During the Microsoft antitrust case, the government's economic expert was asked what the competitive price of Windows was. All he could muster was that the competitive price is "lower than whatever it is."
Figuring out the competitive price for Windows is not easy. In traditional industries like clothing or automobiles, we expect prices to come quite close to incremental production costs. But such expectations make little sense for Microsoft and a host of other companies that have grown out of what we call the third industrial revolution.
The first industrial revolution lasted from l760 to 1850 and was responsible for widespread innovations ranging from steam engines to iron production. Between 1890 and 1930, the second industrial revolution brought us electricity, telephones and the internal combustion engine. Today's industrial revolution stems from two innovations from the second half of the 20th century. First, a rudimentary international computer network (later to be called the Internet) and second, the development of the world's first microcomputers (better known today as PCs).
The first two revolutions grew through the expansion of physical capital like railroads, factories and electric networks. Today's revolution is driven by an ever-growing number of firms that operate in health care, technology, telecommunications and financial and consumer services. Some of these firms are providing Internet services, while many others are taking advantage of the Internet through electronic commerce. While many of the old economic rules still apply to these new companies, others need to be reconsidered. In particular, we should be wary of policymakers in the competition field who are eager to apply yesterday's rules indiscriminately to today's changing industries.
Traditional economics has always been quite useful for understanding the market forces that shaped industries and governed competition among firms during the first and second revolutions. The principle of marginal cost, the theories of supply and demand and profit maximization and a little bit of game theory sufficed to study competition in industries with just a few firms. These tools helped economists and policymakers analyze markets in which firms like Procter & Gamble, General Motors and U.S. Steel competed. But those firms are significantly different from Yahoo, America Online and Microsoft.
The Theory of Competitive Races
The race to develop the next killer product is the name of the game in many high-technology industries today. In the software industry, "killer apps" are those applications around which an entire new industry or category grows. Many firms invest heavily to develop a product that creates a new category. Visicalc did that years ago when it came out with the first electronic spreadsheet. As a result, Visicalc "owned" that category for several years.
While some firms attempt to create and lead new categories, others spend money trying to leapfrog existing leaders. That's how Lotus 1-2-3 stole the lead from Visicalc back in the mid 1980s. That's the way Excel came and ate Lotus' lunch in the early 1990s. And that's how Nintendo, Sega and Sony have been competing for the last few years. Just recently, it looked like Sega's Dreamcast might put Sega back in the game console lead. Too bad Sony's new Playstation 2 has a 128-bit processor, which runs games three times faster than Intel's Pentium III.
Economists have developed a rich array of "racing" models to describe dynamic technological competition among firms. Firms take large risks by investing to win tremendous payoffs. The larger the potential payoff from winning a race, the more firms are attracted to that race, and the more vicious is the competition to land up on top. Many of these races have only one winner. As a result, winners take all (or most), while losers often lose everything (or almost everything). Indeed, some of the racing models suggest that there can be "too much" competition. Firms altogether invest too much in trying to win a prize that is rarely shared among runners-up.
It is challenging for investors to identify the winners in these races. Unlike a tennis tournament, where all participants are seeded according to their individual likelihood of winning, participants of third industrial revolution races can be in the heat of battle, yet have no idea of their future competition. Many of their competitors won't even weigh in until tomorrow's ringing of the Wall Street IPO bell.
Additionally, competitive races are plagued by tremendous incentive and information problems associated with the financing of innovation. Innovative projects are typically associated early on with a great deal of uncertainty. Information gaps between researchers and investors are commonplace. These projects typically have substantial intangible assets, which are difficult to value and may be impossible to resell if a project fails. Similarly, market conditions in many of these industries are highly variable. The nature and magnitude of these problems generate many opportunities for unsuccessful investment decisions. Not surprisingly, venture capitalists bet on only a small fraction of the proposals that are brought to them and make money from only a small fraction of the proposals they fund.
Racing models predict that in high-technology industries, most firms will lose money and only a few will win. There is considerable empirical evidence of this. One such study tracked more than 12,000 software firms between 1990 and 1998. During this time, less than half experienced growth in sales or employment.1
Network Externalities
High-technology industries are often shaped by "network externalities." Such externalities exist when a particular product or technology gains value as a result of other consumers' acceptance of that technology. Economists' favorite example is the fax machine. Fax machines are virtually useless unless many people have them. The first consumer to purchase one was essentially investing in a conversation piece and betting that in the future, many people would make similar purchases. Now that virtually all businesses and many households own them, fax machines are valuable precisely because of their ubiquity.
Network externalities can help a firm quickly gain dominance in a category. Each consumer who chooses one firm's product makes that product more valuable and increases the likelihood that the next consumer makes the same purchase. As a result of this positive feedback loop, the majority of consumers can end up choosing one firm's product. Of course in reality, the value of network externalities is usually more complicated and less direct. For example, Consumer A's Visa card does not become more valuable when Consumer B obtains one. Yet, the more people who own Visa cards, the more attractive card acceptance becomes to merchants. To take another example, no two consumers benefit directly from using the same computer operating system. However, if many people use the same operating system, that system becomes more attractive to applications developers who want to write to a platform that has many users.
Pricing When It's Cheap to Copy
Many high-technology industries are based almost entirely on intellectual property. This type of capital is usually expensive to create, yet cheap to duplicate. Consider computer software. Creating software is a highly labor-intensive process. Yet once the software has been written, it can be duplicated onto a CD for less than a dollar or posted on the Internet for download at even lower cost. As Netscape's Jim Barksdale noted, "We can distribute globally at the blink of an eye." Besides just software, the Internet can be used as a distribution channel for content too. Once a Web site has been created, the cost of maintaining that site does not grow from incremental increases in visitors. As a result, content distribution also incurs virtually zero marginal costs.
Drugs provide another example of an industry characterized by high fixed costs and low incremental costs. In its race to develop one of the first successful treatments for impotence, Pfizer invested more than a half billion dollars. Yet now that Viagra has been developed, it is quite inexpensive for Pfizer to stamp out thousands of additional pills.
High-technology companies have many more considerations than other companies in setting prices. Most companies determine prices based on their production costs. They mark up those costs, taking into account consumer demand and industry competition. Indeed, in many industries there are even rules of thumb for marking up costs. Not so for high technology. Owners and investors spend millions of high-risk dollars funding these ventures. For any firm to secure such high-risk capital, it must set prices high enough to compensate investors for the risks they take on. Of course, firms cannot set prices too high, particularly if they are concerned about the effect of prices on network externalities. Low prices encourage consumers to try their products and, through the positive feedback effects described above, may create additional demand. On the other hand, many high-technology firms are less concerned about losing sales to other competitors. They may not have any significant ones if they have a patent or copyright that prevents short-run imitation.
The Government's Case Against Microsoft
Some of the complaints against Microsoft2 illustrate the importance of understanding the economics of companies that are warriors in the third industrial revolution. In the antitrust lawsuit against Microsoft, the government (actually the U.S. Department of Justice and 19 states) has said that Microsoft charges too much for its Windows operating system. Most people either get Windows as an upgrade for about $89, or they receive a copy installed on their new computers; the computer manufacturers pay about $65 per copy to install Windows on new computers.
How would one know whether Microsoft's price is "too high?" With a second industrial revolution company, economists would ordinarily look at the incremental cost of production. We expect that price should roughly equal incremental cost (plus a return just high enough to attract capital to that industry). Unfortunately, such a simple calculation does not make any sense for a firm like Microsoft. The incremental cost to Microsoft of each additional copy of Windows installed by manufacturers is literally zero. Microsoft sends the computer manufacturers a master disk, which the manufacturers then copy onto their machines. It is therefore not surprising that the government's economist had no idea what the competitive price for Windows really is.
Microsoft's price actually does not appear to be "high" when compared with other operating systems. Apple charges $89 to upgrade to their new Mac OS 8.5. An upgrade to IBM's OS/2 Warp costs $140. And while Red Hat Linux costs less than $40, a consumer who wants any other Unix-based operating system can expect to pay $450 for Sun's new Solaris 7.0 or up to $795 for SCO's Unixware 7. Just like Microsoft, all these firms face the difficulty of setting a price based on many considerations. But unlike textbook discussions of competition, not one of them makes this decision based on incremental production costs.
Economists don't have a simple theory to explain pricing behavior among firms that sell intellectual capital at prices far above marginal cost. It is clear that these firms cannot be expected to charge a price of zero for their product. No firm would enter any competitive race if the trophy awarded to the winner were really only a booby prize. If we want firms to continue making high-risk investments (a proposition that no doubt benefits consumers and the economy as a whole), we have to realize that these firms will require high returns on their investments—and therefore high margins on sales.
The government has also said that Microsoft charges too little for its Web-browsing software, Internet Explorer (IE). Microsoft decided to include IE as part of Windows, and therefore does not charge separately for it. It has also designed versions of IE for other operating systems like the Mac. It gives those away. According to the government, Microsoft gave away IE to help destroy Netscape, which used to have the most popular Web browser. The government maintains that Microsoft wanted to destroy Netscape for fear that their browser could provide a potential rival to Windows. There are, however, innocuous explanations for giving away browsers for free. Interestingly enough, Netscape chose to give away most of its browsers for free, too.
Two considerations make this zero price economically sensible. The first is that Microsoft and Netscape could expect to earn revenues from other sources—complementary revenues—if people used their browsers. Microsoft could expect that more software developers would write software for Windows, and this would increase the value and sales of Windows. Additionally, both Microsoft and Netscape could expect to earn revenues from advertising and other sources if they were able to direct users to their Web portal sites (sites that serve as gateways to the Internet). All else being equal, the more people who use their browsers, the more likely people are to visit their portal sites. When America Online finally purchased Netscape, it spent more than $10 billion of its own stock to do so. Obviously, the company saw sufficient potential for complementary revenues in Netscape, despite Netscape's inability to earn revenues from sales of its browser.
The second reason why a zero price for browsers is economically sensible is the existence of network externalities. The more people who use a browser, the more valuable it becomes. Again, the externality is indirect: more users encourage more applications which encourages more users and so forth. A low price gets more users, more network externalities and more complementary sales.
The government isn't just concerned that Microsoft charges too little for IE. They don't like the fact that it is always sold as a part of Windows and consider it to be a case of illegal bundling, or "tying," of two separate products. While economists think that tying is seldom a bad deal for consumers, the antitrust courts have made it tough for companies with large market shares to engage in this practice. One of the most famous cases involved a hospital just outside of New Orleans that required patients to use its staff anesthesiologists; the Supreme Court thought the hospital had tied the two products together but let the hospital off the hook because it did not have a large market share.
Unlike the tying of hospital services, the bundling of software is tricky to define. Without passing judgment on the merits of the practice, it is easy to see that surgery and anesthesiologists are two quite distinct products. Software code is far more complicated. Just like the words in a book, software code can be organized in various ways to achieve a multitude of effects. Over the course of time, software producers add features to their programs that previously were sold as separate products. Spell checkers, for example, are now a standard component of most word processing packages. People also used to buy separate "utilities" programs to enhance their operating systems. Many of these are now integral parts of most operating systems. Integrating browsing capabilities into an operating system seems the next logical step—operating systems read files from many sources and the Internet is just the latest.
Microsoft and the government have taken quite opposite views of the right way to think about competition in the software industry. The government sees Microsoft as making more than 85 percent of the sales of operating systems for personal computers. By traditional standards that is a high share. And it sees Microsoft as facing few competitors. Computer manufacturers have few alternatives available today for personal computers for home or business use. As a result, Microsoft can price high and wield power.
That's a sensible analysis for second industrial revolution companies. But using traditional economic models to analyze industries characterized by high risk network externalities and zero marginal costs can be a mistake. In the case of computer software, consider Microsoft's position.
Microsoft sees itself in a perpetual competitive race—always looking over its shoulder to see whether known rivals like Apple or come-out-of-nowhere rivals like Linux are catching up. Because software development requires only limited resources, Microsoft is constantly threatened by behemoths like IBM as well as tiny startups like Red Hat that have bright new innovative ideas. Microsoft, therefore, can only view its market share as fragile and transient. It believes that it prices low, because if it doesn't, it will be replaced.
Microsoft and the government have two opposing views of competition in today's high-tech industries. The prevailing view will prove important not only for Microsoft, but for many other companies as well.
--------------------------------------------------------------------------------
1 Lerner, Josh, "The Returns to Investments in Innovative Activities: An Overview and Analysis of the Software Industry," Harvard Business School and National Bureau of Economic Research, September 1998. Microsoft funded the research for this paper.
2 We are working for Microsoft on this litigation. See http://www.nera.com for other papers related to this case.
Viewpoint, Number 1, 1999
Copyright © 1999 by Marsh & McLennan Companies, Inc. All rights reserved.
--------------------------------------------------------------------------------
Back to Top | MMC Views home page
Past Editions Menu
Table of Contents
April 25, 2000
--------------------------------------------------------------------------------
Editorial Page
Why the Case for a Breakup Breaks Down
By Richard B. Mckenzie and William F. Shughart II. Mr. McKenzie, a professor at the University of California, Irvine, is author of "Trust on Trial: How the Microsoft Case Is Reframing the Rules of Competition" (Perseus, 2000). Mr. Shughart, a professor at the University of Mississippi, is author of "Antitrust Policy and Interest-Group Politics" (Quorum, 1990).
The Justice Department is awfully busy this week. By Friday its Antitrust Division is due to submit a proposed remedy in the case of U.S. v. Microsoft. The department reportedly will ask Judge Thomas Penfield Jackson to order Microsoft to break itself up into two or more separate companies, so-called Baby Bills. The advocates of a breakup contend that dissolution is preferable to a "conduct decree" that would leave Microsoft intact but impose restrictions on its behavior.
Advocates of a breakup argue that it would make software markets more competitive and would avoid the continuing judicial and regulatory oversight a conduct decree would entail. They point out that after Standard Oil was dissolved in 1911, the collective market value of the 30 successor companies quickly rose above the oil trust's pre-breakup market value, making John D. Rockefeller Sr. into history's first billionaire.
The implication is that the same antitrust medicine for Microsoft will be beneficial all around -- to consumers, who would get more competitively priced software, and to Microsoft's owners, who would become wealthier than they already are. How could anyone object to such a win-win solution?
***
The problem is that the underlying argument is wrong on both the facts and the logic. No wonder that news of a possible breakup helped send the Nasdaq Stock Market into a tailspin Monday.
The Standard Oil comparison is based on a half-truth. Yes, the total market value of the 30 "Baby Oils" tripled between 1911 and 1913. But the stock market was rising rapidly across the board. A new study by VMI economist Atin Basu Choudhary and University of Mississippi economists Robert Tollison and one of the authors (Shughart) finds that, contrary to conventional wisdom and independent of other market forces, the dissolution of Standard Oil had no persistent impact on the wealth of the company's owners. More to the point, Rockefeller's stock-market gains were not proportionately greater than those of ordinary investors in other oil and industrial companies. Hence, the rise in the market value of Standard Oil after its dissolution can't be used to justify a breakup of Microsoft. Indeed, these economists argue that the Standard Oil case is a classic example of antitrust policy failure.
The logic underlying the contentions of Microsoft's critics is also faulty. If Microsoft were truly acting like a predatory monopoly -- restricting sales with the intention of raising product prices and garnering profits in excess of competitive levels -- then a breakup would be expected to lead to more intense price competition and, therefore, lower profits and a lower stock-market value for the Baby Bills. By arguing that a breakup would increase the Baby Bills' total stock-market value, the critics must believe that in some way the breakup will make software markets less competitive, with the prospects of higher prices and profits.
What about the argument that a breakup would press the Baby Bills to become more efficient, enhancing their values? That is sheer speculation, not well grounded in theory. After all, even monopolies have strong incentives to hold their costs in check.
As for the claim that a breakup would reduce software prices, the reverse is probably true. According to economist Stan Liebowitz, a breakup of Microsoft into two operating-system companies could increase software-development costs by $30 billion in the first three years, given that programmers would have to adapt their applications to more than one version of Windows. Application companies surely would pass these costs on to their customers.
A true monopoly doesn't have to compete, and Microsoft is nothing if not a fierce competitor, using its market prowess to win competitive races with rivals, such as Sun, Netscape (now part of America Online) and IBM. Microsoft's interest in doing that is a plus for consumers. By not acting like a monopolist, it holds its prices (and profits) below what they would otherwise be. While such a pricing strategy might make life difficult for Microsoft's rivals, consumers benefit by getting higher quality software products and lower prices than they would otherwise.
***
Under these circumstances, the proposed breakup of Microsoft amounts to a plan to force consumers to pay higher prices than they are now. This is not exactly the kind of outcome the Justice Department and Judge Jackson should want or seek -- if their real goal is to benefit consumers.
What is telling -- and disturbing -- about the Microsoft case is that it is the company's rivals, not consumers, who are pressing for a breakup. Microsoft's rivals should want Microsoft to act like a monopolist -- that is, restrict its output for the purpose of raising its prices. They would then be able to sell more of their own products at higher prices. The rivals would not rationally support any remedy that will make their markets more competitive, given that more-competitive markets would mean lower prices and profits for them. The rivals' support of the proposed remedy suggests that the effect of a breakup will be to do what antitrust enforcement should never do, help competitors at the expense of consumers.
--------------------------------------------------------------------------------
Return to top of page | Format for printing
Copyright © 2000 Dow Jones & Company, Inc. All Rights Reserved.
Copyright and reprint information.
REASON * March 1999
The New Trustbusters
What's behind the resurgence of antitrust activism--and why it's bad news for consumers.
By James V. DeLong
Joel Klein is a famous man. The head of the Antitrust Division at the U.S. Department of Justice usually toils in anonymity, known only to the in-groups of the bar. Not Klein. He has sued Microsoft, the most prominent company in America's jazziest industry, and demonized the world's richest human, Bill Gates. He has assaulted the ubiquitous credit card franchises of Visa and MasterCard and blocked important aerospace industry consolidations. He makes speeches extolling the pivotal role of the Antitrust Division in the "new economy" of globalization and information. These acts have earned him serious attention in the national press.
Not to be left behind, Klein's fellow antitrusters at the Federal Trade Commission are equally active. They have beaten up on superstores by stopping the merger of Staples and Office Depot and by knocking down important marketing practices of Toys "R" Us. Like the DOJ Antitrust Division, the FTC has gone high tech. It is challenging Intel, the grandee of computer chips, and probing Cisco Systems, the dominant company in Internet switching hardware. In a recent suit against a drug company, the FTC asserted a heretofore unknown authority to force an alleged monopolist to refund to consumers $120 million in allegedly ill-gotten profits.
Professional discussion of this surge in antitrust activism is proceeding at two levels. One is the highfalutin language of bar association meetings and academic conferences, where sessions are given titles like "Antitrust Enforcement and High Technology Markets," "Networks, Lock-In Effects, and the New Economy," and "New Approaches to Reviewing Horizontal Agreements." In this world, Platonic guardians mull over the implications of technological change and devise optimal policies to safeguard American enterprise. They "open up a dialogue with the bar and the academic community about antitrust doctrine," as Klein put it in one of his speeches, on topics such as "the relationship between antitrust and intellectual property, the significance of...network effects and tipping points...
the impact of differentiated products theory, the role of potential competition and innovation markets, and [the meaning of] `agreement.'"
The second level of analysis is more skeptical, grounded in the public choice school of economics. Public choice theory holds that exercises of government power are driven by the material and ideological interests of the people who wield it and by the private parties who can reward them with campaign contributions, money, job security, or whatever else they value. Viewed through the lens of public choice theory, the recent burst of antitrust activity is not primarily about consumers or competition. It is about four kinds of interests: 1) competitors of successful firms who want to hamstring their rivals, appropriate part of their businesses, or turn the clock back so they can re-run the race; 2) companies blown by the winds of technological change and looking for ways to nullify their disadvantage; 3) the personal ambitions of antitrust enforcers, who do not prosper in quiet times; and 4) the class interests of the legal profession, which are served by a combination of activist bias and mushy theories.
Only a taste of this view creeps into the formal talks at antitrust conferences. To get the full flavor, you must hang out in the corridors and hotel bars, and stick around for the after-conference receptions. There you hear jokes like this one: "What is the government's theory in the Microsoft case? That the state of California has more computer companies than the state of Washington, and a hell of a lot more electoral votes." Or this one, about "Gore-Techs," the high-tech entrepreneurs who confer with Vice President Al Gore: "What is Gore-Techs? A new fabric made by combining silicon and money, used for wrapping up politicians."
In these less formal settings, last summer's Senate Judiciary Committee hearings on Microsoft, in which Chairman Orrin Hatch (R-Utah) laid into Bill Gates, are dismissed with, "What else would you expect from a senator who has Novell in his state?" Testimony in the ongoing Microsoft trial about the frequency with which the head of Netscape, the moving force behind the lawsuit, met with Klein and other Antitrust Division representatives is greeted with blasé yawns. So are testimony and e-mail messages showing that high-tech companies regard government antitrust action as simply one more tool in their competitive arsenal, an alternative to price cuts or new products. And as the head of Netscape said when asked why he did not file his own suit, it is much cheaper to use the government's lawyers.
Behind the scenes at the conferences, you hear snickers about the innocence of Microsoft, which thought it could sit out there in Redmond writing software and ignoring Washington, D.C.--as if such a big pot of wealth could go unnoticed by a rapacious imperial capital. The idea that perhaps a company should be able to do its business and ignore Washington is regarded as hopelessly naive. Microsoft is now playing catch-up, adding former congressional aides to its staff and boosting its budget for political contributions.
The suit against Visa and MasterCard, announced in October 1998, also evokes grins. (See "Credit Where It's Due," January.) The Antitrust Division's press release speaks of harm to consumers and of an industry that is "competitively impaired." But the arrangement that the government just attacked has been around a long time and was created in part due to antitrust concerns expressed by the DOJ almost 20 years ago. Prior administrations investigated and declined to take action. The corridor talk notes the close relationship between the Clinton administration and American Express, which has been complaining about its rivals for years. Presidential friend Vernon Jordan is an AmEx director. According to news accounts, AmEx has other ties to the White House: It bought heavy advertising from the 1993 inaugural committee, it won the White House Travel Office account, and it has replaced Diner's Club as the government-issued credit card for federal workers.
In the 1997 Office Depot/Staples merger case, the FTC defined the relevant market as "office supplies sold by superstores," even though the two chains together sell only about 5 percent of all office supplies. Wal-Mart alone sells more than this, and bulk mail-order firms sell another huge chunk. Everyone, except the FTC and the federal district judge who upheld its decision, regards the commission's market definition as a laugher. (See "Pricing Pencils," August/September 1997.) But no one can figure out who had the political clout to persuade the FTC to adopt such a silly position. The Wall Street arbitragers who lost a bundle when the deal cratered think the FTC's action was triggered by complaints from Office Max, the third major superstore chain, which stood to get some assets on the cheap if the government forced the merged entity to spin them off in the name of protecting competition.
Last March, when the DOJ Antitrust Division nixed Lockheed Martin's effort to acquire Northrop Grumman, the formal reasons involved concentration in the defense industry. The corridor talk this time actually made it into The Wall Street Journal. Lockheed had been blindsided by rival Raytheon, which gave the government mountains of negative information. In its June 1998 story, the Journal noted that Lockheed, not expecting trouble, had failed to adopt the usual techniques of the merger game: "lobbying Capitol Hill, working the executive branch and creating a drumbeat of support in the media."
The interests of selected businesses and political figures are not the only ones served by the trend toward antitrust activism. The corridor lounger also notes a certain incestuousness among the conference participants. Almost all the private lawyers and economists on the panels used to work for the government, while the government reps used to work in the private sector, and most will return there at some point. Even the career employees turn free agent if the stars are right. The FTC litigator who led the Staples case left a few months later for a well-paid partnership in a major law firm. In the condottiere world of lawyering, his victory in a case that looked like a dog greatly increased his market value. The arbs who lost big as a result want him on their side next time.
It is a new springtime for antitrust lawyers and economists, promising a return to the good old days of three decades ago. Their memories of that era are bathed in a golden glow. It was a time when antitrust regulators were hyperactive, inhibited only by the need to invent a plausible scenario under which a business arrangement might be regarded as "anti-competitive," and this concept could be given any of several not necessarily consistent meanings. Given such a nonstandard, the government won almost every time it challenged a business arrangement. But the charade required a lot of lawyers and economists.
This era ended in the 1970s. Two disastrous pieces of monster litigation--one against IBM by the DOJ Antitrust Division, the other against the oil companies by the FTC--ground on for years. Each went long past the point where it became obvious that the government lacked a coherent theory of the supposed offenses and long past the point where changes in the industry and in the world turned the case into a joke.
During the same period, the University of Chicago school of antitrust analysis began to convince judges and law students, and eventually even the bar, that many of the government's anti-competitive scenarios were not plausible at all and that arrangements which were valuable to consumers were being outlawed. Antitrust law was suppressing rather than nurturing competition.
This revolution in antitrust thinking was consolidated in the 1980s, when the Reagan administration pared enforcement back to the hard core of attacking agreements to fix prices or divide markets, which almost all antitrust analysts regard as beyond the pale, and preventing mergers among direct competitors in highly concentrated industries. The number of cases dropped off as enforcers ceased to invent new theories, and businesses, with a fairly clear idea of the line between legal and the illegal, could adjust their conduct with little reference to the high priests of antitrust.
Antitrust practice came to mean filing the notification forms the government requires before a corporate merger can be consummated and negotiating the terms on which the government would approve. This often meant selling off a division or two in the name of preserving "competition" in some minor market or product line, but the issues were usually marginal, rarely involving fundamental questions of antitrust purpose and policy. Antitrust became a particularly boring kind of regulatory law, consisting mostly of copying thousands of pages of documents to ship over to the DOJ and the FTC for pre-merger review.
Now, after almost two decades of marginal relevance, the thrill of activism is back. Corporate officers and Wall Street arbs once again furrow their brows when they speak of "The Division" or "The Commission." The telephone rings. Hours are billed. People who say "let's do lunch" actually follow up. Headhunters pass out business cards. Members of the opposite sex no longer flee when you say, "I do antitrust."
It is important to avoid what the English writer C.P. Snow called "the cynicism of the unworldly." One should not underestimate the impact of sincere belief on the actions of government officials. The Antitrust Division, the FTC, and the private bar are full of honest lawyers who would be outraged at the suggestion that they are acting for personal or class interest, or that they are stretching the law in the interests of ideology. But it is also important to recognize that self-interest is the aphrodisiac of belief and that adopting the theories underlying the new activism is much to the benefit of the antitrust professionals on both sides. To those with a bent for public choice explanations, the enforcers, flogged on by the politicians and the competitors, have seized upon the computer, the information revolution, and the Internet as excuses to reverse the cautious "first, do no harm" antitrust policy of the past 20 years and restore the promiscuous activism that characterized the 1960s.
The question, of course, is whether these dour apostles of the public choice view are right. Answering that question requires a return to the basics. The antitrust laws stem from our collective fear of monopoly, a bred-in-the-bone knowledge that a supplier of a good or service who lacks competitors will jack up the price, cut the quality, become arrogant and unresponsive, and in general behave obnoxiously. Economists can give you good explanations for this, complete with charts and graphs of supply and demand curves made out of solid and dotted lines and annotated with lots of alphas and deltas. They can explain with precision why raising the price increases the monopolist's revenue even if it cuts sales, and they can tell you the size of the price hike that will maximize the loot.
The noneconomists among us fear monopolies even without the charts and graphs, based on common sense. After all, raising the price is logical; it is what we would do if we had a monopoly of our own. We also look to our experience. Think of the U.S. Postal Service, cable television companies, state liquor stores, or the local public school. Monopolies all, and legendary for unresponsiveness, ineptitude, and overcharging. But these things are mostly peripheral to our lives, or there are safety valves in the form of substitute goods or direct action, so the lack of competition is an annoyance rather than a disaster. Imagine the impact of a complete monopoly in a truly essential product, such as automobiles or groceries.
The fear of monopoly that triggered the first of the big antitrust statutes--the Sherman Act of 1890, which outlawed combinations in restraint of trade and attempts to monopolize--was far from unfounded. Alfred D. Chandler Jr., the eminent business historian, documented the situation in the late 19th century in his classic book The Visible Hand. The rise of mass production, combined with the railroad, the steamship, the telegraph, and the telephone, created the possibility of nationwide marketing and management. As output exploded, prices dropped precipitously. Many industries responded by attempting to put together cartels intended to reduce output and stabilize prices. The means varied--contracts, holding companies, corporate voting trusts, mergers--but the goal, as explicitly stated by the participants, was to attain the joys of monopoly power over a market.
Considering these expressed intentions, public fear was rational, but history has shown that it was largely unnecessary. It turned out that amalgamation is not alchemy, and the trusts could not achieve their goal of excluding competitors and sustaining prices at monopoly levels. The experience of the National Cordage Association, recounted in Chandler's book, illustrates the vulnerability of monopoly schemes. The NCA could not achieve economies of scale in operation or management. Nonetheless, it was forced to pay premium prices to lure competing companies into selling out, which stuck it with huge capital costs. The bought-out competitors then used their money to start new cordage companies that were more efficient than the NCA, so they could undercut its prices. The association went bankrupt in 1893.
Even mighty Standard Oil, the great bête noire for trustbusters of the late 19th and early 20th centuries, was a symbol more than a true villain. By 1911, when the Standard "monopoly" was broken up by the Supreme Court, eight other large integrated oil companies were competing with it. Before that, Standard never tried to sustain prices at high levels. The history of oil during the late 19th century was one of huge expansion in markets and facilities and steadily falling prices.
Many of the complaints about Standard came from medium-sized refiners that lost cozy local monopolies to Standard's rationalizing and price cutting. The same pattern has been repeated many times. As changes in communications and transportation create possibilities for new forms of business organization that might make things cheaper and better for consumers, firms that are doing well under the old forms fight back. This was true in the late 19th century, and it remains true today.
Often, the beneficiaries of the status quo cry "Monopoly!" Almost always, they are wrong. If you look back at the examples of monopolies given a few paragraphs ago, you will notice that all are public institutions or publicly regulated businesses. It is impossible to find examples of truly private monopolies, except on a minor scale. In areas where the natural technical and economic structure of an industry would tend toward monopoly, such as utilities or railroads, public regulation was imposed early (with dubious results--but that is a different tale). In other sectors, monopolies are rarely developed and never sustained.
Some thinkers, such as Dominick T. Armentano, a professor emeritus of economics at the University of Hartford and a scholar of antitrust history, have responded to this experience by advocating the elimination of the antitrust laws. They argue that the benefits from the few instances in which these laws might do good are outweighed by the harm caused by their inevitable misapplication at the behest of the political system.
Others are less willing to scrap the antitrust laws. They note that it's always possible that the members of an industry might succeed in getting together and setting a monopoly price. Perhaps in 1911 Standard and the other eight big oil companies could have set up regular meetings and fixed prices or merged into one company. Over the long term, such a monopoly would not last, any more than OPEC lasted. As the price remained high, new entrants would want to get a cut and would have to be brought into the cartel. Eventually, shares would get so diluted that some members would find it worthwhile to break off and cut the price, thereby ruining the system.
But a cartel can cause pain while it lasts, as did OPEC, and the wait for its inherent contradictions to assert themselves can seem awfully long for us short-lived humans. So most people who study the problem, whatever their general political orientation, agree that a core of antitrust enforcement is important. In particular, they agree that price fixing among competitors should be outlawed, and so should merger to monopoly and agreements to divide markets and set up a series of monopolies in different sectors of the economy. This was the basic antitrust policy of the Reagan years.
As the abolitionists love to point out, though, when you take this first step you step on an intellectual banana peel that sends you skidding down a slippery slope. The fundamental problem is that the meaning of "monopoly" is amorphous, and the ambiguity leaves unclear exactly what the law is trying to prevent, encouraging ad hoc approaches and governmental mischief.
A true, clear monopoly would be something for which there are no substitutes. If you are drowning in the middle of a lake, and someone in a boat comes along and asks if you want to buy a life jacket, he can fairly be said to have a monopoly. Short of a situation like that, there are only varying degrees of power to charge a premium price. A beef company's power is limited by the availability of chicken and fish, and a local grocery store's power is limited by your ability to shop in a neighboring town. In each case, the power is real but limited. So at what point does this power justify government action? There is no clear answer, and no satisfactory theoretical basis for developing one.
This fuzziness gives those interested in expanding antitrust enforcement, including the enforcers, a lot of leverage. They argue that any power over price, however small, in any identifiable product market is illegal. Then they seek to define the market narrowly, making it easy to find a price effect. In a recent paper for the Cato Institute, University of Mississippi economist William F. Shughart II, one of the most relentless public choice critics of antitrust policy, lists some markets defined by the government: "high-priced, non-ethnic frozen entrees"; "noncarbonated, ready to serve, naturally or artificially flavored fruit drinks, fruit punches, or fruit ades which contain 50 percent or less fruit juice and are customarily sold under refrigeration to the consumer"; "direct contract front-loaded trash removal in Dallas." These market definitions do not have quite the same impact as "oil" or "steel."
The next step in expanding the realm of antitrust is to use static economic models to predict an effect on price, often ignoring strategic responses available to competitors that would erode any monopoly power. In analyzing the Staples/Office Depot merger, for example, the FTC noted that slightly higher prices prevail in some areas where only one of the companies operates. It assumed this pattern was immutable, ignoring the ability of superstores, mail-order houses, and customers to observe it and react.
By the time these steps are taken, the result is ordained. Anyone playing by these rules who cannot find a monopoly on every corner isn't trying.
This game discourages efficiency-enhancing mergers. It also discourages arrangements among firms, such as joint ventures and strategic partnerships, that could create innovative products and reduce costs. Such deals always involve some agreements on price and territory because the partners must decide how to split the proceeds. Each also needs to ensure that its partner does not use it for a time, then muscle it out of the way and take over the whole business.
Banning these contracts outlaws many worthy projects. As Fred Smith, head of the Competitive Enterprise Institute, explains: "Any company wants to do many different things. But it can be only one size, which means it cannot be exactly the right size and have exactly the right mix of skills and resources for each of the things it wants to do. For some it is too big, for others too small." If different and possibly competing businesses can integrate some of their operations without merging entirely, they can form entities that are the right size for the venture at hand. The size that is appropriate for inventing a software product, for example, may be different from the size needed to make it commercially viable. Manufacturing and marketing require still different sizes, talents, and structures. The industry is filled with people patching together the temporary alliances needed to fill the gaps in their own organizations. The result, in software and elsewhere, is a flexible and adaptive economy, which serves consumers well.
For a century, antitrust enforcement has been hostile to such partial integrations among businesses. Ironically, this hostility creates pressure for firms to merge completely. Due to oddities of antitrust law, merger may be allowable under circumstances where partial integrations by contract would be blocked.
In the 1960s, scholars with a public choice perspective began analyzing whether past antitrust actions had actually benefited consumers. It is not an easy question to research, but a variety of ingenious studies have looked at stock prices, scholarly opinions, competitors' reactions, and other indicators. The results are strikingly consistent: Antitrust actions do not help the public, though they may help the special interests that trigger them (while providing well-paid employment for antitrust professionals). CEI's 1997 Antitrust Reader and the 1995 book The Causes and Consequences of Antitrust: A Public Choice Perspective (University of Chicago Press), edited by Shughart and Emory University economist Fred McChesney, are good introductions to this literature.
This research, combined with the dearth of empirical work on the other side, was an important factor in the reforms of the 1980s. Proponents of interventionist antitrust had an embarrassing lack of good examples. As McChesney and Shughart point out, when you have a century of experience with a program and virtually every landmark case looks to have been a mistake, perhaps it is time to stop saying, "well, we'll get it right next time," and start rethinking the basic premises.
The current activists are having none of this. Quite the opposite. The seminar sessions at antitrust conferences--the serious part, where the participants do not joke about Gore-Techs--operate as if none of this rethinking happened. The governing assumption is that basic antitrust policy and doctrine was and is sound. The question is whether this tool that has served us so well can be applied to the growing high-tech economy as it stands, or whether it needs to be adapted and extended to reflect new realities. The possibility that the antitrust emperor might be stark naked is ignored.
This should not really be surprising. More than 20 years ago, when I was a middle manager in the FTC, an exasperating commission action prompted me to say to another staffer: "They tell the story of `The Emperor's New Clothes' a little differently around here. It goes along as usual until the little boy pipes up, `That man is naked!' Then, in the government version, the emperor turns and says, `Kill that kid.' And they do." Public choice analysts know that an agency-emperor will seize any opportunity to off the uppity kid and re-impose the myth of the beautiful robes. It will be assisted enthusiastically by private interests who want to turn the myth to their own advantage.
As a result, there is now a gusher of amorphous theories justifying renewed antitrust activism. For example, theories of "path dependence" and "lock-in" hold that society can become committed to an inferior technology, unable to break free when a superior one comes along. It is an interesting idea with little empirical support, as has been demonstrated by economists Stan Liebowitz and Stephen Margolis in the pages of REASON ("Typing Errors," June 1996) and at greater length in the academic literature. The Dvorak typewriter keyboard was not really significantly better than the old QWERTY version, and now that anyone can go Dvorak with a computer keystroke, almost no one does. The other standard example, the triumph of VHS over Beta in videotape, is also wrong. VHS, with its longer recording time, was regarded by consumers as a better product.
It's true that once companies or consumers buy into a technology they will incur costs if they want to change later. But this is hardly a new development; the same is true for purchases of equipment, constructing buildings, or any other investment. Change occurs only when a new technology is sufficiently superior to justify the switching costs, or when investment is turning over anyway. So what? The fact that a company that introduces a good product gets some first-mover advantages is one of the mainsprings of innovation. Is the government now saying this should be foreclosed?
Another phrase popular among antitrust activists is "network effects." Some things become more valuable as more people sign up, and the company that gains an initial advantage may then sweep the field. The classic example is the telephone, where everyone wants to be on the same network.
Again, an interesting idea, and one with clear merit. But again, not new, and not unique to high-tech industries. Timothy Muris, now a professor at George Mason University law school and, as head of the FTC's Bureau of Competition, an architect of the antitrust reforms of the 1980s, notes: "The fact that network effects are everywhere should give us pause about the utility of the concept [for antitrust]. Many products, not just high-tech ones, have the characteristic [that] the benefits of use increase as the number of users grows. Thus, consumers of products that require post-sale service, such as automobiles and appliances, produce network effects from the growth of service outlets when more consumers purchase the product. Coke and Pepsi drinkers benefit from the network of their fellow consumers in that their drinks are widely available in restaurants and from vending machines. Sports fans benefit when they live in an area with enough other fans that teams find it profitable to locate there."
During the formal presentations at antitrust conferences, the audience listens raptly as phrases such as "path dependence" and "network effects" roll off speakers' tongues. During the receptions the respect drops off, especially because the speakers seem more interested in explaining why the concepts justify an increase in their power than in any real analysis. A typical corridor reaction to the new concepts is "buzzwords and bullshit."
More genteelly, participants note that most of the ongoing antitrust actions lack basic clarity and coherence. The Microsoft litigation started with the idea that the company was illegally tying sales of a Web browser to its operating system. Then, during the trial, the theory shifted to a charge that Microsoft wanted to divide the browser market with Netscape--a pretty silly idea, given the capacity of other companies to make browsers of their own. Now the government seems to be trying to prove that Microsoft is a meanie, which is not an offense under the Sherman Act. It is for the most part simply hard-edged jockeying over the terms of the partial integrations that are crucial to an efficient software industry.
The government also seems to think that high market share alone meets the legal definition of monopoly, even if there is only tenuous evidence of power to raise prices. Microsoft argues, of course, that it has a high share only because it keeps prices low and that this is what really bothers its competitors. They want Microsoft to be forced to set prices high enough to leave room for them.
The incoherence extends beyond the Microsoft case. The FTC accuses Intel of withholding information about forthcoming chips from companies that sued it for patent infringements. This is a novel extension of antitrust law. The relief sought would force Intel to treat all similarly situated customers alike, which would put the FTC in the middle of endless disputes over the meaning of "similarly situated" and "alike." The case looks like intervention in contract disputes combined with casual meddling in difficult intellectual property issues, plus an unarticulated assumption (also apparent in the Microsoft case) that any company with a large market share should be converted into a new breed of public utility by government fiat.
The DOJ's Visa/MasterCard case is also puzzling. Two separate violations are charged. First is the practice called "duality," whereby the two brands of card are issued by the same banks. The government contends that duality causes both Visa and MasterCard to go easy on introducing new products, such as smart cards, that might damage the other. The second alleged violation is that the Visa/MasterCard banks do not issue AmEx, Discover, or any other card.
These theories are confusing. If Visa and MasterCard are agreeing to slow down innovation, then AmEx, Discover, and other potential competitors should be thanking them for leaving open such a valuable market opportunity. Government intervention would be superfluous, and the last thing a competitor would want is for the government to wake Visa and MasterCard up.
As for the other theory, if duality is illegal, then why is the government trying to make it into quadrality? Or is the government saying that the banks must become the partners of anyone who wants to go into the credit card business? Again, this is a theory that converts any dominant company into a public utility.
All of these matters, and more besides, fan suspicions that government policy is blown by the winds of special pleading and political interest, and that the complex intellectual rationales are simply a new cover for old-fashioned rent seeking.
In a 1961 paper later published in Ayn Rand's Capitalism: The Unknown Ideal, Alan Greenspan wrote: "The entire structure of the antitrust statutes in this country is a jumble of economic irrationality and ignorance. It is the product: (a) of a gross misinterpretation of history; and (b) of rather naive, and certainly unrealistic, economic theories." His analysis is still on target. Building a new structure of "buzzwords and bullshit" atop an old one of irrationality and ignorance will not fix the problem.
In the end, the private interests that are so eager to foster this activism will regret it. As more than a few princes of Renaissance Italy could testify, once you bring in the condottieri you have a problem. Before long, you don't own them; they own you. The princes of Silicon Valley will soon have cause to reflect on this lesson of history, as they plead with ignorant but arrogant lawyers for permission to make deals or enter partnerships.
As T.J. Rodgers of Cypress Semiconductor recently warned his fellow tycoons in a New York Times op-ed piece: "Winning by politics is antithetical to the free-market competition that underpins Silicon Valley's success. The Justice Department isn't just attacking Microsoft; it's attacking the way Microsoft does business--and, by extension, the way most successful high-technology companies do business."
Rodgers closed by pleading with his colleagues to go back before it's too late. It is possible they will see their own interest and bring their political weight to bear on the side of free markets. If not, then the weary witness to the predictive power of public choice theory, watching the government re-create the good old days of antitrust disasters, will be left with nothing but the consolations of schadenfreude.
Contributing Editor James V. DeLong is an adjunct scholar with the Competitive Enterprise Institute and the author of Property Matters: How Property Rights Are Under Assault--And Why You Should Care (The Free Press). His Web site is www.regpolicy.com.
--------------------------------------------------------------------------------
How to subscribe
Return to the Reason Magazine home page 칼럼】美 MS社를 위한 변명
1999.06.28
--------------------------------------------------------------------------------
美 법무부와 19개 주정부들이 마이크로소프트(이하 MS)社에 대한 반독점 소송을 제기한 지 9개월이 흘렀다. 그 9개월 동안 MS社는 업계 내부와 여론의 비판으로 만신창이가 됐지만 美 정부가 증명해 낸 것이라곤 MS社가 '정보 고속도로 산업'에서 도태되지 않기 위해 신시장 확보에 필사적으로 달려들었다는 사실 뿐이었다.
MS社의 가장 큰 혐의는 자사 운영 프로그램 윈도 98과 인터넷 검색 프로그램인 MS 익스플로러(Explorer)'를 패키지화 시켜 다른 경쟁업체들의 시장 진입을 사전에 봉쇄했다는 것이다. 그러나 지금도 수많은 소프트웨어 개발업체들이 신생 운영시스템 리눅스(Linux)용 프로그램을 개발하고 있고 MS社 역시 독점 상황을 이용해 윈도 98의 가격을 올린 적도 없다. 10년 전 전세계 운영프로그램 시장을 장악했던 MS-DOS의 가격은 수백 달러를 호가한 바 있다.
美 법무부는 또한 MS社가 컴퓨터업체들에게 싼 가격으로 윈도 98 패키지를 팔아 고객들이 다른 MS社 제품을 사용하도록 종용, 유통상의 이익을 극대화시켰다고 주장했다. 그러나 MS社의 전략은 자사 제품의 판매 촉진을 위해 시리얼 업체들이 수퍼마켓 주인들에게 웃돈을 주고 계산대 근처에 자사 제품 진열대를 마련하게 하는 '합법적이면서 고전적인 판촉전략'에 불과했다.
MS社가 컴퓨터업체들로 하여금 기능상 별다른 진보가 없는 윈도 98을 설치하도록 했다는 주장도 듣기 거북하기는 하지만 불법적인 일은 아니었음에 틀림없다.
MS社사 윈도 95, 98과 인터넷 브라우저를 패키지화한 것은 분명 경쟁상품인 넷스케이프(Netscape)의 시장점유율을 낮추고자 하는 주요 전략이었다. 그러나 그것은 질레트(Gillette)社가 기존의 두 날 면도기를 세 날 면도기로 업그레이드한 것과 하등 다를 바가 없는 기술보완으로 봐야 할 것이다. 똑같은 이유에서 MS社는 컴퓨터 아트에서부터 하드 드라이브 청소기에 이르는 각종 소프트웨어를 윈도 98에 포함시켰던 것이다.
MS社의 전략이 넷스케이프의 시장 진출을 원천적으로 가로막았다는 주장도 사실과 다르다. 넷스케이프는 지금도 널리 유통되고 있을 뿐만 아니라 부지런히 업그레이드되고 있으며 몇 달 전에는 100억 달러에 거래되기까지 했다.
현재 MS社의 윈도 프로그램은 다방면에서 경쟁업체들의 공격을 받고 있다. 강력한 도전자 리눅스가 그 세력을 급속도로 확대하고 있을 뿐만 아니라 윈도 프로그램을 탑재하지 않은 각종 휴대용 통신기기들이 속출하고 있다. 특히 소니(Sony)의 차세대 통신기기 플레이스테이션(Playstation)Ⅱ와 같은 기기는 윈도 98을 운영시스템으로 사용하지 않고도 전자우편은 물론 인터넷 검색 또한 인텔 칩보다 3배 빠른 속도로 수행할 수 있다. 한 마디로 수많은 기업들이 윈도를 대체하기 위해 엄청난 돈을 투자하고 있다는 말이다.
美 법무부는 70년대를 IBM의 독점을 막는데 허송세월했다. 당시 법무부는 사생결단의 자세로 소송에 달려들었다가 IBM의 심각한 경영난에 봉착, 회생의 몸부림을 치기 시작하자 결국 82년에 소송을 취하했다.
지난 9개월 동안에도 법무부는 MS社가 경쟁자들에게 상당히 비열한 회사였음은 증명했지만 소비자들의 권익을 침해하면서까지 반독점법을 위배했다는 사실은 입증하지 못했다. 美 정부는 반독점 소송을 접고 더이상 정보기술산업의 세세한 부분까지 규제하려는 어리석은 노력을 반복하지 말아야 할 것이다.
기고: 로널드 카스(Ronald A.Cass) 보스턴 로스쿨 교수이자 MS社 고문 변호사
저널】반독점 혐의에 대한 마이크로소프트의 항변
1999.01.31
--------------------------------------------------------------------------------
빌 게이츠 마이크로소프트(Microsoft)社 회장은 지난주 '마서 스튜어트 리빙(Martha Stewart Living)' TV 프로그램에 출연, 소프트웨어를 사용해 알파벳을 배우는 자신의 두 살배기 딸에 대한 이야기를 늘어놓으며 미국 어린이들에게 컴퓨터 세계에 빠져볼 것을 권유했다. 빌 게이츠 회장의 이날 TV 출연은 MS社에 대한 반독점 혐의에 맞서 MS社를 홍보키위한 의도에서 이뤄진 것으로 해석되고 있다. 빌 게이츠 회장은 컴퓨터 산업과 그 영역에서의 MS社 입지에 관해 독점 당국과는 완전히 다른 시각을 갖고 있기 때문이다.
우선, MS社는 스스로를 막강한 독점기업으로 여기기 보다는 단지 PC 사용자들에게 기술혁신과 소비자 만족을 제공키위해 노력하는 하나의 기업으로 바라보고 있고 아울러 모든면에서 경쟁업체들의 위협을 느끼고 있다. 이는 정부의 입장과는 상반대는 것이다. 따라서 반독점 소송의 결과는 MS社가 자신들과 같은 시각으로 세계를 바라보도록 법정을 설득할 수 있을 지의 여부에 따라 달라질 것으로 보인다.
현재 세계 PC 운영소프트웨어의 90%를 장악하고 있는 MS社가 스스로 독점기업이 아니라고 주장하는 근거는 과연 무엇일까? 무엇보다도 MS社가 인텔, 선 마이크로시스템, 오라클, AT&T 등을 포함한 거의 대다수 하이테크 업체들을 경쟁업체로 여기고 있다는 점이다.
MS社는 특히 이 기업들이 최근 급속도로 제휴를 추진해나가고 있는데 대해 상당한 위협을 느끼고 있는 상황이다. 아메리카 온라인은 넷스케이프를 앳 홈은 익사이트를 그리고 루슨트는 어센드 커뮤니케이션을 각각 인수했다. 이 거래들은 모두 MS社에 대한 反독점 재판 이후 성사된 것들이다. 이와관련, 윌리엄 뉴콤 MS社 법률고문은 "이 산업은 상당히 불안정하다"며 "상황이 얼마나 빠르게 바뀌는 지를 파악하는 것이 무엇보다 중요하다"고 밝혔다. 물론 이들 M&A 거래중 MS社의 윈도우 시장에 직접적인 위협을 가하는 것은 한 건도 없다. 하지만 그렇다고 해서 MS社의 걱정이 덜어지는 것은 아닌 것 같다.
MS社는 또한 MS社의 독점적 관행으로 소비자들이 피해를 입었다는 정부의 주장역시 도저히 납득할 수 없다며 이에 대한 반론으로 PC 소프트웨어가 꾸준히 향상되어왔다는 점을 제기하고 있다. 지난주 리차드 쉬말렌스 경제학자는 MS社의 증인으로 법정에 출석, MS社가 만약 독점을 했다면 MS社의 윈도우 소프트웨어 가격은 지금의 16배가 됐을 것이라고 증언했다. MS社는 특히 정부측 경제학자인 프랭클린 피셔가 소비자들이 인터넷 브라우저 시장에서 MS社의 행동으로 지금당장 피해를 입은 것은 아니라고 증언한 사실을 강조하고 있다. 피셔는 소비자들이 앞으로 피해를 입겠지만 아직 그 상황이 발생한 것은 아니라고 증언했었다.
MS社의 반론중 가장 흥미로운 점은 향후 PC가 윈도우와 같은 소프트웨어에서 운영되지 않을 수 있다는 것이다. 즉 MS社는 가까운 장래에 컴퓨터들이 무료로 공개된 소프트웨어에서 운영되거나 또는 인터넷을 기반으로 워드 프로세싱이나 E-메일 등과 같은 응용프로그램을 실행할 것으로 전망하고 있다. 더이상 윈도우가 필요없는 세상이 올 수도 있다는 것이다. 한마디로 MS社 관점에서 현재 MS社의 시장내 입지는 단지 하나의 패러다임 변화에 불과하다는 것이다.
업계 전문가들은 과연 이것이 가능할 지 그리고 얼마나 빨리 진행될지에 대해 열띤 논쟁을 벌이고 있다. 그러나 법률 전문가들은 MS社의 이같은 주장이 법정에서 수용하기에는 무리한 것일 수 있다고 지적하고 있다. 윌리엄 코바식 조지 워싱턴大 법학교수는 "법정에 아직 일어나지도 않은 상황을 진지하게 받아들이도록 요구하는 것은 어렵다"고 설명했다.
세계에서 가장 성공한 기업 MS社가 직접 자신의 쇠락을 예측하는 것이 다소 이상하게 들리기는 하지만 사실 월街에서는 어느정도 익숙한 이야기이다. MS社는 월街의 애널리스트들에게 늘상 전망이 암울하다고 말해놓고서는 높은 순익을 발표해 애널리스트들을 놀라게하는 것으로 유명하다. MS社는 지난주에도 작년 4/4분기 75% 증가했다고 발표한 바 있다. 투자자들은 이제 오랜 경험을 통해 MS社의 엄살섞인 전망을 액면 그대로 받아들이지는 않고 있다. 그러나 지금 MS社는 법정에 정말로 향후 상황이 암울할 수 있다는 점을 설득시켜야만 한다.
Figuring out the competitive price for Windows is not easy. In traditional industries like clothing or automobiles, we expect prices to come quite close to incremental production costs. But such expectations make little sense for Microsoft and a host of other companies that have grown out of what we call the third industrial revolution.
The first industrial revolution lasted from l760 to 1850 and was responsible for widespread innovations ranging from steam engines to iron production. Between 1890 and 1930, the second industrial revolution brought us electricity, telephones and the internal combustion engine. Today's industrial revolution stems from two innovations from the second half of the 20th century. First, a rudimentary international computer network (later to be called the Internet) and second, the development of the world's first microcomputers (better known today as PCs).
The first two revolutions grew through the expansion of physical capital like railroads, factories and electric networks. Today's revolution is driven by an ever-growing number of firms that operate in health care, technology, telecommunications and financial and consumer services. Some of these firms are providing Internet services, while many others are taking advantage of the Internet through electronic commerce. While many of the old economic rules still apply to these new companies, others need to be reconsidered. In particular, we should be wary of policymakers in the competition field who are eager to apply yesterday's rules indiscriminately to today's changing industries.
Traditional economics has always been quite useful for understanding the market forces that shaped industries and governed competition among firms during the first and second revolutions. The principle of marginal cost, the theories of supply and demand and profit maximization and a little bit of game theory sufficed to study competition in industries with just a few firms. These tools helped economists and policymakers analyze markets in which firms like Procter & Gamble, General Motors and U.S. Steel competed. But those firms are significantly different from Yahoo, America Online and Microsoft.
The Theory of Competitive Races
The race to develop the next killer product is the name of the game in many high-technology industries today. In the software industry, "killer apps" are those applications around which an entire new industry or category grows. Many firms invest heavily to develop a product that creates a new category. Visicalc did that years ago when it came out with the first electronic spreadsheet. As a result, Visicalc "owned" that category for several years.
While some firms attempt to create and lead new categories, others spend money trying to leapfrog existing leaders. That's how Lotus 1-2-3 stole the lead from Visicalc back in the mid 1980s. That's the way Excel came and ate Lotus' lunch in the early 1990s. And that's how Nintendo, Sega and Sony have been competing for the last few years. Just recently, it looked like Sega's Dreamcast might put Sega back in the game console lead. Too bad Sony's new Playstation 2 has a 128-bit processor, which runs games three times faster than Intel's Pentium III.
Economists have developed a rich array of "racing" models to describe dynamic technological competition among firms. Firms take large risks by investing to win tremendous payoffs. The larger the potential payoff from winning a race, the more firms are attracted to that race, and the more vicious is the competition to land up on top. Many of these races have only one winner. As a result, winners take all (or most), while losers often lose everything (or almost everything). Indeed, some of the racing models suggest that there can be "too much" competition. Firms altogether invest too much in trying to win a prize that is rarely shared among runners-up.
It is challenging for investors to identify the winners in these races. Unlike a tennis tournament, where all participants are seeded according to their individual likelihood of winning, participants of third industrial revolution races can be in the heat of battle, yet have no idea of their future competition. Many of their competitors won't even weigh in until tomorrow's ringing of the Wall Street IPO bell.
Additionally, competitive races are plagued by tremendous incentive and information problems associated with the financing of innovation. Innovative projects are typically associated early on with a great deal of uncertainty. Information gaps between researchers and investors are commonplace. These projects typically have substantial intangible assets, which are difficult to value and may be impossible to resell if a project fails. Similarly, market conditions in many of these industries are highly variable. The nature and magnitude of these problems generate many opportunities for unsuccessful investment decisions. Not surprisingly, venture capitalists bet on only a small fraction of the proposals that are brought to them and make money from only a small fraction of the proposals they fund.
Racing models predict that in high-technology industries, most firms will lose money and only a few will win. There is considerable empirical evidence of this. One such study tracked more than 12,000 software firms between 1990 and 1998. During this time, less than half experienced growth in sales or employment.1
Network Externalities
High-technology industries are often shaped by "network externalities." Such externalities exist when a particular product or technology gains value as a result of other consumers' acceptance of that technology. Economists' favorite example is the fax machine. Fax machines are virtually useless unless many people have them. The first consumer to purchase one was essentially investing in a conversation piece and betting that in the future, many people would make similar purchases. Now that virtually all businesses and many households own them, fax machines are valuable precisely because of their ubiquity.
Network externalities can help a firm quickly gain dominance in a category. Each consumer who chooses one firm's product makes that product more valuable and increases the likelihood that the next consumer makes the same purchase. As a result of this positive feedback loop, the majority of consumers can end up choosing one firm's product. Of course in reality, the value of network externalities is usually more complicated and less direct. For example, Consumer A's Visa card does not become more valuable when Consumer B obtains one. Yet, the more people who own Visa cards, the more attractive card acceptance becomes to merchants. To take another example, no two consumers benefit directly from using the same computer operating system. However, if many people use the same operating system, that system becomes more attractive to applications developers who want to write to a platform that has many users.
Pricing When It's Cheap to Copy
Many high-technology industries are based almost entirely on intellectual property. This type of capital is usually expensive to create, yet cheap to duplicate. Consider computer software. Creating software is a highly labor-intensive process. Yet once the software has been written, it can be duplicated onto a CD for less than a dollar or posted on the Internet for download at even lower cost. As Netscape's Jim Barksdale noted, "We can distribute globally at the blink of an eye." Besides just software, the Internet can be used as a distribution channel for content too. Once a Web site has been created, the cost of maintaining that site does not grow from incremental increases in visitors. As a result, content distribution also incurs virtually zero marginal costs.
Drugs provide another example of an industry characterized by high fixed costs and low incremental costs. In its race to develop one of the first successful treatments for impotence, Pfizer invested more than a half billion dollars. Yet now that Viagra has been developed, it is quite inexpensive for Pfizer to stamp out thousands of additional pills.
High-technology companies have many more considerations than other companies in setting prices. Most companies determine prices based on their production costs. They mark up those costs, taking into account consumer demand and industry competition. Indeed, in many industries there are even rules of thumb for marking up costs. Not so for high technology. Owners and investors spend millions of high-risk dollars funding these ventures. For any firm to secure such high-risk capital, it must set prices high enough to compensate investors for the risks they take on. Of course, firms cannot set prices too high, particularly if they are concerned about the effect of prices on network externalities. Low prices encourage consumers to try their products and, through the positive feedback effects described above, may create additional demand. On the other hand, many high-technology firms are less concerned about losing sales to other competitors. They may not have any significant ones if they have a patent or copyright that prevents short-run imitation.
The Government's Case Against Microsoft
Some of the complaints against Microsoft2 illustrate the importance of understanding the economics of companies that are warriors in the third industrial revolution. In the antitrust lawsuit against Microsoft, the government (actually the U.S. Department of Justice and 19 states) has said that Microsoft charges too much for its Windows operating system. Most people either get Windows as an upgrade for about $89, or they receive a copy installed on their new computers; the computer manufacturers pay about $65 per copy to install Windows on new computers.
How would one know whether Microsoft's price is "too high?" With a second industrial revolution company, economists would ordinarily look at the incremental cost of production. We expect that price should roughly equal incremental cost (plus a return just high enough to attract capital to that industry). Unfortunately, such a simple calculation does not make any sense for a firm like Microsoft. The incremental cost to Microsoft of each additional copy of Windows installed by manufacturers is literally zero. Microsoft sends the computer manufacturers a master disk, which the manufacturers then copy onto their machines. It is therefore not surprising that the government's economist had no idea what the competitive price for Windows really is.
Microsoft's price actually does not appear to be "high" when compared with other operating systems. Apple charges $89 to upgrade to their new Mac OS 8.5. An upgrade to IBM's OS/2 Warp costs $140. And while Red Hat Linux costs less than $40, a consumer who wants any other Unix-based operating system can expect to pay $450 for Sun's new Solaris 7.0 or up to $795 for SCO's Unixware 7. Just like Microsoft, all these firms face the difficulty of setting a price based on many considerations. But unlike textbook discussions of competition, not one of them makes this decision based on incremental production costs.
Economists don't have a simple theory to explain pricing behavior among firms that sell intellectual capital at prices far above marginal cost. It is clear that these firms cannot be expected to charge a price of zero for their product. No firm would enter any competitive race if the trophy awarded to the winner were really only a booby prize. If we want firms to continue making high-risk investments (a proposition that no doubt benefits consumers and the economy as a whole), we have to realize that these firms will require high returns on their investments—and therefore high margins on sales.
The government has also said that Microsoft charges too little for its Web-browsing software, Internet Explorer (IE). Microsoft decided to include IE as part of Windows, and therefore does not charge separately for it. It has also designed versions of IE for other operating systems like the Mac. It gives those away. According to the government, Microsoft gave away IE to help destroy Netscape, which used to have the most popular Web browser. The government maintains that Microsoft wanted to destroy Netscape for fear that their browser could provide a potential rival to Windows. There are, however, innocuous explanations for giving away browsers for free. Interestingly enough, Netscape chose to give away most of its browsers for free, too.
Two considerations make this zero price economically sensible. The first is that Microsoft and Netscape could expect to earn revenues from other sources—complementary revenues—if people used their browsers. Microsoft could expect that more software developers would write software for Windows, and this would increase the value and sales of Windows. Additionally, both Microsoft and Netscape could expect to earn revenues from advertising and other sources if they were able to direct users to their Web portal sites (sites that serve as gateways to the Internet). All else being equal, the more people who use their browsers, the more likely people are to visit their portal sites. When America Online finally purchased Netscape, it spent more than $10 billion of its own stock to do so. Obviously, the company saw sufficient potential for complementary revenues in Netscape, despite Netscape's inability to earn revenues from sales of its browser.
The second reason why a zero price for browsers is economically sensible is the existence of network externalities. The more people who use a browser, the more valuable it becomes. Again, the externality is indirect: more users encourage more applications which encourages more users and so forth. A low price gets more users, more network externalities and more complementary sales.
The government isn't just concerned that Microsoft charges too little for IE. They don't like the fact that it is always sold as a part of Windows and consider it to be a case of illegal bundling, or "tying," of two separate products. While economists think that tying is seldom a bad deal for consumers, the antitrust courts have made it tough for companies with large market shares to engage in this practice. One of the most famous cases involved a hospital just outside of New Orleans that required patients to use its staff anesthesiologists; the Supreme Court thought the hospital had tied the two products together but let the hospital off the hook because it did not have a large market share.
Unlike the tying of hospital services, the bundling of software is tricky to define. Without passing judgment on the merits of the practice, it is easy to see that surgery and anesthesiologists are two quite distinct products. Software code is far more complicated. Just like the words in a book, software code can be organized in various ways to achieve a multitude of effects. Over the course of time, software producers add features to their programs that previously were sold as separate products. Spell checkers, for example, are now a standard component of most word processing packages. People also used to buy separate "utilities" programs to enhance their operating systems. Many of these are now integral parts of most operating systems. Integrating browsing capabilities into an operating system seems the next logical step—operating systems read files from many sources and the Internet is just the latest.
Microsoft and the government have taken quite opposite views of the right way to think about competition in the software industry. The government sees Microsoft as making more than 85 percent of the sales of operating systems for personal computers. By traditional standards that is a high share. And it sees Microsoft as facing few competitors. Computer manufacturers have few alternatives available today for personal computers for home or business use. As a result, Microsoft can price high and wield power.
That's a sensible analysis for second industrial revolution companies. But using traditional economic models to analyze industries characterized by high risk network externalities and zero marginal costs can be a mistake. In the case of computer software, consider Microsoft's position.
Microsoft sees itself in a perpetual competitive race—always looking over its shoulder to see whether known rivals like Apple or come-out-of-nowhere rivals like Linux are catching up. Because software development requires only limited resources, Microsoft is constantly threatened by behemoths like IBM as well as tiny startups like Red Hat that have bright new innovative ideas. Microsoft, therefore, can only view its market share as fragile and transient. It believes that it prices low, because if it doesn't, it will be replaced.
Microsoft and the government have two opposing views of competition in today's high-tech industries. The prevailing view will prove important not only for Microsoft, but for many other companies as well.
--------------------------------------------------------------------------------
1 Lerner, Josh, "The Returns to Investments in Innovative Activities: An Overview and Analysis of the Software Industry," Harvard Business School and National Bureau of Economic Research, September 1998. Microsoft funded the research for this paper.
2 We are working for Microsoft on this litigation. See http://www.nera.com for other papers related to this case.
Viewpoint, Number 1, 1999
Copyright © 1999 by Marsh & McLennan Companies, Inc. All rights reserved.
--------------------------------------------------------------------------------
Back to Top | MMC Views home page
Past Editions Menu
Table of Contents
April 25, 2000
--------------------------------------------------------------------------------
Editorial Page
Why the Case for a Breakup Breaks Down
By Richard B. Mckenzie and William F. Shughart II. Mr. McKenzie, a professor at the University of California, Irvine, is author of "Trust on Trial: How the Microsoft Case Is Reframing the Rules of Competition" (Perseus, 2000). Mr. Shughart, a professor at the University of Mississippi, is author of "Antitrust Policy and Interest-Group Politics" (Quorum, 1990).
The Justice Department is awfully busy this week. By Friday its Antitrust Division is due to submit a proposed remedy in the case of U.S. v. Microsoft. The department reportedly will ask Judge Thomas Penfield Jackson to order Microsoft to break itself up into two or more separate companies, so-called Baby Bills. The advocates of a breakup contend that dissolution is preferable to a "conduct decree" that would leave Microsoft intact but impose restrictions on its behavior.
Advocates of a breakup argue that it would make software markets more competitive and would avoid the continuing judicial and regulatory oversight a conduct decree would entail. They point out that after Standard Oil was dissolved in 1911, the collective market value of the 30 successor companies quickly rose above the oil trust's pre-breakup market value, making John D. Rockefeller Sr. into history's first billionaire.
The implication is that the same antitrust medicine for Microsoft will be beneficial all around -- to consumers, who would get more competitively priced software, and to Microsoft's owners, who would become wealthier than they already are. How could anyone object to such a win-win solution?
***
The problem is that the underlying argument is wrong on both the facts and the logic. No wonder that news of a possible breakup helped send the Nasdaq Stock Market into a tailspin Monday.
The Standard Oil comparison is based on a half-truth. Yes, the total market value of the 30 "Baby Oils" tripled between 1911 and 1913. But the stock market was rising rapidly across the board. A new study by VMI economist Atin Basu Choudhary and University of Mississippi economists Robert Tollison and one of the authors (Shughart) finds that, contrary to conventional wisdom and independent of other market forces, the dissolution of Standard Oil had no persistent impact on the wealth of the company's owners. More to the point, Rockefeller's stock-market gains were not proportionately greater than those of ordinary investors in other oil and industrial companies. Hence, the rise in the market value of Standard Oil after its dissolution can't be used to justify a breakup of Microsoft. Indeed, these economists argue that the Standard Oil case is a classic example of antitrust policy failure.
The logic underlying the contentions of Microsoft's critics is also faulty. If Microsoft were truly acting like a predatory monopoly -- restricting sales with the intention of raising product prices and garnering profits in excess of competitive levels -- then a breakup would be expected to lead to more intense price competition and, therefore, lower profits and a lower stock-market value for the Baby Bills. By arguing that a breakup would increase the Baby Bills' total stock-market value, the critics must believe that in some way the breakup will make software markets less competitive, with the prospects of higher prices and profits.
What about the argument that a breakup would press the Baby Bills to become more efficient, enhancing their values? That is sheer speculation, not well grounded in theory. After all, even monopolies have strong incentives to hold their costs in check.
As for the claim that a breakup would reduce software prices, the reverse is probably true. According to economist Stan Liebowitz, a breakup of Microsoft into two operating-system companies could increase software-development costs by $30 billion in the first three years, given that programmers would have to adapt their applications to more than one version of Windows. Application companies surely would pass these costs on to their customers.
A true monopoly doesn't have to compete, and Microsoft is nothing if not a fierce competitor, using its market prowess to win competitive races with rivals, such as Sun, Netscape (now part of America Online) and IBM. Microsoft's interest in doing that is a plus for consumers. By not acting like a monopolist, it holds its prices (and profits) below what they would otherwise be. While such a pricing strategy might make life difficult for Microsoft's rivals, consumers benefit by getting higher quality software products and lower prices than they would otherwise.
***
Under these circumstances, the proposed breakup of Microsoft amounts to a plan to force consumers to pay higher prices than they are now. This is not exactly the kind of outcome the Justice Department and Judge Jackson should want or seek -- if their real goal is to benefit consumers.
What is telling -- and disturbing -- about the Microsoft case is that it is the company's rivals, not consumers, who are pressing for a breakup. Microsoft's rivals should want Microsoft to act like a monopolist -- that is, restrict its output for the purpose of raising its prices. They would then be able to sell more of their own products at higher prices. The rivals would not rationally support any remedy that will make their markets more competitive, given that more-competitive markets would mean lower prices and profits for them. The rivals' support of the proposed remedy suggests that the effect of a breakup will be to do what antitrust enforcement should never do, help competitors at the expense of consumers.
--------------------------------------------------------------------------------
Return to top of page | Format for printing
Copyright © 2000 Dow Jones & Company, Inc. All Rights Reserved.
Copyright and reprint information.
REASON * March 1999
The New Trustbusters
What's behind the resurgence of antitrust activism--and why it's bad news for consumers.
By James V. DeLong
Joel Klein is a famous man. The head of the Antitrust Division at the U.S. Department of Justice usually toils in anonymity, known only to the in-groups of the bar. Not Klein. He has sued Microsoft, the most prominent company in America's jazziest industry, and demonized the world's richest human, Bill Gates. He has assaulted the ubiquitous credit card franchises of Visa and MasterCard and blocked important aerospace industry consolidations. He makes speeches extolling the pivotal role of the Antitrust Division in the "new economy" of globalization and information. These acts have earned him serious attention in the national press.
Not to be left behind, Klein's fellow antitrusters at the Federal Trade Commission are equally active. They have beaten up on superstores by stopping the merger of Staples and Office Depot and by knocking down important marketing practices of Toys "R" Us. Like the DOJ Antitrust Division, the FTC has gone high tech. It is challenging Intel, the grandee of computer chips, and probing Cisco Systems, the dominant company in Internet switching hardware. In a recent suit against a drug company, the FTC asserted a heretofore unknown authority to force an alleged monopolist to refund to consumers $120 million in allegedly ill-gotten profits.
Professional discussion of this surge in antitrust activism is proceeding at two levels. One is the highfalutin language of bar association meetings and academic conferences, where sessions are given titles like "Antitrust Enforcement and High Technology Markets," "Networks, Lock-In Effects, and the New Economy," and "New Approaches to Reviewing Horizontal Agreements." In this world, Platonic guardians mull over the implications of technological change and devise optimal policies to safeguard American enterprise. They "open up a dialogue with the bar and the academic community about antitrust doctrine," as Klein put it in one of his speeches, on topics such as "the relationship between antitrust and intellectual property, the significance of...network effects and tipping points...
the impact of differentiated products theory, the role of potential competition and innovation markets, and [the meaning of] `agreement.'"
The second level of analysis is more skeptical, grounded in the public choice school of economics. Public choice theory holds that exercises of government power are driven by the material and ideological interests of the people who wield it and by the private parties who can reward them with campaign contributions, money, job security, or whatever else they value. Viewed through the lens of public choice theory, the recent burst of antitrust activity is not primarily about consumers or competition. It is about four kinds of interests: 1) competitors of successful firms who want to hamstring their rivals, appropriate part of their businesses, or turn the clock back so they can re-run the race; 2) companies blown by the winds of technological change and looking for ways to nullify their disadvantage; 3) the personal ambitions of antitrust enforcers, who do not prosper in quiet times; and 4) the class interests of the legal profession, which are served by a combination of activist bias and mushy theories.
Only a taste of this view creeps into the formal talks at antitrust conferences. To get the full flavor, you must hang out in the corridors and hotel bars, and stick around for the after-conference receptions. There you hear jokes like this one: "What is the government's theory in the Microsoft case? That the state of California has more computer companies than the state of Washington, and a hell of a lot more electoral votes." Or this one, about "Gore-Techs," the high-tech entrepreneurs who confer with Vice President Al Gore: "What is Gore-Techs? A new fabric made by combining silicon and money, used for wrapping up politicians."
In these less formal settings, last summer's Senate Judiciary Committee hearings on Microsoft, in which Chairman Orrin Hatch (R-Utah) laid into Bill Gates, are dismissed with, "What else would you expect from a senator who has Novell in his state?" Testimony in the ongoing Microsoft trial about the frequency with which the head of Netscape, the moving force behind the lawsuit, met with Klein and other Antitrust Division representatives is greeted with blasé yawns. So are testimony and e-mail messages showing that high-tech companies regard government antitrust action as simply one more tool in their competitive arsenal, an alternative to price cuts or new products. And as the head of Netscape said when asked why he did not file his own suit, it is much cheaper to use the government's lawyers.
Behind the scenes at the conferences, you hear snickers about the innocence of Microsoft, which thought it could sit out there in Redmond writing software and ignoring Washington, D.C.--as if such a big pot of wealth could go unnoticed by a rapacious imperial capital. The idea that perhaps a company should be able to do its business and ignore Washington is regarded as hopelessly naive. Microsoft is now playing catch-up, adding former congressional aides to its staff and boosting its budget for political contributions.
The suit against Visa and MasterCard, announced in October 1998, also evokes grins. (See "Credit Where It's Due," January.) The Antitrust Division's press release speaks of harm to consumers and of an industry that is "competitively impaired." But the arrangement that the government just attacked has been around a long time and was created in part due to antitrust concerns expressed by the DOJ almost 20 years ago. Prior administrations investigated and declined to take action. The corridor talk notes the close relationship between the Clinton administration and American Express, which has been complaining about its rivals for years. Presidential friend Vernon Jordan is an AmEx director. According to news accounts, AmEx has other ties to the White House: It bought heavy advertising from the 1993 inaugural committee, it won the White House Travel Office account, and it has replaced Diner's Club as the government-issued credit card for federal workers.
In the 1997 Office Depot/Staples merger case, the FTC defined the relevant market as "office supplies sold by superstores," even though the two chains together sell only about 5 percent of all office supplies. Wal-Mart alone sells more than this, and bulk mail-order firms sell another huge chunk. Everyone, except the FTC and the federal district judge who upheld its decision, regards the commission's market definition as a laugher. (See "Pricing Pencils," August/September 1997.) But no one can figure out who had the political clout to persuade the FTC to adopt such a silly position. The Wall Street arbitragers who lost a bundle when the deal cratered think the FTC's action was triggered by complaints from Office Max, the third major superstore chain, which stood to get some assets on the cheap if the government forced the merged entity to spin them off in the name of protecting competition.
Last March, when the DOJ Antitrust Division nixed Lockheed Martin's effort to acquire Northrop Grumman, the formal reasons involved concentration in the defense industry. The corridor talk this time actually made it into The Wall Street Journal. Lockheed had been blindsided by rival Raytheon, which gave the government mountains of negative information. In its June 1998 story, the Journal noted that Lockheed, not expecting trouble, had failed to adopt the usual techniques of the merger game: "lobbying Capitol Hill, working the executive branch and creating a drumbeat of support in the media."
The interests of selected businesses and political figures are not the only ones served by the trend toward antitrust activism. The corridor lounger also notes a certain incestuousness among the conference participants. Almost all the private lawyers and economists on the panels used to work for the government, while the government reps used to work in the private sector, and most will return there at some point. Even the career employees turn free agent if the stars are right. The FTC litigator who led the Staples case left a few months later for a well-paid partnership in a major law firm. In the condottiere world of lawyering, his victory in a case that looked like a dog greatly increased his market value. The arbs who lost big as a result want him on their side next time.
It is a new springtime for antitrust lawyers and economists, promising a return to the good old days of three decades ago. Their memories of that era are bathed in a golden glow. It was a time when antitrust regulators were hyperactive, inhibited only by the need to invent a plausible scenario under which a business arrangement might be regarded as "anti-competitive," and this concept could be given any of several not necessarily consistent meanings. Given such a nonstandard, the government won almost every time it challenged a business arrangement. But the charade required a lot of lawyers and economists.
This era ended in the 1970s. Two disastrous pieces of monster litigation--one against IBM by the DOJ Antitrust Division, the other against the oil companies by the FTC--ground on for years. Each went long past the point where it became obvious that the government lacked a coherent theory of the supposed offenses and long past the point where changes in the industry and in the world turned the case into a joke.
During the same period, the University of Chicago school of antitrust analysis began to convince judges and law students, and eventually even the bar, that many of the government's anti-competitive scenarios were not plausible at all and that arrangements which were valuable to consumers were being outlawed. Antitrust law was suppressing rather than nurturing competition.
This revolution in antitrust thinking was consolidated in the 1980s, when the Reagan administration pared enforcement back to the hard core of attacking agreements to fix prices or divide markets, which almost all antitrust analysts regard as beyond the pale, and preventing mergers among direct competitors in highly concentrated industries. The number of cases dropped off as enforcers ceased to invent new theories, and businesses, with a fairly clear idea of the line between legal and the illegal, could adjust their conduct with little reference to the high priests of antitrust.
Antitrust practice came to mean filing the notification forms the government requires before a corporate merger can be consummated and negotiating the terms on which the government would approve. This often meant selling off a division or two in the name of preserving "competition" in some minor market or product line, but the issues were usually marginal, rarely involving fundamental questions of antitrust purpose and policy. Antitrust became a particularly boring kind of regulatory law, consisting mostly of copying thousands of pages of documents to ship over to the DOJ and the FTC for pre-merger review.
Now, after almost two decades of marginal relevance, the thrill of activism is back. Corporate officers and Wall Street arbs once again furrow their brows when they speak of "The Division" or "The Commission." The telephone rings. Hours are billed. People who say "let's do lunch" actually follow up. Headhunters pass out business cards. Members of the opposite sex no longer flee when you say, "I do antitrust."
It is important to avoid what the English writer C.P. Snow called "the cynicism of the unworldly." One should not underestimate the impact of sincere belief on the actions of government officials. The Antitrust Division, the FTC, and the private bar are full of honest lawyers who would be outraged at the suggestion that they are acting for personal or class interest, or that they are stretching the law in the interests of ideology. But it is also important to recognize that self-interest is the aphrodisiac of belief and that adopting the theories underlying the new activism is much to the benefit of the antitrust professionals on both sides. To those with a bent for public choice explanations, the enforcers, flogged on by the politicians and the competitors, have seized upon the computer, the information revolution, and the Internet as excuses to reverse the cautious "first, do no harm" antitrust policy of the past 20 years and restore the promiscuous activism that characterized the 1960s.
The question, of course, is whether these dour apostles of the public choice view are right. Answering that question requires a return to the basics. The antitrust laws stem from our collective fear of monopoly, a bred-in-the-bone knowledge that a supplier of a good or service who lacks competitors will jack up the price, cut the quality, become arrogant and unresponsive, and in general behave obnoxiously. Economists can give you good explanations for this, complete with charts and graphs of supply and demand curves made out of solid and dotted lines and annotated with lots of alphas and deltas. They can explain with precision why raising the price increases the monopolist's revenue even if it cuts sales, and they can tell you the size of the price hike that will maximize the loot.
The noneconomists among us fear monopolies even without the charts and graphs, based on common sense. After all, raising the price is logical; it is what we would do if we had a monopoly of our own. We also look to our experience. Think of the U.S. Postal Service, cable television companies, state liquor stores, or the local public school. Monopolies all, and legendary for unresponsiveness, ineptitude, and overcharging. But these things are mostly peripheral to our lives, or there are safety valves in the form of substitute goods or direct action, so the lack of competition is an annoyance rather than a disaster. Imagine the impact of a complete monopoly in a truly essential product, such as automobiles or groceries.
The fear of monopoly that triggered the first of the big antitrust statutes--the Sherman Act of 1890, which outlawed combinations in restraint of trade and attempts to monopolize--was far from unfounded. Alfred D. Chandler Jr., the eminent business historian, documented the situation in the late 19th century in his classic book The Visible Hand. The rise of mass production, combined with the railroad, the steamship, the telegraph, and the telephone, created the possibility of nationwide marketing and management. As output exploded, prices dropped precipitously. Many industries responded by attempting to put together cartels intended to reduce output and stabilize prices. The means varied--contracts, holding companies, corporate voting trusts, mergers--but the goal, as explicitly stated by the participants, was to attain the joys of monopoly power over a market.
Considering these expressed intentions, public fear was rational, but history has shown that it was largely unnecessary. It turned out that amalgamation is not alchemy, and the trusts could not achieve their goal of excluding competitors and sustaining prices at monopoly levels. The experience of the National Cordage Association, recounted in Chandler's book, illustrates the vulnerability of monopoly schemes. The NCA could not achieve economies of scale in operation or management. Nonetheless, it was forced to pay premium prices to lure competing companies into selling out, which stuck it with huge capital costs. The bought-out competitors then used their money to start new cordage companies that were more efficient than the NCA, so they could undercut its prices. The association went bankrupt in 1893.
Even mighty Standard Oil, the great bête noire for trustbusters of the late 19th and early 20th centuries, was a symbol more than a true villain. By 1911, when the Standard "monopoly" was broken up by the Supreme Court, eight other large integrated oil companies were competing with it. Before that, Standard never tried to sustain prices at high levels. The history of oil during the late 19th century was one of huge expansion in markets and facilities and steadily falling prices.
Many of the complaints about Standard came from medium-sized refiners that lost cozy local monopolies to Standard's rationalizing and price cutting. The same pattern has been repeated many times. As changes in communications and transportation create possibilities for new forms of business organization that might make things cheaper and better for consumers, firms that are doing well under the old forms fight back. This was true in the late 19th century, and it remains true today.
Often, the beneficiaries of the status quo cry "Monopoly!" Almost always, they are wrong. If you look back at the examples of monopolies given a few paragraphs ago, you will notice that all are public institutions or publicly regulated businesses. It is impossible to find examples of truly private monopolies, except on a minor scale. In areas where the natural technical and economic structure of an industry would tend toward monopoly, such as utilities or railroads, public regulation was imposed early (with dubious results--but that is a different tale). In other sectors, monopolies are rarely developed and never sustained.
Some thinkers, such as Dominick T. Armentano, a professor emeritus of economics at the University of Hartford and a scholar of antitrust history, have responded to this experience by advocating the elimination of the antitrust laws. They argue that the benefits from the few instances in which these laws might do good are outweighed by the harm caused by their inevitable misapplication at the behest of the political system.
Others are less willing to scrap the antitrust laws. They note that it's always possible that the members of an industry might succeed in getting together and setting a monopoly price. Perhaps in 1911 Standard and the other eight big oil companies could have set up regular meetings and fixed prices or merged into one company. Over the long term, such a monopoly would not last, any more than OPEC lasted. As the price remained high, new entrants would want to get a cut and would have to be brought into the cartel. Eventually, shares would get so diluted that some members would find it worthwhile to break off and cut the price, thereby ruining the system.
But a cartel can cause pain while it lasts, as did OPEC, and the wait for its inherent contradictions to assert themselves can seem awfully long for us short-lived humans. So most people who study the problem, whatever their general political orientation, agree that a core of antitrust enforcement is important. In particular, they agree that price fixing among competitors should be outlawed, and so should merger to monopoly and agreements to divide markets and set up a series of monopolies in different sectors of the economy. This was the basic antitrust policy of the Reagan years.
As the abolitionists love to point out, though, when you take this first step you step on an intellectual banana peel that sends you skidding down a slippery slope. The fundamental problem is that the meaning of "monopoly" is amorphous, and the ambiguity leaves unclear exactly what the law is trying to prevent, encouraging ad hoc approaches and governmental mischief.
A true, clear monopoly would be something for which there are no substitutes. If you are drowning in the middle of a lake, and someone in a boat comes along and asks if you want to buy a life jacket, he can fairly be said to have a monopoly. Short of a situation like that, there are only varying degrees of power to charge a premium price. A beef company's power is limited by the availability of chicken and fish, and a local grocery store's power is limited by your ability to shop in a neighboring town. In each case, the power is real but limited. So at what point does this power justify government action? There is no clear answer, and no satisfactory theoretical basis for developing one.
This fuzziness gives those interested in expanding antitrust enforcement, including the enforcers, a lot of leverage. They argue that any power over price, however small, in any identifiable product market is illegal. Then they seek to define the market narrowly, making it easy to find a price effect. In a recent paper for the Cato Institute, University of Mississippi economist William F. Shughart II, one of the most relentless public choice critics of antitrust policy, lists some markets defined by the government: "high-priced, non-ethnic frozen entrees"; "noncarbonated, ready to serve, naturally or artificially flavored fruit drinks, fruit punches, or fruit ades which contain 50 percent or less fruit juice and are customarily sold under refrigeration to the consumer"; "direct contract front-loaded trash removal in Dallas." These market definitions do not have quite the same impact as "oil" or "steel."
The next step in expanding the realm of antitrust is to use static economic models to predict an effect on price, often ignoring strategic responses available to competitors that would erode any monopoly power. In analyzing the Staples/Office Depot merger, for example, the FTC noted that slightly higher prices prevail in some areas where only one of the companies operates. It assumed this pattern was immutable, ignoring the ability of superstores, mail-order houses, and customers to observe it and react.
By the time these steps are taken, the result is ordained. Anyone playing by these rules who cannot find a monopoly on every corner isn't trying.
This game discourages efficiency-enhancing mergers. It also discourages arrangements among firms, such as joint ventures and strategic partnerships, that could create innovative products and reduce costs. Such deals always involve some agreements on price and territory because the partners must decide how to split the proceeds. Each also needs to ensure that its partner does not use it for a time, then muscle it out of the way and take over the whole business.
Banning these contracts outlaws many worthy projects. As Fred Smith, head of the Competitive Enterprise Institute, explains: "Any company wants to do many different things. But it can be only one size, which means it cannot be exactly the right size and have exactly the right mix of skills and resources for each of the things it wants to do. For some it is too big, for others too small." If different and possibly competing businesses can integrate some of their operations without merging entirely, they can form entities that are the right size for the venture at hand. The size that is appropriate for inventing a software product, for example, may be different from the size needed to make it commercially viable. Manufacturing and marketing require still different sizes, talents, and structures. The industry is filled with people patching together the temporary alliances needed to fill the gaps in their own organizations. The result, in software and elsewhere, is a flexible and adaptive economy, which serves consumers well.
For a century, antitrust enforcement has been hostile to such partial integrations among businesses. Ironically, this hostility creates pressure for firms to merge completely. Due to oddities of antitrust law, merger may be allowable under circumstances where partial integrations by contract would be blocked.
In the 1960s, scholars with a public choice perspective began analyzing whether past antitrust actions had actually benefited consumers. It is not an easy question to research, but a variety of ingenious studies have looked at stock prices, scholarly opinions, competitors' reactions, and other indicators. The results are strikingly consistent: Antitrust actions do not help the public, though they may help the special interests that trigger them (while providing well-paid employment for antitrust professionals). CEI's 1997 Antitrust Reader and the 1995 book The Causes and Consequences of Antitrust: A Public Choice Perspective (University of Chicago Press), edited by Shughart and Emory University economist Fred McChesney, are good introductions to this literature.
This research, combined with the dearth of empirical work on the other side, was an important factor in the reforms of the 1980s. Proponents of interventionist antitrust had an embarrassing lack of good examples. As McChesney and Shughart point out, when you have a century of experience with a program and virtually every landmark case looks to have been a mistake, perhaps it is time to stop saying, "well, we'll get it right next time," and start rethinking the basic premises.
The current activists are having none of this. Quite the opposite. The seminar sessions at antitrust conferences--the serious part, where the participants do not joke about Gore-Techs--operate as if none of this rethinking happened. The governing assumption is that basic antitrust policy and doctrine was and is sound. The question is whether this tool that has served us so well can be applied to the growing high-tech economy as it stands, or whether it needs to be adapted and extended to reflect new realities. The possibility that the antitrust emperor might be stark naked is ignored.
This should not really be surprising. More than 20 years ago, when I was a middle manager in the FTC, an exasperating commission action prompted me to say to another staffer: "They tell the story of `The Emperor's New Clothes' a little differently around here. It goes along as usual until the little boy pipes up, `That man is naked!' Then, in the government version, the emperor turns and says, `Kill that kid.' And they do." Public choice analysts know that an agency-emperor will seize any opportunity to off the uppity kid and re-impose the myth of the beautiful robes. It will be assisted enthusiastically by private interests who want to turn the myth to their own advantage.
As a result, there is now a gusher of amorphous theories justifying renewed antitrust activism. For example, theories of "path dependence" and "lock-in" hold that society can become committed to an inferior technology, unable to break free when a superior one comes along. It is an interesting idea with little empirical support, as has been demonstrated by economists Stan Liebowitz and Stephen Margolis in the pages of REASON ("Typing Errors," June 1996) and at greater length in the academic literature. The Dvorak typewriter keyboard was not really significantly better than the old QWERTY version, and now that anyone can go Dvorak with a computer keystroke, almost no one does. The other standard example, the triumph of VHS over Beta in videotape, is also wrong. VHS, with its longer recording time, was regarded by consumers as a better product.
It's true that once companies or consumers buy into a technology they will incur costs if they want to change later. But this is hardly a new development; the same is true for purchases of equipment, constructing buildings, or any other investment. Change occurs only when a new technology is sufficiently superior to justify the switching costs, or when investment is turning over anyway. So what? The fact that a company that introduces a good product gets some first-mover advantages is one of the mainsprings of innovation. Is the government now saying this should be foreclosed?
Another phrase popular among antitrust activists is "network effects." Some things become more valuable as more people sign up, and the company that gains an initial advantage may then sweep the field. The classic example is the telephone, where everyone wants to be on the same network.
Again, an interesting idea, and one with clear merit. But again, not new, and not unique to high-tech industries. Timothy Muris, now a professor at George Mason University law school and, as head of the FTC's Bureau of Competition, an architect of the antitrust reforms of the 1980s, notes: "The fact that network effects are everywhere should give us pause about the utility of the concept [for antitrust]. Many products, not just high-tech ones, have the characteristic [that] the benefits of use increase as the number of users grows. Thus, consumers of products that require post-sale service, such as automobiles and appliances, produce network effects from the growth of service outlets when more consumers purchase the product. Coke and Pepsi drinkers benefit from the network of their fellow consumers in that their drinks are widely available in restaurants and from vending machines. Sports fans benefit when they live in an area with enough other fans that teams find it profitable to locate there."
During the formal presentations at antitrust conferences, the audience listens raptly as phrases such as "path dependence" and "network effects" roll off speakers' tongues. During the receptions the respect drops off, especially because the speakers seem more interested in explaining why the concepts justify an increase in their power than in any real analysis. A typical corridor reaction to the new concepts is "buzzwords and bullshit."
More genteelly, participants note that most of the ongoing antitrust actions lack basic clarity and coherence. The Microsoft litigation started with the idea that the company was illegally tying sales of a Web browser to its operating system. Then, during the trial, the theory shifted to a charge that Microsoft wanted to divide the browser market with Netscape--a pretty silly idea, given the capacity of other companies to make browsers of their own. Now the government seems to be trying to prove that Microsoft is a meanie, which is not an offense under the Sherman Act. It is for the most part simply hard-edged jockeying over the terms of the partial integrations that are crucial to an efficient software industry.
The government also seems to think that high market share alone meets the legal definition of monopoly, even if there is only tenuous evidence of power to raise prices. Microsoft argues, of course, that it has a high share only because it keeps prices low and that this is what really bothers its competitors. They want Microsoft to be forced to set prices high enough to leave room for them.
The incoherence extends beyond the Microsoft case. The FTC accuses Intel of withholding information about forthcoming chips from companies that sued it for patent infringements. This is a novel extension of antitrust law. The relief sought would force Intel to treat all similarly situated customers alike, which would put the FTC in the middle of endless disputes over the meaning of "similarly situated" and "alike." The case looks like intervention in contract disputes combined with casual meddling in difficult intellectual property issues, plus an unarticulated assumption (also apparent in the Microsoft case) that any company with a large market share should be converted into a new breed of public utility by government fiat.
The DOJ's Visa/MasterCard case is also puzzling. Two separate violations are charged. First is the practice called "duality," whereby the two brands of card are issued by the same banks. The government contends that duality causes both Visa and MasterCard to go easy on introducing new products, such as smart cards, that might damage the other. The second alleged violation is that the Visa/MasterCard banks do not issue AmEx, Discover, or any other card.
These theories are confusing. If Visa and MasterCard are agreeing to slow down innovation, then AmEx, Discover, and other potential competitors should be thanking them for leaving open such a valuable market opportunity. Government intervention would be superfluous, and the last thing a competitor would want is for the government to wake Visa and MasterCard up.
As for the other theory, if duality is illegal, then why is the government trying to make it into quadrality? Or is the government saying that the banks must become the partners of anyone who wants to go into the credit card business? Again, this is a theory that converts any dominant company into a public utility.
All of these matters, and more besides, fan suspicions that government policy is blown by the winds of special pleading and political interest, and that the complex intellectual rationales are simply a new cover for old-fashioned rent seeking.
In a 1961 paper later published in Ayn Rand's Capitalism: The Unknown Ideal, Alan Greenspan wrote: "The entire structure of the antitrust statutes in this country is a jumble of economic irrationality and ignorance. It is the product: (a) of a gross misinterpretation of history; and (b) of rather naive, and certainly unrealistic, economic theories." His analysis is still on target. Building a new structure of "buzzwords and bullshit" atop an old one of irrationality and ignorance will not fix the problem.
In the end, the private interests that are so eager to foster this activism will regret it. As more than a few princes of Renaissance Italy could testify, once you bring in the condottieri you have a problem. Before long, you don't own them; they own you. The princes of Silicon Valley will soon have cause to reflect on this lesson of history, as they plead with ignorant but arrogant lawyers for permission to make deals or enter partnerships.
As T.J. Rodgers of Cypress Semiconductor recently warned his fellow tycoons in a New York Times op-ed piece: "Winning by politics is antithetical to the free-market competition that underpins Silicon Valley's success. The Justice Department isn't just attacking Microsoft; it's attacking the way Microsoft does business--and, by extension, the way most successful high-technology companies do business."
Rodgers closed by pleading with his colleagues to go back before it's too late. It is possible they will see their own interest and bring their political weight to bear on the side of free markets. If not, then the weary witness to the predictive power of public choice theory, watching the government re-create the good old days of antitrust disasters, will be left with nothing but the consolations of schadenfreude.
Contributing Editor James V. DeLong is an adjunct scholar with the Competitive Enterprise Institute and the author of Property Matters: How Property Rights Are Under Assault--And Why You Should Care (The Free Press). His Web site is www.regpolicy.com.
--------------------------------------------------------------------------------
How to subscribe
Return to the Reason Magazine home page 칼럼】美 MS社를 위한 변명
1999.06.28
--------------------------------------------------------------------------------
美 법무부와 19개 주정부들이 마이크로소프트(이하 MS)社에 대한 반독점 소송을 제기한 지 9개월이 흘렀다. 그 9개월 동안 MS社는 업계 내부와 여론의 비판으로 만신창이가 됐지만 美 정부가 증명해 낸 것이라곤 MS社가 '정보 고속도로 산업'에서 도태되지 않기 위해 신시장 확보에 필사적으로 달려들었다는 사실 뿐이었다.
MS社의 가장 큰 혐의는 자사 운영 프로그램 윈도 98과 인터넷 검색 프로그램인 MS 익스플로러(Explorer)'를 패키지화 시켜 다른 경쟁업체들의 시장 진입을 사전에 봉쇄했다는 것이다. 그러나 지금도 수많은 소프트웨어 개발업체들이 신생 운영시스템 리눅스(Linux)용 프로그램을 개발하고 있고 MS社 역시 독점 상황을 이용해 윈도 98의 가격을 올린 적도 없다. 10년 전 전세계 운영프로그램 시장을 장악했던 MS-DOS의 가격은 수백 달러를 호가한 바 있다.
美 법무부는 또한 MS社가 컴퓨터업체들에게 싼 가격으로 윈도 98 패키지를 팔아 고객들이 다른 MS社 제품을 사용하도록 종용, 유통상의 이익을 극대화시켰다고 주장했다. 그러나 MS社의 전략은 자사 제품의 판매 촉진을 위해 시리얼 업체들이 수퍼마켓 주인들에게 웃돈을 주고 계산대 근처에 자사 제품 진열대를 마련하게 하는 '합법적이면서 고전적인 판촉전략'에 불과했다.
MS社가 컴퓨터업체들로 하여금 기능상 별다른 진보가 없는 윈도 98을 설치하도록 했다는 주장도 듣기 거북하기는 하지만 불법적인 일은 아니었음에 틀림없다.
MS社사 윈도 95, 98과 인터넷 브라우저를 패키지화한 것은 분명 경쟁상품인 넷스케이프(Netscape)의 시장점유율을 낮추고자 하는 주요 전략이었다. 그러나 그것은 질레트(Gillette)社가 기존의 두 날 면도기를 세 날 면도기로 업그레이드한 것과 하등 다를 바가 없는 기술보완으로 봐야 할 것이다. 똑같은 이유에서 MS社는 컴퓨터 아트에서부터 하드 드라이브 청소기에 이르는 각종 소프트웨어를 윈도 98에 포함시켰던 것이다.
MS社의 전략이 넷스케이프의 시장 진출을 원천적으로 가로막았다는 주장도 사실과 다르다. 넷스케이프는 지금도 널리 유통되고 있을 뿐만 아니라 부지런히 업그레이드되고 있으며 몇 달 전에는 100억 달러에 거래되기까지 했다.
현재 MS社의 윈도 프로그램은 다방면에서 경쟁업체들의 공격을 받고 있다. 강력한 도전자 리눅스가 그 세력을 급속도로 확대하고 있을 뿐만 아니라 윈도 프로그램을 탑재하지 않은 각종 휴대용 통신기기들이 속출하고 있다. 특히 소니(Sony)의 차세대 통신기기 플레이스테이션(Playstation)Ⅱ와 같은 기기는 윈도 98을 운영시스템으로 사용하지 않고도 전자우편은 물론 인터넷 검색 또한 인텔 칩보다 3배 빠른 속도로 수행할 수 있다. 한 마디로 수많은 기업들이 윈도를 대체하기 위해 엄청난 돈을 투자하고 있다는 말이다.
美 법무부는 70년대를 IBM의 독점을 막는데 허송세월했다. 당시 법무부는 사생결단의 자세로 소송에 달려들었다가 IBM의 심각한 경영난에 봉착, 회생의 몸부림을 치기 시작하자 결국 82년에 소송을 취하했다.
지난 9개월 동안에도 법무부는 MS社가 경쟁자들에게 상당히 비열한 회사였음은 증명했지만 소비자들의 권익을 침해하면서까지 반독점법을 위배했다는 사실은 입증하지 못했다. 美 정부는 반독점 소송을 접고 더이상 정보기술산업의 세세한 부분까지 규제하려는 어리석은 노력을 반복하지 말아야 할 것이다.
기고: 로널드 카스(Ronald A.Cass) 보스턴 로스쿨 교수이자 MS社 고문 변호사
저널】반독점 혐의에 대한 마이크로소프트의 항변
1999.01.31
--------------------------------------------------------------------------------
빌 게이츠 마이크로소프트(Microsoft)社 회장은 지난주 '마서 스튜어트 리빙(Martha Stewart Living)' TV 프로그램에 출연, 소프트웨어를 사용해 알파벳을 배우는 자신의 두 살배기 딸에 대한 이야기를 늘어놓으며 미국 어린이들에게 컴퓨터 세계에 빠져볼 것을 권유했다. 빌 게이츠 회장의 이날 TV 출연은 MS社에 대한 반독점 혐의에 맞서 MS社를 홍보키위한 의도에서 이뤄진 것으로 해석되고 있다. 빌 게이츠 회장은 컴퓨터 산업과 그 영역에서의 MS社 입지에 관해 독점 당국과는 완전히 다른 시각을 갖고 있기 때문이다.
우선, MS社는 스스로를 막강한 독점기업으로 여기기 보다는 단지 PC 사용자들에게 기술혁신과 소비자 만족을 제공키위해 노력하는 하나의 기업으로 바라보고 있고 아울러 모든면에서 경쟁업체들의 위협을 느끼고 있다. 이는 정부의 입장과는 상반대는 것이다. 따라서 반독점 소송의 결과는 MS社가 자신들과 같은 시각으로 세계를 바라보도록 법정을 설득할 수 있을 지의 여부에 따라 달라질 것으로 보인다.
현재 세계 PC 운영소프트웨어의 90%를 장악하고 있는 MS社가 스스로 독점기업이 아니라고 주장하는 근거는 과연 무엇일까? 무엇보다도 MS社가 인텔, 선 마이크로시스템, 오라클, AT&T 등을 포함한 거의 대다수 하이테크 업체들을 경쟁업체로 여기고 있다는 점이다.
MS社는 특히 이 기업들이 최근 급속도로 제휴를 추진해나가고 있는데 대해 상당한 위협을 느끼고 있는 상황이다. 아메리카 온라인은 넷스케이프를 앳 홈은 익사이트를 그리고 루슨트는 어센드 커뮤니케이션을 각각 인수했다. 이 거래들은 모두 MS社에 대한 反독점 재판 이후 성사된 것들이다. 이와관련, 윌리엄 뉴콤 MS社 법률고문은 "이 산업은 상당히 불안정하다"며 "상황이 얼마나 빠르게 바뀌는 지를 파악하는 것이 무엇보다 중요하다"고 밝혔다. 물론 이들 M&A 거래중 MS社의 윈도우 시장에 직접적인 위협을 가하는 것은 한 건도 없다. 하지만 그렇다고 해서 MS社의 걱정이 덜어지는 것은 아닌 것 같다.
MS社는 또한 MS社의 독점적 관행으로 소비자들이 피해를 입었다는 정부의 주장역시 도저히 납득할 수 없다며 이에 대한 반론으로 PC 소프트웨어가 꾸준히 향상되어왔다는 점을 제기하고 있다. 지난주 리차드 쉬말렌스 경제학자는 MS社의 증인으로 법정에 출석, MS社가 만약 독점을 했다면 MS社의 윈도우 소프트웨어 가격은 지금의 16배가 됐을 것이라고 증언했다. MS社는 특히 정부측 경제학자인 프랭클린 피셔가 소비자들이 인터넷 브라우저 시장에서 MS社의 행동으로 지금당장 피해를 입은 것은 아니라고 증언한 사실을 강조하고 있다. 피셔는 소비자들이 앞으로 피해를 입겠지만 아직 그 상황이 발생한 것은 아니라고 증언했었다.
MS社의 반론중 가장 흥미로운 점은 향후 PC가 윈도우와 같은 소프트웨어에서 운영되지 않을 수 있다는 것이다. 즉 MS社는 가까운 장래에 컴퓨터들이 무료로 공개된 소프트웨어에서 운영되거나 또는 인터넷을 기반으로 워드 프로세싱이나 E-메일 등과 같은 응용프로그램을 실행할 것으로 전망하고 있다. 더이상 윈도우가 필요없는 세상이 올 수도 있다는 것이다. 한마디로 MS社 관점에서 현재 MS社의 시장내 입지는 단지 하나의 패러다임 변화에 불과하다는 것이다.
업계 전문가들은 과연 이것이 가능할 지 그리고 얼마나 빨리 진행될지에 대해 열띤 논쟁을 벌이고 있다. 그러나 법률 전문가들은 MS社의 이같은 주장이 법정에서 수용하기에는 무리한 것일 수 있다고 지적하고 있다. 윌리엄 코바식 조지 워싱턴大 법학교수는 "법정에 아직 일어나지도 않은 상황을 진지하게 받아들이도록 요구하는 것은 어렵다"고 설명했다.
세계에서 가장 성공한 기업 MS社가 직접 자신의 쇠락을 예측하는 것이 다소 이상하게 들리기는 하지만 사실 월街에서는 어느정도 익숙한 이야기이다. MS社는 월街의 애널리스트들에게 늘상 전망이 암울하다고 말해놓고서는 높은 순익을 발표해 애널리스트들을 놀라게하는 것으로 유명하다. MS社는 지난주에도 작년 4/4분기 75% 증가했다고 발표한 바 있다. 투자자들은 이제 오랜 경험을 통해 MS社의 엄살섞인 전망을 액면 그대로 받아들이지는 않고 있다. 그러나 지금 MS社는 법정에 정말로 향후 상황이 암울할 수 있다는 점을 설득시켜야만 한다.
'형설지공 > 경제경영' 카테고리의 다른 글
| MS 독점 찬반론으로 분류하기 어려운 글 모음 (0) | 2001.03.08 |
|---|---|
| 미 법무부의 MS 독점금지법 규제가 타당하다는 입장의 글 모음 (0) | 2001.03.08 |
| Global Capitalism: Can it be made to work better? (English-K (0) | 2001.03.08 |
| 한국외환위기당시 외국신문보도 자료 (한글) (0) | 2001.03.08 |
| 한국외환위기당시 외국신문보도 자료 II (영문) (0) | 2001.03.08 |