American Bar Association, Section of Antitrust Law, 44th Annual Spring Meeting, Omni Shoreham Hotel
I am delighted to provide an FTC perspective on hot topics in antitrust and consumer protection. But I don’t want to overstate the temperature at the tip of the Federal Triangle. FTC enforcement activity over the last few years has corrected some imbalances that characterized the 1980’s, as Chairman Pitofsky often points out, and has returned to the mainstream. So, while some of the topics I will discuss today are quite warm, you need not worry that the FTC building is ablaze.
Let me begin with antitrust. Many have noticed that there are fewer decisions in fully-litigated cases than there were a couple decades ago. The reduction in guidance from the courts makes it more important that government enforcement agencies use the tools available to us — including Guidelines, analyses to aid public comment on consent settlements, and speeches — to fill the vacuum. Today I will provide my perspective on some recent Commission enforcement actions. Even more than usually, I want to emphasize that the views I am presenting are my own, and not necessarily those of the Commission or of any Commissioner.
I will not bore you with a recital of every case the Commission has brought since the last Spring Meeting. Nor will I share insiders’ dope about the ones I disagreed with, because, of course, there weren’t any. Actually, you might find it interesting to learn how much the managements of the Bureaus of Competition and Economics are in synch: I looked at all the antitrust matters on which I sent memos to the Commission, and found that Bill Baer’s office agreed with my recommendation more than three-quarters of the time.
I begin with mergers because that is where most of the enforcement action still is. I will talk about unilateral theories of competitive effect, and about making divestitures more effective.
Several recent merger cases demonstrate the importance of theories of unilateral competitive effects. We routinely employ all the unilateral theories set forth in the Guidelines. It is striking that four years after these theories were first laid out in the 1992 Horizontal Merger Guidelines, the competitive problems we now find with mergers more often involve the removal of direct competition between the merger partners than the threat that coordinated behavior will become more likely or more effective.
One case involved the bidding model set forth in footnote 21 of the Merger Guidelines. The merging firms sold an input to downstream manufacturers. The merging parties pointed out that many manufacturers had qualified only one supplier. They argued that suppliers not presently qualified could not constrain pricing much, so the deal would not affect competition. But customers told us that the other two potential suppliers “waiting in the wings” were constraining the price charged by existing suppliers. If so, the transaction reduced the number of bidders from three to two, or two to one, in many of what were effectively auctions by buyers seeking specific types of the input.
A merger we analyzed last year involved an interesting variation on the Guidelines’ unilateral competitive effects theory for differentiated products. We asked whether the transaction would give the merged firm a unilateral incentive to raise price even though the merger partners were arguably not the closest competitors for any customers. To see the issue, suppose there are three hospitals, labeled for present purposes as A, B and C, with B sitting between the other two geographically. Suppose further that a hospital is only a significant competitor with its immediate neighbor. Before the merger, hospital A’s pricing is constrained by hospital B only; hospital C’s pricing is constrained by hospital B only; and the hospital in the middle, hospital B, is constrained by both A and C.
Hospitals A and C now merge. Our theory was that in a plausible model of one- time oligopoly interaction — this story is not about coordination among the three hospitals in the sense of repeated games — B becomes less of a constraint on the merged firm than it was on hospitals A and C individually. Before the merger, A sees that if it raises price, B is not free to follow very far because it will lose too many customers to C. And C thinks likewise. By merging, A and C can raise price simultaneously; under such circumstances, B is more likely to accommodate that price rise by raising price itself than it would have before. With knowledge of B’s likely reaction, the merged A-C firm would have an incentive to raise price unilaterally.
In another unilateral effect case, the merger appeared to enhance the market power of a dominant firm. The transaction raised the acquiring firm’s market share to nearly 70 percent, and left only one other non-trivial producer. Such a merger may lead the merged firm to raise price, as the Merger Guidelines point out, because it provides the merged firm a larger base of sales on which to enjoy the resulting price rise and also eliminates a competitor to which customers otherwise would have diverted their sales. In our case, the remaining firm would have had the capacity to expand output if the dominant firm tried to raise prices post-merger, and thus the ability to undermine the feared anticompetitive effect. The staff investigation properly focused on whether such a strategy would be profitable, and thus on whether non-party expansion would solve the competitive problem.
My other merger topic involves effective remedies for otherwise anticompetitive acquisitions. The Bureau of Economics is in the middle of an ambitious study of a large number of past Commission divestiture efforts, to learn what works and why. During the past year, we combined forces with the Bureau of Competition on some preliminary analyses, to see what we could learn about improving our prospects for success in restoring competition through divestitures before the results of the full study are in. Although our preliminary sample was quite small, and our review of those cases limited, the early results were eye-opening. We saw how respondents undermined our ability to find a strong potential buyer or hampered the buyer’s ability to succeed — in short, we discovered ways in which the Commission had been snookered.
As Bill Baer described at breakfast this morning, you will see us respond by doing business differently. We will seek to shorten divestiture times and encourage merging firms to identify a buyer when proposing settlement. These steps will reduce the time in which the merged firm will have control over the assets it must divest. They will also help the Commission ensure that the buyer is viable. If the buyer appears weak, and the Commission staff has learned about potentially superior alternatives during the investigation, the merging firms can expect some hard questions about their settlement proposal. In addition, we will be more skeptical of relief that requires discretionary post- divestiture actions by respondent, as may arise when the divestiture involves technology transfer or a supply contract, and we will more often insist on crown jewel provisions in such cases. In order to enhance the stand-alone viability of divestiture packages, we will be more skeptical of proposals to divest a narrow package of assets. In consequence, you should expect to see us more frequently insisting that divestiture packages go beyond the narrowest conceivable set of assets that would restore competition where there is a competitive overlap.
I will now turn to antitrust enforcement in the non-merger area, first vertical restraints then horizontal. Antitrust has long been interested in the possible anticompetitive consequences of vertical restraints, although government enforcement in this area ebbed during the 1980s. Vertical arrangements can and often do achieve efficiencies. But vertical restraints can also impair horizontal competition. When they are harmful, it is because they have horizontal effects. For example, vertical restraints can facilitate horizontal coordination, or they can create what I think of as "involuntary cartels" by foreclosure or raising rivals' costs. A recent Commission action involving a "most- favored-customer" or "most-favored-nations" clause illustrates the revival of vertical antitrust enforcement to address reductions in horizontal competition.
A "most-favored-customer" clause is a promise by one party, for example a supplier, to treat a buyer as well as the supplier treats its best, "most favored" customer. If the supplier lowers price to someone else, then the buyer's price will be lowered to match. Earlier this year the Commission issued a complaint and consent order against the use of such a contractual provision by RxCare, the leading pharmacy network in Tennessee. The Commission concluded that a most-favored-customer clause in RxCare's contracts with participating pharmacies tended to keep reimbursement rates high by discouraging selective discounting and the development of rival networks. The primary theory of the case was that the most-favored-customer provisions facilitated horizontal coordination by the pharmacists; this "facilitating practices" theory is distinct from the equally interesting "raising rivals' costs" theory behind some recent Justice Department cases involving most- favored-customer provisions.
The most-favored-customer provisions in the RxCare case discouraged pharmacists from joining discount networks promising additional business but offering a lower reimbursement rate — that is, they discouraged pharmacists from cutting price to attract new business — even though RxCare was nominally non-exclusive. The reason? The pharmacists would then be required to cut price to the RxCare network, which is the largest source of third-party business for Tennessee pharmacies. Third party payors could not plausibly avoid the high price implications of the most-favored-customer clause by assembling a rival network that excluded RxCare pharmacists, because almost all the state's pharmacists were RxCare members. And because RxCare was owned by a pharmacists' association, it had less incentive than would an independent network to bargain aggressively with its pharmacist members in order to offer low reimbursement rates to third party payors. To the contrary, RxCare actually sought to persuade third party payors to raise their reimbursement rates to the RxCare level. These anticompetitive incentives were significant: according to the complaint, reimbursement rates were higher in Tennessee than in other states.
Just this week the Commission issued a decision that may become a landmark in the evolution of the rule of reason analysis of horizontal restraints. The first thing to notice on picking up the California Dental Association decision and order is the heft of the package. The order plus Chairman Pitofsky’s majority opinion plus Commissioner Azcuenaga’s dissent plus Commissioner Starek’s statement add up to more than 100 single-spaced pages. The pile of pages is too thin to be used as a doorstop, but too long actually to read without a good reason, so my job today is to supply the reason.
The case reviewed advertising restrictions adopted by a state trade association of dentists. Let me highlight a few of the decision’s headlines. First, agreements among competitors creating broad, categorical bans on truthful and nondeceptive price advertising — such as restrictions on advertising “low” or “reasonable” fees — are now illegal per se, even when imposed by professional associations. Twenty years ago, in AMA, such restrictions were reviewed under the rule of reason. Here the Commission announces that it has learned enough since to place these horizontal restraints in the per se category. This holding applies the Catalano principle that horizontal restraints on particular forms of price competition are illegal per se even if other forms of price competition are available.
Second, the majority devotes substantial effort to harmonizing the Commission’s decade-old Mass. Board approach to analyzing horizontal restraints with the relevant Supreme Court precedents. The dissenting Commissioner believes that the majority has overruled Mass. Board, and the other Commissioner writing separately believes that the majority has almost done so. To understand this issue, we must examine the majority’s analytical scheme.
The majority recognizes two approaches to analyzing horizontal restraints, a per se rule and a rule of reason. The majority’s per se rule is not the rigid classification rule of the pre-BMI era; it recognizes, as did Mass. Board, that a horizontal restraint that initially appears to fall in a per se category should be reviewed under the rule of reason if sufficient competitive benefits are shown.
And the majority’s rule of reason isn’t necessarily full-blown. When it comes to analyzing non-price restrictions on advertising, such as restrictions on quality claims, the majority applies the rule of reason with a “quick look.” “Quick look” is something of a misnomer here, because the majority spends 15 pages on the topic, but the term does capture the way the majority found a violation without proving a market in the way the full rule of reason would require. Antitrust law needs some kind of structured inquiry for at least some important classes of horizontal restraints in order to make the legal rules practically administrable; courts shouldn’t — and don’t — demand a merger-style analysis in every horizontal restraints case. In consequence, Chairman Pitofsky’s roadmap for quick look review under the rule of reason may be his opinion’s most influential contribution.
The majority’s rule of reason analysis has three steps, each applied quickly. First, the Commission found that the restraint was likely to produce anticompetitive effects. It relied primarily on testimony that consumers valued the advertising that was prohibited, and that the advertising attracted new patients to dentists. Second, the Commission found that the trade association which imposed the restraints on its members had market power. The majority did not prove market power by defining a market and examining entry and shares within it. Instead, and consistent with the Supreme Court’s Indiana Federation of Dentists approach, the majority inferred market power from evidence demonstrating the combination of actual anticompetitive effects and the trade association’s ability to enforce the restraints on its members. This may have been enough to prove market power, although the majority also noted evidence of the high market share held by trade association members, and evidence that it would be difficult for non-member dentists, who would be free to advertise, to undermine the advertising ban through entry or expansion. Third, the majority examines the proffered efficiency justifications, and finds, without much elaboration of the analytical framework, that the restraints are not limited to advancing those goals. This approach emphasizes that the question of a less restrictive alternative is not necessarily a separate step in the rule of reason analysis when the restriction appears “not even arguably tailored” to the justification, as the Court put it in NCAA. The efficiencies discussion, which turns on ways of preventing deceptive and unfair advertisements, is an interesting application of the Commission’s consumer protection expertise to illuminate an antitrust problem.
Commissioners Azcuenaga and Starek, in their separate opinions, express concern with the way the majority conceives of the per se rule and the rule of reason. They describe Mass. Board as emphasizing the analysis of efficiencies in determining whether to apply a per se rule, and suggest that the majority, despite the majority’s professed allegiance to BMI, has tilted back toward the rigid characterization approach of the past and devalued the importance of efficiencies. If they are right about what the majority is up to, the result could be, perhaps paradoxically, to increase the importance of market power in the analysis of horizontal restraints. After this decision, it may be tempting to take a “quick look” — with proof of market power as an element — at a broader range of restraints than before, including some that would previously have been considered illegal per se after an examination of efficiency justifications but not of harm to competition.
Commissioner Azcuenaga also employs a fuller review under the rule of reason than the majority finds necessary, including defining markets. After applying what might be called her “slower look,” she concludes that the restraints have not been shown to harm competition, in significant part because entry would solve any competitive problem. Neither she nor the majority, however, truly instruct us as to when a “quick look” is sufficient and when the rule of reason review must be full-blown; the majority merely notes that the restraint must be examined in the detail necessary to understand its competitive effect. Yet if appellate courts do not give the Commission and district courts deference to decide how detailed a review is required, the rule of reason may inevitably be pushed toward a full-blown inquiry. And if that is possible, cautious plaintiffs may conduct discovery and argument as thought the rule of reason review will be full-blown.
Let me turn now to some hot topics in consumer protection. I could spend the rest of my time discussing this area without exhausting the many important initiatives Jodie Bernstein and her Bureau have undertaken, including attacking fraudulent telemarketers, protecting commerce and privacy in cyberspace, challenging deceptive advertising, and repealing old and unnecessary Commission rules. Instead I'd like to concentrate on two issues in consumer protection that concern Commission economists.
The first issue emerges from the way we economists think about penalties for violations. The law-and-economics literature instructs us that if we want to get the right amount of general deterrence in the economy — to deter harmful conduct by inducing an efficient level of compliance expenditures by firms — and if we do it only through levying fines, we should set the penalty by reference to the injury the lawbreakers have caused third parties. The right penalty, conceptually, is the injury to victims inflated by a multiple. The multiple is highest for violations most difficult for us to detect and convict, and lowest, near one, for violations easily detected and convicted. The multiple is chosen so that potential lawbreakers would expect to pay a penalty equal to victim injury, after taking into account the probability of getting away without conviction.
This approach is usually more of a conceptual benchmark than a method of setting penalties for a host of reasons. For example: we often do not have the authority to assess penalties, although we can require consumer redress; we have the ability to substitute fencing-in relief for monetary assessments, but it is often difficult to place a dollar value on the deterrent effect of such provisions; many defendants would lack the financial resources to pay the penalty this calculation would require; and it may be difficult to estimate the dollar value of injury to victims in some kinds of consumer protection cases. But despite these problems, computing the penalty benchmark can be instructive, and my staff routinely attempts to do so.
I think I am learning one lesson from these benchmarks: I am coming to the tentative conclusion that we have not always been tough enough in fraud cases when the lawbreakers are not bankrupt. You can expect me to point out such cases to the Commission, and recommend that the Commission litigate rather than settle unless it receives higher amounts of consumer redress, or tougher fencing-in provisions (possibly including bans on continuing in related businesses or bonds to provide the means for paying consumer redress in the event violations continue), than might have been suggested by past practice.
The conceptual benchmark of the economist’s penalty model is also useful in thinking about when to seek relief against agents as well as principals for deception — for example, when to go after the advertising agency in addition to the advertiser in a deceptive advertising case. The economist’s initial instinct is that it should not matter who in the agency relationship is assessed the penalty. The threat of penalty to either principal or agent can be expected to lead the contracting parties to arrange their affairs to deter lawbreaking by monitoring compliance by each other. But when one party in the chain can most cheaply detect and prevent deception, and its actions cannot cheaply be monitored by the other party, we can expect to deter deception at lowest social cost by targeting some relief, monetary or injunctive, specifically at that party. So in a deceptive advertising case, we should focus on the advertiser when it provides the ad agency with unsubstantiated information. But when the advertiser and the ad agency develop ads together with correct information, but the ads were misleading due to the way the message was conveyed, the ad agency might reasonably be named as well as the advertiser.
The second issue also involves deceptive advertising. We economists generally prefer to use copy tests to determine what claims consumers take from ads, rather than relying on expert opinion unaided by consumer survey evidence. And because we focus on estimating the extent of consumer injury — we need to know consumer injury in order to estimate the deterrent penalty benchmark I just described — we prefer copy tests that control for consumers' preexisting beliefs. Consumers may believe something to be untrue after viewing an ad, but if they believed the same thing before exposure to the advertisement, the ad did not alter purchasing behavior and cause consumer injury (unless it reinforced or exploited such beliefs). In addition, copy tests must control for prior beliefs to ensure that consumer responses to survey questions elicit ad meaning rather than those beliefs.
Role of the Bureau of Economics
Let me spend a moment on what I think of, with complete objectivity, as the hottest topic about the Federal Trade Commission: the role of the Bureau of Economics. When I first met with the Commission’s economic staff, nearly one year ago, I suggested something I still believe: that the Bureau of Economics ought to be as influential as any bureau in the Commission. After all, economic analysis has become the essence of antitrust, in the courts and the enforcement agencies, independent of the political party in power. Republican presidents appointed economically-oriented antitrust scholars like Judges Posner, Bork and Easterbrook to the circuit courts, and President Clinton has done the same by elevating Justice Breyer to the Supreme Court. On the consumer protection side of the house, for over a decade Congress and the last three Presidents, again of both political parties, have insisted that enforcement agencies test most if not all regulations with a careful cost-benefit analysis — again calling on the expertise of economists to shape agency decision-making.
I am pleased to report that the Bureau of Economics is well on the way toward achieving this position. I cannot count how many times Commissioners have complimented my staff on their analysis of cases, and told us that the economists’ views were influential in their decision-making. And when I surveyed the antitrust matters on which I made recommendations to the Commission, I noticed that when asked to choose between Bureau recommendations, the Commissioners have taken the economists' side a good fraction of the time. We've also seen that the better the legal and economic bureaus cooperate in conducting investigations, the more seriously our views are taken by the lawyers and the Commission. Bill Baer and I in particular have worked very hard over the past year to institutionalize cooperation between our bureaus.
The Bureau of Economics contributes to the success of the Commission's mission in many ways beyond participating in case investigations. Commission economists are in demand as expert witnesses. Our advocacy program remains active. During the past year, for example, the Bureau of Economics shared our competition expertise with both the Federal Energy Regulatory Commission and the California Public Utilities Commission on what is the most important deregulatory issue in the nation now that telecommunications legislation has been enacted: how to bring competition to electricity generation and distribution. Bureau of Economics staff, jointly with the staff of the Bureau of Consumer Protection, advised the Food and Drug Administration on two subjects involving advertisement and promotion of prescription drugs: how to deal with advertisements directed to consumers, and how to treat claims made to managed care organizations. And Commission’s economists worked with Commissioner Varney at the OECD to promote increased competition in international satellite markets through the restructuring of INTELSAT.
I am also pleased to report that the antitrust community has been supportive of our active research program. During the Commission's recent Hearings on Global and Innovation-Based Competition, witness after witness called on the Bureau of Economics to conduct studies. We have a large number of ongoing mission-related research projects. We are sharing our expertise in competition and consumer protection with our counterpart organizations in other nations, and two economists have spent the past year on Susan DeSanti's team organizing the Commission's hearings and working on the staff's upcoming report.
As I hope I have demonstrated, the Pitofsky Federal Trade Commission is an exciting place. There is plenty to do — even for the lawyers. And we're always learning something new.
Let me conclude my tour of hot topics with a description of a meeting with the parties on a merger case at which I learned something new: some new, and perhaps insightful, jargon. The merging firms’ lawyers had the unenviable task of explaining away a corporate counsel’s nightmare, the document so hot it singes and smokes. You won't believe this: the document specifically calculated how much prices would rise after the acquisition through the exercise of market power. And it was prepared expressly for the purpose of assessing the value of the deal. And the parties gave it to us.
Until recently, the standard ploy for outside counsel was to dismiss awkward documents like these as merely the product of “brain-damaged middle management.” This ploy is no longer credible. After waves of corporate downsizing, there are no middle management left, brain-damaged or otherwise. A new explanation is needed.
Outside counsel were up to the task. They looked us in the eye and confidently explained that the document at issue should be ignored because it was merely “2 am babbling,” a nice phrase which evokes the new corporate organizational structure with no more middle management, only chiefs and newly-hatched MBAs, all losing perspective after working 22-hour days.
Any firm that creates this kind of document can expect our full court press, and the Commission did not disappoint the merging parties here. But when we investigate, we are thorough. And here our investigation demonstrated that the document’s claims were fantasies. Concentration was not high, customers indicated they could readily turn to alternative suppliers if the merging firms attempted to raise price unilaterally, and coordinated competitive effects appeared unlikely because deviation would have been difficult to detect. In addition, the acquisition appeared likely to produce significant cost savings. Against this background, the Commission properly resisted the temptation to wave the hot document at trial, as much fun as that would have been.
The moral of the story is that we don’t yell “fire” every time we see smoke. The converse is true, and will wrap up my presentation: even though the FTC building is not ablaze, the pots are bubbling productively over the flames of our current “hot topics.”
(1)*The views expressed here are those of the author, and not necessarily of the Federal Trade Commission or any Commissioner.