The Evolving Approach to Merger Remedies

As published in Antitrust Report

Date:
By: 
David A. Balto, Former Assistant Director of Policy and Evaluation;
Richard G. Parker, Former Director

Probably no single issue currently is receiving as much attention as the topic of relief in merger cases. The question of whether there is a remedy to an anticompetitive merger and what that remedy should be is perhaps the single most intriguing and complex issue faced by the Bureau of Competition of the Federal Trade Commission.

In this article, we seek to provide an overview of how the Bureau of Competition approaches the issue of merger enforcement and remedy. We begin by outlining the important responsibilities of an enforcement agency in fashioning relief. We then discuss how the FTC's approach toward merger remedies has evolved over the past two decades. After that, we describe several cases in which the Bureau of Competition chose not to accept various remedies proposed by parties to a merger. These examples illustrate why certain forms of relief, both structural and non-structural, may be inadequate to resolve certain types of competitive problems. We close with a series of difficult and important questions regarding merger remedies.

the merger wave: new challenges

The merger wave continues at a rapid and breathtaking pace. Each week there are announcements of new mergers, many of which appear to restructure industries or create firms of a size that was unimagined a few years ago. A recent article characterized the merger wave as "a frenzy of merger madness, capping a dramatic wave of global corporate consolidation that has been gaining momentum through much of the decade."1 In terms of simple numbers, reported Hart-Scott-Rodino transactions have tripled since 1991, from 1,529 to 4,642 in fiscal year 1999. More important, the total value of these mergers has increased eleven-fold during this period, from $169 billion to over $1.9 trillion. The pace in the first six months of fiscal year 2000 promises another record year, with transactions averaging over 20 percent more than last year.

Of course, the vast majority of mergers are procompetitive or competitively neutral. Some mergers bring together firms in complementary relationships, or involve markets that appear to be converging. That is why at both the FTC and the Justice Department only a small handful, less that 3 percent a year receive some type of in-depth investigation. At the FTC, the vast majority (over 60 percent) of these investigations result in enforcement actions. In fiscal year 1999 there were thirty merger enforcement actions. So far in fiscal year 2000 there have been ten enforcement actions.

There are several aspects to the ever-intensifying merger wave that directly impact the issue of merger remedies. The problem of designing and securing effective relief is an increasingly complex and challenging problem. Why is that? The primary reason is that mergers are increasingly strategic in nature. Many of the investigated mergers are motivated by strategic concerns, such as the desire to become dominant in a market. Unlike the mergers of the 1980s, which were frequently motivated primarily by financial concerns, today's mergers are based on a desire to strengthen competitive position. Thus, they are more likely to involve substantial horizontal overlaps, some of which are much larger than those the FTC dealt with in the past. Replacing a competitor with 30 percent of the market is far more daunting than replacing one with a 5 percent market share. Moreover, as each merger occurs the number of remaining firms diminishes and, in turn, so does the pool of potential acquirers of divested assets. Often, when presented with problems of substantial relief and few remaining competitors, the parties propose putting the FTC in a regulatory position, monitoring remedies short of a clean divestiture.

There are other factors that increase the challenge of remedy. The sheer size of the mergers and the number of markets involved is far greater than the past. Since technology and information are assuming primacy as driving forces in the economy, relief often must include these assets. But crafting relief for intangible assets can create tough challenges. Some transactions are in regulated or newly deregulated industries where the antitrust agencies must determine whether to rely on regulation to protect competition.

Finally, as described below, the Bureau of Competition recently completed an important study of the divestiture process. The Bureau has learned from the success and failure of remedies in the past, and approaches merger remedies with a renewed sense of humility and caution. Unlike other agencies which possess expertise in a specific industry, the FTC has general jurisdiction. We are not experts in any particular industry. Therefore, we have increasingly recognized the need for more thorough examination and care before any particular remedy is adopted.2

the range of remedial options -- how do we choose?

There are a variety of approaches to curing anticompetitive mergers. First and foremost, the agencies may simply decide that no remedy, short of blocking the transaction, will fully resolve the competitive concerns. Second, the agencies may decide that the resolution of competitive concerns will require the divestiture of an entire ongoing business and related assets. Third, they might conclude that some form of partial divestiture incorporating various aspects of a business would be acceptable, because it could facilitate the entry or expansion of a replacement competitor. Fourth, a merger might be resolved through contractual arrangements, such as the licensing of intellectual property or perhaps a supply agreement. Fifth, the agencies may decide to use some form of behavioral relief such as a non-discrimination provision. Some mergers can be resolved with a combination of these forms of relief.

The Commission has broad discretion in deciding whether any one of these possible remedies is acceptable in a particular case, so long as the remedy will cure the competitive problem.3 So how does it decide which approach is most suitable for a given case?

The foremost obligation of antitrust enforcers is to make sure that a merger does not reduce competition to any significant extent. As Justice Brennan recognized over forty years ago in DuPont: "The key to the whole question of an antitrust remedy is of course the discovery of measures effective to preserve competition."4 Consumers should benefit from the same degree of competition after a merger as before a merger. Thus, our first objective is to determine which remedies will effectively and fully preserve competition.

A second objective is to select a remedy that will preserve competition with as much certainty as possible. The risk of inadequate relief, or the burden of untimely relief, should not be borne by consumers.

The third objective is to preserve the efficiency-enhancing potential of a merger, to the extent that is possible without compromising our obligation to preserve competition. If there are two remedial options, both equally effective (based on experience) and both equally likely to achieve their objective, but with different implications for preserving cognizable merger efficiencies, we should choose the one that is more likely to preserve efficiencies. They must be effective based on experience--theory alone may not be enough for the risk of a failed remedy to be shifted to consumers.

Our approach to remedies evolves, as does our approach to merger enforcement generally. We learn from each case what works and what doesn't work. Our past actions provide guidance, but there are no absolute rules. We evaluate remedies based on the facts in each individual case. We also go back and evaluate our remedy process, as described below, to see if expectations are borne out and the remedies are effective.

We should also keep in mind what our objectives do not include. The FTC is not a market regulator. Apart from enforcing the prohibitions that are contained in the antitrust laws, our job is not to regulate or prescribe the market behavior of firms. That is a function of the competitive process. Nor are we industrial planners. Our obligation is straightforward and simple--make sure that the post-merger world is every bit as competitive as the one that existed before the merger. Of course, nothing in the real world is ever that simple. Tradeoffs and judgment calls need to be made.

is there a preferred merger remedy?

One way to assess the FTC's approach to merger remedies is to determine whether there is some benchmark or preferred remedy it should be trying to achieve. Generally, there is. In most cases divestiture is the preferred remedy. As Justice Brennan stated in DuPont: "Divestiture has been called the most important of antitrust remedies. It is simple, relatively easy to administer, and sure. It should be in the forefront of a court's mind when a violation of § 7 has been found."5 Many courts have followed that guidance for the past several decades, as have the enforcement agencies.

The facts in DuPont illustrate why divestiture is preferable. The parties had proposed various forms of behavioral relief (e.g., barring DuPont from influencing the selection of GM officers or directors and prohibiting preferential trade relationships) in favor of divestiture. The Court found that enforcing such a decree likely would be cumbersome and time-consuming, that framing an injunction to address all forms of anticompetitive conduct would be impossible, and that policing the order "would probably involve the courts and the Government in regulation of private business affairs more deeply than administration of a simple order of divestiture."6

Of course, saying that divestiture is the preferred remedy somewhat begs the question: Divestiture of what? The entire acquired entity? A complete, ongoing business? A partial divestiture of assets that might provide the basis for starting a business? In markets where technology is a key to success, is a divestiture of "soft" assets such as intellectual property sufficient, or is a broader asset package, even an ongoing business, needed to ensure successful entry? The Supreme Court's characterization of divestiture in DuPont as "simple, relatively easy to administer, and sure" applies most clearly to a clean separation of two ongoing businesses--essentially, undoing the merger or acquisition. That is what was ordered in DuPont, where the Court stated that "complete divestiture is peculiarly appropriate in cases of stock acquisitions which violate § 7."7 DuPont was a post-acquisition case, of course. Today, thanks to Hart-Scott-Rodino, we more typically look at the remedy issue in the premerger context, and the lesson of DuPont would be to prevent the two businesses from combining in the first place.

One issue that arises where the divested facility produces several products is whether divestiture of the entire facility is necessary. Occasionally, parties argue that they should be able to retain those portions of a facility that produce products which do not raise competitive concerns. This argument will not carry the day where those other portions of the facility are necessary to ensure the viability of the divested entity. For example, in Olin Corporation, which involved a chemical plant that manufactured certain swimming pool sanitizers, the respondents sought to exclude from the Commission's order part of the plant that manufactured cyanuric acid. The Commission rejected that request because there was no evidence that the part of the plant that manufactured the swimming pool sanitizers could operate independently. Thus, the Commission concluded that divestiture of the entire facility was necessary "to give its acquirer a real chance at competitive success."8

The teaching of DuPont and Olin is that it is clearly within the Commission's power to require divestiture of a greater set of assets than those which participate in the overlap markets in order to effectively replace competition. Often the buyers of the divested assets will need other ancillary assets in order to effectively restore competition. Sometimes without these ancillary assets the buyer will not be able to replicate the economies of scale or scope of the firm that has been acquired. In other cases, these additional assets will be necessary to give the buyer both the incentive and ability to fully restore competition.

This principle was applied in both of the recent oil mega-mergers, Exxon/Mobil and BP/ARCO. In Exxon/Mobil, there was a direct overlap in California between the two firms in oil refining, but a far less significant overlap downstream (in gas stations). The FTC required divestiture not only of Exxon's refinery, but also of all of Exxon's downstream assets. The Commission required a "clean sweep" of all assets in order to assure the buyer had the same level of economies of scale and scope that Exxon possessed prior to the merger. A vertically integrated refinery would be a far more significant competitive force.

Similarly, in BP/ARCO, there were significant competitive overlaps in the production, sale, and delivery of Alaska North Slope crude oil. The parties entered into a separate agreement with the State of Alaska which would have combined various assets of BP and ARCO. This mix-and-match approach at best only partially cured the direct overlaps, but failed to create a firm with the efficiencies possessed by ARCO. The Commission rejected the proposed consent and sought to enjoin the merger. Ultimately, after extensive negotiations the parties agreed to the divestiture of all of ARCO's complete, free-standing Alaska businesses, including oil and gas interests, tankers, pipeline interests, exploration data, and selected long-term supply agreements. Again, a clean sweep approach was necessary to provide the acquirer of the assets (Phillips) with ability to fully restore competition.

When we depart from the kind of divestiture remedy the Court spoke about in DuPont, it is not always clear that divestiture is "simple, relatively easy to administer, and sure"--at least not retrospectively. That is the lesson of the Bureau's divestiture study. But first let us explain how our policy to merger remedies has evolved.

the government's approach to merger
remedies over the years

The government's position towards remedy has evolved over the past several decades. Prior to enactment of Hart-Scott-Rodino, the government typically was faced with seeking to remedy a merger several years after it had been consummated. That was a daunting, almost always hopeless, task because the assets had been intermingled and the acquired firm typically dissolved. Usually there was relatively little left to divest, at least little that resembled the acquired entity. Sometimes the agencies would require the merged firm to scrabble together various plants and other assets into something that vaguely resembled the acquired entity before the merger. Most often these divestitures were nothing more than "pyrrhic victories."9 DuPont was an easier case, since it involved a partial stock acquisition and it was not difficult (apart from tax considerations) to spin off the acquired stock.

With the enactment of the Hart-Scott-Rodino Act and its mandatory waiting period before merger consummation, the agencies became able to enjoin a merger before the assets were scrambled. That placed the agencies and the merging firms on equal footing in terms of finding the appropriate resolution to a problematic merger.

The agencies' initial response to Hart-Scott-Rodino was to almost always seek a preliminary injunction. Sometimes the agencies would seek to enjoin the entire merger even when the amount of overlap was relatively small. That policy began to change in the early 1980s, when the agencies were more willing to allow firms to restructure transactions to avoid competitive problems ("fix-it-first"), or to engage in partial divestitures. In some cases, such as the Alcan/Arco joint venture, or the GM/Toyota joint venture, the agencies were even willing to resolve their concerns solely on certain forms of behavioral relief.10

During the mid 1980s, there was a shift back toward litigation and away from settlement, at least at the FTC. From 1986 to 1988, for example, of the thirty merger enforcement actions authorized by the Commission, only seven, or 23 percent, were settled prior to litigation with some form of divestiture or behavioral relief. In the vast majority of enforcement actions the Commission chose to litigate. Typically that resulted in the parties' dropping the transaction. In the cases that were litigated, the FTC often prevailed and the merger was preliminarily enjoined.

During the late 1980s and early 1990s, the FTC began to take a more flexible view of merger relief. While divestiture of a plant or facility was typically the most common remedy, the Commission increasingly considered a variety of alternative solutions to competitive problems. In a number of cases, the Commission was willing to accept licensing arrangements (which might eventually result in partial structural relief), supply agreements, and certain forms of behavioral relief, such as firewalls and nondiscrimination provisions. A number of these cases involved high-technology markets in which licensing was used to convey intellectual property rights to bring a new entrant into the market. We had come a long way from DuPont. Not surprisingly, this broader remedy policy resulted in a greater number of settlements and far fewer litigated cases.

When Bill Baer and George Cary arrived at the Bureau of Competition in 1995, they wanted to take a fresh look at the question of merger remedies. There had been a perception, both in the private bar and within the FTC staff, that some merger relief orders had not worked as well as expected. As part of the Baer-Cary initiative, the Bureau of Competition and Bureau of Economics staffs were directed to study what happened as a result of a number of merger consent orders issued from 1990 through 1994.

the ftc divestiture study

Issued in August 1999, the study ("Divestiture Report") has both reinforced some of our approaches to remedies and caused us to rethink others.11 The Divestiture Report provided new insight into the divestiture process, and understanding its lessons is vital for all merger lawyers. It found that in the majority of cases the acquirer of the divested assets was able to enter the market. An important detail, however, was that the likelihood of successful entry was much higher if an ongoing business was divested. A divestiture of selected assets to facilitate entry was significantly more problematic. The Divestiture Report also observed that a number of factors can complicate the divestiture process and lessen the likelihood of success, unless they are adequately dealt with. For example, respondents have incentives to offer a divestiture package that is too narrow, to propose a weak buyer, and to engage in strategic behavior to impede the success of the buyer; and even if they don't affirmatively try to impede the buyer, respondents normally don't have incentives to assist or cooperate with the buyer during the transition phase.

One particular problem identified by the Divestiture Report was continuing relationships between the seller and buyer of divested assets after divestiture, such as a supply arrangement or technical assistance requirement, which may increase the buyer's vulnerability to the seller's behavior. Of the nineteen divestitures where a seller had a continuing relationship with the buyer of the assets, in six cases the ongoing relationship was so detrimental that the buyer could not operate effectively, and in seven cases the ongoing relationship was competitively harmful. Yet those ongoing relationships may be critical to the buyer's success, particularly if less than a separate complete business is divested.

Another significant finding was that buyers sometimes--too often, in fact--have a serious informational disadvantage. They may not fully know what assets they need to succeed in the business, or whether the assets offered by respondents are up to the task. That finding came as somewhat of a surprise, since it was generally assumed that purchasers of divested assets would be informed buyers who could protect their own interests. That assumption isn't necessarily valid when much of the key information is principally held by the respondent. Unfortunately, we face the same informational disadvantage. While we try to learn as much as we can about the industries and businesses we investigate, we don't presume to know how to operate the business. The Divestiture Report also revealed that buyers may not have the same objectives as the Commission, so the remedial purposes of the order may not be met.

Finally, divestitures that include technology transfers present serious difficulties and challenges. They bring together many of the problems already mentioned--respondent's incentive to limit the asset package, the buyer's informational disadvantage, the buyer's reliance on the respondent for technical assistance and transfer of know-how, and the respondent's incentives to engage in strategic behavior. Another difficulty, because technology transfers often involve the divestiture of less than an ongoing business, is that the buyer may be at the bottom of the learning curve and thus starts with a disadvantage.

So it became increasingly evident from a relative early stage of that study that we needed to rethink and modify our approach to merger remedies. If fact, the Bureau of Competition began to incorporate many of those lessons into its remedy approach while the study was still being completed. In 1996, the Bureau adopted several reforms based on initial findings of the Divestiture Report:

  • More frequent use of up-front buyers;

  • Shorter divestiture periods, to minimize the risk of interim harm and dissipation of asset value. The divestiture periods were shortened from twelve months to typically three or four months;

  • Increased use of full or structural relief. The closer the divestiture package is to an ongoing business--better yet, if it is an ongoing business--the greater the likelihood that competition will in fact be restored;

  • The use of interim trustees, especially where technology transfers are involved.

The value of up-front buyers and short divestiture periods is illustrated by the Schnucks supermarket case, where the consent order did not require an up-front buyer. Schnucks Markets acquired its chief competitor in St. Louis, Missouri, and the Commission required the divestiture of twenty-four stores within twelve months. But before the stores could be acquired Schnucks failed to maintain them properly, resulting in a relatively unattractive set of assets.12 The Commission filed a civil penalty action, and Schnucks agreed to pay a $3 million civil penalty and divest two additional stores.13 While that was a substantial penalty, we cannot rely on civil penalty actions alone to ensure that respondents do it right. Obviously, the prospect of substantial civil penalties did not deter Schnucks from engaging in strategic behavior, and it may simply have been an investment or cost of doing business to preserve market power. Moreover, by the time we can bring a civil penalty action, the damage to the market will have already been done. So we have to make sure up front that the remedy really will work.

Up-front buyers are probably the most vital tool in assuring a successful divestiture. It enables us to better determine (a) whether a proposed package of assets that is not a stand-alone business is viable in the real world, (b) whether there is a buyer for the proposed divestiture assets, and (c) the likelihood that the proposed buyer will restore the competition that otherwise would be lost through the merger. This last factor is receiving careful scrutiny. The FTC seeks to assure not only that the buyer will successfully enter, but also that it can restore competition fully.

Up-front buyers are now used in over 60 percent of the cases in which there is some form of non-behavioral relief. There might have been an impression that the up-front buyer policy is reserved for cases where assets may waste quickly, such as supermarkets. That is not the case. The Commission has used up-front buyers in pharmaceutical cases, in other health care products, industrial products such as refractories, acrylic polymers, lead smelters, industrial power sources, and consumer products. (See the appendix for a representative list of cases and markets.) In many cases where the parties have identified an up-front buyer at the beginning of the investigation, the Commission has been able to resolve its concerns and enter a proposed consent order in less than sixty days after the investigation began. The message is straightforward: parties must consider and be able to identify an up-front buyer as part of the merger planning process.

recent application of remedy reforms

The application of remedy reforms over the past few years, especially the greater focus on effective structural relief, has led to claims that the FTC has raised the bar for resolving merger concerns. That characterization is not entirely accurate. We have always insisted on the kind and quantum of relief necessary to protect competition. We evaluate what it takes to preserve or restore competition, based on our experience and the evidence. But as our experience with divestitures grows, so does our understanding of what it takes to successfully remedy the potential anticompetitive effects of a proposed merger. We are more willing today to consider non-litigated resolutions to merger concerns, but that is no more a lowering of the bar than the recent reconsideration of merger remedies has resulted in a raising of the bar.

In reality, the vast number of mergers raising competitive problems are resolved through consent orders that include a wide variety of approaches to relief. In most cases, structural relief involving divestiture of an ongoing business is required. In many cases, a partial divestiture is appropriate, often because it is clear that the acquiring firm has sufficient assets to replicate the efficiencies of the acquired firm and fully restore competition. In other cases, even more refined relief, such as behavioral relief or licensing arrangements, may be used. Again, cases of more limited relief will require a careful assessment of whether the relief can fully restore competition.

One illustration of the Commission's flexible approach is its evaluation of the merger between Ciba and Sandoz. Although divestiture is the "preferred remedy," that does not mean it will be invariably used, especially where it might diminish procompetitive aspects of a merger. This can be a tough issue, particularly in high-technology markets where research and development rights and scientists work together on a number of projects. In the Ciba/Sandoz merger, the Commission chose licensing over divestiture because of the problems of separating ongoing R&D projects.14 Commissioner Azcuenaga dissented as to the licensing aspect of this order, noting that divestiture would cure the anticompetitive problem in a "simple, complete, and easily reviewable" manner. While divestiture is certainly an easier remedy to impose and monitor, it may not always be the most effective way of restoring competition. Because licensing is more flexible and can more easily be tailored to unusual fact situations, it may be the preferred remedy in innovation cases where divestiture could interrupt potentially successful research efforts. In this case, the majority of the Commission determined that the gene therapy research efforts, which contained a number of joint efforts with third parties, would be too difficult to disentangle from the merging firms, and would thus "not only … hamper efficiency but also could be less effective in restoring competition if it led to coordinated interaction or left the divested business at the mercy of the merged firm."15

Another consideration to keep in mind is that many mergers are taking place in a changing market environment. As noted earlier, many mergers we review are large and complex, they involve strategic combinations of businesses, and they may involve new forms of competition. Complex cases are difficult to resolve, and we must be careful that the remedy we accept really will do the job. Not surprisingly, parties are presenting the Commission with proposed orders that are increasingly complex and regulatory. As Chairman Pitofsky has observed, our recent experience is that parties are often presenting "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."16

That said, our approach to merger remedies may affect the resolution of particular cases. Compared to the practice in the late 1980s and early 1990s, we are somewhat more careful in the use of non-structural and partial structural relief. During the mid and later 1990s, the Commission was faced with a number of cases in which the parties proposed relief short of divestiture that was simply insufficient to remedy the competitive problem. The nature of the competitive problem has a lot to do with whether there is an acceptable fix. Some of the mergers during the last few years presented new competitive issues that were not easy to fix, short of blocking the merger. Others posed particularly complex issues of relief. Here are some examples of these types of situations.

partial divestiture of the overlapping assets

Often the competitive problems from a merger can be resolved through the divestiture of some assets in overlapping markets. This is frequently the approach in retail markets where we require divestiture of all the stores in markets where there are overlaps and significant levels of concentration. For example, in Exxon/Mobil we required the divestiture of all gasoline stations from Virginia to Maine to ensure that there was no increase in concentration in northeastern gasoline retail markets. This "clean sweep" approach resolved the competitive concerns in those markets.

Drug wholesalers

Sometimes, such an approach will not be sufficient, especially where competition is not solely local. For example, in the drug wholesalers cases, FTC v. Cardinal Health, Inc. and FTC v. McKesson Corp.,17 the FTC challenged two mergers of the four largest drug wholesalers. As anyone who followed the trial knows, the court explored every opportunity with the parties to find a settlement that could permit the proposed mergers to go forward. The parties suggested that a divestiture of several drug wholesale distribution centers in the Northwest, where there were clear overlaps, would be sufficient to restore competition. As the FTC told the court, that divestiture would have been severely inadequate because, in the FTC's view, customers demanded firms that could provide national service and divestiture of a handful of distribution centers could not compensate for the loss of two national competitors that would have resulted from the proposed transactions. The court found that regional firms did not offer the same level of competitive restraint as the national firms. Thus, the proposed settlement was appropriately rejected and the court issued a preliminary injunction.

Rite Aid/Revco

Rite Aid's proposed 1995 acquisition of Revco would have resulted in a single pharmacy chain of over 5,000 stores. In thirty MSAs in twelve Midwestern states, the firm would have had over a 35 percent market share, and in most of these markets it would have been more than twice as large as its closet rival. Rite Aid proposed to divest some stores where there was an immediate geographic overlap. Had we been concerned only about those retail overlaps, we might have been able to reach a satisfactory resolution. In previous cases where the relevant market was viewed as direct retail sales to consumers, the Commission had agreed to accept divestitures in towns where the firms had immediate overlaps.

Although that remedy might have been satisfactory in the past it was not in this case because the markets had evolved. The competitive problem was not simply the elimination of competition in direct retail sales to consumers but also in a parallel market, the provision of network pharmacy services to pharmaceutical benefit managers ("PBMs") and other managed care providers. These firms use networks of pharmacies to deliver pharmaceutical benefits to consumers. From the perspective of these PBMs it was necessary to form a network of pharmacies in geographically diverse locations. Rite Aid and Revco were direct competitors in providing PBMs with a suitable network, and they often competed to be the "anchor" of the managed care network.

The nature of the competitive concern complicated the remedy issue; it was more difficult to make the divestitures necessary to replace a network than it was to eliminate some local overlaps through the divestiture of a few stores. The proposed divestitures offered by Rite Aid would have eliminated the direct local overlaps, but were simply insufficient for an acquirer of those assets to fully restore competition in this managed care market. Ultimately, the Commission refused the proposed divestiture and authorized staff to seek an injunction. Rite Aid dropped the acquisition, and Revco was acquired by CVS, which currently competes aggressively with Rite Aid in the markets where competitive concerns were raised.

Mediq/UHS

Mediq and UHS are the two largest firms in the country that rent durable, movable medical equipment--such as respiratory devices, infusion devices, and monitoring devices--to hospitals on an "as-needed," short-term basis. Much of the contracting for durable medical equipment is done on a national basis, and hospital chains and group purchasing arrangements require a national network for this equipment. Mediq Inc.'s proposed acquisition of Universal Hospital Services ("UHS") in 1997 would have given Mediq a near monopoly in the national market, and a near monopoly in numerous local geographic markets as well. Competitive concerns were heightened because earlier acquisitions by Mediq had led to higher prices.

In an attempt to forestall litigation, the parties presented a purported "fix-it-first" solution involving Medical Specialties, a firm in the business of renting infusion pumps to home healthcare customers. The parties proposed to sell rental equipment to Medical Specialties and provide it with an option to lease several facilities. Our assessment--and that of customers--was that Medical Specialties would not have been an adequate replacement for UHS. The new firm would have had a substantially smaller inventory than UHS, which itself was considerably smaller than Mediq. Customers--particularly national ones, like hospital buying groups--testified that Medical Specialties would not have the amount and breadth of equipment necessary to replace UHS. Moreover, much of the business that Medical Specialties claimed it needed in order to compete successfully in the hospital rental market was under long-term exclusive contracts with UHS and MEDIQ.

The Commission found the proposed relief inadequate and authorized the staff to seek a preliminary injunction.18 The defendants attempted to short-circuit the litigation by asking Judge Sporkin to approve the proposed settlement, but the judge was unwilling to second-guess the FTC. On the eve of the preliminary injunction hearing, the parties dropped the proposed acquisition.19

behavioral relief

Of course, behavioral relief is typically a less satisfactory solution than structural relief, since it often involves some sort of ongoing regulation. But that does not mean that it is never used. In appropriate cases, the Commission has used behavioral relief such as firewalls and nondiscrimination provisions, particularly to remedy vertical concerns. For example, in the Time Warner/ Turner transaction, the Commission approved the merger based on a wide variety of behavioral rules. In other cases, a behavioral approach may be inadequate.20

Questar/Kern River

The proposed Questar/Kern River transaction in 1995 involved a situation in which a monopolist sought to acquire a 50 percent ownership interest in a firm that was on the verge of entering its market. Questar was the only transporter of natural gas to the Salt Lake City area, and Kern River Transmission Corp. had a gas pipeline that ran past Salt Lake to points further west. Kern River, which was jointly owned by Tenneco and Williams, planned to build a lateral pipeline to serve Salt Lake customers as well. The focus of the case was on transportation service to industrial customers, which could bypass the local utility and purchase gas directly from suppliers and pay separately to have it transported to their facility. Kern River had begun to solicit customers and was already having an effect on the market. Because of Kern River's marketing efforts, Questar sought and obtained a tariff to lower its rates to certain industrial customers, to persuade them not to switch to Kern River. Questar then sought to acquire Tenneco's 50 percent interest in Kern River, with the other 50 percent to be retained by the Williams Companies. The transaction obviously raised concerns because it would eliminate the current price effect of Kern River's presence in the market and prevent future competition and the erosion of Questar's monopoly.

Questar proposed what was in effect a competitive rules joint venture in which it would be permitted to have a 50 percent interest in the Kern River pipeline but Williams would have a large degree of independence on its decisions where to enter. There were several problems with the proposal. First, the agreement undermined Questar's incentives to discount on its own pipeline since it had a 50 percent interest in its only competitor. Questar's 50 percent interest in Kern River would have diminished its incentives to engage in unfettered competition with Kern River; even if Questar lost a bid, it would still have a big share of the business through its interest in Kern River, so it was less likely to bid aggressively. Second, Kern River shipped all the way to California, and the remedy would not diminish Questar's incentives or ability to direct Kern River's capacity away from Salt Lake City. Third, although the "competitive rules" had a "capital forcing" mechanism in which Williams theoretically could have secured Questar's commitment for capital expansion projects, it was unclear this mechanism could work. The Bureau rejected the remedy as inadequate and too regulatory. The Commission authorized a preliminary injunction action, and Questar abandoned the transaction.21 Ultimately, Tenneco sold its share of the pipeline to Williams, which competes aggressively with Questar today.

Barnes & Noble/Ingram

Barnes & Noble's 1999 attempt to buy Ingram Book Group raised a different set of issues. Barnes and Noble is the largest book retailer, and Ingram is the largest wholesaler of books in the United States. Thus, it was largely a vertical transaction.22

The transaction raised concerns principally under the "raising rivals cost" theory. The Bureau was concerned that the acquisition of an important upstream supplier such as Ingram might enable Barnes & Noble to raise the costs of its bookselling rivals, such as independent book retailers or Internet retailers, by foreclosing access to Ingram's books and services or denying access on competitive terms. The rivals would be less able to compete, and Barnes & Noble could increase its profits at the retail level or prevent its profits from being eroded as a result of competition from new business forms such as Internet retailing. We were concerned that the combined Barnes & Noble/Ingram could do that in a number of ways, including strategies short of an outright refusal to sell to the non-Barnes & Noble bookstores. For example, Barnes & Noble/Ingram could choose to (1) sell to non-Barnes & Noble bookstores at higher prices, (2) slow down book shipments to rivals, (3) restrict access to hot titles, (4) restrict access to Ingram's extended inventory or back list, or (5) price services higher or discontinue or reduce these services.

The parties did not present a complete settlement proposal, which makes a discussion of remedies hypothetical. There were reasons to be skeptical that the deal could have been fixed. The nature of the competitive problem would have made it very difficult to address from a remedy standpoint. Structural relief would seem to require the creation of a substitute for Ingram, but that didn't seem to be a realistic possibility. The only remedy that might have addressed the situation is a set of behavioral rules--essentially, a set of non-discrimination or "fair dealing" provisions. But those kinds of rules can be problematic. They are susceptible to evasion and difficult to monitor, particularly in a transactional setting where discrimination could be exercised in subtle ways on several different variables. While the Commission has on occasion accepted some forms of behavioral relief in mergers, those approaches may not have worked in this context. Recall the Supreme Court's admonition in DuPont that "the public interest should not in this case be required to depend upon the often cumbersome and time-consuming injunctive remedy" to enforce behavioral rules.23

Another concern about the merger was that Barnes & Noble could use Ingram to obtain competitively sensitive information about its bookselling rivals. Independent booksellers raised concerns about two types of information they provide to Ingram in the course of their supplier-customer relationship: the financial information they supply to obtain credit, and the titles and quantities of books they purchase from Ingram. Barnes & Noble might use this information for such purposes as targeting promising store locations, identifying competitors' weaknesses, and reaping the fruits of others' marketing efforts. Whether or not the fears were realistic, the fact that they were out there could have had its own dampening effect on competition. For example, independents may have less incentive to develop a market for special interest books if they believe Barnes & Noble would simply free-ride on their efforts, or might have returned their usage of Ingram and been forced to rely on other higher cost book wholesalers.

This concern has been addressed in other cases by obtaining a remedy commonly called a "firewall." Could a firewall work effectively in this case? Most of the cases in which a firewall has been used are situations, such as defense mergers, where there is a regulator which can identify violations of the firewall.24 Even if a firewall could address the information access problem, there was the discriminatory access problem discussed earlier. In the end, we did not have to decide these remedy issues--there was no proposal on the table--but it would have been difficult to find a satisfactory solution. The parties chose to abandon the transaction following press reports that Bureau staff would recommend a preliminary injunction.25

divestiture of an ongoing business

Divestiture of an entire business will usually resolve the FTC's competitive concerns, since there will be some evidence that the business unit has operated effectively and efficiently. But that will not always be the case, as illustrated by the review of the Ahold/Pathmark merger.

Ahold/Pathmark

Last year, Ahold, the fourth largest supermarket in the United States with over 1,000 supermarkets in fifteen states, attempted to acquire Pathmark stores, a regional supermarket chain of 135 stores in metropolitan New York, Philadelphia, and New Jersey. The acquisition was valued at approximately $1.75 billion. Unlike most of the supermarket mergers the Commission had reviewed over the past several years, this deal involved a much more dramatic and clear geographic overlap. Previous supermarket mergers were resolved through consent agreements primarily because the acquisitions enabled the acquiring firm to gain entry into new markets that did not pose competitive problems, and the limited overlap areas that in fact did present competitive problems were resolved through divestitures. The competitive concerns raised by the Ahold/Pathmark transaction were much more serious. Ahold was acquiring a supermarket chain that competed head-to-head with Edwards, a chain that Ahold already operated in the same geographic areas. This was not a geographic extension merger, but rather the elimination of a direct competitor.

The parties' initial proposal of relatively modest divestitures of individual stores in various overlap markets did not meet the standards of recent consent orders in the industry. Based on our concerns from prior supermarket mergers, we typically require divestiture of a single chain's stores to an up-front buyer to resolve competitive concerns. In almost all cases, we require a "zero delta" approach. That is, we require divestiture of a sufficient number of stores to maintain competition at the pre-merger level.

The parties eventually proposed to divest all of the Edwards stores. While that would eliminate the competitive overlap at least nominally--i.e., zero delta--a serious question remained whether a suitable purchaser existed that would fully restore competition. Edwards was a strong rival to Pathmark to no small degree because it was funded by a much larger parent organization, Ahold. Many efficiencies of being part of Ahold would have been lost if Edwards was divested to a smaller rival. Our assessment was that divesting the entire Edwards chain still might not be sufficient to adequately restore competition because another firm might not be able to provide the level of support necessary to keep this same level of rivalry.

We insisted on a high probability of success in the Ahold/Pathmark matter because there is some sense that many of the divestitures in our supermarket merger orders do not succeed. In retail markets, a chain's assets consist of far more than just the individual stores and the fixtures inside (assets that clearly can be divested). Customer and supplier relationships are critical assets that cannot be conveyed in a divestiture. Thus, even where large numbers of stores have been divested, if the stores are not an entire ongoing business, frequently they do not succeed.

mix-and-match approach

Sometimes parties will offer to divest a combination of assets selected from both of the merging firms. This mix-and-match approach requires a more careful review by the agencies than the divestiture of a single firm's business, because the agencies must determine whether the mixed assets can function effectively as an ongoing business. The agencies also must determine whether the mixed assets will be capable of producing comparable efficiencies and economies of scale and scope as the acquired firm. Merging parties must recognize that this type of evaluation will delay the merger review process, and take that into account in their merger planning. The Commission has accepted a mix-and- match approach in some cases, such as the Albertsons/American Stores merger, where the divestiture included stores from both firms.26 In other cases, such as BP/ARCO, a mix-and-match approach was rejected because the proposed divestiture could not have replicated the competitive significance of the acquired firm.

Federal-Mogul/T&N plc

The merger between Federal-Mogul and T&N plc27 is an example of why a mix-and-match approach may not work. Both firms were leading producers of a wide range of automotive parts in Europe and the United States, and the merged firm would have accounted for 80 percent of sales in the worldwide market for thin-wall bearings used in car, truck, and heavy equipment engines. The merger was investigated by multiple jurisdictions. Rather than offering to divest an ongoing business unit, the parties initially proposed to divest a package of assets from both Federal-Mogul and T&N, in both Europe and the United States; they even presented an up-front buyer. Upon close examination, this offer, while substantial, was found wanting. The divestiture package included some of the parties' least efficient production facilities. More important, they offered insufficient research and development assets. We concluded that the up-front buyer's ability to maintain competition in the United States with these assets was questionable at best. We ultimately obtained the divestiture of T&N's entire thin-wall bearings business, which consisted of the assets and plants that T&N used to make thin-wall bearings, as well as the assets, including intellectual property, that T&N used to develop and design new bearings to meet the bearings needs of engines that OEMs will develop in the future. The assets were ultimately divested to the Dana Corporation.

ongoing relationships between merged firm and
acquirer of the divested assets

Many of our consents require ongoing relationships between the merged firm and the acquirer of the divested assets. Often ongoing relationships will be required, especially in pharmaceutical cases, where the acquirer has to undertake a regulatory approval process and may need an interim source of supply during that period. Although the Divestiture Report observes that these relationships can be problematic, often they are successful. The Abbott/Alza merger illustrates where ongoing relationships may raise concerns.

Abbott/ALZA

Many of the FTC's recent enforcement actions have involved pharmaceutical markets. Last fall, it reviewed the proposed acquisition of ALZA Corporation by Abbott Laboratories. The investigation revealed that the proposed merger would lead to serious anticompetitive effects in the market for palliative hormone drug treatment for advanced prostate cancer. At the time of the proposed merger, Abbott already had an 80 percent market share in a two-firm market. ALZA was not yet in the market but was poised to enter within a relatively short period of time, and the investigation confirmed that ALZA would provide vigorous competition when it entered. ALZA was planning to enter with an innovative delivery mechanism providing longer drug deliveries for patients.

Over the course of the investigation the parties presented several settlement proposals that involved selling various assets related to Viadur, ALZA's product, which was still in development, to another pharmaceutical company. The staff had serious concerns about competition being restored based on this arrangement for several reasons. First, the completion and commercial scale-up of Viadur would depend upon the research and development know-how associated with individuals from throughout ALZA's organization for several years as Viadur and its manufacturing processes are optimized and made most efficient. Ascertaining the necessary ALZA individuals was impossible before the product or process variables are known. Second, the acquiring party was a pharmaceutical company that was not in the business of developing innovative drug delivery systems the way ALZA is; the potential acquirer had experience transferring technology associated with ongoing pharmaceutical businesses, not those still in development. Third, the acquirer would have taken several years to be approved by the FDA at its own facility after trying to replicate facilities and processes of Abbott/ALZA's that are not yet even in place, and would have been dependent upon Abbott/ALZA's supplying the product for several years after it completed the development and commercial scale-up process. With Abbott controlling 80 percent of the market, and having such a critical role in the success of any buyer of the assets, it was uncertain whether any divestiture could effectively work. In addition, the length of the supply contract, which would have had to be more than two years, posed significant competitive concerns.

An alternative upon which the acquisition might have been approved would have been for Abbott to divest its own cancer product. That could have resulted in something that resembled the pre-acquisition state of the market. Because this was not a viable option for Abbott, the transaction was terminated by the parties.

potential competition mergers

Increasingly, the elimination of potential competition is a concern in mergers, especially in telecommunications, energy, and grocery markets. In many cases, where the scope of potential competition is relatively modest, divestiture may be sufficient relief. For example, competitive concerns in several supermarket mergers have been resolved through the divestiture of various land sites that were purchased in order to enter new markets. In other cases, where the scope of potential competition is far more substantial, divestiture may be inadequate as illustrated by the Staples/Office Depot merger.

Staples/Office Depot

Staples' proposed acquisition of Office Depot in 1997 involved the two largest office supply superstore chains in the United States. In many geographic markets, the merger would have resulted in a monopoly, and at most there was only one other superstore competitor, Office Max. The parties sharply disputed that office supply superstores were a relevant market, but suffice it to say that the district court ultimately agreed with us. As with the Rite Aid/Revco merger, the parties offered to divest stores in local markets where they had direct overlaps; they proposed a divestiture of sixty-three stores, primarily in merger to monopoly markets.

There were two problems with the proposed solution. First, it did not address a significant potential competition issue. Both Staples and Office Depot had been rapidly expanding into each others' geographic markets where they did not already have a store. The evidence in the case clearly showed that prices were lower in markets where there were two competing superstores, rather than a single superstore, and lower still in markets where there were three superstore competitors. The merger would have eliminated the likelihood of lower prices as Staples and Office Depot continued to invade the other's backyard, and the proposed divestitures did nothing to cure that.

The second problem with the parties' proposed remedy was that the most likely purchaser of the divested stores probably was Office Max, which was already in the market and could provide a basis for achieving reasonable scale economies. But a divestiture to Office Max would result in a duopoly in the overlap markets. It was clearly better to have three competitors than two. Consequently, the Commission rejected the proposed divestitures28 and sought a preliminary injunction, which the court granted.29

The enforcement action has clearly led to substantial benefits to consumers. Both Staples and Office Depot have expanded at a rapid rate, and within three years after the merger was abandoned each firm has surpassed the size that the merged Staples/Office Depot would have achieved. Both firms are competing aggressively, invading each other's markets and driving prices down to levels not even seen before the merger was proposed. Both firms compete aggressively on the Internet, where Office Depot is the clear leader.

coordinated interaction

As markets are becoming more concentrated, there are increasing concerns over mergers that may enhance the ability of firms to engage in coordinated interaction. Almost invariably these mergers are resolved through significant divestitures, typically of ongoing businesses. But where there is no acquirer with the incentives and ability to fully restore competition, even a substantial divestiture may be insufficient.

DuPont/ICI

DuPont's proposed acquisition of the Tioxide division of Imperial Chemical Industries in 1998 was structured in a way that sought to avoid antitrust problems, but in our view it fell short of a satisfactory solution. DuPont was the leading supplier both in the United States and the world of titanium dioxide ("TiO2") pigments, which are used in paints, plastics, paper, inks, and other products to provide whiteness, enhance brightness, and improve opacity. ICI was the second-largest supplier in the world, with plants located both in the United States and abroad. The deal was structured so that DuPont would acquire ICI's TiO2 facilities outside North America, and NL Industries, another competitor, would acquire ICI's TiO2 assets in the United States.

The DuPont/ICI transaction therefore avoided a production overlap in North America. But it did not avoid a competitive overlap, because ICI also was a significant importer of TiO2 into the United States, especially for use in plastics and architectural coatings. In fact, imports accounted for a majority of ICI's sales to North American customers. ICI was also developing new sulfate-based TiO2 products to compete with DuPont's chloride-based products. Consequently, the acquisition would still give DuPont control over a very substantial percentage of the supply of TiO2 for North American customers. Our concern was that the elimination of an important import competitor like ICI could facilitate or increase the likelihood of coordinated behavior.30

DuPont tried to address our concerns by proposing a supplemental basket of other arrangements: It would exclude from its acquisition one of ICI's European plants, which instead would be acquired by NL Industries; DuPont would supply TiO2 products to NL for two years; DuPont would not compete against NL for North American customers by sourcing them from plants acquired from ICI; and DuPont would divest ICI's North American customer lists, current contracts, and customer information. There were several problems with these proposals--the plant that DuPont proposed not to acquire was a relatively minor supplier to North America, and the non-competition agreement would be an oddity for an antitrust order--but the most critical deficiency was that the proposal did not address the elimination of a competitor that stood in the way of coordinated behavior. The parties abandoned the transaction in January 1999.

an observation about these actions

Occasionally, some people question whether mergers should be challenged in court, because they expect that once a firm is "on the block" its days are numbered and it will inevitably cease to be a competitive force. That observation is not supported by the cases discussed in this article. In the ten cases in which the FTC authorized an injunction action, in only one case--Rite Aid/Revco--did the target of the acquisition cease to be an independent competitor.31 In none of the other cases, have any of the firms exited from the market--they continue to be direct competitors.

going forward

Having described in detail which proposed remedies the Bureau did not find acceptable in some recent cases, let us describe the basic contours of the Bureau's approach to remedies:

  • The divestiture of an ongoing business is strongly preferred over more limited forms of divestiture;

  • The use of up-front buyers is critical in making sure that a divestiture package is adequate;

  • We appropriately have a healthy dose of skepticism about proposals that seek to "mix and match" assets from the two firms;

  • Often we won't have sufficient expertise to determine how the divestiture of specific assets will work, and so we will need the assistance of interim trustees; interim trustees will also play a vital role in making sure the acquirer can seek and secure necessary regulatory approvals; and

  • Other forms of relief, such as hold separate orders, will also play an important role.

We welcome your views on these or any other issues involving merger remedies.

Endnotes:

1. Sandra Sugawara, Merger Wave Accelerated in '99; Economy, Internet Driving Acquisitions, Wash. Post, Dec. 31, 1999, at E01.

2. For a more elaborate discussion of many of these factors, see Robert Pitofsky, The Nature and Limits of Restructuring in Merger Review, Prepared Remarks before Cutting Edge Antitrust Conference (Feb. 17, 2000).

3. FTC v. Ruberoid Co., 343 U.S. 470, 473 (1952). Courts have long and consistently held that the Commission's authority to enforce Section 7 includes the ability to condition approval of a merger on the parties' divestiture of certain assets or interests, either by negotiating a consent decree or through litigation. See, e.g., Lieberman v. FTC, 771 F.2d 32, 34 (2d Cir. 1985); Yamaha Motor Co. v. FTC, 657 F.2d 971, 984-85 (8th Cir. 1981), cert. denied, 456 U.S. 915 (1982); United States v. Beatrice Foods Co., 493 F.2d 1259, 1273 (8th Cir. 1974), cert. denied, 420 U.S. 961 (1975).

4. United States v. E.I. du Pont de Nemours & Co., 366 U.S. 316, 326 (1961).

5. 366 U.S. at 330-31. See also California v. American Stores Co., 495 U.S. 271, 285 (1990) (divestiture is "the remedy best suited to redress the ills of an anticompetitive merger"); Ford Motor Co. v. United States, 405 U.S. 562, 573 (1972) (divestiture is "particularly appropriate" in merger cases).

6. 366 U.S. at 334.

7. Id. at 330.

8. 113 F.T.C. 400, 619 (1991), aff'd, Olin Corp. v. FTC, 986 F.2d 1295 (9th Cir. 1993), cert. denied, 510 S. Ct. 1110 (1994).

9. See K. Elzinga, The Antimerger Laws: Pyrrhic Victories?, 12 J.L. & Econ. 43, 65 (1969).

10. General Motors Corp., 103 F.T.C. 374 (1984)

11. Federal Trade Comm'n, A Study of the Commission's Divestiture Process (1999), available at www.ftc.go v/os/1999/9908/ind ex.htm#6.

12. Schnucks was required to divest 24 supermarkets in the St. Louis area as a result of its 1995 acquisition of National Food Markets and was subject to an asset maintenance agreement pending divestiture. As soon as it closed on the National Foods acquisition, it began treating the divested stores as second class citizens. It closed departments, failed to keep others adequately stocked and staffed, unlisted store phone numbers, and referred customers to Schnucks stores that were not being divested. During the year it had to sell the stores, the sales for those stores declined approximately 35%.

For further discussion of the Schnucks case and other supermarket mergers, see David A. Balto, Supermarket Merger Enforcement, Antitrust Rep., Aug. 1999, at 2.

13. FTC v. Schnucks Markets, Inc., Civ. No. 01830 (E.D. Mo., filed Sept. 5, 1997).

14. Novartis AG, C-3725 (Apr. 8, 1997) (consent order) (Commissioner Azcuenaga concurring in part and dissenting in part).

15. Statement of Chairman Pitofsky and Commissioners Steiger, Starek, and Varney at 2.

16. Robert Pitofsky, The Nature and Limits of Restructuring in Merger Review, Prepared Remarks before Cutting Edge Antitrust Conference (Feb. 17, 2000).

17. 12 F. Supp. 2d 34 (D.D.C. 1998).

18. See FTC v. Mediq, Inc., Civ.Act. No. 97-1916 (D.D.C. Aug. 22, 1997).

19. FTC Press Release, Mediq Informs FTC That It Will Abandon Merger With UHS in Face Of Challenge, Sept. 22, 1997.

20. For a discussion of remedies in vertical merger cases, see Richard G. Parker & David A. Balto, The Merger Wave: Trends in Merger Enforcement and Litigation, 55 Bus. Law. 351 (1999).

21. Questar Corp./Kern River Gas Transmission Co., FTC File No. 961 0001 (preliminary injunction action authorized, Dec. 27, 1995); FTC v. Questar Corp., No. 2:95CV 1137S (D. Utah 1995) (transaction abandoned).

22. Although the firms stood principally in a vertical relationship, the transaction also had horizontal implications. At the horizontal level, there were two competitive concerns. First, Barnes & Noble, which had its own distribution centers, could compete directly with Ingram by wholesaling to other bookstores. In fact, Barnes & Noble had announced publicly that it was considering providing wholesale services to other book retailers. Second, Ingram wanted to retain Barnes & Noble as a customer and so offered competitive prices, expanded its range of titles, and improved service. All of Ingram's customers, including independent bookstores, were beneficiaries of this competition, and there were concerns that the acquisition would have eliminated that stimulus to competition.

23. 366 U.S. at 333-34.

24. E.g., Martin Marietta Corp., FTC Dkt. No. C-3500, 117 F.T.C. 1039 (1994); Eli Lilly & Co., FTC Dkt. No. C-3594, 120 F.T.C. 243 (1995).

25. See, e.g., Stephen Labaton, Staff of FTC Is Said to Oppose Barnes & Noble Bid to Wholesaler, N.Y. Times, June 1, 1999, at A1, C9; Patrick M. Reilly & John R. Wilke, FTC Staff to Fight Barnes & Noble Bid for Wholesaler, Wall St. J., June 1, 1999, at B16.

26. For a description of the Albertsons-- American Stores merger and the mix and match approach, see Balto supra note 12.

27. Federal Mogul Corp., No. C-3836 (Dec. 9, 1998).

28. FTC Press Release, FTC Rejects Proposed Settlement in Staples/Office Depot Merger, Apr. 4, 1997.

29. FTC v. Staples, Inc., 970 F. Supp. 1066 (D.D.C. 1997).

30. The investigation revealed that ICI had a unique incentive to import substantial quantities of TiO2 into North America, because of the configuration of its extensive European facilities. ICI in fact had demonstrated a commitment to supply U.S. customers during peak demand periods, and it had been attracting increasing sales. Given its incentive to import, ICI was a potential disruptive force in any scheme to coordinate output or prices in North America. By removing that threat, it could become much easier for DuPont and remaining suppliers to engage in coordinated behavior. Concerns about coordinated behavior were sharpened by the presence of a number of factors that generally facilitate collusion--e.g., inelastic demand and substantial information flows between competitors. Firms had considerable knowledge of their competitors' capacity, pricing, and sales to individual U.S. customers. Thus, firms were capable of monitoring pricing and output and detecting cheating. In addition, DuPont already played a strong price leadership role in the industry, with other firms taking their cues from DuPont. The elimination of ICI's import competition could only strengthen that role. Those concerns were heightened by evidence that North America's price declines during slack demand periods already were shallower relative to other regions. FTC staff also were concerned that with a more commanding position wordwide, DuPont would have increased incentives to close some of the capacity acquired from ICI to demonstrate its resolve to promote higher prices and encourage investment restraint by other suppliers.

31. Revco was acquired by another firm.