‘Good administration’ and the ‘Good’: The normative foundation for the protection of legitimate expectations

Abstract

An idea that has gained significant traction in both case law and academic commentary as a justification for the protection of legitimate expectations is the concept of ‘good administration’. Going beyond the usual criticisms of the concept’s ambiguity, this article aims to highlight an additional set of difficulties with the invocation of ‘good administration’ as the normative justification for the doctrine. This article’s central argument is that the concept of ‘good’ invoked by the idea of ‘good administration’ inevitably falls to be substantiated by a particular conception of what the ‘good’ requires as a matter of political philosophy. And given that there are multiple competing conceptions of what ‘good’ law and government are, this magnifies the challenges of coming to a landing on the precise content of ‘good administration’. This article will illustrate that the various formulations of the normative foundation of the doctrine track closely with four different conceptions of ‘good’ law and government and will explore the implications of this diagnosis for the formulation of the proper justification for the protection of legitimate expectations.

Keywords

public law good administration Hong Kong Singapore legitimate expectations

Introduction

The idea that the promises, policies or practices of public authorities can generate legitimate expectations on the part of affected applicants who ought to have remedial recourse through judicial review if the authority subsequently violates these expectations is an accepted feature of several common law jurisdictions around the world. Indeed, in the United Kingdom and Hong Kong, for example, courts have illustrated a willingness to protect such expectations even if they extend beyond the proper procedure an applicant is entitled to and include expectations of substantive benefits.

Yet, the normative foundation of the legitimate expectations doctrine, especially in its substantive form, remains unsettled. Judges and commentators have offered various possibilities in this regard. One of the most commonly invoked justifications for the doctrine—indeed, invoked most recently by the UK Supreme Court itself—is that the doctrine of legitimate expectations ‘is underpinned by the requirements of good administration’.¹ However, this justification for the doctrine has been subject to sharp critique, including, inter alia, the fact that the concept of ‘good administration’ is so broad that it provides scant principled guidance as to when the doctrine should be invoked.

Going beyond these criticisms, this article aims to highlight an additional set of difficulties with the invocation of ‘good administration’ as the normative justification for the protection of legitimate expectations. This article’s central argument is that one’s conception of what ‘good administration’ entails is inextricably linked to one’s conception of the ‘good’ as a matter of law and government. In other words, the concept of ‘good’ invoked by the idea of ‘good administration’ will fall to be substantiated by a particular conception of what the ‘good’ requires as a matter of political philosophy. And given that there are several competing conceptions of what ‘good’ law and government are, this magnifies the challenges of coming to a landing on the precise content of ‘good administration’ as a normative justification for the doctrine. A bare invocation of ‘good administration’ does not do justice to these layers of contention and will consequently be unhelpfully vague. Indeed, cognisance of this additional level of disagreement in relation to the idea of ‘good administration’, insofar as it has been frequently invoked as an important normative justification for the doctrine, serves to explain at least in part the protracted disagreement over the proper normative justification for the protection of substantive legitimate expectations, in particular.

It is worth noting that the central argument of this article, directed as it is at highlighting a conceptual difficulty with substantiating the concept of ‘good administration’, is not necessarily limited to the doctrine of legitimate expectations. Indeed, to the extent that the entire domain of administrative law is said to be justified on the basis of ‘good administration’, the central argument in this article is capable of broader application across the entire realm of administrative law.² This article aims, however, to focus on the law of legitimate expectations as a useful lens through which one can study the difficulties surrounding the concept of ‘good administration’, with special attention to the protection of substantive legitimate expectations. Indeed, the doctrine of substantive legitimate expectations is at present characterised by significant normative ambiguity, and it is in this specific legal context that the concept of ‘good administration’ has been often discussed, making the law in this regard a particularly useful resource for analysis of the concept of ‘good administration’. Cases relating to procedural legitimate expectations will also be drawn upon where relevant. The concept of ‘good administration’ has also been referenced in relation to procedural legitimate expectations, and the distinction between the protection of procedural and substantive legitimate expectations can often be hard to draw in practice—the UK Supreme Court decision in In the matter of an application by Geraldine Finucane for Judicial Review (Northern Ireland) is an example in this regard.³

Accordingly, this article will proceed in the following parts. First, the article will describe the doctrine of legitimate expectations. Second, it will outline the issues that have been raised in case law and academic commentary with respect to the idea of ‘good administration’ as the core normative justification for the doctrine. Third, this article will study the doctrine of legitimate expectations as it has been discussed in several common law jurisdictions to reveal that the various formulations in judicial discussions and academic commentary as to the proper normative foundation of the doctrine track closely with at least four different conceptions of the ‘good’ in relation to law and government. The article will go on to explore the implications of this diagnosis for the formulation of the proper justification of the doctrine.

As a matter of methodology, the article will draw upon discussions of the doctrine in several common law jurisdictions—specifically, the United Kingdom, Hong Kong, Singapore and Australia. While the precise normative basis underlying the protection of legitimate expectations may indeed vary across these jurisdictions, and may be shaped by the unique constitutional or cultural context of these jurisdictions, these jurisdictions have grappled with the same question of the doctrine’s normative foundation as a shared concern. Indeed, the various interpretations of the doctrine that have been proffered across these jurisdictions to this question helpfully illustrate the breadth of the potential normative justifications for the protection of legitimate expectations.

The doctrine of legitimate expectations

Judges and commentators have suggested that several considerations can be distinguished in the application of the doctrine of legitimate expectations:⁴ first, when a legitimate expectation arises; second, when it will be unlawful for a public body to frustrate such an expectation; and third, what the proper judicial remedy should be for the unlawful frustration of a legitimate expectation. Each consideration will be discussed in turn.

In relation to the first consideration, a legitimate expectation must be grounded upon a representation by a public authority which is sufficiently clear and unequivocal.⁵ Jason Varuhas has argued that the ‘paradigm case’ for the protection of legitimate expectations is

the case where an authority makes an express promise, assurance or undertaking to an individual or group of individuals that the authority will act or omit to act in some way, this act or omission having a bearing on the individual or group’s interests.⁶

Case law, however, suggests that such representations can also be made through policies describing how a power will be exercised or consistent past practices in the exercise of the power.⁷ Legitimate expectations generated by such representations are not limited to expectations as to the procedure a decision-making body will follow in the exercise of its power—they extend also to the substantive decisions of the decision-making body.⁸ There is generally no need for the applicant to demonstrate detrimental reliance upon the representation, but such reliance will be a strong indicator in favour of the finding that there was indeed a legitimate expectation.⁹

If a legitimate expectation has indeed arisen, the analysis will move to the second consideration: the court will consider whether there is an overriding public interest which justifies the frustration of the expectation.¹⁰ This will require a context-sensitive determination resting on the nature of the legitimate expectation and its relation to broader public interest considerations.¹¹ Should the court decide that there are indeed no overriding public interest considerations, the court will move to the third consideration—the proper remedy that should be accorded to the applicant. Scholars have argued that the proper protection to be accorded to a legitimate expectation should depend on the nature of the expectation, to strike a good balance between safeguarding the applicant’s interest and the public interest in preserving the public authority’s discretion.¹² The courts have sometimes required the public authority to deliver the substantive content of its representation to the affected individual(s).¹³ However, substantive fulfilment of the legitimate expectation is not the only remedial option.¹⁴ The courts have other times found that the appropriate remedy would be to require the decision-making authority to take the legitimate expectation into account in decision-making or to consult the affected parties before departing from the representation.¹⁵

Framed as such, it will be readily apparent that the application of the doctrine can lead a court to come very close to the merits of an administrative decision—especially if the doctrine is taken to require the enforcement of substantive legitimate expectations.¹⁶ This feature of the doctrine has attracted some criticism.¹⁷ For example, the Singapore Court of Appeal, in expressing some hesitance to accept the doctrine of substantive legitimate expectations in Singapore law, suggested that the doctrine would cohere uneasily with the legality–merits distinction in administrative law.¹⁸ Another recurring criticism of the doctrine is that its normative basis remains unclear.¹⁹ This is the issue which will be the focus of the rest of this article.

The normative foundation of the doctrine: The invocation of ‘good administration’

The courts have vacillated with respect to the proper normative justification for the protection of legitimate expectations. Three key concepts have recurred in case law—fairness, abuse of power and good administration.²⁰ The relationship between these concepts is ambiguous. On first impression, one might perceive these concepts as all competing on the same plane as justifications for the doctrine. There are some indications in case law, however, that these concepts are closely related.²¹ The connection between these concepts will be returned to later in this article.

For present purposes, the point to highlight is that there are indications in case law of a judicial inclination towards the concept of good administration as the central normative justification of the doctrine. The idea of ‘good administration’ as a key justification for the doctrine began to gain traction following Laws LJ’s suggestion in R (Nadarajah) v Secretary of State for the Home Department (‘Nadarajah’) that the doctrine of substantive legitimate expectations is an expression of ‘a requirement of good administration’, ‘by which public bodies ought to deal straightforwardly and consistently with the public’.²² Indeed, in his Lordship’s view,

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.²³

Laws LJ subsequently had the opportunity to elaborate upon this proposed normative foundation of the doctrine in R (on the application of Bhatt Murphy (a firm)) v Independent Assessor. In this case, Laws LJ’s view of the centrality of ‘good administration’ as the key normative justification of the doctrine became even more apparent—his Lordship suggested that the substantive legitimate expectations doctrine required public authorities to act fairly and refrain from abuses of power—grounded ultimately by the requirement of ‘good administration’.²⁴ Accordingly, it is clear that in Laws LJ’s view, it is the concept of ‘good administration’ which forms the core justification for the doctrine of substantive legitimate expectations, incorporating fairness and prevention of abuse of power as substantiations of the requirements of good administration. This idea has gained traction beyond Laws LJ as well. Indeed, the centrality of the idea of ‘good administration’ as the normative justification for the doctrine was recently reinforced by Lord Kerr in the UK Supreme Court decision of In the matter of an application by Geraldine Finucane for Judicial Review (Northern Ireland), where his Lordship held that the doctrine of legitimate expectations was ‘underpinned by the requirements of good administration’.²⁵ Scholars have also argued that good administration does and indeed ought to feature as a conceptual basis for the doctrine of legitimate expectations.²⁶

This trending centrality of ‘good administration’ as the normative justification for the doctrine will be the focus of attention in this article. A key issue with the invocation of ‘good administration’ as a justification for the doctrine is that there is significant ambiguity as to what exactly ‘good administration’ entails. The ambiguity of the concept is widely acknowledged in case law and academic commentary alike.²⁷ Indeed, bare references to the concept of ‘good administration’ as the justification for the doctrine have been criticised as vacuous and empty of normative content.²⁸ The vagueness of such justifications has been acknowledged by the courts themselves—the Court of Appeal in R (on the application of Bibi) v Newham LBC noted that the case law on the doctrine was replete with references to ‘fairness’, ‘abuse of power’ and ‘good administration’ and cautioned that ‘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’.²⁹

Indeed, ‘good administration’ is a capacious concept that can incorporate a variety of different norms. Judges and commentators have offered many possibilities as to what these norms are—preventing abuses of power, giving effect to expectations, consistent application of policies, promoting rules-based management, fairness and fostering trust in public authorities, inter alia.³⁰ To add to the confusion, these norms can overlap with each other—for example, as will be suggested later in this article, fairness and abuse of power are normative justifications which are conceptually closely related to each other.³¹ The outcome of this conceptual ambiguity is that ‘good administration’ has become what Matthew Groves colourfully described as a ‘motherhood statement’—statements which are ‘often used as a shield to divert attention and criticism from the highly contested concepts they seek to implement and the many practical problems that these concepts might introduce’ and which imply an imaginary consensus over the concept where none actually exists.³²

This ambiguity as to the proper normative content of ‘good administration’, with the consequent ambiguity for the normative foundation of the legitimate expectations doctrine, has real implications for the clarity of the doctrine. Indeed, depending on what precisely ‘good administration’ is taken to encompass, the doctrinal weight to be accorded to each element in the legal test for the doctrine can shift significantly.³³ For example, if one conceives of the doctrine as principally directed at upholding trust in public administration, this would justify a doctrinal reduction of emphasis in the relevance of the scope of the representation and would cast doubt on the hesitation that judges have expressed against upholding representations directed at large classes of people.³⁴ Indeed, Paul Reynolds argued that while the concept of ‘good administration’ provides ‘a clue as to the doctrine’s ‘moral impetus’’, it is ‘too vague and overarching to provide any concrete delimitation or guidance’ as a matter of doctrine.³⁵ In a similar vein, Knight argued that the loose way in which various justifications and concepts have been proffered as bases of the doctrine suggests that ‘the different terms are being used interchangeably and are about as useful as one another’, as a matter of providing principled guidance to the resolution of cases.³⁶

In relation to the invocation of ‘good administration’ as a justification for the doctrine of legitimate expectations, a significant amount of attention has been paid to the ambiguity as to what exactly ‘good administration’ entails. On this view, the proper solution to rectify such ambiguity would be to articulate the normative values behind the doctrine with the greatest precision possible. Indeed, the bulk of literature dedicated to this issue is directed at this goal.³⁷

This article suggests, however, that while such work has achieved admirable results, there is a deeper level of ambiguity to the concept of ‘good administration’ which complicates such efforts. This is because the concept of ‘good’ invoked by the idea of ‘good administration’ will fall to be substantiated by a particular conception of what ‘good’ law and government are³⁸—indeed, different conceptions of the ‘good’ in political philosophy will translate into different visions of just what exactly ‘good administration’ is. If this is correct, then the very invocation of ‘good administration’ as a justification for the doctrine is inextricably tied to a concept in political philosophy which is deeply contentious, with the consequence of introducing an additional layer of ambiguity in the proper normative justification for the doctrine. The rest of this article will be dedicated to exploring and substantiating this proposition.

The connection between ‘good administration’ and the ‘good’

One can discern at least four different conceptions of the ‘good’ in law and government that underlie academic commentary and judicial reasoning on the normative foundation of the legitimate expectations doctrine. The first suggests that the fullest expression of ‘good’ law and government requires law and government to further the common good and uphold substantive justice. The second proposes that ‘good’ law and government should be primarily measured by its coherence with procedural requirements of the rule of law. The third is the idea that law and government serve primarily to protect individual rights and entitlements to the maximum possible extent. The fourth suggests that ‘good’ law and government are primarily directed at promoting a community characterised by mutual interpersonal trust.

As a preliminary matter, it should be noted that these four conceptions of the good are not necessarily mutually exclusive. There can be overlaps between these ideas. For example, one may validly view the ideas of trust and the rule of law as subsets of the idea that good law and government ought to further substantive justice. The doctrine of legitimate expectations may thus straddle multiple conceptions of the good.³⁹ Yet, it is suggested that such overlaps should not obscure important differences between these conceptions of the good—while a broader conception of the good can indeed encapsulate several others, each conception of the good remains distinct as a matter of emphasis and scope.

It will be argued that different conceptualisations of the normative foundation of the doctrine correspond closely to these competing conceptions of the ‘good’ in law and government. Each conception of the good will be discussed in turn.

Fairness and preventing abuses of power

The first category of justifications for the doctrine of legitimate expectations emphasises the requirements of fairness and substantive justice in relation to the dealings between public authorities and individuals. This category of normative justifications for the doctrine captures the strain of thinking among judges and commentators that the doctrine of legitimate expectations is focused on upholding the duty of public authorities to act fairly and avoid abuses of power.

This category of justifications has achieved significant traction. Judicial support for the concept of fairness as the normative foundation of the doctrine can be found all the way back in Council of Civil Service Unions v Minister for the Civil Service—while discussing the concept of legitimate expectations, Lord Roskill suggested that the fundamental concern in such situations was ‘the duty to act fairly’.⁴⁰ In R v Inland Revenue Commissioners, ex p MFK Underwriting Agents Ltd, Bingham LJ held that ‘the doctrine of legitimate expectation is rooted in fairness’.⁴¹ A clear expression of this proposed normative foundation for the doctrine can be found in R v Inland Revenue Commissioners, ex p Unilever Plc.⁴² In that case, the Inland Revenue sought to enforce strictly against Unilever its time limit for claims for tax relief, despite its longstanding practice over 20 years of accepting Unilever’s claims for tax relief out of time, with the effective consequence of denying Unilever tax relief for that relevant year. Even though the Court of Appeal considered that the relevant long-standing practice of the tax authority was not sufficient to constitute a clear and unambiguous representation, the court nevertheless held that ‘the categories of unfairness are not closed’ and upheld Unilever’s legitimate expectations claim. It is telling, from the Court of Appeal’s treatment of the doctrine, that it was of the view that the doctrine was ultimately founded on the concept of fairness and could thus be legitimately invoked to give effect to this central normative concern.

Looking beyond the United Kingdom, courts in other common law jurisdictions have also perceived the doctrine as conceptually founded on fairness. In their discussion of whether the legitimate expectations doctrine ought to apply in Australian law, the Australian courts have characterised developments in English law in this regard as premised upon the concepts of substantive unfairness and abuse of power.⁴³ In Hong Kong, where the doctrine of substantive legitimate expectations has been accepted as an independent ground of judicial review, a similar concern with fairness can be discerned in the various judgments in the Hong Kong Court of Final Appeal’s (CFA) decision in Ng Siu Tung v Director of Immigration (‘Ng Siu Tung’).⁴⁴ In that case, the CFA described the doctrine as grounded in ‘the principle of good administration and the duty to act fairly’⁴⁵ and invoked fairness as the reason for the invocation of the doctrine against selected classes of applicants in the case—the CFA held that in relation to the class of applicants who had received specific representations that the Immigration Department would follow the final decision of the courts, thus encouraging them not to mount their own applications for judicial review, the disappointment of the expectation that this representation had engendered amounted to ‘a very substantial degree of unfairness’ justifying the court’s intervention.⁴⁶ Bokhary PJ, in his dissent in the same case, similarly drew upon the idea of fairness—in his words: ‘whether they are enforcing legitimate expectations procedurally or substantively, the courts are acting to accord fairness’.⁴⁷

This normative justification for the doctrine was termed by Farrah Ahmed and Adam Perry the ‘fairness account’ of the doctrine. They described the fairness account of the doctrine as expressing ‘the idea that promises, practices, and policies generate legitimate expectations because of their relationship to fairness’.⁴⁸ Indeed, Trevor Allan has suggested that the doctrine of substantive legitimate expectations is merely a restatement of the general principle of fairness in a more technical form.⁴⁹ Paul Craig and Soren Schonberg have described the doctrine as functioning to alleviate ‘severe hardship’ caused by public authorities to individuals.⁵⁰

Falling within the same category of normative justifications for the doctrine are those which suggest that the doctrine is ultimately based on preventing abuses of power. The prevention of abuses of power has been often invoked in itself as a justification for the doctrine.⁵¹ For example, the Court of Appeal in Coughlan described the doctrine of legitimate expectations as ‘a distinct application of the concept of abuse of power’.⁵² Building upon Coughlan, Laws LJ in R v Secretary of State for Education and Employment, ex p Begbie (‘Begbie’) suggested that the three categories of legitimate expectations cases described in Coughlan were all informed by the concept of abuse of power.⁵³

Given the prominence of abuse of power as a normative justification for the doctrine, the prevention of abuse of power has been described as a competing justification vis-à-vis the justification based on fairness.⁵⁴ However, it is suggested that the justification based on preventing abuses of power is substantially similar in kind to the justification based on fairness—avoiding abuses of power is simply the negative component of the positive duty of public authorities to act fairly. Indeed, it is telling that fairness and preventing abuses of power are often mentioned in the same breath.⁵⁵ For example, the Court of Appeal in R (on the application of Association of British Civilian Internees—Far East Region) v Secretary of State for Defence, in the course of discussing whether a clear and unequivocal representation was necessary for a legitimate expectations claim to succeed, held that ‘it will only be in a rare case where, absent such a representation, it can be said that a decision-maker will have acted with conspicuous unfairness such as to amount to an abuse of power’.⁵⁶ In a similar vein, the Privy Council in Rainbow Insurance Company Ltd v Financial Services Commission (Mauritius) held that ‘the courts have developed the principle of legitimate expectation as part of administrative law to protect persons from gross unfairness or abuse of power by a public authority’.⁵⁷

The invocation of fairness in relation to the doctrine of legitimate expectations has been criticised.⁵⁸ Nevertheless, the important point for present purposes is that the normative justification for the doctrine based on fairness and prevention of abuse of power reflects a certain conception of good administration.⁵⁹ And this vision of good administration tracks a certain conception of how the ‘good’ should be instantiated in law and government—that is, law and government should ultimately further the common good of the political community, defined in substantive moral terms. On this view, good administration should be concerned with substantive fairness. Indeed, good administration would be perceived as a means by which the common good of the political community can be realised and would encompass both the fairness of administrative processes and the justice of administrative outcomes.

A theorist who has proffered a closely related conception of the ‘good’ in law and government is the legal philosopher John Finnis. Finnis argued that the common good in a political community referred to

a set of conditions which enables the members of a community to attain for themselves reasonable objectives, or to realize reasonably for themselves the value(s), for the sake of which they have reason to collaborate with each other (positively and/or negatively) in a community.⁶⁰

Justice specifies in outline what is required for the common good⁶¹—for example, in the furtherance of distributive justice, public authorities have a duty to pursue the common good, and not their own private advantage.⁶² Further, Finnis suggested that public authorities owe duties of commutative justice to those subject to their authority, which would require them to carry out lawful and regular administration to accord each person his or her due rights and entitlements.⁶³ Unlike Rawls, however, who would have viewed this as a comprehensive principle of justice in itself, Finnis perceived this merely as a component of what the common good, defined in substantive terms, requires. In a similar vein, Finnis viewed the idea of the rule of law, as expressed by Lon Fuller’s eight requirements of internal morality, as a component of justice and fairness,⁶⁴ intended to secure for individuals the human dignity of self-direction and freedom from manipulation.⁶⁵ Indeed, in Finnis’s view, the entire enterprise of law—encompassing the moral good of the rule of law—derives its moral force from basic principles of justice and basic human goods.⁶⁶

Accordingly, such a conception of the ‘good’ in relation to law and government posits a deep connection between law, government and substantive justice. On this view, the protection of rights and entitlements, upholding procedural fairness, and ensuring consistency and rule-based management are not the ends of law in themselves but are ultimately referable and justified by broader conceptions of fairness and justice directed at securing human flourishing. This conception of the ‘good’ would translate into a certain vision of ‘good administration’—the fundamental concern of good administration should be to further the common good and secure substantive fairness and justice.

Articulating the conceptual justification of the legitimate expectations doctrine by reference to the common good brings with it an advantage which should be noted at this point: it would serve to explain and justify the interaction between upholding legitimate expectations and the rule against fettering. The two doctrines may be perceived as diametrically opposed. The rule against fettering requires administrators to exercise their power flexibly; yet, the doctrine of legitimate expectations requires administrators to be held to their previous policies. This tension can be negotiated in a principled manner if one sees the requirements of ‘good administration’ as linked to the idea of the common good. While ‘good administration’ may generally require legitimate expectations to be upheld, ‘good administration’ may sometimes itself require administrators not to be held rigidly to their previous policies to accord them decision-making flexibility, which would in turn further the common good overall. In other words, ‘good administration’ can itself justify the dashing of a legitimate expectation, when viewed through the lens of this normative justification.⁶⁷

Upholding law-like, rules-based administration

A second category of normative justifications for the doctrine of legitimate expectations suggests that the doctrine is fundamentally concerned with ensuring that executive power is exercised in a law-like fashion. On this view, the doctrine of legitimate expectations is justified by a normative concern with the consistency and procedural regularity of public decision-making.

There is ample evidence in case law in support of this justification as the key normative concern of the doctrine. Lord Carnwath in United Policyholders Group v Attorney General of Trinidad and Tobago suggested that the doctrine of legitimate expectations ought to be justified on the ‘basic rule of law and human conduct that promises relied on by others should be kept’.⁶⁸ In a similar vein, Sedley LJ in Begbie held that a key concern of the doctrine is that of legal certainty—that public authorities should be held to the representations they had made.⁶⁹ Mark Elliott argued powerfully that R (Rashid) v Secretary of State for the Home Department ⁷⁰ (‘Rashid’) ought to be explained on the basis of the principle of consistency—in his view, the principle of consistency captures the idea that ‘administration bodies should not arbitrarily decline to apply the same, self-proclaimed norms to all relevant cases which come before them so long as the policy remains in force’.⁷¹ Given that the court in Rashid had applied the doctrine of legitimate expectations to decide the case, Elliott’s analysis of the heart of the reasoning in Rashid is an important indication of the court’s perception in Rashid of the normative content of the doctrine.⁷²

Judicial invocation of the doctrine of legitimate expectations to require public authorities to provide principled transitions between changes of policies also suggests a perception of the doctrine as fundamentally concerned with consistency and rules-based administration.⁷³ For example, in R (Patel) v General Medical Council,⁷⁴ the Court of Appeal upheld the substantive legitimate expectation of the applicant that his relevant qualification would be sufficient to qualify him for medical practice in the United Kingdom, despite the medical council’s subsequent change of policy to deny recognition of his qualification. In coming to its decision, the Court of Appeal suggested that the medical council should have carried out its change of policy in phases to account properly for individuals affected by the shift between policies. Indeed, the courts appear to have taken the position that the existence of such transitional provisions would limit the applicability of the doctrine—as long as public authorities have set in place such transitional provisions, the courts will not inquire too deeply into the details of such provisions.⁷⁵ Such judicial attitudes indicate a view of the doctrine of legitimate expectations as chiefly concerned with ensuring rules-based and law-like administration—as long as this obtains, the scope of application of the doctrine is accordingly limited.

Commentators have also suggested that consistency and certainty lie at the heart of the doctrine. Philip Sales and Karen Steyn defined the interests protected by the doctrine as ensuring ‘a high degree of predictability as to the behaviour of a public authority’ as a requirement of the rule of law⁷⁶—indeed, on this view, the ‘ideal of legal certainty’ lies at ‘the foundation for the protection of legitimate expectations’.⁷⁷ In a similar vein, Craig and Schonberg have argued that the protection of substantive legitimate expectations is justified by the rule of law, as an important component of ensuring the consistency, predictability and certainty of administrative action.⁷⁸ Paul Daly has also argued that consistency is crucial to good administration—accordingly, policies and decisions should be departed from only where there is a reasonable justification for doing so.⁷⁹ Indeed, Ahmed’s and Perry’s account of the doctrine as focused on ensuring that agencies abide by social rules which may not have legal status⁸⁰ and which are goal-dependent⁸¹ is an account which emphasises the normative concern of the doctrine in upholding good rules-based and law-like administration—on this account, ‘promises, practices, and policies generate legitimate expectations because they are all ways a public body binds itself with a non-legal and goal-dependent rule’.⁸²

The principle of consistency as the normative foundation of the legitimate expectations doctrine has been criticised—among other things, it has been criticised as providing a normatively impoverished account of the doctrine, in comparison to substantive fairness and abuse of power.⁸³ Further, given that the principle of consistency has been developed as a separate ground of review in English law,⁸⁴ the adoption of consistency as the normative foundation for the doctrine of legitimate expectations puts the doctrine at risk of being redundant. Indeed, Varuhas has argued that for the sake of greater clarity as to the contours of the doctrine of legitimate expectations, several situations which have thus far been analysed under the doctrine of legitimate expectations are better categorised as applications of the principle of consistency instead.⁸⁵ Nevertheless, the key point for present purposes is that this category of normative justifications for the doctrine represents a certain conception of what ‘good administration’ entails. On this view, ‘good administration’ represents law-like behaviour on the part of public authorities—authorities should make decisions in a principled, consistent, non-arbitrary and rules-based manner. This conception of ‘good administration’, in turn, tracks a particular vision of the ‘good’ in relation to law and government—that is, that the coercive power of the state should be exercised through law in accordance with procedural rule of law principles such as consistency, rationality and clarity.

This conception of the ‘good’ resembles Lon Fuller’s well-known account of law as order, expressed in his eight desiderata of law’s internal morality. Fuller argued that there are eight minimum requirements of the rule of law: there must be rules; the rules must be promulgated; there should be no retroactivity in law; rules must be understandable; non-contradictory; not require the impossible; be sufficiently constant; and there must be congruence between the rules and their actual administration.⁸⁶ These eight requirements of internal morality represent both what Fuller termed the morality of duty and the morality of aspiration—in other words, they demarcate the minimum standard that rule of law must be achieved for it to be properly called law, while also presenting an aspiration of perfect legality.⁸⁷

It should be noted that Fuller’s eight desiderata were directed at law—Fuller argued that different considerations ought to apply to administrative action, especially in relation to polycentric decision-making which would be much less amenable to law-like constraints.⁸⁸ Nevertheless, it is suggested that this does not pose a problem for the reference to Fuller in the context of the argument being made here. Indeed, it should be noted that Fuller’s hesitation to apply the eight desiderata to administrative action is not a suggestion that administrative action ought not to be law-like.⁸⁹ Rather, his argument that certain administrative decisions are not amenable to the idea of internal morality of law closely tracks the doctrine of non-justiciability—that there are some provinces of administrative decision-making that are not susceptible to judicial review. This implies that the provinces of administrative decision-making which are not excluded from judicial intervention by the doctrine of non-justiciability ought to be amenable to Fuller’s eight desiderata.

Protection of rights and entitlements

The third category of normative justifications for the doctrine of legitimate expectations can be described as focused on individual rights and entitlements—this category of justifications focuses on upholding the integrity of promises, meeting individual expectations and safeguarding the rights of citizens vis-à-vis government agencies.

As a preliminary matter, one might be inclined to view such justifications as conceptually independent from ‘good administration’—such justifications may be perceived as based on the conceptual foundation of rights-protection, and thus focused on giving effect to an individual’s rights and entitlements, rather than the conduct of an administrative body. It is suggested, however, that such justifications are not necessarily distinct from the concept of good administration. One can explain the existence of elements of rights-protection analysis in public law doctrine by observing that the protection of rights is an important element of good administration—indeed, such protection gives effect to commutative justice, an aspect of the ‘good’.⁹⁰ As such, this category of justifications can be validly perceived as a specific instantiation of what ‘good administration’ requires.

This normative justification for the doctrine suggests that the doctrine is principally about upholding the integrity of promises made by public authorities and has been described as ‘almost-contractual’ in nature.⁹¹ This justification proposes that the doctrine primarily seeks to protect the rights or interests of individuals that have been legitimately generated by the conduct of a public authority. On this view, the doctrine is closely analogous to the private law doctrine of promissory estoppel—just as private law rights can be generated by the conduct of a contracting party through the doctrine of estoppel, the doctrine of legitimate expectations can confer analogous public law rights on individuals who have been the recipients of similar conduct by public authorities.⁹²

Support for this justification as the key normative justification for the protection of legitimate expectations can be found throughout case law. In Coughlan, the Court of Appeal held that the health authority could not renege on its promise to provide a ‘home for life’, especially since the relevant promisees had acted upon the promise to move into Mardon House.⁹³ The health authority’s promise, followed by detrimental reliance on the part of the promisees, effectively generated a quasi-contractual exchange of obligations between the authority and the affected individuals, thus forming the basis for the doctrine to give effect to these rights and obligations. Beyond Coughlan, other judges and commentators have expressed a view of the doctrine as focused on rights-protection. For example, Simon Brown LJ in R v Devon County Council, ex p Baker invoked the language of rights in his holding that the doctrine of substantive legitimate expectations is capable of conferring substantive rights upon individuals where there has been a clear and unambiguous representation.⁹⁴ In a similar vein, Abhijit Pandya argued that ‘the doctrine of legitimate expectations in English law protects individuals from changes to representations made by government bodies. This protection can arise by giving individuals either due process rights or substantive rights’.⁹⁵

Beyond the United Kingdom, while the Singapore courts have expressed significant reluctance to accept the doctrine of substantive legitimate expectations,⁹⁶ their analysis of the doctrine to date evinces a perception that upholding the integrity of promises that have resulted in an exchange of obligations lies at the heart of the doctrine. In Chiu Teng @ Kallang Pte Ltd v Singapore Land Authority, the Singapore High Court suggested that the doctrine should apply in Singapore, explaining the doctrine by using an analogy to private law estoppel: ‘If private individuals are expected to fulfil what they have promised, why should a public authority be permitted to renege on its promises or ignore representations made by it?’⁹⁷

This normative justification of the doctrine is a particular conception of what ‘good administration’ requires—that is, good administration requires the proper protection of rights and entitlements generated by the conduct of public authorities pursuant to a quasi-contractual exchange of obligations. This conception of what ‘good administration’ requires tracks closely against a particular conception of the ‘good’ in law and government—that is, that the primary ‘good’ that political institutions and law should be concerned with is the protection of rights and entitlements in the pursuit of upholding liberty for individuals. In other words, this position would conceptualise the ‘good’ as primarily about giving effect to entitlements fairly.

A well-known political theory closely related to such a conceptualisation of the ‘good’ is that proposed by the renowned political philosopher John Rawls. Rawls argued that the overall aim of the political community is to secure maximal liberty for every person.⁹⁸ Rights in a political community should be justified in accordance with principles of cooperation which citizens can be expected to acknowledge ‘when each is fairly represented as moral persons’⁹⁹—highlighting the importance of autonomous choice and individual liberty in Rawls’s vision of the political community. Accordingly, law and government should regulate the community in accordance with such principles of cooperation. It is by doing so that government would be able to satisfy the citizens’ public conception of justice.¹⁰⁰ Rawls went on to describe the rule of law as ‘justice as regularity’ and associated the rule of law with precepts of justice such as having like cases treated alike, due process and maintaining clarity of legal rules.¹⁰¹ Crucially, Rawls justified the importance of these rule of law principles by reference to the protection of individual liberties. In his view, laws ‘constitute grounds upon which persons can rely on one another and rightly object when their expectations are not fulfilled’.¹⁰² Indeed, Rawls argued that the rule of law is founded on ‘the agreement of rational persons to establish for themselves the greatest equal liberty’.¹⁰³

On such a conception of the ‘good’, it is crucial to uphold individual rights and entitlements. Justice will indeed be measured principally by the ability of the legal framework to secure individual rights and entitlements. Accordingly, this translates into a certain vision of what ‘good administration’ requires. ‘Good administration’ would safeguard liberty by protecting the legitimate expectations of individuals through rules directed at according to individuals due rights and entitlements. Indeed, ‘fairness’, on such a conception of the ‘good’, would be tied to giving effect to principles of cooperation that each would reasonably agree to—the very idea of fairness is closely linked to individual liberty. Described as such, the normative justification for the doctrine of legitimate expectations based on rights-protection is closely related to a vision of ‘good administration’ which draws upon the conception of the ‘good’ articulated in the Rawlsian view of the proper aim of law and government.

Fostering trust in public institutions and authorities

A final category of normative justifications for the doctrine of legitimate expectations suggests that the doctrine is principally directed at fostering trust in public institutions.

There is some support for this normative justification in case law and academic commentary. Its most notable invocation in case law can be found in Bokhary PJ’s dissenting judgment in Ng Siu Tung. In his discussion of the reasons to recognise the doctrine’s applicability in Hong Kong, he raised fostering trust in public administration as one of the key reasons.¹⁰⁴ Aside from this, however, commentators have noted that the ‘judicial recognition of the role of trust has been limited’.¹⁰⁵ Nevertheless, commentators have suggested that the promotion of trust in public authorities is a key principle of administrative law and can be an apt justificatory foundation for the doctrine of legitimate expectations.¹⁰⁶ Iain Steele pointed out that the regular frustration of legitimate expectations can lead to the erosion of trust in public authorities and damage to their institutional reputation.¹⁰⁷ This will have detrimental consequences both for affected individuals and public authorities—individuals will suffer injustice, while the efficacy of administration would ultimately be affected by systemic distrust in public administration.¹⁰⁸ Accordingly, preventing such erosion of trust has been suggested to be a good normative foundation of the doctrine that can provide greater principled constraint to the doctrine’s application.¹⁰⁹

This normative justification for the doctrine reflects a certain vision of good administration. Indeed, Reynolds argued that good administration requires the political legitimacy of administrative action, defined as the idea that ‘public institutions are valued for themselves and considered proper and right’.¹¹⁰ This occurs when citizens are able to trust public authorities. This conception of ‘good administration’, in turn, reflects a certain conception of the ‘good’ in law and government—that law and government are primarily concerned with promoting relationships of trust, between citizens as well as between citizens and the government.

Such a conception of the ‘good’ in law and government is reminiscent of Roger Cotterrell’s vision of law as centred on trust. Cotterrell argued that perceiving law principally through the lens of state law—identified with a particular society and state—limits our perspective to law.¹¹¹ He suggested that one can arrive at a broader perspective towards law by viewing law as linked to the idea of community—as a social phenomenon, rather than simply the law of a political society.¹¹² A community, in his view, is a set of relations characterised by a high degree of mutual interpersonal trust between its members.¹¹³ Trust is the basic element of a community.¹¹⁴ Given this definition of community, fostering trust within a community becomes a central regulatory concern of law.¹¹⁵ Specifically, a central concern of law is to foster trust-based interactions within the community, with a view to promoting a community characterised by a high degree of endurance and mutual interpersonal trust. In Cotterrell’s view, law can do so by ‘sustaining and encouraging patterns of trust embodied in ideal typical forms of collective involvement or interaction’.¹¹⁶ While Cotterrell’s main focus was to expand the regulatory possibilities for law to regulate citizen–citizen relationships, the concept of trust is easily extended to relations of trust between the governed and the government—indeed, Cotterrell described ‘judges or other decision-makers’ as ‘in affective community with litigants and perhaps the wider public’.¹¹⁷

This conception of the ‘good’ in law and government focuses on their role in fostering a community characterised by relationships of trust among citizens and between citizens and the government. This translates into a certain conception of ‘good administration’—good administrative decision-making should ultimately promote such relationships of trust. Described as such, it should be apparent that the normative justification for the legitimate expectations doctrine based on trust reflects these conceptions of ‘good administration’ and the ‘good’ in law and government.

Implications of the connection between ‘the good’ and ‘good administration’

Thus far, this article has argued that there are different levels of ambiguity which make it challenging to articulate the concept of ‘good administration’ as the normative foundation of the legitimate expectations doctrine. On one level, there is ambiguity as to what ‘good administration’ entails—which conception of good administration justifies the doctrine of legitimate expectations? Going further, the central argument of this article has been that there is another level of ambiguity which magnifies the challenges of using ‘good administration’ as the normative foundation of the doctrine—that is, ambiguity about what the ‘good’ is, in relation to law and government. There are multiple competing conceptions of what ‘good’ law and government are. Given then that there is a contest among rival conceptions of the ‘good’ and that each conception of ‘the good’ translates into different conceptions of ‘good administration’, it is surely unsurprising that bare references to ‘good administration’ as the normative foundation of the doctrine are unable to provide significant principled guidance as to the proper shape of the doctrine. Disagreements over the content and scope of ‘good administration’ reflect and are indeed traceable to deeper disagreements over underlying conceptions of ‘the good’ in law and government. Accordingly, this potentially forms at least part of the explanation for the continued disagreement and lack of consensus over the proper normative foundation of the legitimate expectations doctrine.

One might argue that since these differing conceptions of the ‘good’ in law and government may not necessarily be conceptually mutually exclusive, a possible solution to this problem would be to adopt all of them as the normative justifications for the doctrine. It is suggested, however, that this would be an unsatisfactory way forward. Each different conception of the ‘good’—and as a corollary, ‘good administration’—has specific doctrinal implications. Accordingly, adopting all the proposed visions of ‘good administration’ as the normative foundation of the doctrine would lead to divergent developments of the doctrine.

For example, with respect to the legal significance that should be accorded to individual representations as opposed to representations made through policies, different conceptions of the ‘good’ will lead to divergent answers. The courts’ application of the doctrine thus far has indicated that individual representations will be accorded greater legal significance as compared to general representations made through policies—accordingly, it is generally easier to succeed under the doctrine if the relevant representation is an individual promise made by a public authority.¹¹⁸ Yet, the conceptions of the ‘good’ which focus on the law’s role in promoting mutual interpersonal trust, or in the upholding of justice and the common good defined in substantive moral terms, would not necessarily agree that a specific representation carries greater moral force than a general representation made through a policy—indeed, since the failure to give effect to representations made through policies may have a larger impact on the common good and/or systemic trust in public institutions, one might take the view that such representations should be accorded at least equal, if not more, legal significance as compared to representations made through individual promises.¹¹⁹ A similar outcome would follow applying a conception of the ‘good’ focused on upholding law-like decision-making—the scale of the implications of a procedural defect would strengthen a claim for relief under the doctrine.¹²⁰ On the other hand, a conception of the ‘good’ that focuses on the protection of individual rights and entitlements would be indeed likely to reach the conclusion that specific representations to individuals do carry greater moral force and ought to be accorded greater legal significance.

Another doctrinal issue implicated by varying conceptions of the ‘good’ is when an expectation is to be considered legitimate for the purposes of the doctrine.¹²¹ As was noted earlier, detrimental reliance has not been considered a necessary condition for a legitimate expectations claim to succeed. However, a conception of the ‘good’ focused on the protection of individual rights and entitlements might suggest, to the contrary, that detrimental reliance ought to be a legally significant condition for the doctrine to take effect, as a requirement for the exchange of obligations which can consequently give rise to legal rights for the applicant.¹²² A conception of the ‘good’ that focuses on the upholding of procedural rule-of-law norms might, however, not view detrimental reliance as a necessary requirement for the doctrine to take effect—the focus would be on the specific conduct of the public authority, rather than an exchange of legal rights and obligations between the relevant individual and the public authority.¹²³ Conceptions of the ‘good’ based on fairness and fostering trust in public administration would take a broader view of the relevant factual matrix. On such conceptions of the ‘good’, detrimental reliance would be simply an element in a broader and wider-ranging consideration of whether injustice has been done, or whether trust in public administration has been impaired.

Accordingly, the specific conception of the ‘good’ and ‘good administration’ taken to be the normative foundation of the doctrine has the potential to shape quite significantly the contours of its doctrinal framework. Given then the central argument made in this article—that ‘good administration’ as a normative foundation for the doctrine introduces an additional layer of contention over conceptions of the ‘good’—and if amalgamating different conceptions of the ‘good’ as the normative foundation of the doctrine is an unsatisfactory solution, what ought to be the way forward for thinking about the normative foundation of the doctrine?

One possible implication to draw from this discussion is that ‘good administration’ simply should not be invoked as a normative justification for the doctrine of legitimate expectations. Indeed, if it is true that such invocation introduces additional layers of ambiguity into the conceptualisation of the doctrine’s normative foundation, one may justifiably conclude that ‘good administration’ is simply an unhelpful normative justification that does little to constrain and shape the development of the doctrine in a principled manner. This does not mean that one is foreclosed entirely from invoking ideas of the ‘good’ as normative justifications for law—the argument would simply be that such invocations of the ‘good’ in relation to a doctrine in dire need of clarity over its normative foundation to shape its doctrinal content are unlikely to be helpful.

Beyond this implication, however, it is suggested that there are other possible implications to be drawn from the central argument in this article. Indeed, identifying a deeper layer of contention with respect to the invocation of ‘good administration’ as a normative foundation for the doctrine has value beyond explaining the protracted disagreement over the proper normative foundation of the doctrine—it presents an opportunity to get to the bottom of the dispute by addressing directly the deeper underlying level of disagreement. Should one recognise the existence of this underlying level of disagreement, a possible way ahead for thinking about the doctrine would be to articulate the most justifiable concept of the ‘good’ in relation to good administration.

Such thinking could proceed in the following fashion. One might argue that a rights-protection-focused account of the ‘good’, translated into ‘good administration’ as a normative foundation for the doctrine, risks losing sight of the fact that beyond the protection of rights and entitlements, public law is concerned with broader public interest considerations such as justice, fairness and the legitimacy of public authorities. Varuhas has criticised a rights-focused conception of administrative law, arguing that ‘significant features of doctrine tell against a wholesale recalibration of administrative law around rights’.¹²⁴ Indeed, in specific relation to the substantive legitimate expectations doctrine, such a conception of the ‘good’ risks not giving sufficient effect to judicial pronouncements that the doctrine should be ready to stand on its own feet, separate from the private law concept of promissory estoppel.¹²⁵ Beyond these arguments, one could go on to critique such a conception of the ‘good’ as a governing ideal for law and government more broadly—conceivably on the basis that such a conception of the ‘good’ unduly prioritises liberty and autonomy as ultimate values, foreclosing the possibility that there may be other goods and values worth facilitating as a matter for law and government.¹²⁶

Going further, one could argue that a conception of the ‘good’ emphasising the role of law and government in fostering substantive fairness and justice is similarly problematic when translated into a conception of ‘good administration’ serving as the foundation of the legitimate expectations doctrine. As one moves further away from classical due process norms and into the realm of substantive norms of justice, the possibilities of disagreement over the relevant content of these norms increase markedly.¹²⁷ Accordingly, as a normative justification for the doctrine, such a conception of the ‘good’ risks being so broad as to be unable to provide much principled guidance. As such, while this may be considered an appealing conception of the ‘good’ in view of its robust substantiation of what law and government are for in substantive moral terms, as a principled normative justification for the doctrine of legitimate expectations, this conception of the ‘good’ might be an unsatisfactory solution.

Yet, the idea that substantive conceptions of the ‘good’ should provide moral substantiations of what law and government are for in the context of the legitimate expectations doctrine can remain very appealing for several reasons, including, inter alia, the fact that this provides a useful framework within which one can negotiate the interaction between the legitimate expectations doctrine and the rule against fettering, as mentioned earlier. Holding on to this idea, one may consider that a conception of ‘good’ law and government focused on procedural rule-of-law norms provides an appropriately principled normative justification for the doctrine, when translated into a vision of ‘good administration’. Indeed, such a conception of the ‘good’ recognises that the ‘good’ should be substantiated in moral terms and yet provides a more constrained account of such an incorporation of morality by limiting the concern of law to internal morality, in Fuller’s terms. In response to the criticism that this results in a normatively impoverished account of the doctrine, this conception of the ‘good’ can be augmented with a conception of the ‘good’ suggesting that law and government ought to promote mutual interpersonal trust. Indeed, in contrast to a broader reference to substantive fairness and justice, such a conception of the ‘good’ would be capable of providing a more specific substantiation of the normative content of justice, thus translating better into a vision of ‘good administration’ that can provide a principled normative foundation for the doctrine.

Identifying this additional layer of disagreement over the normative foundation of the doctrine gives rise to another opportunity. When one perceives that the substantive legitimate expectations doctrine is capable of reflecting different conceptions of the ‘good’ in law and government—and has indeed done so—this suggests that the doctrine is sufficiently malleable and responsive to different conceptions of the ‘good’. Accordingly, it is possible for the doctrine to be suitably adjusted to take into account different conceptions of the ‘good’ in law and government that may obtain in different jurisdictions. The recognition of this possibility is of some importance with respect to jurisdictions which have thus far expressed scepticism over the introduction of the doctrine as a new ground of judicial review—Singapore, for example.¹²⁸ The Singapore courts have expressed fears that the doctrine represents a red-light conception of the proper institutional relationship between the courts and public authorities—a conception which the Singapore courts have considered to sit uneasily with the Singapore context.¹²⁹ Yet, it may be possible to frame the doctrine against a background conception of ‘good’ law and government which the Singapore courts would be more amenable to—for example, the promotion of rules-based administration and promotion of trust—thus presenting an opportunity for the doctrine to be accepted in some form in Singapore administrative law.

Conclusion

In sum, this article has sought to argue that the concept of ‘good administration’, as a normative justification for the doctrine of legitimate expectations, is more problematic than has been commonly appreciated. While the existing literature has focused on the ambiguity as to what exactly ‘good administration’ entails, this article has proposed that beyond this level of contention, there is yet another deeper underlying level of contention—indeed, conceptions of ‘good administration’ are inextricably tied to conceptions of the ‘good’ in relation to the proper role of law and government. Accordingly, contentions in relation to these conceptions of the ‘good’ are introduced into the doctrine of legitimate expectations insofar as ‘good administration’ forms the normative justification for the doctrine.

This article has further sought to draw out the implications of this argument—if correct, this does not necessarily entail giving up upon ‘good administration’ as a justification for the doctrine. Indeed, greater clarity as to the true crux of disagreement provides an opportunity to resolve the debate at a different level. Instead of being limited to articulating norms of ‘good administration’ with as much clarity as possible, one can seek to articulate the most suitable conception of the ‘good’ in law and government that can in turn ground a vision of ‘good administration’ justifying the doctrine. The article has provided a preliminary attempt in this regard. It is hoped that this article will be a useful stepping stone for further analysis of the normative foundation of the legitimate expectations doctrine.

Footnotes

Acknowledgments

The author is very grateful to Professor Mark Findlay, Professor of Law, Singapore Management University, for his insightful and valuable comments on an earlier draft. Any mistakes or omissions remain the author’s own.

Conflict of interest

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Kenny Chng