A. Introduction
On April 27, 2024, it was thirty years since South Africa went to the polls in its first democratic elections. Its widely acclaimed constitution, drafted by the democratically elected Constitutional Assembly, came into force just under three years later, on February 4, 1997. Its preamble declares that the Constitution is to “heal the divisions of the past and establish a society based on democratic values, social justice and fundamental human rights” and to “improve the quality of life of all citizens and free the potential of each person.”Footnote 1 Equality is a central leitmotiv of the Constitution. Not only is the achievement of equality a founding value of the Constitution,Footnote 2 but equality, together with human dignity and freedom, is also one of the three democratic values that are the guiding principles for the interpretation of the Bill of Rights,Footnote 3 and that guide the determination whether the limitation of rights is justified and proportionate.Footnote 4 The Bill of Rights also contains a clause, subtitled “equality,” Section 9, which contains three key elements: (a) The principle of equality before the law and the right to equal protection and benefit of the law,Footnote 5 (b) the prohibition on unfair discriminationFootnote 6 and (c) the affirmation that steps taken to remedy disadvantage arising from unfair discrimination may be taken.Footnote 7 Section 9 is often referred to as “the equality clause” and it has been invoked in many cases before the Constitutional Court in the last three decades.
Yet despite this strong equality provision and the active jurisprudence arising from it, South Africa continues persistently to have one of the highest rates of inequality in the world. The Gini coefficient of incomes, which measures inequality on a scale from 0 to 1, where higher values indicate higher inequality, was 0.67 in 2006, 0.65 in 2009 and 0.65 in 2015,Footnote 8 one of the highest rates of income inequality in the world. There is no academic disagreement that South Africa has an extremely high rate of income inequality, and the South African government has recognized that reducing the income Gini coefficient should be a national priority, with the National Planning Commission’s National Development Plan targeting a reduction in the Gini coefficient to 0.60 by 2030.Footnote 9
The disconnect between the evident ambitions of the Constitution to foster the achievement of equality and the troubling persistence of inequality in South Africa warrants further examination by constitutional lawyers. And this Article seeks to commence that examination by reflecting on the purpose and reach of Section 9 of the Constitution, which entrenches equality before the law and a right against unfair discrimination. This exercise should be of interest not only to those interested in South Africa, but to those interested in equality law more broadly, because despite the widespread adoption of anti-discrimination laws and a burgeoning jurisprudence on them in jurisdictions across the world, there is a growing global scholarly debate, and disagreement, about the purposes of equality and anti-discrimination law.Footnote 10
But not only is there a divergence amongst legal scholars both across and within jurisdictions as to the purposes of anti-discrimination law, there is a divergence too between how law defines discrimination and how ordinary people understand discrimination, as Tarun Khaitan observed in his theoretical work in the field.Footnote 11 Khaitan notes that what counts as discrimination in law is not the same as what constitutes discrimination in the eyes of ordinary people.Footnote 12 For example, discrimination is ordinarily understood as being intentional and direct. Yet in terms of discrimination law, discrimination may arise even where discrimination is not intended, and even where it is indirect.
A similar disjuncture, I suggest, lies in how the law—and Section 9 of the South African Constitution is a useful example here—uses the word “equality,” and how the word “equality” is used by ordinary people. Ordinary people, I suggest, may think of equality in economic or material terms, as Richard Wilkinson and Kate Pickett use it in their ground-breaking work, The Spirit Level: Why Equality is Better for Everyone. Footnote 13 Wilkinson and Pickett argue that the quality of social relations in any society is built on material foundations and, using income inequality as a key measure, they argue that reducing inequality particularly in developed societies would be beneficial to all inhabitants of those societies. They argue too that reducing material inequality in developing societies would also be beneficial but for slightly different reasons.Footnote 14 Their conception of equality is based on material equality, a usage widely followed in ordinary parlance. That usage is different to the concept of equality that lawyers ordinarily use when coupling it with anti-discrimination law. The use of the word “equality” in anti-discrimination law draws on the principle of equality before the law, and the principle that differentiations should not be drawn in law or practice between different groups of people in a harmful manner. Its focus is on how differentations are drawn, for what purposes and with what effects. It is a different concept to material equality or inequality.
The use of the sub-title “equality” in Section 9 of the Constitution might thus lead to the assumption that the primary mission and purpose of Section 9 is to address deep seated forms of material inequality, such as the income inequality that is so pervasive in South Africa. Instead, a review of the text of the clause and the jurisprudence under Section 9 makes it clear that its primary focus is to ensure that differentiations as to how benefits and burdens are conferred upon different groups of people within South African society, whether by the state or by private actors, are not unfair or unjustifiable. Given South Africa’s history where racial discrimination was constitutionally and legally entrenched with devastating social and economic consequences, this purpose is of crucial importance.
I pause here to note that it does not follow that patterns of material inequality are irrelevant to the interpretation of Section 9. They will be centrally relevant in determining whether a differentiation is fair or justifiable and to the interpretation of the affirmative action clause. A full discussion of these points is not possible here, but in short, the more likely a differentiation is to further disadvantage an already disadvantaged group, the more likely it will be found to be unfair. It also does not follow that the South African Bill of Rights is uninterested in questions of poverty and inequality. The Bill of Rights entrenches economic and social rights which provide, amongst other things, that everyone has the right of access to adequate housing, healthcare, education, social security, and food. The positive obligations upon government that are the correlative of these justiciable rights are of central importance in ensuring that government acts to address those living in poverty across South Africa.
In this Article, I suggest that the patterns of apartheid geography that persist into the democratic era are not easily addressed by reliance on the equality clause, Section 9. The roots of contemporary racially segregated neighborhoods lie, of course, in a raft of racially discriminatory laws—which will be briefly outlined below—that have now been repealed. It is important to note that if they had not been repealed, they would undoubtedly have been held to constitute a breach of Section 9. Yet it is also important that the repeal of the racially discriminatory laws did not end racial segregation. Today racial residential segregation is maintained by a complex intersection of social and economic factors. And the consequences of that segregation are severe, as I shall show. Racially segregated neighborhoods in South Africa maintain and even exacerbate inequality along a range of axes: Educational, employment, income, health and life expectancy.
To examine the question more fully, I will first provide a brief overview of poverty and inequality data in South Africa, and then explain how apartheid geography remains a key driver of both poverty and inequality. I will then turn to the South African Constitutional Court’s jurisprudence on equality and outline the approach that has been adopted by the Court, and discuss some of the leading cases, showing that the decisions have had an important impact, especially on family law. I will conclude by suggesting that although Section 9 may in some circumstances be important in addressing government policies or practices that discriminate against people living in segregated neighborhoods, it will not be easy to litigate. I will illustrate this point by outlining the challenges that arose in a recent case concerning the allocation of police to a poor and racially segregated neighborhood in Cape Town, Khayelitsha.
B. Patterns of Poverty and Inequality in South Africa
During the apartheid years, reliable data on poverty in South Africa were unavailable. Data on household poverty rates nationally were produced for the first time by a survey conducted in 1993.Footnote 15 Using a relatively high measure of the poverty line—ZAR840 at the time, which was equivalent to approximately 250 U.S. dollars at the time—the survey calculated that about one half of all South Africans lived in poverty, although the survey also showed that people living in poverty were not evenly spread, as in some of the more rural provinces two-thirds of people lived in poverty.Footnote 16 Even on a less generous poverty line measure, $1 a day, 25% of all South Africans lived in poverty.Footnote 17 The alarmingly high levels of poverty in the country are perhaps not surprising given the legacies of both apartheid and colonialism, nor was their uneven spread. It is perhaps worth noting that South Africa’s poverty rates were not as severe in 1993 as those of many of its neighbors, for example Zambia, but they were higher than in other middle-income countries, such as Chile, Mexico and Indonesia.Footnote 18
The racial dimensions of poverty were also starkly apparent in South Africa in 1993. 57% of Black African people were classified as poor in 1993, 20% of Colored people, 7% of people of Indian origin and only 2% of white people.Footnote 19 Life expectancy was approximately ten years longer for white South Africans than Black Africans and African babies were seven times more likely to die in their first year than white babies.Footnote 20
Seekings and Nattrass identify three reasons for the deep poverty, and inequality, of Black African people: First, over the previous decades, and centuries, colonial and apartheid policy deprived Black Africans of access to land and the independent African peasantry had been destroyed; second, Black Africans struggled to gain access to the labor market, for reasons of skills and location, both of which had been exacerbated, indeed driven, by apartheid labor market and education policy, and when they did get jobs they were generally poorly paid. And third, Black African households had until the closing decade of apartheid, few claims on the state for social security payments, such as pensions, and were relatively under-serviced by other state services, such as health care and education.Footnote 21
Thirty years later the picture of racialized poverty and inequality in South Africa remains largely unchanged and is deeply concerning. If one looks at the three drivers of poverty and inequality identified by Seekings and Nattrass, the first and second remain strong drivers of poverty and inequality. There has over the last three decades been little change in relation to access to land, and unemployment remains extraordinarily high. The only factor that has declined is the increase in the payment of grants to poor South Africans. As Seekings and Nattrass argue the labor market remains arguably the primary cause of the deep patterns of inequality and poverty that characterize South Africa.Footnote 22 South Africa has one of the world’s highest unemployment rates, and one of its highest rates of inequality as measured by the GINI coefficient. According to Harvard’s Growth Lab, unemployment rates have been growing in South Africa on average by 0.5 percentage points a year since 1994, reaching 33.5% in 2022, from 20% in 1994.Footnote 23 The overall unemployment rate is now 39% if one includes those who have ceased seeking employment. Youth employment is especially high at 61.5% in 2022. Even more concerning is that the widening gap in unemployment rates between Black and white South Africans—the unemployment rate of Black Africans is just under 40% while the rate for white South Africans is under 10%.Footnote 24
Scholars debate why poverty and inequality have been so persistent in South Africa. Some scholars argue that it was the democratic South African government’s adoption of neoliberal policies that resulted in the further entrenchment of racial poverty and inequality in South Africa,Footnote 25 but other scholars have argued that this is too simple an account that often ignores the important non neo-liberal elements of government policy, such as the refusal to privatise state-owned enterprises and deregulate labor markets.Footnote 26 The purpose of this paper is not to discuss what have been the major drivers of the failure to achieve the vision of South Africa’s constitution, to improve the quality of life of all. Instead, it is to look at a particular characteristic of South African society that makes equality law less effective in addressing patterns of inequality. That characteristic may be described as the long legacy of apartheid geography.
C. Apartheid Geography
South Africa’s current patterns of poverty and inequality reflect the historical processes of exclusion and marginalization set in train by the process of colonial dispossession in the seventeenth century.Footnote 27 By the early 20th century, large portions of the country were occupied by white settlers. One of the most significant pieces of legislation in the early years of union—which took place in 1910—was the Natives Land Act 1913,Footnote 28 which together with together with the Native Land and Trust Act of 1936Footnote 29 preserved 87% of South Africa’s land for ownership and occupation by white people, who at that time constituted about a fifth of the population. One of the direct consequences of this legislation was to deprive peasant farmers of access to land, with severe consequences for poverty in rural areas. Most of the land that was available to Black African people for ownership and occupation fell within what came to be known during the period of apartheid as “homelands” or “bantustans,” demarcated according to the nine main African language groups in South Africa. Over three and a half million South Africans were forcibly removed from “white” South Africa, into the homelands during the 1960s and 1980s.Footnote 30 The terms of the entry of Black African people to and residence in “white” South Africa were strictly regulated by strictly enforced pass and migrant labor laws.Footnote 31 The homelands were governed by government appointed traditional leaders under the Black Authorities Act of 1951, legislation which sparked fierce political resistance in the 1950s.
Urban areas, too, were the subject of sharp racial geography. Legislated segregated areas for Black Africans commenced under Prime Minister Smuts in the 1920s with the 1923 Native Urban Areas Act which empowered municipalities to provide for segregated residential areas. The process of urban segregation was accelerated once the National Party came to power in 1948 when it introduced its policy of apartheid, or separation. Under the Group Areas Act enacted in 1950,Footnote 32 the government was empowered to declare neighborhoods in urban areas for occupation and ownership by specified racial groups. Thousands of people were displaced under the legislation as racialized patterns of urban segregation were strictly enforced.
Apartheid urban development policy was to build segregated neighborhoods expressly to prevent urban integration. As one civil society activist describes in the context of Cape Town:
The archipelago of Cape Town was not created by accident. It was done so by design. Its extreme segregation was … primarily engineered over the course of the twentieth century. There were many reasons for this, but one of the most important was to control and repress political organisation. Apartheid’s urban architects needed to contain people, corralling them away from power.Footnote 33
These racial patterns persist thirty years into South Africa’s constitutional democracy, and they remain a profound driver of inequality and poverty. Figure 1 maps current rates of unemployment—the higher the rate the darker and bluer the color—against a map of the former homelands.
Figure 1. Unemployment rates across municipal areas in South Africa, overlaid with maps of the former homelands.Footnote 39
In the areas of the former homelands outside of the country’s major metropolitan areas, unemployment rates are nearly 80%, twice as high as for the country generally.Footnote 34 Nearly a third of the population lives in these areas. The problem of unemployment, however, is not only acute in rural areas that previously formed part of the homelands.
A similar pattern is evident in urban areas. The state has built a very large number of houses since 1994. The most recent census in 2022 now suggests that nearly 90% of all South Africans live in formal housing, rather than informal housing or traditional forms of housing.Footnote 35 If accurate, and the newly released census data has been subjected to scholarly criticism,Footnote 36 this is a substantial achievement for government. Yet, challenges arising from complex planning laws, the government’s ongoing commitment to providing four-room houses on individual plots rather than more dense forms of accommodation, and the cost of land in the inner-city areas, amongst other factors, have resulted in most houses being built on the periphery of cities, further embedding the patterns of apartheid geography and highly dispersed cities.Footnote 37
Jobs in the formal sector are located near central business districts, whereas as a direct result of apartheid spatial planning and post-apartheid housing delivery, most Black African people continue to live in neighborhoods at the periphery of the city. The consequence is that Black Africans who live on the periphery of the cities bear very high transport costs. A recent study by Harvard’s Growth Lab found that for the lowest quintile of workers, direct transport cost equate to a third of wage income on average.Footnote 38 As the authors state, given “these exceptionally high costs of getting to work, the logical decision of workers is therefore to remain out of the workforce.”Footnote 40 This pattern is not unique to South Africa, but one might expect low levels of participation in the formal economy to be accompanied by higher levels of informal economic activity.Footnote 41 South Africa, however, not only has very high levels of unemployment but also relatively lower levels of informal economic activity, for reasons that are not entirely clear, although persistent apartheid geography which results in distant and low density residential areas may well be a factor.Footnote 42
Material inequality in South Africa is thus driven, at least to some significant extent, by the ongoing patterns of racial residential segregation both in urban and rural areas. These patterns of segregation have their roots in colonial and apartheid policies, but despite the repeal of those policies, they persist. I turn now to an overview of the equality clause of South Africa’s Constitution.
D. The Approach to Section 9 of the Constitution
The doctrinal approach to the prohibition on unfair discrimination was established in a decision in 1997, Harksen v Lane NO, which concerned provisions of the insolvency legislation which vested the property of solvent spouses of an insolvent in the official managing the insolvent estate.Footnote 43 The Court set out the test to determine whether a provision is inconsistent with Section 9.Footnote 44 The core doctrinal approach requires two separate enquiries. The first is whether legislative differentiations have a rational connection to a legitimate government purpose. Under this test it does not matter whether the differentiation is based on one of the seventeen prohibited grounds,Footnote 45 any differentiation must be rational. The second enquiry is whether the differentiation, whether legislative or by conduct, is based on one or more of the prohibited grounds—or a different but cognate ground—and if it is, whether it is unfair. Determining whether discrimination is unfair requires an assessment of the impact of the discrimination on the complainant and others in their situation.
The approach in Harksen v Lane NO is thus focused on differentiation, whether arising from a legislative provision or conduct, and whether it has happened, and if it has, on what grounds and with what impact. The approach has been applied relatively consistently since 1997 across a range of areas of law. Perhaps most notably, it has been important in family law. South Africa has a pluralist system of family laws: The common law system based on Roman Dutch law, now substantially codified by legislation which regulates divorce, matrimonial property, intestate succession and provides a duty of maintenance for surviving spouses, the system based on African customary law, which also has been developed by legislation, including provision for registration of marriages, the regulation of polygynous unions, and matrimonial property. In addition, there are several systems of family law based on religious systems of law, notably Muslim personal law, Hindu marriages and Jewish marriages. Section 9 has been a key driver extending recognition and protection across these systems and extending protections to life partnerships that are not formalized by marriage under any of these systems.Footnote 46
For example, a recent decision held that the key legislation regulating marriage that originally applied only to common-law marriages, was inconsistent with the Constitution because it failed to provide protection to wives in marriages solemnized according to Muslim personal law and to children born of such marriages, and the Court gave the legislature two years to address the unconstitutionality, but the Court also ordered that for the meantime, the protections of the legislation should be afforded, with effect retrospective to 2014, to spouses and children in such marriages.Footnote 47 In order to comply with this order, an amendment to the Divorce Act, 1979 was adopted in May 2024, which recognized Muslim marriages and regulated divorce between people married accorded to Muslim rites.Footnote 48 A similar series of important decisions have declared invalid rules and practices that prevent Black African women from holding property.Footnote 49
The equality clause too was the basis for ruling that the prohibition on the marriage of same-sex partners was inconsistent with the Constitution,Footnote 50 and also the basis on which a range of other legislation withholding protection or benefits from same-sex couples was declared unconstitutional.Footnote 51
Harksen v. Lane NO has been employed in other areas too, notably social security law. For example, in an early case, the court held that legislation that the exclusion of permanent residents from means-tested social grants constituted unfair discrimination in breach of the equality clause.Footnote 52 In another recent case, the court held that the exclusion of domestic workers from legislation that provides compensation for workers who are injured at work was, amongst other things, indirectly unfairly discriminatory because the majority of domestic workers are Black women.Footnote 53 The order of invalidity of the legislation was backdated to April 1994.
These decisions have been important and have resulted in the provision of benefits to those who would not have otherwise received them. There have been disappointing decisions under Section 9 as well. Although early cases extended the protection of Section 9 to some non-citizens,Footnote 54 in other cases, the Constitutional Court has not afforded refugees and non-citizens remedies under the clause.Footnote 55
There is an established, though perhaps less well developed, jurisprudence in the field of affirmative action. The leading case established the principle that measures aimed at remedying past discrimination should be neither punitive, nor retaliatory, should not infringe the dignity of others and should advance the position of those who have suffered previous discrimination.Footnote 56 Although the application of this test has led at least on one occasion to sharp disagreement on the court, the principles themselves are generally accepted.Footnote 57
E. The Reach of the Constitution’s Equality clause
Understanding the drivers and the patterns of racial inequality and poverty in South Africa is important to understanding the reach of the Constitution’s equality clause in addressing inequality and poverty. The enduring character of apartheid geography means that Black South Africans who live in those rural areas, originally designated as homelands or bantustans, are not similarly situated to those who live in urban areas, and especially not to those who have jobs in the formal sector in urban areas. Similarly, those who live in the periphery of the cities in what were originally neighborhoods segregated by law, designed for exclusion and control, struggle to gain access to the formal labor market. It is hard to overstate how deep these patterns of inequality run and how multi-faceted, and how powerfully they drive processes of exclusion and inequality. Yet to those who do not experience them they may be scarcely visible.Footnote 58 But these patterns of inequality are not caused by current discriminatory legislative provisions. They are a result of long complex historical processes, including now repealed discriminatory legislation, that have not been arrested or reversed in the first thirty years of democratic rule.
During the first thirty years of democratic government relatively little has been done to address the consequences of apartheid spatial planning. The National Development Plan Vision for 2030 does note that South Africa’s inherited spatial structure perpetuates exclusion and asserts as a key principle the need to reverse the “historic policy of confining particular groups to limited space (ghettoisation and segregation).”Footnote 59 Although there has been some small reduction in the pattern of segregation, it remains very high.Footnote 60
We have seen already that unemployment is deeper in segregated neighborhoods, where poor people live, both in the cities and the rural areas. And there is clear evidence that racial segregation in schools, especially for Black African learners, remains very high. According to a recent study, the average Black African student attended a school that comprised 96.4% Black African learners and only 0.9% white learners, whereas the average white student attended a school that was 68.5% white, 3.3% Indian, 8.5% Colored or mixed race, and 19.6% Black.Footnote 61 The same study confirms that the data shows that a large majority of South Africans from all backgrounds favor racially diverse schools, but that residential segregation makes this very difficult, especially for Black African families.Footnote 62
Equality and discrimination law have thus not yet had a material impact on the patterns of disadvantage that accompany the persistent apartheid geography of South Africa. Racially segregated neighborhoods, whether in urban or rural areas, may at times be subject to policies or practices that are racially discriminatory, but even then, identifying and proving that government policies or actions are discriminatory is hard. Establishing that legislation is discriminatory is easier, which is why much of the key equality jurisprudence has been directed at legislative provisions, rather than conduct.
Just how hard it is to establish that government policies as they affect neighborhoods that are rooted in apartheid geography are unfairly discriminatory is illustrated by the recent decision of the Equality Court in Social Justice Coalition v Minister of Police.Footnote 63 This case concerned whether the South African Police Service (SAPS) discriminates in the human resources—that is, number of police officers—it allocates to poor segregated neighborhoods in Cape Town that are occupied, almost exclusively, by Black African people.
The case arose following the report of a commission of inquiry tasked with investigating complaints of inefficiency against the SAPS at the three police stations in Khayelitsha, Cape Town and whether there was a breakdown in the relationship between the local community and the SAPS in that neighborhood.Footnote 64 The Commission of Inquiry was established following a long campaign by six civil society organizationsFootnote 65 active in Khayelitsha to achieve better policing in the neighborhood. When their efforts directed at SAPS proved unsuccessful, the organizations wrote to the premier of the Western Cape province asserting that there were widespread inefficiencies, apathy, incompetence and system failures of policing in Khayelitsha and requesting her to establish a Commission of Inquiry, which she eventually did.Footnote 66
Khayelitsha is a large neighborhood on the eastern urban edge of Cape Town, with a population of approximately 400,000 people in 2014. It was established in the last years of the apartheid government and its residents are overwhelmingly poor Black African people. The median household income in Khayelitsha is half that of the median household income for the City of Cape TownFootnote 67 and 75% of households in Khayelitsha live below the poverty line.
Khayelitsha has a very high crime rate.Footnote 68 According to expert testimony received by the Commission, if one combines the crime rates in the three police stations in the area, greater Khayelitsha is the neighborhood with the highest number of murders, attempted murders, sexual offences, serious assaults and aggravated robbery in South Africa.Footnote 69 The Commission concluded from its examination of the expert evidence on crime statistics that, ‘the residents of Khayelitsha are faced with extremely high levels of contact crime, which included murder, sexual offences and aggravated robbery’ and it also concluded that reported crime rates probably significantly understate the prevalence of crime, by as much as 40%.Footnote 70
Members of the complainant organizations testified before the Commission as to why they had sought its establishment and what outcomes they hoped for from the Commission. Simply put, the organizations wanted the police to provide a fair and competent service to the people of Khayelitsha. The police were not seen as an enemy but rather as a service that was deeply unsatisfactory, something that the organizations wished to correct. For example, Ms Funeka Soldaat of Free Gender, an organization established to defend the rights and interests of the lesbian and bisexual community in Khayelitsha, explained that for ‘gays and lesbians the police are the most important people in our lives, because you must know that we have a problem in our community, we also have problems in our homes, and therefore that is the most important place to us because they are objective, they don’t take sides so the SAPS or police are very important to us.”Footnote 71
Expert evidence received by the Commission showed that the most understaffed police stations in the Western Cape, calculated on the basis of police personnel per capita, were in the poorest neighborhoods, often those established under apartheid on the periphery of the city for Black African people. Two of the three police stations in Khayelitsha were among the ten most understaffed police stations in the Western Cape province. The SAPS sought to rebut this evidence but without success. The Commission concluded that, “the residents of the poorest areas of Cape Town that bore the brunt of apartheid are still woefully under-policed twenty years into our new democracy.”Footnote 72 The Commission also noted that the mechanism for human resource allocation was neither accessible to the public nor monitored by key oversight agencies and that there appeared to be no check to ensure it did not result in anomalies.Footnote 73 The Commission recommended an immediate urgent overhaul of the system for allocating human resources by SAPS and also recommended that there was no reason for the system of human resources allocation by the police to be kept secret, and that whatever mechanism SAPS chose to replace its existing system of human resource allocation be subjected to oversight by the key oversight agencies.Footnote 74
The data examined by the Commission that resulted in its conclusions and recommendations, were not available in the public domain. The evidence was provided to the Commission, following subpoenas issued by the Commission to senior personnel in the South African Police Service (SAPS) requiring them to produce the data, amongst other data. After an initial challenge to the subpoenas failed, SAPS did provide an extensive amount of information, which the Commission estimated to be more than 50,000 pages, including the information concerning the allocation of human resources within SAPS.Footnote 75
It appeared from the documentary evidence, as well as the oral testimony provided by a senior SAPS officer, Brigadier Rabie, who was responsible for the design of the system, that the system SAPS used for allocating human resources between police stations was based on a calculation of what SAPS called the Theoretical Human Resource Requirement (the THRR). The THRR calculation is complex. Put simply, each year information is gathered from the more than 1,100 police stations in South Africa relating to a range of measures including crime rates for the previous four years within the precinct of each police station, as well as a range of environmental data including population density, the number of businesses, schools and other institutions within the precinct, and the condition of infrastructure in the precinct, which is then used to calculate a baseline figure. That baseline figure is then adapted in light of the demographic analysis of each police station, and then a process of allocation takes place.Footnote 76
The provincial commissioner of SAPS, Lieutenant General Lamoer commented that he continuously disputed the human resource allocations provided to police stations in his province and he testified that the other provincial commissioners had done so as well.Footnote 77 He said there was something “fundamentally irrational” with how the resource allocation was calculated.Footnote 78
An expert witness, Ms. Jean Redpath, analyzed the police evidence on human resource allocation.Footnote 79 She argued that if there were marked differences in police to population ratios in the police system of human resource allocations, the police needed to explain and justify them.Footnote 80 She suggested that the marked and apparently irrational variations in human resource allocations probably arose from a range of factors.Footnote 81 These included the poor quality of data provided by individual police stations, the wide range of factors included in the calculation that may result in “double counting” in a manner that might prejudice poor neighborhoods, and the failure to account adequately the human resource challenges that arise in policing informal neighborhoods.Footnote 82
Having considered the evidence, the Commission concluded as noted above, that the system of human resource allocation used by SAPS was unacceptable and recommended that the system be overhauled as a matter of urgency. The Commission also recommended that the Minister of Police establish a task team to investigate the question of human resource allocation. In the meantime, the Commission recommended that additional uniformed police be allocated to the Khayelitsha police stations.Footnote 83
Following the Khayelitsha report, several of the civil society organizations that had driven the establishment of the Commission sought to engage with SAPS to ensure that the Commission’s recommendations be implemented, but to little avail. When their efforts proved fruitless, they turned in March 2016, to the Equality Court in the Western Cape, seeking an order declaring that the system used by SAPS for the allocation of human resources in the Western Cape unfairly discriminates against Black and poor people on the basis of race and poverty.Footnote 84 The Equality Courts are set up in terms of the Promotion of Equality and Prevention of Unfair Discrimination Act, 4 of 2000, legislation mandated by s 9 of the Constitution. The evidence placed before the Equality Court was drawn from the evidence placed before the Commission, outlined above. In December 2018, the Equality Court upheld the challenge, finding that the system for human resource allocation adopted by the police discriminated unfairly on the grounds of race and poverty.Footnote 85 The Equality Court issued a declaratory order stating that the allocation of police human resources was unfairly discriminatory on the grounds of race and poverty, and then postponed the question of remedy for hearing on a date to be arranged between the parties.Footnote 86
SAPS initially sought to appeal the order made by the Equality Court, and the complainant organizations sought a cross appeal, but sometime later both withdrew their applications for leave to appeal. The parties then sought to find a date for the hearing on remedy. A date was provisionally set in March 2020, but the hearing did not proceed, and despite several attempts, no date was set by the Equality Court for the hearing on remedy. Frustrated by the failure of the Court to determine a date for the remedies hearing, the Social Justice Coalition approached the Constitutional Court in April 2021 for an order declaring that the Equality Court had “constructively refused to grant them a remedy” and to grant them direct access to the Constitutional Court.Footnote 87 The Constitutional Court, by a majority, refused relief on the basis that the Court could not entertain an appeal against a decision of the Equality Court when it had failed to give a decision, as the matter remained pending before the Equality Court. As at the time of writing in July 2024, the Equality Court has still not formulated a remedial order flowing from its decision that the human resource allocation system in the Western Cape is unfairly discriminatory on the grounds of race and poverty.
This account of the Social Justice Coalition case illustrates how difficult it may be to succeed in discrimination claims on behalf of communities who live in the periphery of the cities, or in rural areas, where it is claimed that the discrimination arises from government conduct or practice. The case suggests that there are at least four reasons for the difficulty. First, and perhaps most importantly, it will often be hard to obtain the evidence necessary to establish that the conduct of a state agency, such as SAPS, is discriminatory as between poor neighborhoods and wealthy ones. The evidence will often not be in the public domain, and it may require significant efforts to obtain it. In this case, even when subpoenas had been issued, SAPS was unwilling to make the evidence available, and it was only obtained after litigation to the highest court, on the part of the complainant organizations. Other avenues for obtaining such evidence may be found in the Promotion of Access to Information Act, 2 of 2000, which gives effect to the constitutional right to freedom of information, or alternatively using questions in Parliament.Footnote 88 But all these avenues will present challenges. In this respect, challenging a legislative provision is more straightforward, especially where the discrimination is direct, and relatively little evidence is required, as was the case in many of the discrimination cases discussed above. These cases are therefore easier to litigate than the factually dense cases seeking to declare that an internal policy or practice of a state department has an unfairly discriminatory effect.
Second, once the evidence has been obtained, the question arises whether unfair discrimination has been established. One of the challenges is that the differences between wealthier residential areas and poorer areas are significant and so differences in practices and policies in relation to those areas are to be expected. The question will be whether those differences are appropriate or not. In the policing case, for example, one of the key issues is how the police assessed the human resources requirements of policing informal neighborhoods such as Khayelitsha. Informal neighborhoods have few, if any, access roads for vehicles, often no street addresses to locate places where crimes have occurred and poor street lighting, if any. These difficulties had not been given sufficient weight in the human resource calculation. The differences between poor and wealthy neighborhoods in this respect are often stark, evade simple comparisons and warrant different policy interventions. Assessing whether the different interventions are unfairly discriminatory may be challenging. In the Khayelitsha case, following expert analysis, it became clear that the police population ratios in poor neighborhoods were far higher than in the wealthier neighborhoods. That differential was counter-intuitive given the high crime rates in Khayelitsha. The Equality Court concluded that the human resourcing allocation system was unfairly discriminatory given that no justification was provided by SAPS for the differential between poor and wealthy neighborhoods. However, assessing whether any differential is justified may be challenging in some cases.
Third, even if a court concludes that a state policy or practice is unfairly discriminatory between different neighborhoods, the question of appropriate remedies might not be straightforward. As mentioned in the Social Justice Coalition case, the Equality Court has not yet formulated a remedy. The challenge the Court faces in formulating a remedy will be to ensure on the one hand that the remedy removes the unfair discrimination, but at the same time the remedy must be practicable and enforceable, and result in appropriate levels of human resourcing across a wide range of neighborhoods.
Finally, even once a finding of unfair discrimination has been made and a remedial order made, further difficulties may arise with implementation. There is a difference between declaring an internal policy of the police to be in breach of Section 9 of the Constitution or of the Equality Act, and declaring a legal provision to be in breach of Section 9 of the Constitution. In the latter case, as happened in all the discrimination cases discussed earlier, the effect of an order of invalidity is that unless the Court tempers the effect of the order of invalidity, attempts to implement or rely on the provision may be challenged in court. There is less clarity with a remedial order relating to a policy or practice.
These four difficulties illustrate why reliance on the constitutional principle of equality will often be challenging when seeking to assert that government policies or practices are unfairly discriminatory in the context of long-established apartheid geography.
F. Conclusion
The equality clause, Section 9 of South Africa’s Constitution, has been an important instrument for addressing unfair discrimination that arises from legislative provisions and it has led to profound changes in several areas of the law, perhaps most notably the system of family law. However, it has so far proved less effective for addressing the material inequalities that are the legacy of apartheid geography, apparent in the lasting, deep racially based segregation of both South African cities and rural areas. That legacy is no longer enforced by law, but persists as a result of deeply embedded social, economic and historical forces. In this respect, calling Section 9, the “equality” clause may lead to a disconnect between what people think the provision can and should achieve and its actual purpose and effect.
Even when material inequalities are reinforced by government policies or practices, it may be challenging to seek to remedy those inequalities by litigation. Proving that a government policy or practice is unfairly discriminatory will require an assessment of the relevant policy and its effects. It may often be hard for citizens or civil society actors to obtain the detailed information necessary to conduct that assessment, and even where that is possible, and it is found that the policy is unfairly discriminatory, it may be a challenge to determine an appropriate remedy, as the Social Justice Coalition case illustrates. Unravelling the deep material inequality produced by apartheid geography that persists in South Africa requires robust and responsive attention from government as the National Development Plan Vision 2030 recognizes. It also requires courts when interpreting and applying the Constitution to be alert to the ways in which those material inequalities have been constructed and watchful in seeking to ensure that laws, policies and practices do not entrench them.
Acknowledgements
The author declares none.
Competing Interests
The author discloses that she served as the Chairperson of the Commission of Inquiry into Policing in Khayelitsha.
Funding Statement
The author declares no specific funding associated with this Article.
Open access
