The Cutting Edge Antitrust Conference
New York, N.Y.
In recent years, the enforcement agencies in their merger review have been offered more ambitious and complicated restructuring proposals to address overlaps and other potential anti-competitive effects. The question I will address today is what factors the FTC should rely upon in deciding whether and to what extent restructuring can save an otherwise anti-competitive transaction. My intent today is not to describe a new or revised policy, but only to describe what we are doing, and have been doing, for some time. I do intend to set out more fully than before the various factors that influence, in my mind, decisions to restructure.
Let me frame the issue with a simple example. Able seeks to acquire all of the stock or assets of Baker, a rival firm that is neither failing nor in a distress condition but that sells the same product as Able in some overlapping geographic markets. As a result, there is a clear anti-competitive problem. Able proposes to divest 20 percent of Baker's assets to X in overlapping Zone 1, 20 percent to Y in overlapping Zone 2 and 20 percent to Z in overlapping Zone 3. In the remaining geographic areas, it proposes to retain assets because there is no overlap and therefore no anti-competitive effect. One could complicate the transaction immensely - as regularly happens in the real world - with proposals by the acquirer that the company to which the assets are divested is newly created,(2) the divestee needs Able's raw materials or staff in order to initiate its business venture, must on a continuing basis rely on Able's licenses or other intellectual property in order to compete,(3) conduct limits need to be placed upon Able after the merger or upon how the acquirer will conduct its new businesses,(4) and so forth.
Ignoring complications for the moment, if all overlaps are eliminated through restructuring, and any other competitive effects are addressed, what is the problem? There are several. First, the only sure consequence of the restructuring is that Baker disappears, and any efficiencies of scale or scope, any experience it has, or its established trade relations with consumers or suppliers may be lost in the process. Second, the companies to which the assets are divested (or a single company) may or may not integrate the divested resources successfully. Experience shows that even when companies invest substantial resources and therefore have incentives to succeed, they may not successfully replace the eliminated firm. Third, the likelihood of success of restructuring often depends on sophisticated predictions about technical and financial factors in the market, turning on a detailed knowledge of particular industries, and antitrust lawyers and economists may not be the best people to undertake those tasks.
By stating the problem, I do not mean to suggest that extensive restructuring should never or even rarely be an acceptable solution to merger problems. The FTC Divestiture Study released last year concluded that most divestitures do establish a viable competitor.(5)When proposed mergers can be restructured in a feasible way that fully protects competition, the FTC will work with the parties to eliminate the problems and then clear the transaction. Often the proposed merger, after restructuring, will introduce substantial efficiencies, rationalize existing assets, turn assets over to superior management, or allow the successful integration of complementary resources and technologies. In other words, mergers and restructuring are a consequence of an immensely dynamic economy and should be acceptable to an agency like the FTC - at least where the threat to consumer welfare is not unduly great.
I believe there is no more important set of policy questions facing the antitrust community than defining the nature and limits of appropriate restructuring in merger review. Partly that is true because merger review has become so much a regulatory effort. Relatively few merger cases are fully litigated in the United States courts or the FTC's administrative process. Some of that is inevitable, but it makes the need for transparency in the process all the more important.
B. Background Statistics.
The United States is in the midst of an almost unprecedented merger wave. Slightly more than 4,700 mergers were submitted for review under the Hart-Scott-Rodino process in fiscal 1998, and slightly less than 4,700 in fiscal 1999.(6) That is almost three times the total of merger filings four years ago. Merger filings are up another 13 percent, in the first three months of this fiscal year.
Despite the merger wave, the FTC and the Antitrust Division of the Department of Justice still issue Hart-Scott-Rodino second requests (i.e. thoroughly review) somewhere between 2 to 3 percent of all mergers filed each year. The FTC issued about 90 second requests in fiscal 1998 and fiscal 1999 . Of that total:
- 19 mergers were withdrawn and transactions abandoned.
- 23 mergers resulted in the FTC ending its investigation and the transaction was cleared.
- 38 mergers resulted in consent agreements that involved restructuring.
- 3 mergers were litigated in court.
- 7 mergers remain pending.
While several challenges to proposed transactions led to elaborate statements of law and fact in court,(7) almost all of this merger work consisted of agency review without judicial participation.
C. Some General Observations
From the beginning, merger restructuring to address competitive problems and preserve otherwise legitimate proposals has been an accepted part of antitrust enforcement. In Brown Shoe Co. v. United States,(8) the first Supreme Court case to address the policy implications of Congress' 1950 revision of Section 7 of the Clayton Act, the Court discussed briefly the role of equitable relief:
"To illustrate: if two retailers, one operating primarily in the eastern half of the Nation, and the other operating in the West, competed in but two mid-Western cities, the fact that the latter outlets represented but a small share of each company's business would not immunize the merger in those markets in which competition might be adversely affected. On the other hand, the fact would, of course, be properly considered in determining the equitable relief to be decreed."
Restructuring efforts have come a long way since Brown Shoe. Indeed, the Brown Shoe footnote seems almost quaint in light of the restructuring proposals the FTC is called upon to consider today.
Because settlement negotiations are confidential, I am not comfortable describing specific matters in which proposed elaborate restructurings were rejected by the Commission or its staff as inadequate to preserve competition. Generally, those mergers were larger and involved more, and more complex, product overlaps. I can report that there have been increasing instances in which parties to a proposed merger have indicated they would accept restructuring of well over 50 percent of the acquired assets, have proposed mix and match divestitures that combined selected production and distribution assets and intellectual property (instead of divestiture of a going business) that might be divested to one or more acquirers, or proposed ongoing contractual or licensing arrangements with one or more acquirers that would have required continuing supervision by the Commission to ensure that the acquiring party had the resources to compete effectively in the marketplace. Even those approaches, in appropriate circumstances, might be acceptable if they satisfied the fundamental concerns that I will shortly describe.
All of this does not mean that we will not work with the parties on restructuring where appropriate in the circumstances. As noted earlier (see page 4), almost 60 percent of mergers investigated with a second request by the FTC in the last two years (38 of 67), where the FTC did not withdraw its objections, were settled by consent order. Some examples:
- In Ciba-Geigy's merger with Sandoz, a proposed restructuring was accepted. It called for the licensing of gene therapy technology and patents to Rhone-Poulenc Rorer Inc. in order to replace lost competition between the merging parties in innovation in that field, and for divestitures of Sandoz's corn herbicide and pet flea-control products to replace competition in those product markets. Essentially, licensing was preferred because divestiture would have impeded efficient, on-going research.(9)
- In the merger of Federal Mogul and T&N, the two firms allegedly would have accounted for 80 percent of sales in a worldwide market for thinwall bearings used in car, truck and heavy equipment engines. The parties originally proposed to divest a mix and match package of assets in Europe and the U.S., including some assets from the T&N research facility. Upon close examination, the offer was found wanting because it included some less than efficient production facilities and insufficient research and development assets. The Commission ultimately obtained a divestiture of T&N's entire thinwall bearings business which consisted of the assets and plants that T&N used to make thinwall bearings, including intellectual property that T&N used to develop and design new bearings in the future.(10)
Most restructuring proposals accepted by the FTC are successful and consumer welfare is protected. Last year, the agency's Bureau of Competition, in collaboration with the Bureau of Economics, published the first systematic review of orders that required divestiture of assets as a result of Commission action. The staff interviewed most of the buyers of divested assets and a few third parties. Of 35 orders entered between 1990 and 1994,(11) about three-quarters created viable operations in the relevant market. In those instances in which divestiture did not work out, it usually was because the seller engaged in strategic conduct to seek out marginally effective buyers (not entirely surprising), or buyers, because of informational disadvantages and lack of experience in the particular markets involved, were unduly optimistic about their ability to compete effectively with the acquired assets (somewhat surprising).(12) To address problems that arise in connection with restructuring arrangements, the Commission has adjusted its procedures modestly - for example, by insisting more frequently upon up-front identification of a buyer(13) and by reviewing the buyer's business plans with respect to the assets being divested.
The Divestiture Study findings confirm that feasible divestitures play a constructive role in the overall merger review process. That is especially true if the merger itself is likely to lead to efficiencies that ultimately benefit consumers. It remains true, however, that some kinds of proposed divestitures and other restructuring are unlikely to preserve competition in the market where the proposed merger would produce anti-competitive effects.
D. Factors Relating to Decisions to Restructure.
At the outset, let me be clear that the list of factors that I will discuss is not complete and no single factor is controlling. In some matters, the weight attributed to a particular factor will vary because of the impact it will have on achieving the goal of restoring competition lost as a result of the proposed merger.
1. How effective is the proposed solution?
The law is clear that divestiture and other restructuring remedies should not be adopted unless they are likely to restore fully the competition lost as a result of the merger.(14) But what factors should an agency like the FTC look to in assessing the effectiveness of the proposed solution?
With respect to the buyer or buyers, key questions include whether they will obtain assets sufficient to operate an effective competitive business, and whether they have sufficient incentives, competence, resources and experience to restore competition. The Divestiture Study concluded that often the buyers, because of lack of reliable information and experience in the market where the divested assets operate, may underestimate the assets they need, and may overestimate their ability to be effective competitors. As a result, the Commission in recent years has often insisted on knowing who the buyer or buyers are likely to be, and on seeing the buyers' business plan. While we certainly will be influenced by the buyer's predictions of competitive viability, those predictions cannot be controlling; thus, the Commission often will make a separate analysis of the feasibility of the proposed restructure. Even where the buyer has appropriate incentives, can it restore the scale and scope economies that the acquired company possessed? If not, its cost structure may make it a relatively ineffective competitor.
One way to ensure that scale and scope economies are not entirely lost is to prefer - but not always insist upon - the divestiture of a going business, with staff, supplies and customer relations intact, to divestiture of a package of selected assets.
With respect to sellers, the most important question is whether it divested itself entirely of its ability to influence or control the divested assets. Again, the Divestiture Study is instructive. Of 19 divestitures where the seller had a continuing relation with the buyer of the assets - for example, because of proposed IP licenses, supply contracts, or know-how arrangements - in six cases the continuing relationship was so detrimental that the buyer could not operate effectively, and in seven cases the ongoing relationship was competitively harmful.(15) These findings should not mean that continuing relationships are never acceptable, but they should be approached with caution.
The most challenging issue that agencies must address in the area of effectiveness of the remedy is how certain the enforcement officials and the courts must be that the restructuring will adequately protect competition. My view is that there is no simple answer to that question. Consider the following two hypotheticals:
- Suppose in a $30 billion deal, the competitive problems relate only to one percent of the deal, but with respect to that one percent there is a somewhat qualified but not totally acceptable buyer.
- To add complexity to an already complex set of issues, consider further the same deal and add that there are likely to be substantial efficiencies in the part of the deal not affected by the overlap.(16)
Does it make sense to try to block a proposed merger in those circumstances? What would be the consequence of a rigid policy? One possibility is that the acquirer, in the face of any such policy, will spin off the offending assets before making its merger filing - perhaps to a less than effective competitor - and the market will be less protected than if the enforcement agencies were willing to entertain, in those circumstances, a restructuring proposal.
My conclusion is that approval of such restructuring is a matter of degree, and no simple bright-line rule can be applied. But the bottom line is the divestiture must be effective and consumer welfare should not be asked to bear an unreasonably high risk that accompanies an uncertain and questionable undertaking.
The FTC's first responsibility is to preserve free and open competition. Nevertheless, as with the merits of a proposed merger, the FTC should also take into account pro-competitive efficiencies in deciding whether to approve a proposed restructuring. For several reasons, analysis of efficiencies in the context of settling merger cases can be even more complicated than in litigation.
If the efficiencies are claimed in the market where the restructuring is proposed , analysis is relatively easy. The agency will have investigated the claimed efficiencies in the course of reviewing the transaction.(17) The key questions then would be whether the efficiencies are clearly verified and are likely to be sufficiently substantial, and sufficiently likely to be passed on to consumers, to outweigh the risks that the restructuring will not fully restore pre-existing competition. Also, efficiencies asserted as a justification for restructuring must be merger specific. Experience with claims of efficiency defenses, since the Guidelines were amended in 1997, indicates that efficiencies are often exaggerated and, perhaps even more often, can be achieved through non-merger routes.(18)
Where the claimed efficiencies relate to portions of the deal not affected by the restructuring, analysis becomes more complicated. First, the staff ordinarily would not have investigated the other efficiencies and will have little more than the parties' assertions that such efficiencies will occur and will produce pro-competitive results. As many have noted, efficiencies are much easier to assert than to prove. Second, in the market where the assets overlap, and where by definition the restructuring does not produce a fully adequate restoration of competition (if it did there would be no need to turn to the efficiencies issue), consumers will be deprived of the benefits of competition so that consumers in a different market, or the shareholders of the merging corporations, can profit. I would not say that efficiencies in these other markets would never be taken into account, but surely they should be viewed with a skeptical eye.
3. How Complicated is the Restructuring?
The FTC Divestiture Study concluded that a clean divestiture of assets comprising a going business was most likely to create viable competitors.(19) This conclusion is consistent with longstanding judicial practice.(20) Complications arise, however, when some smaller, selected package of divested assets is made available to a buyer, and even more complications arise when there are continuing business relationships between the merging parties and the acquirer of the divested assets. In those situations the Commission is called upon to play a role as a continuing monitor of the arrangement, in a situation in which the divesting party has incentives to cut corners, delay compliance or interpret the divestiture order in a grudging manner.(21)
4. Effect on Future Transactions.
Just as a decision not to challenge a particular transaction may initiate a trend towards similar transactions in the same industry, a decision to accept a certain kind of restructuring may lead firms in subsequent transactions understandably to demand the same treatment. The answer by an enforcement agency to the effect that restructuring was acceptable at an earlier time in a moderately concentrated market, but is not acceptable in a more concentrated market, usually leaves the sponsors of the later transactions unpersuaded.
The legislative history of Section 7 of the Clayton Act makes clear that the responsibilities of enforcement officials and courts is to weigh not only the anti-competitive effects of the particular deal at issue, but also the possibility that the transaction is part of a merger wave. See Brown Shoe, 370 U.S. at 332-34. Our responsibility is not just to examine the merits of a particular transaction, but to take into account where the industry, as a result of similar transactions, might be going. That is as true with respect to equitable remedies as it is to the merger itself.
Of course the naked fact that there is a merger wave occurring is not dispositive on our obligations to approve of proposed restructuring. The movement toward consolidation, and attendant restructuring, may be occurring because economic factors have changed and the firms are pursuing efficiencies, or because subsequent proposed mergers are an effort to duplicate the market power achieved in the earlier combinations. Distinguishing among motives such as an aim to adjust to changed market conditions as opposed to an aim to augment anti-competitive market power, especially when many deals present elements of both, is one of the most difficult decisions antitrust enforcers and ultimately judges are called upon to make.
5. Experience with Past Transactions.
Enforcement agencies should not be expected and would not be justified in making the same mistake over again. If restructuring in a particular industry and in similar circumstances has been unsuccessful, enforcement officials have a responsibility to determine why. If the same or similar problems are likely to arise again, past experience should not be ignored.
6. Creation of New Law Through Settlement.
I started this discussion by noting the obvious fact that relatively few merger matters are fully litigated, either in court or in the FTC's adjudicative process. While disposition of controversial matters through negotiation and consent is undeniably efficient for both the government and the private sector, there is a down side to leaving the courts out of the game. I believe that the bipartisan consensus that strongly supports vigorous antitrust enforcement in this country depends in no small part on the view that agency enforcement decisions ultimately will be reviewed by an independent judiciary.
Obviously it is unfair and impractical to force parties to litigate otherwise easy-to-settle matters just to give the judiciary an opportunity to review agency decisions. But what about consent orders that in effect rely on new law? For example, both the Department of Justice and the FTC have settled cases in recent years that depended upon a finding of anti-competitive effects in an "innovation market" - that is a lessening of competition as a result of the merger in innovation rather than the production and sale of actual products.(22) Where the other factors I have mentioned today are adequately addressed, and particularly where the government is satisfied with respect to the effectiveness of the remedy, settlements should be pursued even where they depend on new interpretations of law. If the new interpretation is sufficiently ill conceived, at least some parties will choose to litigate. Even where that does not occur, the bar and the academic community can be relied on to direct a spotlight on faulty initiatives.
I hope I have been clear that the FTC has been and remains willing to consider restructuring proposals, even those that are fairly extensive and complicated, if it is likely that the restructuring will preserve competition. In recent years, however, we have seen more frequent proposals that are so extensive and complex that it is impossible to predict with any confidence that competition will be restored and consumer welfare protected.
One final point. It should be clear that the burden of coming forward with adequate restructure proposals should be on the sponsors of the merger. They are likely to have more extensive information about their industry and the effect of the proposal than any set of enforcement officials. The FTC's job is to make clear what its reservations are about the proposed transaction and to be available for a constructive dialogue on how any problem can be adequately addressed. It is not the FTC's job to propose solutions - indeed I have reservations about whether the FTC is always qualified to play that role.
1. Chairman of the United States Federal Trade Commission. The views expressed are my own and do not necessarily reflect the views of the Commission or other Commissioners.
2. See, e.g., In re Global Industrial Technologies, Inc., Dkt. No. C-3825 (F.T.C. Sept. 10, 1998),(consent agreement requiring divestiture of acquired company's glass-furnace silica refractories business to a newly formed company).
4. See, e.g., In re McDermid, Inc., Dkt. No. C-3911 (F.T.C. Dec. 22, 1999), </os/1999/9912/macdermidd&o.htm> (proposed consent order requiring termination of certain distribution agreements and cessation of participation in any horizontal agreements that have as their effect any allocation, division or illegal restriction of competition).
5. Federal Trade Commission Bureau of Competition Staff, A Study of the Commission's Divestiture Process 9-10 (1999) ("FTC Divestiture Study").
6. See Pub. L. No. 94-435, 90 Stat. 1383 (1976) (codified as amended in scattered sections of 15 U.S.C.).
7. See, e.g., Federal Trade Commission v. Tenet Health Care Corp., 186 F.3d 1045 (8th Cir. 1999); Federal Trade Commission v. Cardinal Health, Inc., 12 F. Supp. 2d 34 (D.D.C. 1998); United States v. Long Island Jewish Medical Center, 983 F. Supp. 121 (E.D.N.Y. 1997); Federal Trade Commission v. Staples, Inc., 970 F. Supp. 1066 (D.D.C. 1997).
8. 370 U.S. 294, 337, n.65 (1962).
9. See supra, n.3.
10. See In re Federal-Mogul Corp., File No. 981-0011 (F.T.C. March 6, 1998), </os/1998/9803/9810011.agr.htm>.
11. The 1990-94 period was selected so that enough time would have passed to allow an informed view of the success of the divestiture.
12. See generally FTC Divestiture Study at 8-29.
13. A buyer up-front is now required in about 60% of Commission divestitures. See Richard G. Parker, Global Merger Enforcement 8 (Sept. 28, 1999), </competition/global_merger_enforcement.htm>.
14. See, e.g., Ford Motor Co. v. United States, 405 U.S. 562, 573 (1972).
15. See FTC Divestiture Study at 12.
16. This hypothetical is somewhat similar to In re Roche Holding Ltd., Dkt. No. C-3809 (F.T.C. May 22, 1998), </os/1998/9805/9710103d.htm>. In that consent order, the Commission allowed Roche to acquire Corange Limited in an $11 billion transaction. One of two competitive concerns raised by the merger was that a Roche subsidiary and BM, a subsidiary of Corange, were both engaged in the research, development and manufacture of cardiac thrombolytic agents (CTAs). The consent responded to this concern by requiring the divestiture of BM's CTA business to Centocor, Inc., a substantially smaller biotechnology company. In order to ensure that Centocor could be a viable competitor despite its limited resources, the Commission carefully scrutinized Centocor's business plan for the acquisition and, in consultation with Centocor, required full divestiture of employees, know-how, and facilities, and an interim product supply commitment from the merged entity to Centocor, before it approved the merger.
17. A summary of the circumstances in which the enforcement agencies will take efficiencies into account can be found in U.S. Department of Justice and Federal Trade Commission, Horizontal Merger Guidelines § 4 (as revised in 1997).
18. An interesting example involved the Commission's successful challenge to the proposed merger between Staples and Office Depot.See FTC v. Staples, 970 F. Supp. 1066 (D.D.C. 1997). The merger would have involved about 1,000 office supply superstores, and would have consolidated the office supply superstore market from three firms to two, or from two to one, in approximately 40 geographic markets. A major line of defense was that the merger would allow the firms to own a larger number of stores which in turn would produce better quantity discounts from suppliers. The district court found, however, that the parties' efficiency claims were "in large part unverified," id. at 1089, and not merger-specific. Both parties were expanding rapidly by opening new stores at a rate of about 100 to 150 each per year, so their buying power was destined to increase without the merger, albeit perhaps a little less rapidly.
19. FTC Divestiture Study 11-12, 16-17, 27-28, 38, 42.
20. See, e.g., United States v. E.I. du Pont de Nemours, 366 U.S. 316, 330-31 (1961) ("Divestiture has been called the most important of the antitrust remedies. It is simple, relatively easy to administer, and sure. It should always be in the forefront of a court's mind when a violation of § 7 has been found.").
21. For example, in In re Cooperative Computing, Inc., Dkt. C-3757 (F.T.C. June 20, 1997), </os/1997/9706/ccitri~2.htm>, the Commission approved the merger of the two dominant providers of auto-parts databases, subject to the requirement that the acquirer, CCI, divest a perpetual, royalty-free, transferable, assignable, and exclusive license to its auto-parts database to a competitor, McDonald. In the course of the licensing arrangement between CCI and McDonald, disputes arose as to the effectiveness of the required technology transfers. In order to resolve these disputes and ensure that competition was maintained in accordance with the Commission's order, the parties agreed, during a Commission investigation, to the appointment in July 1998 of an Independent Auditor to monitor and supervise the transfer and to report to the Commission, and to certain procedures to facilitate and verify CCI's compliance.
22. See, e.g., In re Ciba-Geigy, Ltd., supra, n.3; United States v. Halliburton Co., No. 98-CV-2340, Complaint (D.D.C. filed Sept. 29, 1998), <http://www.usdoj.gov/atr/cases/f1900/1964.htm> (challenging merger because of competitive concerns in product and innovation markets for LWD oil and gas drilling tools; consent required divestiture of LWD business). See also U.S. Department of Justice and Federal Trade Commission, Antitrust Guidelines for the Licensing of Intellectual Property § 3.2.3 (April 1995) (discussing the analysis of innovation markets in the context of intellectual property licensing).