FEDERAL COMMUNICATIONS COMMISSION In the Matter of: ) ) ROUND TABLE ON THE ECONOMICS OF ) MERGERS BETWEEN LARGE ILECS ) Docket No. CC-98-141 HELD ON FEBRUARY 5, 1999 ) ) LIVE VIDEOTAPE PROVIDED TO ) HERITAGE REPORTING CORPORATION ) ON FEBRUARY 8, 1999 ) LIVE TAPE The following transcript was transcribed from an audio cassette tape provided by the FCC from a Live Videotape provided to Heritage Reporting Corporation on December 8, 1998. Pages: 1 through 150 Place: Washington, D.C. Date: February 8, 1999 Before the FEDERAL COMMUNICATIONS COMMISSION Washington, D.C. 20554 In the Matter of: ) ) ROUND TABLE ON THE ECONOMICS OF ) MERGERS BETWEEN LARGE ILECS ) HELD ON FEBRUARY 5, 1999 ) Docket No. CC-98-141 ) LIVE VIDEOTAPE PROVIDED TO ) HERITAGE REPORTING CORPORATION ) ON FEBRUARY 8, 1999 ) Federal Communication Commission 445 12th Street, S.W. Washington, D.C. Monday, February 8, 1999 The parties met, pursuant to the notice. BEFORE: WILLIAM ROGERSON ATTENDEES: WILLIAM ROGERSON, Chief Economist DENNIS CARLTON University of Chicago ROBERT CRANDALL Brookings Institute JOSEPH FARRELL University of California-Berkeley ROBERT GERTNER University of Chicago RICHARD GILBERT University of California-Berkeley ATTENDEES: (CONT'D) ROBERT LITAN Brookings Institute ROGER NOLL Stanford University JEFFREY SHEPERD University of Massachusetts MICHAEL KATZ HAROLD FURCHTGOTT-ROTH, Commissioner Federal Communications Commission I N D E X PAGE OPENING REMARKS WILLIAM ROGERSON 4 SESSION ONE: DENNIS CARLTON, UNIVERSITY OF CHICAGO 10 ROBERT GERTNER, UNIVERSITY OF CHICAGO 14 ROGER NOLL, STANFORD UNIVERSITY 18 SESSION TWO: JOSEPH FARRELL, UNIVERSITY OF 31 CALIFORNIA-BERKELEY ROBERT CRANDALL, BROOKINGS INSTITUTE 40 DENNIS CARLTON, UNIVERSITY OF CHICAGO 42 COMMISSIONER HAROLD FURCHTGOTT-ROTH 77 SESSION THREE: ROBERT LITAN, BROOKINGS INSTITUTE 83 JEFFREY SHEPERD, UNIVERSITY OF MASSACHUSETTS 84 RICHARD GILBERT, UNIVERSITY OF 101 CALIFORNIA-BERKELEY SESSION FOUR: MICHAEL KATZ 118 DENNIS CARLTON, UNIVERSITY OF CHICAGO 127 P R O C E E D I N G S MR. ROGERSON: Good morning, and welcome to the FCC round table on the economics of mergers between large incumbent local exchange carriers. In 1996 when the Telecom Act was passed, there were eight large regional telephone companies providing local telephone service, seven Baby Bells and GTE. Since this time the FCC has approved two mergers between these firms, reducing the numbers to six, and right now the FCC is faced with proposals for two more mergers that would further reduce the numbers down to four; specifically Bell Atlantic is proposing to merge with GTE, and SBC is proposing to merge with Ameritech. Well, the two companies resulting from these mergers would control two-thirds of the local telephone lines in the United States. Furthermore, the arguments that proponents of these mergers are making to support them are the type of arguments that if the FCC accepted them might well also cause the FCC to accept further consolidation among the remaining four, perhaps to bring us down to two some people have suggested. Therefore, I think that the FCC really is at a defining moment for telecommunications policy and that the decisions that the FCC is going to make about these mergers could well affect the structure of the telecom industry for years to come. The issues that we have to evaluate, and I was actually getting very close to the climax here, Mike. I promised them all that you would show up at about this point. The issues that the FCC has to evaluate really are very complicated, the FCC has discovered going through all of these. Critics of the mergers have told us at the start well, it is not all that complicated at all. You used to have eight, and you are on your way to two. That is on the face of it anti-competitive. The proponents of the mergers have come back to us and told us well, these six current ILECs basically do not compete with each other at all. They are each in their own region and serving their own people, and there is very little competition between them. They have shown very little interest in competing out of region, and even when they have shown interests in competing out of region they have shown very little aptitude. To the extent that there has been competition in local telephone markets, proponents of the mergers tell us that it has basically just been supplied by long distance companies trying to become local companies as well and by start up companies. Proponents of the mergers have argued to us that most competition in local telephone markets is being provided by long distance companies and by independent start up companies. They argue that plenty of competition is being provided by these types of firms, and, in any event, whether Baby Bells merge or not is irrelevant to the issue of competition in local markets because they do not compete with each other. In fact, they make a stronger argument than this. They claim that although these Baby Bells have been too small and anemic to compete out of region thus far, if only they were allowed to bulk up a little more, in fact, they might become very dynamic competitors out of region and actually increase competition if we allow these mergers. Well, critics of the mergers do not take this lying down. They say yes, it is true that in fact the Bells may not compete with each other that much right now, but there is every indication that they would have begun to compete with each other tomorrow or six months from now or surely by next year. The problem with these mergers, critics say, is not that they are going to destroy competition that is occurring today, but they are going to destroy competition that surely would have occurred but for these mergers. Well, proponents come back and say this is highly speculative, and they do not think it is true. Then we will see what the critics say to that. I think they say they think it is true. We will find out. The issues do not stop here. It turns out there are a number of other distinct possible benefits and harms of these mergers that economists and other interested parties have raised. What I have tried to do today is to organize the round table that we are going to have into sessions that will kind of go systematically through the arguments that have been presented as the FCC sees them and expose each one to the full light of day and have some vigorous argument and debate about each of them. We have a very distinguished panel of economists who are joining us today to help us go through these arguments, and I now would like to take a moment to introduce them. I am Bill Rogerson, and I am the chief economist. Starting to my right is Dennis Carlton from the University of Chicago. He has submitted an affidavit on behalf of SBC-Ameritech in favor of the merger. Robert Crandall from the Brookings Institute submitted an affidavit on behalf of BA-GTE, and he was in favor of that merger. Joe Farrell from UC-Berkeley submitted an affidavit on behalf of Sprint in opposition to the mergers. Rob Gertner from the University of Chicago submitted an affidavit on behalf of BA-GTE. Rich Gilbert from UC-Berkeley submitted an affidavit on behalf of SBC-Ameritech. Michael Katz from UC-Berkeley submitted an affidavit on behalf of Sprint. Bob Litan from the Brookings Institute and Roger Noll from Stanford, these guys apparently could not find anyone to get them to pay them to write a paper, but they went ahead and wrote a paper anyhow, so they have published a Brookings research paper jointly co-authored that is critical of the mergers. Finally, Jeff Sheperd from the University of Massachusetts has submitted an affidavit in support of a consumer group on behalf of a consumer group called the Texas Office of the Public Utility Council that is generally critical of the mergers. The agenda for today. Do people have copies of it? Were they available? This has been a deep secret. None of our panelists do either. Apparently I can change this as I go along. What the agenda in front of me says is we have divided the round table into four equally long sessions. Really what I have tried to do is focus on what seemed to be the four big issues that there has been a lot of debate about. We are going to consider them each one at a time. The first issues are what are the potential benefits of these mergers, and how big are they? The second, third and fourth sessions are going to each focus on separate theories of harm the different parties have advanced that are critical of these mergers, so there are three quite distinct different theories of harms that have been put forward to the FCC that we have to evaluate and deal with. Session Two will be devoted to the issue of whether or not these mergers will affect the FCC's ability to do benchmarking. Then we will have a break. Session Three will be devoted to whether or not these mergers will have effects on actual or potential competition. Finally, Session Four will be devoted to whether or not these mergers will increase either the incentive or the ability of the large ILECs to act anti-competitively against their rivals. Then we will have a very brief concluding session where we can wrap up. The way I am going to run each session is I have asked two to three people to each take five minutes at the start of each session to basically set out the framework of issues for us to give us a basic pro, what the argument is being made, and the basic con, what the structure on the other side is. After we have had each of these people give us a basic framework for what the issue is, I am going to turn it loose to the entire panel, and I hope they will all question each other vigorously and so on. If they are not questioning each other vigorously enough, I am going to turn it loose to the audience. In fact, I am going to turn it loose to the audience even if you are questioning each other vigorously, so I should admit that now. Probably about five or ten minutes before the end of each session I am going to turn to the audience and ask the audience if they have any questions that they would like to address to our panelists. Let's get started then. The first session is on potential benefits of the mergers. Do these mergers have benefits? If so, how big are they, and why do we know they exist? I have asked Dennis Carlton, Robert Gertner and Roger Noll to each make five minutes of opening remarks. What I will do is when you have one minute left, Dennis, I will tell you you only have 30 seconds left, okay? Go ahead, Dennis, if you would like to start. MR. CARLTON: Thank you. It is a pleasure to be here. The merger of SBC and Ameritech will create a national competitor that can quickly provide a broad range of services for both residential and business customers. The benefits of stimulating competition in the provision of these services are undeniable and large. SBC plans to offer a wide variety of services, including local, long distance, Internet and customized data services for both residential and business customers in a one stop shopping environment. The national/local plan is SBC's response to the rapid changes in demand and supply for telecommunications services. As far as I am aware, no one has seriously disputed that the national/local plan is a sound business strategy whose implementation will significantly increase competition. The validity of the national/local plan is confirmed by the fact that the other major providers of telecommunications services are heading in precisely the same direction, the most prominent being the three major interexchange carriers, MCI WorldCom, AT&T and Sprint. It is no surprise that the objections to this transaction principally are being made by the very firms that SBC plans to challenge in the marketplace. Opponents claim that the national/local plan should not be considered a merger specific efficiency. They have made two basic arguments which are glaringly inconsistent. On the one hand, they claim that SBC will not really carry out the plan. MCI WorldCom goes so far as to call the plan a rouse. On the other hand, opponents also claim that the plan is not merger specific because either SBC or Ameritech would carry out similar plans absent the merger. Opponents are wrong. The claim that SBC's commitment is not credible can be easily dismissed. There is simply no reason to believe that SBC would willfully risk misrepresenting itself before consumers, investors, Congress and the FCC. Just yesterday on the front page of the Boston Globe there was an announcement of how SBC would offer new local service in Boston. The claim that SBC or Ameritech would carry out the plan in the absence of the merger is simply unsupported speculation. The issue is not whether SBC could finance the plan by itself. The issue instead is whether in the absence of the merger SBC would have the necessary economic incentive to undertake such an aggressive plan in such a short time. There is no evidence whatsoever to support the position that it would be profitable for SBC to undertake the national/local plan absent the merger. Acting alone, SBC would face higher costs and greater risk of failure in pursuing the national/local plan than under the merger. For example, either firm would need to deploy more managers proportionately, more engineers. In the absence of the merger, they could not carry through with their follow to headquarters, follow to a home customer plan, as easily as they could after the merger. In the absence of the merger, the plan would be less attractive financially, and it would be perfectly rational for SBC or Ameritech to decide not to pursue this risky strategy. Given the race now underway to offer packages of services on a nationwide basis, delays in establishing a national footprint translate into a reduced likelihood of a project success and, therefore, the reduced likelihood that a project of this scope and speed would be undertaken. The fact that mergers can create a national footprint should be no surprise. WorldCom's acquisition of MCI and AT&T's acquisition of TCG had similar motivations to accelerate the deployment of packages nationally of end to end service. Furthermore, in addition to the national/local plan, the merger is expected to bring additional efficiencies. SBC expects to realize significant savings from the Ameritech transaction, including more than $1 billion in annual cost savings by 2003. SBC has a proven track record in achieving projected cost savings. In the Pac Tel merger, they are ahead of schedule in achieving more than $1 billion in annual cost savings by the year 2000. In sum, the substantial benefits from this merger are indisputable. MR. ROGERSON: Dennis, you have 30 seconds. MR. CARLTON: Okay. MR. ROGERSON: I am not joking. MR. CARLTON: In sum, the substantial benefits from this merger are indisputable. The merger creates a more potent national service provider for business and residential customers. The notion that this merger should be stopped because someone hopes that each company would on its own embark on a similar plan is faulty. Consumers should not be deprived of the benefits of this transaction on the basis of unfounded speculation. Thank you. MR. ROGERSON: Thanks, Dennis. Now Rob Gertner from the University of Chicago. MR. GERTNER: Thank you, Bill, for the opportunity to participate in today's round table. I look forward to discussing the economic impact of these mergers. In my remarks, I will address the pro-competitive benefits of the mergers, focusing on the Bell Atlantic-GTE merger. The telecommunications market is changing rapidly. Deregulation and new technology are transforming the industry. Not surprisingly, other industries facing such fundamental shifts have seen major changes in the identity, scope and scale of competitors. These changes are characteristics of deregulated industries, such as airlines, trucking and energy, as well as technologically dynamic industries such as computer software and hardware and telecommunications equipment. Many of these changes include significant consolidation through mergers and acquisitions. Competitive adaptation to such a change in environment is fundamental for achieving economic efficiency. This is especially true in industries such as local telecommunications where the geographical and product scope of the companies has been determined by regulation rather than market forces. Certainly proposed mergers must be analyzed carefully by regulatory authorities for potential anti-competitive effects, but regulators should be mindful of the value of competitive responses to a changing environment. These mergers are between large companies. Although this may make some people worry, it is widely accepted that big is bad is a flawed way to think about mergers. Instead, we must evaluate carefully the likely impact of the mergers on competition and consumers. Opponents of the mergers present a variety of objections to both proposed transactions, but their economic arguments lack empirical support. A careful analysis of the institutional and competitive environments in which these firms compete show that the opponents' concerns are not economically significant. On the other hand, the pro-competitive strategic rationales for the mergers are strong. The most significant benefit from the Bell Atlantic-GTE merger follows from two simple premises that are widely accepted by all parties, including regulators and companies opposing these mergers. The first premise is that the ability to provide facilities based bundle services on a wide geographic scale is an important strategic asset for telecommunications providers. Indeed, the major opponents of these transactions are pursuing similar strategies in similar ways by acquiring firms that are allowing them to offer portfolios of telecommunications services on a national or near national basis. For example, AT&T has recently completed several major acquisitions and announced a new business strategy based on offering bundled telecommunications services. The FCC, in proceedings on these mergers, has acknowledged the importance of bundle services, and the pleadings include statements from many business customers that they value such services. The second premise is that existing customer relationships provide an important competitive advantage in the evolving market. Wide ranging evidence supports this view. The evidence includes the cost incurred by interexchange carriers and wireless carriers to induce customers to switch service, the difficulty GTE has had in selling services out of its local exchange region, consumer surveys and the strategies adopted by numerous companies to sell new services to their existing customers or to make acquisitions to gain access to an expanded customer base. The Commission also agrees with this premise. For example, in the Bell Atlantic-Ninex Order, the Commission argued that the major interexchange carriers are among the most important potential competitors in local markets because of their existing customer bases and brand recognition. The merger of Bell Atlantic and GTE will have significant pro-competitive benefits. GTE's GNI and Internet backbone and Bell Atlantic's customer base are strongly complimentary assets. The combination of these two assets will create a strong facilities based bundle services competitor. Furthermore, the merged firm will use GTE's existing presence in or near many geographical dispersed markets to facilitate timely and efficient entry. The benefits to consumers will include the presence of another national or near national provider of bundled telecommunications services. This increased competition should result in lower prices and greater consumer choice. Businesses will be able to receive the same set of advanced services at all locations. They will be able to coordinate upgrades and service throughout their organizations with a single provider that understands their telecommunications needs. Consumers will be able to reduce transaction costs and coordination costs by having a single provider. MR. ROGERSON: You have 30 seconds left, Rob. MR. GERTNER: In addition to these benefits, the merger will result in significant cost savings. Bell Atlantic and GTE estimate that the merger will lead to $2 billion annual cost savings within three years of the merger. There is an important reason to not be skeptical about these benefits, given the experience that they have had in their previous mergers in meeting these targets. The pro-competitive benefits of the merger is clear. It would be unwise to forego these benefits because of potential harms that are unlikely and for which there is no empirical support. Thank you. MR. ROGERSON: Thanks, Rob. Roger Noll from Stanford University. MR. NOLL: I am still in awe of the new building. What did you guys have to do? Usually in the Silicon Valley when you walk into a new building like this you know that the company is successful and about to have an IPO, and can I buy stock. I do not have a prepared statement because I did not have to clear it with any lawyers. That is actually one of the advantages of not coming here representing somebody. What I want to do is step back and say how should we think in general about merger policy in the context of the past 15 years of history in the telecommunications industry and then put these things in perspective, the arguments about benefits. It seems to me that the entering premises here have to be two. The first premise has to be in the best of all possible worlds, if a firm believes it has a superior business strategy and wants to undertake a series of agglomerations, whether horizontal, vertical or adjacent, to achieve that business strategy we would normally just get out of the way and let them sink or swim on placing their own bets. That is the whole point of having a decentralized market based system. Then we ask the question what is there about the history of telecommunications that might cause us to say this is not the right way necessarily to think about the problem? That is to say that we might want to look beyond the kinds of statements we have just heard about the largely firm specific benefits that would arise from this activity. Now, of course, that is not necessarily bad because the premise is the first specific benefits arise because they are somehow more able to please consumers. Therefore, they make more money by doing good, as well as by doing well. The answer here is quite simple, and that is that we have a history of precisely these arguments defending the presence of ubiquitous monopoly in the industry not only in the U.S., but everywhere in the world. That is to say that the history of this industry is one in which we have been told throughout the lives of everyone in this room that the nature of this industry is you are better off if there is just one guy out there who does everything, and you just sit back and do not worry your pretty little head about which particular alternative is offering better services, that it is just something that you should not worry about. We know what you want, and it is to get the complete bundle of telecommunications services from a single ubiquitous provider, which is a different argument than the national monopoly argument. It is an argument about complexity, information impactedness and, on the supply side, the integratedness of the whole telecommunications enterprise. We have a long history of looking back at things like the introduction of competition in the U.S. and other countries, things like the divestiture which created seven RBOCs instead of one, all of which prior to the act being taken were predicted to impose substantial cost. The flip side of the argument about benefits is that the introduction of competition and the introduction of divested RBOCs in multitudinous numbers was that there was going to be a big, positive cost impact. Indeed, the majority of state regulators at the time immediately following divestiture gave emergency rate relief to the RBOCs on the grounds that the act of undertaking divestiture was going to make them less efficient. These then emergency rate reliefs, within a matter of about 18 months, were all rescinded because they were unnecessary. The is the first important background point. As a subsidiary of this, the only point that I think one needs to keep your eye on all throughout this argument -- MR. ROGERSON: Roger, in 30 seconds -- MR. NOLL: Right. MR. ROGERSON: -- I am going to tell you that you have 30 seconds. MR. NOLL: Okay. MR. ROGERSON: I think you needed this extra warning. MR. NOLL: Bill, I can still rescind that Ph.D. Now that I thoroughly lost my train of thought, the right level of analysis is the industry, not the firm. That is extremely important to bear in mind. The issue is what is happening to consumers in all markets, those who want to bundle their own packages, as well as those who get them by themselves. What is happening to the cost of the firms that have not merged, as contrasted with the firms that have? We know that the long run cost trend in local carriers is in fact that real costs are declining. To say that the merged entities have had lower costs since the merger is not to say very much in an industry where costs are falling. I am not saying it is not true. I am just saying the right level of analysis is not the firm. It is the industry. Finally, one more point before Bill gets here, which is think about this in the context of the Telecommunications Act. If it is the case that the vision of the Telecommunications Act of having ubiquitous, vertically integrated competition is true, then within a few years we are not going to care about these mergers. Within a few years, if these benefits are real, the companies that will succeed will be the vertically integrated ones, and we will not care if mergers take place. If the vision of the Act is not true, then we are going to care a great deal if we in fact de facto recreate the old AT&T and undo all the competition and all the benefits from the competition that we have observed in the last 15 years. MR. ROGERSON: Thanks, Roger. Okay, guys. Let me turn you loose. Rich, go ahead. MR. GILBERT: I just want to respond to Roger, if I may. This is not about creating a monopoly. This is about creating firms that can compete on a national and on a global scale with other integrated telecommunications providers, such as AT&T-TCI-Time Warner, Alliance and MCI and Sprint, and the global players as well, the French Telecom and British Telecom and NTT and such. This is the playing field that we are dealing with. This is not about creating a monopoly. On the merger benefits side, I just want to focus on the in region benefits, since others have talked about the national/local strategy out of region benefits to say that this is not a speculative analysis. This is based on evidence, and the evidence is that we have a track record now from the acquisition of Pacific Telephone, Pacific Telesys, by SBC, and we have for this merger we have $1.4 billion annually in projected cost savings. At SBC-Pac Tel we had something similar, about $2 billion, and the evidence right now is that those efficiencies are right on track. We have $50 million in annual cost savings on tandem and trunk design. We have $88 million in operator services, $134 million in directory publishing. These are coming in proven, demonstrated efficiencies in region, so we do not have to speculate. This is fact. MR. ROGERSON: Okay. Rich has said that if you have two comptrollers you can fire one, etc., and there is real efficiencies then to having two firms combine into one, right; just plain, old horizontal efficiencies because there are all sorts of things you do not have to do twice. Is that true? Is there evidence to support it? Is that a significant reason for allowing this merger? Bob? MR. CRANDALL: I think I agree with Roger that you cannot find evidence of the technical economies of scale in something like that that would drive these mergers; that in fact there has to be something else. I think what is driving it is a desire to reach out and become a national presence and compete against the Sprints and AT&Ts of this world, but we should not underestimate the impact of these mergers in creating efficiencies; that is, shaking out inefficiencies. Right within this room there are people who have been estimating cost models that show that forward looking cost models give you much lower cost than the embedded costs of these companies. There is a reason for that. These companies have been subject to numbing regulation for decades, and that is part of the problem. In every industry that I know of that has gone through a major change of the increase in competition, whether it be from foreign trade or from deregulation, there has been a huge shake out of management. I remember in the airlines we had to retire Frank Bohrman to sell used cars in Arizona. In the case of the steel industry, the only large steel companies that have survived and will survive are the ones who brought in new management. You need to shake up these organizations. You get enormous efficiency gains from doing it, and one of the ways you do it is through merger. Even in this industry, AT&T has not really gotten its act together until it brought in a CEO from outside. I think you really need to allow these firms to sort things out. You need to allow these mergers in order to get these efficiencies. You have gotten a lot of them so far, and I think you can get a lot more. MR. ROGERSON: Bob? MR. LITAN: Is this on? Can people hear me? Yes? Okay. On efficiencies, and then I want to go back to potential competition in the national plan. You can get a change in management without a merger, so I do not see how the merger is a condition for achieving these efficiencies. We will still have numbing state regulation even after these mergers if that is causing inefficiencies. I am going to go back to the local plan and address several points that Dennis raised. First, he said that the firms I guess in the case of both of them, that they will both have enhanced incentives by merging the two of them to enter out of region. To me, that is not self-evident. I would like to see that explained, especially in light of my second point, which is you would presume that those incentives were operational after SBC bought Pac Bell, and I did not see the combined SBC-Pac Bell running around with a national plan. Why is it that they need another RBOC to make the national plan a reality? The third point that Dennis raised is well, why would SBC misrepresent itself about this plan? I do not think anyone has to accuse them of misrepresenting. The fact is that plans change all the time in this industry. AT&T rolled out a resale plan, only to basically rescind it and switch strategy to go to the cable strategy, realizing that resale at least was not going to work. The point is that the FCC in advance has no way of knowing how credible any particular promise is. The promise may be credible at the time. It may be advanced in full faith, but it may change because business strategies change. After all, in the wake of these mergers SBC may say look, we want to spend all of our attention concentrating on in region, and we will get to outer region later. They could do that maybe four or five months after the merger, which leads to the final point, and that is I was trying to think creatively. The FCC could take SBC at its word and say all right, if you are going to promise 30 new cities, we will make that a condition of the merger. All right. I mean, SBC in a way is taking an enormous risk by making this plan and inviting the FCC to attach this condition. In my outline that I have handed out and I will address later, I point out that there are problems in imposing such a condition, and that is it is hard to operationalize. SBC could enter and then withdraw six months later and say well, we tried; it did not work, so there would be no way to enforce it. Alternatively, you could impose a hold separate Order on the companies and then say well, you can basically merge after you go ahead and enter, but holding separate for several years may eliminate the efficiencies and probably eliminate the appetite of the management for doing the deal. All of this is to say that the FCC should not dismiss, though, out of hand the notion that you attach this as a condition if in fact SBC is really serious about this, but I question the premise to begin with. MR. CARLTON: May I respond? MR. ROGERSON: Yes. Right. Jeff Sheperd wants to take a shot at you, too, but first I would like to have you specifically respond to the question why is it that SBC-Ameritech will have an increased incentive to enter and pursue a national strategy after this merger? MR. CARLTON: Okay. I would like to actually respond also to something related that Roger said. I would like to make three points. First, there is a distinction between ubiquitous monopoly and one stop shopping. If you have many people providing a bundle of products, you do not call that ubiquitous monopoly. I would call that competition among many people to provide a bundled product that is desirable, and that is what seems to be going on in people's pursuit of a national strategy. Second, a merger like SBC's that is promising competition out of region strikes me as the antithesis of a merger creating a monopoly. Now, to go directly to Bob's two points, what are the increased incentives -- MR. ROGERSON: Get to Bill's one question, too, at some point. MR. CARLTON: Well, the first point was yours, Bill. What are the increased incentives as a result of this merger? I think there are two or three points here. First, there is absolutely no evidence that absent the merger, either company on its own would undertake the same investments, the same project. Now, what are the increased incentives? The national/local plan is based on the premise that following your customer gives you an advantage. What does following your customer mean? You have customers with headquarters in your territory. You have relations with them. You can then follow them. If you are merged, you have more customers in your territory that you can then follow. Therefore, you have more customer contacts. That is the first point. It is a slightly different strategy than what the other say three large IXCs are following. It is a different strategy, and it would be materially affected if there were not a merger. Second, and in fact I have a table in my report that shows precisely that if each company tried to do it on their own, say if Ameritech tried to expand into 15 extra cities, it would only cover what they believe is around 31 percent of their in region customers, while with a merger if they could merge and then expand into the national/local plan they will cover 80 to 81 percent of the right coverage for their in region customers, which then makes it easy for them to expand out of region. That is the first point. The second point. Why is it easier? If you look at what happens when a smaller company has to expand rapidly compared to a larger company, if you require a proportionately larger increase, which you would if SBC on its own had to undertake the same national/local strategy of going into 30 cities, that is a much larger proportional increase. You would have to expand more managers, more engineers. It is more costly. We know adjustment costs rise with the speed of expansion. What does that mean? That means that the speed with which this strategy will be deployed is certainly going to be much slower if you require the companies to do it on their own. What does a slower strategy mean in a race to capture customers with switching costs? It means the strategy will be likely less successful, so you have lowered the profitability of the strategy, and you have reduced the speed of the policy. Both of those are reasons why the incentive is greater and would bring benefit immediately to consumers. This is precisely why you saw the large IXCs or one reason why you saw the large IXCs while they are involved in mergers and acquisitions in order to build up not de novo, but take existing assets, put them together quickly in order that they can get a national footprint. You could have made the same argument with respect to those acquisitions. The point is when you start with an inefficient industrial structure dictated by regulation, not by marketplace efficiencies, you don't have the efficiently sized firms, and acquisitions can get you to that place quicker. MR. ROGERSON: Okay. Wait one second. I just want to summarize in 30 seconds what I believe the answer was. If SBC merges with Ameritech, the business plan for entering out of region will be more profitable than if SBC did it by itself. Now the question I want to ask all of these panelists is will the business plan for SBC to enter out of region be profitable; maybe not as profitable, but still profitable? MR. CARLTON: At the same speed? At the same speed? MR. ROGERSON: Joe, go ahead. MR. FARRELL: Good morning, everybody. We are present today at a historic occasion because -- MR. ROGERSON: I think you are not answering my question, Joe. MR. FARRELL: I am. I am. We are present at a historic occasion because Dennis has just revolutionized merger policy by giving arguments which prove very generally that creating larger firms always enhances competition because it is easier and more profitable for a larger firm to expand and take more customers than it would be for the component smaller firms to do so. The argument seems to prove a bit too much. Now, one way of thinking of all this is to step back and say what is the right geographic market definition for analyzing this kind of merger? Clearly I think as far as most customers numerically are concerned, the big issue, of course, is the last mile bottleneck, and that is an issue that is defined in the customer's own local market. The claims about the national/local strategy and wanting to become a national, if not international, competitor are really a statement that it is important also, or perhaps even instead, to analyze a market for customers who have multiple locations and so the question then becomes phrased in that way, and this is just a rephrasing, but I hope it is a useful one. If you take that seriously, and obviously you have to also remember the other customers, but if you take that seriously should we say because the geographic market scope is now national, if not international, rather than local, should we say that means that it is tremendously important to allow this company or this pair of companies to become better at serving that market? Obviously, yes. Should we say, on the other hand, that it is really important not to allow all of the potential entrants into that national market because there is nobody serving that national market yet, to remain potential entrants into it? We have a situation where there are zero companies currently who are in this alleged geographic market. You have a number of potential entrants, namely those who are in the smaller geographic markets and could expand. The question is how do you think about a merger as a step towards having somebody in that market, which at the same time reduces the number of firms that could come into that market? MR. SHEPERD: Since I was promised or mentioned at some point -- MR. ROGERSON: Right. Okay. Go ahead. MR. SHEPERD: I am Jeffrey Sheperd. I am much the oldest person on this panel. I have seen regulation working for so many decades. I would like to speak as a colleague, not as another advocate among this group. The economists use usually much the same logic, all of us. It is just the judgements of amounts that differ more or less. That is why we get into these traps of advocacy arguments, speculation against speculation. I would like to suggest instead that we think of a different concept or, as Roger suggests, get a bigger perspective on these issues. The points I would like to stress at this point are two. One, the sector is in a very unstable period right now, and companies rightly have a sense of risk that they are about to be blindsided by somebody else's merger if they do not do theirs. That is really the FCC's economic problem. They are facing a cascade of mergers, some of which I think are plainly irrational, but driven by fear, driven in an arms race. There is actually a literature on arms races which would help understand why we are here and why, as you say, this is a critical moment. If the FCC draws the line now and says let's hold back, we will not let anybody merge at least if they dominate their markets for awhile until competition is established on the strength of these powerful companies, then, as has been said, it will not matter once competition is going, mergers. We can let them rip. I am afraid if we spend instead the time today debating this specific merger or any of them, of course we will match two sided speculation against each other, and we will not get anywhere. I think we all know most of the details of both of these plans. Instead, the FCC needs some more collegial economic guidance on how to sort things out. MR. ROGERSON: Richard? MR. GILBERT: My first suggestion is that we should put these things up if we do not want to speak. (Laughter.) MR. GILBERT: It would be more efficient. Just a standard of convention. I actually just have a brief comment, which is on what happened after the SBC-Pac Tel, why, you know, did we not stay out of region entry there. The fact is that SBC did have a plan around that time, and their plan was to use the wireless platform of their out of region wireless services as a platform to enter into local service. They tried that in Rochester, and it was extremely unprofitable. It did not work at all. I think you can see that that is a need for change, a need for revaluing how they are going to deal with a national entry strategy. MR. LITAN: But does that not discount then any promises now? MR. GILBERT: Well, it is true that there is change. Of course there is change, but -- MR. LITAN: It just says that the FCC should not give a lot of weight to that. MR. GILBERT: -- they tried it, and it did not work. MR. LITAN: I mean, that point says if I was the FCC, I would not give a lot of weight to that. MR. GILBERT: You have to evaluate each proposal. What it is saying is that what we thought might work, which would have been an easier strategy with need for less capital, less scale, did not work. You have to go to the next step. MR. ROGERSON: Michael? MR. KATZ: Now I feel like I should go out in the audience and do a bad TV talk show. Maybe that is what this is turning into. Jerry Springer will be for the afternoon session. We will get to see the real action. I just want to comment on a few things that have been said before. When Bob Crandall talked about looking at the internal efficiencies or shaking out the inefficiencies, and he talked about shaking it up through merger as Bob Litan addressed, he also mentioned in passing saying that competition is a great way to shake things up, and I agree with that. I think that experience in a variety of different industries shows that, which I think, though, is not something that argues in favor of these mergers, but it is a source of concern. I would just note for now it is something we should come back to. When we discuss the loss of benchmarks, I think the focus in that session will be on the loss of regulatory benchmarks, but there is also the loss of benchmarks, this is competitive benchmarks, used by industry. I think when we are going to talk about internal efficiencies we need to think about it there. Similarly, when we get to the session on the issue of local competition, we are going to have this same issue. I do not want to go into it now. Some people have already started the later sessions. That is something that we need to note, and it is by no means clear that this is an argument in favor of the mergers. We will hear more about that later. I want to build on something that Joe Farrell said about arguments in favor of these mergers being too strong. I think the question maybe to ask everybody, particularly the proponents, though, is do they advocate having a single ILEC for the country? Do they think it would be fine to put the Bell system back together, and we can debate the part about long distance, but at least at the local exchange level? It seems to me there are two reasons to ask that question. One is I think it brings out the positions more sharply. Second, it may be a relevant question for the Commission. I think there is going to be the question of after these mergers, if they are allowed, then the next merger will come along, and the argument will certainly be made well, it is not that big a deal, given the mergers that have taken place. In fact, if nothing else, they should break up the remaining ILECs into little pieces and just merge one little piece at a time, each one being not that big a deal. In the end, of course, it is going to be a huge deal. It seems to me that is something that needs to be thought about. Finally, I just wanted to mention something else. Joe was talking about local markets and the focus on customers and to what extent there were national customers. I think it is important to recognize. He was talking about end user customers, but also what is going on is there are a lot of carriers that are customers. We keep hearing about how Sprint and AT&T and MCI are the rivals of the ILECs that are proposing to merge, but, of course, the other thing that needs to be taken into account is they are big customers of them. They are quite concerned about buying access services at a national level. Now, obviously they cannot do that from any one ILEC at this point, but there is a relevant national market there, or at least it is something that needs to be taken into account at the national level, and that is true even if the individual customers are not national. MR. ROGERSON: Okay. This is the moment we have all been waiting for. Questions from the audience? Would anyone from the audience like to ask any of our panelists a pointed, witty, thoughtful question? Okay. How about other questions? I think there is someone at the back. Stan Newman from the FCC? MR. NEWMAN: Yes. Could the panelists who are representing the merging companies explain how much cap X you plan to spend out of region and how many subscribers you think that cap X will capture for you? MR. ROGERSON: Who would like to handle that? Okay. None of them will. Roger? MR. CARLTON: Well, I can say that those figures are in the testimony, the exact numbers. The plans have been filed. My understanding is the initial investment that they are talking about is something around in SBC's case something like $3 billion plus all the expenditures on managers and engineers and the like, and also that does not include all the expenditures that have been made to date. My understanding of the penetration they hope to achieve, which was your question, in those cities that they are in, you know, they do have projections over time. The exact numbers I cannot recite off the top of my head. I think they are somewhere between penetration rates of ten and 15 percent in those cities. MR. GILBERT: I mean, the numbers I have seen are $23.5 billion in operating costs over ten years, I think $2 billion in up front expenditures, 80 switches in 30 cities, not counting 14 foreign locations and 2,900 miles of fiber internally in the U.S. MR. CRANDALL: Let me just make one point. We are here to discuss the economic issues in this case. We are not here with fiduciary responsibilities to the stockholders of these companies. Therefore, we do not know precisely what the capital expenditure plans would be, nor can we attest to them over a long period of time. MR. ROGERSON: Okay. Are there more questions? Pat DeGrabi? MR. DEGRABI: Thanks. Here is an economics question about the national/local strategy. The theory of the national/local strategy is basically one of incumbency advantage. Following the customer out of region means because I serve the customer in region, say Dallas, I am more likely to serve them in Atlanta for whatever reason it is that you want to announce. The question here is how big is that incumbency advantage? The argument was made earlier that we ought to think about ways of measuring some of the variables involved here, and I would argue that we have a potential test of how big incumbency advantage is, and that is all of the RBOCs are incumbents, and they are all competing against entering CLECs. Ameritech has been very proud of the fact that it has actually lost a measurable number of its access lines to CLECs, announcing that there is in fact competition, and this in fact suggests that the incumbency advantage in region is not big enough and that CLECs are competing on a level ground. How can I square the notion that to go out of region I need to have an incumbency advantage by serving one building of that potential customer in region when in fact I see that I cannot even hold that customer in region because CLECs can come in and compete? MR. ROGERSON: Okay. Rich Gilbert would like to explain, as I understand it, why it is that CLECs did not need a whole bunch of existing customers to chase, yet have been doing quite well. MR. GILBERT: Pat, this is not about incumbency advantage. This is about a network advantage. In fact, the concern that the ILECs have is that the interexchange carriers have these relationships, and they have the relationships on a national level and are promoting the ability to provide this end to end competition. They are quite concerned about being able to compete for all their customers. To do that, you have to have a similar presence, so I think it is not about incumbency at all. MR. CARLTON: Can I just add one thing to that? The fact that one person believes that there is some advantage to incumbency while other firms, MCI, AT&T, may have other strategies does not mean that any one of these advantages is absolute. It means that one firm thinks it has an advantage over another firm on some dimensions. Maybe on other dimensions it does not. Therefore, my view is you should let them exploit what they think is their desirable business strategy. Your question is will it be a success? My view is I would let the marketplace determine whether it will be a success. SBC has put in an enormous effort to this national/local plan. MR. ROGERSON: I cannot resist but to ask the question that Bob Litan asked in his remarks to the panelists. Suppose we approve these mergers, and then suppose next year we are here at a panel, and there are two more proposed mergers before us. Bell Atlantic wants to buy Bell South, and SBC wants to buy U.S. West, and they tell us that Bell South and U.S. West are not providing much competition anyhow. They are too little. There would be efficiencies if the mergers occurred, and in fact they could do a more dynamic national competition strategy if we allowed those mergers. Would those be good mergers? More to the point, would the arguments you are making today apply equally well to those mergers? MR. CARLTON: I think the answer is simple. MR. ROGERSON: Yes. MR. CARLTON: I think the answer is you can ask abstract questions, but you would obviously have to evaluate it at the time the merger occurs, and you would have to look at the circumstances at that time. MR. ROGERSON: Yes. MR. CARLTON: I think it is quite clear that right now there is a very clear answer to the question you and Michael asked. Would a merger of all the RBOCs be good? The answer from these two mergers that are proposed is quite clear that these guys are going to each pursue some out of region policy. It is clear it is horizontal, in my view, and -- MR. ROGERSON: Right, but I am not -- MR. CARLTON: -- we would allow it. MR. ROGERSON: -- asking would a merger between SBC and Bell Atlantic be thinkable. I am willing to believe that the arguments that the proponents are making today indicate that that would be a bad idea because you are both claiming you are going to compete against each other. What I am asking you is are the arguments you are making today consistent with making an argument next year that Bell Atlantic should be allowed to buy Bell South and that SBC should be allowed to buy U.S. West? MR. CARLTON: I would say it is not inconsistent and may be consistent. It depends on the circumstances at the time. MR. LITAN: Bill? Bill, can I just add? I just want to go to the moving your customer point. MR. ROGERSON: Okay. MR. LITAN: I just want to make one 30 second intervention. All right. I can understand how SBC wants to follow its Dallas customer that moves out of region. All right. Ditto with Ameritech having an incentive to follow a Chicago customer who goes out of region. They each have incentive to follow. Now, the thing that I do not understand is that when you put them together, the combined entity now has more of an incentive than each one of them separately had to begin with. I do not understand that. MR. ROGERSON: Okay. Roger? MR. NOLL: They do have an incentive because contrary to the assertions that have been made, competition in access is more prospective than it is real. It is more profitable to be a monopoly. Most of the arguments we are hearing are it is more profitable and cheaper to form the ubiquitous interconnected network that will track all the customers if there is only one wire line base carrier than if there are two. That is the essence of the argument. The point that we should bear in mind is that as analysts, we should not care who the first ubiquitous national network is, and it is intriguing and it is probably true that if all the ILECs merged together they could ubiquitously be one, and they are saying but we need to be allowed to do that because if you do not, AT&T will be there first with its cable companies. Our view about that should be A, AT&T and the TCI thing is prospective, not real. It is about the nineteenth idea in the last ten years about how to create the ubiquitous firm. None of them have worked yet, and because there are downside risks to creating the single ubiquitous wire line carrier, we should not do it until we know that is in fact how the market is going to work. MR. ROGERSON: Okay. I am going to cut the discussion off here and move us to our next session. Our next session is asking the question will these mergers have an effect on the FCC's ability to benchmark across firms? I have asked Joe Farrell to speak for five minutes to explain what this possible harm is, and then I have asked Robert Crandall and Dennis Carlton both to critique Joe's presentation. Joe, go ahead. MR. FARRELL: Thanks, Bill. Regulation inherently involves holding a regulated firm to some kind of performance standard or pricing standard that the regulated firm has not freely chosen. That raises some risks obviously. The goal is to make this performance standard or pricing standard or whatever it is efficiently challenging for the firm, but feasible. If the regulator does not know what is feasible, then the results are likely to be bad in a variety of ways. Either the demands on the firm will be infeasible, or the firm will be cut too much slack and prices will be allowed to go too high, or bad incentives will be created one way or another. We are familiar with all this kind of stuff. As regulation moves, we hope, yet slowly and gradually away from kind of traditional rate regulation or more clearly moves into new areas such as interconnection, the prospect of regulators having a hard time knowing enough to do the regulation they need to do seems to be more and more of an issue. How do regulators find out what is feasible? How can regulators find out what is feasible? It seems to me like there are three generic methods. Maybe there are more. I do not know, but here are three. One is what could politely be described as making an independent assessment or rudely described as trying to run a shadow business, so trying to know the technology, trying to know the structure of demand, trying to do what you would do if you were a conscientious member of the board of directors. That is pretty hard to do well, and it is pretty hard especially to do well if you are dealing with thinking about imposing an interconnection duty, let's say, that has never been imposed before in that form. The second thing that a regulator can do, which is the traditional thing that regulators do, is to use information from the firm's past to get an estimate of what the firm can do in the future. That is the traditional approach. In some sense it works, but in some sense it works rather badly. We are very familiar with some of the bad incentive effects that are created and notice that this, too, does not do you really a bit of good when you are trying to figure out whether sub-loop unbundling in three days at a reasonable price is feasible or not. The third thing you can do is to use information from other firms. Notice that this is fundamentally how competitive markets do it. That is to say the standards to which a competitive firm are held are the standards given by the performance of the most successful other firms in the market, and that should clue us in to the idea that this probably has some pretty good features. Well, it does have some pretty good features. It also has some defects and it has some problems, but those defects and problems surely are not perfectly correlated with the defects and problems of the other methods that regulators can use to figure out what is feasible. What I mean by that is even though benchmarking relative performance evaluation has its problems, it is surely true that the arsenal or tool kit of information tools that regulators have with it is a heck of a lot better than the arsenal or tool kit that they have without it. Now notice, and actually Michael made this point a little earlier. Even private firm managers who surely have a much better chance with knowing the technology, knowing the market, independent assessment approach, and who have a much better chance with using good information from the firm's past than do regulators, the FCC, even private managers often use what is called benchmarking. It is the hot thing of the late 1990s in business management is to go off and find out what your competitors are doing by way of responding to customer complaints or whatever it is. This really suggests that relative performance evaluation of various kinds is a very useful tool in finding out what is feasible. As I suggested at the beginning, that is in some sense the key problem of trying to do efficient regulation. Now, as some of you know and others of you probably are not going to bother to find out, but maybe some will, Bridger Mitchell and I submitted a paper, and there was also an attachment to the paper written by Wilkey Farr. In this paper, we gave lots and lots of examples where the FCC has explicitly used performance comparisons and benchmarking, and we somewhat arbitrarily talked about average practice benchmarking, as in setting a uniform X factor for price caps, best practice benchmarking as in various interconnections -- MR. ROGERSON: Joe, if you had 30 seconds left, what would you say? MR. FARRELL: Okay. Benchmarking of regulated firms, therefore, really is a used and useful technique for relatively efficient regulation. So what is the effect of mergers on all of this? First of all, a number of people have made the point, so I will forestall them before they make it again. As usual, not much effect if there are plenty of firms left to benchmark against. For some purposes, that probably will be true. I do not think it is really likely to be true for all purposes, especially if economies of scale are significant in this new national market. The comparisons with small ILECs and with CLECs may not do you a lot of good. There is a loss of pure diversity and a loss of information even if behavior does not change. This point raises some real subtleties, and I think it may be a mistake to spend too long on the subtleties because the real point is -- MR. ROGERSON: It would be at this point, yes. MR. FARRELL: -- incentives do change. Just as with product market competition, there are a lot of decisions that a firm can make that have opposite side effects on other firms and on consumers. If this firm merges with one of those other firms, then those cross effects on the other firm are going to be internalized, and consumers will lose. MR. ROGERSON: Okay. Joe has said multiple ILECs mean you can have competition within regulation, and that is useful to the regulator. Robert Crandall will now tell us why that is not so. MR. CRANDALL: I am not going to tell you necessarily that benchmarks are not useful to the regulator, but I am also not going to read a prepared statement because Roger has scared me into thinking that that looks like something the lawyers went over. Given that I have already compared the people that have hired Rob and me to Eastern Airlines and Bethlehem Steel, I guess there is not much risk. First of all, it seems to me that you have to keep in mind that what we are involved with here is a transition away from regulation to a situation which market forces and competition between CLECs and ILECs is supposed to dominate the landscape, not regulation from Washington. Even after your victory in the Supreme Court, I mean you only provide guidelines to the states, and even that should wither away over time. One should not think that benchmarking off a set of firms who grew up in a regulated environment provides you necessarily with efficient benchmarking. Otherwise the GOS plan might have argued you should not privatize Russian steel companies or, you know, at the CAB they might have held on not allowing Slow Hawk and Agony to merge into U.S. Air and drive an airline which today is offering service at one-half the price of the domestic passenger fare investigation standard, which is based on benchmarks. Secondly, there is nothing in the record here, and I have seen nothing from Joe or Bridger, about how much benefit these benchmarks provide. We have some estimates of what the potential benefits from the mergers are. If we stop the mergers in order to maintain a couple of inefficient benchmarks, how much benefit will that provide, and will it offset the benefits from the mergers? Third, the mergers themselves, if they work, will generate more CLEC activity out of region and provide more CLEC/ILEC sort of benchmarks. Over time, presumably there is going to be more efficient benchmarks as CLECs and ILECs negotiate with each other over the terms of interconnection. It might be terms of interconnections and networks that do not now exist at the ILECs. It might be packets which networks of the sort that AT&T-TCI claim they are going to build now that they have apparently abandoned Project Angel. Finally, as we move towards a more competitive environment, the whole 271 process has to come to an end at some time soon. Paul Macaboyd claimed it would take ten years for RBOCs to get 271 permission, but it looks as if this process is beginning to move, particularly in New York state, and should spread fairly rapidly after that, at which point the benchmarks necessary for implementing 271 and OSS seem to me to go away. In addition, looking just parochially at the GTE-Bell Atlantic merger, it is hard to consider GTE as an appropriate benchmark for Bell Atlantic or some of the other RBOCs. They are not involved in the 271 process. Their entire structure, the dispersed operating systems around the country, are really very different from the RBOCs, and it is hard to argue you are losing a very important benchmark there. Finally, Joe's point on benchmarks for the X factor, the productivity factor, which is more in your filing than in your oral comments today, suggested you run the risk of internalizing the efficiency gains, which would then have this ratchet effect on providing disincentives for pursuing productivity enhancing investments at the ILECs. Two things need to be said about that. First of all, there are lots of benchmarks for that around the world. We should not just be looking at U.S. ILECs. Secondly, the Commission has never used the same approach twice, so it would be very hard for an ILEC to try to predict how the X factor is going to be adjusted in the future, given what they have done in the past. Third, chances are the entire benchmarking thing for X factors should look at a longer period of time because as has been shown in the case of British Telecom and the case of the railroads, productivity gains initially are very rapid when you begin to unleash the regulatory restraints because of the efficiency gains moving off an inefficient production technology towards a more efficient one. If you look at deregulated industries in this country, the railroads probably have the greatest rate of productivity gain, hardly not because of enormous technological change, but because they simply are able to move from inefficient operations. I would not put much stock in the notion that you need to preserve independent large ILECs in order to reset the X factor. MR. ROGERSON: I would like to commend you on getting done before my 30 second warning, Bob. Thank you. MR. CRANDALL: I am so scared of your tyrannical approach. MR. ROGERSON: Dennis Carlton? MR. CARLTON: Thank you. The relevant question in considering the effect of these mergers on benchmarking is whether the mergers will so significantly impede the ability of regulators to do their job that it will overwhelm the substantial benefits from these mergers. The question is not whether there is going to be one or two fewer data points for some hypothetical comparison. There is no empirical evidence to support the claim that these proposed consolidations are going to make regulations significantly more difficult. There is no evidence, for example, that the previous RBOC mergers resulted in significant impediments to regulators' ability to do their job. The critics of these mergers who are relying on benchmarking are ignoring ways in which trends in the industry are themselves right now creating more and more benchmarks. One of the key concerns of regulators today is how an ILEC is going to interact with a CLEC. The most useful tool in evaluating such potential discrimination is to compare the service that ILECs provide to themselves as compared to CLECs in the territory. That is, the ILECs provide an internal benchmark to measure their own performance, and this key benchmark will certainly remain after these mergers. Moreover, there are new benchmarks constantly emerging in this industry. Just look at what this transaction is going to do. A benchmark is not valuable if everybody is similarly situated. Benchmarks get more and more valuable as people are pursuing different strategies. SBC is moving out of region. SBC will be interacting with an ILEC out of region. SBC will now have very different incentives than other people in making sure that connections with its out of region ILEC are proper. That, of course, will mean it is pursuing a different strategy. That is when benchmarks start getting more and more informative. If you look, for example, about a concern, which are benchmarks concerning how new technologies will be hooked up, well, now you have the possibility that we have vertically integrated firms. You can look at how Sprint as an ILEC treats itself as a CLEC. What you are having in this new environment are smart CLECs able to monitor ILEC performance. You have CLECs that are ILECs in other regions, and you have vertically integrated firms. All of those are new benchmarks that are becoming available. In Joe's statement, he mentioned that there would be a reduced incentive to engage in productivity enhancing investments because of what he called the rachet effect; that is, because regulators are going to respond in the future and may lower your prices. In the future, you will have a lower incentive to respond. That, of course, ignores an opposite incentive, which is there may very well be economies of scale in investment. If that is the case and efficiencies result, you are going to get more, not less investment. Finally, let me just point out that if the concerns about benchmarking are accurate, SBC has embarked on a strategy in which you would say it is subjecting itself to this cost of benchmarking that Joe was describing. That does not strike me as a reasonable business strategy to be engaged in if you really think it is a serious problem. In sum, there is no evidence suggesting that past reductions in the number of ILEC benchmarks have had a significant adverse effect on the ability of regulators to regulate. Given the industry trends and the new information generated not only by this merger, but also by trends in the telecommunications industry, the concern about reducing the number of data points by one or two seem over exaggerated to me. Consumers should not be denied the tangible benefits of these mergers based on unsupported speculation about theoretical harms. MR. ROGERSON: Okay. First, Roger, does that really mean you want to speak? MR. NOLL: Yes, I want to speak. MR. ROGERSON: Okay. Go ahead. MR. NOLL: I have distinctly mixed feelings about benchmarking, so I will sort of give a critique of everything I have heard. The first obvious point to say is, Dennis, the value of benchmarking in the competitive industry is zero for the reason that Bob said. Benchmarking is interesting only if you have regulated monopoly. It is not interesting if you have competition. If the basic premise of the proposals for the merger is true, then it is pro-competitive because it introduces substantially more competition in long distance services and advanced telecommunications services. Then there is no value to the benchmarking that will arise from the vertically integrated firms because of the fact that we will want to regulate it. Bob's basic point that you want to think in the long run about what is going to be regulated and what not and evaluate benchmarking on those terms is exactly right. Now, the place that I think that Bob is wrong is that indeed there is a lot of regulation out there; not only regulation by states, but regulations by the FCC of interconnection, so the idea that there are people out there with serious market power in some aspects of this industry is something that in the short run at least we have to bear in mind. The logic then of Joe's argument about benchmarking as a useful potential tool is completely valid. The only trouble is regulators in the past, notwithstanding Joe's examples, have not really taken advantage of the opportunities for benchmarking, and that is my main concern with the benchmarking argument. Notice that the FCC in the late 1980s and early 1990s went through this strategy of trying actually to enforce the concept of uniform accounting principles across ILECs and having them all produce quarterly accounting cost estimates that segregated their costs into that which is in the FCC's jurisdiction, that which is in the state's jurisdiction and that which is unregulated. It is the case the FCC did on occasion use that information to in fact pull out some numbers from the rate bases for interstate rate setting purposes of local exchange carriers of the interconnection part of the basis of this benchmark information, so it was used. The flip side was the resources of the FCC to actually make use of this information were infinitesimal compared to what they would have had to have been to use it completely. That was the subject of not one, not two, but three GAO reports which said how can you possibly have the FCC make use of this information, go to this enormous expense to collect it, when there is no staff to analyze it. Now we are talking about the FCC here, which is the singularly most sophisticated regulator. North Dakota's regulators are going to make use of this? Uh-uh. I think that is the problem. Now, there is one point in which I think Joe's argument is absolutely solid, which is not the accounting cost efficiency policing/non-discrimination policing part, but it is in the technical part. That is to say when Company A says it is technically impossible to do X, but Company B is doing it, then the presence of benchmarking is absolutely essential. In terms of interconnection, a lot of the issue is about technology. It is about the feasibility of number portability, the feasibility of certain kinds of unbundling, the feasibility of having access to software inside switches. These are not issues that require enormous staff time to do accounting. These are issues is it true, or is it not. It seems to me there the advantages of multiple local exchange carrier providers are really great, and I think the argument has the most force there. MR. ROGERSON: I actually gave a much less eloquent version of the last few minutes of your speech at a breakfast meeting the other day. Allan Campasero, who is actually here in the audience, stood up from GTE and said Bill, that is really a lovely theory. Why do you not give me 15 examples? Why not give me two or three good examples at least, right? Since this is such an important theory, certainly this has been happening in the last while, and you could give me some. I want to turn Allan's question over to anyone on this panel who would like to deal with it. MR. FARRELL: Well, there is a voluminous document full of examples. Let me just mention a couple that I remember. The LRN number portability issue, which Bridger and I discussed in some detail, where a number of large ILECs claimed that LRN was not reasonably implementable in the foreseeable future. I forget the claims, but Ameritech said oh, no problem. We can just do it. There was, on the other hand, just to say that this is not about Ameritech being particularly special, in a shared transport proceeding Ameritech claimed that it was impossible to do the bookkeeping and billing. Bell Atlantic, on the other hand, was just doing it. There are a couple of examples. If you want me to pull out my document and leaf through and read some more, I can do that. MR. ROGERSON: Well, I do not know. Is that enough to convince you, Dennis, or do we need more? I mean, are those good examples? MR. CARLTON: You know, I think there are two things you can say. You can look at the details of those examples and ask what would have happened but for, okay. That is an interesting experiment about what has happened in the past. I think there are two responses, though, you should be keeping in mind, two concerns. The first is things are changing. We are getting more competition. We are getting more CLECs. We are getting more ILECs out of region. That is going to give you more information. Second, if you self-select just one or two bad examples, even if they turned out to be true, and I am not suggesting they necessarily are, but even if they were you have ignored the other side of the coin. What about all the benefits? What about the increased information you are getting from benchmarking because now you have SBC negotiating with an ILEC out of its territory saying I want you to do this, and I know you can do this because I do it. I do it in my territory, so do not give me any baloney that you cannot do it. I am going to tell the regulator you can do it. That seems like an enormous amount of information. In fact, the FCC in the Bell Atlantic-Ninex decision has exactly such a statement that they understand the benefit that comes about when one ILEC in one region is a CLEC in another. Pursuing different strategies is giving you new information and giving you new incentives. That is when you are getting information from benchmarking. If everybody stuck in their own territory doing all the same thing, you are going to get very similar responses. It is when you start mixing them up, have some ILECs competing outside their territory against and being a CLEC outside of their territory competing against ILECs. That is when you start getting lots of information, so that would be my response. Let me just end by saying if you were asked in an anti-trust case to ask if prices are going to go up as a result of a merger, I do not think you would find one customer or two customers who said yes, my price is going up. You would ask on average if price is going to go up. On average, are prices going to go up in a new environment? I think that is the danger that you fall into if you rely on one or two examples, the danger being that you reach a conclusion about average overcomes, the overall outcome to the consumer, based on one or two anecdotes. MR. ROGERSON: Michael? MR. KATZ: Well, Roger, I think you got your wish. I do not think that Dennis cleared that last remark with his lawyers because my understanding has always been that the RBOCs have argued vigorously that they are not particularly good at negotiating interconnection agreements out of region, and that is why you should not consider them potential competitors of each other. That is just a statement about their lawyers, not about Dennis. I do want to address some of the points that Dennis raised and that Bob Crandall raised. Just a couple things on that. One is this issue of well, are there really going to be a lot more benchmarks? I think Bob hit it on the head with exactly what the problem is going to be. He said well, GTE is not a very good benchmark. They are too much different. Well, how is some little CLEC that is going to be a hundredth the size and in a very different market position and certainly not under the strictures of 271 then going to be a good benchmark? We talked about having a lot of CLECs that will monitor. I think CLECs do monitor today, and they complain vigorously to the states and the FCC. The question is how do you tell which one is right? I am sure there would be a lot of people in the industry who would be happy if the RBOCs would delegate all the authority to make these decisions to COVAD. However, COVAD said we are an intelligent CLEC. We figured out you are doing something bad. Fix it. If you want to stipulate to that, I think it will be fine. It may also bring the industry to a crashing halt, but that is another issue. Similar to that is this question of internal benchmarks of the LECs serving themselves. The problem there again is I cannot believe the ILECs would want to be held to the standard that says the same service you provide yourself is what you provide to everybody else. I just do not see how they actually believe them when they would say they cannot make their OSS system, make order entry, work the same today for someone outside of the organization as for inside. I certainly have not seen any willingness that they would grant the same access to the central offices or the same access to the software and the switches. I do not think in fact it is probably reasonable in some of those cases to demand the same access, so I do not see how again you can say that these internal benchmarks are going to be a powerful force because I think there are legitimate differences. I think some of those differences have been overstated, but I think fundamentally there are legitimate differences and so these are not going to be benchmarks that are going to be these great substitutes for having separate RBOCs and separate ILECs. MR. ROGERSON: Jeff? MR. SHEPERD: Very briefly. In the literature back in the 1930s, this was all tried in electric industry indirect competition between public and other enterprises. Here, as there, it is not so much what the regulators know that they can order. It is rather what is being done in a diverse way as the theory of innovation teaches us, a variety of things being tried which then not only teach each other, but bring pressure upon each other to try them also. That is what this merger is likely to stamp out. Therefore, in general do not forget the basis on which we need to think. MR. ROGERSON: Rob Gertner? MR. GERTNER: I think it is useful when focusing on benchmarks to actually think fairly specifically. It seems essential to think about how the mergers affect the amount of information that is out there. It varies a lot from all across the different things people talk about using benchmarks for. I mean, you really cannot ignore doing sort of the analysis of thinking what are the sources of variation? Why is there different information out there? Technical feasibility, for example. You want to think carefully about why does one firm think something is technically feasible and others do not? Is it because they had incentives to invest more in R&D that makes it learn that way, or is it because they have a diversity of interests and, therefore, some firms might therefore notice the feasibility in a different way or push towards the feasibility? If it is diversity of interest, then mergers which lead to greater diversity, like these mergers do, can actually increase the amount of information. The other thing that is really relevant for all this is the way in which this all feeds back into the incentives of the firms. Again, I think you need to look very carefully across the specific ways in which benchmarks are being used. In X factor type analysis, mergers that enhance efficiencies, those effects will swamp the rachet effect that exists there. Again, it becomes very important to think carefully through about how the benchmarks actually impact the incentives. MR. ROGERSON: Joe? MR. FARRELL: Thank you. I hardly know where to start. Let me just take a couple of these points. First on the creation of new benchmarks and the use of internal benchmarks. I think Michael is right that in many cases the ILECs will argue, and sometimes rightly, that the strict version of this equal provision standard is not reasonable, and that raises the whole question of whether you can reasonably do it. That is something that again regulators need to have more information to do. Let me also point out that even if you can and do do it, it is not enough. The equal provision standard solves a static version of a certain leveraging problem. It does not solve the dynamic version of leveraging problems, so suppose that an ILEC is providing input, A, and ILECs and others are providing in a complementary business B, and suppose that somebody other than the ILEC wants to innovate in B and needs cooperation from the ILEC's provision of A, for example, as with long distance and with access service. Then the ILEC can stymie that competitive innovation without violating the equal provision standard because it does not need to provide this cooperation to itself because it does not have that innovation. By imposing the equal treatment standard, not only are you providing only low powered incentives at best for the ILEC's provision of this monopoly service, but you are also allowing it to monopolize innovation in the competitive segment. Secondly, if you say that the ILEC has to equally provide A to itself and to others, you have done absolutely nothing to get efficient provision of A. You may have solved some leverage problem into B, or you may not, but you have done nothing for the provision of the monopoly input, A, and that is exactly, of course, where you may need regulation and where you may need benchmarking to provide good information so you can do not too bad regulation. Let me also come back to a point Bob made about the so-called unpredictable adjustment of the X factor. I think it is true that the way the Commission has behaved and the way other people have behaved elsewhere and just thinking about the political economy, nobody can really know for sure how the X factor will be adjusted and when. It is absolutely not true that that implies that each ILEC is going to assume that the X factor is completely exogenous to its actions. Just to give you an example, back in 1997 when there was the access reform proceeding, there were parties who argued that there should be company by company level re-initialization. In other words, take away any so-called excess profits company by company. That did not happen. I am glad it did not happen, but it obviously was not common knowledge that it would not happen. Yes, there is uncertainty about how X factors get adjusted, but, no, that does not imply that everybody, therefore, makes decisions as if the X factor is going to be completely exogenous to their technical process. MR. ROGERSON: Have I reached that point where I should turn the audience loose? Rich? Rich Gilbert? MR. GILBERT: I want to say one thing, which is it seems that what you said about the dynamic access story is also a reason in favor of the merger as a better benchmark. Let me explain. If ILECs are confined to essentially their local focused strategy focused on their own territories paired to a national local strategy where you are anticipating providing integrated services, primarily a bigger package of services and direct competition with the IXEs, you have an incentive to do more things and to offer more services and to, therefore, provide those services nationwide that you might not be providing with a different strategy. It really also speaks to this point that it is different behaviors, it is different business strategies, that can produce different benchmarks. MR. ROGERSON: Joe, do you want to directly respond to that? MR. FARRELL: Yes, that is what I want to do. Look, if you really believe that these mergers are both necessary and sufficient conditions for a really vigorous out of region entry, then I think they should go ahead, and the sooner the better. MR. GILBERT: We do. MR. FARRELL: But I think there is a lot of skepticism and there is substantial grounds for skepticism about perhaps whether and certainly how much of that there really is. Then you have to start focusing on well, what if not? MR. CARLTON: Do you agree, Joe, that these mergers will accelerate out of region entry and, therefore, the benefits to consumers from new products, as well as more competition, will come faster? If it is a benefit and we know the benefits from new products are tremendous, is that not a fact that should weigh in your thinking as you think about stopping the mergers to preserve benchmarks, as well as the fact that in the future there are going to be more competition and more vertical integration? MR. FARRELL: Yes. I think that is exactly the question that I just answered. If you believe that these mergers are both a necessary condition and a sufficient condition for these LECs plunging wholeheartedly into out of region facilities based competition, then I think that is great. If you are not convinced of that, then you obviously have to discount your consideration. MR. CARLTON: But does that mean you would stop a merger in a non-regulated context on the same grounds, hoping that the companies would do whatever it is they are going to do together on their own? In other words, what you have here when you have a merger is you have a business plan. You have investments in a business plan, statements to their investors, a lot of money spent already on these plans. I do not understand how you can not say that that means they are prepared to go faster than if I say no, you cannot do the merger. You figure out what you want to do. Go back to the drawing board. MR. LITAN: Dennis, even if what you say is true discounted with some probability, it is still something that you put on the scales, and you compare it to the other potential harms that we are going to talk about, and then you decide, you know, where the overall balance is. MR. CARLTON: I agree. I am just saying -- MR. LITAN: We are debating how heavy we want to put the scale of the pro-competitive arguments. It is certainly a relevant factor. MR. CARLTON: Right. That is the point I wanted to make. MR. CRANDALL: Can I just make one point in response to Joe, and that is I do not know how much we are going to lose in the way of benchmarks even over the traditional issues that Joe raises, and he raises a couple of examples. He did not mention the British Water and Sewerage Administration example, which I do not think will bear much on our problem here today, but the fact is that a lot of these issues are arbitrated between parties at the state by state level. These mergers may not reduce benchmarks at all in that regard, number one. Number two, this issue of the fact that we are going to have more efficient, different technology companies entering out of region, for instance, Bell Atlantic and GTE, with an Internet backbone, with a much more complete array of services, something which nobody has responded to in Rob's presentation. It seems to me you are likely to get some better benchmarks, and you are not going to lose that much in the way of benchmarks from the traditional negotiations anyway, given the state by state regulation. MR. ROGERSON: Joe? MR. FARRELL: Okay. Let me say for the third time if serious, vigorous, effective out of region entry, that would be a great thing. If the mergers are both a necessary condition and a sufficient condition for that, then that may well dominate any other considerations. Now back to your other point about the state level, the state level analysis. This gets a little bit to the subtleties that we talked about before. Do you really get the same information? I think again the main point there is you get different incentives, so the state level analysis says, of course, if you have mergers of holding companies that does not change the operating company level behavior, then you get no direct impact on the information flow, but because there is a change in incentives you will get a change in the information flow and a change in the efficiency of behavior as a result of that. I think that is really the robust point to focus on. MR. ROGERSON: Okay. We have time for one or two questions from the audience. This time I will not insist that they be thoughtful or witty. Allan Campasero? Okay, Allan. MR. CAMPASERO: I had to take this chance. This is a very interesting general discussion of benchmarks. Of course, for us it is a specific deal that is at stake here. I cannot think of and I wonder if anybody can think of an instance where GTE has been used as a benchmark for the RBOCs? MR. ROGERSON: These are the same kind of questions he was asking me at breakfast the other day, guys. Who would like to take a shot at that? Anyone? MR. NOLL: There actually is one example. The cost study that was done in California about ten years by Bridger Mitchell -- MR. ROGERSON: Roger, we cannot hear you. MR. NOLL: Okay. The cost study that was done about ten years ago for the California Public Utilities Commission on the extent of what actually is local access cost and at what size of an exchange do you exhaust the economies of scale was actually based upon both GTE and Pac Bell cost studies. In fact, most of the information about the smaller cities came from GTE, and it was used by the California Public Utilities Commission for a generic regulatory proceeding that affected both of them. To think of it the other way, of course, is that GTE, and I do not know where the biggest GTE investment in the whole world is, but certainly Los Angeles has got to be right up there, so GTE in California is more like an RBOC than it is like a little, tiny guy, which it is in lots of other places. MR. ROGERSON: Mario Schwartz? MR. SCHWARTZ: Yes. Just a quick clarifying remark. Even if GTE was not used as a benchmark against the RBOCs, as long as the RBOCs were used as a benchmark against GTE you would still expect the merger to make a difference in the parties' incentives and, therefore, get less benchmarking. MR. ROGERSON: There is a question over there. Would you mind telling us your name and your affiliation before you ask your question? MR. CLARK: I am Rich Clark with AT&T, and I guess I would like to address about three or four years ago there was a proposal by Pacific Bell that it was going to replace its copper pair network with a hybrid fiber co-ax network. These plans were cancelled after SBC took over Pacific Bell, and I guess we are now waiting to see if AT&T and TCI can make that work. I would suggest that there are situations where, you know, these mergers have already had the effect of eliminating diversity or possibility for benchmarks of different technologies. MR. ROGERSON: Rich Gilbert, do you want to respond? MR. GILBERT: I would just like to respond that the merger with Telesys refocused their efforts on new services, and they have had a really accelerated roll out of high band with DSL services. MR. ROGERSON: Bob Crandall? MR. CRANDALL: It seems to me that the fact that SBC re-evaluated their architecture and decided not to go ahead with that is not any more damaging to the notion that now they are going to try another strategy to enter out of region than perhaps denying your merger with TCI because you never really got fixed wireless going very well. MR. ROGERSON: Okay. We have one more question from the back, and then we will break. We will take our break. MS. BLOOMENFELD: Sue Bloomenfeld with Wilkey Farr. I just wanted to fill out I think Rich's question, which is the point was that in seeking merger approval, SBC and Pac Tel promised the Commission that that video experiment and build out would continue post merger, and then the plans were changed. I think maybe that may be relevant here. MR. ROGERSON: Okay. Who would like to give me an economic analysis of that question? MR. GILBERT: I would still like to respond by saying from a consumer's point of view, the concern is what is the availability of high band services. On that measure, the merger has done very well. There is no question that now SBC with Telesys has the highest DSL roll out I think of any ILEC. MR. ROGERSON: Great. Okay. We will take a 15 minute break, and we will resume at 11:00 a.m. with Session Three. (Whereupon, a short recess was taken.) MR. ROGERSON: No FCC round table involving economic analysis would be complete without us parading out our own Commissioner, who is in fact an economist, Commissioner Furchgoti-Roth. He has kindly agreed to make a few remarks to us all prior to starting our third session. Commissioner Furchgoti-Roth, please go ahead. MR. FURCHTGOTT-ROTH: Thank you, Bill. I first want to thank all the panelists for taking time out of their busy schedules to come here to the Commission. We are very honored to have you here today. It is a rare privilege to have such eminent economists all come together at one time, and I think it is a great tribute to you, Bill, for having organized this and having very thoughtfully brought together all these wonderful people. I am sure we are all learning a great deal about the proposed mergers. As in a lot of topics here at the Commission, I come at this with a slightly different perspective than some other folks at the Commission. I have no idea whether these proposed mergers or any of the dozens of other proposed mergers that might be here at the Commission present any anti-competitive problems, and I am sure that the panelists here have some very strong view of it and many of you in the audience as well. I am very interested in finding out what the proper role of the FCC as an institution is in reviewing these mergers or really any other mergers. I very much look forward to reading the transcripts or seeing the videotapes of these proceedings. They may not be quite as interesting as Monica Lewinsky, but I think there may be something here to be learned. As many of you know, the FCC's authority to review mergers comes in two areas. One is directly out of the Clayton Act, and I am not quite sure if the Commission has ever used that. Secondly is through the licensing process through Sections 208 through 210 or 214, and that is how the Commission has chosen to review mergers and is reviewing these. The difficulty we have, as far as I can tell, and I have requested and have yet to find them, is the specific rules by which the Commission reviews these mergers. We certainly have some case history that we do not have specific written rules. I would be very interested if any of you have seen any such thing or, to the extent you have not, if you could give us any guidance as to what such rules might look like under 208 through 210 and 214 and in particular how those would differ from the DOJ/FTC merger guidelines. All of you on this panel I suspect have had some interest in these mergers and have made presentations either to DOJ or possibly to state regulators, and I am curious if the issues that you think the FCC should consider are in any way materially different from the issues that the Department of Justice should consider in its merger reviews and if the standards that the FCC should apply are in any way different from those that the Department of Justice would apply. If they are different, should they be written down, codified, memorialized in some way that would set some clear guidance to the public about how this Commission will review mergers. I think we can get very quickly to the issue of whether there is anything in the presentations that you have made here today that are different from the information that you have already presented to the Department of Justice and, if so, what triggered that. Why is that different? How have you come to that conclusion? I do not know what the right answers are to any of these questions, but I have been asking these questions now for the past few months. I have been asking a lot of lawyers and have not been getting any answers. The nice thing about economists is economists will actually tell you something and not worry about who it offends. VOICE 1: Whether it is legal. MR. FURCHTGOTT-ROTH: Right. MR. ROGERSON: We will test this, Commissioner, if you would like. MR. FURCHTGOTT-ROTH: Thank you. MR. ROGERSON: Joe Farrell is dying to answer. You know, I think you have raised very interesting questions, and I would love to spend five minutes turning the panel loose on it if it is all right with you. MR. FURCHTGOTT-ROTH: Please. MR. ROGERSON: Joe, go ahead. MR. FARRELL: Okay. Thank you. To answer your questions, or try to, I think the issues are overlapping, and the standards should potentially be different. Let me try to explain that. My understanding is the Commission is supposed to consider a fairly broad public interest test of whether mergers are a good thing or a bad thing. That is not necessarily the same question as DOJ/FTC are supposed to consider, which is whether they substantially reduce competition. I think the differences are in two areas. One is the FCC is trying to help along a process of increasing competition rather than just preventing it from being diminished, and that raises, of course, the whole issue of potential competition, how it is dealt with, what the evidentiary standards are in the Courts at the DOJ and FTC and whether those are the right standards for an industry where competition in some segments has been illegal until relatively recently and is still not going very far. Then, of course, you will not be surprised to hear I think the benchmarking type issues, which I talked about this morning, are probably of more direct concern to the Commission than they would be to DOJ or other anti-trust agencies. MR. FURCHTGOTT-ROTH: If I could just very quickly make a comment, Joe? It is true that the Commission has some public interest obligations under 208 through 210 -- actually, we have public convenience and necessity under 214, but not a public interest obligation -- but those are for the transfer of licenses, not for mergers. This agency handles over 10,000 license transfers every year, the vast, vast majority of which, more than 10,000 license transfers a year, which are never subjected to any kind of public interest test. We have no written rules about how we decide which license transfers are going to be subjected to a greater degree of scrutiny than some other license transfers. At least the Department of Justice has specific rules about which mergers it is going to review and by statute which will trigger a process and which do not. We at the Commission do not have that sort of clear, written guidance. It is a little loose right now, I think. MR. ROGERSON: You know, we are very fortunate to have two former chief economists from the Department of Justice with us today. I certainly would love to hear what both of them have to say about this. Rich Gilbert, why do you not go first? MR. GILBERT: The Department of Justice has guidelines for the review of mergers, but does not really have what I would call rules. Each merger evaluation is a very fact specific exercise, and the ultimate question is in my view really a public interest test, although the analysis says first is there a threat to competition, is there a risk of harm to competition, and then if there is then do the benefits and efficiencies outweigh that risk of harm to competition. I think the issues before the Commission should be very similar. There may be some regulatory issues as well. In principle, you could actually say that that is also relevant to a DOJ analysis as well to the extent that the change in competition, if there is some, affects regulation, but I think that there should be a substantial convergence in the two standards, even if they may not be identically the same. MR. ROGERSON: Bob Litan? MR. LITAN: Actually, this was the very first thing that I was going to address in my presentation, and so I am going to carve it out from my presentation and preserve my time. MR. ROGERSON: No, no. You are too quick for me. MR. LITAN: Look, I had presumed that the FCC operated under a public interest standard because if it did not, there would be no role for the FCC. I mean, why even have the FCC rule on mergers? My all around presumption is that is public interest, and I wrote down on my outline, which is out there for people to look at, at least three alternative tests that have never been made explicit, by the way. In fact, in answer to your question, you are in the process or you can make case law in this area. I do not know the case law on the definition of public interest, but you can have three alternatives. One is you could set the bar at a reasonable likelihood that the merger will just lessen competition, as opposed to substantially lessen competition, which is the Clayton 7 standard. That will be a slightly different standard that will be a little more strict than the DOJ standard. You could then make it even tougher by saying that you had to find that the mergers may be pro-competitive, or you could go even further and try to find that the mergers are actually likely to be pro-competitive. I think, frankly, when I discuss the outcome of potential competition analysis, the outcome depends heavily on which of these standards you actually apply, but I think you are in the process of writing the rules. MR. ROGERSON: You know, I mis-spoke myself. I am betraying my great youth, I guess. A little middle aged joke there. We have, I believe, a third former chief economist of DOJ in the room, which is, of course, Jeff Sheperd. I certainly want to hear from him as well. MR. SHEPERD: I will be very brief. Collective memory is so short. In 1967-1968, I was the third of Don Turner's special economic assistants, so I am only half of the chief economist, but I was there when this all started, and I helped draft the guidelines on the mergers that he put out in 1968. I would only say in this matter that I think after some careful reflection that whatever anti-trust does is a help, but it should not set the parameters for what the Commission does. MR. ROGERSON: I have asked all of the chiefs except one more chief economist, Michael Katz. Go ahead. MR. KATZ: Actually, I want to address this problem from a slightly different angle, which people have been talking about should the standards be different between the Department of Justice and the Commission. I think there is another difference because I think the standards should largely be the same. I think there is scope for some difference, but largely they should be the same. The analysis should largely be the same. I think there is a big difference between the two agencies in terms of their ability to impose and to implement remedies. The Department of Justice rightly is loathe to get into regulatory solutions, and so when the Department of Justice evaluates -- you think it is not right, or you think they got into them? MR. LITAN: We got into them all the time. I spent a year and a half negotiating them from deal to deal. That is what I did. MR. KATZ: Fortunately, the people I deal with are trying to stay out of them. In any case, it would not be right for them to try because they are not an expert agency. I mean, there are some industries they obviously are involved in quite a bit, but their role is very different than that of the Commission and so I think that is what I see as I think the biggest difference is the sorts of remedies that are available and proper for the Commission are I think much more intrusive and expansive. I know those are bad words these days in regulation, but I think the fact is that regulation is going to be around for awhile and I think that that is appropriate, although I share everybody's hopes that -- well, everybody hopes that regulation will go away. I think we differ on why, whether we hope there is also competition as the driver of it going away. Let me also mention one other thing, and I know you are not supposed to be rude to your host, but I know a lot of things that I then do not listen to, and that is I think that the information that goes to the Commission is different than what goes to the DOJ because I think that unfortunately the Commission it is much more cumbersome to provide confidential information and for the Commission to act on it. I do not know what the solution to that is, but I think it is something that I saw when I was chief economist and I see now that I think does hinder the Commission's ability to analyze some of these things because these issues that we are discussing today, and Bob Crandall brought this up when people start asking about the business plans. First off, we are the wrong set of people to ask about the details of it because we are not the business people, but also this is the wrong place to discuss it. These are very sensitive issues. It is sensitive for the merging parties. It is a sensitive issue for the parties that are concerned that they will be denied access, and that will force them to alter their plans. I think, Commissioner, I am glad you are looking into this, and I think it is an excellent thing to do. I would add that to part of what you look at is how to deal with confidential information in a more streamlined or more effective way. MR. ROGERSON: Okay. Now that we have heard from all of the chiefs -- MR. NOLL: Let me be an Indian. I just want to make the obvious comment. A group of economists is not the right group for Harold to ask the question simply because no matter how we slice it, we are always going to come down to the same set of criteria, right? We are trying to do things efficiently. Both agencies offer in a political and legal environment. That means the constraints, what they can do. Neither agency maximizes economic efficiency. All right. There are other constraints operating upon what they do from the legal environment in which they have been created and from the political environment in which they have to get their budgets and their staffing increase requests and all the rest. Just the obvious point is the point about the way potential competition is taken into a place, the way the FCC worries about universal service, which is not a concern of the Department of Justice. As economists, we have a hard time cognizing exactly how we are going to take into account these constraints, but it seems to me the real reason why the FCC has a separate and independent authority to review mergers is in fact a public interest issue that in part goes beyond the economic efficiency criteria. It means two things. It means that we as economists are not going to write all the rules, but, secondly, it means that within the domain of efficiency maximization there are going to be actions and strategies and decision rules available to the FCC that are not available to Justice and vice versa, and so I think part of what should be going on here is taking into account the feasible set of policies and rules, as opposed to the best of all possible worlds set, which I agree are largely the same in the two environments. MR. ROGERSON: Bob Crandall? MR. CRANDALL: One brief point that is kind of ironic. Rich talked about enforcing the Clayton Act at Justice. In fact, fortunately, Justice does take into account efficiency gains from a merger for two reasons. One is to balance it against any potential cost, and, secondly, it wants to see if this is really the reason why firms are merging or might it be an attempt to monopolize. In fact, under the Act such a balancing is not permitted. The Act says any lessening of competition that tends to create a monopoly in any line of commerce has to be stopped. That means you do not bring some cases you could win. When Jeff worked for Don Turner, Don Turner was accused of treating the Supreme Court like a bunch of C students in his course, but the irony here is that you can take benefits into account, whereas formally in a Court of law it is more difficult for Justice to do it. To the extent that you can take them into account, it is hard to imagine benefits being negative. This would suggest that you are more likely to approve a merger than DOJ is. MR. ROGERSON: Dennis? MR. CARLTON: I will be very, very brief. I have never been a chief economist at the Department of Justice, although I worked there a little bit. If I did fractions, I would be below a half. I did help write the recent merger guidelines, though, so maybe I am epsilon above zero. I guess I agree with a lot of what Michael Katz said. I think I disagree a bit with what Joe Farrell said and Bob Litan. I think it would be a mistake if the FCC adopted different standards for potential competition or for benchmarking. The Department of Justice should be concerned if a merger will cause a harm in an industry by impairing regulators and that will adversely affect competition. That strikes me as something that should be taken into account. I am glad I am seeing someone from the Department of Justice shaking their head and agreeing. I think that the standard that the Department of Justice uses in evaluating mergers is the correct one. Exactly how the Courts interpret it or not, I think the way it has been implemented, how the Courts would interpret it, the way it has been implemented at the Department of Justice, my understanding is, they do take account of total benefits and total cost. I think it would be a mistake to have a different standard at the FCC. However, I think there may be special categories of customers, such as those you want to protect through universal service, that raise special issues that the Department of Justice would find outside their realm of expertise and may not pay attention to. That is something the FCC should be concerned with. I do not see that arising in any of our discussions this morning, but if there were such a group of individuals that the FCC is charged with making sure they are protected in some way, that would be a special difference. The other difference is you have special expertise that gives you the ability perhaps to analyze things either differently, not with a different goal, but just to come to different conclusions because of your past experience or, as Michael said, the reluctance to impose regulatory solutions, complicated regulatory solutions. They may not seem so complicated to you guys as to the Department of Justice. I think those are the differences, but in terms of fundamental goals under anti-trust, the standards I think should really be the same. MR. ROGERSON: Rob, I have to call on you. MR. GERTNER: Just to complete the picture. MR. ROGERSON: Right. MR. GERTNER: I think that I basically agree with what Dennis just said. When you think about the public interest standard, you know, the guidelines and the way the Justice Department analyzes mergers has seemed to be and have turned out to be a very effective way I think of promoting the public interest in merger analysis, the notion that, you know, absent evidence of substantial anti-competitive harm, we want to allow mergers to go through. That is a standard that works well because in fact it does promote the public interest in general, so I think that by and large I agree with what Dennis said. MR. ROGERSON: Commissioner, would you like to raise any final issues? MR. FURCHTGOTT-ROTH: I thank all of you for your comments. They are all very thoughtful. I have learned a lot and will try to take those into account as we move forward in reviewing these. I am glad you are here on a telecom matter. We are not here on Exxon Mobil or Amoco BP. I still have a lot of questions and look forward to reviewing the entire session. Thank you very much. (Applause.) MR. ROGERSON: All that repressed, you have only 30 seconds I am going to take out on Bob Litan now. So now back to business. No more running on. We are going to turn to the issue of effects of actual and potential competition of this merger. Bob Litan is going to begin, then Jeff Sheperd will make remarks, and finally Rich Gilbert will make remarks, and then we will turn it loose. Go ahead. MR. LITAN: Okay. Just one last word on the standard because it is a lead in to my discussion. The question is should the standard of the FCC be that you have to have a winnable Clayton 7 case in order to stop a merger on competitive grounds, or should you be able to stop a merger on something short of a Clayton 7 standard? All I am going to say is that my analysis or the outcome of the analysis rests heavily on which of those two things you believe. All right. Now, on actual competition my belief is, subject to being corrected, that there is very little actual competition between any of these parties except perhaps in some regions in wireless. Where that is true and where that is a potential problem it is easily fixed with divestitures, so I do not think it is a big deal. The only thing that I think is interesting here is potential competition. Now, the Justice Department, to my knowledge, has never won a case on potential competition. On the other hand, they have never had a monopoly situation where potential competition should matter more. You should care more about the presence of contestability where you have a monopoly to begin with. Therefore, you want to look at the number of potential entrants that are out there before the merger, after the merger and whether the merging parties, any of them, are the most likely successful potential entrants. Those are the things you want to look at. What I did in my piece is I prepared a little chart for the three different markets, telephones, TV and advanced services, and I tried to tabulate who were the potential competitors there. The bottom line of that complicated chart is this. I had assumed that except for the merging parties that there are no other RBOCs that are likely potential entrants in these markets. I do not believe that they are significantly likely entrants, so you have one other potential RBOC plus three main long distance companies is what it comes down to in telephones. I know you have the CLECs. I know the arguments about electricity and cable and all that. I view all that down the road, and the CLECs are minnows. All right. So really in the telephone market you are talking about going from basically four to three is what it comes down to. Would that be a Justice Department case? No. I do not think Justice could win on that. Could it be a case here? That is an open question. It depends on what your standard is. The second market, TV. There you have fewer potential competitors. My argument there is you have AT&T. They are marrying with TCI and Time Warner. It is not clear how any other long distance guys are doing it. We have wireless companies' satellites already in this, but they are inhibited because they cannot show local broadcast, although the FCC could change that rule and make them into real competitors, which, frankly, they should. A side commercial. In any event, in the absence of that, there are fewer potential competitors in TV, and there appears to be some evidence Ameritech is already in TV in its local area. I have not seen the corporate documents, but if there are corporate documents that show that Ameritech was planning to get into TV outside a region, that could be more significant. All I am telling you is there are fewer potential competitors in TV. There is more likely to be a problem under any standard you look at. Finally, in advanced services it is anybody's guess. I have basically a lot of unknowns, a lot of questions, and I am really not going to make much of a call in that area. The final point I will raise is that if you believe there is a problem, conditions can fix problems. One condition that the FCC could seriously consider is what they did in Bell Atlantic-Ninex. Remember, they originally had their unbundling rule, their uni unbundling rule. It was challenged in the Courts. We all know what happened eventually in the Supreme Court, but the FCC went ahead despite the challenge and imposed its original rule in Bell Atlantic-Ninex. So what the FCC could do now in the wake of the Supreme Court decision, which, as I understand it, basically said that it was okay for the FCC to have a rule on one uni platform, but on multiple uni platforms they had to go back and do their homework, so theoretically the FCC could just go ahead and impose their original multi multiple uni platform rule as a condition for both mergers. Open question though as to how much additional pro-competitive effect you get relative to offering just one uni platform. I do not know what the answer to that is, but all I know is that if I was the FCC, I would be seriously considering adding that as a condition if I was going to go ahead and approve the mergers. MR. ROGERSON: I have now got to tell you you only have 30 seconds left. MR. LITAN: I talk fast. MR. ROGERSON: What I would like to do is give you 30 seconds to tell us what a multi uni platform versus a single uni platform is. MR. LITAN: Well, this opinion was a mess, as far as I was concerned. The way I interpret it is the Supreme Court said that the Bells had to offer at least one platform or one combination of unbundled network elements at essentially incremental costs, long run incremental costs, but that the Supreme Court questioned. That is the way I read the opinion, and I could be wrong. It questioned whether or not the FCC could force the RBOCs to offer multiple platforms so that other people could pick and choose in effect which pieces they wanted and that the FCC had to go back and do its homework. Now, I could be wrong about that interpretation, but I am still quite confused about what the Supreme Court said. I may not be the only person in the room about that. A lot of people were confused. MR. ROGERSON: Thanks, Bob. Jeff? MR. SHEPERD: Thank you. I am here as an economics colleague of people at the table and some of you. I have an outline of two pages of my main points. The copies are probably gone, but I could try to give it to you if you asked me later. In general, I do not think regulation is so bad. The compulsion to get rid of it should not be so strong. I do not think anti-trust is so good, particularly on mergers. I am not impressed with the concepts that are cobbled together in the merger guidelines now, and I think the division on the whole has weak enforcement of those rules. I think the research basis about mergers, both in business and economic research, is that most of them do not work out even for their shareholders, as well as for the public. Earlier I said that this is a difficult and an unstable period in which managers of telecom firms may well feel compelled to merge in a self-protective way. Solving this arms race is a major problem for the FCC. In fact, I think of this merger, both of them in fact, as trial balloons. They sent them up thinking well, let's try it and see if it works. We do not really expect them to sail through, but the other feature perhaps is that these mergers would tend to nullify the basis the FCC put forth for approving the Bell Atlantic-Ninex merger. That is, they counted on conditions which now would no longer be true if these mergers go through. However, rather than focus on these two mergers, I would like to just remind us of basic economic points that are relevant. First, this is not just a matter of static efficiency. Society wishes for innovation and other good things from mergers, not just a narrow price equals margin of cost consumer surplus maximizing result. Second, effective competition requires three main things. One, you need five, or maybe more or maybe one less, actual comparable competitors, not just two or three. If you want a horse race, you have to have horses. Also, dominance tends to suppress or distort competition, and also you need easy entry, so the basic target and the basis, I suppose, on which 271 might be decided is are there enough competitors so that they will not collude? Is there dominance, which does not permit effective competition, and is entry really easy? I would stress dominance really matters. It is not just a neutral condition, and deregulation has tended and does tend to get detoured into a dominance trap. The monopolist learns to live with ten, 15, maybe 20 percent of the market as competitors, but then says no and from then on expects to live with most of the market. In fact, that is my reading of the business press. The Baby Bells expect to keep 80 percent of the market right on into the future. I do not think that is good enough for the FCC or for anti-trust, for that matter, and in the process towards effective competition mergers are the main danger. They are the thing by which companies can directly stop the progress toward effective competition. Now, whether these mergers are that way is open to debate. Everything among economists is a two sided issue. There is a balance to be struck, and it may tend one side or the other. Of course, we can differ on these, but to say everything is all one way or the other is lawyer talk, not economics talk. As for barriers, there are many sources of barriers. I usually discuss 18 or 20 of them. It is not a matter of just -- MR. ROGERSON: In 30 seconds, though. MR. SHEPERD: -- a few. I will be very quick. Among them are not just the exogenous conditions of size and money needed to enter and so forth, but endogenous strategies and tactics that companies can play to defeat entrants. In general, the burden of proof should lie very strongly against self-interested claims about the balance of the goods and the bads, and that is very important for the FCC that it not just treat everything as kind of everybody has the same amount of credibility. Finally, any gains must be net; not just something you can show, but it would have to be strictly net. Now, finally in my outline I go through the reasons why competition would on balance, I think, be hurt by these mergers, but that is what I was hoping to gather, the data on these things, not just announcements of positions. MR. ROGERSON: Rich Gilbert? MR. GILBERT: There is no potential competition case here period. Now, I still have five minutes, right? MR. ROGERSON: Well, four and 50 about. MR. GILBERT: Okay. Let me show a list of the existing competitors in St. Louis. You are not going to be able to read from this, so I will read from it. Facilities based. MCI WorldCom is in St. Louis, and this is one of the key issues, I think, on a potential competition analysis. Are we talking about possibility of Ameritech moving into St. Louis? AT&T through TCG Teleport, Intermedia, Digital Teleport, Frontier, Birch Telecom, Winstar, Sprint has announced its Ion entry, Telegent has announced that it will enter, and we have AT&T-TCI, resellers. There are a couple of resellers as well. The list that I mentioned is facilities based. If you look into Chicago, that issue has been raised. The list is probably twice as long in Chicago. If we are talking about what I call conventional CLEC style entry, which for an RBOC or any ILEC would mean somebody going out of region, setting up as a standard CLEC, there are lots of folks who can do that. There are a lot of folks who have been doing that. If that is the issue, there simply is not a potential competition case here. What do you have to show for potential competition? You have to show that you have a firm that is a likely potential entrant. You have to be able to say that someone actually is going to enter, as opposed to well, maybe somehow somewhere someone could enter. If that is the standard, everybody is a potential entrant. You have to show that if entry occurred, there would be a substantial deconcentrating effect, which translates into an effect on prices and the market, and you have to show that the potential entrant is one of a few because if there were many potential entrants, then taking one of them out really does not do anything. The issues that have been raised in terms of what types of potential entry would be eliminated -- could be eliminated -- in this transaction are Ameritech's entry into St. Louis and a possibility of SBC into Chicago. I mentioned before the issue about SBC in Chicago was under the old plan of a possible wireless platform based entry. SBC tried that in Rochester. It was very unsuccessful. They abandoned it and decided that that is just not a good way to go. Ameritech into St. Louis. The history, the documents show, and I think these are public documents. I want anyone to stop me if I am divulging confidential information. The documents show that there was a contemplation of offering resold SBC local service. It would be advertised -- this is SBC local service, along with Ameritech's wireless service in St. Louis -- as a defensive measure because Ameritech was concerned that they were going to lose wireless customers to other wireless providers, PCS and cellular providers, who were about to provide bundled services. There was an experimental entry. It had not happened. The merger came along. It is off the table. Any contemplation of entry in St. Louis is as a reseller to the extent that there is any entry at all. There are plenty of people who can do resale. There was no intention of doing any facilities based entry, and it was not an attempt to get into the local exchange business. It was an attempt to protect the wireless customers. For that matter, if these assets are divested as they would have to be -- MR. ROGERSON: Rich, I was so swept away by this. You only have 20 seconds left. MR. GILBERT: All right. I got carried away too. Someone else can take over these assets and do the same thing. The bottom line, and I am going to finish, there is not a potential entry issue in this merger. MR. ROGERSON: Robert Crandall, is that vertical name plate meant to be? MR. GILBERT: I am talking, by the way, about SBC-Ameritech obviously. MR. CRANDALL: I think the reason why the Courts have been reluctant to accept potential competition arguments is they are, of course, speculative. The notion that even though the Commission is a specialized Commission with expertise in telecommunications, the notion that any of us in this room today can predict how competition is going to unfold in this industry is, it seems to me, presumptuous. I think those who are in the process of writing opinions, which unfortunately will not disappear in the next few years and will have to be revisited, might be in an embarrassing position if they tell us that the list of potential competitors is limited only, as Bob said, to four or five. The last time around in Bell Atlantic-Ninex, we were told that the potential competitors were only the two relevant ILECs plus the three interexchange companies. Cable companies were excluded, yet since that time it turns out that AT&T, in order to compete apparently, has to have the largest cable company in the country and sign up agreements with other cable companies. It seems to me that tells us that the cable companies, at least by AT&T's admission, and they may be wrong again, are in the market for potential competition. It may also be true that other media; for instance, I brought this along not because I plan to annoy you and use it in the room today while we are talking, but I bought this service when the cost had fallen to ten cents a minute. Since I bought it three months ago or two months ago, the cost of my service has fallen to 8.3 cents a minute. By the time this case is on appeal, if somehow these mergers are reversed, I will be perfectly willing to explain to you why it is that I have torn out my copper wires at home because the futures price of copper has gone up a bit, and I just melted it down, and I am using this alone. It seems to me that we are getting to the point where the bottleneck is contestable from wireless at a rapidly accelerating rate. It does not have to be Project Angel. It does not have to be any other fixed wireless companies. It just is this little handset, which I can buy for $100 and can change my service on a moment's notice. I think it is a very dangerous proposition to try to limit the number of potential competitors and base on opinion on that. Even if it gets through the Courts, you are going to be embarrassed by it in the future. MR. ROGERSON: Joseph Farrell? MR. FARRELL: Thanks. Well, actually I agree with quite a lot of what Bob Crandall just said; not all of it. I think it is obviously true in this industry that predicting the course of competition is pretty tough, and I think the question that that should lead to is what should you do about the fact that there is so much uncertainty? It seems to me one almost immediate reaction that an economist should have is you want to behave in such a way as to sustain the existence of as many options as possible, and it seems to me that does have an implication for what we are talking about here because it is likely to be much easier to allow these mergers after a year or two if it turns out that wireless really does bypass the local bottleneck than it would be to undo them if the opposite state of the world turned out to be true and if it turned out that there were bad consequences from the mergers conditional on the continuing local bottleneck. I would also like to just offer one comment on the way that people list potential competitors and treat the number of entries in the list as the relevant thing. That is not really quite right. What we should be looking at is whether there is a substantial probability that the merging party who is not currently in the market would turn out to make a big difference to the state of competition in the market. You could have an adjacent ILEC who is a potential competitor and 963 other potential entrants, but a situation in which the entry probabilities are not all that well correlated, and still have a substantial probability that the adjacent ILEC would make a big difference to the state of competition. I do not know enough to say that that is the case, but if you think that, for example, the argument that other ILECs, perhaps particularly adjacent ILECs, have particular expertise in negotiating interconnection or have particular forms of brand image or something, if those arguments are at all plausible then pari passu the argument is plausible that this precluded potential entrant would make a big difference. It does not matter how many other potential entrants there are who would be a competitive force in the other state of the world where those particular assets turn out not to be the key thing. I think I agree with Bob. We cannot really predict very well which of those states of the world is the case, so I think we need to think about how to make decisions, taking into account the uncertainty. It is unfortunate, I think, that the Commission seems to believe it is required -- whether it is or not, I do not know -- to write its decisions in a way that suggests we understand everything, and we know that this is the right decision. I think it would be a lot better in many cases, and this may be one of them, if the Commission were more comfortable with saying there is an awful lot we do not know. This seems like the prudent thing to do at this point, given that fact. MR. ROGERSON: Okay. I thought that was a good question. Joe said he is not sure that the ILECs are necessarily the only potential significant competitors, but it could turn out that way and so why do we not wait a few years and see at a minimum, right? Go ahead. MR. FARRELL: Wait and see also whether the existing local bottleneck continues to be a local bottleneck. MR. ROGERSON: Yes. MR. FARRELL: That is also relevant. MR. ROGERSON: Okay. MR. GERTNER: I think that that would be the wrong way to look at it. I think the reason why the Justice Department is wary about using potential competition is because when you are uncertain about the effects that whether or not the potential entrant will actually come in, then you tend towards allowing markets to operate. Now Joe wants to add another layer of uncertainty here, which is the layer we do not really know what characteristics of potential competitors are actually going to be the relevant ones that are going to determine who is going to be successful competing in these markets. Therefore, what we are going to do is we are not going to allow companies to do what they think in fact is in their interest and, therefore, most likely absent tangible anti-competitive effects to be in consumers' interests as well to go forward. I think if you look at it, you have to say what is it that are the unique characteristics of the merging parties that give them some potential benefits I think when you go down that list. The fact that they are an ILEC; well, there are lots of other ILECs, and in fact others trying to enter this market are doing it by buying CLECs rather than buying ILECs, so that does not seem to be it. Is it proximity? Well, there are lots of companies that are nearby, lots of other competitors nearby. Why does proximity matter anyway? You say maybe it has to do something with brand name. You have to ask how important is that, given that they are not actually serving the customers. In fact, it is knowledge or having existing customers that is most likely to be the source of an advantage. I think when you look at it and you try to identify the unique characteristics of merging parties outside of their region, there really are not any that would make you conclude that it is likely that they have some unique position to have an impact. MR. ROGERSON: Okay. I want to pursue this just for one more minute. I think Rob has said it is not very likely at all that ILECs are likely to be one of a small number of significant potential competitors. He cannot think of too many things that they have that are likely to really matter. Could I give Joe a chance to respond to that? MR. FARRELL: Well, I would come back to my confidence statement that we do not know. I think, you know, the statements that have been made from both sides of this debate that large ILECs can be more efficient in various ways than small ILECs have some implications for whether little CLECs are going to be a full replacement for the kinds of things that a large ILEC might decide to do in some national/local strategy. I think Dennis has harped on -- let me take that back. Dennis has stressed a view of the national/local strategy that says there is no evidence that the firms would undertake it individually, and then he has given some arguments why that might be less plausible, but I think the bottom line from that is not we should assume it will not happen, but we do not know whether it will happen. Again, I think, you know, I have not looked into this particular question deeply, so I cannot really give you a bottom line, but I think the right thing to do is to investigate with a very strong consciousness of how much is unknown not whether it is more plausible or not or not whether it is convincing or whether we should assume, but whether there is a sufficiently big chance that it would be imprudent to ignore that any particular potential entrant will turn out to be important. I think that is the right way to frame the question. I do not know what the answer is, but I think that is what the staff and the Commission and, for that matter, the Department of Justice probably should be doing. MR. ROGERSON: Okay. My neck is getting sore turning this way, so I am going to now turn to Robert Litan for a moment, who has been patiently waiting over here for his turn. MR. LITAN: Yes. Thank you. You do not know whether an ILEC is likely to enter until you look at the documents, all right, at least when we are looking at adjacent ILECs because I do agree that if you are not adjacent I do not see any evidence at all from the last two years that non-adjacent ILECs are interested in entering. We come down to documents. We know in the case of Ameritech there was documented interest in crossing boundaries and going into SBC's territory. We know that. I do not know the record in Bell Atlantic-GTE as well, but at least on the surface it seems to me they have less of a problem than certainly Ameritech would because GTE is dispersed all over the country, and it seems to me less likely that Bell Atlantic would be interested in going into little pieces of GTE's territory. In any event, this comes down to documents. That is point one. Point two is in a way that is all irrelevant because I do agree with Rich that if you count the three big interexchange carriers, you have that plus the adjacent RBOC. You have four already, and that is going to at least eliminate any Clayton 7 standard right there, a Clayton 7 challenge, and it is going to be hard to make the argument that the FCC but for one fact, which is the third point. That is we talked about it earlier. What happens when we get down to one RBOC, all right, or one ILEC? I mean, the thing that still would trouble me very much if I were sitting at the FCC is these mergers go through. What rationale would the FCC then use to stop the next mergers? If there is no such rationale, then why not go the limit and go all the way to one? If that is where we are going, then it seems to me Joe was asking exactly the right question, that knowing that it is legitimate, it seems to me, for the FCC to take into account the uncertainty about all this, and you really then have to believe that there is a very strong likelihood of a pro-competitive effect. You have to believe Dennis' story that they are going to go into 30 other cities in order for you to overcome this nervousness you have about collapsing to one ILEC. I mean, I have been persuaded that that is the case, but if I were in the Commission that is the balance that I would have on my head because the end game here, it seems to me, cannot be ignored. MR. ROGERSON: I am going to Dennis Carlton and then to Roger and then to Rich. MR. CARLTON: I have three quick points. The doctrine of potential competition has fared poorly because it is very hard to predict the future. To give up certain benefits for something that might occur in the future just turns out to be very difficult. Now, I would like you to especially recognize that in a rapidly changing industry predicting who is going to even be the leader in that industry is not so easy, let alone who is going to be the participants. I once had occasion to work on a merger that was unsuccessfully stopped in part on the grounds that the two companies would engage in new innovations and compete harder against each other than if they were merged. That was five or six years ago, and I am still waiting for those new innovations to occur. I will not embarrass the person at the Department of Justice at the time who I told this to. When you go through rapidly changing industries and try and predict, just go through five years ago and look at the list of people who are in telecommunications today and ask yourself would you have predicted some of these names? I think the answer is it is pretty speculative. It is pretty hard. Second, I think it is easy to say let's be careful. Let's wait. That sounds like you are being careful, and there is no cause. You have to recognize that by being careful and waiting, what you are really doing is making life easy for a regulator. You are not making life good for consumers. My interest is in making life good for consumers, which may require regulators to make difficult decisions. It is an easy decision to say wait, let's see in the future. What you are depriving consumers of, though, in the meantime could be significant, very significant, benefit. Finally, to get to Bob's question about one, I think the answer is one. I believe what these companies are saying. These two companies that are merging are going to be horizontal competitors. I would say no to that. I mean, I do not know. It may get me in trouble in the future. They may not hire me, but that is life, you know. One is too few. They are horizontal competitors right now. MR. NOLL: The lawyers did not read that one. MR. ROGERSON: Roger Noll? MR. NOLL: I actually just want to ask some questions because it seems to me that by revealed preference we know that Ameritech wants to be in St. Louis and Southwest Bell wants to be in Chicago, or they would not be proposing to merge. Number one, if that is true then each one wants to merge in the profit maximizing way, wants to enter in the profit maximizing way. Combining and having roughly 80 percent of the customers be monopolized is going to be better for them than going in and sharing those 80 percent in two 40 percent hubs, assuming that were feasible. So, it seems to me that we know that they want to be in other territories by virtue of the fact everybody wants to merge with everybody. Secondly, we know that by far from the company point of view, the most attractive way to be in other places is to actually acquire someone who is there. The absence of direct entry by RBOCs into other RBOCs' turf strikes me as fairly understandable from our normal strategic theories that they think that it is very possible that in the long run they will be able to go to the two or three ubiquitous ILECs, each of which in its own service area has 80 percent of the market, and that as long as that strategic possibility is available they do not want to do anything that makes it less likely to happen. If the FCC were to say for certain we are done with large ILEC mergers, there will never be another one as long as we live, then conceivably Ameritech's attitude about acquiring one of the 973 CLECs in St. Louis, that that calculus would change. In particular, the question that I have is if there is this natural efficiency advantage of a large ILEC providing service in St. Louis by merging with Southwest Bell, say Ameritech, why is there not also the same efficiency advantage for them buying one of these 973 guys who are already there and having the integration, the technical sophistication and the original advantages that the proponents of the merger say apply to a much littler company where they might in fact be substantially greater? It strikes me that what is really the reason that Ameritech does not want to acquire one of these little, tiny guys in all the big cities in Southwest Bell turf is not because of some relative efficiency advantage. It is because they would prefer to have a monopoly position throughout both territories than to be a competitor, a large, significant competitor in those territories. They are not going to do the competitive strategy unless the monopoly strategy is foreclosed. MR. ROGERSON: Rich Gilbert? MR. GILBERT: Saying no is not without cost. What are the costs? The costs are that you give up the range of services that could be provided under this national/local strategy that SBC-Ameritech have and a version of that that Bell Atlantic has as well, so you give up those benefits. Also, there are other costs as well, which is you do not have a good horse race when you let three horses out of the gate first and then you tell the rest of the pack to wait and see what happens. It would be best to have competition and end to end services occur simultaneously with everybody throwing as much at it as they can. Now, it is important in my view, very important, to distinguish what I call conventional CLEC style entry from the national/local strategy. I think this is a key issue in the merger. If you look at conventional CLEC style entry, that just means somebody goes into an area, tries to capture some profitable customers, may put in a switch, may put in a few switches. There are lots of folks that can do that. There is no evidence that I see that adjacency is particularly important for that because none of the RBOCs have ever said that they would go into these adjacent territories because they have excess capacity in their networks and they are adjacent networks and they have, therefore, low costs, incremental costs of going into new areas. The fact is they do not. If they went in, they would either go in as resellers, or they would go in with new facilities and usually without much brand name recognition in any case. What you really have to contrast here is the benefits of a national/local strategy type entry against what are probably insignificant issues in terms of conventional CLEC opportunities without the merger. That would be just essentially no effect on the conventional CLEC competition that would occur without this merger. MR. ROGERSON: I am going to take two short comments from Michael Katz and Rob Gertner, and then we will take a couple questions from the audience. MR. KATZ: I agree with people who are saying that potential competition analysis is difficult. There were two points they were making about that. One, that plans can change. It is impossible to predict the plans of even a single carrier. The other is that there in fact could end up being a lot of rivals in the future. The only point I want to make here is it seems to me that that applies then with equal force to the earlier arguments we heard this morning about the benefits of the national/local strategies and the expansion. Even if we take it as given that the mergers are necessary and sufficient to get that expansion, it seems to me the arguments being raised here then have to be raised there as well of wait a minute, is it not possible there are going to be all sorts of rivals coming in so in fact that these parties are just a little blip on the radar screen of no real significance and so we should not give very much weight at all to the fact that they will be able to do this? It seems to me that is just a mirror image of saying taking them out, even if it were true, would not matter. I am not saying that I have analyzed that issue. As I said, I agree that potential competition analysis is difficult, but it means we have to do the full analysis, and those questions have to be addressed. MR. ROGERSON: Rob? MR. GERTNER: I just wanted to make a quick comment about some aspects that are specific to the Bell Atlantic-GTE merger. It is important to remember it is not just an ILEC buying an ILEC, but in fact a large part of the benefits from the merger come from the complementary assets of GTE's GNI fiber optic network and their Internet backbone, which is an important part of their entering into local markets. To some extent responding to what Roger Noll said, I mean to some extent GTE represents a relatively small presence in a wide geographic variety of markets, which is sort of getting closer towards his alternative entry strategy that he was suggesting. MR. ROGERSON: Okay. Are there any questions from the audience? Quinn Trong? MS. TRONG: I would just like to hear some comments on what appears to be some inconsistencies between the different positions taken by the applicants. On the issue of benchmarking, applicants say that the entry out of territory by these BOCs and by GTE allow increases in approaches because they are particularly well qualified to negotiate better interconnection agreements, and then with regard to public benefits in general there is a claim that having this base of anchor customers will facilitate the CLEC strategy and also particularly with regard to GTE that they are well equipped to expand into adjacent areas. Now, these and other claims from the applicants would seem to indicate that they are particularly well qualified, uniquely well qualified competitors as compared to other companies, other CLECs, so how does that cut to the position on this issue where they say, you know, there is no worry about eliminating these potential competitors? MR. ROGERSON: Okay. Rich? MR. GILBERT: I cannot speak for GTE, of course. I can tell you that I have not seen any evidence that, for example, Ameritech moving into SBC's territory or vice versa would result in extraordinary advantages for that entrant. Neither Ameritech nor SBC would, if they did that, serve those areas with facilities that they have in region. It would be either reselling existing services or setting up a new plant entirely. Neither one has a particularly good brand name. It is still the case. It is still the case that most people think that AT&T is their local exchange provider, and so the IXCs have, if anything, better brand name recognition. At the very least, if that is your view that there is an advantage there in the SBC-Ameritech case, you have to consider all RBOCs, all local exchange carriers, as potential entrants into those markets. MR. ROGERSON: Okay. We are going to wrap up that session and move on to Session Four then. Session Four is will these mergers have any effect on the ability or incentive of ILECs to raise rivals' costs. I have asked Michael Katz to first explain what these mysterious code words mean, raise rivals' costs, and then Dennis Carlton will comment on Mike. MR. KATZ: Thank you, Bob. I want to be clear about one thing because we are talking about raising rivals' costs, and the rivals part I hope is clear, but I think there will be agreement here that the important point is not actually are the firms, the competing carriers themselves, better or worse off. Obviously the carriers care about that quite a bit. The issue that the Commission should concern itself with is not the harm to competitors. I think we all agree it should be whether there is harm to competition because that is what is going to end up harming consumers. I hope that that is not a point of debate. What I would like to do is go very briefly to the factual and logical analysis underlying the conclusion that these mergers do pose a threat, and we expect it to lead to a harm to competition. As I go through it, I want to address a couple of things I think are counter arguments that really are misunderstandings of some of the claims. Let me, as I said, walk through the steps. I think step one is that incumbent LECs possess significant market power in the provision of access services to actual and potential rivals. I want to be clear that in talking about access, I mean that broadly. I mean things like interconnecting the networks, but also making the OSS work a cross system, various forms of originating and terminating access, unbundled network elements. The fact is the networks have to work together. I think it is also a fact that they do have market power unless they are worried maybe that that Linex guy is a threat to them, too. He is apparently very powerful, despite being very small. Maybe he is actually a CLEC in disguise. The other thing I want to point out, because I think there has been some confusion on this, is there are two ways this could happen when they exercise market power. One is by raising prices, and then, of course, you run into the issue of well, is most of this stuff not regulated? I think that is right. The other is by either denying or delaying or degrading access, and that is the part I think there has been some confusion about because at least personally I am not that worried about access arrangements suddenly getting much worse than they have been. I think the concern is really what happens going forward with whether new forms of access are made available as quickly as they should or at the quality levels they should, so I think the things to think about really are something like say XDSL and the whole question of the problems with roll out. Are some of these problems strategic rather than inherent in the technology? Okay. It is not about, which I think has been accused of claiming, that as a result of the merger they are suddenly going to get, you know, white noise generators and start attaching them to people's lines because I think a lot of that would just be too obvious. The second leg in the argument is I think that regulation is an imperfect check on the exercise of ILEC market power. I actually had not expected that one to be controversial, but it has. I want to address again I think something that really is a misunderstanding or just as illogical. The following claim has been put forth by several parties. They said wait a minute. The way this would have to work is the consumers and rivals would see service quality worse than it should be. Wait a minute. If that is true, if everybody in the world can see it, even the lowly end user consumer, then surely the regulators could see it. What is fundamentally wrong about that argument is it misses the point. Everyone would be able to see poor performance. I think everybody today who looks into it can see the problems with CLEC/ILEC OSS interfaces, or people can see the problems with rolling out XDSL, but that is not the issue from the regulators' point of view. That is what consumers care about. They just said wait a minute. It does not work very well. That is going to affect our choices. What regulators have to concern themselves with is why is it happening, and that is not something consumers care about. No one is going to go to Sprint or AT&T or MCI and say, you know, your service is really terrible, but since you have explained to me that you believe it is actually Bell Atlantic's fault, we will stick with you. Okay. It is not going to work that way, but that is the kind of thing that regulators have to look at that, and I think that is the really hard problem that is inherent in this and that makes it difficult and in fact impossible for regulation to fully constrain ILEC market power. I think really this is just a corollary to the existence of market power. The next step that can be exercised I think is to significantly weaken competition. Local and long distance carriers and carriers providing bundles of services are going to be dependent on ILECs for significant portions of their access and that they will be weakened as rivals if that access provided by ILECs is degraded. Now, you ask what does any of this have to do with the merger, which is certainly a fair question and one I am sure we will be debating for awhile in this session. The reason it has something to do with the merger is there are significant competitive spill overs across ILEC regions. Now, a couple things that need to go into that. One is that national rivals are important. In fact, we have been hearing that from the proponents of the merger, and I think there is probably agreement on that that national rivals are the strongest competitive threat to everybody in the market. I think there also would be agreement that there are significant benefits to national scope. Those benefits come in because there are network effects at the subscriber level. There are network effects in terms of third party application vendors. They want to see a broad network with a broad potential market. It is harder to put one's finger on, but there are also word of mouth networks that people in marketing will certainly talk to you about. Lastly, there are economies of scale and scope that arise both in development of systems, in development of marketing programs and being able to take advantage of national advertising. For all of those reasons, you have that competitors want to compete at a national scale, the really significant ones, the other ones that are important, so what that means is that if you are weakened -- MR. ROGERSON: You are going to have to wrap up -- MR. KATZ: Okay. MR. ROGERSON: -- in 30 seconds. MR. KATZ: I have one point left. What it means, though, is if you are going to be harmed in one region, that is going to weaken your ability in the other. That actually brings up another criticism that has been made of the model or the logic, which is to say well, wait a minute. If you harm someone in one region, will they not just run over? Will that not speed them up in other regions? I call that sort of the have switch will travel model, which is sort of the little CLEC that in the back of a van or whatever they have their switch. If it does not work one place, it goes to another. Now, the fact is that is a legitimate, logical argument. It is something that needs to be addressed. I think the answer to that is, though, that the big competitors want to go in nationally. They feel they need to as a viable business, so it really is if they are weakened in one place it is slowing them down and weakening them overall. Finally, just the last step in this is you do have these competitive spill overs. What the mergers do is they help internalize those. They allow greater coordination among the parties, and that is the mechanism to which you see the harm coming. MR. ROGERSON: Dennis? MR. CARLTON: Thank you. I think Michael has laid out the issues fairly clearly, so let me try and respond to several of them. There is just no reason, in my view, that the fear that this discriminatory behavior arising should stop the SBC-Ameritech merger. The claims that significant discrimination will arise, exists now or will arise, as a result of the merger just have no factual support. I would like to explain that. I would like to at the outset, though, state quite clearly that the issue is not whether regulation can fully constrain all possible instances of discrimination This is a merger case. The question is whether this merger will raise the incentive to discriminate. The whole question of discrimination that Michael raised is something that has nothing to do in particular with this merger or the other one under discussion. It is argument about really whether you think the Telecommunications Act of 1996 was a good idea or a bad idea. In my view, I think the decision was made correctly that appropriate regulatory safeguards exist not to prevent all possibilities of discrimination, but sufficient safeguards exist so that on net RBOC entry into other services can be expected to benefit consumers. Again, I want to stress the purpose of this proceeding or any decision is not to make life easy for regulators. It is to benefit consumers. Really the primary objection that Michael raised about this merger raising concerns about discrimination is, as he articulated it, that as an ILEC's area gets larger it will have a greater incentive to discriminate. I would like to submit to you that there is absolutely no evidence presented in either his opening remarks or his testimony with Professor Salup to support that theory. First, let me just return to a point he made. I think it is a correct point. The problem with a lot of arguments about discrimination is that they are detectible. If consumers know they are being taken advantage of, they can certainly report it to the regulators. Now, I agree that means the regulators have to investigate it, but that is a much different problem than having something secretly occur in the switches that disadvantages one person relative to another. I think it is clear, and I am glad, you know, it is now clear, that that is not the concern. The concern is a detectible harm to a customer, and now a regulator must investigate. Second, let me turn to evidence for a second. If you look at the evidence about discrimination, what you see is massive entry of CLECs and IXCs. That strikes me as a vote that they are willing to bet on the regulators to protect them from discrimination, that it is not that serious a problem. If you look at an area where the Katz-Salup theory should apply in which you should see discrimination, say intralata toll, you do not find it. If you look at another area where you should see the Katz-Salup theory apply, cellular, again you do not find it. The FCC is on record as saying they find the business split about equally between wire line and non-wire line carriers. Specifically the Katz-Salup theory says after a merger there should be an increased incentive to prevent entry. Well, we have had some mergers. Has anyone investigated whether that has occurred? I do not see any empirical analysis in the Katz-Salup affidavit. I have done some preliminary work analyzing precisely that question, and the answer is strikingly clear. The answer is no. A statistical study that I have done shows that the amount of entry you see of CLECs, the number of CLECs, is no different than what you would otherwise have expected. The bottom line is there is no support at all for the theory that as an ILEC gets larger it will discriminate more, just no support at all for that. As far as the theory that new technologies have to be hooked up and that is where questions will arise, I believe correctly, as Michael has pointed out, that is something you have to be concerned about. The best example I know of that is the Sprint Ion case. That actual example I think belies the concern. It is my understanding that Sprint has announced that they are signed up to go forward with Ion. They are not concerned about relying on third parties, and, therefore, I do not see this as a serious concern. Finally, let me make one point. The national/ local plan that SBC has embarked on requires SBC to provide in region long distance service. That means it will have to satisfy the 271 checklist. If SBC is found to have discriminated, what a penalty it will have to suffer. It will not be able to provide in region service. That will be a disaster to SBC's national/local plan. SBC, therefore, will have a lower incentive to discriminate, not a higher incentive, as a result of this transaction. I guess simply put -- MR. ROGERSON: When you said finally, then I did not say 30 seconds. MR. CARLTON: Okay. This will be 30 seconds. Okay. MR. ROGERSON: You want to simply put one thing? MR. CARLTON: Yes. Simply put, the goal of regulation is not to make life easy for regulators. It is to benefit consumers. The goal is not to eliminate all theoretical possibilities of discrimination, no matter how insignificant, without regard to the benefits of the transaction. There is simply no evidence whatsoever here that this merger, the mergers, will raise the incentive to discriminate against rivals. MR. ROGERSON: Roger Noll? MR. NOLL: Let me just briefly say that I want to make two points. First of all, the Telecommunications Act of 1996 does have a 14 point checklist and, in addition to that, a public interest standard. It is by no means clear the Telecommunications Act is all about the RBOCs expanding into other things. I think it is mainly about introducing competition into the local service or else the checklist and the public interest standard would not be there. It strikes me this is a beautiful example of a difference in policy perspective that the FCC might have than the Justice Department would have. The Justice Department would never ever in a million years put an anti-trust case about well, if we prevent somebody from being in Market A it makes it more likely they will be in Market B. The best you can hope to in a potential competition argument is to talk about the same market, not a completely different market. In the case of the FCC, the strategy here is more I think having competition in local service than it is having additional competition in long distance and enhanced services. Then when you get to what Michael's point is all about, Michael's point is really not, Dennis, about does the incentive of Ameritech to discriminate against CLECs in Chicago increase if Southwest Bell is part of the same company. I cannot imagine. Maybe it is true there is an effect, but that is not what people are worried about. What people are worried about is in fact the increased ability and incentive to engage in discrimination if you are in the enhanced services long distance business and you are both the originating and terminating carrier. We do not have any examples of that because we do not have the RBOCs in the long distance based enhanced service area and so there obviously is not going to be any evidence one way or the other. That means this is another important point. Who bears the burden of proof here determines the outcome. If the burden of proof is on those who oppose the mergers to demonstrate the existence of long distance discrimination being enhanced by the mergers, they are going to be unable to prove it because the opportunities for that discrimination in principle do not exist because the RBOCs are not in that business. If the burden of proof is on the RBOCs to prove that indeed they have competed successfully in long distance without discriminating against AT&T and MCI and everybody else, then they are going to fail that burden to prove, so you cannot really put the question that way. You have to put it, I think, the way that Michael has put it. I think his way of analyzing it is absolutely right, and it comes down to the case that there is this one important fact. It does enhance the incentive to engage in discriminatory behavior in a dynamic sense, the one that Michael described, if you are both the originating and terminating carrier and if you are in the long distance business, so you can in fact be the person doing the interconnection. That is a world we are not in yet and so we have to base it largely on theoretical, as opposed to experiential, bases. MR. ROGERSON: Jeff? MR. SHEPERD: Very briefly. Two short points. One, I heard a lot of nos and none whatevers just now from the other end of the table. With all respect, let me say I am the editor of a journal in this field in which we look for evidence and factual testing of hypotheses. When I hear no and none I think well, that is theory. That is not economic testing based on evidence. MR. NOLL: I was referring to my empirical study. MR. SHEPERD: Well, I will stick with what I said. The other point is that price discrimination, whenever it occurs, and done by a dominant firm does tend to be anti-competitive. It may be detectible, but after some lag. It may not be easy to cure it, but in general it tends to be anti-competitive. MR. GILBERT: If I could respond? MR. ROGERSON: Yes. Rich Gilbert? MR. GILBERT: Roger said that we have no experience with RBOCs on the origination and terminating end of long distance calls. I think that is wrong. We have lots of experience. They are called intralata toll calls. That is the case where the RBOC is both on the origination end and on the termination end. The experience that I am aware of, for example, in the Pac Bell region the Pac Bell share of the intralata business toll market is under 50 percent, and Ameritech, I understand their share of the entire intralata toll market is around 65 percent. In Pac Bell, I say business because we do not have pre-selection for residential yet. That is I think a regulatory issue. I think that is very, very strong evidence that there is not a problem. That is where you would expect it to occur. There have been no complaints in that area. Another thing I would like to mention is that Sprint has recently announced that they have secured access for their broad band metropolitan access networks in a number of states, including SBC states. They seem to be quite happy with it. For what it is worth, I think it is worth noting that the Texas Public Utility Commission has announced that SBC has satisfied ten of its 14 checklist items. I am not sure that is directly related to this discrimination issue, but I think it would be very hard to look for any evidence here. MR. ROGERSON: Okay. I cannot resist giving Roger a chance to respond to the point that there is evidence in the intralata market, interstate intralata markets. In particular, the evidence is that 50 percent of the market goes to someone else. MR. NOLL: It is obviously true that the fraction of the market accounted for by the interexchange carriers intralata has grown substantially, particularly since the passage of the Telecommunications Act of 1996. Rich, it is just completely false to say there has never been a controversy about intralata access. I mean, I do not know how to evaluate it. I do not play the game of advocating one side or another. I do know that in almost every state there has been constant battles for ten years on this issue. It is also the case that the local exchange carriers have had to be forced through long-term regulatory proceedings to do anything to accommodate intralata competition. I mean, it is one thing to say, and I agree with you completely, has it been possible through a guerrilla war that has lasted for 30 years for the interexchange carriers to get a significant fraction of the intralata market, especially in some states that have relatively large and sophisticated regulatory commissions. The answer is yes. Is it true that this all just sort of happened easily and ubiquitously without a fight and that the ILECs and the RBOCs in particular were just real, real happy to accommodate this and do whatever the IXCs liked? That is completely fatuous. No one could possibly believe that is true. MR. GILBERT: That is not the point. The issue is does the merger contribute to a regulatory problem? Does the merger make it worse? I think if you look at intralata toll, for example, I would be very, very surprised if you saw any correlation between the size of the intralata toll calling that is controlled or on which there is control of origination and termination by an ILEC and any pattern of disputes over the terms. MR. LITAN: Could I just interject? We have a battle of speculations here. We started out this morning with the claim that with a nationwide footprint there were enhanced incentives without evidence that there would be a pro-competitive effect. We are now ending the discussion with an argument that there are enhanced incentives with an argument about evidence that you will have the opposite effect. I suggest at a minimum they cancel each other out and that right now where we are left with is no shred of evidence that this merger will be pro-competitive. Then the question is is that enough to get you through, and that becomes a legal question. MR. CARLTON: Could I just say something? Can I just address that? I would like to actually agree with something that Jeff said. When he was saying no, no, no, it is speculation on both sides, I think he was referring to you just do not want to have a battle of theory saying on the one hand, on the other hand. The point I was trying to make, and maybe I was not clear, is that if you look at the evidence, if you do the evidence and look at the prediction of the Katz-Salup theory based on the mergers that have occurred in the past, I can say there is no evidence. I underline no. I was not saying based on theory on the one hand, on the other hand. I think there is evidence. I think intralata toll provides evidence. I think cellular provides a counter example that discrimination is not to sever as to impede. There is no allegation I think by anyone here -- I do not know; I will speak just for myself -- that regulators should go home. No one is saying that regulation is not important still in these areas. That is not the issue. The issue is does this merger raise the likelihood of discrimination, and is there any evidence to support it. The evidence I have looked at, which I think is exactly the relevant evidence to look at, says no. MR. ROGERSON: Right. Okay. Michael Katz, what about that? MR. KATZ: I actually want, and I have been waiting a long time, to just address a bunch of the things that Dennis said, some of them very quickly. The point about this is harm that people are going to see again I think comes back to people will see the poor performance. How are they going to know? How are people going to know what are the problems with OSS making it work? How are people going to know the problems with XDSL? Now, on a couple of these things, the intralata toll, the cellular and the points about Ion, I am tempted to say especially to Rich, since we are both from Berkeley, that is your reality. (Laughter.) MR. KATZ: I will talk about mine. Apparently Dennis talks to different Sprint people than I do. I thought I probably talked to more of them since they paid me to come here. One of the reasons they paid me to come here is in fact they are worried about making sure that they get good access. Rich talked about signing up agreements for the BMANs. I think that is not so surprising. I think there is more competition if you are talking about that kind of access. I think the big issue, though, is XDSL, and I want to point out that XDSL is not just about saying that that is small customers and it does not apply to large customers because in fact what we are talking about here is national entry, and a lot of large enterprises, a lot of large customers, have a lot of small sites so access services like XDSL are relevant for those markets as well. I will not go into the details of Sprint's concerns about these things. I mean, the appropriate way to address that, it seems to me, is for the Commission to meet with Sprint executives and, as I mentioned earlier, figure out a way to talk to them about it that is also kept confidential since it obviously involves sensitive strategic things. The point about intralata toll and cellular. I disagree with Roger. It certainly has not been an uncontroversial issue. Certainly our example I think of problems with cellular interconnection and certainly with intralata toll, I mean, Rich has labeled some of it as just a regulatory issue, but you could say that about all of these threats. They are just a regulatory issue. The fact is the ILECs have not embraced intralata toll competition with open arms. In terms of Dennis' point about massive entry, the fact is there may be a bunch of them, depending on how you count the numbers, but it has still been on a small scale. Dennis made the point about that they do appear to be betting on regulators. In my notes I make a point the other way, that the ILECs also seem to be betting on regulators because in numerous instances they have tried various forms of discrimination and gotten caught where they have settled the cases, so apparently they are also betting they can fool them. The 271 checklist. I actually think that is an example. The other way the argument has been made, why would the ILECs dare do bad things and keep out of interexchange? Well, apparently they have not been too tempted so far. None of them have satisfied the checklist. I think it is not so surprising; at least the data I have looked at have shown that the margins are much bigger if you can stay in local and avoid competition than if you get into long distance, but have to give that up. Now the really big issue. Actually, one thing before I get to that so this way Bill cannot shut me up. On the balancing point, I actually think that Bob is wrong in one sense. If you are going to talk about stopping the merger or letting it go through and those are the only choices, then it is absolutely right. I think everybody here agrees you have to balance all the effects, and you have to weigh which ones you find plausible, but that is not the only option open to the Commission. VOICE 2: No. There are conditions. MR. KATZ: There are the conditions, which then can address things without balancing them. You have to do a balancing sort of within it to make sure the conditions do not cause their own problem, but it is a narrower inquiry. As to the stuff about the evidence, I want to raise a bunch of points, and then I will stop. First off, I would point out that we have been asking for evidence. Well, is it really true that these mergers are going to allow you to go national? Everybody has said no, no. The previous mergers were too small. Well, they make just the same argument here. If you look, the number of access lines controlled by one party, just to throw a couple numbers in SBC's case, it is going from 19 to 30 percent, Bell Atlantic from 22 to 33. There is significant increases, and people are debating what constitutes significant, in the amount of traffic that will terminate in region if these new expanded regions go through. I would point out that, you know, the mergers have been comparatively recent. It makes time series difficult just because you have limited data. It is also difficult because to do it right you need to take into account industry trends. Rich Gilbert and Bob Harris had a filing that said look. If I got it right, it said the number of co-location agreements has gone way up post merger. That is true, but it has also been going up in the rest of the industry. What one needs to do is correct for industry trend. Now, I have looked at some stuff very preliminarily, and to try and get around this problem of having a short time frame we have looked at some cross sectional comparisons to ask well, do small ILECs seem to perform better or worse than large ones? Is there evidence? I will tell you the studies are preliminary and some of the results are mixed, but it has also been coming out that by some rather imperfect measures the small ILECs, for instance, net before the merger of Cincinnati Bell, have had more entry on adjusting for market size. I think that doing the empirical work here is very difficult, but I think it is wrong to say oh, this all supports the merger and shows it is fine. I think in fact there is evidence there. I think there is reason to believe that these mergers really are different from what we have seen in the past because they are getting so much larger, and that matters both because of the internalization and also because of the concerns the rivals face because the national rival needs to be able to cover a given percentage of the market. I think it is hard to predict these effects. I agree with that. It makes it a difficult task for regulators, but I do not think that means you can avoid doing that and forming the theory and I think then testing every part of it you can. I think the other way to do that is to ask each of the steps in the logic. Is that right? Do they have market power or not? Talk about the forms of access where they do. Is national entry important? How worried are the competitors? Those are all things that are being done, and the FCC can do more of talking to people in the industry. MR. ROGERSON: Bob Crandall? MR. CRANDALL: First of all, the intralata debate. It seems to me it is obvious that all of the LECs would like to keep their intralata monopolies as long as they could and, therefore, are opposed to one plus equal access. The question really is once you have one plus equal access, is there any evidence that subtly discriminatory policies allow them to obtain a disproportionate share? Secondly, Roger, why is the wireless example, the cellular example, not dispositive evidence or at least good evidence that even when they originate and terminate the calls they are unable to engage in subtle discrimination that, for instance, in the most recent Spectrum cap filing I just looked at the other day one of the advocates pointed out that Ameritech, after all these years, is number two in cellular in Chicago? You would think that given the number of people it has working in Chicago, just their own subscriptions would put them over the top there, but they are still number two. Why is that not evidence that they are not engaging in subtle discrimination? MR. NOLL: Okay. With regard to the subtle discrimination point, the obvious point is sort of just like Michael said. To do this thing right, you have to take into account everything it affects, market shares and prices of the firm. You know, actually I think it is the FCC's staff that is the only hope here because they have all the data. They have a lot of resources, and they do not have to be advocates. Here is sort of what we know. Number one, we -- MR. CRANDALL: You have graduates students, though, Roger. MR. NOLL: Yes, but I have to pay them, and that is harder. Number one, let's look at long distance per se. The argument in favor of allowing the RBOCs into long distance is that in an industry in which the former monopolist has approximately half of the market and a couple of other large competitors have about 40 percent of the rest and there is a bunch of little, tiny guys that account for the rest, it still is the case we do not have real competitive pricing. It still is the case that about a third of the customers accounting for something on the order of five to ten percent of the calling pay the high book price instead of the actual price you can get from minimal effort of calling around for competitors. The argument of the RBOCs, which is correct, is that probably if two or three RBOCs were also in the long distance business as facilities based carriers or as really aggressive resellers that there would be more efficiency. Yes, competition has had benefits for consumers in long distance, but the market structure is not sufficiently competitive that it has driven the price down to something like a competitive price. Then we say okay, what is going on in intralata toll? Well, in intralata toll the best we can find, the very best we can find, is something that looks sort of like long distance. That is to say the incumbent local exchange carrier has half. In the vast majority of the country, it is much more than that. Secondly, on a mileage based basis price cost margins are higher for intralata toll than they are for long distance, so if anything that market is less competitive than a long distance one. Now we ask the magic question why? Well, it could be that it is just a superior and wonderful efficiency, right, of the local exchange carriers. That is possible. What we need to do is actually figure it out if they really do have some gain here, but the reality is the prices are higher, not lower. If they really did have superior efficiency, we would expect the intralata toll prices on a mileage based basis to be lower, not higher, than long distance. Finally, with regard to the cellular story, the crucial fact here is yes, it is the case. You can find examples of specific RBOCs that in specific cities are not the dominant cellular carrier, but in most cases it goes the other way. The vast majority of cases it goes the other way. It is in fact the local exchange, the large local exchange carrier cellular company that has the largest amount of market share and that makes the most profits. MR. ROGERSON: Rich Gilbert? MR. GILBERT: Just briefly. I do not see, Roger, why intralata toll or any telecommunications service should be priced by the mile. The wires are there. Electrons move. They do not -- MR. NOLL: The reason I said per mile is because -- MR. GILBERT: -- consume a lot of gas. MR. NOLL: -- it actually favors the RBOCs to make it on a mileage based basis. If I do it on a total call base basis, then the difference is even larger. MR. ROGERSON: Okay. I am beginning to get worried that if we wait any longer we will be debating to an empty room. We have reached nearly 12:45 p.m., and I think the discussion has been extremely productive. We have all learned a lot. I would like to thank all of the panelists for their really insightful comments and I think for really objective comments, too. I think there was a lot of agreement among all sides here on all sorts of issues, and it really clarified, you know, what it is we have to investigate. Finally, before we close I want to particularly thank two of our senior staff economists who are working full-time on these ILEC mergers, Pam Magna and Marilyn Simmon, who are at the front here who very graciously took a number of days out of their schedule analyzing these mergers to help me organize the round table. They did just a first rate job, and I really appreciate it, so thank you very much. Thank you all for coming. (Applause.) (Whereupon, the round table was concluded.) // // // // // // CERTIFICATE OF TRANSCRIPTION I do hereby certify that the foregoing transcript was typed by __Karen Stryker______ from a tape recording furnished by the FCC, and that said transcript is an accurate record of the tape recording provided by CACI to the best of my ability. Date: February 8, 1999 Karen Stryker Transcriber Heritage Reporting Corporation Suite 600 1220 L Street, N. W. Washington, D. C. 20005