It is a pleasure to address the Corporate Counsel Institute here in Atlanta. I have just completed my first year as the Deputy Director of the Bureau of Competition for mergers. I chose a particularly interesting and busy year to return to the FTC after a hiatus of a dozen years. This last year set a new record for the number of mergers filed with and reviewed by the FTC, and I am pleased to have the opportunity to describe our merger enforcement program of the past year. Not only has the year been busy, but it has provided some particularly interesting transactions to review. As the current merger wave has been characterized by strategically-driven transactions rather than the financially driven deals of the last merger wave, the antitrust issues have also become increasingly complex and challenging. Perhaps no other deal exemplifies the current wave as much as the transaction involving your home-town company, Turner Broadcasting. Today, I will talk about the three key initiatives driving our merger enforcement policy, and will describe several of our more significant merger challenges.
The past year has seen the implementation of three key initiatives to improve the enforcement of the antitrust laws governing mergers and acquisitions: (1) not hesitating to recommend to the Commission aggressive enforcement action where the facts revealed in an investigation so warrant; (2) improving the efficiency and quality of the merger investigation process; (3) and insisting on effective relief where a merger threatens competition. These goals are a tall order, particularly at an agency that was widely (and rightly) acknowledged as one of the best in Washington for many years, and one that has strived to constantly improve both its process and its substantive analysis. Nevertheless, I am pleased to report that we have made substantial progress toward each of these goals.
I'd like to illustrate our efforts towards implementing each of these initiatives by describing some of our merger enforcement efforts over the last year.
I will start first by looking at the way we have tackled the substantive review of mergers, as that review played out in three recent merger matters. The key here is that we are committed to the serious economic and legal analysis necessary to figure out whether a merger would likely reduce competition. This requires a certain openness to new mechanisms of competitive harm which flow from the facts of a particular case, rather than limiting our thinking to anticompetitive mechanisms that we have seen before. Antitrust must be a fact-based analysis: rather than attempting to fit the facts into preconceived theoretical models, we need to apply economic theory creatively to understand and explain the facts revealed in an investigation.
Once having found a problem, the Commission must decide whether to block the deal outright, or whether lesser relief, such as divestiture in overlap areas, will adequately resolve the competitive problem. Before we choose to settle, we must carefully analyze any prospective relief to ensure that it really does promptly remedy the competitive problem identified and that it will in fact work. We therefore have been looking carefully at crafting relief that will have a greater effect in restoring competition. Bureau of Competition Director William J. Baer and Bureau of Economics Director Jonathan Baker initiated a divestiture retrospective to help us evaluate whether the divestiture orders entered in the past to remedy anticompetitive mergers have fully restored competition, and if not, what we should be doing to improve upon those orders. As a result of that effort, we announced a number of steps to improve the efficacy of our orders. Among those were:
- We have shortened the divestiture time period which we have included in our orders to no more than six months from twelve months;
- We have taken steps to make the appointment of a trustee swifter and as automatic as possible in the event that the parties fail to perform their obligations under the consent decree;
- We are more careful in determining precisely what assets must be divested; requiring the divestitures of complete businesses rather than incomplete product lines which may not be viable standing on their own, and creating incentives for key employees to stay with the divested business when necessary; all of this is to ensure as much as possible that the divestiture package will be saleable and will in fact restore competition in the affected markets;
- We favor "crown jewel" provisions whereby a respondent promises to divest additional assets making the divestiture package more valuable if the original divestiture package proves to be unsaleable.
- Finally, and perhaps most importantly, we have insisted on buyers in hand before closure of the transaction if we have doubts about whether a buyer can be found or if a divestiture is viable. At the same time, a buyer in hand will speed divestiture since the public comment period can be used both to comment on the consent agreement as well as on the strength of the divestiture candidate.
These efforts have substantially reduced the average time for divestiture from 15 months in 1995 to 10 months in 1996. The average should continue to decrease as a larger proportion of the orders in the sample include these provisions. I will illustrate some of these points in the context of our resolution of the Royal Ahold/Stop and Shop merger. Of course, where we do not see a viable settlement option (as in the Rite-Aid/Revco merger I'll be talking about shortly), no amount of relief will be sufficient short of a full-stop injunction.
Now let's turn to an examination of three major cases to get a sense of substantive merger review and relief policy at the FTC.
Rite Aid - Revco:
I first will talk about the Commission's challenge of the Rite Aid/Revco acquisition, which was abandoned by the parties after the Commission voted to seek to enjoin it.(2) The underlying economic theory of the Commission's case applied a variant of an "auction model." This theory, which is not typically applied in merger analysis, nonetheless seemed to provide a model that matched the competitive dynamic described by the participants in the market. (For a detailed description of this economic model I commend to you the released text of Bureau of Economics Director Jonathan Baker's speech to the ABA annual meeting).(3) The case is also notable in that it broke from the mold established in several other drug store merger cases the Commission considered over the last few years. This case, therefore, is a good illustration of my theme that the economic theory of competitive harm should follow the facts revealed in the investigation.
I will not repeat Dr. Baker's remarks, but will talk briefly about some of the more interesting aspects of this case. I will start with product market, describe geographic market and then discuss the evidence behind the unilateral competitive effect that drove the Commission to take action to stop this merger.
The merger involved two large operators of retail pharmacies in the United States, each with a substantial market share in numerous metropolitan areas. The Commission was concerned that the combined firm would be able to exercise market power in the sale of retail pharmacy services to managed care providers offering pharmacy benefits to their subscribers. In the past, the Commission's challenges had been in the market for cash paying customers. Over time, though, more and more prescription drugs are being purchased through pharmacy benefit plans by third party payers as the market shifts to managed care. Third party payers, often acting through pharmacy benefit management programs ("PBMs"), now are purchasing services for a geographically dispersed group of subscribers.
The first question, of course, is why this shift in the market should make a difference for antitrust analysis. We found that the third party payor market works quite differently from the cash paying customer market. Let me explain. First, whereas an individual cash paying customer can choose from the stores near his home or work, third party payers must offer a network of pharmacies geographically dispersed throughout the area where the employer's covered workforce lives. This kind of extensive coverage can only be provided cost effectively by chains with multiple locations in various geographic areas. Second, prices to third party payers and their PBMs are effectively established by competition between the lowest cost suppliers, which are inevitably the larger chain drug stores who benefit from economies of scale. Only after a price level is set through this competition do smaller, higher cost retail operations join the third party payor's network. However, these higher cost independents would not themselves compete prices down to that same level in the absence of chain store competition. Consequently, if the low cost chain is large enough so that it represents such a large percentage of the pharmacy counters in the area where the employer needs coverage that it becomes indispensable to the network, it can hold out for a higher price. Similarly, if the largest two competitors merge, leaving only smaller less cost effective competitors, the merged entity can also withhold its participation in the network except at higher prices than each would have been willing to participate at premerger. The network simply will not get enough participation at a low price because higher-cost stores will not participate; the network will therefore have to increase the price it is offering in order to induce the participation of the large merged chain, or higher cost competitors.
Let's say that third party plans and their subscribers demand the convenience of broad coverage of their market -- say, hypothetically, participation by 60% of all local pharmacies. Because Rite Aid and Revco have significant market shares, the merger would allow Rite Aid to make a higher all-or-nothing offer for their total share -- say, hypothetically, 35% of an area's pharmacy counters -- than either would have been able to premerger. This would force a PBM to do one of two things: either it could reject the higher Rite Aid all-or-nothing offer and go deeper and deeper into higher cost independent pharmacies in order to retain its 60% coverage, which would, in effect, raise the PBM's costs of doing business. Or it could cave in and accept the higher Rite Aid price. Either way, PBMs pay more and consumers lose.
The geographic market analysis is also different for third party payers than for cash paying customers. For third party payers, the geographic market quite literally is the area that each employer offering a prescription benefits plan must cover for his or her employees. Of course, the map for each employer would be a different amoeba-shaped service area representing the residential pattern of each employer. The pharmacy benefit plan managers that organize the networks would need to contract with pharmacies covering the overlapping areas of each of its third party payers.
Obviously, defining such geographic markets with any precision is impossible. For pleading purposes, we settled on numerous representative Standard Metropolitan Statistical Areas ("SMSA's") and states. We used SMSA's because many employers in a metropolitan area have employees throughout the area. We used states because there are many state-wide employers that contract statewide: for example, public school teachers, Blue Cross, and the state government employees health plan. Because most employers are focused in one metropolitan area, we do not believe this shorthand does violence to the underlying economics of the market.
The numerous geographic markets and the fact that some of the markets overlapped others made it very difficult for the parties to propose a settlement that gave us any confidence that the anticompetitive problems identified could be resolved. A few local divestitures along the pattern of previous drugstore mergers simply would not solve the problem. Even if the parties had divested within an SMSA, the competitive problems would not have been solved because many statewide employers (and their third party representatives) need a statewide network of pharmacies that only a chain can provide economically. Because statewide customers need a low-cost chain that can "anchor" the network, local divestitures in discrete areas simply would not do the trick. In this market, dismantling a functional and efficient competitor could be as anticompetitive as its acquisition. According to press accounts at the time, Rite Aid complained that the FTC was "raising the bar," by rejecting various proposed divestitures. Nothing could be further from the truth. Rather, given the complexities that I've described in defining economically meaningful geographic markets, even divestitures in certain highly concentrated cities would not necessarily resolve the problem in broader regional or statewide markets corresponding to particular employers geographic markets. At the same time, given the importance of efficient chains in bidding down prices, we could not accept a divestiture which could eliminate a strong chain competitor in favor of a series of less efficient and therefore noncompetitive smaller chains or independents.
Finally, the impact on competition of proposed divestitures could not be known unless we could know exactly who the purchaser was in various markets, and what the competitive situation would look like after the divestiture. Unlike in many markets, the specific identity of the competitor and its size in various markets was relevant to judging the competitive results of the divestiture. In short, this case may, as much as anything, stand for the proposition that not all cases can be settled consistent with our mission to maintain competition. In such circumstances, we will be prepared to litigate rather than accept inadequate relief.
Rite Aid/Revco shows our substantive focus on analyzing competitive facts in a particular transaction and then fitting those facts to theory, rather than forcing facts into predetermined theory. As such, it fits within one of the goals I set forth at the beginning for our merger
enforcement program. It also accomplished another goal that Director Baer and Chairman Pitofsky set for the Bureau, which is working more cooperatively with State Attorneys General. In this matter, we shared a good deal of information with the six states and anticipated litigating the case cooperatively with several of them. We found the relationship quite beneficial since, by dividing responsibilities to avoid duplication, we were able to save resources.
Questar/Kern River
The next case to illustrate my theme that we will pursue anticompetitive mergers even if the legal or economic theory is unusual is the Questar/Kern River case. The case involved the merger of two natural gas pipelines. This case is unusual because it is an actual potential competition case. Our investigation revealed that Utah customers were actually getting lower prices because of their ability to threaten the local gas company with competition from the California-bound pipeline.
The acquiring company, Questar, had a pipeline that ran into Salt Lake City and serviced customers via a wholly owned local gas distribution company.(4) The to-be-acquired company, Kern River, ran a pipeline by Salt Lake City, on the way to California. The deregulation of natural gas by the Federal Energy Regulatory Commission created the potential for Kern River to supply Questar customers. Because of the additional cost of building lateral pipelines from the Kern River pipeline to industrial customers to bypass the local gas company, however, Kern River had never actually won a sale. It did however, keep on trying to actively solicit industrial customers in the Salt Lake City area. Given Questar's large fixed costs and low marginal costs, it is likely that Questar would not have underbid Kern River at any price high enough to justify building the bypass. Thus, one might ask, why we would bring such a case. Well, the evidence showed that the mere potential for customers to switch to Kern River had a tremendous restraining effect on Questar's ability to raise prices to Salt Lake City customers. Industrial customers were desirous of a way to avoid being tied to the local gas distribution company that was wholly owned by Questar, and thus continued to pursue Kern River as a potential alternative source. Now, some in the rational expectations school of economics might have argued that, at some point, the California pipeline would have quit beating its head against the wall and stopped trying to sell in Salt Lake City. But the reality -- as demonstrated by the evidence in that case -- showed that prices were significantly lower as a direct result of customers' knowledge that the California pipeline could be tapped. It was therefore quite likely that, after the acquisition, prices would have gone back to monopoly levels.
Moreover, in the near future, there were other opportunities for Kern River to serve Salt Lake City: in 1997, industrial customers would be able to access Kern River through Questar's wholly owned local gas distribution company; and, in 1999, the local distribution company's exclusive contract with Questar expires and the Utah Public Service Commission could order that future arrangements be subject to competitive bidding from Kern River. All of these factors led the Commission to challenge the transaction in federal court. We believe that preserving this competition saved customers millions of dollars annually.
Ahold/Stop & Shop
The third case I want to talk about today, the Ahold/Stop & Shop merger,(5) is a traditional one involving supermarket merger analysis, but it demonstrates the Bureau's emphasis on real relief that will solve competitive problems. We believe that these new standards will both improve our divestitures and benefit the parties through faster resolution of divestitures. As I have mentioned, our divestiture study reveals good reason to be concerned that some of the Commission's prior orders have not restored competition to the extent anticipated when entered into. Once the transaction closes, the parties' incentives to devote energy to lining up a buyer for the to-be-divested assets and otherwise beginning the process of divesting the assets are diminished. Consequently, there is a tendency for divestitures to take longer than promised. There might even have been, unfortunately, attempts to "game" the process by selecting the least competitive purchaser, by dragging out the divestiture process, and even by outright refusing to shop the assets as required under the order. For these reasons, or because the package of assets included in the order was insufficient to create a viable competitive entity, many divestitures simply have not resulted in a restoration of competition. Consequently, the Bureau of Competition has pursued shortened divestiture periods of, as a general matter, six months (or shorter). If there is uncertainty about whether a narrow product line is saleable or viable, the Bureau will insist that respondents produce a buyer in hand, divest the whole business, or put their money where their mouth is and agree to a "crown jewel" alternative divestiture of a greater asset, in the event that the parties are unable to divest as promised.
This brings us to the Ahold case. Because of competitive overlaps in various parts of Connecticut, Rhode Island and Massachusetts, the proposed order required Ahold to divest a total of 30 supermarkets in 14 communities. In this case, the parties came to us early and energetically sought good quality buyers for these stores. They rounded up four proposed buyers with letters of intent before the Commission accepted the consent agreement for public comment. Consequently, the public comment served the twin purposes of giving the public a chance to comment on the substance of relief obtained as well as the opportunity to comment on the suitability of the buyers. The order required that the sale be completed within 30 days after the Commission gave the settlement final approval. If the divestitures are not completed on time, the order permits the FTC to appoint a trustee who, with the FTC's approval, could divest these 30 supermarkets, different stores in the same markets, and even additional stores, if he or she believes it necessary. This added kicker gives Ahold a strong incentive to divest in a timely fashion -- since failure to do so would allow the trustee to sell additional stores that Ahold doesn't want to lose.
This process was good for consumers, and good for the parties. Because the parties came in early, staff could advise as to which buyers were the most acceptable from a competitive standpoint and the divestiture process moved quickly, saving the parties from distractions after-the-fact or needing to undo a sale that the FTC would disapprove. For our part, the compliance burden on the Bureau is substantially reduced. Finally, this procedure provides a level of comfort that competition will be unaffected by the merger.
In the realm of remedy policy, I should note that the Bureau has become more aggressive in appointing trustees where the parties failed to divest on time. In Rite Aid,(6) Revco(7) and Cooper,(8) all involving overdue divestiture, the Commission has appointed trustees to divest the assets.
Improving The HSR Process
We must do all this at a time when merger activity is at record levels, but when our resources have actually declined. That means we have to do it more efficiently, and more effectively than in the past. Significant changes in the process of merger review are designed to help us do that while simultaneously easing compliance burdens on business. The FTC, working with the Antitrust Division of the Department of Justice, has sought to address the HSR process and concerns about burden and reporting requirements. Several types of transactions that are unlikely to raise antitrust issues -- such as certain real property asset sales -- have been exempted from the reporting requirements all together. This change is anticipated to reduce filings by 7%. We also sought to accelerate clearance times through streamlined joint review with DOJ. The average time period has been shortened by almost 40% from 17 days to 10 calendar days. That extra time for investigation and review gives us a greater opportunity to determine whether a second request is really necessary and, if so, what markets it should cover. The result, as one would imagine, have been fewer second requests. Since we adopted the expedited clearance procedures, we have increased the percentage of transactions we have investigated during the initial 30 day waiting period, while simultaneously decreasing by 40% the number of transactions requiring a second request. These improved statistics have also benefitted from an annotated, uniform second request that both agencies are using, along with continued use of the "quick look" policy.
Of course, HSR reporting requirements remain the major tool that permits us to do our job and, consequently, we have come down hard on companies for failing to file as required or failing to produce all of the materials required to be produced by item 4(c) of the premerger notification form. The bottom line is that we take these statutory requirements seriously, since they are critical to doing our job, and we will enforce those requirements vigorously. In Sara Lee,(9) the Commission obtained a record $3.1 million civil penalty to settle charges that it violated the HSR Act by failing to file notification in connection with its acquisition of Griffin and other shoe polish brands from Reckitt & Colman. In that matter, we believed that the parties had deliberately understated the value of U.S. assets (vis-a-vis foreign assets) that they were acquiring in order to hide from our scrutiny a highly problematic transaction. (In 1994, we required the divestiture of some of the assets wrongly acquired by Sara Lee.) In Titan Wheel International, Inc.,(10) we alleged that Titan acquired beneficial ownership of certain assets of Pirelli Armstrong Tire Corp. prior to the expiration of the waiting period. To rectify this violation, Titan agreed to pay $130,000, a 100% civil penalty representing $10,000 for each day after the parties transferred control of the plant in question before the Commission granted early termination of the waiting period. Titan also agreed to change the effective date of the transfer of control so that it would follow the expiration of the waiting period -- in effect, transferring possession and control back to Pirelli until closing. Finally, in another recently announced failure to file case, Foodmaker, we sought and received substantial civil penalties, of $1.25 million, consistent with the respondents financial obligations to creditors.
But failure to file is not our only concern. For some time, we have been concerned about whether private parties were taking seriously Item 4(c)'s requirement that they produce competitively-sensitive studies and analyses of the deal. Since coming back to the agency from private practice, I have seen repeated instances where significant 4(c) documents showed up only in response to the Second Request and were not contained in the initial HSR filing. When we proved to learn the reasons for the violation, parties would sometimes assert privileges while simultaneously asserting inadvertence and good faith. In such a situation, we have taken the position that parties are estopped from arguing inadvertence, and we have treated the violation as intentional. If we discover the violation before the transaction is consummated, our policy is to automatically deem the filing incomplete and start the waiting period over again. Staff is entitled to all of the statutory time with complete information so that our investigation can progress at the strict pace set by the HSR Act.
A recent case, Automatic Data Processing Inc.,(11) shows what can happen when we discover the violation after the transaction has closed. In that case, we did not get any 4(c) documents, and we thus did not issue a second request. After the deal closed, we received a raft of customer complaints, reopened the matter and issued a subpoena. What we found shows the importance of 4(c) documents to proper antitrust analysis. One document read: "[]he acquisition of AutoInfo would enable [ADP] to monopolize the salvage industry in an expeditious[] and timely manner." Another document states that the acquisition "would give us a virtual monopoly in salvage information services . . . . Very good for ADP!" In settling this egregious HSR violation, we demanded and got the maximum penalty of $2.97 million-- $10,000 per day for each day that they had failed to comply with the HSR Act.
On November 14, the Commission filed suit challenging the transaction as creating a monopoly.
I think this record demonstrates that we are making substantial progress in improving both the substance and the process of the FTC's merger program.
Efficiencies
As a final note, I want to mention a substantive question which is just beginning to take shape -- the role of efficiencies analysis in merger review. I am sure that most of you know that efficiencies was one of the topics touched on in the May 1996 FTC Staff Report,Anticipating the 21st Century: Competition Policy in the New High-Tech, Global Marketplace. I am sure that most of you also know that we are beginning to dialogue with our counterparts at the Antitrust Division over how to deal with efficiencies in merger review. Although much of this dialogue is still evolving, let me give you my sense of where we should be going. As the Staff Report suggested, the question is how to go about analyzing in specific cases the relatively uncontroversial premise that merger-generated efficiencies affect competition between firms, often to the benefit of consumers. I can at least report some propositions which are guiding my own thinking in this endeavor, more from the point of view of what we should not be trying to do. First, this effort is not designed to turn merger analysis into a full blown rule of reason inquiry. To do so would do violence to Congress' goal in enacting the Clayton Act of preventing anticompetitive market structures in their incipiency. Second, it is not designed to substitute the presumption that the competitive market serves consumers best with an agency determination of when and whether consumers will be better off without competition because of efficiencies. Finally, efficiencies are most relevant to our understanding of whether, on balance, the market will be more competition after the merger, thereby reducing prices for consumers. This effort is not designed to allow mergers resulting in price increases at the expense of consumers in order to achieve efficiencies that benefit only merging parties. The only efficiencies that count in a competitive effects framework are those that inure to the benefit of competition, i.e., consumers.
We are aiming at a realistic articulation of what the agencies actually do with respect to efficiencies analysis. In other words, we are seeking to give the outside world a sense of the internal agency "common law" and analysis that has developed with respect to efficiencies. We also want to put some evidentiary content to our guidance, and not just leave it on a conceptual level. In essence, we want to develop a road map as to what should be taken into account, and set forth a methodology for taking those things into account.
In sum, I think the Bureau has made some significant advancement on the goals Director Baer and Chairman Pitofsky set forth last year. We've still got a ways to go, though, and we've got some issues -- like efficiencies -- that are only now beginning to develop. Hopefully the result is better enforcement of our competition laws to the benefit of consumers, while not needlessly burdening business or interfering with unobjectionable transactions.
1. These remarks are my own, and not necessarily those of the Federal Trade Commission or any individual Commissioner.
2. FTC Will Seek to Block Rite Aid/Revco Merger, FTC News, Federal Trade Commission, April 17, 1996; Rite Aid Abandons Proposed Acquisition of Revco After FTC Sought to Block Transaction, FTC News, Federal Trade Commission, April 24, 1996.
3. Jonathan B. Baker, Unilateral Competitive Effects Theories in Merger Analysis (Prepared Remarks at the ABA Annual Meeting, August 6, 2006).
4. FTC to Challenge Questar Acquisition of Kern River, Alleging Monopoly over Natural Gas Transmission into Salt Lake City Area, FTC News, Federal Trade Commission, December 27, 1995. The parties abandoned the transaction shortly thereafter.
5. Ahold/Stop & Shop, File No. 961-0052 (July 15, 1996) (proposed consent agreement issued for public comment).
6. FTC Appoints Trustee to Divest Pharmacy Assets in Maine and New Hampshire after Rite Aid Misses Deadline, FTC News, Federal Trade Commission, Feb. 8, 1996 (Rite Aid Corporation, C-3546 (1994)).
7. FTC Appoints Trustee to Divest Revco Pharmacy Assets in Marion and Covington, Virginia, FTC News, Federal Trade Commission, Feb. 8, 1996 (Revco, C-3540 (1994)).
8. FTC Appoints Trustee to Find Licensee for Start Up Business in Industrial Fuse Manufacturing, FTC News, Federal Trade Commission, Feb. 13, 1996 (Cooper Industries, Inc., C-3469 (1993)).
9. United States v. Sara Lee Corporation, No. 1:96 CV 00196 (D.D.C., Feb. 9, 1996).
10.United States v. Titan Wheel International, Inc., Civ. Action No. 96 1040 (D.D.C., May 6, 1996).
11.United States v. Automatic Data Processing, Inc., No. 1 :96CV00606 (D.D.C. March 27, 1996).