U.S. v. Microso, Corp., 253 F.3d 34 (D.C.Cir.2001)

Transcription

1 U.S. v. Microso, Corp., 253 F.3d 34 (D.C.Cir.2001) Hanno Kaiser Silicon Valley An?trust v.2, U.C. Berkeley Boalt Hall School of Law February 2013 This work is licensed under the Creative Commons Attribution 3.0 United States License. To view a copy of this license, visit licenses/by/3.0/us/ or send a letter to Creative Commons, 171 Second Street, Suite 300, San Francisco, California, 94105, USA. Contact me at: hanno [at] wobie.com 1

2 From MicrosoM I to MicrosoM III 1994 U.S. v. Microsoft ("MS1") MS maintained OS monopoly vis-a-vis IBM's OS/2 through exclusive and "per processor" OEM licenses Consent order (1995) 1997 U.S. v. Microsoft ("MS2") Violation of consent order through technologically tying IE to OS to the detriment of Netscape Decision: No violation of consent order (1998) 1998 U.S. v. Microsoft ("MS3") Findings of fact (1999) MS's efforts to suppress Netscape Navigator's (NN) and Java's emerging middleware threat to its OS monopoly Mediation w/ R. Posner (failed) Briefing on legal conclusions, w/ L. Lessig Exclusive dealing w/ OEMs, 1 Tying IE to OS, 1 Monopoly maintenance (OS), 2 Attempted monopolization (browser), 2 Conclusions of law (2000) 1(-) 1(+), per se illegal tying 2(+) 2(+) Decision on appeal (2001) 1(?), ROR applies (remand) Yes, 2(+) No, 2(-) 2

3 U.S. v. Microso, III Table of contents I. INTRODUCTION A. Background B. Overview II. MONOPOLIZATION A. Monopoly Power (+) 1. Market Structure a. Market defini?on b. Market power 2. Direct Proof B. An2compe22ve Conduct (++) 1. Licenses Issued to Original Equipment Manufacturers a. An?compe??ve effect of the license restric?ons b. MicrosoM's jus?fica?ons for the license restric?ons 2. Integra?on of IE and Windows a. An?compe??ve effect of integra?on b. MicrosoM's jus?fica?ons for integra?on 3. Agreements with Internet Access Providers 4. Dealings with Internet Content Providers, Independent SoMware Vendors, and Apple Computer 5. Java a. The incompa?ble JVM b. The First Wave Agreements c. Decep?on of Java developers d. The threat to Intel 6. Course of Conduct C. Causa?on III. ATTEMPTED MONOPOLIZATION A. Relevant Market B. Barriers to Entry IV. TYING (+) A. Separate- Products Inquiry Under the Per Se Test B. Per Se Analysis Inappropriate for this Case. C. On Remand V. TRIAL PROCEEDINGS AND REMEDY VI. JUDICIAL MISCONDUCT VII. CONCLUSION 3

4 US v. Microso, Scorecard Claim Market Who won? The U.S. v. MSFT case is about MSFT s various ac?ons to defend its PC OS monopoly against the emergence of interoperability layers. 1 Monopoliza?on PC OS DOJ 2 Afempted monopoliza?on Browser MSFT MSFT had a stable PC OS monopoly as a result of indirect network effects ( applica?on barrier to entry ) Netscape (NN) and Sun (Java) threatened to erode the entry barrier through middleware MSFT used contracts, threats, decep?on, technological tying, etc. to keep NN and Java from gaining cri?cal mass 3a Tying to leverage PC OS power into browser market Browser MSFT Who won? As a legal mafer DOJ. But in prac?ce, MSFT avoided harsh remedies under the Bush administra?on. 3b Tying to protect PC OS market against entrants PC OS DOJ Why is MSFT so important? Network effects are important barriers to entry Nascent threats are placed under the an?trust equivalent of the endangered species act Plain?ffs should focus on the market in which there already is monopoly power IP is not an absolute defense ( baseball bat analogy ) Per se tying does not apply to plagorm technologies 4

5 A first look at direct and indirect network effects Direct network effects: The more users join a network, the more valuable the network becomes for each user. (E.g., telephone, IM). (Same- side effects) Indirect network effects: The more users there are on one side of a plagorm (e.g., users of an OS), the greater the value of the plagorm to another cons?tuency (e.g., applica?on developers for the OS). And vice versa. (Other- side effects). A mul?- sided plagorms omen has both direct and indirect network effects, e.g., Facebook. Direct effects = deeper pool of friends. Indirect effects = greater adver?ser value. 5

6 Why is middleware such as Java a threat to MSFT s OS monopoly? Application barrier to entry feedback loop application (win) Java (win) "talks to" application (MacOS) "talks to" application (Java) Java (Mac) application (Linux) Java (*nix) All else equal, developers want to develop for all users, irrespec?ve of their OS That s where Java comes in. It provides a common layer of abstrac?on on top of otherwise incompa?ble OSs os (win) "talks to" hardware API driver os (MacOS) hardware API driver os (Linux) hardware API driver OS- agnos?c, cross- plagorm apps break the OS apps posi?ve feedback loop Why buy Windows if your Java app runs on Linux, OSX, FreeBSD, etc.? Today s OS- agnos?c apps are primarily Java (e.g., ERP) and/or browser based (e.g., Google Docs, Facebook) 6

7 Claim 1: Monopoliza?on 1. Licenses to OEMs ( don t bundle NN with Windows ) 2. Technological integra?on of IE and Windows ( scrambling the IE and Windows eggs ) 3. Agreements with Internet Access Provides ( don t include NN on the AOL CDs ) 4. Dealings with Apple ( drop NN or we will kill Office for Mac ) 5. Java ( embrace, extend, ex?nguish ) 6. Course of conduct ( all of the above and more ) 7. Download the full- size version of the MSFT overview here: hfp://hannokaiser.com/ other/2007_msm_exclusion.pdf 7

8 Weak contribu?ng factor causa?on requirement to protect nascent threats Causa?on may be inferred when exclusionary conduct is aimed at producers of nascent compe??ve technologies. In that case, the ques?on is not whether the nascent technology would actually have developed into viable compe??on but whether they reasonably cons?tuted nascent threats. US v. Microso, Corp., 253 F. 3d 34, 79 (D.C. Cir. 2001). Put differently, once the exclusionary effect is established ( power to exclude ), the effect on the consumer ( power over price ) may be inferred. Danger of a freewheeling biodiversity approach to an?trust where poten?al compe??on is involved, i.e., protec?on of technological possibili?es spaces, not necessarily poten?al compe??on. 8

9 Claim 2: Afempted monopoliza?on of the browser market (- ) (2) Attempted monopolization of the browser market, 2(-) Anticompetitive conduct Specific intent to monopolize Dangerous probability of success (-) Relevant market Entry barriers DOJ and court failed to define browser market (82) RealPlayer and Acrobat may be substitutes for a browser (!) (82) No evidence of entry barriers (84) The court makes short work of the afempted monopoliza?on claim, finding no dangerous probability of success. Is there an inconsistency? MicrosoM monopolized the OS market by way of stun?ng the growth of NN and Java As to NN, MSFT pushed IE into the browser market before it could?p in NN s favor How come that MSFT was able to achieve that goal without a dangerous probability of success in monopolizing the browser market? 9