This document is available in two formats: this web page (for browsing content) and PDF (comparable to original document formatting). To view the PDF you will need Acrobat Reader, which may be downloaded from the Adobe site. For an official signed copy, please contact the Antitrust Documents Group.

U.S. Department of Justice Seal DEPARTMENT OF JUSTICE




Antitrust Federalism : Enhancing Federal/State Cooperation



CHRISTINE A. VARNEY
Assistant Attorney General
Antitrust Division
U.S. Department of Justice





Remarks as Prepared for the
National Association of Attorneys General
Columbia Law School State Attorneys General Program



Columbia Law School
New York, NY



October 7, 2009


  1. Introduction

I am delighted to be here today and discuss the relationship between the Department of Justice's Antitrust Division and State Attorneys General. Before I begin, I would like to take a moment to convey the best wishes to all participants in this conference from Attorney General Eric Holder. As many of you know, Attorney General Holder has made clear from the outset of this administration that the Department of Justice is committed to working hand-in-glove with our partners in State Attorney General offices. He attended and spoke at the NAAG annual meeting in March of this year, and made a point of introducing himself to all 50 State Attorneys General during that conference. At his direction, the Department of Justice has reinstituted the Executive Working Group, which brings together representatives of State Attorney General offices, local District Attorney offices and Department of Justice officials to collaborate and coordinate their efforts on issues of joint concern. Over the last nine months, the Department has initiated several programs to work actively with State Attorney General offices on a whole host of issues 8211; from mortgage fraud, to information sharing, to financial crime. We in the Antitrust Division look forward to expanding and building upon Attorney General Holder's efforts in the area of antitrust enforcement.

I know from my experience as a Federal Trade Commissioner and as a private practitioner, State Attorneys General play a critical role in promoting competition in markets and protecting consumers. I look forward to working with the new NAAG Antitrust Committee head, Ohio's new Attorney General Richard Cordray. I am sure that together we can forge a strong working relationship on competition issues, with the hallmarks of transparency of process while ensuring consumer welfare.

In my talk today, I would like to discuss not only strengthening the Division's relationship with the State Attorneys General, but I would also like to focus on two discrete areas: competition in agriculture markets and resale price maintenance. While these two areas are somewhat unrelated, the Division needs the help of State Attorneys General to police agriculture markets, and I would like to offer the Division's help in thinking through the Supreme Court's recent ruling in Leegin on resale price maintenance ("RPM"), as States consider enforcement of their laws related to RPM. Both of these issues were raised at my confirmation hearing by a number of Senators, and I would like to engage with you in discussing both of them here today.

  1. The Importance of an Open and Cooperative Relationship Between Federal and State Antitrust Enforcement

An open and cooperative relationship between the federal and state layers of antitrust enforcement is important because they complement each other. In 1890, Senator Sherman himself explained that the Act that would bear his name was meant "to arm the Federal courts that they may co-operate with the State courts in checking, curbing, and controlling the most dangerous combinations that now threaten the business, property, and trade of the people of the United States."1 Since passage of the Sherman Act, the federal government has supported a distinct role for state antitrust enforcement by, for example, providing seed money to states for increased antitrust enforcement and allocating to State Attorneys General the unique power to enforce the Sherman Act with parens patriae treble damages actions on behalf of state residents.2

The Antitrust Division cooperates with states in active matters on a regular basis. I understand that in the last two years, the Division has been involved in 31 joint investigations with State Attorneys General from 46 states, DC and Puerto Rico. For example, 15 different states participated in the successful investigation and challenge of the Google and Yahoo! advertising agreement.3 During that same period, we filed six civil cases either jointly with states or in parallel actions after a joint investigation. Notably, those cases led to consent decrees or settlements in cases involving proposed combinations of companies involved in health care insurance, wireless telecommunications, solid waste disposal, and beef packing.4

When there is such coordination and transparency with the states on civil litigation, the parties benefit from efficient use of resources and expertise while minimizing the burden and complexity of duplicative investigations or litigation. The case for coordination is most compelling and manageable in matters involving local markets (like asphalt paving), which are known well by state officials, or where local governments are significant purchasers of the relevant product (like school milk contracts). As for bigger and more complex matters, the challenges in managing coordination grow more difficult, although they can, with effective leadership and processes, be managed. It is worth noting that the Antitrust Modernization Committee affirmed a valuable and complementary role for antitrust enforcement by states, specifically calling on federal and state enforcers "to improve the coordination of their investigational and enforcement efforts" and recommending that such cooperation be enhanced going forward.5

In addition to enforcement matters, the Division supports thoughtful state policymaking on competition policy issues. Consider, for example, that the Division has filed comments in or issued opinions to several states on certificate of need (or CON) programs in the health care field, explaining the potential for CON programs to impose costs through diminished competition that outweigh any purported advantages. In Michigan, Governor Granholm relied publicly on the Division's analysis to veto the legislature's proposal to issue a CON to one joint venture to open the only proton beam therapy facility in the state.6 Similarly, the Division wrote to at least nine states that were considering statewide cable franchising legislation and encouraged states to use franchising efforts to support legitimate functions like public safety and regulation of rights of way, but not to build unnecessary regulatory barriers to entry.

In the area of real estate multilists, the Division has actively worked with the states and with state regulatory bodies both on the litigation front and as part of our competition advocacy. On the litigation front, the Division sued and obtained relief against groups of brokers in Columbia and Hilton Head, SC who operate multiple listing services and created rules to exclude competition from innovative and internet brokers.7 On the competition advocacy front, the Division has provided its opinion to several states regarding proposals preventing real estate brokers from offering rebates or proposals mandating a required level of service by brokers (i.e., service requirements that prevent homebuyers from negotiating a more limited arrangement that might better suit their needs). In so doing, the Division highlighted that such proposals often serve to stifle competition, increase prices for homebuyers, and slow entry.

In short, when states confront important competition policy issues, we are interested in sharing our experience and expertise, working with State Attorneys General wherever possible. On our end, we will endeavor to work with you when such issues reach us. On your end, please keep us in mind as you confront and work through opportunities for competition advocacy.

Finally, the Antitrust Division actively supports and engages in training activities and learning opportunities with the states through joint workshops and conferences like this one. Over the last several months, we have worked closely with State Attorneys General to develop training programs on a variety of topics, particularly focused on the opportunity for abuses of federal stimulus funds arising out of the Recovery Act. Going forward, we will continue to look for opportunities to work together on areas of mutual interest. Next month, for example, the Department of Justice, the FTC, and a number of states will co-host a workshop for enforcers in Washington, D.C. addressing energy markets and the changing nature of the electric power industry.

To facilitate our work with the states, we recently added Mark Tobey from the Texas Attorney General's office as Special Counsel for State Relations and Agriculture. He is a dedicated liaison who will assist our interaction with the states. To that end, he is working with the various litigating sections within the Division to develop a set of best practices for making requests of or coordinating with State Attorney General offices. If there are any requests, suggestions, or questions you have for the Division, Mark is the interface to the Division and will respond quickly and effectively to them.

  1. Agriculture Markets

I would like to turn to a brief discussion of the Division's thinking about agriculture markets. We are cognizant that competition in agricultural markets is frequently local or regional in nature, meaning competition-related issues are of specific concern to State Attorneys General. Over the last generation or two, changes in technology and the marketplace have revolutionized agriculture markets, producing some substantial efficiencies as well as concerns about concentration. These factors have facilitated the emergence of large firms, which may present challenges for new firms trying to enter this market.

The Antitrust Division plans to look into the general state of competition in agriculture markets through a number of workshops hosted jointly with the USDA. We welcome and encourage your involvement in these workshops to be held in 2010 and, wherever possible, in partnership with interested State Attorneys General. This undertaking will touch on important questions that will include an evaluation of the state and nature of competition in a range of agricultural markets.

The workshops are designed to be broad in their focus and we are still developing our agenda for them. There are, however, at least three industry segments within agriculture that have captured our attention. First, we are aware that farmers are facing a number of issues of interest and concern, including concerns regarding the levels of concentration in the seed industry (particularly for corn and soybeans), the level of transparency of market transactions and the role of the exchanges, and an increasing degree of vertical integration and buyer power. Second, we recognize that the dairy market has experienced considerable consolidation over the past decade and there are questions about the state of competition in that market. The Division has considerable expertise in this area and we continue to look very carefully at developments in dairy markets. And finally, the Division has been heavily involved in reviewing proposed mergers in livestock markets involving pork and beef,8 working with the states, for example, to stop the JBS/National Beef merger. Going forward, we will maintain a watchful eye on the marketplace – and we need State Attorneys General helping us do this – so that we can use the workshops to increase our understanding of this sector and look for opportunities to address concerns on the points noted above as well as any other that arise.

  1. Resale Price Maintenance
    1. Leegin's Invitation to Develop an Approach to the Rule of Reason for RPM

I am aware that one of the most important legal challenges facing State Attorneys General is how to proceed in light of the Supreme Court's decision in Leegin,9 which overruled Dr. Miles10 and held that minimum resale price maintenance (RPM) agreements would be analyzed pursuant to the rule of reason. Prior to that decision, RPM practices were considered per se illegal under federal law and were the subject of many consumer protection actions brought by states. In the wake of Leegin, many states are reevaluating their legal oversight over RPM11 arrangements and considering whether state law may treat them as per se illegal. As for federal law, however, it is clear that Leegin calls for a rule of reason inquiry and leaves open what exact form of inquiry is appropriate.

The Leegin majority recognized the need for lower courts to "establish a litigation structure to ensure that the rule [of reason] operates to eliminate anticompetitive restraints from the market and to provide more guidance to businesses."12 Because it found that the use of RPM can either serve legitimate business purposes, on the one hand, or lead to creation or maintenance of market power, on the other hand, the Leegin majority invited lower courts to separate the wheat from the chaff and "devise rules . . . for offering proof, or even presumptions" "to make the rule of reason a fair and efficient way to prohibit anticompetitive restraints and to promote procompetitive ones."13

I recognize that Leegin leaves many questions unanswered. In light of the Court's invitation, however, I would like to provide some thoughts about how the courts might apply a structured rule of reason analysis for many RPM arrangements. To provide the "litigation structure" suggested by the Supreme Court, a lower court could require a plaintiff to make a preliminary showing of "the existence of the agreement and scope of its operation"14 as well as the presence of structural conditions under which RPM is likely to be anticompetitive. Such a showing might well be sufficient to establish a prima facie case that RPM is unlawful. Under such an approach, the burden would shift to the defendant to demonstrate either that its RPM policy is actually—not merely theoretically—procompetitive or that the plaintiff's characterizations of the marketplace were erroneous. A court adopting such an approach could impose a burden on a defendant that would vary with the strength of the showing made by the plaintiff. At a minimum, the defendant would have to establish that it adopted RPM to enhance its success in competing with rivals and that RPM was a reasonable method for accomplishing its procompetitive purposes.

    1. The Modern Rule of Reason

The use of a structured rule of reason is consistent with modern development of antitrust analysis under Section 1 of the Sherman Act, which has shifted from a binary choice between the per se rule and a full-blown rule of reason analysis to a far more focused inquiry. In cases like NCAA, and Indiana Federation of Dentists, the Supreme Court made clear that the rule of reason did not open the field widely to include any argument in favor of a challenged restraint, but permitted the Court to engage in a truncated review when, for example, the practice at issue was plainly anticompetitive and did not appear to have any "countervailing competitive virtue."15

    1. A New Structured Rule of Reason Approach for RPM

      1. The language of Leegin

Consideration of an appropriate standard for evaluating RPM practices begins with the language of the Leegin opinion, which provides substantial guidance with respect to developing a structured rule of reason for RPM. First, the Court stressed the importance of interbrand competition and described its protection as "the primary purpose of the antitrust laws."16 In the Court's view, interbrand competition will ordinarily place limits on the exploitation of intrabrand market power.17

Second, the Court identified four circumstances where the use of RPM might be anticompetitive: (1) when used by a manufacturer cartel to identify members that are cheating on a price-fixing agreement; (2) when used to organize a retailer cartel by coercing manufacturers to eliminate price cutting; (3) when used by a dominant retailer to protect it from retailers with "better distribution systems and lower cost structures," thereby forestalling innovation in distribution; and (4) when used by a manufacturer with market power to give retailers an incentive not to sell the products of smaller rivals or new entrants.18 These scenarios provide a foundation for identifying conditions under which the burden of justification should be imposed on defendants.

Third, Leegin identified five potential procompetitive effects of RPM, including: increasing interbrand competition, preventing free riding, promoting competition on customer service, permitting a cost effective alternative to service contracts, and "facilitating market entry for new firms and brands" by guaranteeing favorable margins to retailers.19 These rationales help to identify permissible justifications for RPM.

Based upon these underlying competitive forces, the Supreme Court identified three important factors in applying the rule of reason. First, "more careful scrutiny" is merited if RPM has widespread use by a number of manufacturers in an industry. Second, the source of the restraint is important because there is a greater likelihood that the restraint is anticompetitive if retailers coerced the manufacturer to adopt it. And third, the degree of market power enjoyed by the manufacturer or retailer is important.20 These factors are likely to and should play a central role in any proposed structured rule of reason.

      1. Applying the guidance from Leegin

A careful reading of Leegin suggests a structured application of the rule of reason tailored to the plaintiff's theory of how RPM is anticompetitive in the case at hand. There are four generally accepted theories: manufacturer collusion, manufacturer exclusion, retailer collusion, and retailer exclusion. I will address how a structured rule of reason could be applied to each of them in turn by discussing elements that a plaintiff could use to establish a prima facie showing to shift the burden to Defendants. In so doing, I am not suggesting that these scenarios are exhaustive or the only approach that a court could take.

        1. Manufacturer-driven RPM

Manufacturer-driven RPM, which was less concerning to the Court, can be anticompetitive in at least two scenarios. First, RPM might be used to facilitate manufacturer collusion by helping a cartel police their agreement because transparent retail prices would reflect wholesale price cuts inconsistent with the agreement.21 In this situation, a prima facie case could consist of three elements: (1) a majority of sales in the market are covered by RPM, (2) structural conditions are conducive to price coordination, because such coordination is unlikely in an unconcentrated market, and (3) RPM plausibly helps significantly to identify cheating, which would not be the case if wholesale prices are otherwise transparent. The first two requirements provide evidence that the market is structured such that collusion would be feasible because a few manufacturers control a majority of the market; and the third requirement would show that RPM is a plausible method for policing the cartel.

Manufacturers may also use RPM improperly as part of an exclusion strategy. A dominant incumbent manufacturer may use RPM to guarantee large margins to retailers and make them unwilling to carry the products of small incumbents or new entrants.22 In contrast, the use of RPM by small incumbents or new entrants, who would use RPM to guarantee favorable margins to retailers to gain retail exposure, would be procompetitive. Even the dissenters in Leegin acknowledged that RPM used in this way would be procompetitive.23 A prima facie case under this theory could consist of three elements: (1) the manufacturer has a dominant market position, (2) its RPM contracts cover a substantial portion of distribution outlets, and (3) RPM plausibly has significant foreclosure effect that impacted an actual rival. The first two requirements would provide evidence that use of RPM would have sufficient exclusionary effects, and the last requirement would show the alleged harm is not merely theoretical.

        1. Retailer-driven RPM

The greater concern with RPM for the Leegin Court and many economists is when RPM results from retailer coercion. All five potential procompetitive uses of RPM identified in Leegin involve benefits to manufacturers, not retailers.24 As the Leegin majority stated: "If there is evidence retailers were the impetus for a vertical price restraint, there is a greater likelihood that the restraint facilitates a retailer cartel or supports a dominant, inefficient retailer." Consequently, a plaintiff presenting substantial evidence that retailer coercion was responsible for RPM has gone a long way toward making a prima facie showing of anticompetitive effects.

There are two primary scenarios in which retailers could use RPM anticompetitively. Under an exclusion theory, a retailer with significant market power, or several retailers acting in concert, could coerce important manufacturers to institute RPM and thereby frustrate price competition from discount or internet retailers. Under a structured rule of reason approach, a plaintiff pursuing a retailer exclusion theory might well be able to shift the burden to defendants by showing three elements: (1) that the retailers involved had sufficient market power, (2) that coercion by retailers resulted in RPM covering much of the market, and (3) RPM plausibly has a significant exclusionary effect that impacted an actual rival. The presence of market power and coercion are essential to the retailer's power to exclude. The third requirement would show the alleged harm is not merely theoretical.

Retailers can also use RPM as a cartelization device. An agreement by retailers to fix prices can be implemented and policed by coercing sufficient manufactures to use RPM consistent with the retailer cartel agreement. A plaintiff proceeding on a retailer collusion theory might well be able to shift the burden to defendants by showing three elements: (1) that RPM is used pervasively (e.g., at least 50% of the sales in the market), (2) that RPM was instituted by retailer coercion (not merely persuasion), and (3) retailer collusion could not be thwarted by manufacturers. As the Leegin court noted, manufacturers are generally aligned with consumers in seeking to limit retail margins. There must be some explanation for why manufacturers could not thwart retailer collusion by moving to other retailers, integrating into retailing or sponsoring new dealers. Extensive reliance on well-established retailers carrying the products of many manufacturers should suffice for this third element.

  1. Conclusion

To conclude, let me put our views on RPM in context. By addressing this topic, we are not seeking to disrupt the traditional preeminent role of the FTC and the States in this area. Rather, our interest is in supporting a jurisprudential effort to aid the development of federal law under Leegin. On the jurisprudential front, the approach suggested herein takes up the Court's invitation to "make the rule of reason a fair and efficient way to prohibit anticompetitive restraints and to promote procompetitive ones."25

Finally, I should note that I am not ruling out the possibility that Leegin's dissenters were right in thinking the effort to develop a new analytical framework will not succeed or that evidence will show that the actual uses of RPM are almost always harmful.26 The Division looks forward to analyses of any data that becomes available as a result of RPM practices implemented in the wake of Leegin and appreciates that the states will serve as important laboratories for obtaining this data. With respect to the natural experiments in the years ahead, we urge courts, commentators, and enforcers to keep an open mind because, as has occurred both in the antitrust and other contexts, accumulated experience on the effects of RPM and the litigation of RPM cases will be instructive.27

At its best, antitrust enforcement is a collaborative enterprise and the states are a key part of that enterprise. Consequently, the Division is committed to effective communication and coordination with state agencies on agriculture, RPM and any other antitrust issues. As a result, we will continue to support efforts like this conference and the related activities of the NAAG. More generally, we welcome continued involvement either in the form of requests for assistance or reports of potential cases that may be appropriate for federal enforcement. Thank you for all of your important work and for participating in this conference.


FOOTNOTES

1. 21 Cong. Rec. 2457 (1890).

2. Stephen Calkins, Perspectives on State and Federal Antitrust Enforcement, Duke L. Rev. 2006, Vol. 53:673, p. 677 (citing Crime Control Act of 1976, Pub.L.No. 94-503, § 116, 90 Stat. 2407, 2415 (codified as amended at 42 U.S.C. §§ 3701-96c (1976)), and Hart-Scott-Rodino Antitrust Improvements Act of 1976, Pub. L. No. 94-435, 90 Stat. 1383 (codified as amendment in scattered sections of 15 U.S.C.)).

3. See Yahoo! Inc. and Google Inc. Abandon Their Advertising Agreement (Press Release dated November 5, 2008) (available at http://www.usdoj.gov/atr/public/press_releases/2008/239167.htm).

4. See Justice Department Reaches Settlement With the Arizona Hospital and Healthcare Association and Its Subsidiary (Press Release dated May 22, 2007) (available at http://www.usdoj.gov/opa/pr/2007/May/07_ at_379.html); Justice Department Requires Divestiture in UnitedHealth Group's Acquisition of Sierra Health Services (Press Release dated February 25, 2008) (available at http://www.usdoj.gov/opa/pr/2008/February/08_ at_140.html); Justice Department Requires Divestitures in Verizon's Acquisition of All Tel (Press Release dated October 30, 2008) (available at http://www.usdoj.gov/opa/pr/2008/October/08-at-970.html); Complaint and Final Judgment in U.S. v. Verizon Communications Inc., Case No. 1:08-cv-00993 (D.D.C. June 10, 2008 and April 24, 2009) (available at http://www.usdoj.gov/atr/cases/f233900/233928.pdf and http://www.usdoj.gov/atr/cases/f245300/245394.pdf); Justice Department Requires Divestitures in Republic's Acquisition of Allied Waste (Press Release dated December 3, 2008) (available at http://www.usdoj.gov/opa/pr/2008/December/08-at-1061.html); Department of Justice Statement on the Abandonment of the JBS/National Beef Transaction (Press Release dated February 20, 2009) (available at http://www.usdoj.gov/atr/public/press_releases/2009/242857.htm).

5. Antitrust Modernization Commission, Report and Recommendations (April 2007) 186-87.

6. "Proton Beam Rule Reversed" The Detroit News, June 20, 2008 (available at http://www.detnews.com/apps/pbcs.dll/article?AID=/20080620/BIZ/806200363).

7. See Justice Department Reaches Settlement With the Multiple Listing Service of Hilton Head Island Inc. (Press Release dated October 16, 2007) (available at http://www.usdoj.gov/atr/public/press_releases/2007/2268376.htm); Justice Department Reaches Settlement With Consolidated Multiple Listing Service Inc. (Press Release dated May 4, 2009) (available at http://www.usdoj.gov/atr/public/press_releases/2009/245505.htm).

8. See Statement of the Department of Justice Antitrust Division on its Decision to Close its Investigation of Smithfield Inc.'s Acquisition of Premium Standard Farms Inc. (Press Release dated May 4, 2007) (available at http://www.usdoj.gov/atr/public/press_releases/2007/223077.htm); Department of Justice Statement on the Abandonment of the JBS/National Beef Transaction (Press Release dated February 20, 2009) (available at http://www.usdoj.gov/atr/public/press_releases/2009/242857.htm).

9. Leegin Creative Leather Products v. PSKS, Inc., 551 U.S. 877 (2007).

10. Dr. Miles Medical Co. v. John D. Park & Sons Co., 220 U.S. 373 (1911).

11. Throughout this speech, "RPM" refers only to agreements setting minimum prices. Some manufacturers instead set the maximum price that retailers may charge, but maximum RPM agreements do not pose the same potential anticompetitive or consumer harms as minimum RPM and are not considered here. Maximum RPM agreements have been evaluated under the rule of reason since State Oil Co. v. Kahn, 522 U.S. 3 (1997).

12. Leegin, 551 U.S. at 898.

13. Id. at 898-99.

14. Id. at 898.

15. See Nat'l Soc'y of Prof'l Eng'rs v. United States, 435 U.S. 679 (1978); NCAA v. Bd. of Regents of the Univ. of Okla., 468 U.S. 85, 110, 113, 115-20 (1984); and FTC v. Ind. Fed'n of Dentists, 476 U.S. 447, 459-64 (1986).

16. Leegin, 551 U.S. at 890, 895 (citing State Oil Co. v. Khan, 522 U.S. 3, 15 (1997)).

17. Id. at 895, see also Business Elecs. Corp. v. Sharp Elecs. Corp., 485 U.S. 717, 725 (1988) ("so long as interbrand competition exist[s], that would provide a 'significant check' on any attempt to exploit intrabrand market power"); Continental T.V., Inc. v. GTE Sylvania Inc., 433 U.S. 36, 64-65 (1977) (White, J., concurring); Graphic Prods. Distribs. v. Itek Corp., 717 F.2d 1560, 1571-72 & n.20 (11th Cir. 1983).

18. Leegin, 551 U.S. at 892-94.

19. Id. at 889-92.

20. Id. at 897-98.

21. Of course, such horizontal agreement by manufacturers would be per se illegal and, if detected, most likely prosecuted criminally by the Department of Justice. Such agreements, however, can be difficult to detect and prove. A challenge to pervasive and suspicious use of RPM could frustrate the cartel even if it could not be directly attacked.

22. A. Douglas Melamed, Exclusive Dealing Agreements and Other Exclusionary Conduct—Are There Unifying Principles?, 73 ANTITRUST L.J. 375, 403-05 (2006). Although sharing of monopoly profits between a manufacturer and its retailers may not harm consumers directly, by creating distribution obstacles to entrants it may harm consumers over the long term.

23. Leegin, 551 U.S. at 917-18 (Breyer, J., dissenting).

24. Id. at 897-98 (majority opinion).

25. Id. at 898-99.

26. Id. at 916-17 (Breyer, J., dissenting).

27. The "per se rule is appropriate only after courts have had considerable experience with the type of restraint at issue" so "courts can predict with confidence that it would be invalidated in all or almost all instances under the rule of reason." Id. at 886-87 (majority opinion). Cf. Garcia v. San Antonio Metro. Transit Auth., 469 U.S. 528, 546 (1985) (overruling nine-year-old precedent as "unsound in principle and unworkable in practice"); Continental T.V., Inc. v. GTE Sylvania Inc., 433 U.S. 36, 47-49 (1977) (overruling United States v. Arnold, Schwinn & Co., 388 U.S. 365 (1967), after ten years because the "great weight of scholarly opinion has been critical of the decision, and a number of the federal courts confronted with analogous vertical restrictions have sought to limit its reach. In our view, the experience of the past 10 years should be brought to bear on this subject of considerable commercial importance").