More than any other case, Microsoft brings to a head the emerging debate over competition law's proper place in rapidly changing high-tech industries, where encouraging innovation may be much more crucial than keeping prices low. The case also raises thorny questions about the interface between competition law and intellectual property law in the burgeoning knowledge-based economy.
The Microsoft Proceedings in a Nutshell
While attention has been directed to the Department of Justice's (DOJ) current battle with Microsoft, the saga dates back to 1994. The DOJ's initial case alleged that Microsoft unlawfully maintained a monopoly in the market for PC operating systems through three main anti-competitive practices:
. penalizing original equipment manufacturers (OEMs) that did not pre-install Microsoft software on their computers,
. forcing unreasonably long contracts on OEMs, and
. tying the early access of other software companys to test versions of Windows 95 to their not working extensively with Microsoft's competitors.
The case was settled by a Consent Order, which was approved in August 1995 by Judge Thomas Pennfield Jackson, the same US District Court judge sitting on the current case. The Order prohibited Microsoft from continuing with these anti-competitive acts and from taking other actions that could have similar results. Although the Order specifically barred Microsoft from conditioning the licensing of Windows on OEMs purchasing another Microsoft product or agreeing not to purchase from competitors, it expressly allowed Microsoft to develop "integrated products."
The Consent Order was negotiated before the Internet took off. As the Internet's popularity grew, Microsoft realized its potential to threaten Windows as the dominant operating system for PCs. Microsoft took decisive action. The company's decision to bundle its newly developed browser software with its Windows 95 operating system led the DOJ to launch contempt proceedings. The US District Court issued a preliminary injunction in 1997, but the US Court of Appeals stayed and later removed the injunction on the basis that Microsoft's operating system and browser constituted an integrated design, which was permitted under the Order.
Undaunted, the DOJ (supported by twenty state attorneys general) filed a fresh action against Microsoft. Considering its size, the case moved through trial in record time. To hasten resolution, Judge Jackson decided to issue his decision in three parts: findings of fact (issued in November 1999), conclusions of law (anticipated any day now), and a separate remedy phase if needed. In the meantime, Judge Jackson has urged the parties to settle. However, the appointment of a celebrated jurist and antitrust guru, US Court of Appeals Judge Richard Posner, as mediator has produced no breakthroughs thus far.
The Findings of Fact
In 207 pages of findings, Judge Jackson paints a picture of Microsoft's unrelenting campaign to maintain its monopoly in PC operating systems at all costs. In blunt language, the findings describe how Microsoft used bullying tactics to trample its competitors.
Judge Jackson focused on a relevant market defined as the licensing of Intel-compatible PC operating systems worldwide. He concluded that an "applications barrier to entry" protects Microsoft's existing dominance in this market. This "chicken and egg problem" surfaces because consumers generally buy the operating system that has the most applications written for it, and developers write first (and at times only) for the system boasting the most users. The outcome is a "positive feedback loop" (or, what economists would call "network effects") which creates an applications barrier that makes it difficult for a new player to enter the market for PC operating systems successfully.
According to the findings, Microsoft systematically targeted two major threats to the applications barrier: Netscape's Navigator browser and Sun Microsystems' Java programming language. Each opened the possibility for cross-platform applications, able to run on non-Microsoft operating systems without conversion or "porting".
Netscape's Navigator browsing software, which had become synonymous with the Internet, alarmed Microsoft. When Netscape rebuffed Microsoft's invitation to divide the market, Microsoft retaliated by withholding information needed to develop a Windows 95-compatible version of Navigator in time for Windows 95's release. Next, Microsoft embarked on a plan to increase the market share of its own Internet Explorer browser at Navigator's expense. Microsoft distributed Explorer for free, forcing Netscape to do the same. Microsoft also excluded Navigator from important distribution channels by bundling Explorer with Windows using various techniques.
Because it was designed to function on all major operating systems, the Java language also disturbed Microsoft. In response, Microsoft sank substantial resources into creating Java implementations for Windows that undermined portability and were not compatible with other implementations. Microsoft also bundled its "Java virtual machine" (which makes Java comprehensible to the underlying operating system) with Explorer.
Judge Jackson found that Microsoft's anti-competitive conduct effectively reduced Navigator's market share in favour of Explorer's share, and greatly impeded the progress of Java implementation. Microsoft's actions had no business purpose except to protect the applications barrier to entry and the Microsoft operating systems monopoly. Notably, however, Judge Jackson did not conclude that Microsoft's market power would have been eliminated but for its conduct. This may have important implications during the remedy phase and the almost inevitable appellate review of the decision.
While acknowledging one positive effect of Microsoft's actions (free access to both Navigator and Explorer), the findings focused on potentially significant injuries to consumers. These included forcing OEMs to ignore consumer preferences for Navigator and demand for a browserless version of Windows, as well as depriving the public of software innovation that might otherwise have occurred. At this point, it seems very likely that Judge Jackson will rule that Microsoft violated US antitrust law. The key question is what will the court do about it?
In Search of Remedies
Assuming Microsoft does not settle, Judge Jackson will face the formidable challenge of fashioning a remedy from three main options: behavioural orders, compulsory licensing of intellectual property, or structural relief.
The first Microsoft case embedded behavioural remedies relating to contractual practices in an injunction, with an enforcement mechanism attached. Unfortunately, behavioural orders may require ongoing monitoring and enforcement, which can be complex and burdensome, particularly in a rapidly changing industry. For example, neither Microsoft nor the DOJ was certain about how the Consent Order applied to Explorer (released just months later) or updated versions of Windows. The Order also left other creative competitive practices open to Microsoft. The second case can be seen as the result of these failures to impose effective behavioural remedies.
If the court chooses to use intellectual property licensing as a remedy, it could require Microsoft to license the source code underlying Windows, and possibly other software, to competitors. Such a remedy should allow others to design compatible operating systems and break the applications barrier to entry. Settlement negotiations to date appear to be focussing on this approach, although Microsoft understandably is reluctant to agree to license core intellectual property rights.
The most dramatic remedy would be structural relief involving a break up of Microsoft. There are three intriguing possibilities:
. a "vertical split" that would separate the company's operating system business from its applicationsbusiness;
. a "horizontal split" that would create two or more smaller companies having access to all Microsoft products and competing on all fronts; and
. a combination of vertical and horizontal segmentation to create multiple operating system and applications competitors.
While many commentators see this form of relief as draconian, Judge Jackson might well conclude that it is the only approach likely to effectively restore competitive markets in these circumstances. Intrusive structural relief is uncommon, but not unprecedented. Though controversial at the time, the break up of AT&T in the 1980s is now generally regarded as having been successful in fostering long-term competition (albeit in a more highly regulated industry).
Will Microsoft Undermine Intellectual Property Rights and Innovation?
Intellectual property advocates have portrayed the Microsoft case as an attack on sacrosanct intellectual property rights, which could undermine incentives to innovate. The intellectual property/competition debate is of growing importance. (In Canada, for example, the Competition Bureau issued controversial draft Intellectual Property Enforcement Guidelines last year. See side bar.) However, the Microsoft case does not advance the debate much because it really focuses on contractual and other marketplace practices designed to block competitors. Although the facts are complex and interesting, the applicable legal framework is well-established US monopolization jurisprudence. The DOJ intervened because it perceived anti-competitive activity and intent extending beyond the enjoyment of statutory intellectual property rights. Indeed, Judge Jackson found that many of Microsoft's actions were not designed to protect or enhance its own innovations (some actually reduced Windows' speed and quality), but to prevent others from advancing competitive innovations.
Many commentators urge competition authorities to adopt a lenient rather than aggressive stance with high-tech industries, partly because it is so difficult to understand and predict marketplace dynamics, and partly because of the potential for further technological change to erode an industry leader's market power. (For instance, by the time the infamous IBM case neared completion in the early 1980s, IBM no longer dominated the computer industry.) Yet, the stakes are high. Misplaced deference can have devastating consequences when network effects are able to tip an entire market rapidly in one direction at the expense of potentially beneficial competing products.
At a minimum, the Microsoft case demonstrates that in high-tech industries, competition law must shift its emphasis from the impact a particular business practice has on price or quality to the effect it will have on innovation. Judge Jackson correctly accepted innovation as an integral component of economic welfare by focussing on the potential innovation-deterring impact of Microsoft's behaviour, even though it also benefited consumers by allowing them to obtain browsers free of charge.
Extraterritorial Impact on Canada
For Canadian lawyers, one of the most interesting aspects of Microsoft is its extraterritorial impact. No parallel proceeding has been brought under Canada's Competition Act, even though our reviewable practice of "abuse of dominant position" could comfortably encompass the conduct covered in both US cases. In an increasingly globalized economy, this phenomenon is not unusual. For example, on cross-border mergers, remedies with effects that spill over into other jurisdictions are now fairly common. The result is a "highest common denominator effect," where the jurisdiction with the most stringent antitrust laws or most rigorous enforcement mechanisms shapes the outcome beyond its own boundaries. In practice, actions taken in the U.S. and the EU typically will be more significant than those of smaller jurisdictions like Canada. Whatever remedy the US courts ultimately impose on Microsoft will likely shape its activities and the evolution of the computer industry worldwide.
A. Neil Campbell practises competition and trade law at McMillan Binch in Toronto. The assistance of Ginevra Saylor and Mark Opashinov in the development of this article is gratefully acknowledged.