Skip to main content Skip to navigation

JILT 2008 (2) - Hlatshwayo

The Challenges of IP Protection and Competition Enforcement: An Analysis of the Microsoft Decisions (US and EU) and their Implications for South African IP and Competition Law

Nkonzo Hlatshwayo
Webber Wentzel incorporating Mallinicks, Johannesburg
Lex Informatica Conference, 21st -23rd May 2008
Pretoria, South Africa

This a conference paper published on 15 January 2009.

Hlatshwayo, N., ‘The Challenges of IP Protection and Competition Enforcement: An Analysis of the Microsoft Decisions (US and EU) and their Implications for South African IP and Competition Law’, 2008(2) Journal of Information, Law & Technology (JILT), <>

1. Introduction

Microsoft Corporation’s (Microsoft) pervasive influence in contemporary life, especially in the software industry, is beyond question.1A great majority of the papers presented at this conference were produced through various Microsoft software products. Those who had the occasion to undertake internet research most likely used Internet Explorer which came as part of the operating system bought with their computers. Indeed, many attending this conference may be forgiven for thinking that only Microsoft operating systems and programmes are readily available for use in the business world today. This is far from the truth.

As a result of this growth in influence and importance, software developers have sought to protect their control over their products. In this connection, protection is achieved through a variety of means, including intellectual property law (patents and copyrights). As a measure of control, software developers preclude those who purchase their software either as Original Equipment Manufacturers (OEMs), direct consumers or competitors from reverse engineering the software. Even in terms of use, those who purchase software only acquire the licence to use the software, which comes with a plethora of limitations on the way in which the software could be used. Importantly, software developers are particularly careful to shield the details of their creations from competitors or potential competitors.

The enforcement actions taken by the anti-trust agencies in the United States of America (US) and the European Union (EU) in the last few years underscore Microsoft’s influence over the software industry and the quest by the aforementioned agencies to preserve and/or maintain competition. What is, however, intriguing is that the Anti-trust Division of the Department of Justice in the US and the European Competition Commission (EC) in the European Union (EU) took divergent decisions in relation to essentially the same competition issues that arose in their respective jurisdictions, especially in respect of the remedies that were prescribed. The question is whether these differences flowed from mere jurisdictional approaches to these matters, or on the contrary, reflected deep seated ideological differences.

Given the stature and influence of both institutions in competition enforcement generally, and their influence over the enforcement approaches and policies of emerging markets such as South Africa, the question that arises is which enforcement policies are more suitable for South Africa.2The answer to this question will not only be dictated by the policy choices that South Africa may want to make, but are likely to be strongly influenced by the structure of the South African Competition Act itself.3Further, a number of regional developments could find South Africa looking to Europe for guidance on these matters.

April 2008 saw the inauguration of a regional competition authority in Africa, the COMESA Competition Commission. Although South Africa is not a member of COMESA, there is no doubt that its regulatory oversight will have a profound influence on South African companies doing business in Africa as well as on the South African Competition Commission.4More importantly, COMESA’s competition regime draws heavily from the EC model. For this reason, it must be expected that its enforcement policies, especially around IP and Competition issues are likely to be influenced by EC jurisprudence.

At a national level, South Africa has always had competition regulation from as early as 1955. However, the last piece of legislation that is probably most familiar to some is the Maintenance and Promotion of Competition Act of 1979. This is the legislation that was replaced by the current Competition Act, No. 89 of 1998. The review of the South African competition regime came at a time when the whole SADC region was adorning itself with competition legislation. The emergence of competition legislation in SADC gained momentum in the early 1990s. This was largely sponsored by the international lending institutions such as the World Bank and the International Monetary Fund. Consequently, Kenya, Tanzania and Zambia came up with competition legislation between 1993 and 1994. They were subsequently joined by Zimbabwe in 1996 and then Malawi and South Africa in 1998.

More recent converts include Namibia, 2003 and Swaziland 2007 and Botswana, which already has a Competition Bill. Although Namibia’s Competition Act went through Parliament in 2003, it only came into force on 3 March 2008. Lesotho is apparently still in the process of drafting its legislation which is likely to be passed within the next two or so years. Angola, the Democratic Republic of Congo and Mozambique have individually expressed an interest in drafting competition legislation for their respective jurisdictions. Whilst all this was happening, both Tanzania and Zimbabwe had the chance to review their legislation. While the Zimbabwean review was limited to a number of piece meal amendments, the Tanzanian review was far more extensive. In South Africa, there are clear indications that the current legislation will undergo a major review within the next few months. Government appears to be more concerned with what it calls ‘complex monopolies’ which are apparently responsible for some ‘uncompetitive outcomes’ in some markets. It is conceivable that one or both of these classifications may apply to IP related industries.

Of course, IP has assumed an increasingly important role in modern economies because of the significant technological advancements fuelled by innovation and creativity. This development prompts concerns about how the intellectual fruits of those who have taken the trouble to invest vast resources into Research and Development (R&D) in order to invent new products or discover new ways of doing things through the exploitation of their creative faculties are to be protected from the threat of free-riders and the demands of competition enforcement in its quest to advance consumer welfare through low or competitive prices.

Historically, those in possession of intellectual property rights (IPRs) have always insisted that the monopoly and exclusivity granted to them by law is beyond the reach of anti-trust enforcement. At a minimum this involved granting access rights to whoever they chose as well as the enforcement of their exclusive rights against anyone who attempted to use those rights without their express permission. This has created a private licensing regime in which private parties licence others to use their IP. This argument was set out in Microsoft’s case in the US in these words:

‘[T]o promote creativity, innovation, and competition, the federal copyright laws provide copyright holders such as Microsoft with broad and well recognised rights, rooted in the Constitution, to license their intellectual property as they see fit…’5

In fact, this is at the heart of the debate between IPRs holders and competition enforcement agents. This view was well articulated by the Chairman and founder of Microsoft, Bill Gates, when he said:

‘Microsoft is an intellectual property company. We have no factories of any consequence or natural resources. Indeed, we have no physical assets of any kind that are important to the success of the company. Our products instead consist almost entirely of information we create…Absent protection for intellectual property, there exists little reason to invest in developing software’.6

2. The Microsoft Cases in the US

In order to appreciate the differences in enforcement policy between the EU and the US when it comes to IP, we have decided to look at the Microsoft cases. We start with Microsoft's enforcement problems in the US. To be fair, Microsoft has faced an avalanche of law suits and investigations in its home country. In 1988, Microsoft faced a law suit from Apple about the look and feel of Windows. This was at a time when Microsoft was still smaller than Apple. The law suit was concluded in favour of Microsoft. The FTC itself was nursing an investigation into alleged collusion between Microsoft and IBM in terms of which Microsoft apparently agreed to limit the features of Windows in favour of OS/2. This investigation only became public in 1991 but collapsed in 1993 when the FTC failed to secure a majority ruling to initiate court proceedings against Microsoft.7

At the beginning of that year, 1993, Stac Electronics sought to enforce its IPRs against Microsoft in respect of planned disk compression software in DOS 6.0 on the basis that this violated Stac Electronics’ patents. At the beginning of 1994, Microsoft was ordered to remove disk compression software from DOS 6.0. This lawsuit was settled later in 1994 for $83 million.

Sometime in 1993, the DOJ started pursuing an investigation against Microsoft which looked at OEM licensing of Windows. In July 1994, Microsoft signed a consent decree with the DOJ which was turned down by the Court in February 1995. It was subsequently reconfirmed by the Court in August 1995. When rejecting the consent decree, the Court justified its decision in the following terms:

‘The issue before this Court is whether the entry of the proposed antitrust consent decree between Microsoft Corporation and the United States is in ‘the public interest.’ The Court cannot find the proposed consent decree to be in the public interest for four reasons. First, the Government has declined to provide the Court with the information it needs to make its proper public interest determination. Second, the scope of the decree is too narrow. Third, the parties have been unable and unwilling adequately to address certain anticompetitive practices which Microsoft states it will continue to employ in the future and with respect to which the decree is silent. Thus, the decree does not constitute an effective antitrust remedy. Fourth, the Court is not satisfied that the enforcement and compliance mechanisms in the decree are satisfactory'.

This created some complications for the DOJ in that whilst it was trying to rein in Microsoft’s competitive excesses, it was forced to appeal the court’s decision to turn down the consent decree together with Microsoft. Eventually, however, the consent decree was confirmed in August 1995.

In 1996 a fresh round of law suits against Microsoft started, beginning with Caldera which sued Microsoft on the ground that it used Windows to frustrate DR-DOS. That same year, the DOJ launched a new investigation against Microsoft in respect of unfair competition against Netscape. The following year, 1997, the DOJ revisited the consent decree of 1995, accusing Microsoft of violating its terms by bundling Windows with Internet Explorer. In 1998, the groundswell of opposition against Microsoft culminated in the filing of a law suit by the DOJ together with 20 other states for various violations of the anti-trust statutes in the US.

It is the latter case that is of interest to us in this paper. As already indicated, the case against Microsoft in the US was brought simultaneously by the Anti-trust Division of the DOJ together with twenty states as well as the District of Columbia. The principal allegations against Microsoft were that it sought to monopolise the market for operating systems in respect of personal computers. In this regard, it pursued a number of anti-competitive strategies aimed at maintaining this monopoly. Further, it sought to monopolize the market for internet browsers as there was an expectation that the internet browsers’ market could be the launch pad for a competitive onslaught on Microsoft’s operating system. Consequently, Microsoft decided to bundle its dominant Windows operating system with its internet browser, Internet Explorer. The strategies adopted by Microsoft were also aimed at frustrating the development of another layer of software called middleware, which worked between Microsoft’ s operating system and software applications. Microsoft allegedly adopted this position because it apparently realised that the development of middleware software posed a competitive threat to its operating system.

One of the clearest examples of Microsoft’s tactics to frustrate the emergence of competition was its conduct towards Netscape’s internet browser, Navigator.8In order to dismantle the threat of Navigator, Microsoft simply decided to bundle Internet Explorer with Windows. Consequently, a PC that came with a Windows operating system also came with Internet Explorer pre-installed, which made it unnecessary for the purchaser to look for an additional internet browser. In order to accomplish this, Microsoft imposed contractual obligations on OEMs which forbade them from modifying or deleting any part of its operating system. In fact, it is alleged that Microsoft threatened ‘ PC manufacturers with revoking their licenses to distribute Windows if they removed the Internet Explorer icon from the initial desktop, something that Netscape had requested of its licensees’. Eventually, Microsoft improved on this scheme by simply making it more difficult for customers to delete or uninstall IE without adversely affecting the vital components of the operating system. In addition to the initial IE scheme, Microsoft also placed restrictions on OEMs from modifying certain aspects of Windows. For instance, they were precluded from altering the boot sequence, appearance of the Windows desktop, removing desktop icons and folders and removing start menu entries (First, 2006).

Microsoft’s principal argument before the US District Court when it applied for summary judgment was that the federal copyright laws in force in the US allowed it to place the restrictions contained in its OEM agreements. In other words, these agreements simply confirmed the rights already recognised by the federal laws referred to; that is, the rights contained in these agreements were consistent with the IPRs owned by Microsoft and were recognised by the federal laws.

Microsoft had a number of sympathisers. An open letter written to President Clinton by some 240 Economists on Antitrust Protectionism stated in part that:

‘Consumers did not ask for these antitrust actions - rival business firms did. Consumers of high technology have enjoyed falling prices, expanding outputs, and a breathtaking array of new products and innovations. ... Increasingly, however, some firms have sought to handicap their rivals by turning to government for protection. These cases are based on speculation about some vaguely specified consumer harm in some unspecified future, and many of the proposed interventions will weaken successful U.S. firms and impede their competitiveness abroad' <>.

In response to Microsoft’s arguments, the Court reasoned that the critical question was not the validity of the copyrights as such, but rather, what precisely, the copyrights were meant to protect (First, op cit.). In particular, the Court pointed out that ‘Microsoft’s actions had nothing to do with protecting the integrity of its artistic work; they had everything to do with suppressing the competitive threat that middleware presented’ (Ibid). In this regard, Judge Jackson stated that Microsoft's dominance of the personal computer operating systems market constituted a monopoly. He also pointed out that Microsoft had engaged in conduct that sought to crush any competition to the monopoly it enjoyed. The competitors that were at the receiving end of Microsoft’s anti-competitive conduct included Apple, Java, Netscape, Lotus Notes, Real Networks, Linux as well as many others. The most drastic aspect of the case was the remedy that was imposed by Judge Jackson, which required Microsoft to be broken into two separate units, one to produce the operating system, and the other one to produce other software components.

After the Judge Jackson’s decision, Microsoft sought to appeal directly to the Supreme Court, which refused to entertain the appeal and therefore referred the matter back to the DC Circuit Court of Appeals. Although the Court of Appeals overturned Judge Jackson's decisions, especially his remedy against Microsoft, it is important to state that it did not overturn Judge Jackson’s findings of fact. The DC Circuit Court then remanded the case for consideration of a proper remedy and this task fell on Judge Colleen Kollar-Kotelly. However, before the new remedies could be considered, the DOJ abandoned its initial remedy to break up Microsoft. Instead, it accepted an agreement with Microsoft that required the latter to share its application programming interfaces with third-party companies. The agreement also provided for the appointment of a panel of people to have full access to Microsoft's systems, records, and source code for five years in order to ensure compliance. Importantly, Microsoft was not required in terms of this agreement to change any of its code. Microsoft was also not precluded from tying other software with Windows in the future. At the end of 2002, Judge Kollar-Kotelly issued a judgment that embodied most of the settlement agreement between the DOJ and Microsoft.

This development marked a significant victory for Microsoft in that whilst it was compelled to share some of its IP with competitors, it was saved from the more drastic remedy prescribed by Judge Jackson. Further, the settlement agreement was by no means unanimous. Nine of the states that had started the case together with the DOJ took the view that the settlement was insufficient to deal with the competition issues that had been identified. In 2004, the US Appeals Court accepted the settlement agreement, brushing aside outstanding objections from Massachusetts that the sanctions were still inadequate.

3. The Microsoft Cases in the EU

In reality, the anti-trust cases in the US and the EU started at about the same time. Whilst the US cases started in 1991, it was only two years later that a company called Novell argued that Microsoft was elbowing its competitors out of the market through a variety of anti-competitive practices. This complaint was founded on Microsoft’s licensing practices of the time, which required royalties in respect of each computer sold by a supplier of Microsoft's operating system, even if the unit sold did not actually contain the Windows operating system. However, this matter did not give rise to anti-trust enforcement action for Microsoft because it chose to reach a settlement in 1994, in terms of which it agreed to abandon the offending licensing practices.

Some four years later, Sun Microsystems, one of the companies that had fought various legal battles with Microsoft in the US, initiated a complaint with the EC regarding Microsoft’s refusal to disclose some of the interfaces to Windows NT. Further, Microsoft faced an investigation in respect of another bundling case, this time involving its streaming media technology, Media Player, which was integrated with Windows. The EC’s main concern was the ‘possible leveraging of the monopoly power enjoyed in one market to other markets through the control of interfaces’ (Correa, 2007). It therefore transpired that Microsoft was actually using its near monopoly in the market for operating systems for personal computers in order to extend its monopoly into the markets for operating systems for group servers and for media players.

At the liability stage, the Commission found Microsoft to have violated the EC anti-trust provisions. In particular, the EC took the view that the information required by Microsoft’s competitors was in fact necessary for effective competition in the market for work group server operating systems. The information sought would provide rivals with the ability to interoperate effectively with Microsoft’s operating systems.

‘Consequently, [the Commission] said that Microsoft’s refusal to supply interoperability information was abusive because 1) interoperability information is needed by competitors in the market for work group server operating systems to viably stay on the market [Emphasis Supplied]: 2) Microsoft’s conduct involved a disruption of previous levels of supply; 3) there was ‘a risk of eliminating all competition in the work group server operating system market’; 4) the refusal to supply had the consequence of ‘preventing innovation in the work group server market and of diminishing consumers’ choice by locking them into a homogeneous Microsoft’s solution;’ and 5) the refusal was not objectively justified because on balance ‘negative impact of an order to supply on Microsoft’s incentives to innovate was outweighed by its positive impact on the level of innovation of the whole industry (including Microsoft) (Czapracka, 2007)'.

Consequently, when the Commission considered the appropriate remedies to deal with the anti-competitive conduct perpetrated by Microsoft, it ordered Microsoft to provide the information necessary for competitors who produced competing networking software to interact fully with Windows desktops and servers. In addition, it ordered Microsoft to provide a version of Windows that did not have Windows Media Player embedded in it. The EC also levied a very heavy fine on Microsoft for the conduct in question.

As expected, Microsoft vigorously objected to this remedy and appealed the decision to the Court of First Instance (CFI). Microsoft argued that the decision compelling it to share information with competitors in respect of its software ignored the fact that ‘it had invested substantial sums of money in developing the interfaces and protocols at issue – interfaces and protocols which it regarded as its intellectual property (IP). Microsoft also contended that the order to share information with competitors would, among other things, reduce its return on investment in respect of its R&D. In addition, Microsoft warned the CFI about the dangers of compulsory licensing of the interfaces and protocols at issue in this case which could ‘set a precedent that would have significant negative effects on future incentives to invest in development of IP.’9

Although Microsoft appealed the EC’s decision, it lost it in September 2007. In the process, Microsoft incurred severe penalties for failing to comply with the Commission’ s initial order on interoperability. As at the beginning of 2008, Microsoft was fined a record E899 for failing to comply with the original order. Microsoft has decided to appeal this latest decision.

4. The Main Differences between US and EU Anti-trust Enforcement in Respect of IP

Whilst on the surface, it may appear that both the US and the EU came to the same conclusion regarding Microsoft’s conduct in respect of its refusal to make its IP available to competitors as well as its practice of bundling its operating systems with some of its less successful products, there is in fact a world of difference between the two jurisdictions. This difference is echoed by the press release issued by the DOJ’s Assistant Attorney General for Anti-trust, R. Hewitt Pate, when the EU issued its decision:

‘The United States’ Final Judgment provides clear and effective protection for competition and consumers by preventing affirmative misconduct by Microsoft that would inhibit competition in ‘middleware’ programs, such as the web browser that was the subject of the United States’ lawsuit and the media player that is the subject of the EC’s action today. The Final Judgment, for example, prohibits the use by Microsoft of exclusive contracts or other provisions that inhibit competition, prohibits anticompetitive manipulation of icons and default settings, and requires Microsoft to provide information to allow ‘ interoperability’ of competitors’ software. The United States continues to be active in its enforcement of Microsoft’s compliance with the Final Judgment, and this work has resulted in substantial changes to Microsoft’s business practices.
‘The EC has today pursued a different enforcement approach by imposing a ‘code removal’ remedy to resolve its media player concerns. The U.S. experience tells us that the best antitrust remedies eliminate impediments to the healthy functioning of competitive markets without hindering successful competitors or imposing burdens on third parties, which may result from the EC’s remedy. A requirement of ‘code removal’ was not at any time -- including during the period when the U.S. was seeking a breakup of Microsoft prior to the rejection of that remedy by the court of appeals -- part of the United States’ proposed remedy.
‘Imposing antitrust liability on the basis of product enhancements and imposing ‘code removal’ remedies may produce unintended consequences. Sound antitrust policy must avoid chilling innovation and competition even by 'dominant' companies [Emphasis supplied]. A contrary approach risks protecting competitors, not competition, in ways that may ultimately harm innovation and the consumers that benefit from it. It is significant that the U.S. district court considered and rejected a similar remedy in the U.S. litigation.’

What emerges from Pate’s statement is that the disclosure requirements imposed by the EC on Microsoft were far-reaching than those agreed to between Microsoft and the US government. In particular, the US government was concerned that the disclosure requirements imposed by the EC were likely to be harmful to the innovative process by lowering innovation incentives. Further, the US government appeared to be concerned about the fact that the EC’s remedy focused on protecting competitors as opposed to competition.

The DOJ was also shocked by the fine imposed on Microsoft:

‘While the imposition of a civil fine is a customary and accepted aspect of EC antitrust enforcement, it is unfortunate that the largest antitrust fine ever levied will now be imposed in a case of unilateral competitive conduct, the most ambiguous and controversial area of antitrust enforcement. For this fine to surpass even the fines levied against members of the most notorious price fixing cartels may send an unfortunate message about the appropriate hierarchy of enforcement priorities’ <>.

As stated above, Microsoft attracted even higher fines from the EC for failing to comply fully with the EC’s decision of 2004. The best way to contextualise these differences between the EC and the US is to appreciate the broad policy differences that the two jurisdictions have in respect of anti-trust enforcement.

One of the US’s clearest principles underpinning anti-trust enforcement is that enforcement action must be aimed at protecting consumers from market failures which may be evidenced by a number of factors such as inferior service levels, prices above the competitive level, restrictions of output, etc. Enforcement action must be taken in order to sort out market failures. It must have nothing to do with the protection of market participants, except where firm conduct has been directed at some market participants with a view to compromising their competitiveness. This principle lies at the heart of US anti-trust enforcement. As a result, where a trade off analysis is required in order to balance the outcomes of anti-competitive conduct by a firm against the efficiencies likely to be realised as a consequence of the anti-competitive conduct, the enforcement agencies give due regard to efficiencies in determining whether consumer welfare would be harmed.

On the other hand, EC anti-trust enforcement, whilst acknowledging the importance of enforcement action aimed at protecting consumers, stresses the importance of the ‘protection of rivalry and the competitive process’ as opposed to pro-competitive efficiency gains (Czapracka, op cit.). Consequently, the EC, unlike the US, is likely to be more sympathetic to a competitor who complains against anti-competitive conduct in respect of another competitor. This is aptly put by Czapracka when he says:

‘[A] US official suggested that whereas the US system supports ‘cowboy capitalism’, allowing [a] monopolist to compete aggressively on the merits even if it entails injuries to its rivals, the Europeans require dominant firms to ‘compete like gentlemen’ (Ibid).

The difference set out above is not merely one of style; it runs to the core of the differences in enforcement approaches between the two jurisdictions. This difference in approach also explains why the two jurisdictions approach the issue of market power somewhat differently. Whilst the Sherman Act seeks to protect competition by prohibiting the acquisition or maintenance of monopoly power, Article 82 seeks to regulate the conduct of companies in dominant positions (Ibid). Consequently, dominant firms bear certain responsibilities towards their competitors. For this reason, conduct that would otherwise be acceptable when pursued by a firm that is not considered to be in a dominant position may in fact violate the EC’s abuse of dominance provisions if perpetrated by a dominant firm.

Central to the issue of monopolization and dominance is the whole question of market power and what it means for competition. Market power is normally defined as the ability to act independently of one’s competitors which is often evidenced by the ability to raise prices above the competitive level or restrict output without inducing any reaction from competitors. How market power is assessed can be quite controversial depending on where you are. In the US market power is assessed on the basis of market shares as well as a whole host of factors that indicate a firm’s ability to raise prices and restrict output. However, in the EC, it appears that a less stringent assessment may be pursued. In fact, market shares in the ECU are a significant indicator of market power. Given South Africa’s dominance provisions that appear to have been drawn from EC jurisprudence, it is not surprising that in South Africa, a firm can be held to be dominant on the basis of market share alone, even if, factually, it has no market power.10

What is clear with the EC’s approach generally is that dominant companies bear certain responsibilities towards their competitors. This is not necessarily the case in the US. Czapracka argues that this approach seems to be premised on the EC’s commitment to the essential facilities doctrine. Whilst this doctrine exists in the US, it is said that no higher court has been willing to fully endorse it:

‘[N]o court of last resort has so far accepted a claim for access to the subject matter of IPRs on the basis of the essential facilities doctrine. Treating IPRs and other forms of property equally would imply that the essential facilities doctrine is applicable to IPRs in principle; but its actual application in an individual case would depend upon the scope of the doctrine in general, which is not clear’. 11

However, the EC position is vastly different:

‘Whereas the essential facilities doctrine has been questioned in the US, it has been steadily growing in significance in Europe…The Commission concedes that forced sharing may have adverse effects on investment incentives, but notes its beneficial influence on competition in the secondary market and investment in follow-on R&D. It identifies four situations in which refusal to deal violates EU competition law: 1) terminating an existing supply relationship; 2) refusing to supply an essential input; 3) refusing to supply information protected by IPRs; and 4) refusing to supply information necessary for interoperability’ (Czapracka, op cit.).

5. South Africa’s Competition Enforcement and IP

In view of the fact that the South African Competition Act is more recent than similar legislation in the US and the EU, it draws inspiration from both jurisdictions. However, in respect of the treatment of IP rights when they collide with competition law, it appears that it is strongly influenced by the EC model. Before we deal with the level of inspiration that it draws from the EC jurisprudence, it is as well to deal with some preliminary matters.

The Competition Act recognises about six pieces of legislation providing for the registration and protection of intellectual property rights:

  • the Perfomers’ Protection Act,12
  • the Plant Breeder’s Rights Act,13
  • the Patents Act,14
  • the Copyright Act,15
  • the Trade Marks Act,16and
  • the Designs Act.17

In terms of section 10(4), any firm may apply to the South African Competition Commission (Commission) for any exemption in respect of any conduct which is regulated by Chapter 2 of the Competition Act. Chapter 2 deals not only with horizontal and vertical practices but also abuse of dominance issues.

The debate on the interface between IP and competition law in South Africa is taking place at a time when our jurisprudence on the right balance between the enforcement of competition law and the exercise and protection of IP rights has yet to emerge. Few cases have been handled by the competition authorities in this regard. Many of the cases have been investigated by the Commission and settled with the respondents, without the issues raised in them being fully interrogated through the adjudicative tribunal proceedings. More importantly, the Commission’s primary function in these complaints is investigative, not adjudicative. So its views are somewhat coloured by its investigative function and may often be on the robust side.

One of the most notable cases raising IP issues to come before the competition authorities was Hazel Tau & others v.GlaxoSmithkline & Boehringer Ingelheim18.This matter, which is dealt with below, was settled by the Commission and the respondents.

The consent orders concluded between the Commission and some of the respondents are not entirely helpful, in the sense that the respondents do not necessarily admit guilt. In fact their motivation to conclude a consent order is not necessarily actuated by guilt. In many instances, other considerations come into play, which convince the respondents to prefer a consent order rather than fighting a prosecution from the authorities. Further, the Tribunal is entitled to deal with a consent order on application by the Commission and the respondents, without the necessity of calling for any evidence in terms of s 49D(1).

The debate is also affected by the fact that there are no guidelines on the enforcement of competition law in respect of the exercise of IPRs by the competition authorities. The absence of such guidelines is obviously problematic as business remains unclear on what factors would be considered by the competition authorities in dealing with the exercise of IPRs. However, the Commission authored a brief article that was published in 2001 in which it attempted to reveal the considerations that would guide the Commission in its assessment of the exercise of IPRs.19

The Competition Act makes it plain in section 3(1) that its provisions apply to all economic activity that has an effect in South Africa, including IPRs and their exercise. However, the exemption provisions of the Act refer to the fact that an IPRs right holder may apply for an exemption from application of Chapter 2 of the Act where the right in question flows from the six pieces of legislation set out above. It is crucial to understand that an exemption application will not automatically be granted, regardless of the fact that it flows from an IPRs. There are clear requirements that have to be met before the Commission could take the view that the application is justified.

As suggested above, the South African position in respect of the enforcement of IPRs in relation to competition law is heavily influenced by the EU position. The Competition Act contains abuse of dominance provisions that are applicable if the relevant thresholds (financial and market share) are met. In fact, some of these provisions have been used in South Africa to challenge some pharmaceutical companies in respect of the pricing of their ethical drugs. The most relevant section in this regard is s 8 of the Act. Many of its subsections apply directly to the exercise of IPRs.

The Act defines dominance in section 7. So, unless the exercise of your IPRs places you within the four corners of that definition, you may not be caught by the Act. For this reason, before a determination of whether the alleged conduct shows the abuse of dominance, a proper market must be defined in terms of which a determination must be made on whether you are in fact dominant. In other words, the ordinary competition assessment in respect of dominance has to be carried out in respect of the exercise of the IPRs. It is therefore important to clarify that ‘IP ownership does not automatically confer a dominant position, but it is possible for an IP right to result in or contribute to a dominant position’.20 Consequently, the fact that you are a holder of IPRs does not invariably signify dominance in South Africa.

Section 8(a) provides that a dominant firm is precluded from charging an excessive price, this being a price that bears no reasonable relation to the economic value of the good or service.21 This is an outright prohibition that allows no defence.

Of course the concept of an excessive price has proved to be one of the most difficult concepts of the Act to enforce, because of the uncertainty in determining its precise economic meaning. As many economists would point out, a high price is not necessarily excessive. In fact, they add, an excessive price is not an economic term but something that probably belongs to the public interest provisions of the Act. However, a number of complaints have been brought before the competition authorities on the basis of this provision, some of which related to the exercise of IPRs.

A complaint was lodged against some pharmaceutical manufacturers on the basis of section 8(a) of the Act. The allegation was that their prices of ARVs were excessive. Unfortunately, the matter was never fully resolved in the sense that, before the prosecution of the parties by the Commission, they negotiated a settlement agreement in terms of which they admitted no liability.

Section 8(b) precludes a dominant firm from refusing to give a competitor access to an essential facility when it is economically feasible to do so. This is a provision that brings us much closer to the EC position. Although the provision is often used in the context of network industries, there is a suggestion that it may also apply to IPRs, especially where the IPRs in question hold answers to the competitive participation of others in the market or a related market. In terms of the Act, ‘ essential facility’ means ‘an infrastructure or resource that cannot reasonably be duplicated, and without access to which competitors cannot reasonably provide goods or services to their customers’.22

On the face of it, this provision permits competitors to compel an IPRs holder to license the rights to them if this is reasonably necessary to allow them to provide goods or services to their own customers. However, this provision has not yet been tested fully, although it is clear that it might be used as a ground for compulsory licensing. As indicated earlier, the pharmaceutical cases that were settled by the Commission in December 2003 do not provide a good example, because the issues were not conclusively determined by the Competition Tribunal. What happened in these cases is that Hazel Tau, an AIDS activist, complained to the Commission about an abuse of dominance by GlaxoSmithkline and Boehringer Ingelheim for charging excessive prices for their ethical drugs. At the time of the complaint, GlaxoSmithkline enjoyed patent protection over three drugs central to the fight against AIDS: AZT, 3TC and Combivir. Boehringer Ingelheim had patent protection over the drug Nevirapine.

At the conclusion of the investigation, the Commission announced that it was referring the matter to the Competition Tribunal (Tribunal) for adjudication.23The parties were found by the Commission to have contravened the abuse of dominance provisions of the Act in that they had24-

  • denied a competitor access to an essential facility;
  • charged excessive prices; and
  • engaged in exclusionary acts

In this connection, the Commissioner of the Competition Commission, Mr Menzi Simelane stated that:

‘Our investigation revealed that each of the firms has refused to license their patents to generic manufacturers in return for a reasonable royalty. We believe that this is feasible and that consumers will benefit from cheaper generic versions of the drugs concerned. We further believe that granting licenses would provide for competition between firms and their generic competitors.’25

However, before the commencement of the proceedings before the Tribunal, the Commission and the parties reached a settlement agreement. In terms of this agreement, the two pharmaceutical companies agreed to license generic manufacturers to produce these drugs for the South African market. Mr Simelane justified the conclusion of the settlement agreement in the following terms:

‘The terms of the agreements are substantially similar to the successful outcomes which we would have hoped to achieve at hearings before the Tribunal, namely the issuing of licenses to generic manufacturers of antiretroviral drugs. It has been a particularly difficult case and we are happy that the matter has been amicably resolved.’26

The reason why this case does not provide a useful precedent is because the Commission’s press release does not provide the analysis that the Commission conducted in coming to the conclusion that the refusal to licence competitors constituted a refusal to licence an essential facility. What is evident, however, is that the Commission seems to be prepared to follow the route of the EC in using this doctrine to give competitors access to IPRs belonging to a competitor.

Another essential facilities case involved Druggist Distributors (Pty) Ltd (DD) and the National Association of Pharmaceutical Wholesalers.27One of the complaints made in this case was that a number of pharmaceutical companies decided to use DD to distribute their products as they had done in the past. The complainants, which included several pharmaceutical wholesalers as well as their association, argued that the use of DD in preference to them amount to denying them access to an essential facility. The Competition Appeal Court (CAC) held that the conduct complained of was not that fell within the purview of an essential facility but was most probably a refusal to supply a competitor with scarce goods or services. In particular, the Court cautioned against the use of the essential facilities provision except in appropriate circumstances:

‘The legislature intended, from the architecture of the Act, that there should be limits to the essential facilities doctrine. To demand that a dominant firm should grant access to its facilities is a substantial intervention on the part of a competition authority. The widening of the application and scope of the essential facilities doctrine can have harmful economic effects such as discouraging investment in infrastructure. An investor might be reluctant to invest for fear of a third party demanding a 'free ride' on the fruits of such investment.’

A case that would have been extremely helpful in this regard is DW Integrators CC (DW) and SAS Institute (Pty) Ltd (SAS).28This is a case that raised the essential facilities doctrine in the context of software licensing. The Tribunal was asked to grant an interim relief order in favour of the applicant, DW, pending the conclusion of the Commission's investigation in respect of a complaint lodged by DW. DW was providing consulting services to the licensees of SAS software programs. It therefore claimed that it could not effectively provide its services to its clients without possessing a licence in respect of SAS software. SAS's refusal to grant it a licence, the applicant contended, was preventing it from participating in the market. In particular, the applicant argued that it was being refused access to an essential facility.

The application was dismissed because the applicant failed to demonstrate that the respondent, SAS, was in fact in a dominant position in the market for information delivery software. The case therefore helps to clarify the position that dominance in our law is a critical element to the essential facilities doctrine.

6. Conclusion

What we have considered about are the circumstances under which the competition authorities in the US and the EU would be willing to compel an IPRs holder to share copyright or patent protected information with third parties, including competitors to allow them to compete. As noted, whilst this is possible in the US, it is a position that is accepted reluctantly. In the EU, however, we noted that the EC is more readily prepared to order an IPRs holder to share its IP protected information with competitors to allow them to compete.

As indicated above, the South African position is not very different from the EU position. However, as noted above, only a few cases have raised the essential facilities doctrine. What is clear is that the Commission is ready to argue for compulsory licensing in respect of IPRs. The Tribunal as well as the CAC appear to have sounded caution on this approach. It is probably still early days before we know for sure how the authorities will deal with a hard core compulsory licensing request in respect of the software industry. Although the DW case was dismissed, it must be stressed that this was based on the fact that dominance had not been proven. Consequently, it was unnecessary for the Tribunal to consider the policy question whether to allow compulsory licensing in respect of IPRs.


1 As observed by Nicholas Economides, Microsoft produces a wide range of software products including its famous operating systems for PCs (Windows) as well as operating systems for local network and internet servers. In addition, it produces what may be called back office products for network and internet serves, internet clients, internet and network servers, its successful desktop applications (Office, Word, Excel, Access, Outlook, Powerpoint, etc).

2 Both the DOJ and the European Competition Commission were founder members of the International Competition Network, a loose association of competition authorities that meets on an annual basis to exchange views about enforcement approaches and options. Of late, the ICN has been instrumental in setting the agenda for the harmonisation of enforcement principles and investigative practices. Both the EU and the US are very influential in seconding their staff members to other agencies to share experiences. For instance, South Africa was a recipient of a number of highly experienced economists and lawyers who came to South Africa to impart their skills and knowledge to staff members of the South African Competition Commission. More recently, other secondees from Europe have been to South Africa to do the same thing. In turn, South Africa has hosted a number of officials from other Competition Authorities in Africa who came to South Africa to learn how to run investigations more efficiently.

3 Act No. 89 of 1998. It has been noted elsewhere that the Competition Act draws inspiration from a number of jurisdictions. For instance, its merger review provisions were borrowed from the Canadian Competition Act of 1986, which the definition of a merger lifted also verbatim from that Act. However, it is noteworthy that a number of key concepts in relation to refusals to deal and essential facilities seem to have been borrowed from Europe. Consequently, the structure of the Competition Act may force the South African Competition Authorities to look to Europe rather than the US for intellectual and policy direction.

4 When the idea of a regional competition authority was first mooted in a regional competition conference held in Livingstone, Zambia, in August 2000, the South African delegation argued that a regional competition authority under the auspices of Comesa, which excludes South Africa, was going to be problematic in view of the fact that it was going to leave out the biggest economy in the region. The conference therefore resolved that both SADC and COMESA should work together to create a regional competition authority. At the time of writing, SADC is said to be working on putting together its own regional competition authority. Unless as protocol is struck by these two institutions, it is likely that many companies in South Africa will be subject to more than two competition authorities at any one time.

5 Microsoft 84 F. Supp. 2d 9 ( Civil Action No. 98-1232 (TPJ)), 1998 WL 34201988 at 29

6 Microsoft, 224 F. Supp. 2d at 228

7 The decision of the FTC commissioners was split 2-2 and thus no action could be taken against Microsoft.

8 A representative of the DOJ presented an interesting argument on TV in the mid 1990s as the case was raging on. He took a quick poll from the audience to determine the number of people who used Navigator as an internet browser. At that time, close to 80% of the audience used Navigator. However, today, very few people in South Africa have heard about Navigator.

9 CRA International – Competition memo: December 2007

10 As will be shown below, the South African Competition Act sets various market share thresholds for the determination of dominance. Market power is only a factor if the company in question has less than 45% of the market. Where the company has over 45% of the market, there is an irrefutable presumption that it is dominant, regardless of the question of market power.

11United Nations Conference on Trade and Development (UNTAD) Report of the Intergovernmental Group of Experts on Competition Law and Policy (3 to 5 July 2002, at 16.

12 Act 11 of 1967.

13 Act 15 of 1976.

14 Act 57 of 1978.

15 Act 98 of 1978

16 Act 194 0f 1993

17 Act 195 of 1993

18 Competition Commission's Media Release No. 33 of 2003, 16 December 2003, accessible online at: <>

19 Competition Commission of South Africa ‘Competition News, Edition 4, June 2001’ at 5, accessible online at: <>

20 Ibid.

21 Sec. 1 of the (South African) Competition Act

22 Ibid

23 Competition Commission's Media Release No. 29 of 2003, 16 October 2003, accessible online at: <>

24 The Competition Commission enjoys investigative powers in respect of investigations under Chapter 2 of the Act. This Chapter deals with restrictive practices and the abuse of dominance provisions of the Act fall under this chapter. Once the investigation is completed, the Commission has one of two options: to refer the matter to the Tribunal, which signals its intention to prosecute the respondents before the Tribunal, or to issue a Notice of Non-referral, which means that the Commission has found no contravention of the Act and that for this reason, it will not proceed with the prosecution of the case. A complainant in the latter case is then entitled to prosecute the matter before the Tribunal if it wishes to do so. Consequently, the finding by the Commission that a respondent has contravened provisions of Chapter 2 and the referral of the matter merely signals the beginning of prosecution proceedings. It is in this context that the Commission proceedings must be understood.

25 Op cit

26Competition Commission's Media Release No. 33 of 2003, 16 December 2003, accessible online at: <>

27 Case No. 15/CAC/Feb 02. This was a case against a number of pharmaceutical companies in relation to their decision to use Druggist Distributors as opposed to pharmaceutical wholesalers. In the past the latter were used for the distribution of pharmaceutical products.

28 Case no. 14/IR/NOV99


First, H (2006), ‘Microsoft and the evolution of the intellectual property concept’, 2006(5) Wisconsin Law Review, pp 1370-1432

Correa, Carlos M (2007), ‘ Intellectual Property and Competition Law’, ICTSD Programme on IPRs and Sustainable Development

Czapracka, K A (2006/2007), ‘Where anti-trust ends and IP begins – on the roots of the transatlantic clashes’, 9(Fall Issue) Yale Journal of Law and Technology, p 44