Draft 3: January 27, 2003
Indirect Purchaser Suits and the Consumer Interest
John E. Lopatka* and William H. Page**
It is now widely accepted that antitrust is about promoting the consumer interest.
Courts in antitrust suits shape the categories of liability and define levels of scrutiny
based upon the likelihood that the challenged practices will injure consumers.1 The
Supreme Court has, for example, repeatedly emphasized that antitrust protects
“competition, not competitors.”2 Injury to consumers is almost always necessary (though
not sufficient) for injury to competition,3 while injury to competitors is relevant only to
the extent it suggests or is instrumental in causing harm to consumers.4
Alumni Professor of Law, University of South Carolina School of Law, firstname.lastname@example.org.
Marshall M. Criser Eminent Scholar, University of Florida, Levin College of Law,
The authors would like to thank Jonathan Baker, Ron Davis, and Tina Miller for helpful comments
on earlier drafts.
See John E. Lopatka & William H. Page, Monopolization, Innovation, and Consumer Welfare,
69 GEO. WASH. L. REV. 367, 373-76 (2001).
Coined paradoxically in Brown Shoe Co. v. United States, 370 U.S. 294, 320 (1962), one of the
more anti-consumer decisions of the Warren Court, the phrase was give new meaning in Brunswick Corp.
v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477, 488 (1977).
Injury to sellers, as opposed to consumers, through an exercise of monopsony power by
purchasers, though rare, can also constitute an injury to competition. See, e.g., Mandeville Island Farms,
Inc. v. Am. Crystal Sugar Co., 334 U.S. 219, 236 (1948); Todd v. Exxon Corp., 275 F.3d 191, 201 (2d Cir.
Given the primacy of the consumer interest in antitrust, it is hardly surprising that
consumers are not categorically excluded from suing in antitrust cases. In Reiter v.
Sonotone,5 the Supreme Court refused to deny consumers standing in part because
“Congress designed the Sherman Act as a ‘consumer welfare prescription.’”6 But the
Court has created a daunting legal obstacle for consumers seeking treble damages for
federal antitrust violations. By its decision in Hanover Shoe,7 the Court gave direct
purchasers from monopolizers or price fixers the right to sue for damages measured by
the entire overcharge they paid as a result of the illegal conduct, even if they passed on
2001). Further, consumers can be injured by conduct that does not violate the antitrust laws. See, e.g.,
NYNEX Corp. v. Discon, Inc., 525 U.S. 128, 136-37 (refusing to hold illegal per se under the antitrust laws
conduct by a monopolist that injures consumers through a regulatory fraud).
Granted, the “consumer interest” is subject to interpretation. For Chicago School analysts it
usually means overall social wealth, or “economic efficiency.” See, e.g., RICHARD A. POSNER, ANTITRUST
LAW 9-27 (2d ed. 2001); ROBERT H. BORK, THE ANTITRUST PARADOX 427 (rev. ed. 1993). Viewed from
this perspective, a practice that expands output and increases social wealth does not give rise to antitrust
concern simply because it transfers wealth from consumers to producers. Others suggest that wealth
transfers from consumers are the primary issue. See, e.g., Robert H. Lande, Wealth Transfers as the
Original and Primary Concern of Antitrust: The Efficiency Interpretation Challenged, 34 HASTINGS L.J.
67 (1982). But this theoretical dispute only rarely has practical significance—monopolistic wealth
transfers normally entail reductions in social wealth, because a monopolistic reduction in output causes
both. The important point is that there is a consensus that the consumer interest is the focus of antitrust.
Reiter v. Sonotone Corp., 442 U.S. 330 (1979).
Id. at 343 (quoting ROBERT BORK, THE ANTITRUST PARADOX 66 (1978)).
Hanover Shoe v. United Shoe Mach. Co., 392 U.S. 481 (1968).
some of the overcharge to their own purchasers by raising prices; and in Illinois Brick,8
the Court made the direct purchasers’ right exclusive, denying indirect purchasers the
right to sue for an overcharge at all.9 The Illinois Brick rule is not aimed at consumers
per se; it allows suits by consumers if they are direct purchasers10 and cuts off claims by
businesses if they are indirect purchasers.11 But in practice the Illinois Brick rule
forecloses many consumers’ ability to sue, because firms that participate in significant
national and international price fixing conspiracies or that monopolize markets of
comparable scope do not often or primarily sell directly to consumers.
So we have an anomaly: antitrust is supposed to protect consumers, and
consumers have the capacity to sue for their injuries, yet Illinois Brick denies most of
them the right to sue for overcharges they suffer as a result of common federal antitrust
violations. This anomaly has provoked a counterattack against the Illinois Brick rule.
Critics point out that consumers suffer much of the harm from most price fixing
conspiracies, because direct and intermediate purchasers pass on a significant part of the
overcharge they pay to price fixers. Yet consumers recover nothing, and the direct
Illinois Brick Co. v. Illinois, 431 U.S. 720 (1977).
See, e.g., Kingray, Inc. v. NBA, 188 F. Supp.2d 1177, 1198-1200 (S.D. Cal. 2002) (concluding
that consumers who purchased NBA League Pass from satellite TV providers were barred from suing for
alleged overcharge from a conspiracy among NBA teams and others).
See, e.g., Goldwasser v. Ameritech Corp., 222 F.3d 390, 398 (7th Cir. 2000) (“As direct
purchasers, [plaintiff consumers] have no Illinois Brick problem.”).
See, e.g., Merican, Inc. v. Caterpillar Tractor Co., 713 F.2d 958, 849-50 (3d Cir.1983)
(concluding that non-factory-authorized dealers, who bought product from authorized dealers, lacked
standing even if they were “direct targets” of the conspiracy).
purchasers recover a windfall—treble damages for an “overcharge” that may not, after
netting out what they passed on, have harmed them at all. To rectify this disparity in
treatment, critics argue that Illinois Brick should be repealed or bypassed.12 Efforts to
repeal Illinois Brick at the federal level have thus far failed.13 But many (though not
all)14 states have authorized indirect purchaser suits under their own laws either by
enactment of new legislation or by judicial interpretation of existing legislation.15 In
most of these states, the arguments in favor of allowing indirect purchaser suits have
focused on the consumer interest, particularly the need to compensate consumers.16
Advocates of repeal also suggest that allowing indirect purchaser suits will benefit
consumers by providing greater deterrence. The Supreme Court in ARC America17 held
See, e.g., Kevin J. O'Connor, Is the Illinois Brick Wall Crumbling, 15 ANTITRUST 34 (Summer
2001); Robert G. Harris & Lawrence A. Sullivan, Passing on the Monopoly Overcharge: A Comprehensive
Policy Analysis, 128 U. PA. L. REV. 269 (1979).
See, e.g., George J. Benston, Indirect Purchasers’ Standing to Claim Damages in Price Fixing
Antitrust Actions: A Benefit/Cost Analysis of Proposals to Change the Illinois Brick Rule, 55 ANTITRUST
L.J. 213, 214 nn. 5-11 (1986).
See, e.g., Vacco v. Microsoft Corp., 793 A.2d 1048, 1059-61 (Conn. 2002) (refusing standing to
indirect purchasers under state law in part because state legislature rejected proposed Illinois Brick repealer
See, e.g., O’Connor, supra note xx, at 35 (counting 36 states that allow some form of indirect
See, e.g., John Duchemin, Businesses Fear Lawsuit Bill, HONOLULU ADVERTISER, May 26,
2002, at G1 (describing Hawaii attorney general’s office advocacy of Illinois Brick repeal bill on the
ground that it “expands consumer protection”).
California v. ARC America Corp., 490 U.S. 93 (1989).
that this kind of state statute does not conflict with the federal antitrust laws, and so may
coexist with them.
The state authorization of indirect purchaser suits has triggered an explosion of
indirect purchaser class actions under state law. In most instances, these suits allege the
same offenses that direct purchasers allege in federal court under the federal antitrust
laws, and many, though not all, follow on the heels of successful enforcement actions by
the federal agencies.18 Because of jurisdictional limitations, the myriad private actions
cannot, at least under present law, be consolidated for discovery or adjudication in a
single state or federal court.19 This embarassing state of affairs fulfills the prophecy of
the Court in Illinois Brick that allowing indirect purchaser suits would lead to “massive
multiparty litigation involving many levels of distribution and including large classes of
ultimate consumers remote from the Defendants.”20
We argue here that experience in indirect purchaser litigation under state law over
the past decade contradicts the claim that indirect purchaser actions benefit consumers.
Contrary to their stated intent, these lawsuits provide little compensation to consumers
and actually undermine the goal of optimal deterrence, which should guide any rational
remedial system. In Part II, we summarize the reasoning underlying the Illinois Brick
On the relationship between public antitrust enforcement and class actions, see Stephen Calkins,
An Enforcement Official's Reflections on Antitrust Class Actions, 39 ARIZ. L. REV. 413 (1997).
See generally Andrew I. Gavil, Federal Judicial Power and the Challenges of
Multijurisdictional Direct and Indirect Purchaser Antitrust Litigation, 69 GEO. WASH. L. REV. 860 (2001).
Illinois Brick, 431 U.S. at 740. The court was primarily concerned with the effect of such
actions on the incentives of direct purchasers to sue, but the problems such actions pose are also relevant to
the certification issue.
rule and describe the emergence of state indirect purchaser suits, focusing on the role of
the consumer protection argument in justifying this kind of action on grounds of
compensation and deterrence. We examine in Part III the real consequences of indirect
purchaser litigation for consumer classes. It turns out that, in two key procedural
contexts, courts have repeatedly found indirect purchaser class actions wanting as
mechanisms for compensating consumers. First, courts have frequently refused to certify
indirect purchaser classes, particularly putative classes consisting of consumers, because
individual issues predominate over class issues.21 Second, courts that have certified these
kinds of classes have found, after the cases have settled, that they cannot in fact provide
compensation to consumers. The remedy provisions instead resort to costly cy pres
measures that pay damage awards to groups who did not pay any overcharge. In the end,
indirect purchaser litigation provides precious little compensation to consumers.
We argue in part IV that indirect purchaser suits make no sense in a system
geared toward optimal deterrence. The leading proponents of state indirect purchaser
actions do not generally justify their position on the ground that allowing multiple
recoveries of the same overcharge by direct and indirect purchasers is desirable. Rather,
supporters assert that multiple recoveries, although theoretically possible, do not occur in
practice. Thus, most would agree that indirect purchaser suits are good policy only if
they result in the imposition of a better approximation of the optimal deterrent penalty
than a pure Illinois Brick regime. In theory, if the expected penalty for antitrust
violations provided by criminal sanctions and civil liability to direct purchasers were
See generally William H. Page, The Limits of State Indirect Purchaser Suits: Class
Certification in the Shadow of Illinois Brick, 67 ANTITRUST L.J. 1 (1999).
predictably inadequate, a second-best argument might be made for indirect purchaser
suits. But we argue in the final part that indirect purchaser litigation fails on this score as
well. We also offer some rudiments of an alternative remedial scheme that would work
toward creating optimal deterrence while minimizing the social cost of litigation.
II. Illinois Brick and the Rise of Indirect Purchaser Class Actions
Hanover Shoe held that antitrust defendants cannot reduce their liability for
monopolistic overcharges by proving that the plaintiffs, who purchased directly from
them, had passed on the overcharge by raising their own prices. The Court reasoned that
it would be impractical to determine what portion of any price increase by the direct
purchaser is attributable to the overcharge, and how that price increase affected the direct
Even if it could be shown that the buyer raised his price in response to,
and in the amount of, the overcharge and that his margin of profit and total
sales had not thereafter declined, there would remain the nearly
insuperable difficulty of demonstrating that the particular plaintiff could
not or would not have raised his prices absent the overcharge or
maintained the higher price had the overcharge been discontinued.22
Hanover Shoe, Inc. v. United Shoe Mach. Co., 392 U.S. 481, 492-93 (1968). See also In re
Brand Name Prescription Drugs Antitrust Litig., 123 F.3d 599, 605 (7th Cir. 1997) (Posner, J.) (“Tracing a
price hike through successive resales is an example of what is called ‘incidence analysis,’ and is famously
The Court relied on similar logic in Illinois Brick23 to deny indirect purchasers the right
to sue for the passed-on overcharge. The problems of estimating relative elasticities of
demand and supply make any effort at effective compensation of indirect purchasers
dubious at best.24 Equally important, given the problems of proof, litigation over
dividing the overcharge among multiple classes would only hinder antitrust enforcement.
The amount of the overcharge is the measure of the deterrent penalty; litigation over how
much was passed on to whom is costly and simply multiplies the number of stakeholders
in a share of the “common fund.”25 This ancillary litigation needlessly increases the costs
of antitrust enforcement. Moreover, because of collective action problems, dividing the
right to sue can undermine incentives to bring suit.26 Problems of proof also significantly
increase the risk of duplicative recovery, should all eligible plaintiffs sue.27 Thus, the
antitrust laws “will be more effectively enforced by concentrating the full recovery for
the overcharge in the direct purchasers rather than by allowing every plaintiff potentially
431 U.S. 720 (1977).
Benston, supra note xx, at 222-23; Illinois Brick, 431 U.S. at 741-42. But cf. Gregory J.
Werden, Demand Elasticities in Antitrust Analysis, 66 ANTITRUST L.J. 363, 363 (1998) (“Demand
elasticities are actually being estimated, and the estimated demand elasticities are being used to delineate
markets, to measure market power directly, and to predict the competitive effects of mergers.”).
Illinois Brick, 431 U.S. at 740-41.
Id. at 745.
Id. at 730-31.
affected by the overcharge to sue only for the amount it could show was absorbed by
The Illinois Brick rule has provoked controversy from its inception. The federal
courts have resolutely preserved the barrier and recognized few exceptions.29 Chicago
School analysts have endorsed the rule as the most effective means of affording optimal
deterrence, the primary economic goal of the treble damage remedy.30 But many other
Id. at 734. The court recognized possible exceptions when the direct sale is under a pre-existing
cost-plus contract or when the direct purchaser is under the seller’s “control.” Id. at 736 & n.16. See
generally Herbert Hovenkamp, The Indirect-Purchaser Rule and Cost-Plus Sales, 103 HARV. L. REV. 1717
The Supreme Court has rebuffed efforts to create additional exceptions to the rule. Kansas v.
Utilicorp United, Inc., 497 U.S. 199 (1990) (rejecting an exception where a regulatory agency permits a
direct purchaser to increase rates by the amount of the overcharge). And most lower courts interpret
existing exceptions narrowly. See, e.g., Dickson v. Microsoft Corp., 309 F.3d 193, 215 (4th Cir. 2002)
(withholding judgment on whether to recognize a “co-conspirator” exception and opining that such an
exception would be narrowly defined in any event); Labrador, Inc. v. Iams Co., 105 F.3d 665 (9th Cir.
1997) (construing the “ownership or control” exception narrowly); McCarthy v. Recordex Serv., Inc., 80
F.3d 842, 855 (3d Cir. 1996) (interpreting “co-conspirator” and “cost-plus” exceptions narrowly). Cf. In re
Brand Name Prescription Drugs Antitrust Litig., 123 F.3d 599, 605-06, 613-14 (7th Cir. 1997) (Posner, J.)
(reversing district court’s finding that that drug wholesalers were within “control” exception to Illinois
Brick, but also reversing summary judgment as to wholesalers on the ground that they may have been co-
conspirators of the manufacturers).
William M. Landes & Richard A. Posner, Should Indirect Purchasers Have Standing to Sue
Under the Antitrust Laws? An Economic Analysis of the Rule of Illinois Brick, 46 U. CHI. L. REV. 602
(1979) [hereinafter Landes & Posner, Indirect Purchasers]; William M. Landes & Richard A. Posner, The
Economics of Passing On: A Reply to Harris and Sullivan, 128 U. PA. L. REV. 1274 (1980).
analysts have disagreed.31 More important, many states and the District of Columbia
have “repealed” Illinois Brick by legislatively authorizing indirect purchaser suits under
their own antitrust laws.32 In addition, some states’ courts have interpreted their pre-
Illinois Brick antitrust and consumer protection statutes to authorize indirect purchaser
actions.33 The Supreme Court legitimized these local repeals of Illinois Brick in
California v. ARC America Corp,34 holding that they are not preempted by federal law,
despite their evident disagreement with the Court’s own reasoning.
States are free, of course, to adopt any rational goal for their antitrust laws, even if
it differs from those of the federal antitrust laws. So states could conceivably choose a
goal of imposing quadruple (or greater) damages on antitrust violators. Interestingly,
however, the state attacks on Illinois Brick do not purport to challenge the consumer-
oriented goals of the Sherman Act. On the contrary, the state rejections of Illinois Brick
See, e.g., Robert G. Harris & Lawrence A. Sullivan, Passing on the Monopoly Overcharge: A
Comprehensive Policy Analysis, 128 U. PA. L. REV. 269 (1979).
See Page, supra note xx, at 2 n.8. See also Thomas Greene, et al., State Antitrust Law and
Enforcement, 1252 PLI/Corp 1129 (2001).
See, e.g., Elkins v. Microsoft Corp., No. 2001-431, 2002 WL 31433092 (Vt. Nov. 1, 2002);
Comes v. Microsoft Corp., 646 N.W.2d 440, 450 (Iowa 2002). See also cases collected at Page, supra note
xx, at 2 n.9. But see. Pomerantz v. Microsoft Corp., 50 P.3d 929, 932 (Colo. App. 2002); Berghausen v.
Microsoft Corp., 765 N.E.2d 592, 596 (Ind. App. Ct. 2002).
490 U.S. 93 (1989). For analysis, see Report of the ABA Section of Antitrust Law Task Force to
Review the Supreme Court’s Decision in California v. ARC America Corp., 59 ANTITRUST L.J. 273 (1990).
explicitly rest on a consumer protection rationale.35 As an Arizona appellate court
Given the consumer protection purpose of Arizona's constitutional
antitrust provision and the strong statements against anti-competitive
conduct that it contains, any Arizona statutes adopted pursuant to this
provision must be liberally construed to carry out the purpose of
protecting consumers from price-fixing agreements. Interpreting [the
Arizona antitrust statute] in this light impels the conclusion that consumers
are best protected when indirect purchasers are permitted to maintain
antitrust actions against members of alleged price-fixing conspiracies.36
In this widely held view, Illinois Brick denies many consumers a remedy and gives direct
purchasers a windfall: “The indirect purchaser rule awards greatly overcompensate
intermediaries and greatly undercompensate consumers in the name of efficiency in the
administration of the antitrust laws.”37 Consequently, the argument goes, a state remedy
is necessary to provide consumers with relief. Even where state antitrust law does not
allow consumers to sue, the consumer interest argument has persuaded some courts to
See generally, Albert A. Foer, A Public Interest Advocate Looks at State Antitrust Enforcement,
Presentation to Nat’l Ass’n of Attorneys General, at 11-14 (Oct. 2, 2002), at
http://www.antitrustinstitute.org/recent2/206.pdf (endorsing state Illinois Brick repeals as necessary to
protection of consumer).
Bunker’s Glass Co. v. Pilkington, PLC, 2002 WL 1461739 (Ariz. App. Ct.).
allow relief under consumer protection statutes.38 In addition, courts have accepted the
argument that indirect purchasers should be allowed to sue because direct purchasers will
not do so for fear of retaliation by the defendants, their suppliers.39
III. Compensation of Consumers in Indirect Purchaser Class Actions
We will argue in the next section that allowing indirect purchaser suits is
inconsistent with a policy of efficient deterrence, the primary goal of the system of
antitrust damages. But we shelve that issue for the moment to address the compensation
rationale on its own terms. In this section, we argue that, if compensation of consumers
is the goal of indirect purchaser class action litigation, that litigation has been a miserable
failure. We are not resting our contention here on the claim some Chicagoans have made
that, even under the pure Illinois Brick rule, consumers actually receive compensation
See, e.g., Ciardi v. F. Hoffmann-La Roche, Ltd., 762 N.E.2d 303 (Mass. 2002); Mack v. Bristol
Myers Squibb Co., 673 So.2d 100, 109 (Fla. Dist. Ct. App. 1996). Other courts have held that indirect
purchasers cannot recover under consumer protection statutes where they are barred from suing under state
antitrust laws. See, e.g., Vacco v. Microsoft Corp., 793 A.2d 1048, 1058, 1066 (Conn. 2002); Davidson v.
Microsoft Corp., 792 A.2d 336, 344-45 (Md. Ct. Spec. App. 2002); Arnold v. Microsoft Corp., No. 2000-
CA-002144-MR, 2001 WL 1835377, at *4, *7 (Ky. Ct. App. Nov. 21, 2001); Keiffer v. Mylan Labs., Inc.,
1999-2 TRADE CAS. (CCH) ¶ 72,673 (N.J. Super. Ct. 1999); Gaebler v. N.M. Potash Corp., 676 N.E.2d 228
(Ill. Ct. App. 1997); Abbott Labs., Inc. v. Segura, 907 S.W.2d 503 (Tex. 1995). Where the state did not
have a specific antitrust statute, one court permitted an indirect purchaser action under the general
consumer protection statute (see Elkins v. Microsoft Corp., No. 2001-431, 2002 WL 31433092 (Vt. Nov. 1,
2002)), and one court rejected it (see Blewett v. Abbott Labs., Inc., 938 P.2d 842, 845-46 (Wash. Ct. App.
Comes v. Microsoft Corp., 646 N.W.2d 440 (Iowa 2002).
indirectly, because direct purchasers will “pass on” their expected antitrust recovery in
the form of lower prices.40 Even if the critics of Illinois Brick are right in arguing that it
denies compensation to consumers, their state-law alternatives have proved to be no
better on this score. Courts have thrown up their hands both at the outset and at the end
of state-law class actions when confronted with the task of determining which consumers
were harmed and by how much. The shortcomings of indirect purchaser class actions are
apparent in class certification decisions and in decisions approving settlements.
A. Class Certification
In Reiter v. Sonotone Corp.,41 which endorsed consumer standing in principle, the
Supreme Court acknowledged the problems that consumer class actions would pose for
the federal courts. But it pointed to class certification under Federal Rule 23 as a
significant and necessary screen to weed out unduly burdensome cases.42 The
certification procedure allows the court to determine at the outset of the litigation if the
lawsuit is appropriate for class treatment. The court is supposed to conduct a “rigorous
analysis” to determine whether the requirements for class certification are met.43
See Landes & Posner, Indirect Purchasers, supra note xx, at 605. The idea is that the direct
purchasers’ expected treble damage recovery offsets the cost increase from the overcharge.
442 U.S. 330 (1979).
Id. at 345. See also MANUAL FOR COMPLEX LITIGATION, THIRD § 30.1 (1995) ("The decision on
whether or not to certify a class . . . can be as important as decisions on the merits of the action and should
be made only after consideration of all relevant evidence and arguments presented by the parties.").
General Tel. Co. v. Falcon, 457 U.S. 147, 161 (1982). For example, the court in Castano v. Am.
Tobacco Co., 84 F.3d 734, 742 (5th Cir. 1996) insisted that “a court cannot rely on assurances of counsel
Some state courts freely certify class actions, including those by indirect
purchasers, with no apparent scrutiny of the specifics of the claims at issue.44 Many
others, however, have interpreted the class certification procedures mandated by the state
counterparts of Rule 23 consistently with the rigorous federal view.45 For these courts,
the certification process has proven to be a significant screen of indirect purchaser class
actions, particularly those involving consumer classes.46 The most significant procedural
hurdle for certification of damage class actions is that issues common to the class
predominate over individual issues.47 The requirement actually has two components, that
there be common issues and that they predominate. In most indirect purchaser litigation,
that any problems with predominance or superiority can be overcome.” Moreover, “[g]oing beyond the
pleadings is necessary, as a court must understand the claims, defenses, relevant facts, and applicable
substantive law in order to make a meaningful determination of the certification issues.” Id. at 744. But cf.
In re Visa Check/Mastermoney Antitrust Litig., 192 F.R.D. 68, 76 (E.D.N.Y. 2000) (“[A] court at the class
certification stage should not delve into the merits of an expert's opinion, or indulge ‘dueling’ between
opposing experts.”), aff’d, 280 F.3d 124 (2d Cir. 2001).
See Page, supra note xx, at 20 n.86.
See, e.g., S.W. Ref’g Co. v. Bernal, 22 S.W.3d 425, 435 (Tex. 2000) (rejecting the “approach of
certify now and worry later"); Execu-Tech Bus. Sys., Inc. v. Appleton Papers, Inc., 743 So.2d 19, 21-22
(Fla. App. 1999) (“[T]he trial court has the obligation to conduct a rigorous analysis to determine whether
the elements of the class action rule have been satisfied.”). The state courts do not consistently follow the
federal model. See, e.g, Linda S. Mullenix, Abandoning the Federal Class Action Ship: Is There Smoother
Sailing for Class Actions in Gulf Waters?, 74 TUL. L. REV. 1709 (2000).
Page, supra note xx, at 31-33. See also Melnick v. Microsoft Corp., No. CV-99-709, 2001 WL
1012261 (Me. Super. Ct. Aug. 24, 2001), at *4 n.5 (collecting cases).
See, e.g., FED. R. CIV. P. 23(b)(3).
the issues of liability, including the existence of the illegal conduct—price fixing by an
international cartel, for example—are common to the class. The sticking point is
injury—both the fact and the amount of injury—to indirect purchasers. Most state courts
that have applied an exacting standard to indirect purchaser cases have concluded that
individual issues pertaining to injury predominate over common issues relating to
liability. Only in a very narrow range of cases in which consumers buy goods that were
subject to anticompetitive pricing at some higher level of the chain of distribution has the
class action device been available.
To obtain certification, any plaintiffs’ class must as a practical matter offer a
theoretical model demonstrating that all class members have been harmed.48 The fact of
injury to direct purchasers from a price fixing conspiracy often can be shown by a
classwide model, because proof that the conspiracy was successful in raising prices
shows that direct purchasers would probably have paid less but for the conspiracy.49
Nevertheless, the nature of the alleged conspiracy, product heterogeneity, variations in
See, e.g., A&M Supply Co. v. Microsoft Corp., 654 N.W.2d 572, 583 (Mich. Ct. App. 2002)
(noting authority for the proposition that “if a plaintiff proposes a method or formula by which the court
could determine that the defendant’s conduct caused actual damages or injury to each class member, even if
the amount of damages due each party is later computed at a separate trial, then a class action is possible”).
See, e.g., In re Linerboard Antitrust Litig., 305 F.3d 145, 148-49 (3d Cir. 2002); In re Northwest
Airlines Corp., 208 F.R.D. 174 (E.D. Mich. 2002); Sebo v. Rubenstein, 188 F.R.D. 310, 314-15 (N.D. Ill.
1999). See also Page, supra note xx at 10-12.
buying power, and regional price differences can complicate matters sufficiently to
preclude certification even in the direct purchaser case.50
Whatever the obstacles to certification confronting direct purchasers, indirect
purchasers inevitably face a more daunting task.51 They must prove on a classwide basis
not only that direct purchasers were injured, but also that direct purchasers in fact passed
on at least some of the overcharge to the plaintiffs. That means coming up with a theory
that allows a reasonable inference that all members of the indirect purchaser class
suffered harm. Intermediate purchasers, however, are often a diverse group of stores,
distributors, or manufacturers subject to a variety of constraints that affect how they
respond to a price increase, and how and in what form they resell the product. Depending
on the circumstances, some or all of the direct purchasers may pass on none of the
overcharge. Establishing injury to indirect purchasers by a generalized method is often
Proponents of indirect purchaser classes most often rely on the tax incidence
model, which treats the overcharge to direct purchasers as an excise tax.52 The “tax”
See, e.g., In re Elec. Weld Steel Tubing Antitrust Litig., No 79- 4628, 1980 U.S. Dist. LEXIS
9604, at *20 (E.D. Pa. Nov. 6, 1980). Cf. Weisfeld v. Sun Chem. Corp., 210 F.R.D. 136 (D.N.J. 2002)
(finding common issues did not predominate on the question of impact of an alleged conspiracy of
defendants not to hire one another’s employees).
Page, supra note xx, at 12-13.
Illinois Brick Co. v. Illinois, 431 U.S. 720, 742 n.25 (1977). On tax incidence analysis, see
PAUL A. SAMUELSON, ECONOMICS 164-65 (11th ed. 1980):
Who ultimately pays a particular tax? Does the burden stay on the person on
whom it is first levied? One cannot assume that the people Congress says a tax is levied
shifts intermediate purchasers’ marginal costs upward,53 and thus induces them to
increase their own prices to their customers. How much they increase their prices
depends upon the relative elasticities of supply and demand in the resale market.54 Only
if supply is perfectly inelastic or demand is perfectly elastic would the direct purchaser be
unable to pass on any of the overcharge.
on will end up paying that tax. They may be able to shift the tax: shift it “forward” on
their customers by raising their price as much as the tax; or shift it “backward” on their
suppliers (wage earners, rent and interest receivers) who end up being able to charge
them less than they would have done had there been no tax.
Economists therefore say: We must study the final incidence of the tax—the
way its burden ultimately is borne, the totality of its effects on commodity prices, factor-
prices, resource allocation, efforts, and composition of production and consumption. Tax
incidence is no easy problem and requires all the advanced tools of economics to help
toward its solution.
For diagrammatic presentations, see ROGER D. BLAIR & DAVID KASERMAN, ANTITRUST
ECONOMICS 74-77 (1985); Joseph H. Anderson, Comment, A Legal and Economic Analysis of the Cost-
Plus Contract Exception in Hanover Shoe and Illinois Brick, 47 U. CHI. L. REV. 743 (1980); Harris &
Sullivan, supra note xx; Elmer J. Schaefer, Passing-On Theory in Antitrust Treble Damage Actions: An
Economic and Legal Analysis, 16 WM. & MARY L. REV. 883 (1975).
The share of the overcharge borne by the indirect purchaser bears the same proportion to the
share borne by the direct purchaser as the elasticity of supply bears to the elasticity of demand. See Illinois
Brick, 431 U.S. at 741; Schaefer, supra note xx, at 893. Thus, if the elasticity of supply is infinite, the
indirect purchaser bears the entire burden; if the elasticity of demand is infinite, the direct purchaser bears
the entire overcharge.
Courts and other observers vary widely in their acceptance of the tax incidence
theory as a basis for classwide proof of injury.55 Some take the view that the model,
along with other generalizations about human behavior, makes harm to indirect
purchasers virtually inevitable. But others—following the Supreme Court’s analysis in
Hanover Shoe—believe that real world factors overwhelm the theoretical predictions of
the model in most cases. In effect, the “sanguine” and “skeptical” views reflect different
legal standards that dramatically affect the outcome of certification decisions:56 “[t]he
most important determinant of class certification appears to be where the case is filed.”57
In addition to legal standards, specific characteristics of the market can limit the
applicability of the tax incidence theory and thus make certification less likely. The
theory assumes, for example, that intermediate levels of distribution are perfectly
competitive, but this assumption is often belied by substantial bargaining power and
product heterogeneities. If the intermediate market is oligopolistic, theory cannot predict
a determinate pass-on rate.58 Some powerful buyers, for example, may be able to
negotiate better deals with the defendants. In other instances, competitive conditions in
local markets may not permit an immediate change, or some stores may sell the product
as a loss leader.
Page, supra note xx, at 17-19.
Id. at 21-27. See A&M Supply Co. v. Microsoft Corp., 654 N.W.2d 572, 580-81 (Mich. Ct.
App. 2002) (recognizing significance of skeptical and sanguine views).
Page, supra note xx, at 21.
Id. at 16.
Other market characteristics can make rates of passing on less predictable.59 The
product may be heterogeneous, and product variations in form or size may make it
difficult to show that indirect purchasers suffered harm in any generalized way. To the
extent that the product is altered at any point in the chain of distribution, the value added
by the change can swamp the price increase. This effect is most apparent where the
product, such as high fructose corn syrup, lysine, or carbon fiber, is an ingredient or input
in the manufacture of other products.60 If the ingredient accounts for only a small
percentage of the price of the product, then the increment in price of the final consumer
product attributable to an overcharge may be too small to measure. One illuminating
survey of the economic literature found that dealers mark prices up at different rates
depending upon whether retailers: (1) pursue an “everyday low price” strategy or relied
on periodic sales; (2) discriminate in price among classes of buyers; (3) offer volume
discounts; (4) use a “focal price” strategy that requires all prices to end in the numeral 9;
or (5) benefit from various discount programs from wholesalers or manufacturers.61 Also
relevant are differences in the costs of changing posted prices; the relative “speed of
adjustment” to “cost shocks” of various sizes; advertising rates; point-of-sale services;
prevalence of private label goods; and informational and locational advantages or
Id. at 27-34.
Id. at 31.
Samid Hussain, et al., Economics of Class Certification in Indirect Purchaser Suits, 10
COMPETITION 18, 40-42 (2001), in http://www.cornerstone.com/pdfs/classcert.pdf.
Id. at 42-46.
Interestingly, consumers are by far the least likely of all indirect purchasers to
satisfy the requirements for class treatment.63 They do have the advantage that they are
the end purchasers, and thus not subject to the defense, if the relevant state law
recognizes it, that they have passed on the overcharge.64 But in all other respects, they
face more obstacles to class certification than secondary intermediate purchasers. First,
they are most likely to be separated from the offenders by more than one level of
distribution. Just as one level of intermediate purchasers complicates the problem of
showing injury by classwide proof, additional levels add that much more complexity.
Where manufacturers are the offenders, normally there will be a class of distributors who
are direct purchasers and a class of retailers who are indirect purchasers. Where the
product is repackaged or used as an ingredient or input in another product, there may be
several levels of distribution before the product reaches consumers. As an overcharge
passes through successive sales layers, it shrinks. If an overcharge to a retailer, divided
among all of its customers, would only be a few cents, the retailer may not bother to raise
prices at all, which means that its customers would not in fact be injured. The difficulty
of proving injury in fact from an overcharge passed on through multiple distribution
levels may explain why most cases in which plaintiffs even seek certification of indirect
purchaser classes involve only one tier of intermediate purchasers. Even a court that
granted certification to a consumer class under an unusual statute that explicitly
Page, supra note xx, at 31-33.
Goda v. Abbott Labs., 1997-1 TRADE CAS. (CCH) ¶ 71,730, at 79,143 (D.C. Super. Ct. 1997).
authorized the proof of harm on a classwide basis noted that the result might be different
if there were more intermediate levels.65
In addition, consumers do not retain documentation of most of their purchases.66
They may not even know they are buying a product if, like corn syrup or lysine, it is
incorporated into another product. Without records, the only way class counsel can
determine the amount and price of consumer purchases, and hence whether class
members suffered injury, would be by questionnaires, a process one court characterized
as an invitation to perjury.67
See, e.g., Goda, at 79,142 (recognizing that case involved a single level, because wholesalers
deemed part of conspiracy).
Durden v. Abbott Labs., No. CV 93-663, slip op. at 12-13 (Ala. Cir. Ct. Jan. 16, 1996) (noting
that one class representative could not remember the brand of formula she purchased); Wilcox, slip op. at 7
(observing that a consumer may not even be aware that ingredient was in product; will have no records);
Fischenich v. Abbott Labs., No. MC-94-6868, slip op. at 5 (Minn. Dist. Ct. May 26, 1995) (“[T]he
plaintiffs in this matter have no idea how much they paid for formula and only a general idea how much
formula was purchased.”).
Durden, slip op. 13 (“[T]o simply declare that the [the Court] had a fund . . . out of which
thousands of retail consumers could essentially help themselves by the filing of a form would encourage
fraud and perjury.”). The court distinguished In re Domestic Air Transp. Antitrust Litig., 137 F.R.D. 677
(N.D. Ga. 1991) on the ground that evidence of dates and points of travel was available for each consumer
class member. Id. See also Ashley, slip op. at 40 (rejecting as “fundamentally flawed” plaintiffs proposal
for creating a “common fund . . . against which farmers could file claims and be paid according to the
extent and reliability of their documentation. Any unclaimed amount would be paid to a public interest
group under the cy pres doctrine.”).
Thus far, we have focused on classwide proof of the fact of injury. But plaintiffs
also have the burden of proving amount of harm, albeit typically under a less demanding
standard.68 Some states expressly permit classwide proof of damages.69 But where states
require individualized proof, the individual issues, relating to amount of harm, may
predominate over the common ones. This condition is especially likely to impede
certification of consumer classes, because consumers are exponentially more numerous
than middlemen, numbering in the thousands or even millions.70 Consequently, the sheer
number of “mini-trials” that would be required is larger than for a class of retailers or
middlemen.71 Furthermore, consumer damages, spread over so many buyers, are
necessarily much smaller, and thus more easily lost in other factors that may influence
the amount by which retailers increase prices.72 The burden of these calculations can
make a consumer class action unmanageable.73
For a discussion of the relevant standards under federal antitrust law, see PROVING ANTITRUST
DAMAGES 31 (William H. Page ed., 1996) (“The standards for proving the fact that the plaintiff was injured
are more stringent than those for proving the amount of that injury.”).
See, e.g., D.C. CODE ANN. § 28-4508(c) (“In any class action . . . , the fact of injury and the
amount of damages sustained by members of the class may be proven on a class-wide basis . . . .”).
Page, supra note xx, at 32.
“These ultimate consumers, in today’s case the buyers of single pairs of shoes, would have only
a tiny stake in a lawsuit . . . .” Hanover Shoe, Inc. v. United Shoe Mach. Co., 392 U.S. 481, 494 (1968).
In Ashley, the court noted that “the cost of lysine comprises a tiny fraction of the total cost of
animal feed, perhaps only 1.5%-2.0% of the cost of feed when it is included in the feed mix.” Ashley, slip
op. at 27.
See, e.g., A&M Supply Co. v. Microsoft Corp., 654 N.W.2d 572, 603 (Mich. Ct. App. 2002)
(holding that a class “numbering in the hundreds of thousands” would be unmanageable); Wilcox v.
Interestingly, in the Microsoft indirect purchaser litigation, most of the courts that
have considered the issue have granted certification. The presentations on both sides in
all of these certification disputes were essentially identical, consisting of affidavits from
the same experts, with no inquiry into the specifics of the software markets in the states
themselves. To a large degree, the results confirm the observation quoted earlier that
“[t]he most important determinant of class certification appears to be where the case is
filed.”74 The courts of four of the eight states in which classes were certified had already
established a consistent pattern of leniency in application of certification standards in
indirect purchaser cases,75 while the courts of both of the states denying certification had
established a pattern of strict application.76 But two of the grants of certification in
Microsoft cases occurred in states that had previously applied standards strictly.77 Of
these, the Florida court seems clearly to have adopted the “sanguine view” of the
probative value of incidence analysis and other methods of proving that the overcharge
was passed on—although it cited the earlier Florida case denying certification, it made no
effort to distinguish it, and failed to acknowledge the decisions in other jurisdictions that
Archer-Daniels-Midland, No. 96-82473-CP, slip op. at 5-7 (Mich. Cir. Ct. Sept. 29, 1997) (holding that a
class consisting of entire population of state was unmanageable).
Page, supra note xx, at 21.
Id. (noting grants of certification in Kansas, California, Wisconsin, and South Dakota).
Id. (noting denials of certification Michigan and Maine).
Id. (noting denials of certification Florida (once) and Minnesota (four times)).
rejected certification of indirect purchaser classes.78 The Minnesota court did identify
characteristics of the Microsoft case that purportedly made certification of an indirect
purchaser class more appropriate than in earlier cases in the same jurisdiction.79 In
contrast, the two courts that denied certification evidently adopted a skeptical view of the
prospects for common proof of harm.80
These differences in approach at the certification stage have important practical
consequences for the parties, because, in general, certification greatly enhances the
probability and terms of settlement.81 But even where classes have been certified and the
In re Fla. Microsoft Antitrust Litig., No. 99-27340, 2002 WL 31423620 (Fla. Cir. Ct. Aug. 26,
2002). The court cited but did not distinguish Execu-Tech Bus. Sys., Inc. v. Appleton Papers, Inc., 743
So.2d 19 (Fla. Dist. Ct. App. 1999), aff’g No. 96-9636 CACE 05 (Fla. Cir. Ct. Dec. 15, 1997).
In Gordon v. Microsoft Corp., No. 00-5994, 2001 WL 366432 (Minn. Dist. Ct. Mar. 30, 2001),
the court distinguished the circumstances in Microsoft from those in several earlier Minnesota cases in
which certification was denied. The court accepted, for example, the plaintiffs’ expert’s characterization of
the distribution market for Microsoft software, usually through OEMs, is “highly competitive,” id. at *9, a
necessary condition for incidence theory to predict a consistent pass-through of cost increases. Moreover,
the court noted that intermediate sellers do not change software significantly, and that consumers are more
likely to have records of their purchases of computers, with installed software, than of other consumer
products. Id. at *6. Acknowledging the diversity of forms of the product, the court concluded that
damages still might be proven for each class member “mechanically.” Id. at *12.
See, e.g., A&M Supply Co. v. Microsoft Corp., 654 N.W.2d 572 (Mich. Ct. App. 2002);
Melnick v. Microsoft Corp., No. CV-99-709, 2001 WL 1012261 (Me. Super. Ct. Aug. 24, 2001).
See, e.g., Barry F. McNeil, Class Actions: A Time for a Change, 23 No. 2 LITIGATION 1, 1
(1997); In re Rhone-Poulenc Rorer Inc., 51 F.3d 1293, 1298 (7th Cir. 1995) (Posner, C.J.) .
parties have reached a settlement agreement, the practical benefits for consumers have
usually been minimal. Microsoft has proven no exception to this rule.
B. Distribution of Settlement Funds
Where indirect purchaser classes of consumers have been certified and have
produced a settlement fund, they have produced little in the way of real compensation for
consumers, while imposing significant costs of fund administration. A 1988 study
concluded that consumer class actions involving small amounts of damages to individual
class members were ineffective in distributing damages to members of the class,
primarily because consumers did not file claims in accordance with the claims procedures
established by the courts.82 It is easy to see that, no matter how efficient the claims
procedures, many consumer class actions will involve individual harms that are less than
the cost of determining and distributing damages, which can be substantial.83 As Judge
Motz observed in reviewing the proposed settlement in the Microsoft federal class
the sheer number of potential class members (it has been estimated that
there are as many as 100 million possible claimants), the transient nature
of the U.S. population, the high rate of software piracy, and the fact that
many individual consumers would not have retained proof of purchase
Gail Hillebrand & Daniel Torrence, Claims Procedures in Large Consumer Class Actions and
Equitable Distribution of Benefits, 28 SANTA CLARA L. REV. 747 (1988).
In re Microsoft Corp. Antitrust Litig., 185 F. Supp. 2d 519, 523 n.2 (D. Md. 2002) (noting that
plaintiffs’ expert testified that “the cost of processing claims in other cases has ranged from $32 to $292 per
claim”; objectors’ expert “estimates that the total processing cost per claimant would be $7.52 to $9.00.”).
documents all point to the unfeasibility of economic distribution of class
Thus, even if the experts in the indirect purchaser class actions are right that it will be
easy to show that Microsoft harmed consumers, it will be virtually impossible for
consumers actually to recover any significant compensation.
Courts in some antitrust consumer class actions have attempted to distribute
settlement funds to consumers, but with little success. At best, these attempts involve
gross approximations of real harm. For example, in the Domestic Air Transport
litigation,85 the court set up a procedure for distribution of money based upon the amount
of air travel, presumably on the assumption that consumers who bought more tickets paid
more of the overcharge. Tellingly, all consumers who purchased any tickets would get a
$100 certificate; those who could prove more travel could get more.86 By definition, the
Id. He continued: “While other factual questions would need to be explored, including (1) the
ease and cost of identifying the business organizations who the parties agree hold the bulk of the potential
claims, and (2) the possibility of using a ‘product key’ written into every item of Microsoft's software as a
means for verifying a claim, the practical difficulties of distributing any class recovery seem immense.”
In re Domestic Air Transp. Antitrust Litig., 148 F.R.D. 297 (N.D. Ga. 1993).
Id. at 309. The court wrote:
To receive certificates, class members must submit claims based on the amount
of class travel purchased. Class members who wish to claim fewer than five round-trip or
ten one-way class tickets must file Claim Form A. Claim Form B claimants must have
purchased at least five round-trip or one-way class tickets and their purchases will be
automatically valued at $2,500 regardless of the actual purchase amount. Class members
who file a long Claim Form C must have purchases in excess of $2,500 during the class
period. The certificates issued by American, Continental, Delta, TWA, United, and
$100 certificate is an average—it bears no necessary relationship to any actual
Even with approximations of various kinds, many settlement funds are not fully
distributed, because consumers do not find it worth their while to file claims.87 In the
Microsoft state court litigation, the California class plaintiffs agreed to a settlement under
which Microsoft would create a fund of $1.1 billion to pay for vouchers that class
members could use to purchase software products from any vendor.88 The vouchers
would amount to $29 per copy of Office and $16 per copy of Windows purchased
between 1995 and 2001. Because of the anticipated low claims rate (historically around
25%), the settlement also provides that Microsoft would donate two-thirds of the
unclaimed vouchers to schools and would retain the remaining third. Though this
arrangement may result in significant benefits to some business purchasers, consumers as
a whole are not likely to receive much compensation, even if the court approves the
USAir will be divided into a base fund of settlement certificates in the amount of
$268,984,000 and will be distributed to all claimants on a per capita basis up to $100 per
claimant. A supplemental fund of these fund of these certificates in the amount of
$127,516,000 will be distributed to B and C form claimants, in addition to their fund
distribution, on a pro rata basis.
See, e.g., In re Motorsports Merch. Antitrust Litig., 160 F. Supp.2d 1392, 1393 (N.D. Ga. 2001)
(noting that in a class action for fixing prices of novelties sold at NASCAR races, “[c]laims were submitted
for substantially less than the cash amount of the settlement.”).
Kim Peterson, et al, Microsoft Settles With California for $1 Billion; Consumers to be Paid in
Vouchers, SEATTLE TIMES, Jan. 11, 2003, at A1.
settlement. The effort of gathering and submitting the documentation necessary to claim
the vouchers would be an overwhelming deterrent.89
In other cases, the plan for distribution of the settlement recognizes from the
outset that members of the class will receive little or nothing.90 For example, in the Toys
R Us litigation, the court approved a settlement that established no procedure for claims
by injured consumers, because the size of individual harms was too small to make
distribution practical.91 In these circumstances, courts often turn to cy pres distributions
of part or even all of the funds to worthy causes.92 In Toys R Us, the solution adopted
Patrick Thibodeau, Microsoft Settlement Could Yield Rebates; But Hassles Await Users Who
Apply for Vouchers, COMPUTER WORLD, Jan. 20, 2003, at 10 (quoting market participants who question
whether the expense of gathering information to file claims will be worth the value of the vouchers).
See generally Geoffrey P. Miller & Lori S. Singer, Nonpecuniary Class Action Settlements, 60
LAW & CONTEMP. PROB. 97 (1997).
In re Toys R Us Antitrust Litig., 191 F.R.D. 347, 353 (E.D.N.Y. 2000) (“The decision to forego
individual recoveries was sensible, given the difficulty of identifying proper claimants and the difficulty,
and especially the costs, that such recoveries and their administration would have entailed. The net
monetary relief for any individual claimant would have been limited.”).
In re Microsoft Corp. Antitrust Litig., 185 F. Supp.2d 519, 523 (D. Md. 2002) (“[Cy pres] has
also been utilized as a means for distributing the entirety of a class fund where the proceeds cannot be
economically distributed to the class members.”) (citing In re Toys "R" Us Antitrust Litig., 191 F.R.D. 347
(E.D.N.Y. 2000); New York v. Reebok Int'l Ltd., 903 F. Supp. 532 (S.D.N.Y. 1995), aff'd, 96 F.3d 44 (2d
Cir.1996); 2 HERBERT B. NEWBERG & ALBA CONTE, NEWBERG ON CLASS ACTIONS § 11-20 (3d ed. 1992);
7B CHARLES ALAN WRIGHT, ARTHUR R. MILLER & MARY K. KANE, FEDERAL PRACTICE AND PROCEDURE
§ 1784, at 84 (2d ed. 1986)). See Motorsports (approving “a cy pres distribution in the amount of $250,000
to each of the following nine charities: Make-a-Wish, The American Red Cross, Race Against Drugs,
was to use the fund to distribute millions of dollars worth of toys to children
nationwide.93 State indirect purchaser suits have resorted to similar expedients. For
example, in one case involving alleged price fixing of sorbates the court endorsed the
opinion of class counsel that “end food products contain very small amounts of Sorbates
(ranging from 0.02% to 0.3% concentrations in final products). The difficulty in
determining the dollar value of such purchases and the Sorbates dollar value contained in
these products makes the calculation of damages prohibitively difficult.”94 Consequently,
the court approved the following settlement term:
To end consumers who purchased products containing Sorbates
and to all business entities who purchased products containing Sorbates
solely for resale to end consumers (such as grocery stores and restaurants),
an amount of $50,000 shall be donated on their behalf through a cy pres
Atlanta Legal Aid, Children's Healthcare of Atlanta, Georgia Legal Services Program, KIDS' CHANCE,
Duke Children's Hospital and Health Center, and the Lawyers Foundation”). But cf. In re Airline Ticket
Comm'n Antitrust Litig., 307 F.3d 679 (8th Cir. 2002) (reversing cy pres award to charities and schools of
undistributed funds in a class action by travel agents alleging conspiracy of airlines to cap their
commissions; requiring payment to travel agencies in U.S. territories).
The distribution was not self-enforcing. The agreement recited that the “toys distributed must
be in good condition, include toys suitable for both boys and girls in a wide age range and be from current
inventory.” Toys R Us, 191 F.R.D. at 354. The defendants had to produce advance lists of toys to be
distributed. The plaintiffs undertook “to verify compliance, for example, by sending investigators to retail
shops” (id.), and the federal court became, in effect, a regulatory agency for the distribution of toys to
Williams Foods, Inc. v. Eastman Chem. Co., NO. 99C16680, 2001 WL 1298887 (Kan. Dist. Ct.
Aug. 8, 2001).
award to the United Community Services of Johnson County, a charitable
organization providing an essential safety net for the most vulnerable
citizens, and to the Children's Therapeutic Learning Center, a charitable
organization providing therapy services to children with disabilities from
ages 2 to 6 years old.95
In Microsoft, Judge Motz considered a proposed national settlement that would have
required Microsoft to donate computer equipment to schools rather than to compensate
consumers directly.96 He rejected the settlement not because it adopted a cy pres
approach—which he found unavoidable—but because the settlement was underfunded.97
These settlements demonstrate that in most instances it will be impossible to give
consumers an award that even approximates their actual harm. Cy pres distributions may
be appropriate as a means of imposing the necessary deterrent penalty. For example, in a
direct purchaser class action, they may be the only practical way to impose a penalty on
the defendant that approximates the overcharge. But they do nothing to compensate
To summarize: In many indirect purchaser antitrust class actions, class counsel
do not even attempt to define a consumer class. Of the indirect purchaser consumer class
actions that are filed, many are not certified. Of those that are certified, most make no
effort to provide actual damages to the consumers who suffered the harm, preferring to
distribute the money “on their behalf” to worthy causes. Of those certified consumer
In re Microsoft Corp. Antitrust Litig., 185 F. Supp.2d 519, 523 (D. Md. 2002).
Id. at 527-28.
class actions that do attempt to award actual damages, the claims rates are low, and the
remaining money must be disposed of in other ways. The conclusion is inescapable that
indirect purchaser class actions do not provide significant compensation to consumers.
IV. Indirect Purchaser Class Actions and Efficient Deterrence
Even if indirect purchaser class actions do not provide compensation to
consumers, they still might benefit the consumer interest by providing better deterrence
than a pure Illinois Brick regime. Deterrence actually is preferable to compensation as a
legal goal, even from the consumer’s perspective, because if it is successful,
compensation is unnecessary. But “better” deterrence in this sense does not necessarily
mean larger penalties. Given imprecision in the scope and application of liability rules,
such as those governing monopolization, an unduly large penalty can deter efficient
market conduct. Even in price fixing cases, where there is little risk of unwarranted
liability, an arbitrarily high damage award can be inefficient.98 If the expected penalty is
too large, firms will make excessive investments in penalty avoidance, such as
monitoring employees, thereby increasing their costs to the detriment of social welfare.99
In short, excessive penalties can deter socially beneficial conduct, induce inefficient
See, e.g., POSNER, supra note xx, at 267; Bruce H. Kobayashi, Antitrust, Agency, and Amnesty:
An Economic Analysis of the Criminal Enforcement of the Antitrust Laws Against Corporations, 69 GEO.
WASH. L. REV. 715, 732 (2001).
See, e.g., Kobayashi, supra note xx, at 732; Daniel R. Fischel & Alan O. Sykes, Corporate
Crime, 25 J. LEGAL STUD. 319, 324 (1996); Marc A. Cohen & David T. Scheffman, The Antitrust
Sentencing Guideline: Is the Punishment Worth the Cost?, 27 AM. CRIM. L. REV. 331, 354-55 (1989).
investments, and inflate the price of goods, producing exactly the kind of deadweight loss
the antitrust laws are intended to prevent.
Rational antitrust policy, therefore, seeks to impose an optimal penalty, one that
deters conduct that reduces social wealth without deterring socially beneficial behavior or
stimulating excessive investments in precaution.100 The system of private antitrust
enforcement points toward optimal penalties through the doctrines of antitrust injury and
standing. These doctrines harness the private right of action to the task of imposing an
appropriate deterrent penalty—that is, one that will provide the optimal, not the greatest,
level of deterrence. As a first approximation of the optimal penalty, antitrust injury limits
recovery to those whose injuries are caused by an anticompetitive aspect of an antitrust
violation.101 Rivals who are harmed, for example, by an illegal merger that makes the
surviving firm more competitive do not suffer antitrust injury. As a second, closer
approximation, the doctrine of antitrust standing further limits recovery to plaintiffs who
can most efficiently sue for antitrust injury.102 The goal is to limit the right to sue to
classes of plaintiffs who are likely to bring suit for harms that, in the aggregate, impose
something like the optimal penalty at the lowest direct cost.
The Supreme Court’s treatment of monopolistic overcharges reflects this
approach. If it were certain to be imposed, the overcharge to direct purchasers would be
See generally William H. Page, The Scope of Liability for Antitrust Violations, 37 STAN. L.
REV. 1445, 1452 (1985).
Id. at 1461-62.
Id. at 1483-85.
a fair approximation of the optimal penalty.103 And one can view the trebling feature as a
means of adjusting for the risk that the offense will not be detected.104 Illinois Brick is an
expression of antitrust standing analysis, because it allocates the full right to sue for the
overcharge to the direct purchasers who, the Court determined, are the best candidates
(on balance) to sue for the overcharge.105
State indirect purchaser suits depart from the federal remedial framework by
conferring a right to sue in addition to the direct purchasers’ exclusive right to sue for
three times the overcharge in federal court. In this Part, we consider whether such a
departure from the Illinois Brick regime can be justified by an economically rational
policy of deterrence.106 We address two lines of argument. In the first section, we
examine the claim that Illinois Brick undermines effective deterrence because direct
purchasers are unlikely to sue, or at least less likely to sue than indirect purchasers. We
show that both theory and experience refute this claim: in most cases, classes of both
direct and indirect purchasers will sue whenever either of them does so. But is this state
of affairs good or bad? In the second section, we examine the assertion that, although
See William M. Landes, Optimal Sanctions for Antitrust Violations, 50 U. CHI. L. REV. 652
(1983) (describing optimal penalty for a cost-saving cartel).
Id. at 1465.
Id. at 1487-88.
Blair and Harrison recognize that the Court’s decision in ARC America allowing states to
permit indirect purchaser suits had “a severe derationalizing effect on antitrust enforcement.” Roger D.
Blair & Jeffrey L. Harrison, Reexamining the Role of Illinois Brick in Modern Antitrust Standing Analysis,
68 GEO. WASH. L. REV. 1, 11 (1999). But rather than limit the ability of indirect purchasers to sue under
state laws, as we would do, they would allow all intermediate purchasers to recover lost profits.
direct purchasers manifestly do sue for the full overcharge, indirect purchasers should
also be able to sue for their injuries.
A. Are Indirect Purchasers More Likely to Sue than Direct Purchasers?
One longstanding argument for indirect purchaser suits is that direct purchasers
lack sufficient incentive to sue.107 The argument is that direct purchasers might refrain
from suing in order to protect supply relationships, a possibility recognized by the Court
in Illinois Brick,108 or that because direct purchasers will have avoided any significant
loss by passing on the overcharge, they will have no reason to sue. Neither claim is
The Illinois Brick Court reasoned that concentrating the right to sue in the direct
purchaser would be preferable to apportioning the overcharge among direct and indirect
purchasers, because fragmentation of the right to sue increases the uncertainty of
recovery and reduces the expected recovery of each class.109 Illinois Brick “repeals” may
reflect the judgment that direct purchasers have a lesser incentive to sue for the full
See O’Connor, supra note xx, at 38:
[T]the notion that indirect purchasers lack a sufficient incentive to sue so as to deter antitrust
violators has certainly been laid to rest. Notwithstanding the procedural difficulties outlined here, private
plaintiffs have increasingly sought such relief under state laws. If a right of action for indirect purchasers
had been available under federal law, we undoubtedly would have seen a substantial development and
streamlining of such practice in the federal system. In fact, it is likely that indirect purchasers are perhaps
more likely to bring suit than direct purchasers, given their lack of a direct business relationship with the
Id. at 746.
Illinois Brick, 431 U.S. at 745.
overcharge than indirect purchasers have to sue for their portion of the overcharge.
Experience in state indirect purchaser litigation certainly suggests that the Court in
Illinois Brick underestimated the incentives for indirect purchasers to sue for a passed-on
overcharge. Plaintiffs’ attorneys have had ample incentive to organize class actions to
sue for the passed-on overcharge in scores of cases despite the tiny individual stakes
involved; they have become the real parties in interest in indirect purchaser litigation.
Certainly, indirect purchaser classes almost always bring follow-on claims where the
government has already established liability in criminal prosecutions, because the
government’s victory greatly reduces their litigation burden. But it does not follow that
direct purchasers are less likely to sue. Direct purchasers are undeniably better situated
actually to prove the overcharge.110 More important, under the present regime, they have
the right to sue for the full overcharge in federal court—an enormous incentive.
A common mistake is to assume that the direct purchasers have a lesser incentive
to sue because they are not really harmed by the overcharge. The Iowa Supreme Court
made just this error in recognizing indirect purchaser standing:
It is the indirect purchaser, not the direct purchaser, who is most
frequently injured. . . . The true incentive to enforce antitrust law lies with
the real victims, like those of the Class, who may lack a direct business
relationship with the antitrust violator. . . . Therefore, to facilitate
enforcement of the policies behind the Iowa Competition Law, indirect
Benston, supra note xx, at 234-35, 245-46.
purchasers, the real victims, must be authorized to bring a cause of action
in state court.111
But the incentive to sue under Illinois Brick is unaffected by whether the direct purchaser
is a “real victim.” Under the Illinois Brick regime, even if the direct purchaser passed on
the entire overcharge, even if the “damages” are a pure windfall, the full incentive
remains to recover them. The failure to sue represents an opportunity cost. Experience
shows that direct purchasers are not reluctant to sue. In recent years, direct purchasers
have sued in the vast majority of cases in which indirect purchasers have sued,112
Comes v. Microsoft Corp., 646 N.W.2d 440, 450 (Iowa 2002) (citations omitted).
During the last several years, most violations challenged by indirect purchasers have also been
challenged by direct purchasers. See, e.g., In re Lorazepam & Clorazepate Antitrust Litig., 202 F.R.D. 12
(D.D.C. 2001) (direct purchasers of certain prescription drugs) and In re Lorazepam & Clorazepate
Antitrust Litig., 205 F.R.D. 369 (D.D.C. 2002) (indirect purchasers); Paper Sys. Inc. v. Nippon Paper
Indus. Co., 281 F.3d 629 (7th Cir. 2002) (direct purchasers of thermal fax paper) and Derzon v. Appleton
Papers, Inc., 1998-2 TRADE CAS. (CCH) ¶ 72,300 (Wisc. Cir. Ct. 1998) (indirect purchasers); Dean Foods
Co. v. Eastman Chem. Co., NO. 00C3675, 2000 WL 1557915 (N.D. Ill. Oct 19, 2000) (direct purchasers of
sorbates) and Williams Foods, Inc. v. Eastman Chem. Co., 2001-2 TRADE CAS. (CCH) ¶ 73,414 (Kan. Dist.
Ct. 2001) (indirect purchasers); In re Vitamins Antitrust Litig., 209 F.R.D. 251 (D.D.C. 2002) (direct
purchasers of raw vitamins) and Ciardi v. F. Hoffmann-La Roche, Ltd., 762 N.E.2d 303 (Mass. 2002)
(indirect purchasers); In re Medical X-Ray Film Antitrust Litig., 1998-2 TRADE CAS. (CCH) ¶ 72,305
(E.D.N.Y. 1998) (direct purchasers of medical x-ray film) and In re Cal. Indirect Purchaser X-ray Film
Antitrust Litig., 1998-2 TRADE CAS. (CCH) ¶ 72,336 (Cal. Super. Ct. 1998) (indirect purchasers); In re Flat
Glass Antitrust Litig., 191 F.R.D. 472 (W.D. Pa. 1999) (direct purchasers of flat glass) and Bunker’s Glass
Co. v. Pilkington PLC, 1 CA-CV 01-0046, 2002 WL 1461739 (Ariz. App. June 28, 2002) (indirect
purchasers); In re Commercial Tissue Prods., 183 F.R.D. 589 (N.D. Fla. 1998) (direct purchasers of
particularly in cases following price-fixing prosecutions.113 In some instances, they have
sued where indirect purchasers have apparently not done so.114
commercial tissue products) and Peridot, Inc. v. Kimberly-Clark Corp., 2000-1 TRADE CAS. (CCH) ¶
72,816 (Minn. Dist. Ct. 2000) (indirect purchasers); In re High Fructose Corn Syrup Antitrust Litig., 295
F.3d 651 (7th Cir. 2002) (direct purchasers of corn syrup) and Holder v. Archer Daniels Midland Co.,
1999-1 TRADE CAS. (CCH) ¶ 72,442 (D.C. Super. Ct. 1998) (indirect purchasers); In re Cardizem CD
Antitrust Litig., 200 F.R.D. 297 (E.D. Mich. 2001) (direct purchasers of a prescription drug) and In re
Cardizem CD Antitrust Litig., 200 F.R.D. 326 (E.D. Mich. 2001) (indirect purchasers); In re Terazosin
Hydrochloride Antitrust Litig., 203 F.R.D. 551 (S.D. Fla. 2001) (direct purchasers of a prescription drug)
and In re Terazosin Hydrochloride Antitrust Litig., 160 F. Supp.2d 1365 (S.D. Fla. 2001) (indirect
purchasers); Florida v. Abbott Labs., Inc., 1993-1 TRADE CAS. (CCH) ¶ 70,214 (N.D. Fla.) (direct
purchasers of infant formula) and Free v. Abbott Labs., Inc., 176 F.3d 298 (5th Cir. 1999) (indirect
Many of the cases cited in note xx were private, follow-on actions brought after criminal
actions. See, e.g., United States v. Nippon Paper Indus. Co., 62 F. Supp.2d 173 (D. Mass. 1999) (thermal
fax paper conspiracy); Williams Foods, 2001 WL 1298887, at *1 (referring to guilty pleas in action against
sorbates manufacturers); Vitamins, 2002 WL 31011261, at *2 (referring to guilty pleas in action against
In some cases, direct, intermediate purchasers have sued, but there are no reported actions by
indirect purchasers; these cases also demonstrate that direct purchasers are willing to sue their suppliers.
See, e.g., Thomas & Thomas Rodmakers, Inc. v. Newport Adhesives and Composites, Inc., 209 F.R.D. 159
(C.D. Ca. 2002) (direct purchasers of carbon fiber and prepeg); In re Linerboard Antitrust Litig., 305 F.3d
145 (3d Cir. 2002) (direct purchasers of linerboard); In re Magnetic Audiotape Antitrust Litig., 2001-1
TRADE CAS. (CCH) ¶ 73,306 (S.D.N.Y. 2001) (direct purchasers of magnetic audiotape); In re Antitrust
Litig., 1995-2 TRADE CAS. (CCH) ¶ 71,184 (E.D. Pa. 1995) (direct purchasers of plastic tableware).
The exception to this pattern appears to be the Microsoft follow-on litigation. As
we have seen, indirect purchasers have filed putative class actions against Microsoft
wherever state law gives them standing to sue. But OEMs such as Compaq and Gateway
have not yet sued Microsoft for damages.115 From this, the Iowa supreme court inferred
that OEMs have not sued for fear of retaliation that would disrupt a unique source of
supply.116 Experience demonstrates, however, that direct purchasers are generally willing
to sue, so this concern is apparently only a factor in exceptional cases. Microsoft may be
one of those cases—after all, the fear of retaliation is greater where the purchaser has no
feasible alternative supply sources.
Nevertheless, it seems unlikely that computer manufacturers have not sued
Microsoft because of fear of reprisals. Microsoft frequently deals with firms, like Apple,
that have sued it.117 And courts may require it to do so, as Judge Motz did recently in the
Sun Microsystems litigation.118 OEMs may be waiting for a final resolution of the
government litigation. More likely, though, OEMs have not sued because they recognize
that proving that Microsoft’s illegal conduct resulted in an overcharge to them will be
extremely difficult. As we argue elsewhere,119 the findings and conclusions in the
government case suggest that Microsoft may well have reduced prices to OEMs in its
Comes, 646 N.W.2d at 450.
See, e.g., Apple Computer, Inc. v. Microsoft Corp., 35 F.3d 1435 (9th Cir. 1994).
In re Microsoft Corp. Antitrust Litig., 2002-2 TRADE CAS. (CCH) ¶ 73,910 (D. Md. 2002).
John E. Lopatka & William H. Page, Who Suffered Antitrust Injury in the Microsoft Case?, 69
GEO. WASH. L. REV. 829, 851-55 (2001).
drive against Netscape. Microsoft is an exclusionary practice case; those most obviously
harmed in the short run in such cases are rivals, such as Netscape and Sun Microsystems,
which have sued. One might object that if Microsoft did not overcharge, the indirect
purchasers would not have sued. But the indirect purchasers’ class counsel are more
likely to sue than individual firms where damages are questionable, because of their
superior leverage in forcing settlements, once a class action is certified.120
Entrepreneurial plaintiffs’ counsel may well have different attitudes toward risk in
B. Do Indirect Purchaser Suits Result in Overdeterrence?
We have shown that, under the present regime, if there is an illegal overcharge,
both direct and indirect purchasers will probably sue. It remains to be considered
whether this state of affairs is good or bad from the point of view of optimal deterrence.
More particularly, does it result in duplicative recoveries, and if it does, are duplicative
recoveries somehow desirable? Most supporters of indirect purchaser suits argue that
multiple liability is a chimera, a mere theoretical possibility that does not occur in
See generally, Jonathan R. Macey & Geoffrey P. Miller, The Plaintiffs' Attorneys Role in Class
Action and Derivative Litigation: Economic Analysis and Recommendations for Reform, 58 U. CHI. L.
REV. 1, 45 n. 10 (1991); John C. Coffee, Jr., The Regulation of Entrepreneurial Litigation: Balancing
Fairness and Efficiency in the Large Class Action, 54 U. CHI. L. REV. 877, 896-917 (1987); John C.
Coffee, Jr., Understanding the Plaintiff's Attorney: The Implications of Economic Theory for Private
Enforcement of Law Through Class and Derivative Actions, 86 COLUM. L. REV. 669 (1986). See also In re
Rhone-Poulenc Rorer Inc., 51 F.3d 1293, 1298-99 (7th Cir. 1995) (Posner, C.J.) (“Judge Friendly, who was
not given to hyperbole, called settlements induced by a small probability of an immense judgment in a class
action ‘blackmail settlements.’”) (citation omitted).
practice, or is at least an unfortunate by-product. As Kevin O’Connor, until recently a
prominent state enforcement official, has observed:
the recent experience with multistate indirect damage actions suggests that
[Illinois Brick’s] concerns were, at least in part, ill-founded. First,
defendants have not been subjected to multiple liability, at least not
beyond the treble damages allowable under federal law and many state
laws. The fear has been expressed that direct purchasers and each layer of
indirect purchasers all the way down to consumers would in theory collect
treble damages, or at least single damages, with potentially no offset for
passed-on overcharges. The recently resolved cases, negotiated at arms-
length by vast arrays of plaintiff and defendant lawyers, suggest such
concerns are significantly overblown. In Mylan, the defendants paid
somewhat less than single damages to the combined federal and state
plaintiffs. In the Vitamins litigation, the six settling defendants have paid
approximately two-and-a-half times a reasonable estimate of the single
damage overcharge when one combines the approximately $875 million in
criminal fines, the $1.1 billion in direct damages, and approximately $400
million in indirect damages.121
Thus, the argument is that, even though direct and indirect claimants have multiple
claims, the settlement process by which virtually all class action litigation is resolved
results in the imposition of no more than treble damages. Whether multiple liability
Kevin J. O’Connor, Is the Illinois Brick Wall Crumbling?, 15 ANTITRUST 34, 37 (Summer
really occurs in practice is an empirical question that we cannot resolve on the present
record. The evidence in the quoted passage is not persuasive, because it assumes the
validity of “reasonable” estimates of damages, which may be quite different from actual
damages that a court would find in litigation. Settlements are rarely made for the full
amount of the plaintiff’s predicted damages; the amount must normally be discounted to
reflect the possibility of a lower award or a defense victory.122
But assume that multiple liability really does not occur. That is, assume that a
single plaintiff with the full right to sue for the estimated overcharge would settle for the
identical amount as would all of the various classes of direct and indirect purchasers
under the present regime. In that case, indirect purchaser suits still cannot be justified on
the grounds of optimal deterrence, because they only impose the same penalty as a pure
Illinois Brick regime, but at a much higher cost. O’Connor argues that the Illinois Brick
court exaggerated the costs of apportionment:
Apportioning damages precisely among many layers of potential direct
and indirect purchasers surely is a daunting task if carried to a high level
of precision. However, most states have not had a sufficient number of
decided cases to develop this area of law. Moreover, most indirect
purchaser statutes and parallel case law were adopted against a backdrop
of procedural standards better suited to other types of litigation. Now that
a critical mass of states have adopted in some fashion an indirect
See FEDERAL JUDICIAL CTR., MANUAL FOR COMPLEX LITIGATION (THIRD) § 30.42
(1995); RICHARD A. POSNER, ECONOMIC ANALYSIS OF LAW 568-69 (6th ed. 2003).
purchaser right of action, it is likely that procedural enhancements will
occur that will make such litigation fairer and more manageable.123
But this passage acknowledges that multistate, multijurisdictional direct and indirect
purchaser litigation, as it stands now, is extremely costly. Andrew Gavil has also
provided extensive evidence of these intractable problems.124 It may be that procedural
enhancements will allow a more precise allocation of the settlement fund. Indeed, Gavil
has proposed federal legislation that would facilitate removal and consolidation of the
litigation in a single court. But even conceding the possibility of formal or informal
methods of streamlining indirect purchaser litigation, it will always be far more costly
than a regime of pure direct purchaser litigation as a means of imposing a penalty on
offenders equal to treble the overcharge. There is an irreducible complexity in the
process of multilevel class actions that will inevitably impose greater costs.
We showed in Part III that the costs of apportionment cannot be justified on the
theory that they compensate consumers. If indirect purchaser litigation does not result in
duplicative recoveries, then the costs of apportionment cannot be justified on a theory of
deterrence either, because the only function of apportionment would be to magnify the
costs of imposing the same deterrent penalty—treble the overcharge. Consequently,
indirect purchaser litigation can only be justified on grounds of deterrence if it does result
in duplicative recovery, and if duplicative recovery is for some reason desirable.
Id. at 37-38.
Andrew I. Gavil, Federal Judicial Power and the Challenges of Multijurisdictional Direct and
Indirect Purchaser Antitrust Litigation, 69 GEO. WASH. L. REV. 860, 887-901 (2001).
Of course, states are free to permit indirect purchaser suits as a means of imposing
much higher multiples of actual damages than the federal trebling factor.125 In ARC
America, the Supreme Court rejected the argument that, because state indirect purchaser
suits “might subject antitrust defendants to multiple liability, in contravention of the
‘express federal policy’ condemning multiple liability,” they should be preempted.126
Illinois Brick, the Court reasoned, is only about federal policy, and does not preclude
states from “impos[ing] liability over and above that authorized by federal law.”127 And
it seems clear that some states have attempted to do just that, authorizing any injured
plaintiff to recover the “full consideration” he paid for the goods, not just the provable
overcharge.128 One distinguished observer contends that “the duplicative recovery
Awarding a multiple of an overcharge in an antitrust case can be viewed as the imposition of
punitive damages for a statutory tort, and the Supreme Court has emphasized that state “legislatures enjoy
broad discretion in authorizing and limiting permissible punitive damages awards.” Cooper Indus., Inc. v.
Leatherman Tool Group, Inc., 532 U.S. 424, 433 (2001). Further, states are not constitutionally bound to
use punitive damages only “to compensate for the underdeterrence of unlawful behavior that will result
from a defendant’s evasion of liability.” Id. at 438.
California v. ARC America Corp., 490 U.S. 93, 105 (1989).
See, e.g., KAN. STAT. ANN., § 50-115 (2001) (“Any person injured or damaged by any such
arrangement, contract, agreement, trust or combination . . . may sue for and recover . . . the full
consideration or sum paid by such person for any goods, wares, merchandise and articles included in or
advanced or controlled in price by such combination, or the full amount of money borrowed.”).
situation is serious and is likely to get worse as more ‘full consideration’ and ‘automatic
damage penalty’ statutes make their way through more state houses.”129
Such a regime can be justified under a system of rational deterrence only if the
federal trebling factor is seriously inadequate. (Of course, it could be justified on any
number of political factors that have nothing to do with rational deterrence.) The
economic function of trebling is primarily to take into account a probability of detection
and punishment less than one.130 If probability in fact is much less than one-third—and
there is some evidence for that conclusion, in the case of price fixing131—treble damages
could systematically underdeter. Whether the trebling factor is too low is another
empirical question. But we are skeptical that the sum of all federal penalties for illegal
antitrust overcharges is suboptimal. Civil liability in the form of treble damages is not
the only penalty for price fixing. Criminal antitrust penalties are available and, as we
noted earlier, actually precede a high percentage of indirect purchaser actions. The fines
Donald I. Baker, Hitting the Potholes on the Illinois Brick Road, 17 ANTITRUST 14, 17 (Fall
POSNER, supra note xx, at 269 (2001).
Peter G. Bryant & E Woodrow Eckard, Jr., Price Fixing: The Probability of Getting Caught,
73 REV. ECON. & STAT. 531 (1991). See also Michael Kent Block et al., The Deterrent Effect of Antitrust
Enforcement, 89 J. POL. ECON. 429 (1981); Sentencing Options: Hearing Before U.S. Sentencing
Commission (July 15, 1986) (statement of Douglas H. Ginsburg, Asst. Att'y Gen., Antitrust Div., U.S.
Dep't of Justice) at 5 ("the probability of detection of price-fixing generally is one in ten"), quoted in Anne
Marie Herron, Comment, Antitrust Sentencing Guideline: Deterring Crime by Clarifying the Volume of
Commerce Muddle, 51 EMORY L.J. 929, 956 (2002).
imposed for price fixing have skyrocketed in recent years.132 In light of a more expansive
corporate amnesty policy that increases the probability of uncovering concealable
antitrust violations, and hence reduces the magnitude of the appropriate fine, the ceilings
today may well be high enough that the optimal penalty can be imposed through criminal
sanctions alone.133 Moreover, the FTC has successfully obtained disgorgement of profits
as an antitrust remedy.134 A rational actor will aggregate the expected punishments from
all sources and discount them by the probability of their imposition.135
Even if an optimal penalty requires a multiple of the overcharge higher than three,
it would not justify the present system of duplicative recoveries, which is far more costly
than available alternatives. For reasons set out above, direct purchaser enforcement is
more efficient than combined direct and indirect purchaser enforcement. Consequently, a
preferable response would be to increase the damage multiple for direct purchasers;
perhaps better yet, ceilings on criminal sanctions could be raised further, or expenditures
See Donald C. Klawiter, After the Deluge: The Powerful Effect of Substantial Criminal Fines,
Imprisonment and Other Penalties in the Age of International Cartels, 69 GEO. WASH. L. REV. 745, 747-56
See Kobayashi, supra note xx, at 716 (“[T]he recent increase in fines may have resulted in
higher-than-optimal fines.”). But see POSNER, supra note xx, at 271 (arguing that the maximum criminal
fines and the damages remedy, “considered either separately or in combination, are inadequate”).
See FTC v. Mylan Labs., Inc., 62 F. Supp.2d 25, 37 (D.D.C. 1999) (holding that the FTC is
entitled to recover disgorgement under § 13(b) of the FTC Act as a remedy for unfair competition); In re
First Databank Antitrust Litig., 209 F. Supp.2d 96 (D.D.C. 2002) (referring to the settlement of an FTC
complaint that included $16 million for disgorgement of unlawful monopoly profits).
See POSNER, supra note xx, at 47.
on public enforcement could be increased, thereby increasing the probability of
punishment. The reallocation of government enforcement resources would not require
We have argued that indirect purchaser class actions—touted as a means of
protecting the consumer interest—have failed to benefit consumers, either by providing
them with significant compensation or by more efficiently deterring monopolization or
price fixing. Instead, they have generated costly and duplicative litigation in numerous
state courts. The unhappy performance of private enforcement today in achieving
optimal deterrence calls for some kind of reform. As the foregoing analysis suggests, we
would welcome federal preemption of state indirect purchaser suits. But additional
changes ought to be considered, and in the context of this essay we can only mention a
A major reform would be to place the primary responsibility to enforce the law
against price fixing on the federal agencies, conferring residual authority on direct
purchasers to sue only when the federal agencies choose not to act. The federal
government would in effect have a right of first refusal.136 The United States would be
allowed to recover damages in a civil action, even if it suffered no loss in its own
property. The shaky statutory authority of the FTC in obtaining disgorgement of ill-
gotten monopoly profits would be legislatively ratified. And the damage multiple for
both agencies would be increased, say to 10, and expressed as a ceiling, with the multiple
See POSNER, supra note xx, at 274.
in any given case set by reference to the probability of detection ex ante under the
Although the political resistance to eliminating indirect purchasers’ right to sue is
strong, it may weaken if coupled with the prospect of more likely and larger recoveries
by the federal government. Those agencies would be free to distribute the recoveries in
any way they saw fit, including some kind of payment to purchasers if the costs of
distribution were not prohibitive. Further, any fear that the federal agencies would ignore
violations should be assuaged by the conditional authority to sue retained by direct
purchasers. Most important, this kind of enforcement scheme offers the potential of
achieving optimal deterrence in a single action, rather than the current reality of
haphazard enforcement in a multiplicity of actions.