Buyer power issues are of increasing concern for competition law and policy. This includes buyer cartels. It is important to distinguish cartels from legitimate buying groups although the later can also cause competitive harms. Buyer cartels are not the mirror image of seller cartels because buyer power arises from smaller market shares, buyers have less incentive to defect from a buyer cartel, such cartels can include more participants, and they can employ tacit collusion more readily. Under most law such cartels are illegal, but the arguments for that conclusion are weaker than they should be because of a failure to consider the impact on the competitive process. Buyer cartels should have more attention and appropriate legal policies.

By Peter C. Carstensen1


Both buyer cartels and buying groups have received increasing notice in the last fifteen years with growing awareness of buyer power’s potential for harmful competitive effects.2 The most recent area of interest is labor markets where buyer power appears to be much more pervasive than conventional labor market theory had postulated.3

Buyer cartels pose serious risks to the market process and have no persuasive justification. Hence, unless regulated, they should be unlawful under all circumstances, despite the potential negative effects of seller power on buyers. But because buying groups can have utility, it is essential to distinguish between two types of horizontal agreements. At the same time, buying groups can result in cartelistic competitive harm if they have sufficient power in the buying market. Hence, there should be stricter oversight of such entities than currently exists. However, that is a topic for a different presentation.4

Buyer power can arise from much smaller market shares than is usually associated with seller power. Moreover, buyer cartels can include more participants and face lower risks of defection because of the strong incentives to comply. Consequently, such cartels can be more informal, or even tacit. Thus, buyer cartel issues require a fuller recognition of the incentives that motivate and facilitate such conduct, as well as the need to modify the conventional criteria to infer the existence of a cartel, considering these economic incentives.



Distinguishing a buyer cartel from a legitimate buying group is easy at a conceptual level. but in application, the two activities exist on a continuum ranging from a pure cartel to a buying group that integrates the purchase of all inputs.

A. Buyer Cartels

A cartel is a group of competitors who have agreed to limit or eliminate their competition in some economically relevant dimension. The objective of such a combination is to create, allocate, and exploit power in the market. While conventional monopsony theory posits primarily exploitive explanations, i.e. collusive, or unilateral conduct to reduce input prices, buyer cartels can also use their collective power to exclude or restrain their competitors in the markets in which they buy goods or otherwise regulate the nonprice dimensions of the supply market as a means of entrenching their own dominance in both upstream and downstream markets.5 A buyer cartel eliminates competition for input purchases. Buyers may collude even though they sell in highly competitive markets where seller cartelization is unlikely.6 Such buyer collusion may eventually reduce the total output in the market.7 A significant reduction in output can affect the downstream market by reducing the volume of the output produced, resulting in higher prices for the remaining production.8

Use of cartel power varies depending on the goals and interests of the participants, but also, importantly, on the nature of the supply market. If supply is relatively price inelastic, then colluding buyers have a strong incentive to drive down price because it will not significantly reduce the supply of the input. In other cases, supply maybe more price elastic and reducing price would result in an inadequate supply;9 but buyers can still use their collective power to compel their suppliers to discriminate in price or refuse to deal with new entrants or marginal buyers.10 Thus, supply elasticity will affect the methods used by a cartel but may not alter its goals.11

B. Legitimate Buying Groups

A buying group is a set of potentially competing buyers that pool their purchase orders and jointly or through an agent negotiate for the inputs they seek. The fundamental distinction between a legitimate buying group and a cartel is that a buying group acts to gain the efficiencies of a joint enterprise. The buyer-participants have integrated some of their input acquisition function by participating in the buying group. Efficiencies can result from longer production runs, reductions in transaction costs, or lower costs per unit of quality control, and can include protection against defective or dangerous products, preferred status with shipping services based on high volume, and improved ability to develop new products. In addition, buying cooperative can bargain for better prices. A buyer seeking a large volume is likely to make a more extensive search of the market for suppliers and generate a more active bidding process as a result. Therefore, in a market where competition is imperfect but workable, buyers can gain a price advantage by employing more sophisticated and effective searches for the inputs they need.

C. Distinguishing Buying Groups from Cartels

There are no fixed parameters for a buying group. A legitimate buying group involves some integration of activities, but it is limited to facilitating the functioning of the enterprise.12 One way to distinguish a buying group from a buyer cartel is to focus on its functional goals. If the participants make an investment in, and consolidate, coordinate, and administer some aspects of their buying activity, then they are prima facie a buying group. Conversely, if the group exists only to agree on how the parties will conduct their own purchases, it is prima facie a cartel.



There are four ways in which buyer cartels present different characteristics from conventional seller cartel models. The differences make buyer cartels more likely and harder to detect.

First, buyer power arises are much lower market shares than is generally assumed to be the case on the selling side of the market.13 Thus, the EU has found that a buyer taking as little as 20 percent of a grocery input has the capacity to distort competition.14 In the UK, power was found in groceries where the buyer took as little as 10 percent of the branded goods.15 Toys R Us had only about a 20 percent share of the toy retail market but was able to coerce its major suppliers into refusing to deal with third parties.16 Because the buyer decides whether to buy and from whom to buy, this creates significant leverage especially when suppliers face significant costs in seeking alternative outlets.

Second, because buyers share an interest in keeping input costs down, the risks of defection from a cartel agreement are low. A defector must bid up the price, process the input into a larger volume of goods that will expand output and so is likely to lower sales prices even as its costs increase. In contrast, in a sellers’ cartel defection can yield an immediate gain to the defector by increasing volume even if the price per unit is lower. Thus, buyer cartels are generally more durable and also require less explicit organization and monitoring.

Third, taken together, the lower market share necessary for power and the greater durability of buyer cartels leads to making such organizations more varied. One result is that they can be more inclusive. Hence even in markets with a substantial number of competing buyers, cartelistic agreements are possible because the mutual interest of the participants gives them a strong incentive to cooperate.17 But for the same reasons, it is also possible to have cartels that have limited participation from competitors, and which can focus on limited objectives. The no-poach agreements in employment are examples of this. The participants agree not to compete for each other’s employees but remain free to compete for new entrants and for employees of non-participants.18

Fourth, again, because of the nature of buyer power, tacit collusion is a significant risk in buyer markets. If buyers respect each other’s suppliers and do not compete head-to-head for their services, each has somewhat more freedom to impose burdens on its suppliers because those suppliers will have few, if any, alternative outlets for their products or services.

Taken together these four factors makes buyer cartels more likely but also harder to identify and more challenging to remedy. Indeed, where the restraints are once established, tacit understandings can maintain them more easily on the buying side than on the selling side.



A. Legal Treatment of Buyer Cartels

Competition laws generally prohibit cartel agreements among buyers. The oldest known cartel case involved a grain buying conspiracy in ancient Athens.19 In the United States it is unlawful, as well as a basis for treble damage liability, for buyers to agree (1) on what they individually will pay for goods or services, (2) that they will not bid against each other for particular items at an auction, or (3) that they will restrict wage or employment competition.20 Canada has similar decisions.21 The European Union (“EU”) has held such conduct to be absolutely illegal.22 EU state competition authorities have also been aggressive in challenging them.23] This is consistent with the express terms of Article 101(1) of the Treaty on the Functioning of the European Union.24 Indeed globally, there is general recognition that buyer cartels are anticompetitive and should be prohibited. This is more consistent with the goal of facilitating the competitive process than with the narrow focus on consumer welfare.25

B. Buyer Cartels and the Per Se Illegal Standard

Conventional analysis says that buyer cartels cause economic harm similar to that caused by seller cartels. But as with seller cartels, some scholars defend buyer cartels as enhancing efficiency in some circumstances.26 The credibility of these arguments depends in large part on the goals of competition policy against which those arguments are measured. Although the conventional explanation of why buyer cartels are undesirable has serious limitations, there remain substantial reasons to object to such cartels and the justifications probably overestimate the potential gains to the competitive process.

    1. Theories of Competitive Harm

The conventional economic welfare argument against buyer cartels rests on the assumptions that buyers purchase in discrete units and resell in a market in which they are the only sellers, and that producers face increasing unit costs as volume increases. Under these assumptions, when the price paid for inputs is reduced, output of that commodity declines. As a result, the static comparison of a world with and without a buyer cartel shows that the cartel causes a reduction in production and a consequent increase in prices to consumers. The conclusion, therefore, is that buyer cartels harm consumers as well as aggregate welfare.27

The force of this economic theory is contingent on the validity of its assumptions. The argument loses force if the cartel uses all-or-nothing contracts. Such contracts compel producers to deliver approximately the competitive output but at a lower price.28 Further, if the cartel members compete with many other producers who have different sources of inputs, they would have no incentive or capacity to raise the prices of their output.29 In such a situation, the cartel’s output will be lower than it would have been, but overall prices will not be noticeably affected because volume in the downstream market will not change appreciably. In either case, there would be no adverse effect on the price-output parameters on which consumer welfare is measured.

Thus, a focus on consumer welfare as the goal of competition law weakens the argument that buyer cartels are necessarily undesirable. But if the primary goal of competition law is the protection and advancement of economic competition as a process, two related policy arguments support a broader condemnation of buyer cartels. First, buyer cartels, like seller cartels, directly distort the market process through private agreement.30 By preempting the market, any cartel undermines the fundamental goal of ensuring the goods and services clear markets at competitively determined prices.

Second, a buyer cartel distorts incentives and causes a misallocation of economic rewards over time, undermining the core goals of a competitive market system. Competition policy should focus on the dynamic interest of providing incentives to invest and develop new, innovative solutions to problems. Buyer cartels, however, diminish the rewards to a producer below the level that a competitive market would have provided,31 which sends the wrong signal to investors and innovators. Indeed, the strategic responses to cartels, such as vertical integration, although rational, can, in dynamic terms, result in further distortions of the market process away from better structural options.32

    1. The Argument for Legalizing Certain Types of Buyer Cartels

The primary argument in favor of certain types of buyer cartels rests on a notion of countervailing power.33 Small buyers facing a monopoly or oligopoly-seller market are individually powerless to bargain for lower prices;34 they are therefore compelled to accept the monopoly or oligopoly price. However, if these buyers together can make a credible threat that they will withhold their purchases unless they receive lower prices, they might succeed. For example, at one point, independent pharmacies in the U.S. sought an antitrust exemption that would enable them to bargain collectively over prices and reimbursement rates with oligopolistic wholesalers.35 These pharmacies claimed to believe that such an arrangement would lead to lower prices and higher reimbursement rates for the prescription drugs they happened to resell.36

This argument was based on a competition policy goal of static consumer welfare. The theory is that the successful buyer cartel will induce the seller to reduce prices and increase output, moving the market toward competitive price and output levels.37 This assumes that the buyer cartel has sufficient power to do so and passes on its gains to consumers. But this outcome is contingent on the relative options of each side, as well as the bargaining skill and sophistication of the parties. Powerful sellers can disrupt the group by offering some participants secret discounts to defect. In the short run, the defector would gain a head start in competing with the other members of the group in processing the input and producing saleable output. Under these circumstances, cartels might prove ineffective.

The case for buyer cartels is plausible in the short run only if no other reasonable alternative exists. Monopoly profits at the producer level ought, over time, to induce other responses, such as creating or sponsoring an entity to produce the input, finding substitute inputs, or creating a legitimate buying group. Entry or innovation would reduce or avoid the need for the monopolized product. Therefore, unless there is a strong argument that the seller’s power is not likely to dissipate even in the long run, the justification for a buyer cartel is weak.38 However, if a buyer cartel is the least bad option for establishing equitable prices and services, then an appropriate public regulatory body should oversee the cartel process—the interests of private parties will not necessarily be congruent with the public interest.



Competition laws relating to buyer combinations need thoughtful reappraisal today.

A. Buyer Cartels More Likely Than Generally Realized

The general paucity of buyer cartel cases stands in stark contrast to data which show that input prices are vulnerable to exploitation.39 This suggests that there is substantial potential for conspiracies among competing buyers to affect prices. The recent empirical work on labor markets showing the scope of monopsony and oligopsony power has highlighted the capacity of employers in these markets unilaterally or collusively to exploit the labor input in their products and service.40 The resulting emergence of litigation focused on such exploitation through “no-poach” agreements underlines the scope of risk.

The experience with labor markets and the underlying data on price effects in input markets generally strongly suggest that more enforcement resources should be invested in examination of buyer side collusion. There are metrics that can provide clues such as unexplained increases margins between input and output prices either in an entire market or in regional input markets.

B. Buyer Cartels Should Remain Per Se Illegal

The weakness of the arguments for a general acceptance of buyer cartels when they provide countervailing power provides strong support for retaining a per se rule except when there is express authorization for a cartel. Any short-term gains are likely to be offset by longer term harms. Moreover, the development and implementation of appropriate criteria for any general basis to allow such cartels would be extremely difficult if not impossible. As William Howard Taft observed more than 100 years ago:

. . .[W]here the sole object. . . is merely to restrain competition . . . there . . . is no measure of what is necessary . . . except the vague and varying opinion of judges as to how much, on principles of political economy, men ought to be allowed to restrain competition.41

Taft warned that courts would “set sail on sea of doubt” if they undertook to decide with respect to agreements with “no other purpose and no other consideration on either side than the mutual restraint of the parties, how much restraint of competition is in the public interest, and how much is not.”42 Hence, a per se rule prohibiting unregulated buyer cartels remains the best public policy.

C. Buyer Cartels Require Buyer Side Criteria

The preceding analysis shows that there is need for criteria appropriate to the buyer side of the market. First, buyer power arises at relatively low market shares in comparison to standard analysis of seller side risks. Hence, a buyer cartel can incorporate a lower share of the overall market and still have the capacity to exploit suppliers or exclude some competitors from market access. This is most evident in the cases involving “no-poaching” agreements. The conspiracy among some, but not all the leading Silicon Valley technology firms is an example.43 The implication of this insight is that enforcers need to be more attentive to evidence of collusion among any set of buyers regardless of their overall position in the market.

Second, buyer cartels can include more participants and are less likely to require detailed policing or overt understanding. The incentives to collude in many circumstances where supply is price inelastic, combined with the disincentives to defect mean that there is less need for the kind of verification that is frequently looked for in seller cartels where the incentives to defect are greater. Hence, the relative fewness of potential participants as well as the absence of overt policing mechanisms are less likely to be evident on the buyer side. This makes detection harder. But in the absence of such indicia, if other elements suggest a cartel, this should nonetheless encourage further investigation.

Indeed, the criteria used to exclude “tacit” collusion from condemnation themselves merit reconsideration on the buying side of the market. Rather, the central issue should be remedy and not the absence of more formal collusion. If a remedy can restore workable competition to the supply side of the market, that ought to justify intervention given the inherent incentives to exploit buyer power.



Both buyer cartels and legitimate buying groups present serious challenges to maintaining a strong and viable competitive process. Despite some plausible arguments that buyer cartels provide countervailing market power to a concentrated sellers’ market, the demands of a workable competitive process preclude accepting this justification as a defense. Naked restraints of competition by buyers should be absolutely illegal unless subject to direct public regulation. Enforcement authorities also ought to look more broadly at the impact of parallel buying practices—especially in markets with relatively few buyers and many sellers.

1 Fred W. & Vi Miller Chair in Law Emeritus, University of Wisconsin Law School. This piece draws substantially on my earlier work on this topic. Peter C. Carstensen, Competition Policy and Buyer Power: A Global Issue, Ch. 7 (2017) (hereafter Competition Policy); and Buyer Cartels versus Buying Groups: Legal Distinctions, Competitive Realities, and Antitrust Policy, 1 William & Mary Bus. L. Rev. 1 (2010) and a chapter on buyer cartels to be included in Research Handbook on Cartels, Peter Whelan, Ed., forthcoming 2022.

2 In addition to my own work, see note 1, other notable examples include: Roger D. Blair, Jeffery L. Harrison, Monopsony in Law and Economics (2012) (hereafter 2012); Id., Monopsony: Antirust Law and Economics (1993) (hereafter 1993); Roger G. Noll, “Buyer Power” and Economic Policy, 72 Antitrust L. J. 589 (2005); John B. Kirkwood, Powerful Buyers and Merger Enforcement, 92 Boston U. L. Rev. 1485 (2012; C. Scott Hemphill, Nancy L. Ross, Mergers that Harm Sellers, 127 Yale L. Rev. 2078 (2018).

3 See, e.g. Alan Manning, Monopsony in Labor Markets, 74 ILR Rev. 3 (2021) (extensive citations to relevant work); Suresh Naidu, Eric A. Posner, E. Glen Weyl, Antitrust Remedies for Labor Market Power, 132 Harv. L. Rev. 537 (2018).

4 I have addressed these issues in the works cited in note 1, supra.

5 See, e.g. Klor’s, Inc. v. Broadway-Hale Stores, Inc., 359 U.S. 207, 212–13 (1959) (buyer induced or coerced suppliers to engage in a collective refusal to deal); Montague v. Lowry, 193 U.S. 38, 45 (1904) (buyer organization induced sellers to refuse to deal with nonmembers and to enforce resale price maintenance); Toys “R” Us, Inc. v. Fed. Trade Comm’n, 221 F.3d 928, 936–38 (7th Cir. 2000) (major toy buyer induced sellers of toys to agree to refuse to sell selected toys to competitors of the buyer).

6 A good example is the timber buyer’s cartel in Alaska that drove down the price of trees in the regions where those companies operated, but which had no effect on the overall price of the lumber produced from those trees because that lumber competed in a much broader geographic market. Reid Bros. Logging Co. v. Ketchikan Pulp Co., 699 F.2d 1292, 1303 (9th Cir. 1983).

7 The likelihood of this effect and its substantiality are functions of the elasticity of supply, see, Carstensen, Competition Policy, supra note 1, Chap. 3.

8 This effect is discussed in Roger D. Blair & Jeffrey L. Harrison (1993), supra note 2, 36–42. See also Richard Sexton & Mingxia Zhang, An Assessment of the Impact of Food Industry Market Power on U.S. Consumers, 17 Agribusiness 59 (2001); Chris Doyle & Martijn A. Han, Expropriating Monopoly Rents through Stable Buyer Groups (Amsterdam Ctr. for L. & Econ., Working Paper No. 2009-03); Martijn A. Han, How Buyer Groups Can Effectively Operate as Stable Cartels, 62 Aenorm 14 (2009).

9 If, however, buyers can obtain additional inputs from another, more competitive market, they may be able to exploit one set of suppliers while still filling their residual supply needs. They will then be somewhat less concerned about supplier reaction.

10 The collective volume of purchases gives a cartel significant leverage over any individual seller because that seller must engage in a costly and time-consuming search for other buyers. See Peter C. Carstensen, Buyer Power, Competition Policy, and Antitrust: The Competitive Effects of Discrimination among Suppliers, 53 Antitrust Bull. 271, 281, 284 (2008).

11 Doyle & Han, supra note 8, posit a buyer cartel that requires suppliers to raise their prices, thus causing some reduction in sales but generating significant profits for the sellers, which the sellers then rebate to the buyers in the form of slotting fees or other kickbacks. See Discon v. NYNEX, 525 U.S. 128 (1998) (supplier got contract with higher price by agreeing to pay parent company a large sum thus evading rate regulation); NicSand v. 3M, 507 F.3d 442 (6th Cir. 2007) (large firm drove small supplier from market by offering large payments to buyers who also raised prices once supply competition was eliminated).

12 See Paul W. Dobson, Exploiting Buyer Power: Lessons from the British Grocery Trade, 72 Antitrust L.J. 529, 543–44 (2005) (discussing integrated buying groups in the British grocery industry).

13 This analysis is elaborated in Carstensen, Competition Policy, supra note 1, at 65-75.

14 See European Union, Guidelines on the Applicability of Article 101 of the Treaty on the Functioning of the European Union to Horizontal Co-operation Agreements, 2011 O.J. (C 11) 1, ¶ 204; see also Case 1v/M.1221, Rewe/Meinl, C(1999) 228 final, at 99 (survey of sellers showed that on average a buyer taking twenty-two percent or more of a commodity had substantial power over the seller).

15 See Dobson, supra note 12.

16 See Toys R Us v. FTC, 221 F.3d 928 (7th Cir. 2000) (major toy buyer induced sellers of toys to agree to refuse to sell selected toys to competitors of the buyer).

17 Todd v. Exxon, 275 F.3d 191 (2d Cir. 2001) (upholding treble damages in class action complaint charging a conspiracy to stabilize wage rates for various classes of technical workers at major oil and gas companies);

18 See Final Judgment, United States v. Adobe Sys. Inc., No. 1:10-cv-01629 (D.D.C. Mar. 18, 2011), available at (computer hardware and software companies consented to decree prohibiting restraints in competition for employees).

19 See Lamabros E. Kotsiris, An Antitrust Case in Ancient Greek Law, 22 Int’l Law 451 (1988).

20 See, e.g. United States v. Crescent Amusement Co., 323 U.S. 173 (1944) (conspiracy to allocate motion pictures); Swift & Co. v. United States, 196 U.S. 375 (1905) (conspiracy to limit the price paid for cattle); Knevelbaard Dairies v. Kraft Foods, Inc., 232 F.3d 979 (9th Cir. 2000) (conspiracy and/or unilateral conduct to reduce price of milk); United States v. Romer, 148 F.3d 359 (4th Cir. 1998) (conspiracy to limit prices on real estate); Reid Bros. Logging Co. v. Ketchikan Pulp Co., 699 F.2d 1292 (9th Cir. 1983) (conspiracy to allocate timber). See John Asker, A Study of the Internal Organization of a Bidding Cartel, 100 Am. Econ. Rev. 724 (2010) (describing the operation of a postage stamp cartel). For cases dealing with employment issues, see, e.g. Todd v. Exxon Corp., 275 F.3d 191 (2d Cir. 2001) (upholding treble damages in class action complaint charging a conspiracy to stabilize wage rates for various classes of technical workers at major oil and gas companies);). See also, Final Judgment, United States v. Adobe Sys. Inc., No. 1:10-cv-01629 (D.D.C. Mar. 18, 2011), available at (computer hardware and software companies consented to decree prohibiting restraints in competition for employees).

21 See, e.g. R. v. Abitibi Power & Paper Co., 36 C.P.R. 1888 (Que. Q.B.) (seventeen pulp wood buyers unlawful conspired to avoid competition in buying); 321665 Alberta Ltd. v. Mobil Oil Can. Ltd., 2011 ABQB 292 (Can.) (unlawful conspiracy to boycott a supplier).

22 See, e.g. Case C-209/07, Competition Authority v. Beef Industry Dev. Soc’y Ltd., 2008 E.C.R. 1-08637; Case COMP/C.38.238/B.2, Raw Tobacco Spain (Commission, 2004).

23 See European Competition Network (ECN), ECN Activities in the Food Sector (2012), available at

24 “The following shall be prohibited . . . all agreements [that] directly or indirectly fix purchase . . . prices or any other trading conditions. . . .” Consolidated Version of the Treaty on the Functioning of the European Union art. 101, May 9, 2008, 2008 O.J. (C 115) 47.

25 Victoria Daskalova, Consumer Welfare in EU Competition Law: What Is It (Not) About? 20–22 (TILEC Discussion Paper 2015-011, 2015), available at (pointing out that despite a nominal goal of “consumer welfare,” EU competition law provided protections against exploitation and exclusion of producers even in the absence of an immediate harm to consumers).

26 See Alan Devlin, Questioning the Per Se Standard in Cases of Concerted Monopsony, 3 Hastings Bus. L.J. 223, 241–43 (2007) Christopher Leslie, Achieving Efficiency Through Collusion: A Market Failure Defense to Horizontal Price-Fixing, 81 Cal. L. Rev. 243 (1993); Roger D. Blair & D. Daniel Sokol, The Rule of Reason and the Goals of Antitrust: An Economic Approach, 78 Antitrust L.J. 471, 487–90 (2012).

27 See, e.g. Herbert Hovenkamp, Is Antitrust’s Consumer Welfare Principle Imperiled? 45 J. Corp. L. 101, 115-117 (2020).

28 Carstensen, Competition Policy, supra note 1, at 42-46, 80-81.

29 See, e.g. Reid Bros. Logging Co. v. Ketchikan Pulp Co., 699 F.2d 1292 (9th Cir. 1983) (sawmills had no capacity to increase downstream lumber prices but could collude on timber purchases); see also, Mandeville Island Farms, v. American Crystal Sugar, 334 U.S. 219 (1948) (sugar beet buyer cartel could not affect national or global prices for sugar, but it could limit prices paid for beets from the narrow geographic market in which they are competing for beets).

30 See, e.g. Swift& Co.v.United States, 196U.S.375 (1905) (conspiracy to limit prices paid for cattle); Todd v. Exxon, 275 F.3d 191 (2nd Cir. 2001) (conspiracy to limit compensation to employees).

31 See Blair & Harrison (1993), supra note 2, at 156–63.

32 See Omega Envtl. Inc. v. Gibarco, Inc., 127 F.3d 1157 (9th Cir. 1997) (upholding a vertical exclusive dealing scheme foreclosing an alternative distribution plan even though serious questions existed as to whether exclusivity served efficiency interests of consumers).

33 John Kenneth Galbraith, American Capitalism: The Concept of Countervailing Power 109–12 (1952); see also Blair & Sokol, supra note 26, at 487-490.

34 See generally Thomas Campbell, Bilateral Monopoly in Mergers, 74 Antitrust L.J. 521 (2007) (small sellers should be allowed to merge to monopoly). This article produced some strong disagreement, see, Jonathan B. Baker, Joseph Farrell & Carl Shapiro, Merger to Monopoly to Serve a Single Buyer: Comment, 75 Antitrust L.J. 637, 637–46 (2008) (contesting the claim that bilateral monopoly is economically more efficient than competitive markets on at least one side of the marketplace).

35 See Impact of Our Antitrust Laws on Community Pharmacies and Their Patients: Hearing Before the Task Force on Antitrust and Competition Policy of the H. Comm. on the Judiciary, ll0th Cong. (2007); and Id. at 82–83 (statement of David A. Balto, representing the pharmacies).

36 Id. at 11 (statement of Rep. Weiner).

37 See Devlin, supra note 26 at 241-43; Blair & Sokol, supra note 26, at 487-90.

38 In addition, a buyer cartel can result in coordination on the downstream side of the market if the buyers constitute a substantial part of the resale market. If that occurs, the buyer cartel would morph into a seller cartel whose function would be to raise price and reduce output to the customers of its participants.

39 C. Edward Fee & Shawn Thomas, Sources of Gains in Horizontal Mergers: Evidence from Customer, Supplier, and Rival Firms, 74 J. Fin. Econ. 423, 424–27 (2004) (major cost savings from lower input prices); Sugato Bhattacharyya & Amrita Nain, Horizontal Acquisitions and Buying Power: A Product Market Analysis, 99 J. Fin. Econ. 97 (2011) (id.). Other studies have found substantial losses to sellers resulting from buyer cartels. See John E. Kwoka, Jr., The Price Effects of Bidding Conspiracies: Evidence from Real Estate Auction “Knockouts,” 42 Antitrust Bull. 503, 503 (1997) (finding that a thirty-two percent price decrease resulted from bid-rigging in real estate auctions); Jon P. Nelson, Comparative Antitrust Damages in Bid-Rigging Cases: Some Findings from a Used Vehicle Auction, 38 Antitrust Bull. 369, 386 tbl.4, 392–94 (1993) (finding that a significant price decrease resulted from bid rigging in auctions for used police cars).

40 See references at note 2, supra.

41 United States v. Addyston Pipe Steel, 85 F. 271, 282-283 (6th Cir. 1898).

42 Id. 283-284.

43 See Final Judgment, United States v. Adobe Sys. Inc., No. 1:10-cv-01629 (D.D.C. Mar. 18, 2011), available at (computer hardware and software companies consented to decree prohibiting restraints in competition for employees).