Microsoft: In Search of the Perfect Remedy


Microsoft's blockbuster antitrust trial resumes on February 22 when the attorneys serve up their oral arguments. By late spring, Judge Thomas Penfield Jackson will issue his conclusions of law, which everyone expects will lambaste Microsoft. Next, barring an unlikely settlement, there will be a hearing on remedies, then final resolution, probably before winter. Not surprisingly, competing legal briefs are out in force. Harvard professor Lawrence Lessig wrote one at Jackson's request; former judge Robert Bork got the nod from the state attorneys general; two opposing trade associations weighed in for the Justice Department and Microsoft. Garnering the award for most prestigious authors, the Microsoft group managed to recruit Lloyd Cutler, C. Boyden Gray, Nicholas Katzenbach and Griffin Bell.

Unofficially, the Progress and Freedom Foundation, a generally conservativethink tank, is making the case for Microsoft's archrivals, like IBM, Oracleand Sun Microsystems. PFF submitted its brief to the media, not the court.It wasted little time on legal nuances but devoted fully 11 of its 32 pagesto a recitation -- uncritical and almost verbatim -- of Jackson's factfindings. Then PFF went straight for the kill -- Microsoft's dismemberment.That remedy, according to author Thomas Lenard, "requires the least amountof regulation."

Wrong! The least and best regulation -- sustained by the facts and thelaw -- is no regulation at all. Instead, PFF crows that it has come up withthe optimal solution, one that answers the critics of draconian structuralremedies like divestiture. For support, Lenard craftily cites appellatejudge Richard Posner, recently appointed mediator by Jackson to see if asettlement could be reached. Posner has written that promoting competitionis "the proper purpose of the antitrust laws." PFF links that statement toits own assertion that dismemberment promotes competition. Ergo, Posnermust endorse dismemberment. Inconveniently, PFF's adopted advocate is thesame Richard Posner who has characterized divestiture as a "luddite solution[that] would entail the sacrifice of known and substantial economicbenefits -- those flowing from efficient scale -- for a conjecturalimprovement in performance from competition."

Whether luddite or pro-competitive, the PFF remedy is a "hybrid" ofvertical and horizontal divestiture. Here's how it would work: Like avertical breakup, Microsoft would be split into an applications company (A),selling products like Word and Excel, and a Windows company (W), selling theoperating system bundled with the Internet Explorer browser. Then, like ahorizontal breakup, W would be cloned into three identical companies, eachof which would share "equally" in employees, existing contracts andintellectual property rights. Never mind that nobody at PFF or elsewherehas the foggiest idea how to share employees and contracts equally.

After the breakup, writes Lenard, if A wanted to do business with any of theWs, it would have to do business on the same terms with all of the Ws. Fora reasonable period -- maybe three to five years -- A and the Ws could notmerge or contract exclusively with one another. Otherwise, any of the fourcompanies could develop or acquire products of any type without restriction.That last feature, contends PFF, ensures that the government won't have tointervene regularly to decide whether a new product is an application orpart of the operating system. Because none of the three Ws will havemonopoly power, PFF insists that it won't matter if they bundle otherproducts or engage in practices that would be off-limits to a monopolist.

Thus, to avoid ongoing regulation -- the bane of vertical divestiture -- PFFfinds it necessary to carve Microsoft horizontally as well. But the problemwith that approach, which compels multiple Ws, is that programmers all overthe globe have developed thousands of compatible programs, thanks to thestandardized platform that Windows affords. If PFF were to get its way,that enormous value would disappear overnight. Government-drivenfragmentation of operating system protocols would wipe out Microsoft's mostimportant contribution to software markets: standardization. Like the Unixoperating system, Windows would end up with many variations -- no commonplatform on which software developers can build. The result would be fewerapplications, increased costs of development and higher prices forconsumers.

Anticipating that criticism, Lenard declares, without any support whatever,that "network effects" -- the desire of users to maintain compatibility withother users -- and high switching costs will prevent fragmentation. If theW companies want to retain their installed base of users, says Lenard, theywon't create incompatible new features. To be blunt, that argument isnonsense. PFF did not -- indeed, could not -- point to a single instance inwhich competing entities didn't struggle to gain market share by addingunique features and functions, not available from rivals and not capable ofreplication. That's exactly what happened with Unix. And that's preciselythe dispute now in court over Microsoft's incompatible enhancements to Sun'sJava language. Before long, new features dominate old features and thestandardized system disappears -- at least until a new leader emerges, atwhich time PFF will undoubtedly call for another divestiture to buy moretime.

Still, bemoans PFF, we have to do something. Without a largelibrary ofapplications programs, there will never be a serious rival for Windows; anduntil there's a serious rival for Windows, applications programs won't bewritten. In Lenard's words, it is "not profitable for developers to devoteresources to developing programs for an alternative operating system thatonly has a small share of the market." Astonishingly, barely three pageslater, Lenard announces that "Microsoft was especially concerned abouttechnologies, such as Netscape's browser and [Sun's] Java, that couldsupport platform-independent computing and thereby erode Microsoft's marketposition." Those two statements cannot logically coexist. If no firm coulddent Microsoft's "applications barrier to entry," why in the world wouldMicrosoft be concerned about software developers flocking to theSun-Netscape "middleware" market?

At a minimum, before we adopt PFF's "perfect remedy," let's be sure there'sa problem to be remedied. When technology given life by our most creativeand successful companies is essentially expropriated by bureaucrats whodecide how, by whom and under what conditions it is to be marketed, weshouldn't be surprised if innovation and entrepreneurship -- twin enginesthat propel economic growth -- sputter to a standstill.