Feeds

Special report: Microsoft's pitch on antitrust

No monopoly power, just market power, right?

  • alert
  • submit to reddit

Choosing a cloud hosting partner with confidence

MS on Trial The plain fact is that Microsoft has monopoly power in the operating systems market for Intel-compatible processors, and now in browsing software, but is not willing to own up to it (although it does agree, nudge, nudge) to having market power. It claims it has not abused its power or suppressed competitive threats, but will not go to the next step of admitting monopoly power (which is not illegal in itself, although it is along the road leading to the crime of monopolisation, or illegally exploiting a monopoly). One of the more astonishing claim in Microsoft's Proposed Findings of Fact is that it says that the antitrust laws don't apply because there's no evidence that "Microsoft's actions have actually prevented other companies from creating new software products". It's not even a clever argument. So why wouldn't a venture capitalist support the development of an alternative to Windows? The answer is clear: they lend no money for competing against Microsoft, not because of its software quality, but because of its monopolisation of markets. When a story is largely fictional, it is hard to remember the purported facts. "Monopoly power is .. the power to control market prices or exclude competition, which cannot exist absent high barriers to entry" and "there is no applications barrier to entry [to the operating systems market]". But then we are told: "MS-DOS eventually became the predominant operating system for personal computers, in large part because of a number of popular applications were written for MS-DOS". Oh dear. Then, a fast-changing world is inconsistent with the notion that Microsoft has monopoly power, the monopolist intones. But if Microsoft did not have a monopoly, why should it be that there are effectively no competitors in the OS market? This is a very strong argument for there being entry barriers. Microsoft says that there were more applications for OS/2 than for Windows in 1990, and that when Windows 95 was released there were few 32-bit applications. It is doubtful whether the court will be impressed with the notion that Microsoft is a competitor for itself, which of course proves again that Microsoft is a monopolist as it cannot identify real competitors. Microsoft ruins its argument about entry barriers when it boasts of how much it spends on R&D. In addition, the tying of Windows to MS-DOS was more important than the applications barrier. Microsoft likes to point to the existence of Netscape as proof that there is no significant entry barrier, but fudges the issue that Netscape created a new market that was not the same as the operating systems market. A novel argument as to why Microsoft could not have had a monopoly was that it had to develop "many" new operating systems in this "stable product category" [careful, that sounds like a euphemism for a "market"]. And of course, Microsoft developed the fin de siecle Windows 9x to create a need for users to buy NT. Microsoft's claim that "it is misleading... to attempt to infer the existence of monopoly power from the absolute level of software prices, or from margins earned from licensing successful software products" goes against common sense. Microsoft's profit levels are obscene, because its margins are too great, and it does not spend enough on ensuring products are reasonably bug free before releasing them. Lock-in and switching costs Tipping is a rapid change in a market that results in a product becoming widely adopted. Contrary to Microsoft's view, tipping has occurred in the IE market, as the DoJ expected, and all Microsoft's protests are in vain. Microsoft claims it has "won" several distinct races to provide "the best operating systems software" between 1981 and 1998, and its dominance is not related to monopoly power. Of course, we all know that in these races, competitors were hobbled by Microsoft's marketing machine, and the ability to run in the race was not therefore assessable. Equally, users are mostly locked-in to the Windows market, whether they like it or not, because of high switching cost for a user to go from Windows to, say, OS/2, Unix, or Apple. Economists Having paid so much to bought-in economist Dean Richard Schmalensee for his opinions on computer science and far-fetched economic views du jour, Microsoft was going to make the most of him. Microsoft decided to attack the competence of the DoJ's rival economists (they "lack an adequate foundation" and "conducted essentially no independent empirical studies or other economic analyses" and "[Franklin] Fisher knew little about the software industry generally"), but this is slippery round, in view of Schmalensee's efforts. For Microsoft to claim that Schmalensee "conducted" detailed analyses does not correlate with his evidence: any work seemed to have been done by ill-informed associates or students, and repeatedly demonstrated flawed methodology. Schmalensee used an average price of $2000 for PCs initially, whereas Fisher used $953. Microsoft criticised the fact that Fisher had used data from PC Data, "a source on which no party previously had relied" [meaning it was unfavourable to Microsoft's arguments]. The MDC data set used by Microsoft for its browser claims (a small household study where respondents often did not know what browser they used) was so discredited during the cross-examination of Schmalensee that claims made by Microsoft using the data cannot be given credibility. Microsoft even paid for an opinion by Eugene Erickson that the data were collected "by methods consistent with high standards of survey research", but it is unconvincing. Microsoft tried to pick holes in the AdKnowledge data set used by the DoJ, and while the data do have problems, they did stand up to cross-examination. It is interesting that Microsoft did not use the MDC data internally: for its own studies, but preferred "digital dial calldowns of Web surfers" in the US, and hits to top sites for International data. Incredibly, Microsoft claims that its actions have "promoted increased usage of the Internet" since the number of users has gone up from 3 million to more than 100 million, according to Schmalensee. At least we know who to thank. Market definitions Microsoft squirms in a section entitled "There is no fixed definition of an operating system". The same could be said about the definition of a bicycle or a ham sandwich, and merely showed that Microsoft wanted to use its own definitions for an OS and a browser. Microsoft wants competition to be examined for platforms rather than for operating systems, because this would broaden the market definition. Besides, it has dealt with the rival operating system threat with OEM licensing tricks. So now, to Microsoft, the Web is a platform, so is Java, and so is Netscape's browser. All markets are "purported markets" as Microsoft prefers not to recognise the existence of any market in which it is dominant, since to do so would condemn itself as being a monopolist. Microsoft then wails that in the "narrowly-defined operating systems market for Intel-compatible processors", Microsoft has no monopoly because of Linux. What a laugh. Of course Microsoft always wants to add in Apple to increase the size of the OS market, and claims that Apple is a serious rival: "the MacOS has been a competitor to Windows for more than a decade" Microsoft says, which is absurd, especially since Microsoft is discussing the "alleged market" for Intel-compatible operating systems. Microsoft tries to make the case that the Netscape browser and Java threats to Windows are incompatible with the plaintiffs' market definitions, and put Windows in a different market: this is just wishful thinking by Microsoft. Schmalensee, Microsoft's economist, is credited with claiming that product markets should not be defined because they exclude "companies and technologies that plaintiffs acknowledge pose competitive threats to Microsoft". His confusion will not help: it is not the threat to Microsoft that is of concern, it is Microsoft's threats, and in particular its anti-competitive actions against competitors, and as a consequence, the harm that this does to consumers. Schmalensee wanted the inquiry to be whether Microsoft is constrained in its behaviour by the activities of its competitors. He claimed that Microsoft constrained its behaviour with respect to pricing and product development because of the "intense competition" it faced. His conclusion, that Microsoft is not a monopolist, does not follow from his premises, and so the argument is fatally flawed. OEMs and tying Microsoft criticises the evolving argument by the DoJ with respect to tying, but does not acknowledge that since the Complaint in May 1998, Microsoft has changed how it tied IE to Windows 9x. Microsoft complains that the DoJ was unable to identify a tied product. If, says magician Microsoft, there is a tied IE, where is it then? Really Microsoft's lawyers should look at Microsoft's product history and even read the trial transcript: everybody else in the world knows the trick Microsoft played, including Judge Jackson. Never mind, Microsoft continues, "the tied product [whoops] ... has benefited software developers, computer manufacturers and user". That's a surprising claim. Many people report being seriously inconvenienced by having to suffer IE popping up all the time, even when their browser of preference is not IE. Anyway, the magician continues, isn't it better to have HTML parsing and rendering in Windows rather than in developers' products? But why not in the browser, one might reply? Well, says Microsoft, we've added "new features" [techno-sabotage] so that users must have our version of IE to get the HTML help system working. No prizes for guessing why Microsoft changed to HTML help, and messed up developers who depended on the old system. There is a 12-page special section reserved for an attack on DoJ witness Edward Felten. Of course, we all know that Microsoft sprinkled the code into Windows 98 to make the separation of browsing functions difficult, rather than for reasons of efficiency. This was described in detail in the evidence. Microsoft deliberately misunderstandood what Felten showed, and was caught cheating as well. Not surprisingly Microsoft has nothing to say about its cheating, but we doubt if Judge Jackson will forget the attempt to mislead the court. It was a pity that the DoJ relied so much on their lawyers trying to understand the situation, because their questioning of technical witnesses such as Farber and Felten was frequently not well enough informed. Microsoft also reminded us of the Court of Appeals utterances about penetration: the "interpenetrating" (which Microsoft calls "integrated") design of Windows 98 giving "facially plausible benefits". Come off it. The appellate court got it wrong because it was obsessed with the politics of the case, and certainly did not understand the computer science of the matter. Microsoft has not shown any benefits for the IE -Windows tie that could not have been obtained in ways that were more convenient for users Microsoft quotes its witness Jim Allchin as saying that things work better if they are designed and built as a unit - but of course Windows 3.x and Windows 9x are both unintegrated products, with MS-DOS substitutable by DR-DOS. The user benefits claimed for the "integration" of IE to Windows are said to be better value for users, fewer problems [but not for those who prefer another browser], greater dependability [so why does Windows 9x need to be rebooted so frequently?]; ease of use ["integration" makes no difference as it is transparent to users with non-Windows versions]; and unique capabilities [which many users do not want, such as the requirement that IE be present to use the Help command in Windows 98, and for the filing system]. We cannot imagine Judge Jackson bothering to read ten pages of minutiae about the Windows 98 interface, but perhaps he would be interested to know that right-clicking of a mouse on an icon is at least ten years old, even if it was new to Microsoft. Netscape, Microsoft claims, "wasn't 'foreclosed' either from a single distribution channel or from the marketplace as a whole [so there IS a market, eh?]" and there "were no constraints on Netscape's distribution [except that Microsoft effectively locked Netscape out of the best channel by threatening OEMs and others]". "There was remarkably little evidence about the impact of Microsoft's actions on consumers", the tome opines. For once, that is true, but there is enough on the record concerning how users suffer if there is a lack of choice, whether they know it or not. For consumers, the code remains seriously buggy and unstable; prices are high; there is less innovation; but Microsoft makes an obscene monopoly profit. OEMs and licensing A new and untrue claim is made that "Microsoft's agreements with OEMs are designed to protect the integrity of Microsoft's copyrighted intellectual property by limiting the scope of what OEMs are licensed to do". There then follow two references to the direct testimony of OEM chief Joachim Kempin, but neither reference links the copyright claim to OEM licensing in this way, so Microsoft is trying to mislead the court. Microsoft also says that Windows 9x is covered by "registered copyrights". Microsoft has invented this argument to counter claims that the plaintiff States are seeking "to deprive Microsoft of rights guaranteed to it by the Copyright Act and Constitution", and want to "impose state-created limitations on Microsoft's federally guaranteed rights to license and distribute Windows". In fact, it is the OEM licence agreement that sets out the onerous terms concerning Windows 9x, such as the right to delete icons, and it has nothing to do with copyright. Microsoft says that OEMs may install "whatever software products they like and place icons for those products on the Windows desktop and list them in the 'Start' menu". As we saw in Garry Norris' evidence, Microsoft penalises OEMs who include competitive software, and would prevent an icon being present if its purpose was to enable users to disable certain Microsoft functionality. Microsoft claims that "OEMs are free to install any non-Microsoft software" but of course they are not really free to do so because they have to pay a higher price if they do, in that they cannot qualify for MDA (market development agreement) discounts for not loading certain software. "The existence of older versions of Windows, so-called Windows 'clones' [Sun's WABI and OS/2, apparently, but no mention of X-Windows] and pirated copies of Windows .. limits the price Microsoft can charge for Windows". Microsoft says that its operating systems are licensed at "typically less than 5 per cent of the price of a new computer", but this is untrue. The cost of its operating systems have increased over time as a percentage of systems cost (as Microsoft's own data showed), and are now typically over 10 per cent of systems cost, and increasing. This is a real problem for Microsoft and accounts for its concern about Linux. There are a few differences between the DoJ case and the plaintiff States' case. The States are seeking monetary penalties under their respective antitrust statutes, but it seems unlikely that they will achieve this through the present action, or Microsoft would have been entitled to a jury trial. The States also claimed that Microsoft has indulged in monopoly leveraging, specifically that Microsoft used its OS monopoly to gain control of the browser market, but Microsoft succeeded in having this claim summarily dismissed in September, which was surprising. Dynamics of software market competition Microsoft tries to make a case that the software industry competition is different from that of other industries. Virtually all costs of producing software are "fixed costs", it argues, sweeping aside the fact that marketing and sales costs will be considerable and greatly exceed the fixed costs. The argument that even a small firm can satisfy market demand is not unique to the software industry at all: the same is true for publishing, for example. Another false argument is that the value of software increases as more people use it, because of network effects and increasing returns. In fact, of course, it is the value of the market that increases, or where there is competition, as happened for a time with MS-DOS and DR DOS, the value of MS-DOS decreased. Finally, Microsoft argues, software never wears out so that upgrades must be produced. But there is a good argument that users in many cases do not want more and more trivial additional features in software: they want bug-free software that carries on working, and again this is not unique to the software industry, since, for example, people do not expect reference books, garden tools, or chess sets to wear out to the extent that they must be replaced. Predation and patterns of behaviour Microsoft is trying to make the case that Microsoft's "interactions" so far as Intel, Apple, and RealNetworks are concerned are not predatory, that the allegations were not plead, and that Microsoft's "discussions" were "intended to facilitate technical cooperation". Microsoft moans that predatory pricing was not included in the DoJ Complaint, but has been much discussed. The real point is that the DoJ needed to establish a pattern of behaviour, not just have one or two points where Microsoft had erred. Consequently, Microsoft is unlikely to escape this way. Microsoft helpfully established what it thinks are three criteria that must apply to establish predation, but claims that none apply. The judge may think otherwise, since they are: the objective of removing a competitor; an unprofitable action resulting in the reduction of competition; and a reasonable prospect of recouping losses after the competition is reduced. Microsoft of course now claims that it did not regard IE as a separate product, but selling it and producing stand-alone versions for other platforms rather ruins that claim. Even sillier is Microsoft's claim that its investment in IE, including in IE for non-Windows platforms, has not been unprofitable. To be exact, Microsoft says there is no evidence of this, which should be translated as "guilty as charged". The evidence of Microsoft's predation against Netscape is so great that it must have been embarrassing for the lawyers to pen the words of denial. Procedural matters Microsoft leans heavily on the earlier Court of Appeals decisions relating to the present case, but this is probably optimistic. The appellate court "tentatively" concluded that Windows 95 is an "integrated product" under the terms of the consent decree, but the judges had no background in computer science and clearly made an error that, in an equitable world, should be corrected by the Supreme Court. As expected, Microsoft frequently claims that if the plaintiffs did not question on cross-examination something that Microsoft had said in the thousands of pages of written or oral testimony, then it was conceded. This will not be accepted by the court because the judge wisely decided on an expedited trial, and had each point had to be refuted, the hearings would have gone on for many months more. The separate summary produced by Microsoft PR is not a summary at all, but a whitewashed and misleading account of what Microsoft actually said in the Proposed Findings of Fact. The "summary" contains arguments that are not made in the main document. Microsoft's defence of the antitrust claims does not auger well for its case. The lawyers would have been better advised to be more conciliatory in their tone, but their instructions from Redmond would have precluded any retreat from the aggressiveness that characterises what Judge Stanley Sporkin once described as "a rather young company". ® Next part: The three threats to Microsoft Other sections Microsoft's antitrust defence The three threats to Microsoft MS on the Netscape AOL deal The threat from Open Source and Linux Back to intro Complete Register Trial coverage

Internet Security Threat Report 2014

More from The Register

next story
Bladerunner sequel might actually be good. Harrison Ford is in it
Go ahead, you're all clear, kid... Sorry, wrong film
Euro Parliament VOTES to BREAK UP GOOGLE. Er, OK then
It CANNA do it, captain.They DON'T have the POWER!
Musicians sue UK.gov over 'zero pay' copyright fix
Everyone else in Europe compensates us - why can't you?
I'll be back (and forward): Hollywood's time travel tribulations
Quick, call the Time Cops to sort out this paradox!
Megaupload overlord Kim Dotcom: The US HAS RADICALISED ME!
Now my lawyers have bailed 'cos I'm 'OFFICIALLY' BROKE
Forget Hillary, HP's ex CARLY FIORINA 'wants to be next US Prez'
Former CEO has political ambitions again, according to Washington DC sources
prev story

Whitepapers

10 ways wire data helps conquer IT complexity
IT teams can automatically detect problems across the IT environment, spot data theft, select unique pieces of transaction payloads to send to a data source, and more.
Getting started with customer-focused identity management
Learn why identity is a fundamental requirement to digital growth, and how without it there is no way to identify and engage customers in a meaningful way.
How to determine if cloud backup is right for your servers
Two key factors, technical feasibility and TCO economics, that backup and IT operations managers should consider when assessing cloud backup.
Reg Reader Research: SaaS based Email and Office Productivity Tools
Read this Reg reader report which provides advice and guidance for SMBs towards the use of SaaS based email and Office productivity tools.
Security and trust: The backbone of doing business over the internet
Explores the current state of website security and the contributions Symantec is making to help organizations protect critical data and build trust with customers.