Merger conditions alone cannot deliver transformation under the South African Competition Act

Introduction

The Competition Act 89 of 1998 asks the authorities to judge a merger twice. The first question is whether it is likely to lessen competition substantially. The second is whether public interest grounds, which include employment, the position of small and historically disadvantaged firms and the spread of ownership, point towards or away from approval (Competition Act 89 of 1998). Since parts of the Competition Amendment Act 18 of 2018, among them changes to section 12A, took effect on 12 July 2019, that second question has been asked more loudly than before, and Msutu (2025) describes a “second phase” in which the authorities apply the public interest standard more assertively. Mncube and Ratshisusu (2023) place all of this inside a long project of using competition law for economic transformation and black empowerment.

This paper argues that merger conditions are a legitimate but limited transformation tool, and that they cannot carry the weight now being placed on them unless market inquiries and enforcement against cartels and abuse of dominance do the structural work that conditions cannot. It first sets out how the Act writes public interest into merger control, then looks at two cases, Mediclinic and the Burger King merger, that show both the reach and the fragility of the standard. After that it turns to inquiries, excessive pricing and cartel enforcement, and finally asks where the balance should sit. The argument draws on a small body of recent South African scholarship, including Church (2024) and Loxton (2024) on the cases and Bonakele, das Nair and Roberts (2022) on inquiries, and on the decisions themselves (Mondliwa, Goga and Roberts, 2021).

How the Act writes public interest into merger control

Section 12A(3) lists the public interest factors. The authorities must consider the effect of a merger on a particular industrial sector or region, on employment, on the ability of small and medium businesses and firms owned by historically disadvantaged persons to enter or expand, on the ability of national industries to compete internationally, and on the promotion of a greater spread of ownership, in particular by historically disadvantaged persons and workers (Competition Act 89 of 1998). The ownership factor is the one that matters most for this paper. Msutu (2025) notes that the test applies to every merger and traces it from the Walmart/Massmart matter of 2012, where the Competition Appeal Court set the early parameters, through the 2019 amendments to the Commission’s 2024 Revised Public Interest Guidelines. On that account the guidelines widen the considerations and put a positive obligation on merging parties to promote a public interest ground.

That is a big shift, and this whole thing rests on a particular idea of what competition law is for. Mncube and Ratshisusu (2023) treat the Act as part of the wider post-1994 effort to redress racial exclusion, and they review how competition law has pursued black empowerment while pointing to current challenges. I read this as an honest description of the statute, because transformation is written into its purposes and not smuggled in by the Commission. The harder question is whether a merger is actually the right moment to pursue it. A merger review happens once, on a deal that parties chose to bring, and the remedy is whatever the parties will accept. Church (2024) frames her note on the Burger King merger as a warning to investors, which says something about how the standard is now seen by the people who actually do deals.

Writers working from an industrial policy perspective are less impressed by what competition law has done on the production side. Mondliwa, Goga and Roberts (2021) argue that competition law in developing countries, South Africa included, has been shaped by a neoclassical model that privileges static allocative efficiency and largely ignores production. Looking at metals and machinery and at plastics and chemicals, they link the failure to build diversified production capabilities to entrenched incumbent advantages and the lack of a coordinated policy agenda. If that is right, ownership conditions in individual mergers will not on their own reshape rivalry, and the weight placed on them may be misjudged (Mncube and Ratshisusu, 2023).

The ownership factor deserves a closer look because it is where the Act’s wording meets the cases most directly. Section 12A(3)(e) speaks of increasing ownership by historically disadvantaged persons and workers in firms in the market, and in the Burger King matter it became an employee share ownership plan, a sale to historically disadvantaged buyers and a supplier development target (Competition Act 89 of 1998). The Tribunal’s decision records that public interest must be assessed even where no competition concern arises (ECP Africa Fund IV v Competition Commission, 2021). That is consistent with the text, and it explains how a deal with no overlaps could still be prohibited. What the text does not supply, though, is a weighting rule, so in principle the same ownership loss could justify a prohibition in one case and a modest condition in another (Msutu, 2025). The consolidated Act I could access lists the 2018 amendments as the latest to date.

Two cases that test the standard

Competition Commission of South Africa v Mediclinic Southern Africa (Pty) Ltd concerned Mediclinic’s proposed acquisition of control of Matlosana Medical Health Services, which owns two Klerksdorp hospitals with 247 beds. Mediclinic already had a hospital in Potchefstroom, about 50 km away. The Tribunal prohibited the merger on 29 January 2019, the Competition Appeal Court reversed that by majority, and on 15 October 2021 the Constitutional Court set the appeal court’s order aside (Competition Commission v Mediclinic, 2021). Mogoeng CJ’s majority stressed access to health care, the position of uninsured patients and the Act’s concern with inequality and ownership, and it held that an appellate court should interfere with a specialist tribunal’s findings only on misdirection or a clearly wrong result. Msutu (2025) notes that the decision drew on the Act’s objectives and the Constitution to interpret the public interest test.

Loxton (2024) is sharply critical of that reasoning. The note argues that the Court brought constitutional rights into competition law without engaging adequately with competition law itself, that its approach to jurisdiction contradicts existing jurisprudence, and that it did not deal properly with the evidence. I think the criticism lands on the method more than the outcome. The prohibition rested on a finding that tariffs were likely to rise, so competition harm was in play, and the public interest language strengthened a result that could probably have stood without it. Even so, the case shows how readily constitutional values travel into merger review once the public interest provisions are on the table (Msutu, 2025).

The Burger King merger is the opposite situation. ECP Africa Funds proposed to buy 95.78% of Burger King (South Africa) from a subsidiary of Grand Parade Investments, and there were no horizontal or vertical overlaps. The Commission prohibited the deal on 1 June 2021 purely because it would end the black shareholding that Grand Parade’s ownership represented, and it brushed aside the argument that the sale would benefit those shareholders by treating their return as a private gain and noting that Grand Parade had other ways to raise capital (ECP Africa Fund IV v Competition Commission, 2021). Church (2024) says this was the first time an intermediate merger had been prohibited solely on public interest grounds since the Act came into force. The Tribunal approved the deal in September 2021, but only after the parties accepted conditions that included an employee share ownership plan giving an effective 5% interest, the sale of a meat plant to historically disadvantaged buyers, about R500 million of capital expenditure and at least 1,250 additional jobs for historically disadvantaged employees.

Put side by side, the two cases pull in different directions. In Mediclinic the public interest reinforced a competition finding, while in Burger King it stood alone and produced a negotiated package. The Commission’s own reasoning in the second case was that some of the expansion promised was already in the company’s internal plans and so was not merger-specific (ECP Africa Fund IV v Competition Commission, 2021). That worry seems fair to me, and the Tribunal in turn demanded baselines for the ownership plan, the store roll-out and the payroll commitments so that compliance could be checked. None of the sources I could read follows these conditions through to see whether the jobs and ownership arrived, which means the claim that conditions transform anything remains largely untested (Church, 2024).

What inquiries and enforcement add

Market inquiries are the Act’s tool for looking at structure. Section 43A defines one as a formal inquiry into the general state of competition and the levels of concentration in and structure of a market, without necessarily naming any firm (Competition Act 89 of 1998). Bonakele, das Nair and Roberts (2022) review their use in banking, private healthcare, liquefied petroleum gas, grocery retail and data services, and they argue that the inquisitorial process finds and fixes problems faster than adversarial cases. They also find that inquiries reach what they call competition-plus issues, such as barriers to entry and the ability of smaller firms to compete, although the recommendations on those issues usually depend on better regulation, and the practical impact of inquiries on broader public policy questions is less clear.

The speed claim deserves a caveat. The grocery retail inquiry opened on 27 November 2015, was originally expected to finish in May 2017, and published preliminary findings and recommendations only on 29 May 2019 (Competition Commission of South Africa, n.d.). That is hardly quick, in its own right, even if an adversarial abuse case might have taken longer, and it sits uneasily with the speed that Bonakele, das Nair and Roberts (2022) credit to the process. Still, inquiries address features of a market that no single merger or complaint can reach, and in a country where Mondliwa, Goga and Roberts (2021) see incumbent advantage as one brake on new capabilities, that seems closer to the real problem than a shareholding condition on one transaction.

Abuse of dominance enforcement looks weaker on the structural side and stronger on consumer protection. Section 8(1)(a) prohibits a dominant firm from charging an excessive price, and since 2018 section 8(3) asks whether the price is higher than a competitive price and whether the difference is unreasonable (Competition Act 89 of 1998). Ncube and Shumba (2025) say excessive pricing remains hard to establish, largely because the Act does not define a competitive price, the list of factors in section 8(3) is open-ended, and there are few decided cases. The first case under the amended provisions, Babelegi, concerned the price of face masks, and in another COVID-19 case the Tribunal fined Dis-Chem R1.2 million over price increases of between 47% and 261% (Ncube and Shumba, 2025). Bonakele (2022) describes the Commission’s COVID-19 response as a deliberate choice to use excessive pricing tools to protect poor consumers.

Those cases were basically crisis-driven, and I would not read them as evidence that excessive pricing law can restructure markets. What they show is that the Commission and the tribunals can act quickly when a price rise is visible and politically intolerable, and that courts still ask for empirical support before they intervene (Ncube and Shumba, 2025; Bonakele, 2022). Ordinary pricing power in concentrated markets, which is usually less dramatic, is likely to be much harder to challenge, since the open-ended factor list leaves both sides unsure what evidence will count (Ncube and Shumba, 2025).

Cartel enforcement rests on one of the Act’s clearest prohibitions, since section 4(1)(b) outlaws agreements to fix prices, divide markets or rig tenders (Competition Act 89 of 1998). Even so, the foreign exchange cartel litigation exposed a gap. Ndlovu (2025) explains that the Act addresses subject-matter jurisdiction in section 3(1) but is silent on personal jurisdiction over foreign respondents, so the Commission had to satisfy common law requirements and show connecting factors with the Tribunal. Ndlovu suggests that Parliament amend the Act to deal with personal jurisdiction expressly, and adds that enforcing orders against foreign firms is a separate problem. A transformation agenda that relies on opening markets to entrants depends on this kind of enforcement working, because structure and the reach of firms shape who can compete (Nsomba, 2021).

Where the weight should sit

My position is that merger conditions should stay but be disciplined. They are most defensible where the harm or benefit is specific to the transaction, as with the loss of black ownership in Burger King or the effect on uninsured patients in Mediclinic, and where the conditions can be monitored. The Tribunal’s insistence on baselines is the right instinct (ECP Africa Fund IV v Competition Commission, 2021). What is harder to justify is a standard that grows in the guidelines without much evidence about what conditions achieve, particularly given Msutu’s (2025) point that the guidelines now expect merging parties to promote a public interest ground.

The structural work belongs elsewhere. Parliament could define a competitive price and tighten the section 8(3) factors, which is what Ncube and Shumba (2025) recommend, and it could close the personal jurisdiction gap that Ndlovu (2025) identifies. The Commission could keep using inquiries for concentration and barriers to entry, while recognising, as Bonakele, das Nair and Roberts (2022) do, that recommendations need regulators and departments to follow through. None of this is meant to cut against transformation as an aim of competition law, which the statute plainly embraces. The real question when it comes to remedies is which instrument suits which job (Mncube and Ratshisusu, 2023).

A fair objection is that ownership is exactly where markets will not self-correct, so a merger condition may be the only direct lever available. I accept that for ownership specifically. The Act’s own purposes cover it, and the Tribunal in Burger King showed that conditions can be made concrete. But ownership conditions on a few deals touch a small part of a concentrated economy, and the sources suggest that entrenched incumbents, high barriers to entry and weak enforcement tools are the larger obstacles (Mondliwa, Goga and Roberts, 2021; Ncube and Shumba, 2025). Bonakele, das Nair and Roberts (2022) note that inquiries reach barriers to entry, which individual merger conditions are not designed to address.

There is also a legitimacy cost. When the authorities prohibit a deal that raises no competition concern, and then clear it on conditions that Church (2024) calls onerous, the process starts to look like bargaining under threat. Investors may price that in, and domestic firms may find it hard to predict which commitments will be asked for. The Constitutional Court’s willingness in Mediclinic to read the Act through constitutional values adds a second layer of uncertainty, since Loxton (2024) fears that competition principles are being distorted rather than applied. Predictability is not an enemy of transformation. A clearer weighting rule, whether in the guidelines or in amendments, would make the conditions that follow easier to defend and to monitor (Msutu, 2025).

Conclusion

South African merger control now treats public interest as a standing part of every review, and the authorities have shown they will use it, even to prohibit a deal with no competition concerns. The case law has been mixed. Mediclinic looks like a competition decision wrapped in constitutional language, which Loxton (2024) finds troubling, and Burger King ended in conditions whose results I could not find measured anywhere (Church, 2024).

If transformation is the goal, merger conditions can contribute but cannot be the main engine. Inquiries, excessive pricing law and cartel enforcement deal with concentration, entry and price, and each has gaps that legislation and practice could close (Bonakele, das Nair and Roberts, 2022; Ndlovu, 2025). A sensible first step would be a retrospective look at the conditions imposed since 2019, checking whether the promised ownership, jobs and supplier spending actually appeared, because without that the debate stays at the level of assertion (Msutu, 2025). Until that happens, the public interest provisions will keep being asked to do more than a single transaction can bear.

Reference list

Bonakele, T. (2022) The South African competition agency response to COVID19. Journal of Antitrust Enforcement, 10(2), pp. 241-247. doi:https://doi.org/10.1093/jaenfo/jnac010

Bonakele, T., das Nair, R. and Roberts, S. (2022) ‘Market inquiries in South Africa: meeting big expectations?’, in Motta, M., Peitz, M. and Schweitzer, H. (eds) Market investigations: a new competition tool for Europe? Cambridge: Cambridge University Press, pp. 291-319. doi:https://doi.org/10.1017/9781009072007.006

Church, J. (2024) Investors beware: public-interest considerations in merger review are significant: the Burger King matter. South African Law Journal, 141(3), pp. 454-466. doi:https://doi.org/10.47348/salj/v141/i3a2

Loxton, L. (2024) Merging the unmergeable: the distortion of competition law principles by constitutional rights in the Constitutional Court. Stellenbosch Law Review, 35(1), pp. 61-77. doi:https://doi.org/10.47348/slr/2024/i1a4

Mncube, L. and Ratshisusu, H. (2023) Competition policy and black empowerment: South Africa’s path to inclusion. Journal of Antitrust Enforcement, 11(1), pp. 74-90. doi:https://doi.org/10.1093/jaenfo/jnac015

Mondliwa, P., Goga, S. and Roberts, S. (2021) Competition, productive capabilities and structural transformation in South Africa. The European Journal of Development Research, 33(2), pp. 253-274. doi:https://doi.org/10.1057/s41287-020-00349-x

Msutu, T. (2025) From Massmart to Mediclinic (with a drive-thru Burger King): the development of the public interest standard in merger regulation. South African Mercantile Law Journal, 37(2), pp. 198-227. doi:https://doi.org/10.47348/samlj/v37/i2a4

Ncube, F. and Shumba, T. (2025) Defining excessive pricing in South African competition law: a European comparative analysis. Obiter, 46(2), pp. 410-435. doi:https://doi.org/10.17159/dhfyzn56

Ndlovu, P.N. (2025) The jurisdiction of competition authorities over peregrini respondent firms in South African competition law: revisiting the foreign currency cartel case. Journal of Comparative Law in Africa, 12(SE), pp. 267-297. doi:https://doi.org/10.47348/jcla/v12/2025-sea10

Nsomba, G. (2021) The Coca-Cola Company/Coca-Cola Beverages Africa merger: lessons for robust regional competition enforcement. The Antitrust Bulletin, 66(4), pp. 544-555. doi:https://doi.org/10.1177/0003603×211045436

Legislation and case law

Competition Act 89 of 1998 (as amended, including by the Competition Amendment Act 18 of 2018). Available at: https://lawlibrary.org.za/akn/za/act/1998/89 (Accessed: 18 October 2026).

Competition Commission of South Africa (n.d.) Retail Market Inquiry (Grocery Retail Market Inquiry). Available at: https://www.compcom.co.za/retail-market-inquiry/ (Accessed: 18 October 2026).

Competition Commission of South Africa v Mediclinic Southern Africa (Pty) Ltd and Another [2021] ZACC 35 (CCT 31/20), Constitutional Court of South Africa, 15 October 2021. Available at: https://www.saflii.org/za/cases/ZACC/2021/35.html (Accessed: 18 October 2026).

ECP Africa Fund IV LLC and Others v Competition Commission of South Africa [2021] ZACT 99 (IM053Aug21), Competition Tribunal of South Africa, 2021. Available at: https://www.saflii.org/za/cases/ZACT/2021/99.html (Accessed: 18 October 2026).