Juristica International

1. Introduction

In both academic articles and court judgments, a connection is frequently drawn between the idea of substantive equality, developed in the sphere of discrimination law, and social rights. To give just one example, in Grootboom, the South African Constitutional Court ’s leading judgment on social rights, the court observed that the case brought home “the harsh reality that the Constitution’s promise of dignity and equality for all remains a distant dream”.  *1 Generally, however, such statements are left undeveloped. Part of the aim of this article is to explore the connection between social rights and equality, so often drawn but so seldom elaborated. A further aim is to consider the limits of judicial activism in this area. How far can courts go in enforcing social rights, given the clear policy questions that these rights would seem to raise?

This article discusses these issues in light of the social rights jurisprudence of the South African Constitutional Court . The first part introduces the concepts of social rights and substantive equality. In brief, it argues that the substantive approach to discrimination law is distinctive insofar as it allows for preference to be accorded to worse-off groups, thereby accommodating the promotion of equality within the sphere of discrimination law. Thereafter, the article discusses the most obvious connection between discrimination law and social rights — that is, where a social programme distinguishes between people on the basis of a ground of discrimination. This discussion is undertaken in light of the Constitutional Court ’s judgment in Khosa.  *2

There is, however, a less obvious intersection between substantive equality and social rights. This concerns the principle, integral to substantive equality, that preference should be accorded to those who are worse off. This principle finds expression in Grootboom, where the court ordered that the state’s housing programme should be adjusted so as to accord priority to those who are destitute. The court did not, however, order that all such individuals be provided with relief, and the judgment has been criticised for according insufficient priority to those whose needs are most urgent. The final part of the article discusses the extent to which this criticism is valid and argues for an approach that, while not extending individual entitlements to the social right in question, nevertheless steps beyond Grootboom. It is suggested that this approach would adequately reflect the limits of judicial activism in this area.

2. Social rights and substantive equality: An overview

Social rights require governments to provide their citizens with the most basic amenities of life, such as food, water, and shelter. Typically, the state need only meet this obligation progressively and within the limits of the resources available to it. This is, for instance, the case under the International Covenant on Economic, Social and Cultural Rights (ICESCR)  *3 and the South African Constitution.  *4

Although the ICESCR has been widely ratified, only a handful of states have taken steps to enshrine social rights constitutionally. This means that the question of how social rights should be enforced has received relatively little attention. The United Nations Committee on Economic, Social and Cultural Rights (CESCR) has, however, recommended that a ‘minimum core’ approach be adopted.  *5 In essence, the minimum core refers to obligations that should be fulfilled immediately, notwithstanding the fact that social rights are subject to progressive realisation. One such obligation is that the state initially concentrate on the needs of those who are worse off, before moving on to other, less pressing, needs. Thus, in the sphere of housing, for instance, the state should first cater to people who have no shelter whatsoever before making provision for people who already have some form of housing, however inadequate.

The minimum core does not require that those who are worse off be provided with the relevant social right in its most expansive form. Instead, only ‘minimum essential’  *6 levels of the right need be provided, although the expectation is that these will be progressively upgraded.  *7 The minimum core should, however, be made available as a matter of individual right.  *8 In certain circumstances, the state may be able to justify failures to meet its core obligations. However, the argument is that it should do so on the same basis upon which rights are limited generally, which in South Africa would be a limitation clause enquiry under § 36 of the Constitution.  *9 The South African Constitutional Court has, by contrast, formulated a somewhat different approach to the enforcement of social rights, which is discussed in greater detail below.

Substantive equality, on the other hand, embodies a particular approach to discrimination law. In this regard, a good starting point is Aristotle’s view that equality consists in treating like cases alike and unlike cases differently.  *10 Although this formula has been subjected to criticism,  *11 it is submitted that it is not so much wrong as empty.

An initial problem is that almost all legislation draws distinctions between people for one purpose or another. In other words, governments habitually treat people differently from one another on the basis of perceived differences or similarities. However, not all of these distinctions can be closely scrutinised by the courts. To do so would be impractical and would result in the judiciary making determinations for which it is ill-qualified.

This discussion relates to the grounds of discrimination, or the type of distinctions that should attract judicial solicitude on the basis that they potentially give rise to issues of equality. Clearly, Aristotle’s formula gives us little guidance about how to approach this issue. In the US , courts have attempted to resolve this question by focusing upon process-related factors. Thus, in the seminal judgment of CaroleneProducts, Justice Stone held that legislative classifications should be regarded as ‘suspect’ if they are directed at “discrete and insular” minorities or groups that are at a disadvantage in the political process, the classic example of which are racial minorities.  *12

Quite apart from the question of grounds of discrimination, a further difficulty with the Aristotelian formula is that it does not tell us what should count as like and unlike cases. One attempt to lend it content is formal equality. In essence, formal equality takes the view that equality inheres purely in consistency of treatment; that is, all individuals should be treated alike, regardless of their membership of particular groups. In other words, formal equality gives content to the Aristotelian formula by stipulating that, in judging the similarity or dissimilarity of two cases, group membership should never be taken into account.

Formal equality is now widely maligned, and the difficulties with this approach are well-documented. Nevertheless, it bears reiteration that a central objection to formal equality is that it is capable of disallowing measures that are designed to promote equality. For instance, by insisting that individuals should always be treated alike, regardless of attributes such as race and sex, a formal approach appears to preclude positive action in the form of, for instance, affirmative action. Policies such as these recognise that disadvantage frequently tracks characteristics such as race and sex and therefore takes these into account rather than ignoring them completely. What formal equality fails to recognise, in other words, is that it is only in certain contexts that such characteristics are irrelevant and detrimental.  *13

Substantive equality, in contrast, takes account of the position of the individual in society and the impact that the measure is likely to have upon him or her.  *14 In particular, government action that entrenches pre-existing disadvantage is unlikely to be upheld, whereas measures that promote disadvantaged groups are likely to be endorsed.  *15 Unlike formal equality, substantive equality therefore authorises, although it does not require, positive action. We should note, however, that substantive equality does not imply a rejection of the Aristotelian formula. Instead, like formal equality, it lends it content, yet does so in a different way, by stipulating that whether someone is better off or worse off is relevant to whether that person should be considered ‘like’ other individuals and thus whether he or she may be treated differently. A key feature of substantive equality is therefore its commitment to improving the position of worse-off sectors of society.

The substantive approach to discrimination law raises, of course, a host of further questions. Of these, the most obvious is: If preference may be accorded to worse-off groups, how far may such preference extend? Put differently, at what point does the pursuit of equality cease to treat those adversely affected as equals?

To answer this question, we need to understand the centrality of dignity in the South African Constitutional Court ’s equality analysis. In the words of the court, unfair discrimination “principally means treating people differently in a way which impairs their fundamental dignity as human beings, who are inherently equal in dignity”.  *16 Dignity therefore serves as the test, or touchstone, of whether unfair discrimination has occurred. In so doing, it also tells us how far the remedial measures authorised by substantive equality may extend; equality may be pursued to the extent that it does not violate the dignity of better-off members of society.  *17

Dignity is, however, a notoriously vague value. How then should a violation of this standard be determined? The Constitutional Court has attempted to resolve this question by stipulating that various contextual factors must be taken into account. These include the position of the complainant in society, the nature of the discriminatory law or action and the purpose sought to be achieved by it, and the extent to which the rights of the complainant have been impaired.  *18

An exhaustive discussion of the court’s application of these factors is beyond the scope of this article. However, it is submitted that one way of understanding the court’s approach is that it is applying a proportionality standard, albeit in a loose and unstructured form. Put differently, in determining whether the complainant’s dignity has been violated, the court is weighing the impact of government action against the objective that the state is seeking to pursue. This is implied by the court’s emphasis upon the purpose that the discriminatory law or action is seeking to achieve, and the extent to which the rights of the complainant have been infringed.

As for the third factor cited by the court — the position of the complainant in society we have seen already that the court is more inclined to uphold government action that promotes the position of disadvantaged individuals and less inclined to uphold government action that entrenches pre-existing disadvantage. This could be taken as an indication that the proportionality standard is applied more or less intensely depending upon the position of the complainant in society.  *19 In other words, the worse off someone is, the more intensely the standard will be applied and the greater the weight that will be attached to his or her interests. This would accord with the emphasis of substantive equality upon improving the position of worse-off groups.

With this background in mind, we can now turn to the first, and most obvious, connection between equality and social rights — namely, where a social programme differentiates between people on the basis of a ground of discrimination.

3.Khosa v. Minister of Social Development

Khosa, the Constitutional Court ’s most recent social rights judgment, concerned a challenge to those provisions of the Social Assistance Act  *20 that reserved welfare benefits solely for South African citizens. The applicants, who were permanent residents, and who were therefore not entitled to benefits under the act, raised two arguments. Firstly, they submitted that their exclusion was inconsistent with § 27 of the Bill of Rights, the relevant provisions of which read as follows: “(1) (c) Everyone has the right to have access to social security, including, if they are unable to support themselves and their dependants, appropriate social assistance; (2) The state must take reasonable legislative and other measures, within its available resources, to achieve the progressive realisation of each of these rights.” Secondly, the applicants argued that the legislative scheme unjustifiably discriminated against them on the basis of their non-citizenship and therefore violated their right to equality under § 9 of the Constitution.

Khosa is therefore especially pertinent to this article, given that the applicants had been denied access to a social programme on the basis of their citizenship, a ground that the court had previously recognised as analogous to those enumerated in the Bill of Rights.  *21 The case therefore raises questions about the intersection between social rights and discrimination law. As Mokgoro J remarks, “What makes this case different to other cases that have previously been considered by this Court is that […] the social-security scheme put in place by the state to meet its obligations under § 27 of the Constitution raises the question of the prohibition of unfair discrimination”.  *22

How did the court approach this issue? To understand this, we need to take a step back and recall that in Grootboom — the Constitutional Court ’s leading social rights judgment, which is discussed in greater detail below — the court adopted reasonableness as the standard that is applicable in social rights cases. The court did so because, as indicated, the Bill of Rights requires the state to take “reasonable legislative and other measures, within its available resources, to achieve the progressive realisation of each of these rights”.  *23 The term ‘reasonable’ therefore describes the measures that the state should implement.

What, however, is meant by reasonableness? This issue is also explored in greater detail below. However, we can provisionally understand reasonableness as embodying a standard that is pitched roughly midway between rationality and correctness review.  *24 Furthermore, in Khosa the court develops the reasonableness standard introduced in Grootboom by importing a measure of proportionality. In the words of Mokgoro J, “In considering whether that exclusion is reasonable, it is relevant to have regard to the purpose served by social security, the impact of the exclusion on permanent residents and the relevance of the citizenship requirement to that purpose”.  *25 With this in mind, it is submitted that an analogy can be drawn between reasonableness and what the US Supreme Court terms ‘intermediate’-level review. The latter is applied in certain categories of equal protection cases and requires the state to adduce compelling reasons that are substantially linked to the achievement of important governmental objectives.  *26

In Khosa, the question for the court was the nature of the relationship between the reasonableness standard and the standard employed by the court in the sphere of discrimination law. In her lead judgment, Mokgoro J conceptualises the relationship between these standards as follows: “Equality in respect of access to socio-economic rights is implicit in the reference to ‘everyone’ being entitled to have access to such rights in § 27”.  *27 In other words, where the state draws lines, or formulates criteria, that deny some people access to resources while providing them to others, it cannot unfairly discriminate on the basis of a ground of discrimination. To do so would amount to a violation of the equality and social rights guarantees.

For Mokgoro J, discrimination law and social rights therefore overlap, or intersect, with one another. Equality does not exhaust reasonableness; it is merely a component thereof. In this sense, equality is implicated in some social rights decisions — those involving listed or analogous grounds — but not all. In Mokgoro J’s judgment in Khosa, discrimination law therefore forms one aspect of a broader discussion of reasonableness. It was on this basis that the court held that the exclusion of permanent residents from welfare benefits is unconstitutional.

A similar approach has, incidentally, been adopted by the CESCR, the body tasked with monitoring the implementation of the ICESCR. As mentioned above, in the view of the CESCR, states parties are subject to certain minimum core obligations or duties that should be fulfilled immediately, notwithstanding the fact that social rights are subject to progressive realisation. In this regard, the CESCR has emphasised that social rights should be implemented in a manner that is non-discriminatory. Thus the CESCR has, for instance, stated that a minimum core obligation in the field of health is “to ensure the right of access to health facilities, goods and services on a non-discriminatory basis, especially for vulnerable or marginalised groups”.  *28 This understanding of the relationship between social rights and discrimination law resembles that of the Constitutional Court in Khosa because it suggests that discrimination law is an aspect of the state’s social rights obligations that arises whenever a social programme differentiates between people on the basis of a ground of discrimination.

However, is there a deeper connection between social rights and equality? We have already seen that in Grootboom the court suggested that there might be by commenting that the case brought home “the harsh reality that the Constitution’s promise of dignity and equality for all remains a distant dream”.  *29 In a similar vein, in Khosa, Mokgoro J stated that “decisions about the allocation of public resources represent the extent to which poor people are treated as equal members of society”.  *30 These statements seem to imply that concerns about equality are implicated in the realisation of social rights generally, not just when there is differentiation on the basis of a ground of discrimination. Unfortunately, in both Grootboom and Khosa, such statements remain brief and undeveloped. It is precisely these questions that are explored in the remainder of this article.

4. Government of the Republic of South Africa v. Grootboom

The case that best enables us to explore further connections between social rights and discrimination law is Grootboom. The facts in Grootboom are now well-known, but a brief summary is in order. The claim was brought on behalf of a group of people who had no access to shelter whatsoever and were therefore genuinely destitute. They alleged that their circumstances violated § 26 of the Constitution, which reads, in part, as follows: “(1) Everyone has the right to have access to adequate housing; (2) The state must take reasonable legislative and other measures, within its available resources, to achieve the progressive realisation of this right.”

As noted above, in making sense of this obligation the court focuses upon the term ‘reasonable’ in § 26 (2). As for what reasonableness entails, we have already seen that it is analogous to what the US Supreme Court terms intermediate-level review. However, in Grootboom the court also stipulated that reasonableness entails a number of more specific obligations. These include that a “programme that excludes a significant segment of society cannot be said to be reasonable”  *31 and that “Those whose needs are most urgent and whose ability to enjoy all rights therefore is most in peril, must not be ignored by the measures aimed at achieving realisation of the right […] If the measures, though statistically successful, fail to respond to the needs of those most desperate, they may not pass the test”.  *32

On this basis, Yacoob J found the state housing programme to be invalid to the extent that it failed to make provision for people in immediate and desperate need. Although laudable, the programme concentrated unduly on the goal of constructing permanent houses for as many people as possible over time, instead of providing shelter for the desperate in the interim. The court therefore held that the programme would have to be modified so as to include a component catering for those in immediate need, even if this decreased the rate at which permanent housing could be constructed. As for the form of that component, that was left in the hands of the state, as was the exact proportion of the housing budget that should be allocated for that purpose. The court did, however, stipulate that “a significant number of desperate people in need [must be] afforded relief, though not all of them need receive it immediately”.  *33

Grootboom has been predominantly understood within an administrative law paradigm, no doubt because of the court’s use of the term ‘reasonableness’ to describe the standard of review. On this basis, the court’s decision has been both praised  *34 and criticised.  *35 Whatever the merits of this interpretation, it is submitted that it obscures an important feature of Grootboom, which is that the decision can be read in light of principles of substantive equality notwithstanding the fact that, in contrast to Khosa, a ground of discrimination was not expressly implicated in the facts of that judgment.

Firstly, recall that in Grootboom the court found that social programmes may not exclude a significant segment of society.  *36 This principle is a key step in the court’s judgment. It justifies Yacoob J’s finding that those whose needs are most basic should not be excluded — in the sense of not being specifically or adequately catered for — from the state’s housing programme.  *37 They, in the view of the court, constitute a ‘significant segment of society’.

What, however, is meant by the latter phrase? Clearly, this cannot be taken to mean an arbitrary group in society. After all, certain groups are simply not in need of assistance. Rather, it must be taken to refer to people who cannot meet their socio-economic needs independently, on the basis of their own resources. If such a group is excluded, the legality of the state’s social programme will be thrown into doubt.

The mere fact that a group is vulnerable, or unable to meet its needs independently, cannot, however, mean that it automatically has a claim to public resources. What is required, therefore, is an examination of whether the group has a legitimate claim to inclusion in a social programme from which others already benefit. In practice, this means that the state must explain why it has allocated resources in a particular manner. The task of the court is then one of evaluation. It must consider whether the state’s explanation is convincing, or whether the reasons advanced justify exclusion of the group. In this regard, the standard is, as outlined, one of reasonableness.

In light of this, it is submitted that Grootboom accords with a classic justification for judicial review in the area of discrimination law, which is that courts should protect the interests of vulnerable sectors of society who are unable to avail themselves of majoritarian political processes. As previously mentioned, in the US the seminal case in this regard is CaroleneProducts  *38 , in which Justice Stone held that “strict scrutiny” should be applied where legislation appears to burden “discrete and insular” minorities. Like the discrete and insular minorities of CaroleneProducts, the groups whom the court undertakes to protect in Grootboom — people who cannot meet their basic needs — are, one might say, similarly marginalised and unable to draw attention to their plight through conventional means.  *39

A further parallel that can be drawn between Grootboom and substantive equality is that in determining whether the complainants had been unjustifiably neglected a key consideration was their position in society. Had, for instance, the claim been brought on behalf of people who already had some form of shelter and were asking that their shelter be improved, it is likely that the court would have been less sympathetic. As we have seen, substantive equality provides that members of worse-off groups may be accorded priority vis-à-vis more privileged individuals and is therefore similarly committed to improving the position of worse-off sectors of society. In the areas of both social rights and substantive equality, the focus is on groups that are worse off.

These observations point to a relationship between social rights and substantive equality that is more far-reaching, and less obvious, than that identified by the Constitutional Court in Khosa. However, they also lead to further questions. The analogy between Grootboom and CaroleneProducts might, for instance, seem to imply that ‘poverty’, or some such criterion, can be recognised as a ground of discrimination — an assertion that is likely to strike many as implausible.

Unfortunately, this and other such issues cannot be explored here.  *40 Instead, the remainder of the article discusses the chief criticism of Grootboom, which is that, even though the judgment correctly accords priority to those who are worse off, it does not push this principle far enough. As outlined, the Constitutional Court held that the state need not cater to everyone in desperate need of housing. Instead, it need only provide relief to a ‘significant’ number of such individuals.  *41 Critics have argued that the court should have adopted the ‘minimum core’ approach advocated by the CESCR, in terms of which all destitute individuals would have been entitled to relief as a matter of right.  *42

5. The minimum core

Is this criticism of Grootboom persuasive? At the outset, the numerous attractions of the ‘minimum core’ approach should be noted. The most obvious of these has been mentioned already, which is that the minimum core would seem to better accord with the principle that priority should be given to those who are worse off. Secondly, by allowing social rights to be claimed by individuals, the minimum core would seem to do justice to the inclusion of social rights in the South African Constitution as rights instead of, for instance, directive principles of state policy.  *43 Thirdly, the minimum core lends content to the idea of progressive realisation. By suggesting that the state should first focus upon those whose needs are most urgent, before moving on to those whose needs are less urgent, the minimum core would seem to suggest a clear series of steps that should be taken toward the goal of full realisation of social rights.

However, despite these apparent attractions, it is submitted that the minimum core is subject to difficulties that ultimately make it an unappealing approach to the enforcement of social rights. These difficulties can be best approached by noting that the minimum core is especially attractive in the context of housing. This is because in this area it is possible to identify a particular group of people — those without any form of shelter — as worse off than everyone else. It is also possible to posit a spectrum ranging from those who are destitute and therefore worst off to those who live in permanent structures and are therefore best off. In between are various gradations, such as people living in informal settlements. With this spectrum in mind, it seems to make sense to argue that the needs of those who are worst off should ‘trump’ those of all others.

Nevertheless, it is submitted that this logic does not hold true for all social rights. A good example is the right to health care services. Where the minimum core of this right is discussed, it is usually identified with various forms of primary health care, such as the provision of essential drugs.  *44 However, unlike in the context of housing, primary health care is not co-extensive with the category of people who are worst off. Indeed, in the sphere of health, the difficulty is that there is morethan one group of people who are worst off.

Amartya Sen makes this point well. He asks us to imagine two people, A and B, where A has an income lower than that of B. Let us surmise further that A cannot access even primary health care. B, on the other hand, has chronic renal failure and cannot access treatment for his condition. Who, asks Sen, is worse off? Sen suggests that B might be worse off, given that he has a more “restricted capability set” and is therefore less able to achieve “functionings” that he has reason to value.  *45

In the context of health, it is possible to imagine many more such scenarios. Is someone worse off if he or she has cystic fibrosis or has not been adequately immunised against major infectious diseases? Is it worse to be schizophrenic or without access to essential drugs? The difficulty is that, unlike in the area of housing, here one set of needs cannot automatically be assumed to be most urgent and therefore to trump all others.

Similar difficulties arise in respect of the right to social security. Consider, for instance, Sandra Liebenberg’s attempt to define the minimum core of the right to social security. For her, “the most disadvantaged and vulnerable groups [should be] provided with basic levels of social security”.  *46 This includes the “elderly, people living with disabilities and HIV/AIDS, and the primary care-givers of poor children, and more generally those who are destitute and have no other means of supporting themselves and their dependants”.  *47

It should be clear, however, that Liebenberg’s formulation encompasses a number of different groups, with different needs, rather than a single class of individuals. And this, in turn, leads to problems of enforcement. At any time, only one of these groups is likely to be represented before a court. According priority to that group is likely to have repercussions for groups who are not represented. Again, unlike in the area of housing, where people without shelter can be identified as those who are worst off, there is no one category of people whose needs can automatically be assumed to trump the needs of other individuals catered to by the social security budget.

Fundamentally, the difficulty is that the minimum core embodies two competing rationales. On the one hand, it seeks to guarantee certain basic forms of provision, such as rudimentary forms of shelter and primary health care. On the other hand, it seeks to cater to those who are worst off. In respect of housing, these rationales can be made to converge, which is why the minimum core is especially attractive in that context. However, in areas such as health and social security, these rationales diverge or pull in opposite directions. Proponents of the minimum core are therefore faced with a difficult choice between, on one hand, abandoning their commitment to prioritising those who are worst off (in which case a key attraction of the minimum core is lost) or, on the other, discarding their commitment to providing only basic services (in which case the minimum core threatens to become impracticable).

Whether an alternative approach to the enforcement of social rights is available is discussed below. For now, the remainder of this section considers whether the minimum core remains an attractive approach in areas beyond the rights to health and social security, such as the rights to housing, education, and water.

In respect of these rights, I simply wish to sound a cautionary note. Consider, for instance, the minimum core of the right to housing. In the court a quo in Grootboom, in a judgment that cannot be discussed in full here, the South African High Court adopted what was virtually a minimum core approach by stipulating that the claimants should be provided with tents, portable latrines, and a regular supply of water (albeit transported).  *48

It is submitted that the difficulty with these measures, and the type of emergency relief advocated by proponents of the minimum core in general, is that they do not constitute long-term investments. Instead, they constitute an ongoing drain on resources. If measures such as this are to be effective, they might well have to be carefully targeted and, once in place, rapidly upgraded. In short, it can be argued that emergency relief should be linked to programmes — such as the construction of low-cost housing — that constitute durable investments, thereby freeing up resources that can be applied elsewhere. In a rigid form, the minimum core would, however, preclude the allocation of any resources to non-core needs until the core needs of everyone are met. In the long term, this might not constitute the most effective allocation of scarce resources.

Of course, it could be argued that the minimum core does not have this effect. Instead, it simply provides that where resources are devoted to non-core needs at the expense of core needs this should be justified in terms of the limitation clause, in the same manner that limitations of rights are justified generally.  *49 In this way, it is possible to compromise the minimum core in order to, for instance, strike a proper balance between the short- and long-term goals of social programmes.

However, the difficulty with this suggestion is that the limitation clause constitutes a highly structured and exacting proportionality enquiry.  *50 It might not be a correctness standard, but it certainly verges upon that. The proposed approach would therefore place judges in the invidious position of having to decide, via the limitation clause, the exact balance that should be struck between the short- and long-term aims of social programmes. In effect, this would amount to the courts themselves making these determinations, which they are poorly qualified to do.

6. A third way?

We therefore have reasons to doubt whether the minimum core constitutes an attractive and viable approach to the enforcement of social rights. On the other hand, the Constitutional Court ’s current social rights jurisprudence is also not without difficulties. Firstly, the approach adopted by the Constitutional Court in Grootboom, where the court desisted from recognising an entitlement to shelter, makes it extremely difficult for individuals to allege that their social rights have been violated.  *51 Furthermore, the vagueness of the court’s orders in cases such as Grootboom means that its judgments have not always been properly implemented.  *52

In the remainder of this article, I therefore outline an alternative approach that might constitute a ‘third way’ between Grootboom and the minimum core. Put briefly, the argument is that the court should, in accordance with the principles of substantive equality, vary the intensity of its review depending upon whether the claimant is better or worse off. The better off someone is, the less intense the court’s review should be. Thus, if a housing claim is brought on behalf of people who have virtually adequate housing and wish to have that housing upgraded, the court should apply a rationality standard. If, on the other hand, the claim is brought by someone who has no shelter whatsoever, the court’s standard should be more probing but should not exceed the intermediate standard that the court already employs.

This approach overcomes the difficulties inherent in both Grootboom and the minimum core. In contrast to the Grootboom approach, the court’s review would be focused on individuals rather than groups. In other words, a worse-off individual, rather than a worse-off group, would be able to approach the courts and request an explanation — one that satisfies intermediate-level review — as to why he or she has not been assisted.

On the other hand, in contrast to the minimum core, this approach does not seek to identify one group of worst-off people in respect of each right who should be accorded priority. It acknowledges that there may be more than one set of individuals who are worst off and provides that each should be able to approach the courts. Put differently, the chronically ill individual and the individual without access to primary health care, being similarly worse off, should be entitled to comparable explanations as to why they cannot be assisted, if that is the case.

Of course, in reality, a claim brought by a chronically ill individual in a country such as South Africa would be unlikely to succeed. His or her interest in treatment would in all likelihood be outweighed by the interests of a far greater number of others in primary health care. At the very least, however, the proposed approach would ensure that such individuals are treated as equal members of society.

A final, and crucial, distinction between the proposed approach and the minimum core is that the approach advocated here employs intermediate-level review as opposed to a limitation clause enquiry. As mentioned above, the latter verges upon a correctness standard. Consequently, a key difficulty with the minimum core is that it might require courts to make difficult judgments about the exact balance that should be struck between the short-term and long-term aims of social programmes.

In terms of the approach suggested, the courts would not be required to make such determinations themselves. Instead, they would exercise a secondary judgment that would require the state to demonstrate that it has compelling reasons for not catering to those who are worst off. The state could, for instance, demonstrate that the allocation of resources elsewhere is necessary to preserve the long-term integrity of the relevant social programme, or that resources have been allocated to more pressing needs. Fundamentally, however, the state should be required to meet a proportionality test that, while not placing the courts in the position of primary decision-makers, would ensure that appropriate weight is accorded to those whose need is greatest.

7. Conclusions

In conclusion, this article set out to discuss the relationship between substantive equality and social rights. The most obvious connection between these concepts is where a social programme differentiates between people on the basis of a ground of discrimination. This was, for instance, the case in Khosa. There is, however, a further connection between substantive equality and social rights, which concerns the principle that preference should be accorded to those who are worst off. This principle found clear expression in Grootboom, where the state was ordered to prioritise the needs of those who are destitute. However, a key criticism of that judgment is that by failing to adopt the minimum core, in terms of which all destitute individuals would have been entitled to relief as a matter of individual right, the Constitutional Court did not accord sufficient priority to those who are worst off. The final part of the article outlined several criticisms of the minimum core and proposed an alternative approach that, it was argued, would treat social rights litigants as equals while also reflecting the limits of judicial activism in this area.