* The views expressed here are those of the author, and not necessarily of the Â鶹´«Ã½ Trade Commission or any Commissioner.
Introduction
I'm here today to talk about merger review. But I would like to look more broadly than at U.S. merger review; I'd like to address the reality that businesses face -- merger review by multiple jurisdictions. I start from the premise that good antitrust analysis knows no national boundaries. I'm not suggesting that there is only one way to conduct merger evaluations. Nor must all countries give the same weight to the same considerations in the same manner. But we do live in an age when many -- perhaps most -- significant mergers have transnational implications. In such a globalized economy, inconsistent assessments of transactions, or application of conflicting remedies by diverse enforcement authorities, could impose substantial costs on businesses and on the markets that they serve.
In fact, I think that rarely happens, for two reasons. First, the competition policies of nations differ within an acceptable range -- one that reflects reasonable but not drastic differences in legal systems. Second, our experience thus far is that multijurisdictional merger review has virtually always resulted in consistent outcomes and compatible remedies. But we can do better. The concerns raised by multijurisdictional merger review can be markedly reduced by transparent, cooperative, and efficient merger review procedures.
We now stand atop a century of learning about market dynamics, the importance of innovation and efficiency, and the value of lowering trade and market barriers. In conducting merger analysis, competition authorities widely recognize that big is not necessarily bad. Market concentration alone is not the issue any more than is raw size. Equally important is whether or not there are barriers to market entry, expansion, or mobility. A realistic assessment of a merger's impact should also consider whether, by merging, a firm will realize new levels or kinds of efficiency that will enhance that firm's ability and incentives to compete.
It may be premature to say that all merger laws reflect identical goals. But today most merger control laws focus on preventing transactions that create or enhance market power or facilitate its exploitation in anticompetitive ways. By market power, I mean the ability to raise prices above competitive levels or reduce output -- including product quality, variety, or innovation -- below competitive levels. Firms may possess and exploit market power either unilaterally or jointly, through coordinated interaction, and commonly use that power to cripple or exclude rivals that might otherwise act competitively. Thus, as a substantive matter, merger law is converging around a norm that rests on solid economic thinking and hard-earned experience.
I. Merger Review Process from an International Perspective: ICPAC
Today I want to focus on the merger review process from an international perspective, because the end of the 20th century is marked by three striking proliferations:
First, a proliferation of merger and acquisition activity which, in 1999 alone, reached $3.4 trillion worldwide. In the U.S., the dollar value of mergers reported annually increased a stunning eleven-fold over the past decade - from $169 billion in 1991 to $1.9 trillion in 1999. This merger wave is part of a broader, second, proliferation in cross-border trade. This development will only accelerate as tariff and non-tariff barriers are reduced, deregulation opens markets, technical standards are harmonized, international transportation and communication networks improve, information technology matures, and electronic commerce continues its exponential growth.
Third, we face a proliferation of merger control laws, a development reflected in the fact that in just the last decade the number of countries with premerger review authority grew from about 10 to over 60.
A key challenge to any nation's merger control law is that the law is national, but markets - given this proliferation in cross-border trade - are often transnational.(1) Sometimes markets remain national due to environmental, drug or other regulatory laws, but merging firms nevertheless may have assets or sales in various jurisdictions and compete in a substantial number of national markets.(2) And even when merging parties assets are located in only a single domestic market, their business activities - directly or indirectly - may affect markets in many countries, since the parties' customers, suppliers, and potential competitors may be located around the globe.(3)Therefore, many different national competition enforcement authorities reasonably may end up scrutinizing the same transaction.
In 1997, the United States Attorney General established the International Competition Policy Advisory Committee ("ICPAC") to examine, among other things, the issues and problems associated with reviews of mergers by multiple jurisdictions. This month the Advisory Committee issued its report, which offers many thoughtful recommendations for eliminating unnecessary transaction costs on businesses that potentially may arise when one transaction undergoes multiple reviews.
A. ICPAC Suggestions to Minimize Potential Conflicts.
The Committee has three main suggestions for minimizing the potential costs and conflicts associated with multi-jurisdictional merger evaluations.
The first is increasing the transparency of the review processes. Transparency logically would promote discussion, understanding, and (possibly) harmonization of legal policies among competition authorities. It also would give businesses contemplating a merger greater predictability regarding the success or failure of their plans. Among the many ways that competition enforcement authorities can enhance transparency are publishing guidelines and statements that explain the jurisdiction's merger control laws and prosecutorial policies, and publishing significant merger review decisions.(4)
Second, ICPAC proposed that national competition authorities minimize potential conflicts by adhering to certain basic disciplines in the application of their merger review laws. For example, competition policy enforcement officials should apply their merger law in a non-discriminatory manner. Governmental enforcement authorities should not give domestic companies favored treatment over non-nationals, either to protect domestic industries from competition or to foster a "national champion." In addition, antitrust evaluation of mergers should not consider non-competition elements. At a minimum, reviewing authorities should apply non-competition factors (such as job preservation, the promotion of exports, or the creation of national champions) only after the completion of the antitrust analysis and make completely clear how those factors are weighed. Similarly, antitrust agencies should be independent of political pressures. Finally, national competition authorities should, to the greatest extent possible, tailor remedies to cure domestic problems. Inevitably, some mergers that have a significant anticompetitive effect on a local jurisdiction will require a remedy with an extraterritorial impact. The exercise of self-restraint by national authorities in fashioning remedies respects the possibility that competition officials in other jurisdictions might evaluate the need for a remedy somewhat differently.
Third, the Advisory Committee highlights the need for, and importance of, cross-border cooperation and information sharing among competition authorities when reviewing mergers. The Committee proposes that each competition agency publish a Protocol that explains how it coordinates and cooperates with other agencies so that firms are informed about what this entails. It further suggests that competition authorities publish model confidentiality waivers that parties can use to authorize the agencies to exchange documents and conduct discussions that confidentiality laws would otherwise prohibit. I firmly believe that once parties become comfortable with these limited waivers of confidentiality, and appreciate the confidentiality protections that continue to exist, merger reviews will proceed far more efficiently and expeditiously.
Finally, in its most forward-looking proposals, the Committee suggests that enforcement authorities begin thinking about work sharing arrangements. Depending on the particular case, these efforts could focus on jointly developing remedies that are acceptable to all affected jurisdictions, or coordinating investigatory efforts.(5) Looking further into the future, the Committee raises the possibility of limiting the number of investigating jurisdictions by relying on the efforts of those whose markets are most heavily affected by the transaction. But the Committee recognizes that this degree of work-sharing is a distant vision, and that national authorities today must first protect their markets and their consumers.
B. Rationalizing the Merger Review Process: Two-Stage Merger Screening.
The International Competition Policy Advisory Committee also considered specific ways to rationalize the merger review process. Two themes run through this set of proposals. First is the need to ensure that each national system reviews only mergers that have a nexus to it and a potential to create anticompetitive effects in its territory. Second is the need to ensure that no jurisdiction unduly burdens the merger review process.
As an initial matter, the Committee endorsed the use of a premerger screening program - in my view, one of the most positive competition enforcement developments in the last quarter century. It simply is too difficult, after a merger occurs, for antitrust authorities to separate and reallocate the commingled physical assets, employees, and intellectual property to each of the merging parties. In essence, once the eggs have been scrambled, it's impossible to return them to their original shells. It also is far easier and more efficient to prevent a merger that is likely to reduce competition than it is to catch and subsequently correct the misuse of market power made possible by the merger.
Under the "reasonable nexus" theme, the Committee made several recommendations with respect to the notification thresholds used in merger review systems. First, the thresholds should be made only as encompassing as necessary to ensure the reporting of potentially problematic transactions. Those deals that are unlikely to generate appreciable anticompetitive effects within the reviewing jurisdiction need not be notified. Establishing sensible thresholds works to the benefit of both businesses and enforcement authorities, by not encumbering nonproblematic transactions with regulatory review and by enabling the authorities to focus their resources on the problematic ones.
Interestingly, and I think correctly, the Committee also recommended that agencies should retain authority to pursue transactions that do not meet the notification threshold but that have special features which merit further evaluation. For example, a competition agency certainly should be permitted to examine the merger of two parties that, notwithstanding their modest sales and assets, own valuable intellectual property assets that, if combined, might give the entity the power to restrain innovation in a rapidly-developing market.
Second, notification thresholds should reflect a reasonable nexus between the transaction and the reviewing jurisdiction. That is, there ought to be some threshold of transaction-related sales or assets within the jurisdiction. Absent a jurisdictional nexus, notification and review are a waste of business and enforcement resources; at worst, these unwarranted reviews may create conflicts with the evaluations of much more heavily affected jurisdictions.
Third, notification regulations, like substantive merger-review standards, should be clear, transparent, and objectively-based. Basing the threshold on the amount of the merging parties' (individual or aggregate) sales or assets in the jurisdiction will usually be preferable to using a market-share standard, which typically is difficult to assess accurately, particularly at the early stages of the reviewing process.
Lastly, the Committee recommended that agencies harmonize the time of notification so as to permit filings anytime after executing a letter of intent or agreement in principle. By permitting simultaneous filing in the various jurisdictions potentially affected by the merger, agencies can better coordinate their investigations and the parties can navigate the review process more expeditiously.
There are, of course, many ways to design a merger review program. The Advisory Committee concluded, however, that from a burden-reduction perspective a two-step process is likely to be the most efficient. The initial stage is best devoted to separating benign transactions from mergers that clearly raise competitive problems or whose complexity warrants further inquiry. The Committee believed that this initial stage should last only a short period, no more than a month. In addition, merging parties should be permitted to request early termination of the reviewing period for innocuous transactions or those where the parties are willing to resolve competitive concerns through consent agreements. In the U.S., more than 70% of mergers reported to the FTC in 1998 received early termination within an average of 16 days of filing, and approximately 97% of the reported mergers were cleared in the initial 30-day stage. In short, the use of an initial screening stage permits the vast majority of transactions to proceed from agreement through the reviewing process to consummation with a minimum of inconvenience, delay, and expense.
The Committee further recommended that a longer second review stage be reserved for mergers that appear to raise serious competitive issues. To give greater certainty to businesses, the Advisory Committee recommends that this stage not be open-ended. I frankly prefer the more flexible U.S. system, which has no formal second-stage deadlines, to the rigid timetables that apply in jurisdictions such as the European Union. In my experience, reviewing officials and merging parties alike have often found that extra time is needed to analyze a complex transaction correctly and to negotiate a sensible remedy. Recognizing that some transactions are far more complicated than others, the Advisory Committee offered the suggestion that, where strict deadlines do not apply, enforcement authorities should set notional deadlines for each merger review in accordance with that transaction's complexity.
The Committee also suggested that reviewing authorities should require information submissions that parallel and complement the two-stage process. At the first stage, authorities should seek only the information necessary to determine whether the merger warrants further review. These information requirements should not be regimented by statute. For example, if reviewing officials have nearly enough information at the first stage to determine that the merger will be innocuous, those officials and the parties should have the latitude to negotiate a voluntary submission of additional information to resolve quickly any lingering issues and avoid the need for a second stage review. Similarly, at the second stage, officials should articulate the competitive concerns that are driving the investigation and focus their information requests on those issues.
Finally, the Advisory Committee recommended that parties work multilaterally, possibly through the OECD, to further develop a common framework for merger notification. Such a framework can include understandings regarding the timing of notification and the minimum information that is needed at the first stage to conduct a screening review that is expeditious and effective without being unduly burdensome. This sort of a framework can be particularly useful for countries that are just implementing a merger review system, since it provides a template that reflects the experience gained thus far and encourages convergence around "best practice" standards.
C. Successful Efforts to Avoid Incompatible Resolutions.
Although the Advisory Committee report expresses legitimate concerns about differing substantive merger evaluation systems and the possibility of inconsistent or incompatible remedies, in fact the process is working remarkably well. I can't think of an instance where we, the E.C., or Canada imposed inconsistent or incompatible remedies on a party. For example, U.S. and E.C. officials had different concerns regarding the merger of the Swiss pharmaceutical firms, Ciba-Geigy and Sandoz. Discussions between the agencies, however, resulted in complementary remedies that alleviated E.C. concerns regarding Sandoz' monopoly for methoprene while assuring that the divestee of Sandoz' U.S. and Canadian flea control business would have a reliable supply of this critical ingredient.(6)And in the Zeneca/Astra,(7) ABB/Elsag Bailey,(8) and Â鶹´«Ã½ Mogul-T&N mergers,(9) the FTC and E.C. (or in the case of T&N, E.C. member states) officials engaged in extensive discussions to develop compatible remedies that placed no inconsistent burdens on the transacting parties. I should note that a critical element in the effective and speedy disposition of these merger reviews -- approximately two months from notification to agreement in both the Zeneca and ABB mergers -- was the willingness of the parties to waive confidentiality protections, thereby permitting the various competition authorities to share their information, market assessments, and remedy proposals. Even in the Boeing/McDonald-Douglas merger, the E.C. and we were each able to develop a package of remedy provisions that both sides could live with and that did not impose inconsistent obligations on the merging parties.
II. Remedies and Restructuring
Just as a correct prescription is as important as an accurate diagnosis in medicine, we have found in antitrust that imposing the right remedy is just as important as correctly analyzing the transaction. This is particularly true because most merger challenges do not go to litigation these days but are settled by agreement with the reviewing authorities. Frequently the remedy will involve some sort of restructuring of the merging parties' businesses or assets. For example, 38 of the 83 investigations (more than 45%) in which the FTC issued a second request (thus moving them to the second, more intensive stage of our merger review process) resulted in a consent order that involved some restructuring.(10)
But while we have learned a lot about the substantive evaluation of mergers over the past 100 years, we are only now beginning to define the nature and limits of appropriate restructuring. Of course, from the beginning, merger restructuring has been an accepted way of addressing competitive problems and preserving otherwise legitimate transactions. But recently parties have proposed more and more ambitious and complicated restructuring proposals to address anticompetitive problems and overlaps that competition enforcers have identified. How do we decide whether and to what extent merger restructuring can save an otherwise anticompetitive transaction?
To examine the success of restructuring remedies, the FTC's Bureau of Competition reviewed divestiture orders in 35 cases between 1990 and 1994. In its report issued late last year, the Bureau concluded that most divestitures have produced viable competitors in the relevant market but that there are ways to improve merger remedies so as to avoid potential problems.(11)
In light of both the potential promise and problems associated with divestiture solutions, we generally consider several factors in evaluating a proposed restructuring remedy.
The first is whether the remedy is likely to protect, promote, or restore competition in the affected market. To this end we must consider whether the buyer is obtaining sufficient assets, and the right kind of assets, to be able to create and operate a successful, competitive business, and whether that buyer has sufficient acumen, experience, incentives, and resources to accomplish its goal. Because these factors are so important to the ultimate success of a restructuring remedy, the Commission generally prefers to know in advance who will be the purchaser of the divested assets and the details of that buyer's business plan. In about 60% of the FTC restructuring remedies in the past three years, the Commission has known in advance who the purchaser will be. Further, the Commission considers whether the seller will be able to influence the purchaser's ability to effectively use the divested assets, for example, by not providing sufficient technical support. The Divestiture Study revealed that in about one third of the cases where the seller had an on-going relationship with the purchaser of the divested assets, the seller was able to prevent the effective use of those assets. In another third of the cases, the seller was able to inflict some competitive injury on its new rival.(12)
A second factor to consider is whether efficiencies justify restructuring, rather than condemning, a merger transaction. Particularly since the Horizontal Merger Guidelines were revised in 1997, enforcement agencies in the U.S. do consider efficiencies, but only if the efficiencies are produced by the merger under review. Unfortunately, the creation of efficiencies is much easier to claim than confirm and efficiencies can often be achieved by means other than a merger. But if efficiencies from a merger are claimed in the market where restructuring is proposed, analysis is at least manageable. The agency will already have studied the claimed efficiencies in reviewing the transaction. What remains to be determined is whether the efficiencies are likely to be passed on to consumers and whether they are substantial enough to outweigh the risk that restructuring will not fully restore competition. Questions of proof multiply where the efficiencies arise from less-troublesome, and therefore less-scrutinized, aspects of the merger. Then we are forced to balance the risk that a restructuring will not effectively restore competition against the more difficult to prove possibility that the merger will achieve efficiencies in some entirely different market, the benefits of which may not even be passed on to consumers.
Third, the Commission must consider the complexity of the proposed remedy. The Divestiture Study confirms that a restructuring remedy is much more likely to be successful if the divestiture is of an entire, on-going business, replete with its staff, supplies, customer relations, supplier contacts and other tangible and intangible resources.(13) Prospects for a successful restructuring are less promising, however, where only a package of selected assets is to be divested, or where the buyer will depend to some extent on the seller's continuing support, supplies, or other efforts. In such instances, the enforcement agency may be required to monitor the seller's conduct closely to ensure that the seller does not undermine the purchaser's ability to compete. Such measures, however, tend to be expensive, intrusive, and imperfect.
Fourth, the Commission must consider whether its acceptance of a restructuring remedy might have implications for future matters. For example, when the Commission accepts a restructuring remedy, parties to subsequent proposed mergers tend to insist on a similar resolution to the competitive problems raised by their transaction, although the Commission may view the matters as distinguishable on factual or economic grounds. This is because even where transactions appear to be comparable, prior mergers affect subsequent market structure, and different markets have different dynamics. Past experience also can provide valuable learning regarding the types of restructuring provisions to avoid, or the industries that have proven most resistant to efforts to restore competition. We should not be expected to make, and indeed would not be justified in making, the same mistake again.
III. Conclusion.
Evaluating mergers with cross-border effects is a high risk/high reward enterprise. If we do it wrong, we risk losing the benefits of competition in not just one market, but many. If we do it right, we can preserve, or even expand those benefits. And the impact, either way, can be nationwide, or even global. It is also a difficult enterprise. There is a full spectrum of ways in which businesses can combine their efforts and resources, and each variation has different implications for a country's economy and consumers. With cross-border mergers, the task is compounded by different laws, languages, cultures, and practices.
But it is an effort that I believe we have been doing reasonably well. For this, I give the credit to consistent, good faith efforts by competition authorities in many jurisdictions to cooperate in and coordinate their enforcement efforts. With continued mutual assistance, I believe we can further rationalize the merger review process.
1. See, e.g. Rohm & Haas Co., Dkt. No. C-3883 (Aug. 2, 1999)(in merger affecting North American market for water-based floor care polymers, FTC consent order required Rohm & Haas to divest some of the assets it would be acquiring in merger with Morton International, Inc.); Exxon Corp., Dkt. No. C-3833 (Nov. 4, 1998)(in joint venture involving Exxon and Royal Dutch Shell affecting North American market for viscosity index improver, or oil additive, FTC consent order required Exxon to sell its viscosity improver business to Chevron Chemical Corp. or other FTC approved purchaser); Degussa Aktiengesellschaft, Dkt. No. C-3813 (June 19, 1998)(in purchase by Degussa Corp. and its parent, Degussa Aktiengesellschaft, of DuPont's worldwide hydrogen peroxide, consent order preserved competition in concentrated North American market by permitting Degussa to acquire only DuPont's Gibbons, Ontario, plant but not DuPont's plants in Memphis, Tennessee or Maitland, Ontario).
2. In re Ciba-Geigy Ltd., Dkt. No. C-3725 (F.T.C. March 24, 1997).
3. In the Matter of Boeing Company/McDonnell Douglas Corporation, File No. 971-0051 (July 1, 1997).
4. In some high profile cases, an agency's underlying reasons for challenging or not challenging a merger may get lost in the media and political buzz surrounding a transaction. In the Boeing/McDonnell Douglas merger, EC officials found the merger anticompetitive. Although some observers believed that the demand for significant concessions was made solely to protect Airbus, a careful reading of what DGIV did revealed that it was consistent with EC law on abuse of dominance and with the Commission's prior rejection of the use of competition policy to create "European champions." Aerospatiale-Alenia/DeHavilland, 1991 O.J. (L 334) 42 In the United States, the FTC determined not to challenge the merger, but made clear in a separate statement that its decision was not influenced by any desire to create a "national champion." Statement of Chairman Robert Pitofsky and Commissioners Janet D. Steiger, Roscoe B. Starek III and Christine A. Varney in the Matter of Boeing Company/McDonnell Douglas Corporation, File No. 971-0051 (July 1, 1997).
5. It is noteworthy that many of the cases that have gone to second request at the FTC over the past few years have also been subject to review by antitrust agencies in other jurisdictions. Quite a few of those cases have led to coordinated settlements of antitrust concerns arising in one or more of the territories in which the merging companies do business. For example, in fiscal year 1998, out of the 28 merger enforcement actions the FTC brought, 13 involved notification to foreign governments and, of those, 6 involved substantial discussions with foreign authorities also reviewing the matters. From January through July 31, 1999, the FTC issued 38 second requests in merger cases. Of those cases, 21 involved notifications to foreign governments and 12 involved substantial discussions with our foreign counterparts. In addition there have been about a dozen other mergers in which discussions took place between FTC staff and other enforcement authorities where we concluded that no enforcement action was necessary.
6. In re Ciba-Geigy Ltd., Dkt. No. C-3725 (F.T.C. March 24, 1997).
7. In re Zeneca Group, PLC, File No. 991-0089 (FTC March 25, 1999).
8. In re ABB/Elsag Bailey, FTC Dkt. 3867 (consent order April 14, 1999).
9. In re Â鶹´«Ã½-Mogul Corp., FTC Dkt. 3836 (consent order March 6, 1998).
10. Of the 45 remaining transactions, 23 were cleared for consummation, 19 were abandoned, and 3 were litigated in court.
11. Â鶹´«Ã½ Trade Commission Bureau of Competition Staff, A Study of the Commission's Divestiture Process (1999)("FTC Divestiture Study").
12. FTC Divestiture Study at 12.
13. FTC Divestiture Study at 11-12, 16-17, 27-28, 38, 42.