The Relationship Between Constitutional Equality and Substantive Review

Abstract

General equality rights in written constitutions – rights stating the ideal of equality without specifying categories of impermissible differentiation – have often been effected through the idea of equality as rationality. Equality as rationality demands that differentiations between like entities have to be rationally justifiable. Such equality rights are applicable to legislation and executive action. This presents a prima facie overlap with substantive review in common law administrative law, since substantive review is also concerned about the rational justifiability of executive action. This raises three questions: (1) Are both sets of legal principles indeed similar? (2) Have courts managed to distinguish them in practice? (3) If not, then given that both sets of legal principles exist at different levels in the legal order, how can their similarity be rationalised? This article will study these questions, drawing upon Hong Kong and Singapore law as test cases.

Introduction

The idea that humans, despite their obvious disparities, are all fundamentally equal in a moral sense possesses immense political and cultural salience. Indeed, the ideal of equality is regularly invoked in public discourse. This ideal of equality has also been granted constitutional status in jurisdictions around the world, having been enshrined in a wide variety of written constitutions. Such constitutional rights to equality generally come in two main forms. The first type are general equality rights, which state in general terms the ideal of equality – for example, ‘All Hong Kong residents shall be equal before the law.’¹ The second type states specifically the categories of impermissible differentiation – for example, ‘there shall be no discrimination against citizens of Singapore on the ground only of religion, race, descent or place of birth in any law’.²

In common law jurisdictions possessing written constitutions such as Hong Kong and Singapore, such constitutional guarantees of equality exist alongside legal principles which also serve to shape and restrain government power – specifically, common law administrative law principles. One such set of principles can be described as principles of substantive review. The term ‘substantive review’ is capable of various interpretations, but for the avoidance of doubt, the term as it will be used in this article refers to types of judicial review grounds in administrative law that are targeted at executive action made within the proper scope of power, but are problematic on their merits. Put another way, such grounds of review allow judges to go beyond the process by which executive decisions were rendered and to perform a limited review of their substance, in order to evaluate whether such decisions are rationally unjustifiable. The precise rubric by which the rational justifiability of decisions is assessed can differ. Proportionality or Wednesbury unreasonableness are common heuristic devices serving such purposes.

This article is concerned with the following question: what is the relationship between constitutional rights to equality and common law substantive review? The framing of this question may seem rather startling. Indeed, constitutional equality and substantive review appear very different on their face. The relevance of this question may become clearer, however, when one notes that general equality rights in particular have often been interpreted by courts as a requirement that differentiations effected by the government have to be rationally justifiable, and that general equality rights are also applicable to executive action. Stated thus, the prima facie conceptual similarity between both legal principles becomes obvious. Several further issues arise for discussion. Are both sets of legal principles indeed substantively similar? Have courts managed to adequately distinguish them in practice? If they are indeed similar, then given that both sets of legal principles exist at different levels in the hierarchy of legal norms, how can their similarity be properly rationalised?

The objective of this article is to explore these questions, with an overall view towards elucidating the relationship between general equality rights in constitutional law and substantive review in administrative law. In view of this objective, this article will draw primarily upon common law jurisdictions with written constitutions and which possess a body of administrative law principles rooted in the heritage of English administrative law. Specifically, Hong Kong and Singapore law will be studied as test cases in this article's investigation. While the test cases to be analysed are limited to these jurisdictions in view of space constraints, it is worth noting that the implications of this study are capable of extending to any common law jurisdiction possessing general equality rights in constitutional law. Future research can explore the expansion of the analysis in this article to other common law jurisdictions.

The article will proceed in the following manner. It will first briefly discuss how judges have interpreted and applied constitutional guarantees of equality – with a focus on general equality rights – and will highlight the key features of substantive review in administrative law. With the background thus set, it will argue that there is a prima facie conceptual similarity between both sets of legal principles. The article will then move to study how courts in Hong Kong and Singapore have analysed the respective general equality rights in their jurisdictions. This section will illustrate that there is indeed a substantial similarity in practice between the requirements of general equality rights and substantive review in administrative law in both jurisdictions. The article will conclude by proposing and critically evaluating various modes by which the relationship between the two sets of legal principles can be rationalised. It will argue that the least problematic way of rationalising the relationship between the two sets of legal principles would be to effectuate general equality rights primarily by way of the concept of equality as non-discrimination.

General Equality Rights And Substantive Review

To set the background for the analysis to follow, this section will briefly describe the legal requirements of general equality rights in constitutional law and substantive review in administrative law, and will propose that both sets of legal principles are prima facie conceptually similar.

The Legal Requirements of General Equality Rights

Equality is a contested concept. Indeed, the question of what equality requires raises complex and controversial philosophical issues. Eminent thinkers have engaged thoroughly with these issues,³ and it is not my intent to venture into this philosophical debate, interesting as it is. The primary concern of this article, rather, is how the concept of equality has been effected in legal doctrine.

There are two broad approaches that have been taken towards the effectuation of the ideal of equality in law. These can be characterised as equality as non-discrimination and equality as rationality.⁴

Equality specified as a requirement of non-discrimination is primarily focused on the protection of vulnerable social groups.⁵ It has also been described as an anti-subordination vision of equality. Such a vision of equality would be directed at harnessing the law to ameliorate the systemic disadvantages of social groups which lack political power and to protect them from subordination.⁶ This ideal of equality in law is commonly effected by identifying certain bases of differentiation which are particularly odious and rendering them constitutionally impermissible. Such an approach therefore makes obvious sense where the relevant constitutional equality right already spells out specifically the categories of impermissible differentiation – for example, Article 12(2) of the Singapore Constitution forbids classifications on the basis of religion, race, descent, or place of birth only.⁷ However, such an approach has also been adopted even where general equality rights – the focus of this article – are concerned. For instance, although the US Fourteenth Amendment is a general equality right, the US Supreme Court has articulated a set of suspect classifications which attract a higher degree of judicial scrutiny when triggered – most notably, differentiations on the basis of race and alienage in certain circumstances will attract the highest level of scrutiny.⁸

While general equality rights are indeed capable of being effected in law by way of equality as non-discrimination, such rights are most commonly effected through the idea of equality as rationality. Equality as rationality has been described as an anti-classification principle, bearing a close connection with the Aristotelian imperative that like ought to be treated alike – indeed, equality as rationality has been described as the principle that likes ought to be treated alike unless there is an adequate justification for a departure.⁹ Described as such, the aspect of justification is clearly central to this vision of equality, leading some commentators to characterise such a picture of equality as being at its core a requirement of means-ends rationality – a requirement that the means adopted by government actors are rationally connected to the ends or purposes that they are intended to further.¹⁰

Scholars have argued that equality as rationality is central to the idea of equality. Indeed, Jeffrey Jowell argued that equality fundamentally ‘requires government not to treat people unequally without justification’, suggesting that this is ultimately justifiable by way of basic democratic principles: ‘[i]t gains its ultimate justification from a notion of the way individuals should be treated in a democracy.’¹¹ In a similar vein, Trevor Allan drew a link between equality, which requires distinctions to be ‘capable of a reasoned justification’, and the ideal of the rule of law, which in Allan's view requires ‘governmental action to be rationally justified in terms of some conception of the common good’.¹² Peter Westen argued against construing equality in terms of equality as rationality, on the ground that equality as rationality can be described independently of any reference to equality or the requirement that likes ought to be treated alike, since it is simply anchored on the substantive principle that government actors must have rational and legitimate reasons for the way they treat people.¹³ Nevertheless, Westen's argument is indicative of the close connection that scholars and judges have commonly drawn between equality and the idea of rational justification as a matter of practice.

Equality as rationality is often substantiated in legal doctrine by way of a two-stage test: first, has there been a differentiation? Second, is the differentiation a justifiable one?¹⁴ For present purposes, the point to be emphasised is the centrality of rational justification under such an approach towards specifying the legal requirements of equality. Indeed, the requirement that government action must be rationally justifiable is similarly central to substantive review in administrative law, which we will now turn to describe.

What is Substantive Review?

Substantive review is a type of judicial review ground in administrative law. To be clear, for the purposes of this article, the term ‘substantive review’ will be taken to refer to types of judicial review grounds which are directed at government decisions made within the proper scope of power, but which are problematic on their merits. In other words, such judicial review grounds are capable of evaluating the substance of government decisions for their rational justifiability. They can require the court to assess the weight and balance which a decision-maker has accorded to factors relevant to the pursuit of permissible purposes – should the court determine that the decision-maker has accorded an unreasonable weight to certain factors, it can render the decision unlawful.¹⁵ Two specific judicial review doctrines are structured along such lines: irrationality and proportionality. Each of these doctrines will be discussed in turn.

Irrationality is one of the three classic grounds of judicial review described in Lord Diplock's landmark decision of Council of Civil Service Unions v Minister for the Civil Service.¹⁶ It is widely-accepted as a ground of review across the common law world, including in Hong Kong and Singapore.¹⁷ The classic formulation of irrationality is that a decision is unlawful if it is so unreasonable that no reasonable authority would have come to the same decision.¹⁸ While this formulation is stated in general terms, courts have been willing to vary its requirements in accordance with the context.¹⁹ For instance, where fundamental rights are concerned, irrationality has been applied with particular rigour, such that only compelling public interests would be capable of justifying infringements upon rights, with greater intrusions calling for more by way of justification.²⁰ Where economic policy or public finance is concerned, however, courts have taken a less rigorous stance – a decision would be rendered unlawful only if a government authority had acted in bad faith, for an improper motive, or had acted in such an absurd manner that he must have taken leave of his senses.²¹

Conceptually, irrationality review is framed as narrowly as it is because it is capable of rendering unlawful decisions which had been made within the proper scope of power – in other words, decisions which would not otherwise be ultra vires,would have been made for proper purposes, and with regard to the relevant considerations required by statute.²² As a ground of judicial review, it is therefore capable of intruding significantly into executive discretion, and has to be narrowly framed in order to maintain the principle of separation of powers. Notably, Paul Craig has argued that for irrationality to be relevant and distinctive as a ground of judicial review, it has to be capable in principle of going beyond review of the purposes of government action or the relevancy of considerations taken into account – indeed, irrationality review, in his view, fundamentally requires a judicial assessment of the weight and balance which the decision-maker has accorded to relevant factors in order to determine if the relative weight accorded to the various factors is reasonable by reference to the objective of the action.²³ In contrast, other grounds of judicial review – such as relevancy of considerations – would not enter into the weight accorded to various factors, as long as all material considerations have been taken into account.²⁴

As for the doctrine of proportionality, it has grown to become a widely-accepted heuristic by which to evaluate the constitutionality of rights-infringing government action. The doctrine provides for a structured framework of analysis. While the specifics of implementation can differ across jurisdictions, it generally comprises four stages of analysis: (1) is the action's objective important enough to justify limiting a fundamental right? (2) are the measures adopted to meet this objective rationally connected to it? (3) are the measures adopted no more than necessary to accomplish the objective? (4) has a fair balance been struck between individual rights and the interests of the community?²⁵ The framing of proportionality analysis suggests that it is principally directed at evaluating rights infringements – indeed, where proportionality is applied to executive discretion more broadly in the absence of rights infringements, it will require a more abstract weighing of diverse interests and will become more challenging to apply. Accordingly, it has been suggested that the intensity of proportionality review in non-rights-related contexts ought to be calibrated accordingly.²⁶ The doctrine of proportionality has not yet been accepted in UK domestic law as a standalone ground of judicial review of administrative action, although it is the principal analytical tool to be applied where Human Rights Act violations are concerned.²⁷

At first glance, there may appear to be obvious differences between both grounds of review. Indeed, the structured framework of proportionality analysis, in comparison to the more broadly-worded principle of irrationality, has led many to take the view that proportionality is a more exacting test requiring more by way of justification.²⁸ Further, as mentioned above, the precise formulation of proportionality analysis lends itself specifically to analysis within the context of rights infringements, while irrationality is more readily applicable even to non-rights-related contexts.²⁹

The relationship between both grounds of review and their relative normative desirability has been subject to a lively and illuminating academic discussion. But the key point to be emphasised for the purposes of this article is simply that both irrationality and proportionality share certain features which cause them to fit well within the category of substantive review. It is clear from the very terms of proportionality analysis that it requires the court to assess the balance that a decision-maker has struck between relevant competing considerations in the pursuit of his objective. As has been highlighted above, irrationality review also similarly requires considerations of weight and balance. Indeed, the UK Supreme Court in Pham v Secretary of State for the Home Department and Kennedy v The Charity Commission has observed that considerations of weight and balance are taken into account across both irrationality and proportionality.³⁰ Craig has also suggested that as a matter of practice, the manner in which courts have actually applied irrationality review incorporates an assessment of necessity and suitability in determining the reasonableness of government action – components of assessment which can also be found in the proportionality framework.³¹ Rebecca Williams made a similar point through a careful analysis of case law – in essence, she argued that both grounds of review are ultimately about means-ends rationality and possess a similar structure incorporating assessments of suitability, necessity, and fair balance.³² This meant, in her view, that it is inaccurate to think that the intensity of review is tied to the ground of judicial review that is applied – rather, it is the subject matter of the cases which plays a more significant role in shaping the intensity of review.³³

Equality as Rationality and Substantive Review

It is clear therefore that despite the different formulations of irrationality and proportionality, both grounds of review are broadly centred around a requirement that decisions of government authorities have to be rationally justifiable, and that both grounds of review require a form of means-ends rationality analysis as a means of furthering this requirement.³⁴ Returning to our discussion of equality, it should be readily apparent then that there is a similarity in concept between equality as rationality and substantive review. Indeed, equality as rationality is similarly centred around a requirement of rational justification, and it also requires analysis of the rational connection between the means adopted and the ends pursued.

Eminent scholars and jurists have made this connection. For instance, as a conceptual matter, Guy Lurie argued that the Aristotelian formula that like cases ought to be treated alike unless there is a justification for differentiation is ultimately a ‘relational means-ends test or measure, much like proportionality.’³⁵ Accordingly, in Lurie's view, it is at least partly redundant to use a proportionality analysis to assess the violation of an equality right – the determination of whether there had indeed been a violation of an equality right already involves an assessment similar to that of proportionality.

Judges have also expressly acknowledged this close relationship. For example, in the Privy Council decision of Matadeen v Pointu, Lord Hoffmann suggested that ‘treating like cases alike and unlike cases differently is a general axiom of rational behaviour’, noting at the same time that it does not necessarily always fall to the courts to ‘have the last word on whether the principle has been observed’.³⁶ Indeed, Lord Lester has noted that the manner by which the law has effected the idea of equality is to require ‘objective and proportionate justification for apparently unequal treatment’.³⁷ Most directly, Lady Hale suggested that ‘the principle that like cases should be treated alike is part of the public law doctrine of reasonableness or rationality’.³⁸

Scholars such as Jowell and Christopher McCrudden have conducted detailed studies of case law that amply illustrate this point.³⁹ While John Stanton-Ife argued in response to Jowell that the cases that Jowell sought to rely on to make this point were decided on the basis of a means-ends rationality analysis simpliciter rather than any notion of equality,⁴⁰ it ought to be noted that no matter whether Jowell or Stanton-Ife has the more accurate interpretation of case law, the argument made by both authors serves very much to underscore the practical connection between how equality has been effected in law and the central principles of substantive review.

At this juncture, one might argue that the general framework of analysis required by equality as rationality presents an obvious distinction between it and substantive review – it will be recalled that the first step requires a differentiation to be identified. Accordingly, at first glance, one might think that this serves to dispose easily of the suggestion that equality as rationality is conceptually similar to substantive review. Indeed, one could argue that equality as rationality is directed specifically at the rational justifiability of differentiations made by government authorities, while substantive review is not limited in such a manner.

The problem, however, is that this distinction is more tenuous than it may appear. First of all, most, if not all, government action can be characterised as making differentiations. The differentiations are sometimes quite obvious. For example, if a government policy differentiates between individuals who have received Covid-19 vaccinations and those who have not for the purposes of public health-related movement control orders, the relevant differentiation is apparent on the face of the government policy. However, government action necessarily makes differentiations even where the differentiations are not so apparent. Consider a decision by a government authority to compulsorily acquire a plot of land for developmental purposes. It is not apparent on the face of the decision that any differentiation has been made. Yet, there has indeed been a differentiation – the government authority has decided to acquire this plot of land, and not the neighbouring plots, thereby drawing a differentiation between the acquired plot and the others which have not been acquired. The same can be said of a whole plethora of everyday government action – a decision not to grant an applicant a visa entails a differentiation between the unsuccessful applicant and other successful applicants, a decision to ban the public broadcast of a certain kind of material entails a differentiation between such material and other material not subject to such a ban, and so on. The point is that if almost all (if not all) government action can be characterised as making some form of differentiation, then a distinction between equality as rationality and substantive review based on the requirement of a differentiation may be difficult to sustain in practice.

More fundamentally, the identification of a differentiation for the purposes of equality analysis may overlap substantially with an analysis of the justifiability of the differentiation. Charles Maxime Panaccio has argued that while the Supreme Court of Canada has insisted on drawing an express distinction between the identification of a differentiation triggering an equal protection concern and the justifiability of the differentiation, both stages substantially require the same form of analysis – indeed, he argued that the identification of a differentiation triggering an equal protection concern is not a mere factual inquiry but involves an assessment of whether the distinction is morally problematic in a manner that would trigger an equal protection concern.⁴¹ Accordingly, Panaccio argued that both stages of equality analysis in truth require an analysis of the same sort of considerations, and that there is little sense in keeping both stages compartmentalised. In his view, no matter which stage of the analysis is being expressly invoked, ‘the crucial issue is justification all things considered … the government has to justify its actions.’⁴² In a similar vein, Stanton-Ife has argued that ‘[n]o clear advantage emerges to assuming first a right to equality and secondly, a test of justification. The first idea can be excised without loss, which leaves us again with the notions of reasonableness, rationality and the rule of law…’⁴³ If the very exercise of identifying a differentiation triggering the right to equality is inextricably linked to an analysis of its reasonableness, then, again, it will be challenging to draw a distinction between equality as rationality and substantive review on the basis of the need to identify a differentiation. This is an insight that the Hong Kong courts have also arrived at in their analysis of the constitutional right to equality, as will be described later.

In view of the preceding discussion, it is clear that there is a close conceptual connection between equality as rationality and substantive review in administrative law. This bears interesting implications for common law constitutional jurisdictions with general equality rights in their written constitutions. These general equality rights in constitutional law are commonly effectuated by way of equality as rationality. Such jurisdictions also generally possess a body of common law administrative law containing the principles of substantive review. This body of administrative law exists at a lower level in the hierarchy of legal norms as compared to constitutional law. General equality rights and principles of substantive review are both applicable to executive action.

Given then the overlap in application and conceptual similarity between equality as rationality and substantive review in administrative law, the question is whether such common law constitutional jurisdictions have managed to rationalise the relationship between general equality rights and substantive review in a coherent fashion as a matter of practice. A further question also arises: what should be the relationship between both sets of legal principles, in view of their similarity in substance and their existence at different levels in the legal order?

These questions will be explored in the subsequent sections. By way of methodology, two common law constitutional jurisdictions will be studied as test cases – Hong Kong and Singapore. Both jurisdictions are common law jurisdictions with general equality rights in their written constitutions. Both jurisdictions also possess a body of administrative law principles rooted in the heritage of English administrative law. This means that even though there are differences in the specific doctrines of substantive review which are adopted in each jurisdiction, both jurisdictions do accept the broad principles of substantive review.⁴⁴ These traits make both jurisdictions useful test cases for the purpose of evaluating the questions raised in the preceding paragraph. It ought to be noted, however, that the implications of this study are well capable of extending to any common law jurisdiction possessing general equality rights in constitutional law. Future research unconstrained by space can expand the analysis in this article to other common law jurisdictions.

General Equality Rights And Substantive Review In Practice

Hong Kong

Hong Kong is a common law jurisdiction adhering to the doctrine of constitutional supremacy. Indeed, Hong Kong's Basic Law and Bill of Rights Ordinance enjoy constitutional status in Hong Kong.⁴⁵ Hong Kong's Basic Law in particular contains a general right to equality – Article 25 of the Hong Kong Basic Law provides that ‘All Hong Kong residents shall be equal before the law.’

The general approach that the Hong Kong courts have taken towards the analysis of constitutional infringements is to first consider whether a constitutional right had been infringed, and then to evaluate whether the infringement was justifiable through the usage of proportionality analysis.⁴⁶ In applying the test of proportionality, the Hong Kong courts have essentially adopted the four-step framework of analysis described earlier.⁴⁷ The Hong Kong Court of Final Appeal, the final appellate court in Hong Kong, clarified in Fok Chun Wa v Hospital Authority (‘Fok Chun Wa’) that in applying the doctrine of proportionality, the court would not ‘put itself in a place of the executive or legislature or other authority to decide what is the best option’, but that the court will be ‘particularly stringent or intense’ in its review if ‘fundamental concepts or core-values’ are implicated.⁴⁸

The Hong Kong courts have applied the same general approach to analyse purported infringements of the constitutional right to equality. Indeed, the Hong Kong Court of Final Appeal in Secretary for Justice v Yau Yuk Lung (‘Yau Yuk Lung’) held that the proper approach was to first consider whether a differentiation had occurred such that like had not been treated alike, and then to determine whether the differentiation satisfied the test of justification.⁴⁹

A striking point to observe about the manner by which the Hong Kong courts have applied this two-step test is that the courts have emphasised that a sharp distinction ought not to be drawn between both stages of the test. The consequence of this perspective towards the general right to equality in Hong Kong law is that the test governing constitutional equality has substantially centred around the test of justification.

The following cases will serve to illustrate this point. In the Hong Kong Court of Final Appeal decision of Fok Chun Wa, a constitutional challenge was brought against the hospital authority's decision to reclassify certain non-residents of Hong Kong who were spouses of Hong Kong residents as persons not eligible to receive subsidised medical treatment at public hospitals – specifically, Mainland Chinese women who possessed two-way permits and who had applied for one-way permits to stay in Hong Kong.⁵⁰ This challenge was brought under the equality provisions of the Hong Kong Basic Law and the Bill of Rights Ordinance. The applicants argued that Mainland Chinese spouses of Hong Kong residents living in Hong Kong who had applied for one-way permits were for all intents and purposes identical to Hong Kong resident women with Hong Kong resident spouses. Differentiating between these two categories of persons by allowing the latter to enjoy subsidised medical treatment while denying the same to the former therefore amounted to discrimination.

In addressing this challenge, the Court of Final Appeal first affirmed the applicable approach as set out in Yau Yuk Lung – the proper approach was first to ‘identify the comparators’, determine whether the compared persons were in ‘comparable positions’, and then evaluate whether any differences in treatment between comparable persons were justifiable.⁵¹

The Court of Final Appeal, however, expressed concern about the problems that may ensue if such an approach were applied in an overly rigid fashion. Indeed, the court suggested that the main objective that courts should remain focused on in dealing with equal protection challenges is whether there was ‘enough of a relevant difference between X and Y [the comparators] to justify differential treatment’.⁵² This was because in most cases, the two stages of analysis would not be separable from each other, and the first step of whether the compared persons were indeed in like situations would generally have to be answered by considering whether the differential treatment was justifiable. Accordingly, the court thought that ‘it may matter not at all whether the court's approach is seen as a two-stage one or not’,⁵³ and that ‘in most questions involving the right to equality, there will be an overlap in the application of the two-stage test set out in Yau Yuk Lung’.⁵⁴

In assessing the equal protection challenge before it, the court therefore went straight to the question of whether the differentiation between the groups of persons being compared was justifiable.⁵⁵ The court noted the importance of granting a margin of appreciation, ‘particularly in circumstances where the court is asked to examine issues involving socio-economic policy’.⁵⁶ In circumstances involving the allocation of limited public funds, the Court of Final Appeal held that the ultimate decision-maker ought to be the executive or legislature.⁵⁷ Indeed, the court held that decisions to demarcate who would be eligible for publicly subsidised healthcare services were ‘part of the Government's socio-economic responsibilities’, and that ‘it is no part of the court's role to second-guess the wisdom of these policies and measures’.⁵⁸ Accordingly, since the differentiation based on residence status was ‘within the spectrum of reasonableness’, the court ultimately rejected the constitutional challenges.⁵⁹

The landmark equal protection decision of the Court of Final Appeal in QT v Director of Immigration ⁶⁰ (‘QT’) further illustrates the point that the Hong Kong courts have construed the right to equality substantially as a requirement of rational justification. In QT, the applicant had entered into a same-sex civil partnership in England. Her partner was granted an employment visa in Hong Kong, but the applicant was denied a dependant visa for the reason that she was not in a heterosexual monogamous marriage. She subsequently applied for judicial review against the Director of Immigration to challenge this decision, on the basis that she had been unlawfully discriminated against for her sexual orientation.

In addressing this challenge, the Court of Final Appeal discussed once again the two-step test governing equality-based challenges articulated in Yau Yuk Lung. Notably, the court recognised that a critical drawback of relying too heavily on the first step of the test is that the court risked being drawn into the problem of identifying what precisely the proper comparator ought to be to determine whether the compared classes were indeed alike in the first place. The issue with an analysis along such lines is that it is difficult to resolve in a principled manner – indeed, ‘the notion of whether the comparators are analogous or relevantly similar is elastic both linguistically and conceptually’.⁶¹

As such, the court in QT affirmed the approach it had taken in Fok Chun Wa, holding that the two stages of the Yau Yuk Lung test were intricately connected – ‘where an issue of equality before the law arises, the question of whether a measure is discriminatory is necessarily bound up with whether the differential treatment which the measure entails can be justified.’⁶² Indeed, the Court of Final Appeal thought that the proper approach would be to ‘examine every alleged case of discrimination to see if the difference in treatment can be justified’.⁶³ Therefore, in dealing with the equal protection challenge at hand, the Court of Final Appeal addressed directly the question of whether the Director's decision was rationally justifiable, ultimately holding that the problem with the Director's decision was that it was not rationally connected with any legitimate objective and upholding the equal protection challenge against it.

Other Hong Kong Court of Final Appeal decisions have adopted the same approach towards challenges based upon the constitutional right to equality in Hong Kong.⁶⁴ Without belabouring the point further, the foregoing discussion should suffice to make clear that the Hong Kong courts have interpreted the general equality right in a manner practically synonymous with the test of proportionality. This has important implications for the point made earlier regarding the possibility of differentiating the legal approach to general equality rights and substantive review on the basis that the former requires a prior step of identifying the relevant differentiation. The problems with this suggestion are substantiated by the Hong Kong courts’ experience – indeed, the Hong Kong courts have highlighted the problem of getting enmeshed in an interminable debate over what the appropriate comparator ought to be for the purposes of determining whether like cases have been treated alike, and have suggested that such analysis shades anyway into a test of justification.

The jurisprudence of the Hong Kong courts is also instructive for another reason. The Hong Kong courts have generally drawn a distinction between constitutional and administrative law on the basis that constitutional review triggers the test of proportionality, while ordinary administrative law review would trigger only the doctrine of irrationality. The justification for such a bifurcation is the perception that proportionality review requires a more intensive standard of review than the doctrine of irrationality, thereby making it more fitting for application within the context of prima facie infringements of constitutional rights. One might argue therefore that this provides another possible basis for a distinction between the legal approach taken towards general equality rights in constitutional law and substantive review in administrative law, at least in Hong Kong.

The Court of Final Appeal's decision in QT, however, challenges the tenability of this bifurcation. The challenge in QT to the Director of Immigration's decision was framed as a judicial review challenge in administrative law, not a constitutional challenge. Yet, the Court of Final Appeal, acknowledging that this case ultimately revolved around the concept of equality in law, decided to apply the test of proportionality to the case at hand. Indeed, the court held that ‘the proportionality concepts developed in constitutional law … are equally applicable to deciding whether the differential treatment entailed by the Policy is justified or whether it may be impugned as Wednesbury unreasonable.’⁶⁵ This suggests that the proposed boundary between general equality rights in constitutional law and substantive review in administrative law based upon the applicable doctrine of review may be less distinct than one may have originally surmised. This point will be returned to subsequently.

All in all, the picture which one obtains of Hong Kong jurisprudence is that of a very close connection between the legal approach taken towards general equality rights and substantive review in administrative law. While ostensible boundaries have been drawn between both sets of legal principles, these boundaries have proven to be somewhat porous.

Singapore

We turn then to study whether and how the Singapore courts have rationalised the relationship between the legal approach to general equality rights in constitutional law and substantive review in administrative law. By way of background, Singapore is a constitutional democracy, and a right to equal protection of the law is enshrined in Article 12 of its constitution.⁶⁶ Article 12(1) is a general equality right which reads as follows: ‘All persons are equal before the law and entitled to the equal protection of the law.’⁶⁷ In contrast to Hong Kong, the Singapore courts have not accepted the doctrine of proportionality in either constitutional or administrative law.⁶⁸ Accordingly, substantive review in Singapore is for the most part associated with the doctrine of irrationality in administrative law.

The Singapore courts have articulated a variety of doctrinal tests governing the application of Article 12(1) to government action. One of the leading tests was elaborated upon in the decision of the Singapore Court of Appeal in Lim Meng Suang v Attorney-General ⁶⁹ (‘Lim Meng Suang’). In this case, the Singapore Court of Appeal, the final appellate court in Singapore, was faced with a constitutional challenge to section 377A of Singapore's Penal Code, which criminalised homosexual sexual acts between men in Singapore. One of the bases of challenge was Article 12(1) of the Singapore Constitution.

Confronted with this challenge, the Court of Appeal took the opportunity to articulate the governing approach for constitutional challenges to legislation based on Article 12(1). The court affirmed that the prevailing approach was the ‘reasonable classification’ test. This test had two main parts: a statute prescribing a differentiation would be constitutional by reference to Article 12(1) only if the legislative classification was ‘founded on an intelligible differentia’ and if the differentia bore ‘a rational relation to the object sought to be achieved by the statute’.⁷⁰

Describing the test in more detail, the Court of Appeal held that the ‘intelligible differentia’ limb provided for a relatively low threshold in that all that was required was that the differentiation was capable of being understood.⁷¹ The court further held, however, that the requirement of intelligible differentia would also extend to invalidate differentiations which, while capable of being understood, were nevertheless ‘so unreasonable as to be illogical and/or incoherent’, such that ‘there can be no reasonable dispute (let alone controversy) as to that fact from a moral, political and/or ethical point of view’.⁷² For instance, ‘a law which bans all women from driving’ would be intelligible in the sense that the relevant differentiation could be understood, but would nevertheless fall foul of the requirement of intelligible differentia on the basis that it was manifestly unreasonable.⁷³

As for the second limb of the test, what it required was a rational relation between the differentiation in question and the purpose of the statute at hand – there was no need for a perfect relation between the two, as long as it was a rational one.⁷⁴ Notably, the Court of Appeal disagreed with the suggestion of the court below that in circumstances where the object of the legislation was itself illegitimate, legislation could be rendered unconstitutional even if the reasonable classification test was passed.⁷⁵ The Court of Appeal opined that allowing the court to do so would ‘confer on the court a licence to usurp the legislative function’.⁷⁶ Applying all of this to the case at hand, the court ultimately found that section 377A was constitutional – the differentia based on male homosexual sexual acts was found to be an intelligible one, and there was a rational connection between this differentiation and the purpose of section 377A, which was construed as the disapprobation of male homosexual sexual acts even if conducted in private.⁷⁷

The reasonable classification test as articulated by the Court of Appeal has been subjected to detailed academic critique.⁷⁸ Without delving in detail into the merits of the test, the key point to emphasise for present purposes is that both limbs of the test are analogous to the requirements of substantive review. The manner by which the Court of Appeal substantiated the ‘intelligible differentia’ limb of the test bears close resemblance to the central inquiry of the doctrine of irrationality. And the court's elaboration of the second limb of the reasonable classification test makes it essentially identical to a means-ends rationality test – one of the key features of substantive review, as described earlier. Further, even though the reasonable classification test in Lim Meng Suang appears on its face to be distinct from substantive review on the basis that it is specifically directed only at differentiations within legislation, one ought to recall that it is quite possible to frame almost all government action as requiring some sort of differentiation. Indeed, as general rules of conduct specifying legal rights and obligations, much of legislation (if not all) necessarily makes distinctions between persons and conduct captured within their provisions and those which are not.

Moving on from Lim Meng Suang, in the recent Singapore Court of Appeal decision of Syed Suhail bin Syed Zin v Attorney-General ⁷⁹ (‘Syed Suhail’), the court set out another doctrinal approach governing constitutional challenges based upon Article 12(1). This case concerned a constitutional challenge to executive action, unlike Lim Meng Suang. In Syed Suhail, a challenge was made to the scheduling of a prisoner's impending execution. One of the applicant's arguments was that the scheduling of his execution ahead of other prisoners, despite having been sentenced to death later, amounted to an equal protection violation. This argument was based, inter alia, on the ground that a failure to schedule executions in the order that prisoners were sentenced to death would deprive prisoners of having a fair amount of time to adduce new evidence to challenge their convictions.

Confronted with this challenge, the Court of Appeal took the opportunity to set out a new approach governing equal protection challenges to executive action, replacing the hitherto-prevailing ‘deliberate and arbitrary discrimination’ test.⁸⁰ In its place, the court introduced a new two-stage test: the applicant had to first prove that he had been treated differently from ‘equally situated’ persons, and the decision-maker in question would then have to prove that this differential treatment was ‘based on legitimate reasons’.⁸¹ The second stage of the test would involve a consideration of whether the differential treatment had ‘a sufficient rational relation to the object for which the power was conferred’.⁸²

Applying this new approach, the Court of Appeal concluded that there had indeed been differential treatment which needed justification, since the applicant's execution had been scheduled ahead of another prisoner even though the applicant had been sentenced to death later than the other prisoner.⁸³ The Court of Appeal accordingly granted leave for judicial review in relation to this decision to proceed at the High Court.⁸⁴ The High Court's subsequent decision on the merits of the judicial review application concluded that the application failed at the first stage – he was not equally situated with the other prisoners being compared to since he had no realistic expectation of his case being potentially reopened on the merits, while the other prisoners did.⁸⁵

A natural question may arise at this juncture: what is the difference between the Syed Suhail and Lim Meng Suang tests? As a matter of practice, the difference may simply rest in their applicability to different contexts – Lim Meng Suang is targeted at the constitutionality of legislation under Article 12(1), while Syed Suhail is directed at the constitutionality of executive action under the same provision. As a matter of substance, however, the recent Court of Appeal decision in Tan Seng Kee v Attorney-General ⁸⁶ addressed the issue briefly without coming to a firm conclusion, suggesting that a key substantive difference between the two tests, inter alia, lay in the fact that the Lim Meng Suang test allowed an assessment of the reasonableness of the relevant differentiation at the first limb of the reasonable classification test, while the assessment of reasonableness in the Syed Suhail test came only at the second stage.⁸⁷ The even more recent Court of Appeal decision of Attorney-General v Datchinamurthy a/l Kataiah suggested along similar lines that the first stage of the Syed Suhail test was merely ‘a factual one of whether a prudent person would objectively think the persons concerned are roughly equivalent or similarly situated in all material respects’, and that it was only at the second stage that the court would assess whether the differential treatment was reasonable.⁸⁸

Returning to the issue under interrogation in this article, the key point to observe from this discussion is that the Syed Suhail test is also quite difficult to distinguish from substantive review. Indeed, even if the first stage of the Syed Suhail test requires the factual identification of comparators or differentiations which will then be assessed for their reasonableness, this does not serve as a satisfactory distinction from substantive review – most government action can be easily framed as providing for some sort of differentiation. Accordingly, the crux of the analysis required by the Syed Suhail test rests at the second stage – which is simply a straightforward means-ends rationality assessment similar in kind to what is required under substantive review.

The Court of Appeal in Syed Suhail appeared to be cognisant of this strong connection between the doctrinal requirements of Article 12(1) and substantive review. Seeking to distinguish the approach applicable to Article 12(1) and the doctrine of irrationality, the court held that while irrational acts of the executive branch generally would fall to be reviewed through the administrative law doctrine of irrationality, actions which are ‘impermissibly discriminatory in nature’ would fall under the scope of Article 12(1) instead.⁸⁹ The Court of Appeal emphasised that the two approaches should not be conflated.⁹⁰ Indeed, the court was expressly concerned about the consequences of such conflation – specifically, that such conflation would mean that ‘all executive action which could be challenged under Art[icle] 12(1) would only be vulnerable to challenge under the ordinary grounds of judicial review, and this would render Art[icle] 12(1) nugatory so far as it related to executive action’.⁹¹

However, the Court of Appeal's proposed distinction between the approach applicable to Article 12(1) and the doctrine of irrationality in administrative law is, with respect, quite unsatisfactory. Indeed, what is an ‘impermissibly discriminatory’ decision? If it refers to any decision which involves some form of factual differentiation, then, as mentioned earlier, it is easy to frame any government action as requiring factual differentiation. If it refers to decisions involving factual differentiations which are legally problematic, then the problem is that the manner by which the courts have determined which differentiations are legally problematic under Article 12(1) is precisely a test of rational justification.

On either interpretation, therefore, the Syed Suhail approach towards Article 12(1) may be quite challenging to distinguish from substantive review in administrative law. Such a conflation would indeed be problematic, as the Court of Appeal in Syed Suhail suggested, but perhaps in a different sense from what the court had in mind: if Article 12(1) is quite indistinguishable in substance from substantive review, the issue is that any challenge against executive action would be easily capable of being framed as a constitutional challenge based on Article 12(1).

Overall, this survey of Singapore jurisprudence suggests that while the Singapore courts have been cognisant of the need to differentiate the legal approach towards general equality rights in constitutional law from substantive review in administrative law, they have struggled to articulate a clear and tenable distinction between the two sets of legal principles. In other words, a similar situation as that in Hong Kong obtains in Singapore. Indeed, the conflation between both sets of principles may be even more pronounced in Singapore as compared to Hong Kong. While Hong Kong jurisprudence can ostensibly rely on the usage of the doctrine of proportionality where constitutional equality is concerned to draw a bifurcation between general equality rights and substantive review in administrative law – a bifurcation that may in the final analysis be rather tenuous, as highlighted above – the Singapore courts do not even have the option of relying on this apparent bifurcation in view of their reluctance to accept the doctrine of proportionality.

Rationalising The Relationship Between General Equality Rights And Substantive Review

The preceding discussion has highlighted that the conceptual similarity between the legal approach to general equality rights in constitutional law and substantive review in administrative law has indeed been reflected in the jurisprudence of common law constitutional jurisdictions, with Hong Kong and Singapore serving as case studies in this regard.

It is suggested that this conflation as a matter of legal doctrine is problematic and undesirable. The fundamental problem with such a conflation is that it results in both sets of legal principles involving substantially the same mode of analysis, despite the fact that they exist at very different levels in the legal order. Some of the issues arising from such a conflation have already been mentioned in the preceding discussion – for one, as mentioned in the discussion of Syed Suhail, one consequence of such conflation is that any challenge to executive action can be quite easily framed as a constitutional challenge under the general equality right.

Accordingly, in view of the finding of conceptual similarity between the legal approach to general equality rights in constitutional law and substantive review in administrative law, it becomes particularly relevant to consider how the relationship between the two sets of legal principles can be rationalised.

The first possible way by which a distinction between general equality rights and substantive review can be made has already been raised in the preceding discussion. One might argue that a distinction can be made between both sets of principles by way of the nature of review applicable under each – for example, proportionality review attaches to prima facie infringements of constitutional equality in Hong Kong while the doctrine of irrationality applies to challenges brought against ordinary executive action generally. However, such a distinction has proven challenging to maintain in practice. As was observed earlier in our evaluation of the Court of Final Appeal decision in QT, both types of review were elided somewhat in the court's analysis. Also, as a conceptual matter, one might wonder whether proportionality and irrationality review are indeed substantially different at all, as Craig's and Williams’ arguments referenced earlier have pointed out.

Further, even if one assumes that there is a meaningful difference between both types of review, how would one determine when each type of review would be triggered? The ostensible answer would be that proportionality review would be triggered where a distinction has occurred which forms a prima facie infringement of the general equality right, while irrationality review would be triggered in other situations. As was argued earlier, however, this answer is less workable than one might have initially surmised. Most, if not all, government action can be framed as requiring distinctions, and the very exercise of determining whether there has been a distinction prima facie infringing upon the general equality right itself overlaps substantially with a test of rational justification.

A second possible means of distinguishing the two sets of legal principles would be to highlight the difference in burden of proof. One might argue that under a general equality right in constitutional law, the burden is on the applicant to prove a prima facie infringement – thereafter, the burden of proof shifts to the government authority in question to justify the infringement. In contrast, under substantive review in administrative law, the burden is on the applicant to prove that the government action in question is unreasonable. Accordingly, the manner by which the law allocates burden of proof renders a constitutional challenge more favourable to an applicant from the perspective of litigation strategy.

It is suggested, however, that while this approach does indeed provide a distinction between the two sets of legal principles on its face, it is quite unsatisfactory as a meaningful difference between a general equality right in constitutional law and substantive review in administrative law. The advantage accorded to constitutional challenges as a matter of burden of proof presumes that there is a valid distinction between a violation of a constitutional general equality right and an unlawful government act under substantive review in administrative law. But the problem precisely is that the principles governing when a violation of a general equality right has occurred are substantially similar to the principles of substantive review. If this is the case, then the conferral of a procedural advantage to applicants alleging violations of constitutional equality as compared to applicants arguing that government action was irrational is quite arbitrary in nature. Accordingly, this proposed basis of distinction between the two sets of legal principles serves to underscore the problem of doctrinal incoherence resulting from conflation, rather than address it.

A third possible mode by which a distinction between general equality rights and substantive review can be made is to effectuate general equality rights primarily through the idea of equality as non-discrimination, rather than the idea of equality as rationality. This ought to make instinctive sense. Indeed, if it is the concept of equality as rationality that is causing the substantial overlap with the principles of substantive review, then the obvious solution would be to rely more upon the alternative approach by which a general equality right can be effected.

An approach by which this could be done would be how the US Supreme Court has interpreted the US Fourteenth Amendment – as mentioned earlier, the US Supreme Court has spelled out certain categories of impermissible classifications such as race and alienage which would trigger a higher standard of review. On this approach, it would be the court's responsibility to articulate the relevant categories of impermissible classifications, and to specify the proper standard of review that would be applied to each category.

There is scholarly support for the view that such a solution would give better effect to the idea of equality than the concept of equality as rationality. Indeed, equality as rationality can be perceived as at best giving effect to a formal vision of equality – that like cases have to be treated alike – and not doing enough to further the idea of substantive equality.⁹² Scholars have argued that an approach to general equality rights based on equality as rationality would inadequately address issues of substantive equality such as equality of opportunity and systemic discrimination of disadvantaged minorities.⁹³ Insofar as equality as rationality does indeed further a vision of formal equality, scholars have argued that it presents a merely procedural understanding of equality which fails to take account of the ‘substantive social outcomes of a law's formal application.’⁹⁴ Rabinder Singh has argued in favour of the importance of distinguishing equality analysis from regular principles of administrative law, suggesting that it is important to formulate a distinct and rigorous discipline of equality analysis which protects democratic principle by safeguarding the rights of otherwise disenfranchised minority groups.⁹⁵ Indeed, if one adopts the perspective that equality is ‘a process of reasoning or a methodology’, directed at helping us think about the ‘larger social context’ and issues of ‘difference, disparity, disadvantage, disproportion and domination’, as Justice Sheilah Martin of the Supreme Court of Canada argued, then one might consider that a legal approach to general equality rights that is difficult to distinguish from substantive review would protect only an impoverished vision of equality.⁹⁶

This approach raises its fair share of problems, however. First, how exactly should a court decide whether a classification ought to be raised to the level of a suspect classification? Indeed, without a principled framework by which the courts can decide how to designate certain classifications as suspect, the concern would be that it will be difficult to limit the proliferation of suspect classifications.⁹⁷ More fundamentally, one may be concerned that classifications which are designated as suspect classifications will simply be a reflection of the moral preferences of the bench – a point that acquires greater force when one notes that whether classifications such as sexual orientation ought indeed be suspect classifications has been the subject of considerable controversy in the public square. Scholars have raised the concern that arrogating such power to the courts would risk the ‘judicial creation of a constitutional bill of rights’.⁹⁸ Perhaps in recognition of these issues, the US Supreme Court has become noticeably hesitant to expand or develop the categories of suspect classifications.⁹⁹ Given the potential breadth of the notion of equality and its moral salience – indeed, a general right to equality may even be interpreted as placing upon government authorities a positive duty to reduce various forms of inequality – such concerns of judicial overreach are accentuated to the extent that judges are called upon to infuse moral content into the legal meaning of equality.

A second issue with such an approach is constitution-specific – some common law constitutional jurisdictions possess written constitutions containing both general equality rights as well as constitutional provisions proscribing certain categories of impermissible differentiation. Indeed, this is precisely the situation in Hong Kong and Singapore.¹⁰⁰ The issue therefore would be that in such jurisdictions, specifying the general equality right by way of equality as non-discrimination may cause the general equality right to overlap with other constitutional provisions which already articulate categories of impermissible differentiation.

It is clear that none of these approaches are perfect – each one presents unique and difficult problems. In these circumstances, it is proposed that the third approach – effecting general equality rights principally through the idea of equality as non-discrimination – is the least problematic of the three. The problems it raises, while important to take into account, are not insurmountable. Indeed, looking at the first issue with this approach raised earlier, one way of characterising the problem is that it is fundamentally a concern about the capacity for judicial overreach. Judicial overreach is not an inevitable outcome of this approach. Framed as such, this problem can be addressed via the provision of a principled legal framework by which suspect classifications can be identified.

This is an endeavour which has already attracted a vigorous and illuminating discussion.¹⁰¹ A particularly noteworthy contribution in this regard comes from Tarunabh Khaitan in his landmark monograph on discrimination law.¹⁰² Khaitan argued that suspect classifications can be identified as suspect on the basis of two cumulative requirements. First, the ground must classify ‘persons into groups with a significant advantage gap between them’.¹⁰³ Observing that the disadvantages suffered by certain groups can come in various forms – political, socio-cultural, and material – Khaitan argued that the advantage gap does not need to be absolute in the sense that the well-being of the disadvantaged group has fallen below a certain objective standard, but rather that relative advantage gaps between a disadvantaged and advantaged group would suffice.¹⁰⁴ Second, the ground must either be ‘immutable’ or constitute ‘a fundamental choice’.¹⁰⁵ This requirement expresses the idea that the ground ought to be ‘normatively irrelevant’ – in other words, ‘our possession of these grounds should not affect how successful our lives are’.¹⁰⁶ Khaitan argued that protection of characteristics which are the result of fundamental choice rests along the same spectrum of protection of immutable characteristics, insofar as changes to fundamental choices such as marital or religious status would be ‘usually impossible without significant personal costs to the individual’.¹⁰⁷

The reason for this elaboration of Khaitan's proposal is to illustrate the detail by which frameworks to identify suspect classifications can be articulated. The point is that the concern relating to how suspect classifications ought to be identified is not a systematic incoherence in principle with this approach, but rather is of the kind which can be mitigated through the careful provision of principled frameworks of analysis. This provides the basis for a response to the second issue with this approach highlighted earlier. Indeed, if a principled framework can be formulated for the judicial articulation of suspect classifications under a general equality right, this means that a general equality right can be distinguished from other constitutional provisions specifying categories of impermissible differentiation on the basis that it would allow judges the flexibility to incrementally develop the types of constitutionally impermissible differentiations, beyond those which have already been expressly identified in the constitutional text.¹⁰⁸

Effecting a general equality right principally by way of the idea of equality as non-discrimination can still incorporate reference to the idea of equality as rationality. One manner by which equality as rationality remains relevant would be the provision of a defence of justification for differentiations falling under suspect classifications, which would test ‘the desirability of the objective sought to be achieved’ against ‘the proportionality and necessity of the discriminatory means employed’.¹⁰⁹ Where constitutional review of legislation is concerned, taking further reference from US Supreme Court jurisprudence, a lower-level and more permissive rational basis review could be applied for differentiations not falling within suspect classifications. Where review of executive action on the basis of a general equality right is concerned, however, a qualification to the incorporation of equality as rationality may have to be made. Indeed, otherwise, an overlap with substantive review in administrative law would resurface. It is suggested therefore that such rational basis review of executive action ought not to be available under general equality rights applied to executive action. Assessments of the rational justifiability of executive action simpliciter ought to be confined to administrative law. This will serve to preserve the distinctiveness of general equality rights in constitutional law as compared to common law administrative law. This may indeed require a significant readjustment to the way general equality rights have been approached – as mentioned earlier, the weight of authority thus far leans in favour of effectuating general equality rights through equality as rationality even where executive action is concerned. But such a readjustment may be necessary for the sake of doctrinal coherence.

Provided that the judicial formulation of suspect classifications is properly constrained in a principled manner, the key advantage of such an approach becomes clearer – implementing a general right to equality by way of equality as non-discrimination furthers substantive equality most directly by explicitly addressing the most odious forms of discrimination that government authorities can effect. Indeed, such an approach would focus judicial attention on articulating what exactly is odious about certain classifications.¹¹⁰

The analytical advantage of such an approach in terms of getting most directly at substantive equality can be illustrated by way of an example. Returning to the facts of QT, it will be recalled that the Court of Final Appeal approached the case by ultimately considering whether the Director's decision was rationally justifiable under the test of proportionality. If the approach suggested here were to be adopted, the court would have to determine whether a differentiation drawing upon a category of impermissible differentiations had occurred. In this regard, no relevant category concerning differentiations on the basis of sexual orientation existed in Hong Kong jurisprudence at the material time. This would mean that specific attention would have to be paid to the issue of whether such a category ought to be created – in other words, should classification on the basis of sexual orientation be considered odious enough to merit special constitutional approbation, independent of the objectives such classification would be directed at achieving? This, one might think, was the central substantive equality issue at stake here. And this central issue would have certainly benefitted from focused judicial attention. Adopting Khaitan's framework, the relevant questions here would be whether the ground of sexual orientation classifies persons into groups with a significant advantage gap between them, and whether it amounts to an immutable or fundamental choice.

In contrast, proportionality analysis on its own would be merely directed at means-ends rationality and an analysis of the weight and balance accorded to relevant factors – crucially, it would not direct express attention towards the central substantive equality issue of whether some differentiations are particularly odious in themselves, independent of the ends they are directed towards achieving, and whether they therefore ought to be scrutinised particularly carefully. On this approach, should the court indeed be able to articulate reasons for why classifications on the basis of sexual orientation ought to receive special constitutional approbation, the classification in QT would then be assessed on the basis of proportionality review, demanding a tight correlation between the relevant differentiation and its purpose, as well as a sufficiently high threshold of importance for the ends sought to be achieved. If the court determines that such a classification ought not to receive special constitutional approbation, the classification would be assessed on the basis of regular irrationality review in administrative law – thereby preserving a meaningful distinction between the legal approach taken to general equality rights in constitutional law and common law substantive review.

Overall, therefore, it is suggested that this mode of rationalising the relationship between general equality rights in constitutional law and substantive review in administrative law is the least undesirable – while it is by no means perfect, it presents drawbacks which are at least capable of mitigation.

Conclusion

In sum, this article has sought to argue that there is a conceptual similarity between substantive review in administrative law and general equality rights in constitutional law, insofar as they have been effectuated through the concept of equality as rationality. Drawing upon Hong Kong and Singapore as test cases, this article has illustrated that both jurisdictions have struggled to articulate meaningful differences between both sets of legal principles. Pointing out that this state of affairs is problematic, primarily for the reason that both sets of legal principles exist at different levels in the legal order, the article argued that the least problematic approach towards addressing this conflation is to articulate general equality rights primarily by way of the idea of equality as non-discrimination. While this approach is not entirely unproblematic, the problems it raises are preferable to the problems of systemic doctrinal incoherence which a conflation between the two sets of legal principles would bring about.

Even if the proposal in this article were to be accepted, other issues remain to be resolved: How should direct or indirect discrimination on the basis of suspect classifications be analysed? What is the proper standard of review that should be adopted for differentiations on the basis of suspect classifications? What ought to be the appropriate remedy? The article's contribution, nevertheless, has been to raise these issues for discussion even in the context of general equality rights, which have otherwise been commonly effected only by way of the idea of equality as rationality. This, it is hoped, will be a contribution that paves the way for greater doctrinal coherence in the manner by which general equality rights are effected in law.