European and American Approaches to Antitrust Remedies and the Institutional Design of Regulation in Telecommunications

February 23, 2004

Unedited transcript prepared from a tape recording

9:45 a.m.	Registration
10:00	Presentation:	"European and American Approaches to Antitrust Remedies and the Institutional Design of Regulation in Telecommunications" forthcoming in Handbook of Telecommunications Economics v.2
	Speakers:	Damien Geradin, University of Liege and College of Europe
		J. Gregory Sidak, AEI
10:45	Discussants:	Abbott B. Lipsky Jr., Latham & Watkins
		David S. Sibley, Antitrust Division, U.S. Department of Justice
11:30	Question and Answer
Noon	Adjournment

Proceedings:
MR. SIDAK:  Ladies and gentlemen, could we get started?  Good morning and welcome to the American Enterprise Institute.  I'm Greg Sidak.  We have a panel today to talk about antitrust and regulatory issues.  We're addressing this problem from several different perspectives:  from the perspective of industry or sector-specific regulation versus general antitrust law; we're also looking at European versus American approaches to the design of regulation and competition law policy.

Our speakers today are, on my immediate right, Damien Geradin--you have biographies in your materials, but I'll give you some brief biographical highlights.  Damien is one of Europe's leading legal scholars on antitrust and network industries today.  He's professor of law at the University of Liege and also the founder of the Global Competition Law Institute at the College of Europe.

Next to him is David Sibley.  David is the alpha and the omega of economic analysis in this area.  He was formerly the head economist at Bell Labs, and now he's the head economist at the Antitrust Division, so he can tell us everything about both sector-specific telecom economics and general antitrust law principles from an economist's perspective.

Along the way, he has held an endowed professorship at the University of Texas, and he is the inventor of the economic theory that was at the heart of the government's Microsoft case, namely, that the browser could evolve over time into a substitute for the operating system.  So he had a breadth of experience in other network industries besides telecom.

And then on the far right of the panel here is Tad Lipsky, a partner--geographically speaking.  Tad is a partner at Latham & Watkins.  Before that, he was in charge of international antitrust compliance for Coca-Cola.  Early in his career, he was the deputy who was the right-hand man, so to speak, for Bill Baxter at the Antitrust Division at the time of the ending of the IBM monopolization case and the Bell divestiture, the beginning of the modification of final judgment.  So Tad brings a lot of firsthand insight as well.

I'd like to ensure that we have plenty of opportunity for questions and answers, so I think our plan here is to keep our prepared remarks fairly brief and then open it up to discussion as quickly as possible.

I'd like to ask all of you who have cell phones to please put them on silent mode or turn them off at this time so that we don't have that interruption.

Well, I'll start off.  You have a paper in your folders here that Damien and I have written.  This paper is a chapter for the forthcoming second volume of the handbook of telecommunications economics.  I was working on this chapter, and I realized that it's a chapter about remedies within telecom regulation and how regulatory remedies relate to antitrust remedies or competition law remedies, as we'd say in Europe.  And as I was working on this, I realized that Damien was studying exactly the same topics in Europe, and so it made sense for the two of us to join forces on this, with Damien providing the European perspective and I'm providing the American perspective.

Let me say at the outset that we wrote this draft of the chapter before the Supreme Court decided the Trinco case, and so Trinco is only mentioned in passing, and obviously the merits of the case are not discussed.  And that's something that we will revise before the chapter goes into print.

Damien may have a few words to say about Trinco, but I think that we'll probably get back to that case after all the panelists have spoken, and I would expect there would be some questions from the audience on that.

I will simply pose this following question to you about Trinco, and you can think about it between now and then.  What two words are conspicuously absent from the Trinco opinion?

All right.  So getting on to the question of remedies in telecommunications and whether to rely on sector-specific or general antitrust principles, what is a remedy to begin with?  And here I think there is a distinction between the way economists look at the issue and the way lawyers do.  Lawyers think of remedies as something--a remedy is something that is imposed by a court, usually after a finding of liability, or perhaps a finding of some potential violation of law.

An economist would look at a remedy as something that is imposed after there has been a finding of market failure, that maybe there's market power that is unchecked by some kind of regulatory process.  Maybe there's imperfect information.  Maybe there's an external cost that's not compensated.

So the economic and the legal concepts of a remedy are not necessarily coincident.  They're only going to coincide if lawyers frame the idea of a remedy in terms of something that is triggered when a market failure occurs.

What about the difference between sector-specific regulation and what I'll call generalized antitrust law?  I think that in broad terms, sector-specific regulation relies more on a forward-looking approach.  It's more of an ex ante imposition of rules on firms within an industry.  And I would contrast that with the approach of antitrust law, which tends to be more ex post.  Antitrust is something that is administered through the courts, and, therefore, it's the result of the adjudication of some actual case in controversy.  So by its definition, it's something that's retrospective in nature.

Now, obviously there are some antitrust statutes, like Section 7 of the Clayton Act, that talk about the potential lessening of competition, and that's something that's forward-looking.  But, still, the imposition of the remedy, the ordering of injunctive relief, for example, is something that follows an adjudicatory proceeding.

Ex ante regulation is different, though, in the sense that--we'll just take some examples.  Section 271 of the Telecom Act, which governs the entry of the Bell Companies into long distance, says you can't do this unless these following condition are satisfied, and there happen to be 14 items that are satisfied.  Another example would be the approval of a merger subject to a number of conditions that have a prospective application.

Now,, the line between ex post antitrust remedies and ex ante regulatory remedies is, in fact, not as clear as I've just portrayed it, because if you look at the way the consent decree process works in practice, we have had a number of examples of antitrust laws that mutate into a kind of industry-specific or even firm-specific regulation.  The biggest example, of course, is the modification of final judgment.  The court decree that ended the government's antitrust case against the former AT&T, the Bell System, it was a case alleging monopolization, and as part of the settlement, there was wide-sweeping restructuring of the industry with prospective obligations and restrictions placed principally on the operating companies.

The Microsoft decree, I would argue, also has had a kind of prospective application.  So this line between ex ante regulation and ex post antitrust I think has blurred a great deal.  So that, I think, invites a question:  Well, which way is the flow of ideas taking place here?  Are regulatory agencies and the kind of analysis that they do influencing the development of antitrust law?  Or is it antitrust law that's developing this--influencing the development of regulation?

I would argue that 20, 25 years ago, at the time that the AT&T break-up occurred, it was antitrust law influencing industry-specific regulation.  And you can look beyond telecom to see the effects of that.  I think what the Interstate Commerce Commission was doing in the early '80s with deregulation of railroads, the captive shipper cases, explicitly used antitrust concepts of market definition and consumer substitutability to pare back the scope of rate regulation.

I would say that certainly in the last ten years, the flow, the intellectual flow has been in the opposite direction.  It's been much more a case that antitrust has both--as it is practiced, has turned more into a kind of administrative law practice, more guidelines, less of an emphasis on district court litigation, but also substantively, I think the content of antitrust law has become heavily influenced by the economic research that has been done on network effects, principally the economists at Stanford and Berkeley who were writing about this in the '80s, and then in the early '90s became the chief economists at the Justice Department and the FTC.

I think the culmination of that trend was the Microsoft case itself, which very much had the flavor of this body of theoretical work on network effects, that certainly had most immediate application in telecommunications.  Really, it was an idea that I guess, before Stanford and Berkeley, hatched out of Bell Labs at the time David was there.

It remains to be seen, I think, whether the Trinco case will stop that trend and reverse it.  I'm not convinced right now that that will necessarily happen.  I think Trinco will eliminate the duplication of regulation on top of antitrust, or antitrust on top of regulation.  I mean--well, I shouldn't say eliminate.  It will reduce the extent of overlap.  But I don't think it necessary speaks to the question of whether the intellectual foundation of industry-specific, sector-specific regulatory law and policy will cease to have a major impact on antitrust.

Let me give just one example now of how I think we might see antitrust law influenced by a significant telecom regulatory decision.  The Supreme Court decided the TELRIC case a year or so ago.  This is the Verizon case.  And it came out after the D.C. Circuit decided the Microsoft case and the Supreme Court denied cert.  And without getting into the gory details of the pricing of unbundled network elements, let me say, at a 20,000-foot level of abstraction, that what the Supreme Court did was to defer under the Chevron doctrine to the FCC in its adoption of total element long-run incremental costs, or TELRIC, as the pricing rule to be used for unbundled network elements for local loops and so forth.

Now, if I were a district court judge and I was sitting in an antitrust case, and it was some esoteric tying case, a case involving bundling, along the lines of the Microsoft case, and the plaintiff said the tying product A--let's call it the operating system--is bundled with the tied product B--let's call it the browser--and that violates Section 1 of the Sherman Act and Section 3 of the Clayton Act, it's an unlawful tie.

Okay.  So it's not, strictly speaking, unlawful to offer the bundle, but it's unlawful not to offer the two pieces of the bundle separately.

Okay.  So let's say there's a finding of liability that this is a tie.  Then what?  What's the remedy?  Well, I would think that a district court judge at that point would have to say, all right, we have to price this tying product, the operating system, on an unbundled basis.  So now it starts sounding a lot like pricing unbundled loops.  How do you price the unbundled operating system that's been packaged along with various application systems?

Well, in the telecom case, there were two approaches.  You could take a top-down pricing model.  You could start with the retail cost--the retail price, rather, and eliminate the incremental cost of the tied product, and the difference would be the price that could be charged for, in this case, the operating system.  And that's what the Telecom Act essentially says that you do under resale.

The alternative would be a bottom-up pricing methodology, which is what the FCC has used in its TELRIC proceedings, where you say we're going to calculate the incremental costs of the tying product, the incremental cost of the tied product, and then those products, when sold separately, will be priced according to a formula that takes that incremental cost and adds some contribution to the common cost.

I would think that a district judge, when given this kind of pricing methodology, would say:  Well, you know, it sounds good to me.  The Supreme Court has ratified this approach in a regulatory context, giving Chevron deference to an expert agency, the FCC, after years and years of administrative litigation.  And who am I to say that TELRIC isn't the right approach as opposed to some measure of fully distributed cost or something else?

So I could easily see a situation where future esoteric antitrust cases involving bundling would rely on the kind of analysis that we saw in the Verizon case and ratified by the Supreme Court under a Chevron approach.

Now, the problem, of course, is if these products that are bundled together are essentially intellectual property, their marginal cost is zero.  So neither one of these pricing methodologies in practice would be at all easy to perform.  So we may see a situation where the thinking on the regulatory side influences the development of antitrust doctrine, but it's not necessarily going to make it easier to implement in practice.

Let me conclude by saying that the kinds of developments that we're seeing in telecom regulation and also in what I would describe as the cutting edge of antitrust law as applied to network industries is something that the United States would like to see, the U.S. Government would like to see adopted by other countries.  Certainly the U.S. Trade Representative spends a lot of time trying to get other major trading partners of the United States, such as Mexico and Japan, to embrace the FCC's approach to the pricing of unbundled network elements and the like.

This is potentially going to collide with a different approach for sector-specific regulation and competition policy that is being employed in Europe.  And at that point, I'm turning over the microphone to Damien to tell you about that.

MR. GERADIN:  Well, thank you very much, Greg.  I was delighted when you asked me to write the second section of the paper on the European approach to telecommunications remedies.

I think what I found very interesting is that the EU and the U.S. are facing very similar issues.  Let me take two examples.  You had your Microsoft case.  We have our Microsoft case, which will be, by the way, decided or settled within the next few weeks.  It could be next week.  It could be in two weeks or in six weeks.  We don't know, but--and the issues raised are pretty much the same as the issues that were raised in the U.S. case, except that here the bundling was not with the, you know, web browser system, it was with Media Player.  But roughly these are the same types of issues.

The Supreme Court decided in January the Trinco case.  Well, we've got our Trinco case.  If you look at the Deutsche Telekom decision of the European Commission adopted in 2003, I think it raises very similar issues of a combination of antitrust and telecommunications remedies.  The only thing is that the European Commission took the opposite approach of the Trinco case.  I will say a few words about that.

Let me give you one minute of background about telecommunications deregulation in the European Union.  Well, the whole process started in 1987 when the European Commission decided it was time to get rid of state monopolies in the telecommunications sector.

The European approach was based on a progressive market opening of the sector, so the European Union adopted a series of directives, in other words, you know, regulation--some of our regulations are called directives--that were aimed to progressively open to competition the various segments of the telecommunications sector.  And as in the U.S., the whole sector was open to competition in 1996--well, in fact, 1996 was the time of the decision.  The entry into force of the full competition directive was in January 1998.

Then in 2003, the European Union adopted a new regulatory framework, which is called the New Regulatory Framework on Electronic Communications.  So why do we refer to electronic communications?  Well, essentially because there's a belief that there's a lot of convergence between the telecommunications sector, the IT sector, and the media sector, and, therefore, it didn't make any sense to have separate regulation for these different industries.  So this new framework is focusing on the economic regulation issues, you know, like access to the network, be it the telecom network or the cable network.  All the content-related issues are dealt with through separate directives.  So this is basically where we are.

Now, let me go through the list of issues that were raised by Greg in his presentation, and you will see that we're really facing the same issues.  I can basically make most of the observations that Greg made, but on some occasions, I think we took a different route.  And I think it's very interesting to see what happened.

So on this debate between ex ante versus ex post remedies, well, we do have ex ante regulation.  Well, you know, the regulatory framework I've been mentioning a moment ago contains a set of regulatory remedies, with perhaps two major differences with the United States.  The first one is that our regulation, our set of rules, is much less intrusive than the American one.  I mean, if you look at the set of directives we have, we have actually six, six main directives.  I don't think it makes more than 60 pages of written materials compared to 600 pages in the 1996 Telecommunications Act.

There is a second difference which explains why our framework is less incentive.  It's that it is based on a decentralized model.  In other words, we don't have a European FCC.  What we do have is a set, a network of European telecommunications regulator.

So what does it mean in practice?  Well, it means that typically the directives will contain broad principles, like that in some circumstances, you know, access to the network should be given.  But it will not say much about pricing issues.

For example, in these directives, we don't have an indication about the prices at which, you know, unbundled network elements should be priced, except that, you know, they should be oriented toward costs, which is a very vague point.  So that means that all these decisions tend to be taken at the national level.  Very often, regulators will follow similar approaches, but in some cases there might be differences.  I like the idea of allowing member states to follow different approaches.  I think it's good.  But at the same time, it can also create problems, you know, distortions of competition between different jurisdictions.

We also rely on antitrust rules, and I think that in the European Union antitrust rules have played a much more important role than in the United States, besides, of course, the famous AT&T case that was decided 20 years ago.

As Greg mentioned, typically antitrust rules will be applied ex post, and in the European Union, we had no fewer than 20 or 25 important cases that were started by the European Commission.  Most of these cases were Article 82 cases.  That means abuses of dominant position cases.

The interesting thing is that most of these cases were started on the basis of antitrust rules by the European Commission, but then were settled by the national regulatory authorities on the basis of sector-specific regulation.  It's a little bit like if the DOJ was starting procedures at the federal level and then at some stage in the process was sending the case to the public utility commissions.  So that's something very interesting.

One interesting case we had last year was the Deutsche Telekom case.  Well, the Deutsche Telekom case was about the following issue:  In Germany, rates for all sorts of products and services have to be approved by the telecommunications regulator, and Deutsche Telekom had its prices for access to the local loop approved by the German telecommunications regulator.

So what happened then is that the European Commission received complaints from Deutsche Telekom's competitors, basically companies buying the service from Deutsche Telekom, claiming that there was a price squeeze.  In other words, the price for the wholesale service was roughly the same as the price for the retail service; and, therefore, these companies could not enter in the market because they couldn't make sufficient profit.

Then what happened is that the European Commission said it was right and fined Deutsche Telekom 12 million euros.  The reasoning was that Deutsche Telekom still had some freedom to rebalance its tariffs in order to avoid a price squeeze.  In other words, okay, the tariffs for the wholesale service had been approved by the telecoms regulator, but what they could have done was to increase the price of the retail service in order to avoid a price squeeze to happen.

So that's very interesting because it's a little bit the reverse of the Trinco case.  In the Trinco case--well, at least you can read the Trinco case as saying that when there is a regulatory remedy, there should be no space left for antitrust.  What the Deutsche Telekom case is saying is that it is not because there's a regulatory remedy that antitrust rules should not continue to play.  So it's a very different orientation which I think is interesting to point out.

Antitrust rules, of course, in the European Union can also be used ex ante, and we've had a very large number of merger cases--well, not so many merger cases, but at least a fair number of joint ventures that had to be reviewed by the European Commission.

If you look at the difference between ex ante and ex post rules, you will see that the borderline is very thin, as in the U.S.  For instance, for--well, one reason is that many ex ante antitrust decisions like, you know, decisions over mergers will seek to achieve regulatory objectives.  Let me give you one example, which was the Atlas decision.  Atlas was a joint venture between Deutsche Telekom and France Telecom in order to provide services to global--you know, to multinationals, very large companies.  And the Commission said okay, but then the condition is that the French and the German Government have to accept to open the market for network services earlier than what was provided for by the legislative framework.  So the Commission used an antitrust case in order to gain some--in order to pursue some regulatory objectives.  You could see the same in the Telyat-Telenor(?) case.

The second issue, which is antitrust as a form of regulation, as was pointed out by Greg.  Well, one could see the same in the European Union.  The ex ante cases, merger cases are imposing lots of condition on market players.  I mean, some people argue it's a form of regulatory extortion process because very often the operators are in no position to negotiate.  So in order to have their deal cleared, they have to accept 25, 35, 45 conditions.

Another illustration of this is the increased reliance on guidelines.  We have dozens of guidelines.  It's amazing.  Almost every week there is a new guideline on something.  So I suspect that in the future acting as an antitrust advisor will be very much about checking that one practice fits into a set of guidelines imposed by the European--well, not imposed, but produced by the European Commission.  So, again, it's like in parallel with the American situation.

Now perhaps the main difference.  Well, Greg was mentioning that, in his opinion, the driving force for remedies in the telecommunications sector these days is sector-specific regulation, and he mentioned a number of examples.  I think in the European Union it's very much the reverse.

Why is it different?  Well, because our New Regulatory Framework for Electronic Communications participate from a completely different idea.  I think it is based on two central ideas.  The first idea is that the--well, the analysis and the remedies that are provided by this new regulatory framework should be based on antitrust laws and principles.  Let me give you an example.

The trigger for regulatory requirements to be imposed on an operator is the finding that this operator has significant market power.  Well, in the past, significant market power meant that you had 25-percent market share on a given market, right?  That was not a very good idea.  I think the trigger was far too low.  It also was very artificial.  Why 25 percent?  Why not 26 or 24?

The new definition of significant market power is based on the concept of dominance, which is an antitrust law concept.  It is based on the definition of dominance that was given by the European Court of Justice in very old cases like Hoffman-LaRoche or cases like that.  So there's a real penetration of antitrust concepts and principles into this regulatory framework.

The second idea--and this is perhaps the main one--is that antitrust rules should progressively take over.  The central idea of the European Commission is to say, well, regulation is only there for a few more years; we should get rid of it as soon as possible, as soon as markets become competitive.  But, of course, we don't know when they will become competitive.  We have already a fair number of markets that are competitive today, but we still have a number of markets which aren't competitive.

Well, as soon as a market, a market segment becomes competitive, then there's no reason to regulate it, and the discipline in this market segment should be pursued through the application of antitrust rules, which I think conceptually--I'm not sure how it's going to work in practice, but conceptually it is actually very good because it means that there is only a regulatory solution when there's a market failure.  When you don't have a market failure, no more need for regulation.

So how does it work?  Just in 30 seconds let me explain to you how the framework applies to market players.  Well, there are basically three steps.

The first one is that the national regulatory authorities--so the German telecommunications regulator or the Belgian telecommunications regulator--will define relevant markets on the basis of one recommendation and a set of guidelines that were adopted by the European Commission.  So they will say, okay, there are, you know, 22 different markets.  So that's the first step.

Then the second step is to check whether these markets are competitive.  Well, according to this new framework, when a market--well, when there is no company with a dominant position or significant market power--it means the same thing now--in the market, that means that the market is competitive and there is no need for regulation.

So I think there's an interesting difference here, again, with the U.S., where my understanding is that the Telecommunications Act will apply certain remedies to, you know, predetermined categories of operators independently of their market position, like remedies will be imposed on ILECs or, you know--whereas, we don't have that in the sense that regulation will only be imposed depending on your market position.

Then the third step is if a market is seen as non-competitive because there is still a company with significant market power, the regulators can choose a remedy from--one or several remedies from a list of remedies that are provided by the directives.

If the market is competitive, then they cannot impose a remedy.  And if a remedy had been imposed in the past, this remedy should be withdrawn.

This exercise has to be done from time to time by the national regulators, which means that if it works well, every time there is, I would say, a review of the situation in the member states, some rules should disappear because there will be findings probably that some markets have become competitive.  So it's a system that should self-destruct itself if it works according to plans.

So, my conclusions.  Well, if you look carefully at the two systems, I think you can see two common points.  Essentially one--the first one is that we both use ex ante and ex post remedies, and in both cases we have developed what we would call hybrid remedies, remedies that are a little bit in between ex post and ex ante.

The second thing is that antitrust is becoming increasingly regulatory because of these merger decisions imposing a set of conditions.  You know, they look like regulations to some extent, and also because of this set of guidelines.

Now, two major differences.  The first one is that ex ante regulation is much more intrusive in the U.S. than in the European Union.  Last year, I published a book with an economist of the World Bank in which we surveyed five jurisdictions, and we also looked at many more to inform ourselves.  And we came to the conclusion that the U.S. regulatory regime was by far the most intrusive.  Other regimes were much less intrusive.

And the second difference is I think that in the European Union there is a much larger intervention of antitrust authorities, many more cases.  And I suspect that with the Trinco case, which I think will probably reduce the impact of antitrust in the telecommunications sector and other network industries, the gap between the U.S. and the EU might grow even bigger.

Thank you, and I think--well, you're the moderator.

MR. SIDAK:  Tad?

MR. LIPSKY:  Thanks, Greg.

This is such a target-rich environment, I'll have to struggle to stay within the allotted time, which I will, David.

[Inaudible comment.]

MR. LIPSKY:  I've always been very much in favor of taking the long view when approaching some of these competition and antitrust issues.  I think it tends to give you a broader sense of humility in formulating remedies, regardless of whether they're antitrust or regulatory or ex post or ex ante.

My favorite illustration is how in the early part of the 20th century the perceived bottleneck in telecommunications was the Bell System's unique control of long lines, the long-distance element, with the thought being that it was the equipment suppliers and the local exchange carriers who needed access to that bottleneck, and there was a fair amount of public policy and regulatory wrangling that came out of that era.

Then in mid-century, about the time that Attorney General Brownell started thinking about suing or not suing the Bell System, I think the theory at that time is, if I'm not mistaken, that Western Electric, the equipment part was the monopoly and it was the service ends, the long-distance and local, that were dependent on that and needed access to the equipment.  And then, of course, the straw, the bottleneck theory that broke the camel's back, well, it turns out it was local exchange access all along that was the bottleneck, and so that was the focus of the AT&T decree.

Now, don't get me wrong, I'm not saying that it had to be the case that for that entire century there was just one bottleneck and there were wrong guesses and one right one.  I mean, all three guesses might have been right, but if you think about how dynamic the technology has been and you think about how different the role of telecommunications is in the economy and in society over that period of time, it should give you a powerful sense of a need to be skeptical and to question the assumptions that various regulatory methodologies are built up on, whether they fly under the flag of antitrust or otherwise.

The comparison between the American and the European approaches to this type of question is fascinating for a huge variety of reasons.  You've got two very, very different points on the spectrum when you consider all the differences between the political and legal structure of competition rules in the European Union and those in the U.S.

In the United States, you have basically very little commitment within the legal system to the idea of competition.  Now, that may sound a bit paradoxical, but remember, you still have William Rehnquist going around saying, in response to questions that usually pose some kind of angst about overbroad regulation or regulation that stifles business opportunity or what have you, fond of quoting Justice Holmes' famous phrase from his dissent in Lochner that the American Constitution does not enshrine Herbert Spencer's social statics, by which he means there is no preference in the American legal system for competition and market outcomes as opposed to other outcomes that might be achieved by the legitimate exercise of Congress' powers, and specifically the commerce power.

Now, it's quite the opposite in Europe, and for those of you not familiar with European law and their approach to competition rules, I urge you, take along a copy of the Treaty on European Union.  It's probably a lot simpler and somewhat shorter if you take the original 1957 Treaty of Rome version, but what I'm indicating will survive all of the versions of the treaty.  Virtually every single article talks about freedom of something--freedom of movement, freedom of establishment, the freedom to establish a business anywhere in the Common Market, the freedom of individuals to move and sell their labor anywhere.  And, of course, there are some competition rules in Europe--which brings me to another major difference between Europe and the United States, very much on this same theme, and that is, in the United States we have come about to give a fairly broad deference in the antitrust enforcement system to the regulatory system.

Now, you can argue about whether that's factually true.  I think Trinco in a way I'll describe in just a moment really is the high watermark of antitrust deference to regulatory solutions.  But in Europe, it's just the opposite.  Not only do they suffuse their entire legal structure with these fundamental ideas that are either competition based or competition related, unlike in the United States the European Treaty of Union applies competition rules to state enterprises and it obligates the member states to act in accordance with European Union policy.  Whereas, again, in the United States, it's just the opposite.  We have a very, very broad concept of government action lying totally beyond the reach of antitrust remedies here, so that there's a distinction of government action and private action.  If it's government action, the Sherman Act simply doesn't apply to it.  Well, in the European Union, it can be a little different.  If you're a state entity providing essentially a utility service or a public service like a telecom monopoly or like any other kind of what we think of as local public utilities, you are bound by the rules-- subject to some limitations and some carve-outs, you are basically bound by the competition rules that apply to dominant firms.

This is probably a good place to throw in the point -- [tape ends].

-- Trinco won't be any surprise to those of you who have been following utility regulation and the way in which antitrust applies to utility regulation in the United States.

We had a case in the First Circuit, also a price squeeze case, in the electricity industry, much like the one in the telecom industry that Damien described, Town of Concord v. Boston Edison, and decided by then Chief Judge Steve Breyer, now Justice Breyer, who was in the majority in Trinco.  And basically he did a very interesting, rather deft thing to get rid of an antitrust claim of a price squeeze by a regulated utility.  What he did--we have some clear doctrines under U.S. antitrust law.  We have the doctrine of express immunity.  Congress can say the following industry is immune from antitrust law, QED, that's all there is to say about it.  As long as it was authorized under the commerce power, that's the end of the debate.

We do have a doctrine of implied immunity which is said to be applied very narrowly.  The rule is that without an express immunity, a private anticompetitive act is immune by virtue of federal regulation only where there is, quote-unquote, plain repugnancy between the regulatory statute and the antitrust standard, theoretically meaning that you only get off the hook under antitrust law if it's almost literally impossible to have the supervening federal regulatory scheme function if you also apply an antitrust standard.

Very, very interestingly, in this Breyer decision in the First Circuit I'm talking about, Concord v. Boston Edison, it's kind of a gloss on top of that implied immunity doctrine, the effect of which is actually to very much narrow the scope of antitrust and broaden the scope of regulation.  If you phrase the question in the antitrust suit not as whether the conduct is immune but as whether the particular form of conduct ought to be susceptible to antitrust liability under the substantive standard of the Sherman Act, then you can evaluate all of the institution--the comparative institutional advantages and disadvantages of regulation versus antitrust, and you evade that plain repugnancy standard by simply saying, well, we don't think antitrust courts can improve matters here, because regulation is supple and it's detailed and it can take place in real time.  You know, query whether any of that's true.  Antitrust takes a long time, and we have to hire lots of lawyers and have a 13-year case.  And so maybe the application of an antitrust remedy can't even improve things, and we're not going to recognize this price squeeze claim in this particular industry.

Well, that was Concord v. Boston Edison, and I think you'll recognize essentially that the opinion in Trinco is very much a straight shot out of Concord v. Boston Edison.  It does not find immunity per se.  It says--as a matter of fact, I'm not even sure there's--well, the approach to the immunity is very simple.  It says the statute has an antitrust savings clause.  Nothing in this law changes anything under the antitrust laws, QED, now the question is whether there's liability under an antitrust standard.

But, again, by phrasing the question not as one of immunity but as one of whether there ought to be recognized a species of antitrust claim under the substantive standards in light of these various institutional advantages and disadvantages, we find that there is no such claim as the one that Curtis Trinco's law firm has brought in this case.

I've previously written--Greg, if you need to cut me off to leave room for David, please do.  I was just going to make one or two more points.

First, I think this is--this point that Damien has focused on, on what is the endpoint of what we are doing, and the thought that in Europe the endpoint is really the--you know, regulation become vestigial and disappears, I don't hear that same debate, at least not with the same clarity, occurring in the United States, even though you could argue that with the evolution of technology and many different pipes being put into many different places, reading in the paper--I mean, we've had this debate about multi-channel video to the home.  I notice now that the buzz seems to be about telephony through the Internet, through wireless Internet.  So now we've got not only a route around the traditional telecom access system, but now there's a wireless route through the Internet around the wireless alternative to the traditional telecom system.  So our bypasses are being bypassed, I guess.

So it's a legitimate question as to whether this European model of making regulation vestigial ought to be hurried along a bit.

We tend to find, though, unfortunately, that the--and this is a very well recognized phenomenon not limited to telecom, how the legal system tends to plod along at a very slow pace and at a great distance behind where the technology is taking us, particularly in industries with technology that is as dynamic as it is in electronic communication.

It will be interesting to see who gets to the goal line first, who gets to a solution for electronic communications that is considered to be the proper mix of competition and antitrust remedies on the one hand and regulation and regulatory remedies on the other.  There are so many differences in the starting points.  The United States has an integrated market, but yet we have a state fragmented PUC structure, so in that respect, it's got some resemblance to the national regulatory authority structure of Europe.

We have no way of smoothly crossing over between the competition rules and the regulatory rules in the way that the European Treaty provides the Commission the power to make directives and to impose them in a sense on the NRAs.  On the other hand--and this is where I'll close--it's worth pointing out that the substantive rules--although I don't think this difference, at least in telecom, might--has not really been as salient as it otherwise might have been.  But it's worth pointing out that the United States since the early '70s has been involved in this process of conforming all of its antitrust standards, not just the standards of Section 2 in the Sherman Act, the monopolization standard, under which this debate about access to facilities tends to be carried out, but in the area of vertical restraints and the relationship between intellectual property and antitrust, in the relationship of joint ventures and mergers and acquisitions.

Consumer welfare and efficiency and maximizing productivity, it's now accepted in the United States by all but those on the fringe that those are the dominant guides to interpretation of the U.S. antitrust laws, and all of the doctrines we're talking about at some point will rest on principles at that level.

I think in Europe it is very much the opposite.  The concept that an unfair price or an unfair trading condition is specifically written right into Article 82, the law on prohibition of abuse of dominance, and the head of enforcement, Phillip Lowe (ph), the Director General for Competition for the European Union, has said in a speech as recently as October that that's in our law and we've got to interpret things--we've got to prohibit unfairness.  Unfairness is just a part of the European competition law fabric.

And while the United States cases at the appellate level, at the Supreme Court level, have been slowly kind of closing in on what we call the ghost fleet of the U.S. Supreme Court antitrust cases, the cases from the '30s, the '40s, the '50s, and even the early '60s that had these terribly vague and--these themes that are very much inconsistent with the emphasis on productivity and consumer welfare and so forth, we're closing those out.  And Trinco was in some respects a huge step forward.  It knocked out the so-called leverage doctrine.  It minimized the influence of this case called Aspen Skeen (ph), which is thought to impose an obligation to share monopoly facilities.  It very strongly suggested that Section 2 is going to continue to be guided along this productivity-based method of interpretation.

In Europe it's just the opposite.  They started with this very fuzzy statute that enshrined--that talks about abuse of dominance but enshrines very non-economic notions like unfairness.  And, unfortunately, the evolution of the case law in Europe has tended to exacerbate this kind of kaleidoscopic approach, has not given anybody any assistance in saying, ah, when I am advising a client on the risks of abuse of dominance liability in the European Union, here's a case, here's a statement, here's a doctrine that gives me a fair idea of what I should be pointing toward.  It's very much the opposite, that the European cases, particularly the recent decision in Michelin II, have actually taken away some of the hope that earlier cases had provided, that the Article 82 interpretations in Europe, the construction of that statute, that treaty provision might have focused more narrowly in on an objective, quantifiable set of rules that would also be based on efficiency and productivity.  So it's not only a problem for those in Europe who have to comply with European law and guess where it's going.  It's an especially bad problem for companies that operate globally and have to follow the local rules everywhere because, in this respect, I think the U.S. and the European rules are becoming less coherent rather than the opposite.

MR. SIDAK:  Thanks, Tad.

David?

MR. SIBLEY:  Thanks.  I'm going to, for the most part, talk about the convergence in the U.S. of antitrust and regulatory policy.  I will ask a bit about the European case.  I'm actually going to spell out regs even a little bit more than Greg did before I start sort of evaluating it.

The two big points Greg is making is that there is a convergence of antitrust policy and regulatory policy, and to the extent that that convergence takes place, particularly when antitrust starts looking a lot more susceptible to regulatory remedies, he's concerned that there would be a grafting on of some of the very complex regulatory remedies that the FCC, for example, has come up with in the past few years, particularly TELRIC pricing for network industries.

And he has a very simple and appealing way of characterizing antitrust and regulation.  Regulation is characterized, by and large, with some exceptions, as being ex ante remedies, that is to say, you don't wait until something bad has happened; you say at the outset what ought to happen and hope that people follow the rules.  Whereas, antitrust is, with, again, some exceptions, largely ex post.  It only kicks in, with the exception of mergers, when something bad has happened.

And although, you know, Greg is careful to say that the sort of convergence he's talking about comes from both ends, it sounded to me as if he was a little more concerned about antitrust policy becoming regulatory in nature than the other way around.

Now, there is, in fact, given this characterization of the difference between antitrust remedies and regulatory remedies, a fair amount of evidence in support of what Greg is saying.  In no particular order, he mentions the Section 271 activities that the Justice Department until recently has been engaged in is, you know, very much a regulatory sort of process.  One could imagine the FCC doing exactly the same things that the Justice--in fact, it does--that the Justice Department has been doing on certifying compliance with checklists.

Greg and Damien also point out that the landscape is somewhat littered with guidelines of various kinds.  We have horizontal merger guidelines, IP guidelines, joint venture guidelines, and at one time we had vertical restriction guidelines, too.

Greg is concerned about sort of duplication in the work that the Antitrust Division and that the FCC does in merge analysis, I guess, apart from the fact that it represents sort of two actors in a more cumbersome setup, as also illustrating convergence.  Traditionally, the Antitrust Division tends to defer to the FCC on mergers that it could play a role in, but that's not to say that it hasn't worked on the same mergers that the FCC has.

Finally, consent decrees involve conditions which are ex ante in the sense that they say things that ought to happen or ought not to happen if a particular merger or some other activity is allowed to go through.

Sort of building on that, Greg then says that for network industries, he has a particular concern.  To the extent--in fact, he explained that concern in some detail, that to the extent that we have an antitrust issue, say, in bundling involving a network type industry, isn't it likely and dangerous that a district court judge somewhere would say, ah, I've heard this before, this sounds like T-E-LRIC, or TELRIC.

Now, I share Greg's sense of horror at that.  I should say Greg and I were members of--it has to be said--one of the least successful expert witness teams that I know of.  In 1996-97, we were retained along with two other folks to jet around to state capitals all over the country, testifying against T-E-LRIC in favor of something called the efficient component pricing rule.

Now, I asked Greg to check my numbers on this.  He couldn't.  But as far as I know, the won-lost record was 2 in 48.  And one of our wins sounded to me as if it resulted because the judge was deranged.

[Laughter.]

MR. SIBLEY:  So I share his horror at that.

Anyway, so let me go back now and evaluate this convergence process.  Let me say at the outset, when you're talking about convergence, certainly in a mathematical sense, you're really talking about two things:  how close together are the systems now and have they gotten closer together than they were in the past.  So we not just compare them now but ask if sort of things are more likely now than they've ever been before, if that makes any sense.

Now, let me go back over the areas, the items which tend to say that antitrust certainly has its regulatory aspect to it, and, you know, what I'm basically going to say is that if we sort of compare the regulatory aspect of what the Antitrust Division does to FCC regulations, there is a quantitative difference even though there might be qualitative similarities.

Let's start with the Section 271 application checklist.  This certainly was very detailed, very intrusive regulation by the Antitrust Division.  We have had for years several people in my own group who spent a good deal of their time doing nothing else but evaluating the latest billing system from SBC to see whether the checklist has been satisfied.

Nonetheless, though, there was a--not an end date but an end event, I guess, contemplated for this type of work in the Telecom Act.  It was not envisioned that the Justice Department would go on forever scrutinizing whether the checklists, once having been satisfied, were being exceeded at some point in the future.  And, in fact, we're not doing this kind of work anymore.

Now, on the policy statements and guidelines, it is certainly true, as Greg says, that when you look at the merger guidelines, these are ex ante in nature, and they are sort of trying to say what policies are.  And the same is true of the IP and joint venture guideline.

Nonetheless, when you compare these two, what I think of as regulations--and, of course, Greg is correct, by the way, in pointing out that whether or not I'm the omega, I was one of the alphas.  So when I talk about regulation, this comes after having been involved in numbers which won't mean anything to the less middle-aged of us, you know, Docket 18-128 and 2003, things like that.  So, I mean, the kinds of regulations that I think of when people talk about FCC regulations are extremely detailed regulations.  They don't just say in general what sort of analysis you will do.  They are very prescriptive about exactly how you will do it.

There is a gap in the level of detail and specificity between what, for example, the merger guidelines do and what I think of as regulations would do in the context of the U.S.  Now, Damien suggested, I think, at various points that European regulations haven't gotten quite as intrusive and detailed and horrible as U.S. ones have.  But at least in the U.S., you know, regulations are that way for the most part.

What the merger guidelines do, for example, is to say in qualitative terms with--well, no.  It specifies safe harbors numerically, you know, the famous 1,000 and 1,800, which it now turns out haven't been taken seriously for a long time.  But with those exceptions, it really says here are factors we will look at.  It doesn't say, for example, that a firm that we might be dealing with at some point in the future has to look at them that way.  And it leaves lots of room for deciding just how you would look at those things.

So it is ex ante, but it's really not very detailed.  The IP guidelines are that way as well.  I can't say that I know much about the joint venture guidelines.  And I have seen the old vertical guidelines, and they weren't that detailed either.

Now, consent decrees, which are another activity that you might think puts the Antitrust Division in the regulatory business.  Well, it does in a sense, but, again, consent decrees are not thought of as being devices of long-term scrutiny by the Justice Department.  Typically, there is an event that is supposed to happen, a spinoff or a supply agreement or something like that.  And once it's been verified that that event has taken place in a way consistent with the agreement that the parties made with the Justice Department, then that's over.

In fact, to illustrate how over it is, at one point a few months ago I had occasion to try to find out how well certain types of supply agreements had actually worked in the past.  And I talked to the various gray beards in the Antitrust Division who had been involved in these things, and they actually didn't know.  And I found, somewhat to my surprise, that once a supply agreement, for example, or spinoff is inked, that's it as far as the Antitrust Division is concerned.  It might work out in all kinds of crazy ways, but as far as the Antitrust Division, our job is done.

In fact, there is a paper in, I am think, a 1968 Journal of Law and Economics article by Kenneth L. Zinga (ph), which looks at the effect of court-ordered dissolutions of previously approved mergers.  And he found that in most--in a significant number of those cases, the outcome of the sort of win for the Justice Department or the Federal Trade Commission had been one that went directly against the liability finding in the case that was supposed to settle.

So, for example, if it had been decided after the fact that a particular merger increased concentration too much and reduced competition, a significant fraction of the time the outcome of that case, when the assets involved in the merger were ordered divested, the outcome was to sort of reproduce the same level of concentration that caused the government to say the original merger ought to be changed.

MR. LIPSKY:  David, if you'll forgive a brief interruption, when I worked for Baxter in the early '80s, there was actually an expression for this phenomenon that you're observing, which was:  "Everybody likes to catch 'em, nobody likes to clean 'em."

MR. SIBLEY:  Well, that's right.  You know, institutionally, once you finish winning a case as a lawyer, your job is done.  After that point, people with green eyeshades take over, and their only job is to get rid of the assets.  It's kind of like an IRS examiner who has to clear cases.  And, you know, these people, to say the least of it, are not schooled in any sort of sophisticated way in economics.

So consent decrees, again, are sort of--so that really illustrates how limited the regulatory oversight there is.  It's some exaggeration.  It's terribly unfair.  But you could be pardoned for thinking we just don't care after a certain point so we stop looking.  We do care, but we don't act as if we do sometimes.

So it's not clear to me how too close together the systems are at this point.  I'm not--clearly there's some regulation in what the Antitrust Division does.  But I think it's different both in, to some extent, qualitatively, consent decrees being an example--when a certain thing happens, we're done--and also qualitatively, even things that we are going to look at forever, merger guidelines, we don't look at them in as detailed and intrusive a way as a regulatory agency would.

So I am interested by the degree to which the two systems are close together, but I'm not overwhelmed by that.

Now, let me go to sort of the second part of my test, which is:  Are the systems converging?  Right?  Having gauged whatever you think their differences or similarities are now, were they more that way 20 years ago?  In fact, Tad being here, I gather you must have been a deputy in the early '80s.  So I'm about to talk of that about which I do not have firsthand knowledge, so correct me here.

Let's look at the evolution of the merger guidelines over the years.  I guess the first version was in '68--was that the version in effect when you were a deputy?

MR. LIPSKY:  No.  Actually I was there when Baxter wrote the '82 version.  '68 was the incumbent version, but very [inaudible] at that point.

MR. SIBLEY:   Right.  Now, my impression is, if you want to gauge intrusive regulation by commanding adherence to numbers, the '68 was probably as intrusive as things have ever been.  In fact, numerically I think somewhat less so over time.  Also, as time goes on, the guidelines have been added, say, as of the '92 guidelines, with a lot of sort of new suggestions about how to analyze cases.  I'm thinking here of the current emphasis on unilateral effects analysis.  And also, as time goes on, the list of factors that you look at in a coordinated effects analysis has sort of gotten bigger and tracked the economic literature on that point.

So using the merger guidelines as an example, to me that doesn't say that that regulatory--that ex ante aspect of what the division does has sort of gotten more intrusive over time.  If anything, it has sort of legitimized wider and wider arguments away from specific numbers.

Now, the IP guidelines are a fairly recent phenomenon.  I think they were written in--what?  '95?  I think it was the Rich Gilbert era, yeah.  So it's sort of too hard to say whether they've evolved.  They are what they are.

Now, the vertical guidelines actually provide a counter-example.  These guidelines were pulled by Anne Bingaman, so we actually don't use them anymore.  And I have tried to find a copy of them and couldn't.  So I don't think that the guidelines show the evolution of--shows a convergence over time.

I don't actually know--and I'm going to another illustration of convergence--whether joint work between the FCC and the DOJ is sort of more frequent now than it used to be or not.  There's always the potential for it because, you know, when we look at the FCC, they have all kinds of data.  We never get merger cases, which we'd love to get our hands on.  They look at us and say, Ah, you guys have the CID power which we wish we had but don't.  So, you know, those two facts have been around for a long time and, you know, I'm sure will make us work jointly with the FCC at some point, but they probably have in the past as well.

Now, finally, on the network industries point, just to remind you, this is the point in the story in which Greg is really alarmed, and, you know, theoretically I could be, too; namely, do we seek convergence to the point where an antitrust court judge is going to import TELRIC?  Well, they might do that.  I mean, there's no saying what some judge somewhere might decide to do.  But he might equally well say, well, gosh, you know, the sort of network industry argument that I've looked at in this case and under which I have found liability, I could look at TELRIC or I could just look and see what sort of solutions are proposed in the Microsoft case.  It wasn't to sort of propose prices for unbundled browsers or something like that.  It was to require that Microsoft allow equal access to Windows from competing types of middleware.  So it wasn't that intrusive, at least not in the particular way that TELRIC pricing would suggest.

So with respect to what Greg is saying, it's a fascinating point, and I love the way that Greg characterized it, because it is true, regulation by and large is more ex ante than ex post, and antitrust is more the other way.  That's a very clean way of looking at it.  But I don't know that the convergence he's talking about is something that I see yet.

Now, let me talk a little bit about Damien.  You know, Damien has a very appealing characterization of the European system as being one that progressively kind of folds back after market dominance findings are reversed in industry after industry.  And it does so by sort of delegating these decisions to national regulatory authorities, which he calls NRAs, which may have some expert knowledge in their particular country that the folks in Brussels don't.

I would say, though, at this point it's not clear that that system has met what may turn out to be its biggest challenge yet--for example, the prices for unbundled network elements.  It's true that the EC has used a word which allows some latitude.  I think "cost-based" or "cost-oriented" pricing, something like that.  But then the EC--I'm not sure the EC yet has had to judge an actual case where parties have tons of money to gain or lose, depending on how that's interpreted.

Now, most likely I gather from what you're saying that would be delegated to a national regulatory authority.  So at some point, the Germans or the Spanish or the Belgians are going to decide what "cost-oriented" means, and if they don't want to take a very detailed look at that, there will be probably billion-dollar corporations on either side of whatever they might mean pushing them in the direction of becoming very specific indeed.

In my experience, which admittedly is a bad one, having been through 18, 128, 2003, 19, 104 and TELRIC, says to me the minute you start allowing lots of public input into this process, it's going to get as complicated and messy as the U.S., the sort of way--there is a little bit of hope here, though, possibly, and that is that, you know, there is a tradition in Europe of not allowing the amount of public input that the U.S. system does.  You know, it is not unknown, for example, for the British--in fact, I have seen a decision on price caps in the '80s by the British Oftel (ph) which basically said this is what we decided and we did it for good reasons and we don't feel like saying any more.  And that's it.

Well, okay, that was one decision which went into effect without a lot of litigation.  So it may be that there would be European traditions which would sort of come into play here, in which case I would believe that the system would work out in a lot sort of smoother way than it has in the U.S.  But, you know, as soon as you allow a lot of public input into the process, then I suspect you're going to end up in something like the mess we have.

Let's see.  In that vein, I wonder, the sort of rollback of regulation you're talking about here depends on the precise definition of the word "dominance."  And, you know, at the level that the EC has gone so far, it doesn't have to be that precise.  But somebody someday is going to have to be, if only because there's a lot of money riding on exactly what it decides.  So the real test of whether the process in Europe works as smoothly as we'd like it to I think may have--perhaps has yet to come.

MR. SIDAK:  Good.  Thank you very much, David.

At this point I'd like to invite questions from the audience.  I know that the panelists probably have some points they'd like to make in response, but I'm going to ask them to hold back and maybe work those into their answers to questions from the audience.

Could you wait until the microphone comes to you?  And please give your name and your affiliation.

MR. FEIN:  I'm Bruce Fein of Fein & Fein.  I think that there can be adjuration of intensity of regulation of actual private behavior based upon whether you've got an ex ante list, like the four checklists of Section 271 of the '96 act, as opposed to a more generalized antitrust standard that would kick in post hoc, after you have a finding of liability.  And I think Tad may have brought this out when he expressed great chagrin over knowing what the unfairness element of the European Union Treaty means when you combine it with market dominance as well.  He says there really has not been any clarity on that particular critical standard that's essential to lawyers' advising their clients.

And when you have--and we have something similar in the Federal Trade Commission Act that talks about unfair methods of competition and unfairness.  And when you have an antitrust standard that's so vague, the actual impact upon private behavior may be more pronounced than having a specific checklist because the lawyers are advising you, oh, you have to stay clear of these variety of possible practices because we don't know whether you would be found liable, and the stakes are so high whether you have treble damages or things of that sort.

So I'm not so sure that in trying to assess the impact or the handicap you're imposing upon private business planning that there is a lot of difference between an ex ante list of regulatory rules as opposed to a general vague antitrust standard.  It may, in fact, be more welcome to the regulated industry to know exactly what the rules are, even if you don't like them, because you don't have to then steer a mile wide away from those because of some vague unfairness element.

The second observation is I'm not so sure that the European Union preoccupation with antitrust is all that different from that in the United States.  Tad mentioned that he thought that perhaps a more regulatory approach was superseding an antitrust approach.  But I think if you look at the way in which the Commerce Clause is treated, very much like the standards of freedom of capital, freedom of labor to move within the European Union because of the broad sweep that intends to create national markets created under the Commerce Clause, and also with regard to, you know, the antitrust immunity, the Parker v. Brown doctrine, it doesn't apply to municipalities, and there are government agencies that are subject to antitrust laws.  I think there's a current antitrust suit against the Postal Service, which may or may not succeed.

I think it's simply that the problems may be treated in a little different semantical way, but the basic effort to try to coordinate antitrust with regulatory approaches seems to me largely the same with minor variations.  I don't see the overall impact upon private industry being much different in the United States or the European Union.

MR. SIDAK:  Tad, would you like to take a stab at that since you used to be in the business of advising Coca-Cola on a daily basis.

MR. LIPSKY:  Well, I actually think that Bruce's comment about kind of the shadow effect of antitrust rules is one that is very well accepted and is, in fact, part of doctrine.  As a matter of fact, I think in Trinco there's even a reference as a way of kind of supporting the declination to recognize an antitrust claim.  Justice Scalia says that, you know, there's this possible chilling effect of the liability rule.  And, of course, the chilling effect has made some very--it's been a really major actor on the stage of some significant antitrust decisions in this post-Sylvania era that I've referred to.  In Matsushita, I think there's something in there about the recognition of a predatory conspiracy rule in that context would chill, and that was probably based on the opinion in Monsanto, which basically used that chilling effect idea to hold--I guess to establish the evidentiary rule that you can't get to a jury in a Section 1 price-fixing case just by saying that one distributor complained about the price of the other and the discounting distributor was terminated.  In view of the legitimacy of the discussion between suppliers and distributors, that's not enough evidence, because if it were, there would be lots of claims like that, and that would chill.  So I think that's very legitimate.

On your comment on sort of the role of the competition idea or the competition standard, you know, what I wanted to emphasize was not so much that there aren't in the U.S. situations where agencies can be made subject to antitrust rules, and there are federal regulatory arenas in which Congress has determined specifically that competition would be a very major part of the regulatory structure.  I mean, just look at what happened on the Airline Deregulation Act and the Section 411 regulation by the Department of Transportation since.

Congress can make and on a number of occasions has made a very specific commitment to competition in a specific regulatory context.  What I was referring to, though, is the supposed complete neutrality of our political system vis-a-vis competition.  If Congress passed a statute saying that there will hence forward be a cartel in household furnishings, under the Commerce Clause I think that would be upheld, absent some other form of constitutional abuse.  And, similarly, in the state action area, absent some supervening legislation, the action of the state, the action of a governmental entity is not reached by antitrust law.  That is the current interpretation.

So the choice can be made--and I think, you know, oddly enough, in the United States I think we probably have a greater predilection toward competitive outcomes than they do in Europe for a whole lot of reasons.  But if you just look at the constitutional structure and ask must it be so, you would say under the American structure you have--you could have a solution which had very little scope for competition.  But under the EU structure, the idea of the Common Market and the integrated market and the competitive rules for both private and state enterprises, it suffuses the entire structure.  So it's a very different feeling, at least when you look at the constitutional foundations.  Outcomes may not be as disparate and probably aren't, but--

MR. FEIN:  [inaudible, off microphone.]

MR. LIPSKY:  I don't think I agree, but in any event, let's not take up the whole time with that question.  But that's a good point, Bruce.

MR. SIDAK:  Damien wanted to add something.

MR. GERADIN:  I wanted just to react to Tad's comments as well as your comments as well, I think basically on two issues.  During my oral presentation, I did not refer to the institutional differences between the European Union and the United States.  And I think I should perhaps mention two of these to explain this different relationship or interface between antitrust and sector-specific regulation.

The first one is that--the first issue is why does the European Commission like to use antitrust rules so much in all these network industries, or even in other industries.  Well, it's because there is no FCC.  You know, we don't have a federal regulator.  And so it is very tempting for our, I would say, federal executive board, which is the European Commission, to rely on antitrust rules in order to intervene in these sectors.

The second thing is that, unlike in the U.S., you know, antitrust is an administrative regime.  I mean, basically the European Commission doesn't have to go to court.  And what is really striking is that in the U.S. you may fear relying on antitrust because it's going to take a long time.  In Europe, actually we like antitrust because very often it is a way to settle matters pretty quickly.  I mean, you can have a decision from the Commission in 18 to 24 months, and that's basically it, which is a problem in itself because of, you know, rights of defense and so on and so forth.

The situation might be changing in the European Union because, increasingly, the European Court of Justice is reviewing extremely severely the decisions of the Commission, and we may end up with the same thing as in the U.S. where antitrust cases may last several years and involve a lot of lawyers and so on and so forth.

So I think that, you know--I think, of course, it is perhaps unwise to try to compare general issues in Europe and in the United States because of these different institutional questions.  But I think that when you try to compare more limited questions, like, you know, the Deutsche Telekom versus Trinco approach, then it becomes very interesting because it puts things into a different perspective.  Why did we do things differently in Europe than in the U.S.?  So I'm in favor of limited comparisons.

Now, on the issue of unfairness, that's very interesting.  The European Union has always had a very formal approach to antitrust where you have, you know,  (?)   exemptions, for example, saying you can't do this, you have to do this, and it didn't make any economic sense.  It was very legalistic and formal.

We've moving away from this, and I think the Commission has done a pretty good job in Article 81, especially vertical restrictions,  (?)  , you know, cooperation and so on and so forth.

We have a major problem--I agree it's a terrible problem--which is dominance where the case law is a total mess.  It's very incoherent.  And according to what I heard, the Commission is planning to come up with guidelines at some stage in order to clarify its position.  Unfortunately, the Court of Justice came up with this Michelin decision, which is totally unwise, but there's nothing you can do, you know, to prevent these type of things.  Unfortunately, judges sometimes make good decisions, sometimes bad decisions.  And for us the European Court of Justice is like your Supreme Court.  We're bound by their decisions.

Nevertheless, I think that Article 82 and dominance will be the next big challenge for EU antitrust law.

MR. SIDAK:  I'd like to ask a question of David.  Given the number of guidelines that the Antitrust Division has now, what would you think about the division developing a new set of guidelines on remedies specifically?

MR. SIBLEY:  Well, you know, that could happen, and I guess--presumably the reason you're asking the question is whether that would make a difference as to the theme that you're talking about here.  Whether it does or not would depend on exactly what the guideline says.  If they were, you know, very much like FCC regulations, being very intrusive, numerically oriented, and things like that, that would cut towards your position.  If they weren't, then perhaps not.

By the way, I did misspeak myself on one point earlier.  When I talked about supply agreements being one regulatory aspect, I don't mean to say that when the government views the supply agreement as being inked it ceases to care.  What it does at that point is rely on aggrieved parties to come to it if that agreement doesn't work out.

But, yeah, if there are--but going back to your specific point, it would also make a difference sort of what the status of the remedies are.  Are they, you know, called a training manual or guidelines or folklore?  Something like that.

MR. SIDAK:  This gentleman here.

MR. MALONE:  I'm Bill Malone from Miller and Van Eaton(ph).  There's been a lot of discussion here about the Trinco case, and it seems to me that it overreads it, from what I have seen, having read the briefs and having [inaudible] argument and having attempted to read the decision itself.

You know, beforehand it seemed almost inevitable, a sure thing that the court was going to throw the plaintiff out, and it had three grounds on which to do it:  one was pure standing, one was the implied preemption, and the third was, well, it's a savings clause.  And it seems to me that if one reads the opinion in its jurisprudential context, the court chose the narrowest of these grounds, which was the savings clause, and [inaudible] justified the lawyering by Ted Olsen and by the telephone company in terms of their take on the case.  And [inaudible] prepared to argue at great length that a sounder basis of decision would have been [inaudible] -emption, but for jurisprudential reasons I think the court went the way it did and I just don't think you can read it any broader than that.

MR. SIDAK:  Okay.  I'm going to ask now if we have any volunteers on what the two missing words are in Trinco.  Any guesses?  "Consumer welfare."  Search the opinion for that phrase.  It does not appear.

My criticism of Trinco is that there's a lot of discussion by Justice Scalia of the impact on incentives for investment, for example, from forced sharing of facilities.  But I don't see where he takes the next step, which is the one that Frank Easterbrook took in a concurrence in the '80s, where he said in an essential facilities case involving some basketball arena in Chicago, will output be increased if we admit this competitor to the club?  And I think that that's a large analytical hole in Scalia's reasoning.  It's too bad that he didn't consider that because I think that's the critical issue that the essential facilities doctrine presents as a matter of antitrust jurisprudence.  Does the adding of a competitor through a mandated access remedy have any impact on consumer welfare.  Does it expand output?  And if not, it's not really an antitrust issue unless we're defining antitrust in terms of competition fairness.

Tad, you look like you're ruminating.

MR. LIPSKY:  I'm going to think out loud now, which is really dangerous, but I wonder, Greg, if you--what you would think about the idea that perhaps this was almost an assumption of the analysis.  In other words, maybe Justice Scalia regarded it as so clearly established, at least as a matter of antitrust jurisprudence, that you had to -- [tape ends].

MR. GERADIN:  -- whereas, the other aspect of efficiency would probably, you know, take a longer period of time to show.  So that's just one comment I wanted to [inaudible].

MR. SIBLEY:  Yes, this is not actually directly on point for too much of what we're saying here, but perhaps indirectly it is.  Last week, the FTC and the division sponsored a three-day workshop on mergers following the release of some information about how the Trade Commission and division actually handle mergers, what types of Herfindahl indices and changes in Herfindahl indices tend to set us off, and how many cases involve hot documents, and which industries seem to make us a little more aggressive than other ones.

At any rate, at a couple of points in that workshop, participants said that antitrust regulation, a lot less intrusive than regulation regulation, could be still better if a sort of mindless approach were taken, that is, that bright lines, even if they're the wrong bright lines, probably are going to lead to better outcomes than trying to do it right.

Now, I sort of pushed one of the folks on this a little bit, and I said, well, by bright lines surely you mean a safe harbor; you don't actually mean no if it's below the line and yes if it's above the line.  He said, no, actually that is what I mean.  And, you know, the argument was kind of an interesting one.

Suppose that you allow a few too many mergers to happen because your bright line allows some mergers--some bad mergers to take place.  The grand sweep of history suggests that, in fact, blips in market power tend to get ironed out, maybe not as fast as the one- or two-year time period the guideline talks about, but, you know, they tend to be.  And if you allow too few mergers, if there is really a big benefit to the merger, sooner or later it will be done by external expansion from one or both of the parties.

Now, obviously, there will be cases where those two things don't happen, but in any case, a bright line has the advantage that things will work a lot faster.  And it's certainly true that simply the fact that mergers take a while to work through does affect whether mergers that ought to take place don't take place at all.  Timing is very important in many mergers and acquisitions.

Perhaps as you go back to Europe, you might keep that in mind, that sort of trying to get it right starts getting it wrong fairly quickly, at least in the view of a lot of the panelists on that workshop.

MR. SIDAK:  Use the microphone, please.

MR. TILTON:  My name is Mark Tilton (ph).  I'm a professor at Purdue University.  I had a question for Mr. Geradin about, in thinking about Germany, which is the biggest economy in Europe, and correct me if I'm wrong, but last year there was a decision by the German Government to expand the authority of the telecommunications regulator by putting energy as well under that same regulator.  And just from what I read in the press, the process was essentially that all of the important big political parties in Germany, the Social Democrats, the Christian Democrats, the Greens, thought that the telecommunications regulator had done such a great job with telecommunications that energy should be moved--essentially moved from the cartel office to the sector-specific regulator, which looks to me like a big move in favor of moving towards sector-specific regulation away from general antitrust, that the political decision in Germany was the general antitrust approach by the cartel office isn't as effective as what we're seeing that the telecom regulator is doing in telecom, and let's move energy there as well.

MR. GERADIN:  Well, there is one point that you do not mention, and it's probably because, I mean, you really have to look very deeply into recent legislation from the European Union.  It's that the problem for Germany is that they decided a few years ago not to create an energy regulator, thinking that it could be done by the Bundes-(?).  And that created a problem because since we don't have a FERC, we don't have an energy regulator in the European Union, again, like in telecoms, this is done at the national level, the strategy proposed by the European Commission was to force these guys to cooperation, in other words, to have network of energy regulators meeting from time to time.  There's a so-called Florence Forum for Electricity and a Madrid Forum for Gas, where these guys meet and agree on things, especially for, you know, interconnection-related problems like interconnectors, as we call them.

But then Germany didn't have a regulator, and it created all sorts of problems for the other regulators because they had no German counterpart.  So there's a directive that was adopted last year which requires member states to set up an energy regulator.  At least you could read the provision that way.  It requires that regulation in electricity be carried out by an authority that would be, you know, chosen by the member states.

So I think rather than, you know, a recognition of the great work done by the telecoms regulator, it's simply that the German Government has to comply with the new set of rules, and, therefore, instead of perhaps creating another regulator, they would think that, you know, giving more competence to an existing regulator would be a good idea.

As a matter of fact, now that you're asking this question, I think I'm in favor of having agencies covering several industries.  As a consultant in several--as a World Bank consultant in several, you know, emerging economies, we basically--I basically agree--argue that it was better to create one regulator that would oversee energy, transport, and telecoms, rather than having several distinct regulators, in part because I think it addressed the capture question.  If you oversee many different industries, it is less likely that you'll be captured by one of--you know, by operators in these industries.  But also because if you don't have that many experts--I mean, let's face it, lawyers with expertise in network industries are, you know, not that many and are highly demanded.  That's at least the case in Europe.  And, secondly, for the economists it's even worse.  We have very few outstanding experts.  It's better to put all these guys in the same authority and have them work together.

Finally, I think there's perhaps not so much a physical convergence between the different industries, but an intellectual convergence.  Let's face it, energy, telecoms, and transport is always the same story.  It's about, you know, issues of cross-subsidization, access to the network and so on and so forth.  So I think that there's quite a bit of--you know, you can do quite a bit of cross-fertilization by putting these different industries under the same roof.

Of course, there are downsides that I could talk about, but maybe you want to make a comment, or somebody in the room does.

MR.           :  Well, I'd just say it's hard for me to tell--what you say sounds appealing.  It's hard to tell, without knowing a lot more than I do, whether it would actually work out that way.  For example, at some point, even if you have an overarching regulatory authority, you'll have divisions or whatever you want to call them that will do industries, and that just means the capture issue moves to that level.  But I do think, you know, having one overarching regulator might well be an efficient way of getting a single view of competitive problems in antitrust issues diffused.  So the extent there's uniformity--that uniformity has benefits, that may be the way to do it.

MR. SIDAK:  Let me relate an experience that I had in Australia.  In about 1994, I went to Australia because they were debating whether to create industry-specific regulatory agencies for energy and telecom and other industries or, alternatively, to create more of an antitrust authority that would have jurisdiction over these agencies.  And I argued at the time, partly because of concerns of agency capture, and out of concerns for cross-fertilization, that the latter would be the better approach.

So a year later, I go back to Australia, not--I don't think because of anything I said, but the Australians adopted the latter model and they created what they called the ACCC, the Australian Competition and Consumer Commission.  And all the lawyers and economists I talked to were busy trying to decide whether this facility or that facility should be declared.  Now, that's Australian lingo for being deemed an essential facility.

So the idea of network industry regulation had now permeated all of antitrust policy instead of talking about whether--I exaggerate, but they were talking about whether this bowling alley in Melbourne should be declared.  So sometimes you have to, you know, watch what you wish for.

We have time for one more question.  Yes?

MR. DEVER:  I have a question for Damien.  Bill Dever (ph) from the FCC.  Under the EC's more decentralized model, since these decisions are supposed to be made by the individual agencies, what happens if a particular country does not apply the competition principle of the EC?  How is that enforced?  Is there an appeal to the EC?  Does the EC, which I suppose now is supposed to be handling more cross-border type problems, say--would the EC then go into Germany and fix the problem if the German regulatory agency didn't apply the principles that it was supposed to?

MR. GERADIN:  Well, I think it's a very good question, and there are two answers to this.

The first one is that if a member state or a telecommunications agency fails to comply with the principles that are found in the EC legislation, they will be declared noncompliant, and the Commission will start litigation at the European Court of Justice to force them to comply with EC legislation, which has a priority, you know.  So that's the first answer to that.

Now, the second answer is that, of course, there is a danger--a big danger of a decentralized system is that you have authorities taking decisions which contradict each other, like, you know, the Germans taking an approach which is very different from the French approach, where actually the same players are operating on both markets and you have a number of cross-border transactions.

Well, the answer of the European Commission was to create a network of telecommunications agencies whereby these guys have to talk to each other.  They have to meet.  They have to agree on things.  And this network will be dominated by the European Commission, which at some stage can veto a decision of one of the telecommunications agencies, or at least ask this agency to postpone the decision so that, you know, there is more time for building consensus on this thing.

Of course, this network idea is, again, a very appealing and very beautiful idea, but we don't know yet whether it will work or not.  By the way, a big reform was recently made in the European Union about enforcement of antitrust rules.  There's a new regulation called Regulation 1-2003, which, again, you know, is based on this idea of decentralization whereby most of the basic antitrust work would be done by national courts and national competition authorities.  And here, again, there's a danger of lack of coherence, and the answer of the Commission is, again, to propose the creation of a network of competition authorities.

The big difference there is that the Commission at any one time can decide to take over a case and rule on it, and it will do so, I think, every time there is a major interested state, you know, either because it's a new theory that is being developed or because there is a real cross-border problem.

MR. SIDAK:  We've come to the end of our session.  David Sibley gets the last word.

MR. SIBLEY:  I just had a question of Damien here.  I gather from what you're saying that the substantial market power doctrine is being superseded by a dominance criterion?  Is that right?

MR. GERADIN:  Well, that's right, yes, yes.  In fact, before--

MR. SIBLEY:  [inaudible] question following on.

MR. GERADIN:  Okay, so the answer is yes.  Before, the trigger was 25 percent.  Now the trigger is dominance.

MR. SIBLEY:  Okay.  Now, one of the odd things when the trigger is 25 percent is you can imagine three firms all having substantial market power in the U.S.  That would, you know, more likely than not get approved.  With a market dominance criterion, would it be possible to have more than one firm be thought of as dominant?

MR. GERADIN:  Yes, absolutely, because we refer to dominance either by one firm or several firms.  And one of the most popular theories of the European Commission, as you may know, is this idea of collective dominance.  And collective dominance, I mean, I don't know what to think about it because I haven't really spent the time doing research on collective dominance.  But, I mean, it's, of course, a very appealing approach when you have, you know, oligopolistic markets like in mobile telephony where you have typically four or five players and all these guys look at each other, they sell the same services, very often at the same price.  So the answer is yes.

MR. SIDAK:  All right.  Well, we're adjourned.  Thank you very much for coming.

[Applause.]