The Importance Of Antitrust Enforcement In The New Economy

I'm delighted to be here tonight to talk with you about antitrust enforcement from the perspective of the Department of Justice. These are exciting times at the Antitrust Division as well as in the American economy, and preparing these remarks has afforded me the opportunity to take stock of where we are in antitrust and where we're heading. Some of you may recall that, when I first took over as head of the Division, I announced a three part agenda -- reduced to a soundbite, I termed it, "developing legal doctrine and practices for an age of globalization, deregulation, and technological change." In the past year, we've been doing just that and there is a lot to report on all three fronts. After briefly reviewing what's been going on generally, I will then spend the bulk of my time talking about the third of these challenges -- the role of antitrust enforcement in our high-tech, fast-moving, economy.

I've been interested in this last issue for some time and, in the wake of our consent decree enforcement action against Microsoft, many others appear to have become interested in it as well. While several questions have been raised, there are two points that I want to focus on this evening: first, whether our industrial-age antitrust laws need to be rewritten to cope with today's information-based economy; and second, whether there is any role at all for antitrust enforcement, given the fast-moving, innovative nature of these new markets.

1. DEVELOPING LEGAL DOCTRINE AND PRACTICES TO ADDRESS GLOBALIZATION, DEREGULATION, AND CHANGING TECHNOLOGIES
1. GLOBALIZATION

As we anticipated, developments on the international front have been moving extremely rapidly. In the past half dozen years, the portion of the Antitrust Division's cases having an international dimension has increased from less than 5% to more than 30%.

Most notably, we are now engaged in an extensive and very sophisticated international cartel-enforcement effort, building on our significant accomplishments in the Archer Daniels Midland case. Today, we have more than thirty grand juries throughout the nation, looking at cartels involving companies in more than twenty different countries, some in industries that do over a billion dollars of annual commerce in the U.S. alone. In our last fiscal year (ending on September 30, 1997) we brought in more than $200 million in criminal fines -- five times greater than our previous high -- and I expect that we will soon see guilty pleas or prosecutions in some additional, high-visibility cases. Based on our efforts in this area, moreover, the Organsation for Economic Co-operation & Development (OECD) is about to adopt a proposal, supported by all the major industrial countries, endorsing cooperation among competition authorities with respect to cartel enforcement. I view this as a major step forward, one that will help us expand our efforts to get evidence and reach witnesses throughout the world. The crimes we are pursuing are global and, to be truly effective, our territorial reach must be commensurate.

In addition, we're finding that our enforcement agenda increasingly includes mergers that are also being considered by foreign competition authorities -- for example, right now we and DG IV of the European Union are both reviewing the American Airlines/British Airlines alliance and the two Big-6 accounting-firm mergers; and indications are that the Japanese Fair Trade Commission is likely to seek to extend its jurisdiction to cover these kinds of multi-national mergers as well.

In mentioning this matter, of course, I am immediately reminded of the significant problems that arose last summer when the U.S. Federal Trade Commission and the DG IV reached very different conclusions with respect to the Boeing/McDonnell-Douglas merger. While that kind of sharp divergence is unique in our experience, we still must explore ways to avoid any recurrence and, to that end, we and the FTC have been working closely with DG IV. Given the understandable concerns about national sovereignty, navigating these waters -- along with other issues raised by multi-jurisdictional merger review -- will not be easy.

And, finally on the international front, the issues at the intersection of trade and competition are becoming increasingly important and, frankly, potentially the most difficult to solve. Impairment of market access by private business restraints raises significant concerns that are relevant both to trade and to competition policy. Some of these issues, as you know, are swirling around in the shadows of Kodak's trade case in the World Trade Organization (WTO), which specifically involves charges of governmental barriers that have blocked access to the Japanese film and photographic paper markets. At the same time, the WTO has decided to look directly at non-governmental, private markets restraints -- in a very preliminary way, I should add; and I should also make clear that, at least as I see it, these efforts are unlikely to lead in the near future to some form of international dispute resolution, although I suspect that some countries may seek such action.

For our part, we have been pressing ahead with the concept of positive comity -- where one enforcement authority refers a market-access issue to the agency whose market is most directly affected by the denial of access. We will soon execute a detailed agreement with the Europeans, outlining a formal protocol for such cases and, in the meantime, we have referred our first positive comity case to them -- a case involving allegations that several European airlines engaged in anticompetitive behavior designed to thwart competition by Sabre, the computer reservations system affiliated with American Airlines. Positive comity is a very encouraging development and certainly represents a sound conceptual approach. The current problem, which I anticipate will lessen over time, is that many competition authorities are either insufficiently independent or otherwise too weak to bring market access cases that might benefit foreign competitors.

As even this brief summary makes clear, these international issues are as complex as they are important. While I think we have articulated a clear and sensible policy to deal with them -- for now, relying principally on bilateral cooperation agreements with our counterparts, coupled with some multilateral efforts at the OECD and some very preliminary discussions at the WTO -- I think we need to develop a long-term vision as well. To that end, I have recently established a twelve-member Federal Advisory Committee, which I have charged with taking an independent look at these matters and preparing a proposed blue-print for the Division. The group, called the International Competition Policy Advisory Committee, is chaired by two well-known players in the area of trade and competition, Jim Rill and Paula Stern, and has Merit Janow, a distinguished professor from Columbia University, as its Executive Director. The other ten members are outstanding leaders from business, labor, government, and academia. Their first meeting will be February 26 and I expect that their deliberations will continue over the next couple of years. The Committee's meetings will be open to the public and we welcome participation from the bar. I look forward to very significant contributions from this group.

2. DEREGULATION

During the past twenty-five years, the Antitrust Division has played an important role in facilitating competition and assisting deregulatory efforts in a number of monopoly industries, including telecommunications and electric power. In recognition of that background, and given our experience in the AT&T litigation, Congress called on us to assist in the implementation of the Telecommunications Act of 1996. I have said a great deal about what's been going on in this area in published speeches and congressional testimony, so rather than reiterate the Division's views here, I would simply refer you to our web-site for the details.

In the meantime, in order to eliminate the suspense before you get the opportunity to go on line, let me at least summarize our position. Contrary to the views of the skeptics, including many in the popular press, I believe that, in general, the process is moving forward constructively and that, if we stay the course, the Telecom Act will bring real benefits to America's consumers. Despite the overly-optimistic rhetoric sounded by industry participants during the debates leading up to the passage of the Act, it should come as no surprise to sophisticated observers that long-standing monopoly markets take time to open, especially when widespread competition depends, at least for the near-term, on the shared use of an incumbent monopolist's facilities. Lots of money is at issue here, and shared facilities raise complex issues of pricing, access, and ensuring affordable universal service. As a result, we're seeing at lot of litigation by the affected business interests.

But what's less visible is that market forces are moving forward -- in particular, the incumbent monopolists, while not yet losing a lot of customers, are losing some of their most lucrative customers. And those forces, in my opinion, will lead to significant, increased competition and innovation in the next few years, as the Regional Bell Operating Companies (RBOCs) come to conclude that it makes more sense to open their markets fully, so that they can then get into the long-distance market, rather than to litigate, which often leads to increased uncertainty and delay. I should also mention that, in addition to implementing the Telecom Act, the Division is now actively involved in considering possible federal legislation aimed at deregulating the electricity industry. As an historical matter, these are remarkable developments. Two industries, long thought to be natural monopolies, are now undergoing significant structural change, which will lead to real competition over time.

By the way, one of the more interesting issues to have arisen as we have engaged in these deregulatory efforts is how to think about the relationship between merger policy and deregulation. I gave a speech last week before the Federal Energy Regulatory Commission (FERC) in which I raised some of my concerns in this regard -- asking whether, for example, Congress should consider shifting the burden of proof for certain narrowly defined categories of electricity mergers during the early years of restructuring. I hope you will look at those remarks and let us know what you think. The issues are important and I will candidly acknowledge that the solutions aren't easy.

3. TECHNOLOGICAL CHANGES

As I mentioned at the outset, I plan to spend most of my time this evening on the role of antitrust in high-tech industries, but let me first make a few brief comments generally about what's going on in terms of antitrust doctrine and the use of litigation at the Division. A big part of our agenda has been to focus on, and clearly set forth our views with respect to, important doctrinal issues in antitrust enforcement, such as the application of the unilateral effects, coordinated effects, and network effects doctrines. We do this through a variety of vehicles: in speeches, in formal guidelines, in competitive impact statements, and in court filings.

And, as I have also made clear, we are especially interested in pursuing new doctrinal issues by re-engaging the federal courts, which, for the past two decades, have experienced relatively few cases (save in the criminal area) being litigated by the Antitrust Division. Currently, we have four civil non-merger cases in active litigation -- surely the largest number in several decades -- and, in the past six months, we filed three contested merger challenges, also something of a modern-day record. Several of these cases raise important issues relating to high-tech industries -- Microsoft, obviously, but also our case against General Electric, which involves limitations on intellectual property licenses. These are encouraging developments and I believe that they reflect a renewed vitality in the antitrust field.

2. GOING FORWARD -- ANTITRUST ENFORCEMENT IN A HIGH-TECH, INFORMATION-BASED ECONOMY

Although, as I've just been describing, the implementation of our three-part agenda has been robust, and I believe effective, the truth is that interest in these matters has largely been limited to antitrust lawyers and affected businesses. That all changed, of course, when the media began reporting on the Microsoft case. At its core, the public debate that ensued has raised the fundamental question whether there is a role for antitrust enforcement in our current economy. I have been wanting to talk about this issue for some time and figured that if I were to ask the question about relevance here tonight -- and then answered it in the affirmative -- I would receive a warm reception, if for no other reason than that most of you make your living in this field and so I assume that you'd like to be reassured that we have a future.

Well, I believe we do. To be sure, as I will explain, the role for antitrust in high-tech industries is likely to be modest in scope and surgical in application. But, in my view, that hardly makes it unimportant. Quite to the contrary, the economic qualities that tend to characterize market behavior in high-tech industries are such that we will almost certainly see companies come to enjoy very significant market power, which in turn is likely to lead to antitrust scrutiny (though, I should be careful to note, not necessarily antitrust condemnation). More generally, as I will explain in a moment, there is nothing so different about these new technology-based markets that could possibly support abandoning this Nation's longstanding belief -- a belief based on lots of experience -- that competitive markets work best for consumers and antitrust enforcement is essential for sustaining competitive markets.

1. THE U.S. ECONOMY TODAY

Whatever else we may agree or disagree about, we certainly can find common ground by starting with the fact that the U.S. economy is remarkably strong today -- much stronger, I would suggest, than most of us would have predicted a half-dozen years ago. Unemployment has been below five percent for close to a year and below six percent for three years; our growth rate has been almost three percent over the past six years; and, given these two numbers, perhaps the most remarkable fact is that inflation is essentially non- existent. This wasn't suppose to happen. In fact, the economists told us that it couldn't happen. Under various models, first referred to as the Phillips curve and later refined to the Non-Accelerating Inflation Rate of Unemployment, we were led to believe that, at these levels of growth and unemployment, we would have serious inflation. Why haven't the economists' predictions been borne out?

Several factors doubtless help to explain this remarkable set of economic developments, including President Clinton's leadership and the fine work of Secretary Rubin and Chairman Greenspan. But there is one other factor that is sometimes ignored but should also be noted here, and that is that our economy is more competitive today than it has been in a long, long time. Over the past several decades, we have experienced a steadily increasingly national commitment to competitive markets and away from a regulatory approach. Whether it is airlines, surface transportation, energy, or telecommunications, our faith in competition has triumphed time and again. And, to the same effect, we have repeatedly opened our markets to foreign firms, often leading to renewed competitive vigor in industries that were previously characterized by oligopolistic lethargy -- the automobile industry perhaps being the most obvious example. It is no mere coincidence, I would suggest, that at a time when our economy is the most competitive economy in the world, it is also the strongest.

Indeed, when thinking about the relationship between competition policy and the current strength of our economy, it's especially instructive to compare our policies with those of our major trading partners, some of whom -- like the Japanese and Koreans, for example -- are now experiencing rough times. These other economies have been characterized much more by regulatory, sometimes even cartel-like, arrangements, embodied in notions like rationalization cartels and the promotion of a single "national champion" within a given industry so that it could best compete in the international arena; and some of these countries have also experienced structural or even cultural barriers to foreign competition, such as the keiretsu arrangements that continue to dominate significant parts of the Japanese economy, for example.

The effects of these structural differences in the major national economies were thoroughly analyzed by Michael Porter, a distinguished economist at the Harvard Business School, in his landmark book, The Competitive Advantage of Nations. Almost a decade ago, Porter concluded -- again, before it was fashionable to say so -- that "active domestic rivalry is strongly associated with international success," whereas "creating a dominant competitor rarely results in international competitive advantage." In light of recent economic developments, I think it's fair to say that Porter was right and that the United States is very fortunate to have embraced competition policy with as much vigor as we did.

2. ANTITRUST ENFORCEMENT

While Porter's analysis is generally accepted today, it is less clear what antitrust has to do with all this. Porter had little doubt about that question as well. In a part of his work that has received somewhat less attention, he concludes that "a strong antitrust policy . . . is essential to the role of upgrading any economy."

But why should that be? The answer, I believe, is that, contrary to the view in some quarters, the natural state of markets is not to move towards increasing competition. Market power, rather than a competitive market, is something that every business understandably wants -- because it allows a business to increase its profitability at the expense of the consuming public. And, if the antitrust enforcers closed shop, I have little doubt that the competitive structure of our economy would erode significantly -- mergers and other agreements to achieve market power would occur in a heartbeat; and market power that had been legitimately achieved, through the development of intellectual property, for example, quickly would be used to extend or protect a monopoly position.

While often forgotten, none of this is new. If you go all the way back to Adam Smith's seminal work, The Wealth of Nations, you will see that, despite his pro-market, laissez-faire take on the economy, he fully recognized that the government has a crucial role to play in assuring that businesses do not attempt to end-run the competitive process. More recently -- now we're only talking sixty years ago -- Thurman Arnold, one of the great antitrust thinkers in this Nation's history, as well as the head of the Antitrust Division from 1939-43, explained, "[t]he maintenance of a free market is as much a matter of constant policing as the flow of traffic on a busy intersection. It does not stay orderly by trusting to the good intentions of the drivers or by preaching to them. It is a simple problem of policing, but a continuous one."

Arnold was also fond of another metaphor, one that I find especially congenial for today's economy: "The competitive struggle without effective antitrust enforcement," he wrote, "is like a fight without a referee." I like this way of putting it because it highlights what I consider to be the two essential points about effective antitrust enforcement -- first, that it should not be abandoned; and second, that it should not be overdone. In other words, while it's true that you need a referee in a sporting match, it's equally important to understand that the referee's role must be appropriately circumscribed. If you don't let the players play -- or, in market terms, if you try to over-regulate the competitive process -- you can ruin the game. Markets are rough places and, though competition is not always pretty, allowing it to flourish is ultimately in our best interest. As Arnold put it, "[t]he economic philosophy behind the antitrust laws is a tough philosophy. [Those laws] recognize that competition means someone may go bankrupt. They do not contemplate a game in which everyone who plays can win."

To elaborate on this last point, let me be clear in saying that I believe a great mistake is made when the antitrust laws are used to protect competitors rather than competition, as has occurred too often in our history. It may make sense to assign handicaps in a golf game or to require certain horses carry weights in their saddle-bags during a race, but that kind of handicapping is not appropriate in the market. In keeping a watchful eye on the marketplace, we are concerned with consumers, not competitors, and even if it's boring to see the same person win over and over again, as long as those victories are based on economic efficiency, it will be good for consumers and the antitrust enforcers ought to stay out of the way.

In the same vein, it's important to emphasize that big is not necessarily bad when it comes to antitrust enforcement. Bigness can lead to efficiency -- though a synergistic merger, for example -- which in turn is good for consumers. Arnold emphasized this point as well, having explained that "it is as meaningless to say that small [business] units are better than big units as it is to say that small buildings are better than big ones." And, he added, if the antitrust laws were to become "simply a religion which condemns largeness as economic sin," they would soon be "an anachronism." Nevertheless, even today, there are those who would fault us for not doing enough to block large aggregations of economic capital, pure and simple, even when they are economically efficient.

I want to assure you that we will resist any such temptation. These kinds of notions -- antitrust ought to help weaker competitors, or big is bad -- simply have no place in a sensible enforcement program. We didn't challenge the Bell Atlantic/NYNEX merger -- even though it was one of the largest mergers in our Nation's history -- because we concluded that, while it was a difficult case, on balance the merger was likely to benefit consumers in that the resulting efficiencies would lead to improved services.

But, as I said earlier, just as using antitrust law to implement social policy is a mistake, so too is a religious faith in self-correcting markets. There is a need for antitrust enforcement to aid the free market and, at its legitimate core, such a role focuses on assuring that market power doesn't restrain competition that consumers would otherwise enjoy. And a properly focused concern about market power, in turn, requires surgical intervention precisely because businesses benefit from efficiency and market power alike, whereas consumers benefit from the former but not the latter. So our job is to make sure that we take out the fat (market power) without taking out the muscle (efficiency).

3. THE SHERMAN AND THE CLAYTON ACTS

All of that history is well and good, some say, but they then go on to question whether the existing antitrust laws can possibly be relevant to today's economy. The Sherman Act was passed in 1890 in response to the nationwide industrial trusts that the railroads had made possible, and the Clayton Act was passed in 1914 and was aimed largely at retailing and wholesaling practices in localized markets. How, then, can these ancient statutes be relevant to a 21st Century, information-based, economy? I get asked that question, especially by non-antitrust-lawyers, probably more than any other. And I answer, unhesitatingly, that the laws are just fine, precisely because, unlike most contemporary statutes, they are common-law provisions and, therefore, they are not locked in text or time. People today don't fully appreciate the genius of the common law, but I believe that we in the antitrust field are fortunate to be a part of this declining heritage.

To take an analogy that I find apt, the "freedom-of-speech-and-press" clauses of the First Amendment to our Constitution were enacted at a time when speech and print were the only two media. Today, of course, we have radio, t.v., cable, satellite, the internet, etc. Yet no one really thinks that we need a new First Amendment. The core principles of that provision, developed through over two centuries of case law, have been effectively and sensibly applied to these new media, just as they were once applied in a world without them. In this fundamental respect, as Chief Justice Charles Evans Hughes (among many others) correctly recognized, "[a]s a charter of freedom, the antitrust laws have a generality and adaptability comparable to that found to be desirable in constitutional provisions."

When it comes to the antitrust laws, the core principle, as I just mentioned, is to prevent agreements or mergers that create or increase market power, or unilateral actions that use existing market power to protect or expand a monopoly. As you know, of course, that's what the three key statutory provisions actually do: section 1 of the Sherman Act bars anticompetitive agreements, section 7 of the Clayton Act bars anticompetitive mergers, and section 2 of the Sherman Act prohibits the abuse of monopoly power. In combination, these provisions are fully adequate to deal with the contemporary economy. Indeed, as I will now show, many of the so-called "new" economic issues really aren't so new to antitrust enforcement after all. As with the First Amendment, we have a venerable body of case law -- not every one rightly decided, but as a corpus rich in detail and complexity -- that we can and will draw on as we work through antitrust issues in the new economy.

4. ANTITRUST ENFORCEMENT IN THE NEW ECONOMY

Let me then turn to the software industry, which is obviously paradigmatic of the new economy, and discuss some of the issues currently on our agenda. As I have already suggested, the core principle about market power that animates our concern here should be no different from what it has been since the inception of the antitrust laws: competition is good and market power can undermine it. To be sure, in analyzing market power issues, we must take cognizance of any differences that might characterize the specific market under consideration. So, for example, in high tech markets, we need to determine whether the market is likely to experience a tipping point as a result of so-called network effects, or whether it is likely to be marked by price competition, innovation competition, or both. (For those who are not familiar with the latest jargon, the concept of "network effects" refers to those markets in which particular services or products become more valuable as more people use the network, be it a telephone network or electronic mail.)

Although these potential new wrinkles in high-tech markets raise important questions, I want to make clear that they are hardly so novel -- and certainly not nearly as intractable -- as some have suggested. Indeed, in 1936, the Antitrust Division won a tying case against IBM, involving tabulating machines and cards, in what was surely then considered a "new" industry. And some forty years ago, we brought a case involving network effects in the floral-delivery market and secured relief that made competing networks possible.

By the same token, while price competition has generally been the paramount focus of antitrust enforcement, innovation competition, which appears to be very important in the new economy, is no stranger to our field either. On the contrary, if you go back and study some of the earliest monopoly cases, such as Alcoa and Kodak, you'll see that the courts were concerned with technology innovation and suppression as well as with price competition. And almost thirty years ago, the Justice Department brought what was probably the first pure innovation case, concerning a horizontal agreement among auto makers not to develop certain pollution technologies. Soon thereafter, innovation issues were a key part of the resolution of the AT&T case, as reflected in the equipment- manufacturing provisions of the judgment. And today, of course, cases involving innovation (or R&D) markets are quite common -- such as, for example, the FTC's recent action in the Ciba Geigy/Sandoz merger and our action in the General Motors/ZF Friedrichshafen merger.

Nor is there any reason to think that our customary concerns about price competition are somehow irrelevant in high-tech industries. In high-tech industries, as in others, market power often leads to price increases. For instance, just recently, Business Week reported that "[n]ow that [Microsoft's] Office [software] has 87% of the suite market and thousands of businesses rely on it, the cost of a corporate license in most cases is headed up." On the other hand, some commentators have argued that it makes no sense for us to be challenging a practice like Microsoft's tying of its browser to its operating system when it gives away the browser for free. But if the products are tied together and sold at a single price, how can anyone say that the browser is free? More importantly, even assuming Microsoft gives away its browser, that is, of course, a traditional, long-standing antitrust concern -- though not necessarily a violation -- because free is a curious price. After all, Microsoft spends a lot of money developing and marketing its browser. So why would it give it away for free?

There are two potential reasons -- one legitimate, one not. The former is that revenues from an ancillary or future stream of commerce make a for-free strategy economically rational. For example, a newspaper may be given away in order to build up a large base of subscribers, which, in turn, will attract advertisers that pay enough to justify a for-free subscription price. It is also possible, however, as many cases have found, that a no-cost product is one intended to protect or establish monopoly power and, if that's what's going on, then the strategy is predatory and it violates the antitrust laws. This kind of fact- based question about predation is as old as the antitrust laws themselves.

What all of this proves, I believe, is that the issues raised by antitrust enforcement in high-tech industries are not nearly so new as some may think. Ironically, perhaps the most novel of the phenomena that tend to characterize the software industry in particular -- i.e., the strong presence of network effects -- would appear to warrant increased antitrust concern over certain kinds of monopolistic practices, because network effects can make it especially difficult for a new entrant to penetrate the market.

5. THE DEPARTMENT'S RECENT ACTION AGAINST MICROSOFT

Moving beyond some of the general issues raised by applying the antitrust laws to a high-tech industry like software, I now want to address the specifics of the Microsoft case itself. Let me start with the fact that it is clear to us and, I believe, generally agreed by most observers, that Microsoft currently has a monopoly in personal computer operating systems. Operating systems are the kind of products that are characterized by network effects, and Microsoft has such a large installed base of customers that it is not going to be easy for a potential competitor to challenge its monopoly.

Now, beginning from this understanding of Microsoft's market position, let's turn to the issue of new products and how they are bundled or tied to the operating system. And let's think -- as the Antitrust Division must think when it makes policy -- not only about browsers but about other products as well -- for example, personal finance or electronic commerce software. I don't want to analyze all of those issues tonight but, as we move forward, I would like you to be thinking whether there are different competitive considerations relevant to each of these products or whether Microsoft should be allowed to bundle any and all of them with its operating system. I would also like you to think about whether such bundling, especially in a network industry, makes it harder and harder for a new entrant to challenge Microsoft's monopoly position in operating systems.

In the consent decree case that we brought, we identified two, related, anticompetitive effects that concerned us. First, there was a specific browser effect, as to which some people have asked, "what's the big deal, anyone who wants Netscape's browser can get it?" Even aside from the fact that bundling can distort allocative efficiency by forcing consumers to take a more expensive or lesser quality product, the basic premise of this argument can be very short-sighted, in that consumers who prefer the alternative product soon may find that it's no longer available. It's entirely possible, for example, that browsers, like operating systems, will end up tipping and that only one company's product will survive. Think back not so long ago to the days when there were both Betamax and VHS VCRs. Today, of course, consumers can only get VHS.

If you accept that the market for browsers is likely to tip, the initial competitive concern is that, by forcing computer manufacturers to take Microsoft's browser as a condition of getting its monopoly operating system, the market could tip to a product that consumers would not have preferred in the absence of the forced tie. Consider the following hypothetical (which I use simply to illustrate the issue but not to represent the facts of this case): suppose that half the consumers get their browser from the Original Equipment Manufacturer (OEM) channel and half through other distribution channels (such as by down-loading from the Internet). And suppose further that, in the non-OEM channels -- where there is no opportunity for monopoly tying -- browser A is preferred over browser B by 3 to 2, while in the OEM channel, due to bundling, browser B is preferred by 4 to 1. Overall, 60% of all users would get Browser B -- 20% from the non-OEM channel and 40% from the OEM channel -- and, we can then assume if the market were to tip, it would go to B. Under these circumstances, bundling would be decisive, tipping the browser market to browser B even though, without the bundling thumb on the scale, consumers would have preferred browser A by 3 to 2.

The implications of this hypothetical can be especially profound when you consider its impact on potential innovators writ large as well as across the spectrum of the many products that could be bundled with (or tied to) an operating system. And that leads me to the second -- and especially significant -- anticompetitive effect at issue here. The potential innovators that I am speaking about are people who must decide as an a priori matter whether they have a reasonable chance of prevailing when their opposition can achieve a significant advantage in the OEM channel by dint of bundling. The answer seems obvious. In fact, Irwin Stelzer, a conservative economist at the American Economic Institute, emphasized precisely this point in an article in the Weekly Standard titled "Why Janet Reno v. Bill Gates Is Good For Capitalism." Referring to the current Microsoft case specifically, Stelzer concludes, "[t]hese are markets in which we rely heavily on innovation to improve products and continue to drive down prices. If potential entrants and innovators are warned that any product they may develop will be copied . . . and then tied to a monopoly product, they will find something better to do with their energy, time and money. Result: a stifling of innovation."

Although that precise concern -- i.e., stifling innovation in this very dynamic market - - led us to bring our consent decree case, some have suggested that what is really at issue in the case is whether the government or the market should decide what goes into a computer. We disagree with that view. Rather, the competitive question at issue, as we see it, is whether a monopolist that controls the operating system -- and has the incentive and ability to harm competitors, even at the expense of consumers -- or a large number of competitors at the OEM level -- whose only incentive is to give consumers what they want -- should decide which browser goes into a computer. From our perspective, it is clear that consumers will be better off if OEMs decide. If that happens, consumers are likely to get more choices and, in the end, the better product is likely to prevail in the market.

As it turned out, moreover, the principal policy arguments raised against us were meritless in fact. First, Microsoft said the operating system and browser couldn't be separated. But now they have been, with Microsoft's acquiescence. Consequently, if OEMs want to load Netscape's Navigator instead of Microsoft's Explorer -- as our investigation determined some OEMs had in fact wanted to do -- now they can. On the other hand, Microsoft remains free to make its products on a bundled basis for any OEM that wants it that way. In short, in no way did we attempt to retard Microsoft's product development. Nor do we have any views about which browser any OEM should choose. All we have done is create the possibility for increased choice at the OEM level, a result that is good for consumers.

3. CONCLUSION

Let me wrap up by again quoting Irwin Stelzer who, in his piece on the Microsoft case, said that "[f]riends of the free market should not forget that the antitrust laws deserve their veneration because they keep government's role as a regulator to a bare minimum." I share that view. It is a view that I consider to be pro-market and, more importantly, pro- competition, which ultimately means pro-consumer. That's what we're all about at the Antitrust Division and I expect that a hundred years from now that's what we'll still be all about.