.19(3): A Final Word?’ [2007] Judicial Review 135.

[93] Some commentators have attempted to counter this conclusion with the argument that the substantive legitimate expectation doctrine and other recent changes to English public law simply require the Parliament to explain its intentions with greater clarity. According to this view, the fundamental roles of and relationship between the courts and Parliament remain unchanged: see Paul Craig and Nicholas Bamforth, ‘Constitutional Analysis, Constitutional Principle and Judicial Review’ [2001] Public Law 763, 767. This argument assumes, but does not guarantee, that courts will always ultimately defer to the will of Parliament. The argument also does not translate easily to judicial review of administrative decision-making, which lies at the heart of the substantive legitimate expectation cases, because those cases rarely involve the questions of statutory interpretation that Craig and Bamford anticipate.

[94] See Craig, Administrative Law, above n 1, 4.

[95] The argument is well made in Jeffrey Jowell, ‘Parliamentary Sovereignty under the New Constitutional Hypothesis’ [2006] Public Law 562. It is still widely accepted that English courts should, to some extent, defer to the judgement of other branches of government when giving effect to the rights adopted by the Human Rights Act 1998 (UK) c 42: see, eg, Richard Clayton, ‘Principles for Judicial Deference’ [2006] 11 Judicial Review 109. There is, however, ongoing dispute about the extent to which English courts ought to defer to other branches of government, particularly the legislature: see Jeffrey Jowell, ‘Judicial Deference: Servility, Civility or Institutional Capacity?’ [2003] Public Law 592; Richard Clayton, ‘Judicial Deference and “Democratic Dialogue”: The Legitimacy of Judicial Intervention under the Human Rights Act 1998[2004] Public Law 33; Paul Craig, ‘The Courts, the Human Rights Act and Judicial Review’ (2001) 117 Law Quarterly Review 589; Lord Steyn, ‘Deference: A Tangled Story’ [2005] Public Law 346.

[96] [2001] EWCA Civ 607; [2002] 1 WLR 237, 244 (Schiemann LJ for the Court).

[97] This is a point made openly by Collins J in R (A) v Secretary of State for the Home Department [2006] EWHC 526 (Admin) (Unreported, Collins J, 22 March 2006) [29], where he stated that ‘[t]he court expects government departments and, indeed, all officials who make decisions which affect members of the public to honour statements of policy.’ Bokhary PJ reached a similar conclusion in Tung v Director of Immigration [2002] 1 HKLRD 561, 658, when he stated that ‘[t]he doctrine of legitimate expectation involves a duty owed by those who govern to those who are governed.’

[98] [2005] EWCA Civ 1363 (Unreported, Laws and Thomas LJJ and Nelson J, 22 November 2005) [68].

[99] Some commentators suggest that the substantive legitimate expectation doctrine has a stronger quality: see, eg, Jeffrey Jowell, ‘The Rule of Law and Its Underlying Values’ in Jeffrey Jowell and Dawn Oliver (eds), The Changing Constitution (6th ed, 2007) 5, 21, where Jowell describes the substantive legitimate expectation as ‘a right … to the promised benefit’. This statement endows the doctrine with a force that the cases do not.

[100] The possible educative effect that decisions about procedural fairness might have on administrative officials who are granted discretionary powers is explained in Bruce Dyer, ‘Determining the Content of Procedural Fairness’ [1993] MonashULawRw 7; (1993) 19 Monash University Law Review 165. Some commentators have suggested that the articulation of principle or guidance on judicial review decisions is as important to other judges as it is for administrative officials: see, eg, Sidney A Shapiro and Richard E Levy, ‘Judicial Incentives and Indeterminacy in Substantive Review of Administrative Decisions’ (1995) 44 Duke Law Journal 1051.

[101] John McMillan, ‘The Foundations and Limitations of Judicial Review — A Commentary’ (Paper presented at the 2002 Constitutional Law Conference and Dinner, Sydney, 15 February 2002) 4. Many English commentators have similarly accepted that the ‘rule of law’, which is often invoked as a reason for the expanding scope or intensity of judicial review in England, encompasses substantive qualities of the type McMillan suggests: see, eg, Paul Craig, ‘Formal and Substantive Conceptions of the Rule of Law: An Analytical Framework’ [1997] Public Law 467; T R S Allan, Constitutional Justice: A Liberal Theory of the Rule of Law (2001).

[102] See, eg, Tom Hickman, ‘The Substance and Structure of Proportionality’ [2008] Public Law (forthcoming). Hickman is trenchantly critical of the approach taken by Laws LJ in cases such as Nadarajah [2005] EWCA Civ 1363 (Unreported, Laws and Thomas LJJ and Nelson J, 22 November 2005).

[103] The issues are well captured in Thomas Poole, ‘The Reformation of English Administrative Law’ (on file with the author). An important element of that debate is the extent to which a doctrine of deference should qualify the developing ground of proportionality. It has been argued that proportionality may provide a more structured and substantive approach to some existing grounds of judicial review if it is recognised as a ground in its own right: Craig, Administrative Law, above n 1, 630–1. This might enable proportionality to subsume developments such as the substantive legitimate expectation doctrine, but whether that is the case remains to be seen. The doctrine of deference, like many other aspects of contemporary English public law, is the subject of much disagreement: see Steyn, ‘Deference’, above n 95.

[104] Jeremy Waldron, ‘Are Constitutional Norms Legal Norms?’ (2006) 75 Fordham Law Review 1697, 1697. A similar point is made in the context of Australian judicial review in Mark Aronson, ‘Is the ADJR Act Hampering the Development of Australian Administrative Law?’ (2005) 12 Australian Journal of Administrative Law 79, 96.

[105] See ADJR Act ss 5(2)(j), 6(2)(j); Administrative Decisions (Judicial Review) Act 1989 (ACT) ss 5(2)(i), 6(2)(i). These statutes include a ground of judicial review for an ‘exercise of a power in a way that constitutes abuse of the power.’ The ground is not included in the Judicial Review Act 1991 (Qld) or the Judicial Review Act 2000 (Tas). The Administrative Law Act 1978 (Vic) does not include such a ground because it does not codify the grounds of review.

[106] None of the various law reform reports, parliamentary debates or other materials associated with the enactment of the statutory judicial review regimes that now exist in Australia provide any support for the proposition that the statutory ground of abuse of power was intended to encompass the substantive legitimate expectation doctrine (which had not yet occurred). The various reports that gave rise to the current system of federal administrative law are reproduced in Robin Creyke and John McMillan, The Making of Commonwealth Administrative Law: The Kerr, Bland and Ellicott Committee Reports (1996). The references made in those reports to substantive grounds of review do not mention substantive unfairness, or anything that might come close to that doctrine, as a ground of review: see Commonwealth, Commonwealth Administrative Review Committee Report, Parl Paper No 144 (1971) 76–80; Commonwealth, Prerogative Writ Procedures: Report of Committee of Review, Parl Paper No 56 (1973) 3–5.

[107] See, eg, Sunshine Coast Broadcasters Ltd v Duncan (1988) 83 ALR 121, 130–2 (Pincus J); Daihatsu Australia Pty Ltd v Deputy Commissioner of Taxation (2000) 182 ALR 239, 253–61 (Lehane J).

[108] ADJR Act ss 5(1)(j), 6(1)(j); Administrative Decisions (Judicial Review) Act 1989 (ACT) ss 5(1)(i), 6(1)(i); Judicial Review Act 1991 (Qld) ss 20(2)(i), 21(2)(i); Judicial Review Act 2000 (Tas) ss 17(2)(i), 18(2)(i).

[109] Sir William Wade, Constitutional Fundamentals (revised ed, 1989) 90.

[110] This is an issue examined in Aronson, ‘Is the ADJR Act Hampering the Development of Australian Administrative Law?’, above n 104.

[111] See, eg, Daihatsu Australia Pty Ltd v Deputy Commissioner of Taxation (2000) 182 ALR 239, 255–9, where Lehane J discussed the English cases and noted that they had been received cautiously in Australia; Rush v Commissioner of Police [2006] FCA 12; (2006) 150 FCR 165, 185–7, where Finn J flatly rejected the doctrine; Sidhu v Minister for Immigration and Multicultural and Indigenous Affairs [2007] FCA 69 (Unreported, Lander J, 9 February 2007) [126], where Lander J cited the rejection of the doctrine as being part of current Australian law in Rush v Commissioner of Police [2006] FCA 12; (2006) 150 FCR 165.

[112] (2003) 214 CLR 1.

[113] Haoucher [1990] HCA 22; (1990) 169 CLR 648, 652 (Deane J). See also the comments of McHugh J: at 683.

[114] (1990) 170 CLR 1.

[115] Macrae v A-G (NSW) (1987) 9 NSWLR 268, 283 (Kirby P), 309 (Priestley JA).

[116] Quin (1990) 170 CLR 1, 17 (Mason CJ), 33 (Brennan J), 60 (Dawson J).

[117] Ibid 24 (Mason CJ), 41 (Brennan J), 60 (Dawson J).

[118] Ibid 21.

[119] Ibid 47–9 (Deane J), 68–9 (Toohey J). Each justice expressly anticipated the possibility that the government might decline to appoint Mr Quin but made clear that this could only occur after he was informed of, and given the right to respond to, any reasons why he was thought unsuitable for appointment. This reasoning invoked natural justice, which required Mr Quin’s application to be determined according to the policy that he was led to believe would apply to him, but also acceptance that the fact that he might not be appointed effectively modified the policy to enable the government to take into account the issues that were alleged to have motivated the deferral of Mr Quin’s application.

[120] See, eg, Independent Commission against Corruption v Chaffey (1993) 30 NSWLR 21, 28–30 (Gleeson CJ), 60–1 (Mahoney JA), where it was conceded that the requirements of procedural fairness could not cure the unfairness arising from the adverse publicity of an open hearing. Both judges held that the decision on whether to hold public or private hearings was the province of the body granted power to conduct the hearing.

[121] [1995] HCA 20; (1995) 183 CLR 273.

[122] Ibid 291. Gaudron J agreed with Mason CJ and Deane J on the status of the treaty in Australian law, but placed greater emphasis on the citizenship rights of the children who stood to be affected by the decision in question: at 304. Toohey J adopted similar reasoning to Mason CJ and Deane J: at 299.

[123] Ibid 305–6, 313–15. His Honour’s objections were essentially twofold. First, his Honour rejected the finding that ratification of a treaty could be interpreted as a ‘positive statement’ by the executive to the Australian people upon which a legitimate expectation could be formed. Secondly, his Honour did not accept that a legitimate expectation could exist where the person affected did not know of the source of the expectation. In other words, Mr Teoh did not know of the treaty and so he could not hold any expectation about its application.

[124] Ibid 314.

[125] (2003) 214 CLR 1.

[126] The case also attracted considerable attention because several members of the Court strongly doubted the proposition that ratification of a treaty could itself support a legitimate expectation that the executive and its officers would act in conformity with the treaty: ibid 28–34 (McHugh and Gummow JJ), 38–9 (Hayne J), 45–7 (Callinan J). The wider implications of the High Court’s apparent retreat from this crucial aspect of Teoh are discussed in Wendy Lacey, ‘A Prelude to the Demise of Teoh: The High Court Decision in Re Minister for Immigration and Multicultural Affairs; Ex parte Lam[2004] SydLawRw 7; (2004) 26 Sydney Law Review 131; Alison Duxbury, ‘The Impact and Significance of Teoh and Lam’ in Matthew Groves and H P Lee (eds), Australian Administrative Law: Fundamentals, Principles and Doctrines (2007) 299; Matthew Groves, ‘Is Teoh’s Case Still Good Law?’ (2007) 14 Australian Journal of Administrative Law 126.

[127] Lam (2003) 214 CLR 1, 16, 21.

[128] Ibid 36–9 (Hayne J), 46–9 (Callinan J).

[129] The doctrine of legalism is commonly associated with Sir Owen Dixon, and the terminology of formalism is often attributed to the High Court under the recent leadership of Gleeson CJ. An overview of the issues in this area is provided in Jeffrey Goldsworthy, ‘Australia: Devotion to Legalism’ in Jeffrey Goldsworthy (ed), Interpreting Constitutions: A Comparative Study (2006) 106.

[130] A good discussion of this is in Justice J D Heydon, ‘Judicial Activism and the Death of the Rule of Law’ (2003) 23 Australian Bar Review 110. Heydon is especially critical of the use of international law and norms by the High Court during the leadership of Sir Anthony Mason: at 131.

[131] The key issues are explained in Sir Anthony Mason, ‘Legislative and Judicial Law-Making: Can We Locate an Identifiable Boundary?’ [2003] AdelLawRw 4; (2003) 24 Adelaide Law Review 15; Sir Anthony Mason, ‘The Centenary of the High Court of Australia’ (2003) 5 Constitutional Law and Policy Review 41.

[132] The wider issues surrounding the formalism that Lam and other cases exemplify are discussed in Thomas Poole, ‘Between the Devil and the Deep Blue Sea: Administrative Law in an Age of Rights’ in Carol Harlow and Linda Pearson (eds), Administrative Law in a Changing State: Essays in Honour of Mark Aronson (2008).

[133] Aronson, Dyer and Groves, above n 2, 359–64. However, there is some support for the principle that governmental agencies may be estopped from departing from representations they have made regarding procedures to be followed by them: see generally Enid Campbell, ‘Waiver by Agencies of Government of Statutory Procedural Requirements’ [1998] UNSWLawJl 62; (1998) 21 University of New South Wales Law Journal 711.

[134] (1990) 21 FCR 193.

[135] This power was at the time granted by s 12 of the Migration Act 1958 (Cth), but is now granted by s 200 of the same Act.

[136] Kurtovic (1990) 21 FCR 193, 196 (Neaves J), 201 (Ryan J), 208–14 (Gummow J). The decision was upheld on the sole basis that Kurtovic was denied natural justice by the failure to provide notice of and a chance to respond to the information from parole authorities: at 197 (Neaves J), 205 (Ryan J), 222–4 (Gummow J).

[137] Ibid 196 (Neaves J), 201 (Ryan J), 207–8 (Gummow J).

[138] Ibid 210.

[139] See Sir Anthony Mason, ‘The Place of Estoppel in Public Law’ in Matthew Groves (ed), Law and Government in Australia (2005) 160. Mason does not entirely exclude the possibility that estoppel might form a part of Australian public law. He suggests that some form of estoppel against public authorities might have been possible during the period that the High Court appeared to be establishing a unified principle of estoppel: at 160. He concludes that estoppel might in theory be possible against a public authority, but concedes that the circumstances in which this might occur are relatively limited: at 182–3.

[140] Kurtovic (1990) 21 FCR 193, 220–1.

[141] [1976] EWCA Civ 10; [1977] QB 643, 707.

[142] Ibid.

[143] Kurtovic (1990) 21 FCR 193, 221.

[144] Ibid 220.

[145] Ibid 221.

[146] It could be argued that a plea of estoppel does not necessarily involve the weighing of different factors, but Gummow J clearly thought otherwise. The quote extracted above indicates that his Honour believed that any use of estoppel in public law would require the court to balance the particular interests of a person who claimed an estoppel against the different claims of the public official who sought to argue against the estoppel. Any deliberation of those issues seems to involve a weighing of competing factors, which appears similar to the process embraced in Coughlan.

[147] Quin (1990) 170 CLR 1, 18. His Honour also said that ‘[i]t is possible perhaps that there may be some cases in which substantive protection can be afforded and ordered by the court, without detriment to the public interest intended to be served by the exercise of the relevant statutory or prerogative power’: at 23.

[148] Gummow J dealt with this possibility in Kurtovic (1990) 21 FCR 193, 220–2. See also Daihatsu Australia Pty Ltd v Deputy Commissioner of Taxation (2000) 182 ALR 239, 257–9 (Lehane J).

[149] (1990) 170 CLR 1, 60. Dawson J adopted a similar stance in Haoucher [1990] HCA 22; (1990) 169 CLR 648, 663.

[150] [1998] FCA 615; (1998) 84 FCR 154, 167 (Hill, Sundberg and Goldberg JJ).

[151] Ibid. See also Sunshine Coast Broadcasters Ltd v Duncan (1988) 83 ALR 121, 130 (Pincus J). ‘Discrimination’ in this sense does not mean one of the grounds of discrimination prohibited by equal opportunity or human rights legislation.

[152] The classic case is Kruse v Johnson [1898] UKLawRpKQB 101; [1898] 2 QB 91, 99–100 (Lord Russell). More recent Australian decisions in which unequal or inconsistent treatment might provide a basis for relief include Sunshine Coast Broadcasters Ltd v Duncan (1988) 83 ALR 121; Hamilton v Minister for Immigration, Local Government and Ethnic Affairs [1993] FCA 657; (1993) 48 FCR 20, 36–7 (Beazley J); New South Wales Aboriginal Land Council v Aboriginal and Torres Strait Islander Commission (1995) 59 FCR 369; Daihatsu Australia Pty Ltd v Deputy Commissioner of Taxation (2000) 182 ALR 239, 255–61 (Lehane J); Dilatte v MacTiernan [2002] WASCA 100 (Unreported, Malcolm CJ, Wallwork J and White AUJ, 1 May 2002). There are also some English cases to the same effect: Boddington v British Transport Police [1998] UKHL 13; [1999] 2 AC 143, 169–70 (Lord Steyn); R v Secretary of State for the Home Department; Ex parte Zeqiri [2002] UKHL 3 (Unreported, Lords Slynn, Mackay, Hoffmann, Millett and Rodger, 24 January 2002) [56] (Lord Hoffmann). In the Australian cases, the precise basis for the grant of relief in such cases is not clear. Aronson, Dyer and Groves, above n 2, 348–50 note that some cases regard unequal treatment as an ‘abuse of power’ within the meaning of s 5(2)(j) of the ADJR Act. In my view, it is arguable in some cases that natural justice has been denied to an applicant because the decision-maker failed to ‘respond to a substantial, clearly articulated argument relying upon established facts’ in the sense required by Dranichnikov v Minister for Immigration and Multicultural and Indigenous Affairs (2003) 197 ALR 389, 394 (Gummow and Callinan JJ). This argument would be possible if a decision-maker did not explain why they did not follow those other similar cases.

[153] See, eg, SGBB v Minister for Immigration and Multicultural and Indigenous Affairs [2003] FCA 709; (2003) 199 ALR 364, 372, where Selway J held that the RRT was required to exercise an independent judgment in each case, which precludes any obligation on the part of the RRT to follow its previous decisions even in like situations. See also NARI v Minister for Immigration and Multicultural and Indigenous Affairs [2005] FCA 186 (Unreported, Bennett J, 2 June 2005) [16]–[17], [58]. A similar conclusion was reached in NARY v Minister for Immigration and Multicultural and Indigenous Affairs [2003] FCA 1255 (Unreported, Moore J, 6 November 2003) [10], where Moore J held that the RRT was required to consider and rely upon similar previous decisions only in exceptional instances. The decision of Moore J was upheld on other grounds: Applicants S311 of 2002 v Minister for Immigration and Multicultural and Indigenous Affairs [2004] FCA 45 (Unreported, Madgwick J, 3 February 2004).

[154] Support for this proposition can be taken from E v Secretary of State for the Home Department [2004] EWCA Civ 49; [2004] QB 1044, 1078, where Carnwath LJ reasoned that ‘not all (or even most) Court of Appeal decisions in [refugee law] should be seen as laying down propositions of law; the decisions in this area are unusually fact-sensitive’. It is possible that the Court of Appeal may be favoured such a position in order to head off the possibility that disappointed applicants for refugee status might try to seek review or appeal of their case on the basis that another applicant in an apparently similar position was successful. If challenges of this nature were possible, the number of cases before the courts could increase greatly.

[155] [2005] EWCA Civ 744 (Unreported, Pill, May and Dyson LJJ, 16 June 2005). For a discussion of the case, see above nn 71–6 and accompanying text.

[156] There were many earlier decisions, but the turning point in recent jurisprudence of the High Court occurred at the end of the last decade with Abebe v Commonwealth [1999] HCA 14; (1999) 197 CLR 510; Minister for Immigration and Multicultural Affairs v Eshetu [1999] HCA 21; (1999) 197 CLR 611; Re Refugee Review Tribunal; Ex parte Aala [2000] HCA 57; (2000) 204 CLR 82: see D F Jackson, ‘Development of Judicial Review in Australia over the Last 10 Years: The Growth of the Constitutional Writs’ (2004) 12 Australian Journal of Administrative Law 22.

[157] (1990) 170 CLR 1.

[158] 5 US [1803] USSC 16; (1 Cranch) 137, 177 (Marshall CJ) (1803).

[159] Quin (1990) 170 CLR 1, 35–6. A similar statement was made by Gleeson CJ in NEAT Domestic Trading Pty Ltd v AWB Ltd [2003] HCA 35; (2003) 216 CLR 277, 288, where the Chief Justice explained that ‘[j]udicial review is not an invitation to judges to decide what they would consider fair or reasonable if they were given the function’ that is subject to judicial review.

[160] Quin (1990) 170 CLR 1, 36.

[161] Ibid 37.

[162] Ibid. Marshall CJ made similar remarks in Marbury v Madison 5 US [1803] USSC 16; (1 Cranch) 137, 169–70 (1803).

[163] [2000] HCA 5; (2000) 199 CLR 135.

[164] Ibid 152–3 (Gleeson CJ, Gummow, Kirby and Hayne JJ). This is the so-called ‘Chevron doctrine’, taken from Chevron USA Inc v Natural Resources Defense Council Inc[1984] USSC 140; , 467 US 837 (1984). The doctrine is explained in Richard J Pierce Jr, Administrative Law Treatise (4th ed, 2002) 139–91. It has long been argued that the Chevron doctrine runs counter to the principles of Marbury v Madison, but one influential commentator has recently suggested that this may be desirable: see Cass R Sunstein, ‘Beyond Marbury: The Executive’s Power to Say What the Law Is’ [2006] YaleLawJl 14; (2006) 115 Yale Law Journal 2580. When compared with the Australian conception of the separation of powers doctrine, which seems to direct attention to the separation of judicial power, the American version focuses more on the division of power between the executive and Congress: see Harold H Bruff, ‘The Incompatibility Principle’ (2007) 59 Administrative Law Review 225.

[165] Enfield [2000] HCA 5; (2000) 199 CLR 135, 152–3 (Gleeson CJ, Gummow, Kirby and Hayne JJ), citing Quin (1990) 170 CLR 1, 35–6 (Brennan J).

[166] (2003) 214 CLR 1.

[167] The decision of the High Court does not explain why this was not done, though Gleeson CJ suggested that departmental officials may have reviewed Lam’s first letter and annexed materials and decided that there was no need to contact the carer: ibid 6–7.

[168] A related argument was that the failure of the department to contact the carer meant that a relevant and primary consideration (the best interests of the children) was not properly taken into account: ibid 4–5 (Gleeson CJ).

[169] A point conceded by Gleeson CJ: ibid 9–10.

[170] Hayne J simply noted that neither Coughlan nor its reasoning was placed in issue by the parties, so his Honour expressly declined to consider these issues: ibid 37.

[171] Lam (2003) 214 CLR 1, 10.

[172] Ibid 12. See also A-G (NSW) v World Best Holdings Ltd (2005) 63 NSWLR 557, 586 (Mason P). Mason P was ‘troubled’ by the invocation by Spigelman CJ of considerations of ‘unfairness’ or a ‘scale of unfairness or injustice’ in construing a statute. Mason P concluded that the courts had ‘no mandate to construe legislation by reference to perceptions of morality that are not already firmly embedded in fundamental common law doctrines or the statute itself.’

[173] Callinan J agreed with McHugh and Gummow JJ that the legitimate expectation doctrine could ‘on no view … give rise to substantive rights rather than to procedural rights’: Lam (2003) 214 CLR 1, 48. His Honour delivered a separate judgment addressing other issues, notably the principles that might be derived from Teoh and the role of the procedural legitimate expectation doctrine more generally.

[174] Ibid 23.

[175] Ibid.

[176] Ibid 24–5. Their Honours also suggested that much of the reasoning in Coughlan appeared to be directed to the English assimilation of European public law values: ibid 23–4. That assessment implies that the constitutional structures of England and Australia are increasingly moving in different directions.

[177] Ibid 10, 24–5. See also at 34.

[178] Sir Anthony Mason, ‘Procedural Fairness: Its Development and Continuing Role of Legitimate Expectation’ (2005) 12 Australian Journal of Administrative Law 103, 109.

[179] Lam (2003) 214 CLR 1, 25, citing Bradley Selway, ‘The Principle behind Common Law Judicial Review of Administrative Action — The Search Continues’ (2002) 30 Federal Law Review 217, 234.

[180] This problem is regularly acknowledged by judges even as they defend the doctrine: see, eg, Plaintiff S157/2002 v Commonwealth [2003] HCA 2; (2003) 211 CLR 476, 485 (Gleeson CJ); Re Refugee Review Tribunal; Ex parte Aala [2000] HCA 57; (2000) 204 CLR 82, 141 (Hayne J).

[181] A point frankly conceded in Minister for Immigration and Multicultural Affairs v Yusuf (2001) 206 CLR 323, 351 (McHugh, Gummow and Hayne JJ).

[182] See Benjamin O’Donnell, ‘Jurisdictional Error, Invalidity and the Role of Injunction in s 75(v) of the Australian Constitution(2007) 28 Australian Bar Review 291.

[183] See Plaintiff S157/2002 v Commonwealth [2003] HCA 2; (2003) 211 CLR 476. The precise reasoning is more complex: see Enid Campbell and Matthew Groves, ‘Privative Clauses and the Australian Constitution(2004) 4 Oxford University Commonwealth Law Journal 51; Caron Beaton-Wells, ‘Restoring the Rule of Law — Plaintiff S157/2002 v Commonwealth of Australia(2003) 10 Australian Journal of Administrative Law 125.

[184] Sir Anthony Mason, ‘Procedural Fairness’, above n 178, 109.

[185] Ibid.

[186] [2003] HCA 30; (2003) 198 ALR 59, 98.

[187] Ibid 92.

[188] See the text accompanying above nn 6981.

[189] Cases in which Kirby J has criticised the jurisdictional error doctrine include: Re Minister for Immigration and Multicultural Affairs; Ex parte Miah (2001) 206 CLR 57, 123; Re McBain; Ex parte Australian Catholic Bishops Conference [2002] HCA 16; (2002) 209 CLR 372, 439–40; Re Minister for Immigration and Multicultural Affairs; Ex parte Applicant S20/2002 [2003] HCA 30; (2003) 198 ALR 59, 85–6.

[190] Re Minister for Immigration and Multicultural Affairs; Ex parte Applicant S20/2002 [2003] HCA 30; (2003) 198 ALR 59, 96.

[191] Ibid.

[192] Kable v DPP (NSW) [1996] HCA 24; (1996) 189 CLR 51 (‘Kable’). The implications of the case are explained in H P Lee, ‘The Kable Case: A Guard-Dog That Barked But Once?’ in George Winterton (ed), State Constitutional Landmarks (2006) 390; Fiona Wheeler, ‘The Kable Doctrine and State Legislative Power over State Courts’ (2005) 20(2) Australasian Parliamentary Review 15; Peter Johnston and Rohan Hardcastle, ‘State Courts: The Limits of Kable[1998] SydLawRw 10; (1998) 20 Sydney Law Review 216.

[193] It was successfully invoked in Kable [1996] HCA 24; (1996) 189 CLR 51.

[194] See, eg, Kruger v Commonwealth [1997] HCA 27; (1997) 190 CLR 1, 175 (Gummow J). Zines argues that s 73 of the Australian Constitution ‘provides the unifying element of our judicial system’ because it provides an important means by which to secure conformity between state and Commonwealth law: Leslie Zines, Cowen and Zines’s Federal Jurisdiction in Australia (3rd ed, 2002) 182. A similar approach is implicit in the reasoning of Selway, above n 179, 229–37. Selway identifies the Australian Constitution as the ultimate foundation for judicial review throughout Australia and assumes that no significant distinctions can arise between state and Commonwealth law.

[195] (1990) 170 CLR 1.

[196] Ibid 37. See also at 35.

[197] The history and implications of the Victorian Charter of Rights are explained in Simon Evans and Carolyn Evans, ‘Legal Redress under the Victorian Charter of Human Rights and Responsibilities(2006) 17 Public Law Review 264; Priyanga Hettiarachi, ‘Some Things Borrowed, Some Things New: An Overview of Judicial Review of Legislation under the Charter of Human Rights and Responsibilities(2007) 7 Oxford University Commonwealth Law Journal 61.

[198] The nature and scope of the fettering principle in English law is explained in Chris Hilson, ‘Judicial Review, Policies and the Fettering of Discretion’ [2002] Public Law 111.

[199] The distinction between top-down and bottom-up reasoning in Australian judicial review is discussed in Stephen Gageler, ‘The Underpinnings of Judicial Review of Administrative Action: Common Law or Constitution?’ (2000) 28 Federal Law Review 303. The distinction between top-down and bottom-up reasoning has largely been used in constitutional law: see, eg, Justice Keith Mason, ‘What Is Wrong with Top-Down Legal Reasoning?’ (2004) 78 Australian Law Journal 574.

[200] See Dawn Oliver, ‘Is the Ultra Vires Rule the Basis of Judicial Review?’ [1987] Public Law 543; Sir John Laws, ‘Law and Democracy’ [1995] Public Law 72; Paul Craig, ‘Ultra Vires and the Foundations of Judicial Review’ (1998) 57 Cambridge Law Journal 63; David Dyzenhaus, ‘Reuniting the Brain: The Democratic Basis of Judicial Review’ (1998) 9 Public Law Review 98; P Craig, ‘Competing Models of Judicial Review’ [1999] Public Law 428; Jeffrey Jowell, ‘Of Vires and Vacuums: The Constitutional Context of Judicial Review’ [1999] Public Law 448; Paul Craig, ‘Public Law, Political Theory and Legal Theory’ [2000] Public Law 211; Susan Kneebone, ‘What Is the Basis of Judicial Review?’ (2001) 12 Public Law Review 95.

[201] See Christopher Forsyth, ‘Of Fig Leaves and Fairy Tales: The Ultra Vires Doctrine, the Sovereignty of Parliament and Judicial Review’ (1996) 55 Cambridge Law Journal 122; Mark Elliott, ‘The Demise of Parliamentary Sovereignty? The Implications for Justifying Judicial Review’ (1999) 115 Law Quarterly Review 119; Mark Elliott, ‘The Ultra Vires Doctrine in a Constitutional Setting: Still the Central Principle of Administrative Law’ (1999) 58 Cambridge Law Journal 129; Sir Gerard Brennan, ‘The Purpose and Scope of Judicial Review’ in Michael Taggart (ed), Judicial Review of Administrative Action in the 1980s: Problems and Prospects (1986) 19.

[202] The arguments, which are far more complex than the brief statements here suggest, are usefully collected in Christopher Forsyth (ed), Judicial Review and the Constitution (2000).

[203] The most satisfactory is the ‘modified ultra vires theory’, by which Parliament is taken to have impliedly accepted the principles of judicial review. This approach strikes a balance between judicial innovation and parliamentary sovereignty by assuming that the latter can (and has) been exercised to approve the former and that legislation is passed with knowledge of previous judicial decisions: see Christopher Forsyth and Mark Elliott, ‘The Legitimacy of Judicial Review’ [2003] Public Law 286, 287.

[204] See, eg, Bruce Dyer, ‘Legitimate Expectations in Procedural Fairness after Lam’ in Matthew Groves (ed), Law and Government in Australia (2005) 184, 197–200, where the approaches to the legitimate expectation doctrine taken by Mason and Brennan JJ in Kioa [1985] HCA 81; (1985) 159 CLR 550 are distinguished in terms similar to the English theories of judicial review.

[205] See, eg, Stephen Gageler, ‘Legitimate Expectation: Comment on the Article by the Hon Sir Anthony Mason AC KBE’ (2005) 12 Australian Journal of Administrative Law 111, 114, where Gageler concludes that Lam represents a triumph of the ultra vires theory.

[206] See Selway, above n 179.

[207] Sir Anthony Mason, ‘Procedural Fairness’, above n 178, 109.

[208] The phrase comes from Sir Owen Dixon, Jesting Pilate: And Other Papers and Addresses (1965) 247.

[209] See, eg, J J Spigelman, ‘The Integrity Branch of Government’ (2004) 78 Australian Law Journal 724, 733; Selway, above n 179, 229–31.

[210] Selway, above n 179, 235.

[211] Gageler, ‘The Underpinnings of Judicial Review of Administrative Action’, above n 199, 307.

[212] The points made here about the ADJR Act apply equally to the following similar statutes: Administrative Decisions (Judicial Review) Act 1989 (ACT); Judicial Review Act 1991 (Qld); Judicial Review Act 2000 (Tas). Western Australia is also considering adopting the ADJR Act model: see Law Reform Commission of Western Australia, Judicial Review of Administrative Decisions, Report No 95 (2002).

[213] The ADJR Act simplified the test for standing, introduced a right to reasons for the decisions to which it applied and streamlined the remedies available for judicial review. While each of these matters is largely procedural, there is no doubt that they greatly simplified the procedure for judicial review and therefore widened access to judicial review. The history of the Act and associated reforms to administrative law are explained in Robin Creyke and John McMillan (eds), The Kerr Vision of Australian Administrative Law At the Twenty-Five Year Mark (1998).

[214] Re Minister for Immigration and Multicultural Affairs; Ex parte Applicant S20/2002 [2003] HCA 30; (2003) 198 ALR 59, 97.

[215] Ibid 94.

[216] [1985] HCA 81; (1985) 159 CLR 550. The ground considered was denial of natural justice, available under ADJR Act ss 5(1)(a), 6(1)(a).

[217] The High Court has adopted a similar position in its exposition of the grounds of relevancy and unreasonableness, concluding that the ADJR Act was ‘substantially declaratory of the common law’: Minister for Aboriginal Affairs v Peko-Wallsend Ltd (1986) 162 CLR 24, 39 (Mason J). See also generally at 39–42 (Mason J).

[218] The grounds are respectively ss 5(1)(j) and 5(2)(j) of the ADJR Act.

[219] It has been suggested that the inclusion of these grounds acknowledges ‘the common law’s capacity to develop new grounds’ of review: see Aronson, Dyer and Groves, above n 2, 114. However, those authors do not consider whether the statutory grounds will ever facilitate this possibility.

[220] Aronson, ‘Is the ADJR Act Hampering the Development of Australian Administrative Law?’, above n 104, 94.

[221] Ibid 95–6.

[222] Ibid 96.

[223] Constitution of the Republic of South Africa Act 1996 (RSA) s 33(1) (‘South African Constitution’).

[224] The effect of these reforms is explained in Cora Hoexter, ‘The Future of Judicial Review in South African Administrative Law’ (2000) 117 South African Law Journal 484.

[225] Pharmaceutical Manufacturers Association of South Africa; Re ex parte President of the Republic of South Africa [2002] 2 SA 674, 695–8 (Chaskalson P). The effect of this judgment and other decisions is explained by another member of the Constitutional Court of South Africa in Kate O’Regan, ‘Breaking Ground: Some Thoughts on the Seismic Shift in Our Administrative Law’ (2004) 121 South African Law Journal 424.

[226] See Human Rights Act 1998 (UK) c 42; New Zealand Bill of Rights Act 1990 (NZ).

[227] Public statements of the federal Attorney-General to date suggest that this is the model that will be favoured: see, eg, Jonathan Pearlman, ‘Do-It-Yourself Charter to Right Future Wrongs’, The Sydney Morning Herald (Sydney), 1 December 2007, 26.

[228] The issues are considered in the articles mentioned in above n 95.

[229] Proportionality has not been accepted as a freestanding ground of review in Australia: see, eg, Bruce v Cole [1998] NSWCA 45; (1998) 45 NSWLR 163, 185 (Spigelman CJ). See generally Geoff Airo-Farulla, ‘Reasonableness, Rationality and Proportionality’ in Matthew Groves and H P Lee (eds), Australian Administrative Law: Fundamentals, Principles and Doctrines (2007) 212.

[230] The key difference is that, in the absence of a constitutionally entrenched separation of powers, it is possible for English commentators to suggest that there should be little, if any, deference shown by the courts to administrative decision-making. The strongest exponent of this view is T R S Allan, ‘Human Rights and Judicial Review: A Critique of “Due Deference”’ (2006) 65 Cambridge Law Journal 671.

[231] It is for this reason that English decisions must be approached with particular care. Craig concludes that the traditional dichotomy of review/merits is not helpful because the cases in which English courts have incorporated an element of substantive review ‘entail the judiciary in taking some view of the merits of the contested action’: Craig, Administrative Law, above n 1, 589 (citations omitted). This conclusion suggests that many, if not all, English cases adopt an approach that cannot easily sit with Australia’s constitutional arrangements.

[232] See generally Jim South, ‘The Campaign for a National Bill of Rights: Would “Declarations of Incompatibility” Be Compatible with the Constitution?’ (2007) 10 Constitutional Law and Policy Review 2.

[233] On this point, it is difficult to disagree with Adrienne Stone, who concludes that ‘a bill of rights would vastly expand the range of circumstances in which judges decide (and perhaps have the final word on) highly complex and controversial issues. That is a significant change which requires critical analysis’: Adrienne Stone, ‘Disagreement and an Australian Bill of Rights’ [2002] MelbULawRw 25; (2002) 26 Melbourne University Law Review 478, 496.

[234] Judges have stressed that the unreasonableness ground of judicial review does not allow for review of administrative decisions on their merits, and that the cases in which that ground can be established will be rare. On the qualitative elements which may enter into judicial consideration of whether the unreasonableness ground of judicial review has been established, see Aronson, Dyer and Groves, above n 2, 334–43; Mark Aronson, ‘Unreasonableness and Error of Law’ [2001] UNSWLawJl 26; (2001) 24 University of New South Wales Law Journal 315; Geoff Airo-Farulla, ‘Rationality and Judicial Review of Administrative Action’ [2000] MelbULawRw 23; (2000) 24 Melbourne University Law Review 543.

[235] Murray Gleeson, ‘Judicial Legitimacy’ (2000) 20 Australian Bar Review 4, 11.