Transformation versus Constitutionalism: An Investigation into the Affirmative Action Principles and Their Limitations
TRANSFORMATION VERSUS CONSTITUTIONALISM: AN INVESTIGATION INTO THE AFFIRMATIVE ACTION PRINCIPLES AND THEIR LIMITATIONS
Second Year Postgraduate LLB Student, School of Law, University of the Witwatersrand
I INTRODUCTION AND METHODOLOGY
With an unemployment rate of approximately twenty five per cent in South Africa, jobs for the people of South Africa are difficult enough to come by. This difficulty is exacerbated by the emergence of transformation as the focal point of postmodern politics as the South African government seeks to reduce the inequalities that pervade its nation. The apartheid era of South Africa has caused great disparities of wealth and opportunity amongst the nation’s citizens. As a result, the new democratic order (see the Constitution) seeks to remedy these disparities by way of urging the implementation and adoption of measures that will see South African society reflect egalitarian values. However, these transformative measures, noble as they may be, do not operate problem-free.
This paper will commence with an exposition of the highly contested concept of transformation in relation to the vision of equality or disadvantage as articulated by the Constitution, the judiciary and legal scholars. Subsequently, the paper will expound the underlying principles of affirmative action vis-à-vis transformation. Following which, the paper shall identify the criticisms that loom over the implementation of affirmative action in South Africa. Thereafter, the paper will engage the debates before it to determine whether the ends justify the means. Finally, the paper will conclude with an opinionated account regarding the viability of affirmative action measures, as they are, as the route to achieving equality.
II TRANSFORMATION AND EQUALITY/ DISADVANTAGE
The concept of transformation is multi-faceted and therefore highly contested. Much of the meaning attributed to transformation is determined as the situation so requires. To scholars like Klare and Pieterse, and for the purposes of this paper, transformation is understood as a large-scale social overhaul that occurs within the legal boundaries of any particular state. In South Africa, it is a context-oriented concept that seeks a just transition from a deeply divided nation economically, socially and racially into a nation comprised of equals. This approach alludes to the understanding and interpretation of transformation as judicially-dependent in that its fairness and reach ought to be determined by the courts, in light of the circumstances of the time. Therefore, it takes for granted that transformative constitutionalism will lead to an eventual equality as its content is as dynamic as society.
However, others feel that such an understanding of transformation is flawed. Sibanda, for one, opines that the overreliance on the interpretation by adjudicators is transformation’s pitfall as the judiciary is inherently comprised of adjudicators with different feelings regarding transformation. Thus, he argues that transformation will remain an ever-undermined notion that, without any sort of shared consciousness, will continue to elude South African society.
Nevertheless, the objective of the transformational rhetoric is equality. Section 1 of the Constitution provides that amongst the country’s founding values is the achievement of equality. The term ‘achievement’ implies an already accepted, substantive understanding of equality. Moreover, section 9(1) provides that everyone is equal. The meaning of equality, however, is not to be understood literally. The equality clause and its interpretation are rather multidimensional. Section 9(3) places a duty on the state not to discriminate against anyone in any form. Furthermore, section 9(4) obliges the legislature to enact national legislation that will ‘prevent or prohibit unfair discrimination’. Section 9(2) is the transformative dimension as it permits the undertaking of legislative and other measures designed to protect or advance persons, or categories of persons, disadvantaged by unfair discrimination in its endeavour to achieve equality.
It is clear then from the constitutional text that equality means more than the mere prohibition of discrimination. Equality is understood in some ways as non-existent in South African society – it must be achieved. The Constitution therefore views equality and disadvantage as a regulator of relationships in the sense that it elevates equality of wealth and opportunity.
This understanding is mirrored by the courts in various cases. In S v Makwanyane, it was held that the Constitution is radical in that it symbolizes a shift away from a racist and oppressive regime to an epoch committed to egalitarian values. Moreover, in Soobramoney, the court provided that the economic inequalities that have scarred our country require us, as a nation, to submit ourselves to transform into a society that is inclined to uphold human dignity, freedom and equality. Even when dealing directly with affirmative action, the Constitutional Court (‘the CC’) in van Heerden has focused on the achievement of equality. The inference made by the CC is that equality ought to be interpreted substantively to arrive at its meaning, as opposed to adopting a formal understanding.
Formal equality disallows the implementation of legislation that discriminates between persons or categories of persons. It is oblivious to circumstances on the premise that everyone possesses the same rights and entitlements. This logic therefore ignores the entrenched structural inequalities that continue to hinder access, participation and enjoyment of people's rights. The problem with formal equality is that it ignores the role of circumstances in opportunity and, moreover, the role of history in creating circumstances. Hence, it cannot validate transformative measures that differentiate between categories of persons as it views equality as a present, rather than a future, ideal.
Substantive equality, on the other hand, is an understanding of the historical context of persons, or categories of persons, and its relation to previous systemic forms of domination which have had a prejudicial effect on such a group or category of persons in respect of their human and social development. The logic is such that where such systems have created great disparities, a nation must strive to address these issues with the aim of transforming its society into one based on human dignity, freedom and equality, as held in the Soobramoney judgment.
In addition, Moseneke J in van Heerden explains substantive equality as recognising that apart from ‘race, class and gender attributes of our society, there are other levels and forms of social differentiation and systemic under-privilege which still persist. The Constitution enjoins us to dismantle them and to prevent the creation of new patterns of disadvantage’.
Substantive equality is therefore remedial in that it urges us to achieve true equality. Mc Gregor analyses equality beyond the formal and substantive interpretations. She identifies an in-between feature, ‘Equality of Opportunity’. This view perceives true equality as unattainable if individuals do not have the same starting point. Thus, transformative remedies should be rewarded to groups, not individuals. She argues that substantive equality is consequently unacceptable unless it flows naturally from equality of opportunity. Dupper explains that ‘Equality of Opportunity’ is not naïve to individualism. He submits that rather, it is conscious of the fact that systemic institutionalised discrimination has a devastating effect on the opportunities of individuals because of such individuals’ affiliation to a particular group. Thus, he explains that race or gender-oriented policies may operate fairly to facilitate transformation, but only to the extent that they seek to create an equal starting point.
Albertyn, on the other hand, argues for a transformative equality of outcome. She distinguishes between an ‘inclusionary’ approach and a ‘transformatory’ approach to understanding equality. The ‘inclusionary’ approach recognises inequality but it fails to dismantle the structures that founded and reinforce the systemic inequalities. A ‘transformatory’ approach, however, is devoted to effecting change by restructuring the fundamental framework which determines the points of strength in society.
Thus, Albertyn calls for an understanding of equality that is appreciative of context and impact. Therefore, the impact of discrimination on any particular group of persons ought to be determinative. She supplements her view with reference to the judiciary’s consideration of all relevant factors.
Equality of outcome therefore envisages the achievement of equality as far as reducing inequality between categories of persons by recognising the roles of people or persons within society as ‘different groups in society experience a different mix of political, social and
economic inequalities, giving rise to different kinds of equality claims’. For example, one ought to consider that a woman is a mother and a wife. In some instances, she may even be a single mother and therefore a breadwinner, like the many men that surround her and who earn more for it. Therefore, in addition to being included in the workplace, she must be included in a manner that is sympathetic to her various roles.
The problem with this approach is that it relies on stereotypes to the extent that it attaches benefits where they may not be due. Not every woman wants to be a mother. A mother ought to obtain maternal rights as a ‘mother’ (a category of its own), not as a woman. Moreover, it fails to clarify what exactly must be made equal. Is it their terms and conditions of employment, their remuneration, or expectations of them? How does an employer ‘recognise’ every role of every employee? As a result, this paper prefers an understanding of equality of opportunity and will argue primarily in this light.
However, one must bear in mind that equality of opportunity is not a flawless concept. Many adopt a stance similar to that of Albertyn in the sense that they call for considerations of ‘degrees of disadvantage’. The argument is that transformation, and affirmative action, concentrate on group liberation and are, in turn, oblivious to situations wherein members of the same group experience entirely different lives. For example, within the Black group, there are further imbalances. Middle class Black people attend better schools and therefore develop better skills than their poorer counterparts. In effect, middle class Black people are more likely to receive employment opportunities than lower class Black people. As a result, socio-economic disadvantage becomes an issue to ponder within the designated group. The suggestion therefore is that race, gender and disability cannot be the only factors to consider. All factors must be considered. In India, for example, the more privileged of the Other Backward Classes, termed the ‘creamy layer’, are considered ineligible to benefit from affirmative action policies as they already have an economic advantage. 
The obvious issue with adopting a stance that reflects careful consideration of all factors and degrees of disadvantage is purely practical. Rhetorically, yes it is desirable for its sensitivity and devotion to fairness. Practically, however, if every decision required a consideration of all factors, we would find ourselves taking copious amounts of time deciding who ought to benefit in light of the myriad circumstances before each. Furthermore, every decision or appointment is then exposed to judicial review for its possible lack of consideration of a certain factor that is not mentioned in the legislative text. The result would be an overburden on our courts to determine the correctness of a decision.
Nevertheless, in essence, transformation encompasses a mindset that is projective and reflective in that it interprets equality as the situation so requires. However, it does so without a realised future ideal. Some scholars, particularly Pieterse, opine that attaching any sort of finality to transformation would be a reckless abandon of the intricacies that regulate the ‘transformation’ agenda. They suggest that the meaning and content of transformation ought to evolve along with the society within which it operates. This view reflects cognisance of the need to maintain, and not merely attain. It appreciates that equity may be achieved but which equity is then followed by regress and then a new a lifecycle must begin to redress the new inequality of the time.
Therefore, as will be highlighted later, it is uncertain when transformation or equality will be declared or considered to be achieved, or when the ‘new discrimination’, albeit justified today, will cease to be legitimate. Furthermore, it remains to be seen whether the objectives of affirmative action and its application are proving to dismantle the inequalities that plague South African society.
III THE PRINCIPLES AND OBJECTIVES OF AFFIRMATIVE ACTION
The Employment Equity Act (‘the EEA’)  was enacted to address the disparities that are the result of discriminatory practices in the workplace. The EEA therefore prohibits unfair discrimination. A mere prohibition however will not suffice. More is needed if we are to achieve true equality. Consequently, the EEA introduces affirmative action as the means by which to achieve this end.
Affirmative action is an institutionalized preferential empowering measure that, according to section 2 of the EEA, seeks
‘to achieve equality in the workplace by-
(b) implementing affirmative action measures to redress the disadvantages in employment experienced by designated groups, in order to ensure equitable representation in all occupational categories and levels in the workforce’.
The EEA recognises the obvious racial imbalance as well as the less systemic and more subliminal forms of past oppression as it identifies women and the disabled as beneficiaries of affirmative action.
The interpretation by the courts of affirmative action has been much the same. In George v Liberty Life Association of Africa Ltd, the court described affirmative action as a design aimed at eradicating inequality and tackling systemic institutionalised discrimination. The court referred particularly to racial and gender discrimination.
This sentiment was shared by the Supreme Court of Appeal (‘the SCA’) in Barnard. The SCA provided that the EEA purports
‘to assist in the national struggle to achieve an egalitarian society by putting in place measures to overcome historical obstacles and disadvantages and providing equal opportunities for all’.
The purpose of the EEA is therefore twofold. Firstly, it sets out to do away with unfair discrimination in section 6. Secondly, it envisages and, in some cases, obliges the implementation of affirmative action measures to ensure the equitable representation of suitably qualified persons in all occupations and at all levels in the workplace. Thus, it is abundantly clear that the purpose and objectives of affirmative action are consistent with the substantive understandings of equality promulgated by the Constitution and the courts.
The EEA’s aspirations and considerations, however, do not end there. The EEA reserves preferential treatment for the designated groups only – the subjects of previous systemic institutionalized discrimination. ‘Designated groups’ are defined as Black people, women and people with disabilities. Furthermore, ‘Black people’ includes Africans, Coloureds, Indians and now persons of Chinese descent as per case law. Also, ‘disabilities’ is not limited to physical and/or long-term disabilities. It includes even recurring impairments or mental impairments which significantly hinder one’s prospects of securing employment or advancing in the work arena. In addition, all beneficiaries must be citizens of South Africa.
Also, the EEA recognises ‘designated employers’ and in a bid to ensure that affirmative action is carried out, section 20 obliges a designated employer to propose and implement an Employment Equity Plan (‘EEP’). The EEP must state the affirmative action measures that the employer is undertaking.
Moreover, the EEA, cognisant of the need for efficiency and competency in the workplace, states that any beneficiary of affirmative action must be suitably qualified. In determining whether one is suitably qualified, the following factors must be considered: formal qualifications; prior learning; relevant experience; and the capacity to acquire, within a reasonable period, the ability to do the job.
It is important to note that, although the EEA does express a duty to implement affirmative action measures, the duty is restricted to designated employers only. Furthermore, section 20(3) envisages no individual right to affirmative action, as will be explained with reference to the following two cases.
In Harmse v City of Cape Town, the applicant was a Black man who applied for a number of posts with his employer. He was shortlisted for none. Consequently, he sued the employer and alleged unfair discrimination based on race and lack of relevant experience. The Labour Court (‘LC’) held that affirmative action is a right afforded to an employee from a designated group. The LC’s rationale was that if affirmative action was not construed as a right, it would leave unfairly discriminated against employees without any remedial action.
On the other hand, in Dudley v City of Cape Town, a Black female was unsuccessful in her job application to become a director. Instead, a White male was appointed. As a result, the applicant alleged unfair discrimination on the basis of race and/or sex. Moreover, she alleged that her failure to be employed amounted to a breach of the employer’s obligations to practice affirmative action.
The Labour Appeal Court (‘LAC’) held, contra Harmse, that affirmative action is a collective concept. Therefore, it does not permit any individual to disturb the systemic application of measures designed to promote affirmative action. Even though it can be argued that affirmative action was not promoted by appointing a non-designated member, the salient point is that no individual right attaches to affirmative action. This view was shared by the High Court in Stoman, which perceived affirmative action as targeting categories of persons rather than individuals.
The reasoning in Harmse is therefore flawed for its failure to acknowledge the existence and operation of section 9(3) and section 23(1) of the Constitution, as well as the Promotion of Equality and Prohibition of Unfair Discrimination Act 4 of 2000 (‘PEPUDA’).
Considering the constitutional provisions and PEPUDA, it is ludicrous to claim that, if affirmative action were not a right, there is no remedy. The remedy exists irrespective of whether affirmative action affords an individual right in terms of the EEA.
Furthermore, affirmative action cannot, for reasons of practicality, encompass an individual right against employees as it would restrict the process and freedom of hiring a suitable candidate when an employer has already attained an equitable level of representation. The argument that affirmative action should not be assessed on an individual level was emphasized by the CC in the van Heerden judgment. That case concerned the rules of the Political Office-Bearers Pension Fund. The rules differentiated between three categories of members in respect of employer contributions for members who joined Parliament following the new constitutional order.
In casu, the Minister and the Fund, on the logic that the individual ought to show that he or she was personally disadvantaged by previous systemic institutionalized discriminatory laws, put forward that the scheme was created to achieve equality between previous pension fund members and the new members who were previously excluded as a result of race and/or political affiliation. They argued, therefore, that separate indicators should be utilised for each new parliamentarian because not all new parliamentarians were subjected to prejudice and discrimination.
However, a significant majority of the new parliamentary members were excluded by apartheid laws on the basis of race and/or political affiliation. Hence, in the light of the difficulty and impracticality, the CC preferred to abstain from creating an affirmative.
Thus, the CC established a 3-pronged rationality test to determine whether an affirmative action measure is fair or not:
- (a) Do measures target people or categories of people who had been disadvantaged by unfair discrimination?
- (b) Are such measures designed to protect or advance such people or categories of people?
- (c) Do they promote the achievement of equality?
This test has, however, failed to provide certainty on the content of affirmative action. Many find problems with the inferences made by the formulation of the test. It is clear that the test is a breakdown of section 9(2) of the Constitution. Its focus is threefold: the composition of the benefitted group; the appropriateness of the design of the measure; and the promotion of the achievement of equality. Consequently, an affirmative action measure that passes muster under section 9(2) cannot constitute unfair discrimination under section 9(3). Furthermore, such measure cannot be presumed to be unfair in terms of section 9(5).
The first issue regards the first leg of the test. Particularly, it is offended by the requirement of the CC that the overwhelming majority of a benefitting group be from the designated groups. The inference then is that there could be a minority of beneficiaries who receive benefits but who are not envisaged as the targets of such measures. The criticism is that non-members of the designated groups are on the receiving end of a measure undertaken to enhance equitability.
While many view this as problematic, this paper, for one, agrees with the CC’s astute preference for practicality. To devise a scheme that would guarantee benefits to only the primary targets may prove counter-productive and unduly burdensome on the legislature or the employer. The negative outcome identified by critics is a rare occurrence at most and the CC’s reliance on an ‘overwhelming majority’ serves well to ensure that the greater numbers of beneficiaries are members of the designated groups.
However, the CC erred when it required that it must be shown that beneficiaries had been previously disadvantaged by unfair discrimination. Scholars such as Kentridge and McGregor opine that putting one to prove previous disadvantage as a result of unfair discrimination is a wasteful exercise. Moreover, it will only reinforce the racial divisions that pervade South African society instead of serving to dismantle the barriers. The EEA operates on the assumption that all members of the designated groups are victims of systemic unfair discrimination, and thus so should the litigation process.
A further issue is that because the EEA does not define disadvantage, we are left somewhat uncertain as to the determination of whether a measure promotes the achievement of equality – the third leg of the test. Attached to the determination process of whether a measure does promote the achievement of equality is a weighing of interests. However, due to the absence of a definition of disadvantage, we are unclear as to what standard against which we are weighing interests, or at what point disadvantage equates to inequality. For example, the EEA allows for the short-term loss suffered by previously advantaged persons in order to further the long-term ambition of substantive equality, yet it does not pay attention to situations where the short-term effect is devastating enough to violate section 9(2).
Consequently, there have been cries for a reading beyond the rationality test, to an additional or competing test of proportionality. The proportionality test in this regard would entail the balancing of the need for and importance of remedial measures with the likely or possible impact such measures would have on the negatively affected groups. Hence, where race-created measures are so burdensome or prejudicial to White South Africans that they signify a disregard for the dignity of such persons, such measures may be deemed invalid for they violate the fundamental human rights enshrined in the Constitution.
Alternatively, it should work in tandem with the fairness test. In the recent CC judgment in the Barnard case, Cameron J, along with some of his colleagues, suggested that formulating an applicable standard specific to affirmative action cases is imperative. Such cases require a less deferential standard than mere rationality. Adopting the rationality standard alone would hardly prove invalidity as any decision would in some way have a rational connection to the purpose of affirmative action. Furthermore, the rationality standard, by itself, cannot scrutinize an employer’s considerations of the various designated groups and, moreover, the competing interests of these categories of people with those adversely affected by affirmative action.
Therefore, that standard must be fairness. Adding the fairness standard to the rationality test is a means by which to ensure that the implementation of affirmative action measures is consistent with its purpose. The judges concede that fairness is an open-ended notion, but so are the concepts of reasonableness and proportionality. The judges stress that assessing fairness is not identical to determining whether the measure is unfair discrimination. The latter tests constitutionality while the former seeks to find whether the constitutionally compliant measure is properly implemented.
On the other hand, van der Westhuizen J, in the same case and court, opines that the third prong of the existing rationality test includes an assessment of the measure. He says that the term ‘achievement’ suggests that the measure must have some effect in order to be tested. Thus, there is already a distinction between the measure and the implementation in the rationality test. As a result, from a practical perspective, he is sceptical of the introduction of the fairness standard. It feels redundant after the rationality test. In addition, Jafta J also warned that fairness is a double-edged sword. It requires the court to weigh the interests of three persons, or categories of persons: the claimant; the designated groups; and the employers.
Despite the gaps in the above test and the differing opinions regarding the expansion of the test, it is evident that affirmative action purports to reach an equitable level of representation and enhances equality of opportunity in the workplace. It aims to do so by encouraging and, wherever necessary, obliging the implementation of measures that protect and levitate the opportunities and interests of the previously disadvantaged designated groups. Furthermore, it is clear that the judiciary requires that both the measure and its implementation be proper. The confusion is merely under which category the implementation must be tested, problems that will likely iron out their own creases as an approach to such cases will slowly crystallize.
IV THE PROBLEMS WITH AND LIMITATIONS OF AFFIRMATIVE ACTION
Affirmative action, despite its noble intentions, has been the subject of much criticism. This paper identifies the major pitfalls of affirmative action as articulated by various academics.
First, the following quote by the CC in Bato Star produces an argument that affirmative action will inevitably result in prejudice. It said:
‘Transformation is a process. There are profound difficulties that will be confronted in giving effect to the constitutional commitment of achieving equality. We must not underestimate them. The measures that bring about transformation will inevitably affect some members of the society adversely, particularly those from the previously advantaged communities’.
Opponents of affirmative action who adopt this logic thereby submit that all persons should be granted equal treatment, irrespective of gender, race or disability. Thus, they seek a formal understanding of equality. The basis of their argument is that schemes and measures, such as affirmative action, are as recriminative as the apartheid laws that discriminated against Black people, women and the disabled.
The second pitfall or convolution of affirmative action is the merit-equality of opportunity dichotomy. Critics herein suggest that, despite section 20(3) of the EEA referring to suitably qualified beneficiaries, affirmative action tends to benefit people on the basis of race, gender and disability. As a result, they posit that affirmative action undermines the importance of merit and competence. This belief flows from a fear of a strong application of affirmative action. Strong affirmative action entails that one, implementing affirmative action measures, may employ a member of a designated group instead of other, better qualified candidates. In effect, critics argue, in practice, as a result of affirmative action, representation dominates the selection process while individual merit is relegated. These critics will refer to the facts of the Barnard case to reinforce their argument because in that case the most suitably qualified person was not hired on the grounds that she was White and her appointment would affect racial representivity.
Following a similar route, some scholars opine that the implementation of affirmative action measures in the public sector tends to hinder the state’s capacity to act effectively as many employees are not sufficiently qualified or experienced to carry out their tasks. Consequently, due to the restrictions placed on persons not from the designated groups, many qualified employees have resorted to leaving the public service or not entering it at all.
Another aspect that critics draw attention to within the merit-equality of opportunity dichotomy is that section 20(3) merely requires that a candidate must be suitably qualified. This clause, they argue, may lead to a negative attitude regarding progress and initiative as some members of the designated groups may only strive to achieve the minimum threshold qualification as affirmative action almost guarantees their preference. In effect, we may see a situation where a member of a designated group who is merely suitably qualified and rarely excels is preferred to a non-member who excels beyond expectations and is therefore more deserving.
The third shortcoming of the EEA is that it lacks a sunset clause. Brassey suggests that affirmative action is not concerned with disadvantage, but rather with racial representivity. Supporters of this school of thought posit that the absence of a sunset clause which considers the dispensation of affirmative action requirements imposed on designated employers represents a ‘reverse discrimination’.
A fourth criticism of affirmative action is that it takes for granted that it will automatically succeed. It assumes that all employers will adhere to the rules and strive to meet the goals of affirmative action. The EEA is almost naïve to the severity of the repercussions that may last on certain institutions. Critics contend that failure may well ensue as affirmative action is by and large confusing with regard to its implementation. They argue that affirmative action therefore lacks a bit of realism. It expects all employers to implement it willingly and effectively without guidelines.
Fifth, affirmative action is susceptible to resulting in a different form of discrimination against members of the designated groups who appear to have benefitted from affirmative action. Charlton and van Niekerk submit that often beneficiaries of affirmative action are perceived as inferior by their colleagues. The belief becomes such that a beneficiary, Black, female or disabled, is hired only because such person is Black, female or disabled and not because such person is capable and the most suitable.
Thus, affirmative action is a multi-edged sword. It leads to harmful assumptions within a workplace while at the same time stabs at an appointee’s self-confidence if such person overanalyses and ponders the true impetus behind his or her appointment, which may induce undue physical and mental distress.
Sixth, there is strong criticism that affirmative action can be used as a shield against claims of unfair discrimination. Nepotistic or corrupt appointments of members of the designated groups can easily be justified with claims that the appointment is part of a plan that seeks to promote the achievement of equality.
The seventh obstacle to the success of affirmative action in South Africa is the poor state of the South African education system. According to the Global Competitiveness Report of 2013, in a sample of 148 countries, South Africa ranked 146 and 133 in the higher education and primary education categories respectively. This statistic is alarming in light of its impact of affirmative action policies and its prospects of success. According to the Social Policy Research Group, a poor education system that fails to produce sufficient skills will always be a barrier to achieving the desired end of affirmative action. Due to the lack of sufficiently skilled South Africans, and members of the designated groups, many companies resort to hiring persons who have been educated outside of the mainstream South African school system. As a result, we find the poor (those unable to pay for a better education) further removed from the highly-skilled labour market comprised of those with access to better education. In effect, the state of the South African education system is working against reducing inequality as opposed to facilitating the redress.
Finally, there is a strong belief amongst critics that the implementation of affirmative action has placed too much emphasis on race and not enough on gender and disability. In fact, Dupper exposes that the African National Congress considers the country’s national question to be the ‘liberation of the African majority’. It should spark no surprise then that race is at the forefront of the affirmative action policy. As a result we have an influx of Black people into the workplace with little attention as to whether there are a fair number of women and the disabled amongst them. Moreover, within the Black working class, women tend to be appointed to less pivotal posts. As a result, Black women, by and large, earn less than Black males. Furthermore, it lends itself to excluding White women especially who are regarded as members of the designated groups as they are overlooked on the basis of race. This was evident in the Barnard case, as will be exposed below.
Whether these arguments hold water in the greater context will be discussed below. However, notwithstanding any rebuttal, affirmative action is complex and restrictive on certain sections of the community, especially physically able White males. When taking into account the general makeup of the national demographics, physically able White males are the only category excluded from the designated groups. Thus, any physically able White male applying for any post is less likely to secure employment than his fellow candidate, provided that such candidate has the requisite qualifications. As a result of this unfavourable obstacle, affirmative action will henceforth remain controversial.
V BARNARD V SAPS: REPRESENTATION V MERIT
Facts of the case
The applicant, Renate Barnard (‘B’) applied for two posts that would have seen her promoted within the South African Police Service (‘SAPS’). Both times, she was not appointed on the basis that she was White. During both selection processes, she was shortlisted and regarded as the best candidate. In addition, all other shortlisted applicants were members of the designated groups. In respect of the first post, the evaluation panel gave B a rating 17 per cent higher than the next best candidate, Shibambu. The post was revoked and nobody was appointed to that post. When the second post was being shortlisted, B again scored significantly higher than her closest competitors. She scored 85.3 per cent while Mogadima and Ledwaba scored 78 per cent and 74 per cent respectively. Moreover, B was recommended for employment by the interview panel. Again, B was not appointed on the basis that her appointment would increase representation of White people on that salary level and the level of representation was already perceived to be too high. However, her promotion would have made representation at her current level more equitable. As a result, B instituted action against SAPS on grounds of unfair discrimination based on race.
The Labour Court (‘LC’) held that an employer ought not to apply affirmative action rigidly without any consideration of fairness in light of the circumstances of the case. Therefore, the LC held the failure to appoint B to be unfair discrimination. The Labour Appeal Court (‘LAC’), however, found that, because no decision occurred, no discrimination took place.
The SCA, on the other hand, concentrated on the issue of fairness in any particular circumstance. It sought to determine whether the failure to appoint B in these circumstances was a fair decision. In doing so, the SCA adopted the Harksen test. Essentially then, it questioned the impact of the discrimination on B and the hypothetical effect it could have on others in B’s position.
Furthermore, the SCA pointed out that B, as a female, was a member of the designated groups. Race ought not to be the defining criteria. The SCA underlined the fact that B’s closest competitors, Black persons, were not appointed serves as an indicator that B was the only suitable candidate because if representivity was a genuine concern either Mogadima or Ledwaba would have been appointed.
The SCA finally urged employers to be ‘situation sensitive’. The term ‘situation sensitive’ was adopted by the court in Naidoo as well. In this case, Naidoo applied for a job. The EEP targets of the workplace formulated a racial and gender composition of the workforce in that workplace. It reserved 2.5 per cent of the jobs for Indians. In addition, gender composition was split 70-30 in favour of males. Of the 19 posts at the relevant level, calculations revealed that 0.475 posts were reserved for Indians. Furthermore, as a result of the gender split, only 0.1 per cent of the posts were to be occupied by an Indian female. With the possibility at effectively 0, the court concluded that it is unwise to rely on demographics when creating an EEP. It urged employers to be ‘situation sensitive’ in order to ensure fairness. Thus, a ‘situation sensitive’ approach requires employers to be cognizant of an endless number of factors when setting targets.
Despite upholding the appeal, the CC judgment hardly deviated from the SCA, and definitely not in terms of the constitutionality of affirmative action. That was never at issue. The judgment contained four different opinions. Each opinion found the National Commissioner’s decision not to appoint B to be correct. However, the reasoning of each was somewhat different. The bulk of the variance surrounded the standard against which to assess an affirmative action decision, of the National Commissioner or any other decision-maker, as was discussed above.
Nevertheless, each opinion concluded that in light of the circumstances, particularly the over-representation of Whites at that salary level, including White females, the decision not to appoint B was not unfair. The CC also considered the fact that B was not absolutely barred from promotion. In fact, she was promoted to a different post later on. Furthermore, the CC provided that setting numerical targets, and exercising decisions in pursuance of them, do not amount to quotas and can be legitimate in circumstances that permit it.
Moreover, the CC reiterated that affirmative action measures should not affect service delivery. In this case, it was of little concern in the CC as it was pointed out that B herself conceded that any of her competitors would also enhance service delivery. The court did mention that a decision-maker must however explain how he or she balanced representivity and service delivery in reaching his or her decision. Van der Westhuizen J held that ‘it is incorrect to assume that the ideals of representivity and efficiency are necessarily opposed’.
It therefore cannot be said that the judicial rhetoric has shifted between the SCA and the CC. Both courts require the proper implementation of all measures and a consideration of all factors to determine the correctness of the implementation in question. In some ways, much of the CC judgment resembles a ‘situation sensitive’ approach of assessing the implementation of affirmative action measures. Because of its breadth, many have frowned upon a ‘situation sensitive’ approach as it opens the door to challenge any affirmative action appointment on the basis that it failed to display the proper extent of sensitivity.
VI DO/ CAN THE ENDS JUSTIFY THE MEANS?
Firstly, affirmative action, from the outset, perceives equality substantively. It operates to eliminate the disadvantages that exist within society in respect of employment and equality of opportunity. Any suggestion then that affirmative action is flawed due to the collateral damage it may cause by prejudicing previously advantaged groups contradicts the notion of transformation and affirmative action entirely. One must bear in mind that transformative measures will vary in time and place. In the United States (‘US’) for example, affirmative action entails race-based decision making in favour of the minority. In the US, unlike in South Africa, it was the minority groups who were the subjects of oppression. Critics of affirmative action in the US assert that it discriminates against a vast majority of the population. The difference in South Africa is that the absence of affirmative action may well be discriminatory against the majority. The most noteworthy ideological clash between the two models is that the US model merely seeks to address the present effects of previous discrimination – employers are urged to attain and not necessarily maintain racial representivity – whereas the South African approach is significantly more adamant about eradicating disadvantage permanently by obliging the public and private sector to implement affirmative action measures. In fact the US Supreme Court has upheld a ban against affirmative action as a factor when considering college admissions. The historical contexts within which the two models of affirmative action operate are clearly distinguishable. However, what is common cause is that they both seek to include rather than exclude. Affirmative action will not be transformative if it did not prefer one group over another. Further, it is absurd to compare the previous regime’s restrictions on employment opportunities to affirmative action as the purpose beneath each system is wholly distinct. The former sought to exclude members of the community whereas the latter is adamant about rectifying the imbalance in society by ensuring the inclusion of those previously restricted.
Secondly, regarding the issue of representation versus merit, section 20(3) of the EEA recognises the danger of persons, particularly employers, mistaking affirmative action to promote racial representivity at the expense of merit. Hence, section 20(3) specifically requires that any beneficiary of affirmative action be suitably qualified. In the main judgment in Barnard, Moseneke J underlined the ‘suitably qualified’ requirement. He states that abiding by this requirement is pivotal to ensure that efficiency and competence are not compromised in trying to achieve transformation. He explained that the EEA thereby protects ‘itself against the hurtful insinuation that affirmative action measures are a refuge for the mediocre or incompetent’. The CC advocated that in attempting to strike a balance between representation and service delivery and/or merit, various factors must be considered: the nature of the job and its duties; the needs of the workplace; the needs of the employer; and the under-or-over-representation of the grouping seeking advancement. Appointing an unqualified or incapable candidate, or ignoring the qualities of a more than competent candidate, would definitely be unacceptable.
Moreover, suggestions that affirmative action has led to a deterioration of suitably qualified White males in the public sector, or the abandonment, by White males¸ of pursuing employment in the public sector are statistically incorrect. In fact, the number of White managers in the public service almost doubled between 1995 and 2001 from 14 738 to 26 276. However, during the same period, at senior management level, the number of White males decreased by roughly eight per cent. Despite this mild decrease, by March 2005, 30 per cent of senior management was still White. In addition, the latest annual report of the Commission of Employment Equity revealed that White males still hold 24.1 per cent of the top management positions at state-owned enterprises and 26.3 per cent of top management positions at educational institutions.
With regard to private entities, White males still dominate tremendously holding 58.6 per cent of the top management positions in private entities. This figure is astounding when one considers that the remaining 41.4 per cent is shared as follows: (all figures represent percentage) African males, 8.8; Coloured males, 3.2; Indian males 6.6; African females, 3.8; Coloured females, 1.5; Indian females, 1.8, White females, 11; foreigners, male and female, 4.8.
It therefore goes without saying that the effects of the apartheid laws and systemic institutional restrictions still have a burdensome effect on the distribution of wealth and opportunity throughout South Africa, particularly in the private sector. The representation of males in each racial group is at least double that of females. More staggering then is the number of White females in top management positions. White women comprise 11 per cent of the top management positions – more than any male grouping of the non-White races. In light of the fact that women are considered to be previously excluded, the figure highlights the extent of the racial division between Whites and non-Whites. Furthermore, the statistics underline the different experiences of women across races. Questions have subsequently been raised surrounding whether White women ought to be regarded as beneficiaries of affirmative action. 
The fact that a minority population still controls a vast majority of the wealth within the country invokes an appreciation of the pressing need to reform the structure of business in order to achieve equality as envisioned by section 9 of the Constitution. Nevertheless, the application of affirmative action measures to sustain efficiency and attain representivity remains tricky. Mushariwa, in recognition of this conundrum, questions whether the EEA’s cry for representivity equates to efficiency in the exercise of public services, in light of section 20(3). Some opine that the requirement of a suitably qualified person ensures an efficient exercise of services, as opposed to the mere acquisition of representative employee. The aim of employers should then be to attain a balance between promoting representivity and improving service delivery, particularly in the private sector.
As pointed out above, the CC in Barnard clarified the rules surrounding employment equity. It declared that race, and therefore racial representivity, cannot be the only criterion upon which an appointment is made or not made. The CC judgment suggests then that where the most or only suitably qualified person for the job is not a Black person in terms of the EEA, or even not a member of the designated groups at all, such person should be appointed. However, it also stated that race can be a primary factor in circumstances that permit it. The problem with an over-reliance on race as a determining factor was illustrated in Naidoo above. An EEP which created targets based on demographics resulted in a zero per cent chance of employment for Indian females. Nullifying a person’s chance at employment based on racially-induced formulae is discriminatory. The CC in Barnard warned against this.
The concern with the Barnard judgment is that it appears to prefer race. The layman who is not legally sophisticated merely learns that a White woman was not hired despite being the best candidate consequent to an affirmative action measure and that the CC upheld it. In effect, in practice, race may continue to be determinative in selection processes as some will follow what they perceive to be a permissible trend or others adversely and improperly denied employment will accept it, uninformed of the intricacies, and harness hatred towards the measure. Some applicants or employers may not be aware of the true content of this judgment and therefore adopt a prima facie view of affirmative action – that it prefers ‘Black people’ to ‘White people’, or ‘women’ to ‘men’, or the ‘disabled’ to the ‘able’. In effect, the judgment only protects employees and applicants who, or whose employers, are aware of the interpretation of affirmative action given by the CC.
Therefore, the Barnard case, albeit a step in the right direction, is not the answer to ensuring that representivity does not supersede merit or that all measures are properly implemented. The various opinions suggest that the CC itself may be uncertain about what affirmative action requires in situations such as that in Barnard. One might assume then that the CC’s inclination towards a ‘situation sensitive’ approach, albeit unsaid, would result in one of two things. The first possibility is that selection processes may take much longer than we are used to as selection committees and employers are tasked with taking into account endless factors. Alternatively, employers, particularly in the private sector, may abandon their affirmative action targets for fear of being challenged on every appointment for not being sensitive enough.
Finally the subject of an inclusion of a sunset clause into the EEA is moot. What follows such a suggestion is a myriad of questions. The obvious queries include the possibility of ever reaching an equitable workplace at all or, more daunting, how to determine when equality has been achieved. It is widely held that affirmative action is a temporary measure. It seeks to remedy the disparities created by systemic institutionalised discrimination.
Thus, when the disparities cease to exist, the need to apply affirmative action will cease along with it. However, the US experience has taught us that the implementation of a legislative sunset clause may prove to undo the hard work done to achieve equality. In the US, employers are not required to maintain racial representivity. Therefore, if we were to implement a standard sunset clause, we too run the risk of going backwards once targets have been reached. Hence, it may be wiser to allow employers to determine their own sunset clauses rather than imposing a legislated and generally applied sunset clause. Each employer has different challenges specific to its workplace. A sunset clause, like an EEP, should be ‘situation sensitive’.
In Willemse, the LC held that once an employer has reached a level of equitable representation, merit must stand as the only consideration for appointing an employee to a post. This view was shared by the LAC in UNISA v Reynhardt. In this case, the respondent, Reynhardt, a White male, was not appointed to a post for which he was the most suitable candidate. Instead, a Coloured male, Mr Summers was. UNISA, the appellant, had an EEP which provided that upon attaining an equitable level of representation (70 per cent Black; 30 per cent non-Black), no further affirmative action would be implemented.
The LAC found that this represents an express intention not to apply affirmative action when it is no longer necessary. Accordingly, the LAC held that UNISA’s decision not to appoint Reynhardt on the basis of race was irrational. The policy adopted by UNISA appears to be a reasonable sunset clause that can ensure that affirmative action will cease to operate once equitability has been attained.
However, an EEP of such effect can only exist effectively if the employer who drafts it shares the same level of commitment to achieving an equitable level of representation at all. Moreover, one must bear in mind that, once the level of representation sought has been achieved, there is a responsibility to maintain that level. Hence, if an appointment or dismissal affects the level of representation resulting in a decrease below the thresholds set in the EEP, affirmative action should then be reintroduced. Thus, it should commence a new ‘lifecycle’.
In addition, the problem with introducing or inserting a sunset clause is that transformation and equality encompasses much more than just equitable representation in the workplace. It entails reaching equality socially and economically. Therefore, for as long as there remain inequalities in respect of opportunities and accessibility that are resultant of wealth, there will be a need to implement affirmative action because employment and wealth may enhance previously disadvantaged persons’ access to education and equality of opportunity. Also, a noteworthy consideration is the fact that the circumstances before us are the result of decades of systemic oppression. It will take much longer to rewrite the past.
It is clear that equality evades South African society. Affirmative action has been enacted to reduce the inequalities that are the result of previous systemic institutionalized discrimination and to facilitate South Africa’s transformation into an egalitarian society. In principle, affirmative action is a measure that furthers the accessibility and opportunity of persons belonging to the designated groups in order to attain equality within society. In agreement with judicial undertakings, affirmative action must operate as a law of general application which is blind to individual disadvantage and, in turn, mindful only of group disadvantage. To view disadvantage on an individual-scale would be overly onerous and burdensome on employers and the courts.
Thus, we ought to adopt a mindset that affirmative action is more likely to be beneficial than detrimental. Moreover, one ought to realise that affirmative action does not necessarily levitate race, gender and disability to a level of being determinative for appointment. Yes, these factors are important but secti