The five black employees all work as merchandise controllers at Makro. In 2018, they discovered the payslip of a white female merchandise controller on the work printer, and noticed that she earned substantially more than they did.

They lodged a grievance with Makro, who then reviewed the salaries of all merchandise controllers (MC) “against the pay range” and made adjustments “to ensure that all MCs were within the range”.

But as they were still paid less than their white colleague, they approached the labour court with a claim that Makro was guilty of racial discrimination in the workplace.

They relied on section 6(1) of the Employment Equity Act which prohibits designated employers from unfairly discriminating, either directly or indirectly, against an employee, in any employment policy or practice, on one or more grounds, including race. This section must be read with section 6(4) of the Act, which makes clear that paying employees different salaries for doing the same work based on their race constitutes unfair discrimination.

In a disappointingly cryptic and formalistic judgment, the labour court in Saccawu obo Mabaso and Others v Masstores (Pty) Ltd t/a Makro, dismissed the claim on the ground that the differences in salary were not based on race, but were rather the result of the process previously followed by Makro when it recruited new staff. (Until 2018, Makro considered a candidate’s employment history when determining their remuneration, and in order to make the offer attractive, usually offered both internal and external candidates 15% more than their current salary.)

The court also argued that the differences in salary could not be based on race because Makro paid at least two black MCs more than they did the white female.

The court, somewhat confusingly, concluded that it was not sufficient for the five employees to merely allege that the white female employee earned more than them because of her race, and that “something more” was required of them. What they had to do and failed to do, was to show that they were paid less because of their race. 

Visit Daily Maverick’s home page for more news, analysis and investigations

Technically, the court mistakenly assumed that the onus was on the black employees to prove that racial discrimination occurred. But section 11(1) of the Employment Equity Act makes clear that all they needed to do was to allege that such discrimination occurred, after which the onus shifts to the employer to prove, on a balance of probabilities, that such discrimination did not take place as alleged, or that any discrimination that did occur was rational and not unfair.

More substantively, the court failed to consider the possibility that this may have been a case of indirect discrimination, and thus failed to consider the possible negative and disproportionate impact of the company’s policy on the wages of black employees. It seemed to have accepted that the policy did not discriminate on the basis of race because it was couched in neutral terms and did not negatively impact on all black employees doing a similar job.

But as the Constitutional Court explained in City Council of Pretoria v Walker, the prohibition on indirect discrimination “evinces a concern for the consequences rather than the form of conduct. It recognises that conduct which may appear to be neutral and non-discriminatory may nonetheless result in discrimination”.

It is not necessary to show that the employer had the intention to discriminate against employees of a particular race or that it explicitly did so. Instead, one has to ask whether the company’s relevant processes, rules or practices led to a situation in which black employees were on average paid less for doing the same job as their white counterparts. 

This focus on the impact of rules, practices or policies, rather than on the intention behind them, is of considerable conceptual importance. It signals a rejection of the formalistic conception of equality which requires identical treatment of similarly situated individuals, and recognises, as the Constitutional Court noted in Brink v Kitshoff NO, that an equal treatment approach will perpetuate “patterns of group disadvantage and harm” and “build and entrench inequality amongst different groups in our society”.

The South African approach is also consistent with the adoption of affirmative action measures, which our courts do not view as an exception to a general rule requiring equal treatment, but as an inherent requirement for the eradication of systemic discrimination and thus for the achievement of substantive equality. 

Because the court adopted a narrow and formalistic approach to discrimination, it failed seriously to consider the possibility that Makro’s recruitment procedures, in fact, discriminated against black employees. 

To assess whether a large organisation like Makro in fact discriminates, one would have to look at the institutional culture and history of the organisation, as well as the broader evidence of the impact of racial prejudice and systemic racial discrimination on the wages paid to black employees. 

One would also have to look critically at the processes, rules and practices of the organisation to determine to what extent these reflected the particular worldview or life experiences of the historically, socially and economically dominant group within the institution. 

As the Constitutional Court pointed out in MEC for Education: Kwazulu-Natal and Others v Pillay, norms embodied in an institution’s processes, rules and practices are not necessarily neutral, and often reflect the worldview of the “mainstream and historically privileged” group in the institution. 

The processes, rules and practices of an institution historically dominated by white men, for example, may be based on the unexamined assumption that the average employee of that organisation is Christian, heterosexual, gender-conforming, middle-class, white and male, and that the circumstances, life chances and experiences of all employees are more or less the same.

While appearing to be neutral and thus not discriminatory, such rules may have a disparate impact on those who are different from this presumed norm and may in fact discriminate against an employee on the basis of their race, gender, religion or sexual orientation.

An example of this would be a policy that requires employees of a certain rank to have a driver’s licence, where this is not an inherent requirement of the job. This would disadvantage employees who did not have the same opportunities to obtain a licence as the imaginary middle-class white male employee would have had, and would likely amount to racial discrimination. Similarly, a policy that did not reasonably accommodate the needs of primary caregivers of children (who, unfortunately, remain overwhelmingly female), may disadvantage women and discriminate on the ground of sex and gender.

Returning to the case under discussion, it is striking that Makro, in fact, reviewed the salaries of all merchandise controllers after the black employees lodged their grievance, and then adjusted the salaries of its employees “to ensure that disparities are eradicated”. We are not told what these disparities were and to what extent these may have been linked to race, but it does raise a strong suspicion that the company engaged in race-based wage discrimination in the past.

It is unclear from the judgment what justification the company offered for the continued wage discrepancies (the five black employees are still being paid less than their female colleague), but it would be surprising if the company did not use the existing salaries of its employees as a starting point for determining the new salary scales. 

Makro previously determined the salary of new MCs by looking at the employment history of the employee (inside or outside the company) and the salary they had earned in their previous job. This raises alarm bells as it suggests the company may have adjusted its salaries after receiving complaints, but did so without disturbing the racialised patterns of disadvantage created by past practices.

In any event, basing any decision on what salary to pay an employee on their past remuneration will more than likely result in racially discriminatory remuneration practices. This is so as various studies have shown that there is a significant unexplained differential in South Africa between the wages of black employees and white employees doing comparable work (although some studies suggest that this is not the case for the lowest-paid workers). 

In fact, one study, which compared the earnings of black employees and white employees who did comparable work, and were more or less of the same age, education, gender and worked in the same industry, occupation or sector, with the same family situation, found that the average wages of black employees were 30% less than that of white employees.

As the wages of many employees at Makro were directly or indirectly determined by an employee’s previous salary (as determined by a wage market that discriminates against black employees), it would be surprising if the overall effect of the company’s policy was not to discriminate against black employees by, on average, paying them less than white employees doing comparable work. 

As Makro carried the onus to prove that they did not discriminate against black employees, once the employees showed that they earned less than the white female employee, they had the duty to present any evidence to prove that on average white merchandise controllers were paid no more than their black counterparts.

Had the company produced the requisite evidence to show there was no evidence that black employees were on average paid less than white employees doing comparable work, the court would have been justified to rule in favour of Makro.

Makro seemed to have failed to produce such evidence, but because the court wrongly assumed that a seemingly neutral process or set of principles will not amount to racial discrimination, they nevertheless got away with it. DM