Thinking About Merger Policy During the Process of
Electric Power Restructuring
JOEL I. KLEIN
Federal Energy Regulatory Commission
Let me begin by thanking Chairman Hoecker for inviting me to participate in the Commission's Distinguished Speakers Program. I welcome the opportunity because, first of all, it has given me a chance to think through some important issues of mutual concern to the Antitrust Division and Commission during what is an exciting and challenging time in restructuring the electricity industry. As importantly, I hope that the Chairman's kind invitation and my presence here today will signal a commitment by both agencies to work closely as we deal with issues concerning electricity restructuringissues whose difficulty will be outpaced only by their importance.
In the past, I fear, communication between our two agencies has been too infrequent and generally limited to formal public hearing and comment procedures. I understand that there was a useful, informal meeting between our staffs last month and I hope that is also of harbinger of things to come. For my part, I will tell you that I have learned a great deal during the process of implementing the Telecommunications Act of 1996 about the importanceno, the necessityof working closely with our colleagues at the Federal Communications Commission and the various state public utility commissions. The affected industry participants and the public deserve consistency across the governmental spectrum. And, to that end, I hope to bring to bear what I've learned from our telecom experience as we engage with this Commission and the state commissions in addressing electricity restructuring.
Now, turning to the substance of the issues, let me start by pointing out that, regardless of what happens with respect to possible federal restructuring legislation, we know two things for sureor, at least at the level of confidence that passes for certainty in this town: First, we know that many states already are engaged in the restructuring process; and second, we know that the Department of Justice and the Commission will continue to have shared jurisdiction over merger reviews during this time of restructuring. So let me first make some preliminary comments about competition in the electricity industry in general, and then turn to our mutual interest and involvement in merger review.
As you probably know, promoting competition in wholesale electric power markets has long been one of the Antitrust Division's goals. In 1969, five years before the AT&T case was filed, the Division filed the landmark Otter Tail case.1 The Supreme Court's decision in that case eliminated significant doubts about whether the antitrust laws applied to regulated industries, and began what has turned out to be a quarter-century-long process to facilitate competition in wholesale electric power markets by assuring competitive access to electric power transmission networks. To that end, at about the same time that Otter Tail was filed, the Department began a long and successful campaign to condition nuclear power plant licenses on the provision of transmission services.2
Important as these early efforts were, there is no doubt that considerably more progress has been made in recent years. Today, truly competitive wholesale markets are now beginning to emerge as a result of the Energy Policy Act of 1992, coupled with some more recent initiatives, primarily at the state level. The Energy Policy Act made possible the Commission's Order 888 mandating open access, and several states already have begun dramatic restructuring. In terms of the states' efforts, I would note, in particular, that an increasingly popular component of restructuring is the use of an independent system operator, which offers the prospect of a simple yet effective solution to the transmission-access problem. In our view, this is a promising development.
These developments are important, but we believe that other efforts are also likely to be needed as we move down the path from regulation to competition. The antitrust laws provide ample authority for the Justice Department to challenge anticompetitive conduct of various sorts, but we cannot challenge market structure itself. In other words, to whatever extent restructured electric power markets are too highly concentrated to yield pricing at or near competitive levels, the antitrust laws provide no remedy. To address these kinds of structural competitive problems, some states have encouraged or required divestiture as part of their restructuring efforts, and these divestiture efforts have progressed substantially. In support and furtherance of such efforts, the Antitrust Division suggested in testimony before the House Judiciary Committee last June that Congress might want to look into providing authority to order divestiture in any federal restructuring legislation. Such authority, if conferred, would presumably go to the Commission. In the same testimony, we also suggested that the Commission undertake a comprehensive study of market power in a restructured electric power industry. This is a difficult and challenging assignment, but we think it is very important and that it would be worth the significant effort that would be required. We are prepared to work with the Commission in devising and carrying out such a study.
One other preliminary point worth mentioning is that, as competition becomes increasingly important in the electricity industry, so, too, will antitrust enforcement. That is true in the merger area, as I will soon discuss, but it is true in civil enforcement under the Sherman Act as well. Indeed, two of the four contested civil non-merger cases that we currently have in court involve anticompetitive behavior by incumbent or would-be monopolists in electricity markets. The first case, which we filed in April of 1996, sought to enjoin the City of Stilwell, Oklahoma from refusing to extend or connect water and sewer lines to consumers unless they also bought their electric power from the City. We believe that this conduct amounts to a per se unlawful tying arrangement and that it impaired competition on the merits between the City and an electricity cooperative. Next, in June of 1997, we brought suit against Rochester Gas & Electric, challenging its agreement with the University of Rochester. In that case, we charged that RG&E made threats and offered financial rewards to induce the University to abandon its plan to build a new, efficient power plant, that would have competed directly with RG&E.
This whole deregulatory process really is exciting stuff and I know it is receiving a great deal of attention at the Commission as well as in our offices. But, for today, there is a specific issue that I would like to address in some detail. As we move forward with our mutual efforts in electricity restructuring, I think it may be useful to ask whether the resulting flux should require any special treatment for mergers. Before commenting specifically on that area, let me first note that I started thinking about this issue during the Division's investigation of the Bell AtlanticNYNEX merger. Based on a year-long analysis of millions of documentsincluding, significantly, the non-public business plans of many of the affected playersas well as lots of deposition testimony, interviews, expert commentary, and advice, I believed then, and continue to believe, that the merger was not anticompetitive. In fact, the evidence indicated that real efficiencies were likely to result from the mergersome of which have already been realized -- and that, over time, those efficiencies would lead to better service in the affected areas.
Still, in the process of analyzing the Bell AtlanticNYNEX merger, one thought (well, at least one thought) continued to nag at me: "Wouldn't it be nice," I kept thinking, "if we had been able to observe some actual, real-world experience with head-to-head competition between Regional Bell Operating Companies, so we could have actual market data to add to our other evidentiary considerations?" Under the AT&T consent decree, of course, such head-to-head competition had never taken place and, while it is now allowed under the 1996 Telecom Act, none of the regional companies had invaded another's territory at the time we were reviewing the Bell AtlanticNYNEX deal. We frankly regretted not having that kind of information available to us.3
Thinking about this lack of market-based experience while working on the Bell Atlantic merger started me thinking more broadly about merger policy during a period of industry restructuring and deregulation. And, as my subsequent comments will demonstrate, I'm still thinking about the issue. But I've progressed far-enough to at least articulate some potential concerns and some possible solutions. I do so in the time-honored tradition of putting ideas out for consideration so that we can generate discussion and analysis which, in turn, one hopes, will lead to better policy. But lest there be any confusion let me make one thing, as they say, absolutely clear at the outset: I'm not here suggesting that any of this is necessarily better than the status quo, much less that it should be part of any federal legislation; all I'm saying is that I believe these matters merit thoughtful discussion and debate.
Merger Review under the Clayton Act
Let me turn then to the issue of merger review, and start by noting that, while the Justice Department and the Commission have shared jurisdiction in this area, our statutory responsibilities and missions are somewhat different. The principal antitrust statute affecting mergers is the Clayton Act, passed in 1914, which prohibits mergers whose effect "may be substantially to lessen competition" in any relevant market.4 For the most part, merger enforcement under the Clayton Act is prospective: that is to say, we generally challenge mergers before they are consummated.
This kind of prospective enforcement is greatly facilitated by the Hart-Scott- Rodino Antitrust Improvement Act of 1976,5 which requires parties proposing mergers exceeding certain financial thresholds to inform the Department and the Federal Trade Commission in advance and to wait certain specified periods before going forward and consummating their deal. During the statutory waiting period, or such longer period as the parties may voluntarily elect to wait, we do our investigative work, sometimes reach a remedial agreement with the parties or, if we think it appropriate, file suit seeking to block the merger.
As with any other civil plaintiff, the Department is entitled to relief only if it establishes by a preponderance of the evidence that a merger "may . . . substantially . . . lessen competition." Thirty years ago, during what we at the Antitrust Division fondly call the "halcyon days" of antitrust enforcement, the Supreme Court sustained practically every merger challenge that we filed. Recently, however, the lower federal courts have been less willing to uphold merger challenges. While we are not happy about that, I should hasten to add that very few cases are contested; some challenged transactions are abandoned, and most others that concern us are remedied through consent decrees. So only the really hard ones end up in litigation.
Most of our concerns, as I said, result in a consent agreement with the parties. The purpose of such a consent decree is to restore the market to roughly the competitive posture it would have had but for the merger. For us, merger decrees are almost always structural; they normally require divestiture of one or more lines of business or a package of competitively significant assets. For completeness, I should add that, unlike in many other areas of the law, an antitrust consent decree is not routinely entered by the court. The Department first publishes a Competitive Impact Statement explaining the alleged anticompetitive effects of the merger and how the decree would prevent those effects. There is then a comment period, and possibly even hearings before the court. The decree is ultimately entered only after the court concludes that it is in the public interest.6
Finally, in providing some relevant background, I should also note that there is little history of Justice Department merger enforcement in the electric power industry. The reason is partly that there have not been many major mergers in this area, but mostly because, given the historically regulatory nature of the industry, there has been little competition that could have been lost through merger. Competition has been growing in importance in the electric power industry, and mergers are now potentially far more problematic from a competitive standpoint. This may also mean that there will be far more impetus for mergers, although restructuring may provide efficiency rationales for mergers as well. Given all of this, in the past couple of years, we have devoted more resources to merger review in the electricity industry and we are beginning to focus on potential competitive concerns in some of our analyses.
Merger Policy During Electric Power Restructuring
Now, with that primer on the Clayton Act in mind, let me turn specifically to the question of merger policy during electric power restructuring. At a recent hearing before the House Judiciary Committee, I suggested the possibility of restricting mergers during electricity restructuring. Without endorsing the idea, I mentioned it in response to one of the questions I had been asked about lessons we had learned from implementing the Telecom Act. I explained that, in the typical merger investigation, we deal with a reasonably mature industry about which we can learn a great deal from observing and analyzing the recent history of market transactions. The Antitrust Division is very experienced in conducting such investigations, and we are confident that they provide a sound basis for predicting and proving the competitive effects of mergersboth pro- and anticompetitive. But what about the uncertainties stemming from major industry restructuring resulting from deregulation? How does this affect merger review?
To be sure, we deal with uncertainties every day because the Clayton Act requires a prediction about an inherently uncertain future. We must make predictive judgments about what would happen in the event of a post-merger attempt to exercise market power. But while often difficult, these judgments are grounded in experience and hard data. On the other handand this is the key pointthe grounding in experience and hard data from a competitive market may be lacking, at least to some degree, during the early stages of a transition to competition, especially in an industry that has experienced little, if any, prior competition.
To be specific about electricity, there is nowhere in the United States today that can we observe a fully restructured market in actual operation, and, in large parts of the country, we have little idea, for example, of what the basic nature of transmission pricing will be. This lack of experience presents practical analytic challenges for us. For example, the paradigm of the Merger Guidelines requires us to examine the effects of a five or ten percent increase by a hypothetical monopolist above levels likely to prevail in the near future. In almost all cases, currently prevailing prices provide the best indication of prices in the near future, and they are used as the benchmark for analyzing price increases. But there is ample reason to doubt that current electric power prices will prevail in the near future. Similar complications arise is addressing key issues such as whether generation at point A is likely to be competitive at point B. We can make some reasonably educated assumptions about these kinds of issues but, as time passes, experience obviously will help confirm or refine our thinking.
In short, in assessing mergers and other transactions in the electric power industry today, I find unsettling the possibility that a competitive analysis may at timesand perhaps in non-trivial waysdepend on assumptions, rather than empirical, experience-based analysis and modeling. The source of my discomfort is not that I think an analysis without empirical, experience-based input would necessarily provide an insufficiently sound basis for making policy decisions. Rather, I am concerned that a competitive analysis based significantly on assumptions about basic issues may make it more difficult to carry our burden of proof in court, where empirically based fact-finding is the norm and as is clearly favored.
This concern, I hope, will not endure for long. I anticipate that in the next few years the electric power industry will have taken significant steps in its transition to competition; many of the new market institutions will become well established, and there will be important market experience to look to, certainly at least those states that restructured first. I also expect that our modeling capabilities for sophisticated electric power networks will have improved as well. But during the transitional period, we will continue to face the potential uncertainties that I've been discussing.
Given that fact, utilities may see this as a time when they have a window of opportunity in which to consummate mergers. Mergers with little immediate anticompetitive effect can nonetheless frustrate the emergence of competition. For example, incumbent dominant firms could pick off competitors in their infancy, or even before they become competitors. In this climate, even utilities that otherwise might prefer to devote their present energies to other pressing matters, may feel that they must act now or possibly lose a golden opportunity.
Public Policies Toward Mergers During The Transition to Competition
That's the problem, as I see it. And the question is, what, if anything, should we do? First, let me stress that the antitrust laws in general, and the Clayton Act in particular, have served the nation very well, and I certainly would not propose amending them to deal with potential short-term problems associated with electric power restructuring. Having said that, I still believe that it is worth discussing whether, if there ultimately is federal restructuring legislation, it should address merger policy specifically. Although I mentioned the possibility of a temporary moratorium to the House Judiciary Committee, I did not mean to suggest that it is the only possibility or even the best one. A variety of alternatives merit consideration.
The idea behind a moratorium would be to postpone making difficult competitive evaluations for a brief period until we have developed a market-based history of evidence for making them. We can hope that the transition process will take only a few years, and postponing certain mergers for a brief period might not inflict significant hardships. In some cases, a moratorium could even do positive good because some firms would decide during the cooling-off period that the potentially anticompetitive mergers they might be contemplating are not such good ideas after all. The moratorium idea reflects the general proposition that mergers are very difficult to undo after they prove to be anticompetitive and that, during a transition to competition, there is unlikely to be any prospect for meaningful relief after the damage is done. Missed opportunities for the emergence of competition at the outset of the transition are forever lost, with potentially substantial social costs.
Of course, a literal moratorium on all electricity mergers clearly would go much too far. Most mergers are undoubtedly competitively benign or even procompetitive. To be precise about it, the most that I think merits discussion in terms of a moratorium would be the prohibition for a limited period of time of one or more narrowly defined classes of potentially problematic mergers. An example of such a category would be mergers of adjacentthat is, directly interconnectedgenerators, each with very substantial generating capacity.
Given the absolute nature of a moratorium, we at the Antitrust Division have also been considering some other, more flexible, alternatives to the status quo. One approach would be a targeted moratorium with a waiver provision. This is what was done in the Telecommunications Act of 1996 for mergers between local telephone companies and cable operators in the same service area. These mergers are prohibited by the Act unless the FCC makes any of several findings, including that "the anticompetitive effects of the proposed transaction are clearly outweighed in the public interest by the probable effect of the transaction in meeting the convenience and needs of the community to be served."7 As Senator Leahy explained during the legislative debates, the premise of this restriction was that allowing "telephone companies to buy out cable companiestheir most likely competitorin the telephone companies' local service areas . . . would destroy the best hope of developing competition in both local telephone service and cable television markets."8
A potentially softer approach, and the one that I currently find most intriguing, would be a modification of the burden of proof in merger cases during transitions to competition. The basic notion behind this approach is that any uncertainties associated with the transition period should not provide a legal license to merge but rather may be a good reason for increased skepticism about certain combinations. One way to think about such a proposal is to consider the way the antitrust laws treat restraints that are deemed to be facially anticompetitive but that do not fall within the narrow categories of per se illegal conduct. Such restraints are often analyzed under what antitrust lawyers and courts generally refer to as an "abbreviated rule of reason"9 or as I have called it, a "stepwise approach to antitrust review."10 Under this approach, when the ostensible nature of an practice is shown to suggest a likelihood of competitive harm, the plaintiff's initial burden is satisfied. Defendants are then required to offer an efficiency justification in order to subject the practice to a fuller competitive inquiry. If the defendant cannot come forward with a convincing explanation of how the practice will create real efficiencies, it is summarily condemned.
That in a very general way is the basic idea. I recognize, of course, that there are possible variations on the definitions of the parties' respective burdens as well as variations as to which mergers such burden-shifting should apply to. I will not detail these variations, much less discuss their virtues or vices, but I do want to make it clear that modifying the burden of proof during a period of industry deregulation is not meant to amount to a merger ban. Defendants should not be required to prove things they never can, and reallocation of burdens should not materially affect the analysis of mergers that are very likely to be competitively benign.
The Commission And Merger Policy
Until now, I've focussed on the Justice Department's role in merger review. Now let me make a few concluding comments about the Commission's role. First, I should note in this regard that the idea of placing some sort of burden of proof on the merging parties is hardly revolutionary. That is typically what occurs in the regulatory merger approval process and, as I understand it, that is how the merger process works when the Commission itself acts under section 203 of the Federal Power Act.11 More importantly, the Commission reviews mergers under a broad public interest standard, and that standard, I would suggest, might provide the flexibility to allow the Commission to craft special policies or procedures for dealing with mergers during a transitional period.12 Thus, I'd like to take this opportunity to encourage the Commission to consider the matters that I have been discussing. Although I wholeheartedly concur with the Commission's decision in its Merger Policy Statement to adopt the Horizontal Merger Guidelines, and I expect the Commission to look to Clayton Act case law for guidance in implementing those guidelines, I still think the Commission might want to go further and consider possible alternatives for coping with the special temporary problems associated with restructuring.
While on the issue of the Commission's merger authority, let me focus on one other point. The Commission is authorized under section 203(b) of the Federal Power Act to condition its merger approvals and to exercise continuing jurisdiction over merged entities.13 Although this authority certainly is very useful, I would caution against allowing it to result in an overly regulatory approach to merger review during the transition to competition. While I recognize, of course, that the Commission is a regulatory agency, and that the electric power industry has long been highly regulated, restructuring obviously is intended to move away from that paradigm. We at the Department hope and expect that market forces will become the primary determinants of wholesale electric power rates. And, in that context, mergers that substantially lessen competition should be allowed to proceed only if a court-imposed consent decree, or set of Commission-imposed merger conditions, offers a permanent, preferably structural remedy for the anticompetitive effects of the merger. More specifically, I would urge the Commission to reject rate freezes or rate roll-backs as conditions for approval of mergers creating structural competitive problems in generation.14 Such remedies typically are short-term, and do not in any way address the real competitive effects of the merger. Even in the short term, there will often be reason to doubt that the frozen rates would be as low as competitive rates.
Finally, based on a century of experience, I would further emphasize that the Department is also highly skeptical of any relief that requires judges or regulators to take on the role of constantly policing the industry. Relief generally should eliminate the incentive or the opportunity to act anticompetitively rather than attempt to control conduct directly. We are institutionally skeptical about code-of-conduct remedies. The costs of enforcement are high and, in our experience, the regulatory agency often ends up playing catch-up, while the market forces move forward and the underlying competitive problems escape real detection and remediation.
Let me end by reiterating that the Clayton Act has served the nation well, and I do not propose to tinker with it to better handle the narrow time frame in which competition takes hold in a previously regulated monopoly industry. But, for the reasons that I've laid out today, I hope I have persuaded you that it is worth considering whether any supplemental restrictions are appropriatebe they in tailored federal restructuring legislation or by dint of Commission action. If the subject interests you, I am sure that members of the Division's staff would be happy to discuss these issues with members of Commission's staff and I expect that a productive dialogue would ensue. Thank you.
1. Otter Tail Power Co. v. United States, 410 U.S. 366 (1973).
2. 1970 amendments to § 105(c) of the Atomic Energy Act provided that the Attorney General review each nuclear power plant construction and operating license to determine whether "the activities under the license would create or maintain a situation inconsistent with the antitrust laws." Pub. L. No. 91-560, 84 Stat. 1473, codified at 42 U.S.C. § 2135(c)(5).
3. Subsequently, Ameritech announced that it will compete out-of-region. I hope that we will learn something significant from that experience as we go forth with our efforts in deregulating local telephone markets.
5. Pub. L. 94435, title II, § 201, 90 Stat. 1390 (1976), codified at 15 U.S.C. § 18a. The Hart-Scott-Rodino filings for electric power mergers now typically are made long after the FERC filings. Consequently, we investigate these mergers prior to the Hart-Scott-Rodino filings.
10. See Joel I. Klein, A Stepwise Approach to Antitrust Review of Horizontal Agreements, Address Before the ABA Antitrust Section Semi-Annual Fall Policy Program (Nov. 7, 1996) (available at http//:www.usdoj.gov/atr).
11. 16 U.S.C. § 824b. Placing the burden on the applicants to show that a merger is in the public interest was upheld in Pacific Power & Light Co. v. Federal Power Commission, 111 F.2d 1014, 1015 (9th Cir. 1940).
12. "Although the Commission must include antitrust considerations in its public interest calculus under the FPA, it is not bound to use antitrust principles when they may be inconsistent with the Commission's regulatory goals. . . . [I]ndiscriminate incorporation of antitrust policy into utility regulation could undercut the very objectives the antitrust laws are designed to serve." [The Commission's] "analysis must sensitively recognize and reflect the distinctive economic and legal setting of the regulated industry to which it applies." Northeast Utilities Service Co. v. FERC, 993 F.2d 937, 947 (1st Cir. 1993) (internal quotations and citations omitted).
14. Long-term contracts of various sorts may adequately guard against anticompetitive effects when those effects would be of very limited duration. A contractual solution may be satisfactory when a large transaction produces only a small and short-term competitive problem.