The George Mason Law Review Antitrust Symposium: The Changing Face of Efficiency
It is now 18 months since the Department of Justice and Federal Trade Commission published Revised Guidelines for treatment of efficiency claims in defense of mergers.(2) My goal today is to present - from the point of view of a non-objective observer - a progress report on that revised section of the Guidelines.
When first published, I detected two sorts of reactions to the changes:
1. From many antitrust enforcers, it was said that it was difficult enough for a plaintiff to win a merger case now that the courts have backed away from strict structural presumptions; it will be far more difficult and perhaps impossible to win cases once a new and amorphous "efficiency defense" is introduced into merger analysis.
2. From many in the defense bar and many business consultants, it was said that the revised defense is so circumscribed with qualifications - that the efficiencies must be merger specific, clearly substantiated, not a result of reductions in output or service, likely to improve consumer welfare, and virtually never sufficient to overcome mergers to monopoly - that introduction of this "revised" defense will make no practical difference in the results reached in any specific case.
A year and one-half later, it is clear that the enforcement agencies can and will win merger cases in the face of efficiency claims. It is also clear in my view that the defense will make a difference, though I will concede that thus far that difference primarily occurs in connection with the exercise of prosecutorial discretion.
Let me recall the state of affairs before the section on efficiencies in the Horizontal Merger Guidelines was revised in April, 1997. The courts were steadily introducing or attaching greater weight to efficiency claims in many areas of antitrust - for example, as a device to avoid per se treatment in horizontal restraint cases, (3) joint venture analysis (4) and with respect to vertical distribution arrangements.(5) In merger review, however, efficiencies were relevant to the exercise of prosecutorial discretion but, according to authoritative Supreme Court doctrine, not relevant when a transaction was examined in court. (6) Nevertheless, lower courts were uncomfortable with the absolute preclusion of efficiency claims in merger cases and frequently were taking efficiencies into account,(7) though no case can be cited where an otherwise illegal merger was declared legal because of the presence of substantial efficiencies.
In a broad sense, what were the revisions to the Guidelines designed to accomplish?
- The most significant aspect of the 1997 revisions is that they tied efficiencies directly into competitive effects analysis. The revisions recognized that cost reductions may reduce the likelihood of coordinated interaction or the incentive to raise price unilaterally. In these and other market situations, efficiencies are likely to lead to benefits to consumers.
- The revisions refine the concept that efficiencies must be attributable to the merger and could not be achieved in a less anticompetitive way. Instead of requiring proof that claimed efficiencies could not be achieved through some hypotheticalalternatives such as unilateral expansion or competitor collaborations, the agencies committed to evaluate claimed efficiencies against other practical alternatives.
- Efficiency analysis now expressly incorporates a sliding scale approach. The revisions state that the agencies will require proof of greater efficiencies as the likely anticompetitive effects of the merger increase. Thus efficiencies should almost never justify a merger to monopoly or near monopoly.
- The revisions define more clearly and explicitly which efficiencies "count" - what Section 4 now defines as "cognizable efficiencies." In particular, efficiencies must not arise from anticompetitive reductions in output, service or other competitively significant categories, such as innovation.
To my eyes, the early returns are encouraging. Lawyers present their efficiency claims, within the FTC and in court, in a more organized and realistic way. People dealing with the FTC staff - including Commissioners - find that staff reactions to efficiency claims are more consistent. There are now standards that defense lawyers, enforcement officials and judges can and do turn to in arguing efficiency questions pro and con.
And the Commission does take efficiency issues into account. I don't want to oversell the role of claims of efficiencies in influencing prosecutorial discretion, but there have been instances where they played a role. In hospital mergers, it is often (though not always) the case that significant efficiencies can be achieved through a merger, and the Commission has relied on such efficiencies in recommending that some mergers not be challenged. In the Tosco-Unocal merger, involving a combination of refineries in California, the decision not to challenge similarly was influenced by a conclusion that there were real synergies to the deal. And in Chrysler - Daimler Benz, early presentation of evidence of efficiencies influenced the decision not even to issue a second request.
Let me now examine briefly how the revised Guidelines have functioned in four recent government merger cases litigated in the district courts.
1. Staples/Office Depot(8). You will recall that Staples was the second- largest office supply superstore chain in the United States with approximately $4 billion in annual sales and that it proposed to acquire Office Depot, the largest office supply superstore chain with a little over $6 billion in annual sales. The District Court concluded that office supply superstores constituted a separate relevant product market, found that the merging parties had exceptionally high market shares in those markets, and then turned to the respondents' claim of efficiencies. As in all four cases I will discuss, the court did not declare efficiency claims irrelevant under the Supreme Court'sProcter & Gamble decision, but instead carefully examined the efficiency claims that were advanced. In Staples, many efficiency claims were asserted, but the most important was that the combined firm would have augmented purchasing power and could extract better prices from its various vendors. The court took the claim seriously and examined it closely but rejected it on two grounds: first, the claims appeared exaggerated when compared to internal documents not prepared with litigation in mind;(9) and second, assuming increased buying power is an "efficiency," the merger was not necessary to increase buying power because both parties to the merger were expanding rapidly.
I interpret Staples as a decision in which the Judge rejected efficiency claims primarily because he found them grossly exaggerated when measured against internal documents. Even if these efficiencies had not been exaggerated, leading firms in a rapidly expanding market will have difficulty successfully asserting efficiency defenses such as improved purchasing power, increased advertising discounts, better geographic coverage, since the same result may be achieved fairly promptly through internal expansion.
2. Long Island Jewish Medical/North Shore Health(10). In this hospital merger case, the District Court directed judgment for the hospitals primarily because the court found that the government failed to prove its relevant product and geographic markets. It also examined carefully a wide range of claimed efficiencies that totaled $92 million in recurring annual savings and $78 million in capital avoidance. As in Staples, the Court concluded that claimed savings were exaggerated, but nevertheless concluded there probably would be substantial savings - approximately $25 to $30 million annual savings beginning in the fifth year following the merger. The most interesting feature of the case is the way in which the Court found that any savings would be likely to be passed on to consumers. Under the Revised Guidelines, it was anticipated that consumer benefits usually would result from the structure of the market. In the Long Island Hospital case, the Judge inferred consumer benefits from the fact that both hospitals were not-for-profit organizations, had a "genuine commitment to help their communities" and had entered into an agreement with the New York Attorney General that they would pass on to the community the substantial cost savings that would be achieved. These are rather unusual factors, not likely to be duplicated outside the health care field, and not the sort likely to be relied on to persuade enforcers or judges that savings are likely to be passed on to consumers.
3. Lucy Lee/Doctors Regional Medical Center(11). In another hospital merger case, the FTC challenged the merger of the only two hospitals in Poplar Bluff, Missouri and the largest hospitals in Butler County where Poplar Bluff is located. The Court accepted the Commission's relevant product and geographic market definitions which gave the two hospitals roughly an 84% percent market share, and then turned to claims of efficiency. Because the case is on appeal, I will limit my remarks to reporting the conclusions of the District Court.
The Court viewed the primary efficiencies claim of the parties as a contention that the two hospitals were under-utilized and inefficient because of relatively low in-patient occupancy rates, and that the merger would enable the combined hospital to eliminate unused beds and substantially cut costs. The Court rejected that claim on grounds that excess capacity could be reduced without a merger, that some of the claimed efficiencies were speculative, and that the merged hospital would be unlikely to pass savings on to consumers absent competitive pressure from each other to lower prices.
An interesting feature of the opinion concerns the question of whether the Court should regard as an efficiency the claim that the proposed merger would allow the combined hospitals to offer new services not previously available. As noted earlier, the relevant product market was agreed to by the parties as "general acute care in-patient hospital services, including primary and secondary services but not tertiary hospital services." The parties argued that the proposed merger would allow the combined hospitals to offer open heart surgery and other tertiary services. Relying on United States v.Philadelphia National Bank, 374 U.S. 321, 371 (1963), the Court rejected this efficiency claim , concluding that alleged procompetitive consequences in one market do not justify anti-competitive effects in a different market. In that respect, the Court was following the approach of the revised Merger Guidelines which, in footnote 36, explains that Section 7 of the Clayton Act prohibits mergers that may substantially lessen competition "in any line of commerce . . ." and, therefore, the Agencies normally will challenge a merger if it is likely to be anti-competitive in any relevant market. The Guidelines add that the agencies, as a matter of prosecutorial discretion, might take cross-market efficiencies into account, but the language clearly suggests that would occur only in very limited circumstances.(12)
Putting the Lucy Lee/DRMC case aside, I would like to offer some general comments about claims of efficiency that cross relevant markets - comments that generally echo ideas I advocated in articles written before I became a Commissioner in 1995.(13)
As the Court in Lucy Lee/DRMC indicated, the plain language of Section 7 of the Clayton Act, as interpreted in Philadelphia National Bank, arguably precludes any balancing of pro-competitive effects in one market against anticompetitve effects in another. Even if that statutory language is not controlling, I have come increasingly to the view that it is not practical in run-of-the-mill merger cases to trade off pro- and anti-competitive effects across markets. Imagine the measurement problems of an increase in market shares from 20 to 30% in market A compared to a 5% decrease in marginal cost (assuming marginal cost were somehow knowable) in market B. There is not presently a practical formula available to enforcement agencies or courts that would allow that kind of trade-off to be made.
Aside from measurement issues, there is also a fairness question involved. Assuming relevant market is correctly defined and market shares correctly measured, the consumers in the market rendered less competitive as a result of the merger will pay more; consumers in the market rendered more competitive as a result of efficiencies will pay less. But is it appropriate to deny to one group the guarantees of a competitive market in order to provide the benefits of efficiency to a separate group? Antitrust often involves trade-offs - for example, comparing benefits to inter-brand competition and losses in intra-brand competition in connection with vertical distribution arrangements.(14) But in those situations the same group of consumers wins and loses.
Of course, there may be situations in which the anticompetitive effects in one market are slight and the efficiencies in another market enormous, or the same consumers purchase in both markets (as pens and ink). The Revised Guidelines acknowledge that such situations could arise and reserve for prosecutorial discretion the possibility of taking the trade-off between efficiencies and anti-competitive effects into account. After 18 months of experience with purported cross-market trade-offs, I am more convinced than ever that this narrow approach is the right conclusion.
4. Cardinal/McKesson(15). A particularly interesting treatment of claims of efficiency in defense of a merger may be found in Judge Sporkin's recent opinion granting a preliminary injunction to the FTC in its challenge to proposed mergers between Cardinal Health and Bergen, and between McKesson and Amerisource - the four largest drug wholesalers in the United States. If both mergers had been allowed to proceed, the leading national drug wholesalers would have been reduced from four to two, and the combined market share of the two merged companies would range from 63% to 80% of the pharmaceutical wholesale market - depending on whether captive capacity was included in the relevant product market.
Judge Sporkin acknowledged that Supreme Court law probably would preclude giving any weight to claimed efficiencies (citing Procter & Gamble,(16) Brown Shoe,(17) and Philadelphia National Bank(18)). Nevertheless, the Court noted the tendency of lower court Judges to take efficiencies into account, including the Staples decision in its own district, and the fact that the Federal Trade Commission had moved away from the old Supreme Court cases in revising the Merger Guidelines.
The parties argued that the efficiencies would be substantial, ranging from $220 million in the three years after the merger to over $307 million, that at least 50% would be passed along to consumers, and that the efficiencies were a major if not principal reason for the mergers. The range of efficiencies cited included consolidation of distribution centers, increased discounts from vendors as a result of buying power, reductions in corporate overhead and interest, and a reduction in inventory.
The FTC staff did not challenge the assertion that there would be significant efficiencies, or that some of these savings would be passed on to consumers. Rather, the Commission staff argued that the claimed efficiencies were inflated (as usual these days, the efficiencies seemed to grow as the parties came closer to a trial date) and some of these claimed efficiencies were not merger specific - for example, closing redundant distribution centers - since that could have been accomplished without a merger. Finally, the Commission argued and the Court accepted that competition in the past had led the parties to pass along approximately 80% of any cost savings to their powerful customers. Thus, a lessening of competition, combined with a commitment to pass along 50% of savings, really amounted to a net consumer loss on the transactions.
What should we make of this opinion? These were four leading firms in a market merging to two, with combined market shares of up to 80%. My view is that this is an example of a merger to near monopoly and, except in the most extraordinary circumstances, efficiencies should not vindicate an otherwise illegal transaction. The Judge concluded that the defendants had not made their case:
"The critical question raised by the efficiencies defense is whether the projected savings from the mergers are enough to overcome the evidence that tends to show that possibly greater benefits can be acheived by the public through existing, continued competition. The defendants simply have not made their case on this point."(19)
I believe the decision is important and sound, not just for its treatment of efficiency claims but, more broadly, for all of antitrust analysis. In the past several decades, the enforcement agencies and the courts have abandoned reliance on structural presumptions - i.e. on market shares alone indicating competitive concerns - and have become more sensitive to other factors that diminish the likelihood that even high market shares will lead to anticompetitive effects. These include the possibility of new entry if post merger prices rise, difficulties in achieving post merger coordinated pricing, supply substitution in the event of price increases -- and, now, claims of efficiency.
But Judge Sporkin's opinion rightly reminds us that even when these additional considerations are present to some extent, we must not neglect or forget the market shares that represent the initial stage of analysis. When two firms, as a result of merger, will account for as much as 80% of a properly defined relevant market protected against entry, these 'other considerations,' including efficiency claims, should not usually reverse a finding of illegality. It's one thing to examine whether market share numbers really mean what they suggest (as directed by the Supreme Court in General Dynamics(20)); it's another to forget them in the process of further review.
It is of course premature to offer a definitive analysis of the role of efficiency claims in merger enforcement. It is clear however, that some of the extreme predictions - that enforcement agencies will find it next to impossible to defeat efficiency claims, or that the Guidelines' revised treatment of efficiencies will have no effect on merger enforcement - have not proven to be accurate. It is likely that any final verdict on this modification in antitrust analysis will be far more nuanced.
A few interim conclusions can be advanced:
1. Given the many limitations and qualifications on the successful assertion of an efficiency defense, it will be difficult to reverse what otherwise would be a finding of illegality. That is as it should be. The goal of the merger revisions was to open the door to efficiency claims as part of competitive effects analysis in close cases - not to give away the entire enforcement enterprise.
2. Grossly exaggerated efficiency claims do little good for those defending the legality of a merger. Indeed they may do considerable harm in making enforcement officials generally skeptical of proferred defenses and undermining credibility with judges.
3. With increased experience, and with the help of some direction from judges, defense counsel should become more sure and more comfortable in addressing efficiency claims. I believe that is the history of the increased attention to conditions of entry over the last several decades and is likely to be the result in this area as well.
1. The views expressed are my own, and do not necessarily reflect those of the Commission or other Commissioners.
2. Revised Section 4, Horizontal Merger Guidelines issued by the U.S. Department of Justice and the Federal Trade Commission, reprinted in 4 Trade Reg. Rep. (CCH) ¶ 13, 104, April 8, 1997.
3. Broadcast Music, Inc. v. CBS, 441 U.S. 1 (1979); Northwest Wholesale Stationers, Inc. v. Pacific Stationery & Printing Co., 472 U.S. 284 (1985).
4. National Bancard Corp. v. Visa U.S.A. Inc., 779 F.2d 592,599 (11th Cir.), cert. denied, 479 U.S. 923 (1986); Northrop Corp. v. McDonnell Douglas Corp., 705 F.2d 1030, 1050-53 (9th Cir.) cert. denied, 464 U.S. 849 (1993).
5. Continental T. V., Ind. v. GTE Sylvania Inc., 433 U.S. 36 (1977).
6. The most emphatic Supreme Court statement on the point is in FTC v. Procter & Gamble Co., 386 U.S. 568, 580 (1967), where the court said "possible economies cannot be used as a defense to illegality."
7. For example, see FTC v. University Health Inc. 938 F.2d 1206, 1222 (11th Cir. 1991), United States v. Country Lake Foods, Inc. 754 F.Supp. 669, 680 (D. Minn. 1990), United States v. Carilion Heath Sys., 707 F.Supp. 840, 849 (W.D.Va.), aff'd, 892 F.2d 1042, 1084-85 (4th Cir. 1989).
8. FTC v. Staples, Inc., 970 F. Supp. 1066 (D.D.C. 1997).
9. For example, the cost savings estimate submitted in Court exceeded by almost 500% the figures presented to the Boards of Directors of the two firms when they approved the transaction.
10. United States v. Long Island Jewish Medical Center, 983 F. Supp. 121 (E.D.N.Y. 1997).
11. FTC v. Tenet Healthcare Corp., 1998-2 Trade Cas. (CCH) ¶ 72,227 (E.D. Mo. 1998).
12. The footnote reads: "In some cases . . . The Agency in its prosecutorial discretion will consider efficiencies not strictly in the relevant market, but so inextricably linked with it that a partial divestiture or other remedy could not feasibly eliminate the anticompetitive effect in the relevant market without sacrificing the efficiencies in other markets. Inextricably linked efficiencies rarely are a significant factor in the Agency's determination not to challenge a merger. They are most likely to make a difference when they are great and the likely anticompetitive effect in the relevant market(s) is small."
13. Pitofsky, The Renaissance of Antitrust, Publication of the Association of the Bar of the City of New York (1990).
14. Continental T.V. Inc. v. GTE Sylvania, Inc., 433 US 36, 51-52 (1977).
15. FTC v. Cardinal Health, Inc., 1998-2 Trade Cas. (CCH) ¶ 72,226 (D.D.C. 1998).
16. 386 U. S. 568 (1967).
17. Brown Shoe Co. v. United States, 370 U.S. 294, 344 (1962).
18. 374 U.S. 321, 371 (1963).
19. Cardinal Health, 1998-2 Trade Cas. ¶ 72,226 at 82,450.
20. United States v. General Dynamics, 415 U. S. 486 (1974).