APPLE V. PEPPER: RATIONALIZING ANTITRUST’S INDIRECT PURCHASER RULE

APPLE V. PEPPER: RATIONALIZING ANTITRUST’S INDIRECT PURCHASER RULE

Introduction

In Apple Inc. v. Pepper, the Supreme Court held that consumers who allegedly paid too much for apps sold on Apple’s App Store because of an antitrust violation could sue Apple for damages because they were “di­rect purchasers.” 1 Apple Inc. v. Pepper, 139 S. Ct. 1514, 1519 (2019) (internal quotation marks omitted) (quoting Ill. Brick Co. v. Illinois, 431 U.S. 720, 745–46 (1977)). The decision sidesteps most of the bizarre complexities that have resulted from the Supreme Court’s 1977 decision in Illinois Brick Co. v. Illinois, which held that indirect purchasers could not sue for passed- on overcharge injuries under federal antitrust laws. 2 431 U.S. at 746–47. The decision was a logical complement to the Court’s previous decision in Hanover Shoe v. United Shoe Mach. Corp., 392 U.S. 481, 488 & n.6 (1968), which held that a defendant could not have its damages reduced by the amount that the plaintiff passed on to its own customers. The Illinois Brick decision and its complex aftermath are analyzed in 2A Phillip E. Areeda, Herbert Hovenkamp, Rodger D. Blair & Christine Piette Durrance, Antitrust Law ¶ 346 (4th ed. 2014) [hereinafter 2A Areeda et al., Antitrust Law].

I. The Indirect Purchaser Rule Under Illinois Brick

The Illinois Brick decision itself was factually straightforward, al­though its reasoning was controversial. Makers of construction blocks fixed their prices and sold them to contractors who built buildings for the State of Illinois. 3 Ill. Brick, 431 U.S. at 726. No one seriously doubted that when a cartel sells some­thing at a higher price to an intermediary, that intermediary firm will pass on at least part of the price increase to its own customers, and so on down the line. The legal question was how the antitrust laws should rec­ognize that fact in determining the right to collect damages. 4 Id.

Purchaser damages in cartel and monopolization cases are ordinarily measured by the amount of the price increase, or “overcharge.” Roughly speaking, this is the difference between the actual price that the defendant(s) charged and the price that would have prevailed had there not been any price fixing. As Justice Holmes put it in the first Supreme Court decision to confront the issue, the measure of damages should be “the difference between the price paid and the market or fair price that the city would have had to pay under natural conditions had the combi­nation been out of the way, together with an attorney’s fee.” 5 Chattanooga Foundry & Pipe Works v. City of Atlanta, 203 U.S. 390, 396 (1906). On the measurement of overcharge damages in cartel cases, see Herbert Hovenkamp, Federal Antitrust Policy: The Law of Competition and Its Practice (6th ed. forthcoming 2020) (manuscript § 17.5) (on file with the Columbia Law Review) [hereinafter Hovenkamp, Federal Antitrust Policy].

Illinois Brick held that the first purchaser in line, or the direct purchaser, should obtain the entire overcharge as damages, without any reduction for the amount that it had passed on to purchasers beneath it in the distribution chain. 6 Ill. Brick, 431 U.S. at 745–46. Accordingly, indirect purchasers would not be able to claim any damages, since they had already been recovered in full by the direct purchaser. 7 Id. This rule, the Court noted, was not one of “stand­ing” but rather of entitlement to damages. 8 Id. at 728 n.7. Specifically, the Court stated: “Because we find Hanover Shoe dispositive here, we do not address the standing issue, except to note, as did the Court of Appeals below, that the question of which persons have been injured by an illegal overcharge for purposes of § 4 [of the Clayton Act] is analytically distinct from the question of which persons have sustained injuries too remote to give them standing to sue for damages under § 4.” Id. (citation omitted).

In Illinois Brick, the Supreme Court recognized some qualifications, indicating that its concern  was  with  the  computation  of  passed-on  dam­ages  rather  than  standing  to  sue. 9 See id. at 737. For example, purchasers under a pre­ex­isting contract that fixed both the quantity and markup could obtain damages, for in such cases the entire overcharge would be passed on. 10 See id. at 736; see also Hanover Shoe v. United Shoe Mach. Corp., 393 U.S. 481, 494 (1968) (recognizing the same exception); 2A Areeda et al., Antitrust Law, supra note 2, ¶ 346e (discussing the exception for preexisting fixed-cost, fixed-quantity contracts). Other exceptions were not mentioned by the Supreme Court but flowed naturally from the Court’s focus on passed-on damages. For example, the lower courts have held that Illinois Brick does not preclude an action for an injunction because no calculation of passed-on damages is necessary. 11 See, e.g., In re Warfarin Sodium Antitrust Litig., 214 F.3d 395, 402 (3d Cir. 2000); see also 2A Areeda et al., Antitrust Law, supra note 2, ¶ 346d. In its noneconomic approach based on proximate cause, the Apple dissenters doubted standing for indirect purchasers seeking only an injunction. 12 Apple Inc. v. Pepper, 139 S. Ct. 1514, 1525–27, 1527 n.1 (2019) (Gorsuch, J., dissenting); see also infra notes 69–75 and accompanying text.

The Supreme Court also held in Kansas v. UtiliCorp United, Inc. that the indirect purchaser rule applied even when the direct purchaser pass­ed on 100% of its overcharge. 13 497 U.S. 199, 203–05 (1990); see also Simon v. KeySpan Corp., 694 F.3d 196, 198 (2d Cir. 2012), cert. denied, 133 S. Ct. 1998 (2013) (mem.) (holding that a retail electric consumer complaining of price fixing in the wholesale electricity market was an indirect purchaser). The direct purchaser in that case was a price-regulated utility that operated under a state statute that computed rates as including a one-to-one pass-through of all variable costs, which in­cluded fuel costs. 14 See UtiliCorp, 497 U.S. at 205. Under cost-of-service ratemaking, a public utility generally recovers a fair rate of return on its fixed costs plus “pass-through” of variable costs. See Stephen Breyer, Regulation and Its Reform 36–59 (1982); 1 Alfred E. Kahn, The Economics of Regulation 26–57 (1971). Indeed, as the dissenters pointed out, the state regulatory provision required complete pass-through of the overcharge, which appeared as a surcharge on each customer’s utility bill. 15 UtiliCorp, 497 U.S. at 222–23 (White, J., dissenting). Neverthe­less, the Court concluded the utility itself was the direct purchaser and the customers, who paid the entire overcharge, were merely indirect pur­chasers. 16 Id. at 207 (majority opinion) (finding that “the consumers in this case have the status of indirect purchasers” and therefore “any antitrust claim against the defendants is not for them, but for the utilities to assert”). The Utilicorp decision rejected a carefully reasoned opinion by Judge Richard Posner in a related case, who observed not only that there was perfect 100% pass-through, but also that there was very little output reduction resulting from the price fix because the retail elasticity of demand for power was very low. See Illinois ex rel. Hartigan v. Panhandle E. Pipe Line Co., 852 F.2d 891, 895–97 (7th Cir. 1988); cf. E. Air Lines, Inc. v. Atl. Richfield Co., 609 F.2d 497, 499 (Temp. Emer. Ct. App. 1979) (holding that “any functional equivalent of a cost-plus contract exception to the Hanover Shoe ban against defensive use of passing on must be one that is already in existence, in that its impact on pricing decisions must be known in advance”). In that case, the direct purchasing utility was not injured by the cartel. By statute, it passed  on  100%  of  its  overcharge. 17 See UtiliCorp, 497 U.S. at 205. Further,  it  suf­fered  little  or  no  decline  in  output. 18 Cf. Hartigan, 852 F.2d at 896 (noting, in an analogous rate regulated environment, that “an increase in . . . rates was bound to increase the firm’s revenues”). The Supreme Court effectively gave the only antitrust cause of action to a firm that had not been injured.

About half of the states have either amended or interpreted their own antitrust statutes so as to permit indirect purchaser damages actions, and the Supreme Court has approved such provisions. 19 California v. ARC Am. Corp., 490 U.S. 93, 105–06 (1989). For a comprehensive survey of state law indirect purchaser provisions, see 14 Phillip E. Areeda & Herbert Hovenkamp, Antitrust Law ¶ 2412d (4th ed. 2019 & Supp. 2019). Most of the litigation progress we have made in determining how antitrust damages are passed through a distribution chain has undoubtedly occurred in the context of this state antitrust litigation.

II. Complexities and Ambiguities in the Distribution Chain

The facts of Apple v. Pepper were more complex than those of Illinois Brick, and illustrate some of the difficulties that the courts have faced in applying the indirect purchaser rule. The owners of Apple’s iPhones are required to purchase programs, or “apps,” on Apple’s App Store, which is itself an app that can be found on the iPhone screen. 20 Brief for Respondents at 4, Apple Inc. v. Pepper, 139 S. Ct. 1514 (2019) (No. 17-204), 2018 WL 4659225. Apple’s App Store is thus a bottleneck through which the apps’ producers must pass if they are to reach Apple iPhone users. In this consumer class action, iPhone owners accused Apple of monopolizing the market for Apple iPhone app sales, both through this exclusivity requirement and by charg­ing app producers a very high commission of 30% of the app’s sale price. 21 Apple, 139 S. Ct. at 1518–19. The plaintiffs alleged the margin was extremely high in relation to cost because Apple does very little in its role of distributor of these apps sold by others. 22 See Second Amended Consolidated Class Action Complaint ¶¶ 40–41, In re Apple iPhone Antitrust Litig., No. 11-cv-06714-YGR, 2013 WL 6253147 (N.D. Cal. Dec. 2, 2013), 2013 WL 6387366.

In this case, unlike Illinois Brick, the customers paid their money and purchased directly from Apple. “There is no intermediary in the distribu­tion chain between Apple and the consumer.” 23 Apple, 139 S. Ct. at 1521. The Supreme Court ma­jority found this to be “dispositive” of the result—namely, that the plaintiffs paid their money directly to the defendants, and this entitled them to be treated as direct purchasers. 24 Id.

This would all seem clear enough were it not for the fact that in the forty-year history of Illinois Brick jurisprudence other courts had charac­terized this same problem differently. A case in point is Campos v. Ticketmaster, 25 140 F.3d 1166 (8th Cir. 1998). which was very similar to Apple, although neither the district court nor either of the Supreme Court opinions in Apple  dis­cussed  it.  The  Ninth  Circuit  decision,  however,  addressed  it  at  some length. 26 For the Ninth Circuit’s discussion, see In re Apple iPhone Antitrust Litig., 846 F.3d 313, 323 (9th Cir. 2017). For other decisions raising similar issues to Apple, see In re ATM Fee Antitrust Litig., 686 F.3d 741, 750 (9th Cir. 2012), cert. denied sub nom. Brennan v. Concord EFS, Inc., 134 S. Ct. 257 (2013) (mem.) (holding that ATM card users were indirect purchasers, notwithstanding that ATM machine operators, who allegedly fixed transaction fee prices, charged these to banks, who passed them on 100% to the card users); McCarthy v. Recordex Serv., 80 F.3d 842, 848–55 (3d Cir. 1996), cert. denied, 519 U.S. 825 (1996) (mem.) (holding that lawyers who purchased copies of hospital records on behalf of clients were the direct purchasers, even though they passed on the entire price to the clients, who were the plaintiffs); Hyland v. Homeservices of Am., Inc., No. 3:05-cv-612-R, 2006 WL 3498569, at *2 (W.D. Ky. Dec. 1, 2006) (finding that Illinois Brick barred a challenge to real estate broker commission fixing brought by home buyers); Petition for Writ of Certiorari at 44, app. A at 5a–6a, Durkin v. Major League Baseball, 519 U.S. 825 (1996) (No. 95-208) (including the Third Circuit opinion, which held that cable television subscribers were indirect purchasers with respect to their claim of price-fixing among sports teams, cable channels, and cable programmers when these prices were charged to local cable television companies and then passed on to subscribers). In Ticketmaster, the defendant, Ticketmaster, was an online event ticket retailer accused of monopolizing event ticket sales. 27 Ticketmaster, 140 F.3d at 1168. The purchas­ers were event goers who bought their tickets directly from Ticketmaster, paying high processing and handling fees. 28 Id. Ticketmaster itself set the final ticket price. 29 See id. at 1169, 1171–72. The court held, however, that the concert promoters were direct purchasers of “ticket distribution services” from Ticketmas­ter, 30 Id. at 1171. and that the actual ticket buyers were only indirect purchasers of these services. 31 Id. This was true, the court reasoned, because there was “an anteced­ent transaction between the monopolist and another, independ­ent purchaser” who was in a position to absorb part or all of the over­charge. 32 Id. at 1169. Under the plaintiff’s theory of the case, the concert promoters presumedly behaved as most intermediaries behaved: They passed on Ticketmaster’s high markups by charging higher wholesale prices for the tickets. The final purchasers, who actually used the tickets, were the only ones unable to pass anything on.

Ticketmaster produces the perverse result that the buyer who is able to pass on all or part of the overcharge is treated as the direct purchaser and can sue for damages, while the buyer who is at the end of the line and must absorb the entire overcharge passed on to it has no damages claim, even though it purchased directly from the defendant. As in Apple, the Ticketmaster consumers paid the violator directly, but the Eighth Circuit found them to be “indirect” purchasers. The Apple decision effectively overruled Ticketmaster sub silentio. The Apple majority created an apparently categorical rule that whoever pays the money directly to the defendant should be counted as the direct purchaser. 33 See supra notes 22–23 and accompanying text.

III. The Problem of Passed-on Damages

Illinois Brick’s indirect purchaser rule was problematic from the beginning for a number of reasons that the Apple majority did not address and certainly did not fix. First, it was plainly inconsistent with the antitrust damages statute, which gives an action to “any person who shall be injured in his business or property” by an antitrust violation. 34 15 U.S.C. § 15(a) (2012) reads:
“[A]ny person who shall be injured in his business or property by reason of anything forbidden in the antitrust laws may sue therefor in any district court of the United States in the district in which the defendant resides or is found or has an agent, without respect to the amount in controversy, and shall recover threefold the damages by him sustained, and the cost of suit, including a reasonable attorney’s fee . . . .”
That hard­ly sounds like a limitation to direct purchasers.

A. Computing Pass-on

Second, the Illinois Brick Court exaggerated the difficulty of “tracing” indirect purchaser damages. Computing how damages are passed on at each stage of a distribution chain requires “incidence” anal­ysis, which economists use to compute how a tax or other expenditure is passed along through the economy. In general, the more elastic a firm’s demand, the more of an overcharge it will shift to others. By contrast, the more inelastic its demand the more it will have to absorb. 35 See 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, 615–21 (1979). For a critique, see Robert Cooter, Passing on the Monopoly Overcharge: A Further Comment on Economic Theory, 129 U. Pa. L. Rev. 1523 (1981). Indeed, the ear­liest explicit Supreme Court applications of economic analysis to a legal problem occurred in the 1920s and 1930s, and involved computa­tion of how a tax might be passed on from one entity to another. 36 E.g., Indian Motocycle Co. v. United States, 283 U.S. 570, 581–82 & n.1 (1931) (Stone, J., dissenting) (applying economic analysis to the incidence of a sales tax). The economist who contributed nearly all of the citations in these cases was Edwin R.A. Seligman. See generally Edwin R.A. Seligman, The Shifting and Incidence of Taxation (5th ed., rev. ed., Columbia Univ. Press 1926) (1899); Edwin R.A. Seligman, The Taxation of Corporations. III., 5 Pol. Sci. Q. 636, 671–75 (1890); Edwin R.A. Seligman, The General Property Tax, 5 Pol. Sci. Q. 24 (1890); Edwin R.A. Seligman, On the Shifting and Incidence of Taxation, 7 Publications Am. Econ. Ass’n 7, 119 (1892). On the use of early marginalist economics to deal with problems of tax incidence, as well as the courts’ numerous citations to Seligman, see Herbert Hovenkamp, The Opening of American Law: Neoclassical Legal Thought, 1870–1970, at 106–22, 367 n.59 (2015) [hereinafter Hovenkamp, The Opening of American Law]. While the Illinois Brick decision acknowledged its relevance, 37 Ill. Brick Co. v. Illinois, 431 U.S. 720, 741­ n.25 (1977). later decisions held that the effect of Illinois Brick was to forbid the use of incidence anal­ysis in the estimation of passed-on antitrust damages. 38 See, e.g., In re Brand Name Prescription Drugs Antitrust Litig., 123 F.3d 599, 606 (7th Cir. 1997) (stating that Illinois Brick bars the use of incidence analysis in the computation of passed-on damages); In re Beef Indus. Antitrust Litig., 600 F.2d 1148, 1163 n.19, 1166 (5th Cir. 1979) (noting the Illinois Brick Court’s disapproval of the “speculative nature” of incidence analysis); see also Herbert Hovenkamp, Commentary, The Indirect-Purchaser Rule and Cost-Plus Sales, 103 Harv. L. Rev. 1717, 1721 (1990) (“[T]he Illinois Brick exception for pre-existing, fixed-markup, fixed-quantity contracts should be broadened to cover all ‘cost-plus’ contracts.”); Charles E. McLure, Jr., Incidence Analysis and the Supreme Court: Examination of Four Cases from the 1980 Term, 1 Sup. Ct. Econ. Rev. 69, 72–84 (1982) (explaining the basic principles of incidence analysis).

To illustrate the problem, if a farmer is taxed on wheat, she may absorb part of that tax but pass a portion on to the wholesale bread baker, who may in turn pass some of it on to the grocer, who will then pass part of it on to the consumer. The technical quantification of pass-on is quite demanding, requiring determination of the elasticities  of  supply  and  demand  facing  each  individual  firm  in  the  distribution chain. 39 See Landes & Posner, supra note 34, at 620. A further complication is that the calculations are very sensitive to the distribution of fixed and variable costs. Variable costs generally show up in the market price and are passed on; fixed costs generally are not. 40 See, e.g., Cooter, supra note 34, at 1523–24.

However, in most cases antitrust experts can assess damages without computing pass-on. 41 See 2A Areeda et al., Antitrust Law, supra note 2, ¶ 346k. For example, under the “before-and-after” method, the expert looks at a market just prior to the violation, just subsequent, or both, comparing prices during the violation and nonviolation peri­ods. 42 See Hovenkamp, Federal Antitrust Policy, supra note 5, § 17.5b2. The “yardstick” method of estimating damages compares prices in the violation market with those in a similar “yardstick” market where the violation is thought not to be occurring. 43 Id. § 17.5b1. Illustrating the methodology is Greenhaw v. Lubbock Cty. Beverage Ass’n, 721 F.2d 1019, 1026 (5th Cir. 1983). In Greenhaw, the court stated, “By comparing these ‘should have been’ prices to the actual prices charged in 1975, [the expert] was able to estimate the amount of overcharge resulting from defendants’ price-fixing activity.” Id. In both cases, one does not need to estimate pass-on at each stage. Rather, the expert compares the price in the cartel market with the price in some reasonably similar com­parator market that was not affected by the cartel. 44 See Hovenkamp, Federal Antitrust Policy, supra note 5, § 17.5.

For example, suppose that dairies are fixing prices, raising the wholesale price of milk from $2.00 to $3.00 per gallon. The milk is sold to distributors, who sell it to grocers, who finally sell it to consumers. Com­puting the amount passed on by each of these would be difficult. But suppose we can identify a “yardstick” market, similar to the cartel mar­ket but without price fixing. In that market the dairies’ wholesale price is $2.00, as it should be. Ignoring the distributors and grocers, who are not parties, we also observe that retail prices in this competitive mar­ket are $2.90, while retail prices in the cartel market are $3.75. If consum­ers sue, we do not need to know how much of the overcharge was passed on at each level. These numbers tell us that although not all of the over­charge was passed on, consumers in the cartel market paid 85 cents more for their milk. This represents their overcharge, regardless of how much was absorbed at each link along the way. 45 Id. § 17.5b1. For a situation where computation of pass-on was not required, see Drug Mart Pharmacy Corp. v. Am. Home Prods. Corp., No. 93-CV-5148 (ILG), 2002 WL 31528625, at *9 (E.D.N.Y. Aug. 21, 2002). The plaintiff pharmacies were indirect purchasers of pharmaceuticals complaining under state antitrust law that a charge-back system effectively required them to pay more for pharmaceuticals than did favored purchasers, such as HMOs. The plaintiffs’ claimed injury was mainly loss in their ability to compete with the favored purchasers. Id. at *1–3. In this case, overcharge damages could be computed by comparing the net price paid by the pharmacies with that paid by other purchasers at the same distribution level; in any event, diversion of sales rather than the overcharge better captures the plaintiffs’ injuries.

Courts have approved these incidence-avoiding methodologies in cases involving state antitrust laws that permit indirect purchaser recov­eries. 46 E.g., In re Flonase Antitrust Litig., 284 F.R.D. 207, 232–33 (E.D. Pa. 2012) (approving the methodology); In re Terazosin Hydrochloride Antitrust Litig., 220 F.R.D. 672, 699 (S.D. Fla. 2004) (same); see also In re Class 8 Transmission Indirect Purchaser Antitrust Litig., 140 F. Supp. 3d 339, 349–54 (D. Del. 2015) (evaluating incidence-avoiding methodologies, but finding that plaintiffs failed to satisfy their burden), vacated on other grounds, 679 F. App’x 135 (3d Cir. 2017); cf. Melnick v. Microsoft Corp., No. CV-99-709, CV-99-752, 2001 WL 1012261, at *16 (Me. Super. Ct. Aug. 24, 2001) (expressing doubt about the methodology when the defendant sold at a wide variety of prices). As one court observed, “the before-and-after ‘yardstick’ methodol­ogy has been accepted by courts as a means to measuring damages in both indirect and direct purchaser antitrust actions.” 47 In re Flonase, 284 F.R.D. at 232. For an expression of skepticism, see Thomas A. Lambert, Tweaking Antitrust’s Business Model, 85 Tex. L. Rev. 153, 187 (2006) (reviewing Herbert Hovenkamp, The Antitrust Enterprise: Principle and Execution (2005)) (arguing that the indirect purchaser rule “provides a closer to optimal level of deterrence” than incidence-avoiding methodologies). One district court noted and approved the plaintiff expert’s use in what it characterized as:

[A] “bottom across” approach which obviates the complexities Defendants cite in their “top down” approach. “Bottom across” means that the overcharge is determined by examining the price differential between the generic and the brand drug at the retail level only. Thus, there will be no need to review “pass-through” variations. 48 In re Cardizem CD Antitrust Litig., 200 F.R.D. 326, 344 (E.D. Mich. 2001); see also In re Asacol Antitrust Litig., No. 15-12730-DJC, 2017 WL 53695, at *3 (D. Mass. Jan. 4, 2017) (observing but not passing judgment on the plaintiff’s proposed methodology for estimating passed-on damages). The In re Asacol court stated:
“According to the EPPs [“End Payor Plaintiffs,” or the last purchasers in line], they are not calculating “injury or damages by relying on a top-down vertical ‘pass-through’ economic analysis[.]” Rather, “EPPs will use a ‘yardstick’ damages and impact methodology to examine the retail price of the drugs EPPs were forced to purchase” in comparison “to the forecasted price (and volume) of the drug that should have been available but-for Defendants’ misconduct[.]”[].
Id. (alterations in original) (citations omitted). On the use of “comparator” methods rather than estimating passed-on damages, see Ewa Mendys-Komphorst, Assessing Damages in Antitrust Actions, Global Arbitration Rev. (Nov. 29, 2018), https:// globalarbitrationreview.com/print_article/gar/chapter/1177446/assessing-damages-in-antitrust-actions?print=true [https://perma.cc/284J-JNQ7].

B. Who Is Injured, and How?

Third, if we were going to give the overcharge to a single set of buyers it should be the end users, not the direct purchasers. The end user is the only person in the distribution chain who is unable to pass anything on. 49 See 2A Areeda et al., Antitrust Law, supra note 2, ¶¶ 346a–c, k. The impact varies from one situation to another, but in many cases the largest losses are those absorbed by end users, and often they absorb the entire overcharge. Many intermediaries use markup for­mulas that are standardized across products. Consider the dairy example. Suppose that the milk retailer is Kroger, which routinely computes a 10% mark­up on its dairy products, depending on the extent of competition each item faces. If it pays $2.50 at wholesale in a competitive market it adds 10%, or 25¢. However, if it pays $3.50 it also adds 10%, which is now 35¢. Far from “absorbing” part of the overcharge, Kroger actually exacts a higher markup when the price of milk is fixed, reflecting its standard­ized percentage of the higher cartel price. The consumer gets hit even harder. However, the Supreme Court has applied the indirect purchaser rule even to situations, such as price regulated industries, where it is clear that pass-through is 100%. 50 E.g., Kansas v. UtiliCorp United, Inc., 497 U.S. 199 (1990); see supra text accompanying notes 12–17.

This is not to say that Kroger in the above example is not injured by the price fixing. Its injury results from lost volume rather than the over­charge. Under collusion, sales volume goes down. 51 See Peter Asch & J. J. Seneca, Is Collusion Profitable?, 58 Rev. Econ. & Stat. 1, 4–5 (1976) (finding that the presence of collusive behavior is negatively associated with firm profitability, on which the demand for the firm’s products exerts an important effect). This suggests an important principle: The real injury to direct purchasers and other intermediaries in the distribution chain is not from the overcharge at all; rather, it is from the loss of sales volume. As a result, the “overcharge” is not even the theoretically correct measure of damages for an intermedi­ary who passes on at least part of an overcharge.

If we really wanted to award damages based on injury to a buyer’s business or property, as the statute requires, we would compute damages as lost profits for all intermediaries, including the direct purchaser. Only the final purchaser, or consumer, should obtain damages for the amount of the overcharge passed on to it. Lost profit damages do not present problems of computing pass-on. Rather, they require information about the impact of the unlawful practice on both the volume of sales and the reseller’s margin. In response to a cartel, volume virtually always goes down. Margins can go down, up, or stay the same depending on the reseller’s markup practices or the intensity of competition that it faces. 52 See Jorge Padilla, Economist’s Note: Should Profit Margins Play a More Decisive Role in Horizontal Merger Control?, 9 J. Eur. Competition L. & Prac. 260, 261–62 (describing the relationship between profit margins and competition). The before-and-after and “yardstick” methodologies commonly used in cases involving business damages work here as well. 53 See, e.g., Bigelow v. RKO Radio Pictures, Inc., 327 U.S. 251, 257–58 (1946) (using the two damage methods in conjunction); Lehrman v. Gulf Oil Corp., 500 F.2d 659, 667 (5th Cir. 1974) (explaining methods of proving lost profits).

To be sure, in common with all commercial damages formulas, lost profit formulations impose some complexities, but these are not differ­ent in substance from those experienced by the victims of antitrust viola­tions that exclude rivals, where overcharge measures are not relevant. 54 On lost profit damages for exclusionary practices, see 2A Areeda et al., Antitrust Law, supra note 2, ¶ 397. Indeed, a wide variety of commercial legal violations that are subject to private enforcement compute damages as lost profits. 55 E.g., Mission Prod. Holdings, Inc. v. Tempnology, LLC, 139 S. Ct. 1652 (2019) (approving a lost-profit measure in a breach of trademark case); Rite-Hite Corp. v. Kelley Co., 56 F.3d 1538 (Fed. Cir. 1995) (same in a patent infringement case); Biotronik A.G. v. Conor Medsystems Ireland, Ltd., 11 N.E.3d 676 (N.Y. 2014) (same in a breach of contract case); Formosa Plastics Corp. USA v. Presidio Eng’rs & Contractors, Inc., 960 S.W.2d 41 (Tex. 1998) (same in a tort and breach of contract case).

Computing the lost profits that result from a cartel’s higher prices need not be any more complex than computing the lost profits resulting from, say, patent infringement or the breach of a supply contract. For example, the Patent Act permits recovery of lost profits attributable to an infringement, 56 35 U.S.C. § 284 (2012). which requires the fact finder to determine the harm caused by each infringing sale of something that could range from a full substitute for the plaintiff’s product 57 See, e.g., WesternGeco LLC v. ION Geophysical Corp., 138 S. Ct. 2129, 2135, 2139 (2018) (applying a patent damages provision to foreign sales of an infringer’s product that was substantially identical to patentee’s product). to a relatively small component. 58 See, for example, Yale Lock Mfg. Co. v. Sargent, concluding that the infringe­ment damages should equal:
“[T]he difference between his pecuniary condition after the infringement, and what his condition would have been if the infringement had not occurred, is to be measured, so far as his own sales of locks are concerned, by the difference between the money he would have realized from such sales if the infringement had not interfered with such monopoly, and the money he did realize from such sales. If such difference can be ascertained by proper and satisfactory evidence, it is a proper measure of damages.”
117 U.S. 536, 552–53 (1886).
If the patent is for an improvement, the fact finder must disaggregate that portion of the loss of sales attributable to the improvement from that portion attributable to noninfringing conduct. 59 As the Court in Garretson v. Clark, 111 U.S. 120 (1884), stated:
“When a patent is for an improvement, and not for an entirely new machine or contrivance, the patentee must show in what particulars his improvement has added to the usefulness of the machine or contrivance. He must separate its results distinctly from those of the other parts, so that the benefits derived from it may be distinctly seen and appreciated.”
Id. at 121.
There is no reason for thinking that computation of lost-profit damages from collusion would be more difficult than this.

While pass-on must be allocated individually for each successive link in the distribution chain, loss of output is often simpler because the same loss passes from one entity to the next. For example, if a cartel of wheat growers reduces output by 30%, wholesale bakers will bake 30% less bread and retailers will sell 30% fewer loaves. Of course, complexities in the distribution chain, including the ability of intermediaries to vary the proportions of their inputs, can complicate these numbers.

Guidelines from the European Commission to the national courts of member states permit purchaser damages to be computed by all of the methodologies discussed here. 60 Guidelines for National Courts on How to Estimate the Share of Overcharge Which Was Passed on to the Indirect Purchaser, 2019 O.J. (C 267) 4, 24 [hereinafter Guidelines for National Courts] (“When estimating the passing-on related price effect national courts may rely on different types of economic approaches to quantification . . . .”). Injuries for what the Guidelines describe as “price effects,” or overcharge, can be estimated using the conventional pass-on methodologies contemplated in the Illinois Brick decision, or else by comparing overcharges in the violation market with those in markets that are not affected. 61 Id. at 24–25. The Guidelines describe these as “comparator-based” methods, giving as examples:
“[1] price or margin data concerning this market before and/or after the infringement, usually referred to as the before-during-after approach;
[2] data concerning the same (product) market but in a different geographical area, or another product market that is considered to evolve in a similar manner to the market where the direct or indirect purchaser operates, usually referred to as the cross-sectional approach; or
[3] a combination of comparisons over time and comparisons across markets, usually referred to as the difference-in-differences approach.”
Id. at 25; see also Miguel de la Mano & Christopher Milde, Estimating the Pass-on Effect in Antitrust Damage Cases: Relative Strengths and Weaknesses of the “Comparator” Method vs. the “Pass-on Rate” Method (Apr. 30, 2019) (unpublished manuscript), https:// ssrn.com/abstract=3380657 (on file with the Columbia Law Review). For a more com­prehensive look at the problem in a wide variety of European actions, not limited to competition law, see generally Magnus Strand, The Passing-on Problem in Damages and Restitution Under EU Law (2017).
The Guidelines also permit damages to be based on what they call the “volume effect,” or lost profits from loss of sales. 62 Guidelines for National Courts, supra note 59, at 13. Just like § 4 of the Clayton Act, the EU Damages Directive authorizes national courts to award provable damages for injuries sustained, but does not specify a particular methodology. 63 Directive 2014/104/EU of the European Parliament and of the Council of 26 November 2014 on Certain Rules Governing Actions for Damages Under National Law for Infringement of the Competition Law Provisions of the Member States and of the European Union, 2014 O.J. (L 349) 1, 8–9 (stating that each member state has the responsibility to determine its own rules for qualifying the harm, such as the required proof and methods). However, the EU takes the more realistic approach, and one that is more faithful to the relevant provisions, of permitting damages to be estimated by any controllable method for which there is adequate evidentiary support.

C. Addressing Indirect Purchaser Harm: Expert Testimony

The Illinois Brick decision is apparently based on the premise that damages measurement in antitrust cases is exceptional and requires special and more categorical treatment than is used for other types of com­mercial injuries. That is hardly obvious. One possible distinction is the existence of treble damages, but Congress can always change that. As noted above, other types of injuries such as those that result from patent infringement pose complexities that are as great or greater. 64 See supra notes 55–58 and accompanying text.

Rather than being so categorical about the nature of damages, we should assess how damages operate in a particular case, and the varia­tions can be substantial. This means that judges must assess the reliability and applicability of expert testimony. Once a judge has made that assess­ment and admitted the testimony, the testimony still must be subjected to rigorous cross-examination at trial.

The Supreme Court’s Daubert decision, 65 Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579 (1993). which came fifteen years after Illinois Brick, should be the controlling mechanism for evaluating expert models rather than anything as blunt, categorical, and frankly wrong as the indirect purchaser rule. 66 On the use of Daubert in antitrust damages models see 2A Areeda et al., Antitrust Law, supra note 2, ¶¶ 340a, 399. Daubert rulings, which are gener­ally not subject to jury control, should provide judges with an adequate mechanism for ensuring that expert damages reports are based on relia­ble and relevant assumptions, methodologies, and data. 67 See, e.g., Comcast Corp. v. Behrend, 569 U.S. 27, 39 (2013) (Ginsburg & Breyer, JJ., dissenting) (assuming that Daubert applies to expert evidence in antitrust case); ZF Meritor, LLC v. Eaton Corp., 696 F.3d 254, 291–92 (3d Cir. 2013) (applying Daubert to damages evidence of lost profits in an antitrust case); City of Tuscaloosa v. Harcros Chems., Inc., 158 F.3d 548, 563–67 (11th Cir. 1998) (applying Daubert to evidence of an overcharge from collusion). In a particular case, if there is doubt about a particular expert opinion, the court will be re­quired to make a Daubert ruling with respect to the expert’s methodology. 68 See, e.g., Messner v. Northshore Univ. HealthSystem, 669 F.3d 802, 811–13 (7th Cir. 2012).

IV. Proximate Cause, or the Proper Measurement of Damages?

The Apple dissenters adopted a distinctively noneconomic approach that dispensed with the pass-on problem entirely. Indeed, they were not even concerned about who was injured. They reasoned that only the direct purchaser had an injury that was “proximately caused” by the defendant’s antitrust violation. 69 Apple Inc. v. Pepper, 139 S. Ct. 1514, 1530–31 (2019) (Gorsuch, J., dissenting). This view harkens back to a nineteenth-century tort law concept that was used by some courts to limit tort liability, particularly in railroad cases. 70 See, e.g., Blythe v. Denver & R.G. Ry. Co., 25 P. 702, 702–04 (Colo. 1891) (holding that an “act of God” did not meet the requirement of proximate cause); Stone v. Bos. & A.R. Co., 51 N.E. 1, 2–4 (Mass. 1898) (holding that the plaintiff could not recover when oil leaking from barrels on a railroad platform was ignited by a person unconnected to the railroad for a lack of proximate cause); see also Hovenkamp, The Opening of American Law, supra note 35, 106–22. Under this rule, only a single entity could be said to have an injury that was proximately caused by the defendant’s conduct. That approach was rejected by John Stuart Mill by the mid-nineteenth century. 71 See 1 John Stuart Mill, A System of Logic, Ratiocinative and Inductive: Being a Connected View of the Principles of Evidence, and Methods of Scientific Investigation 241, 398 (1843) (“For every event, there exists some combination of objects or events, some given concurrence of circumstances, positive and negative, the occurrence of which will always be followed by that phenomenon.”). A sharp debate ensued in the United States between Mill and Francis Wharton, an orthodox cleric who wrote on legal subjects and adhered to strictly premarginalist views. See Francis Wharton, The Liability of Railway Companies for Remote Fires: Proximate and Remote Cause 731–47 (1876); Francis Wharton, A Suggestion as to Causation 5–14 (1874). The dissent’s approach also detaches proximate cause concerns from the factor that has always been central to proximate cause inquiries—foreseeability. 72 See, e.g., Oliver Wendell Holmes, Jr., The Common Law 95 (1881) (“There is no such power [to avoid evil] where the evil cannot be foreseen.”); see also Mark F. Grady, Proximate Cause Decoded, 50 UCLA L. Rev. 293, 322–33 (2002); Benjamin C. Zipursky, Foreseeability in Breach, Duty, and Proximate Cause, 44 Wake Forest L. Rev. 1247, 1252–54 (2009). Indeed, it is so foreseeable that overcharge injuries will be passed on that no one seriously disputes it. 73 See Bank of Am. Corp. v. City of Miami, 137 S. Ct. 1296, 1305–06 (2017) (finding that proximate cause limited liability for a Fair Housing Act violation, even for certain foreseeable harms).

In the legal system, this narrow approach to proximate cause analysis was properly abandoned in tort cases and should be laid to rest. In eco­nomics, it was very largely upended by the marginalist revolution, which first provided serious tools to investigate how economic disruptions are passed through the economy. 74 See Hovenkamp, The Opening of American Law, supra note 35, 123–55. It makes even less sense in antitrust cases.

The Apple dissent’s view also confuses questions about proximate cause, or legal cause, with questions about the proper way to measure damages. 75 Apple Inc. v. Pepper, 139 S. Ct. 1514, 1528–31 (2019) (Gorsuch, J., dissenting). Illinois Brick was distinctly not a case about causation or proxi­mate cause. Indeed, the majority never used the term at all, and Justice Brennan used it only a single time in his dissent. 76 Ill. Brick Co. v. Illinois, 431 U.S. 720, 760 (1977) (Brennan, J., dissenting). The virtually exclusive concern with Illinois Brick was with computation of damages in situations where an overcharge might be passed on from one party to another.

Even on its own terms it seems hard to justify a conclusion that only the direct purchaser has an injury “proximately caused” by the violation when in many cases it suffered no overcharge injury at all.

Conclusion

Working within the context of the existing Illinois Brick rule, the majority reached the right conclusion about its application in Apple v. Pepper. While that eliminates some of the irrationalities of the indirect purchaser rule as it has been applied, it is hardly a solution to the problem. The correct solution is more consistent with the statutory lan­guage granting an action to (1) “any person who shall be injured in his business or property,” and then measuring the recovery as the (2) “damages by him sustained.” 77 15 U.S.C. § 15(a) (2012). In the process, this solution addresses a serious and widely recognized problem—namely, that the current policy toward price fixing under-deters. 78 See, e.g., OECD, Cartel Sanctions Against Individuals 17 (2003), https://www.oecd.org/competition/cartels/34306028.pdf [https://perma.cc/GA9D-YWM9] (“It is commonly assumed, however, that corporate fines rarely reach a level that would maximize their deterrent effect.”); Peter G. Bryant & E. Woodrow Eckard, Price Fixing: The Probability of Getting Caught, 73 Rev. Econ. & Stat. 531, 535 (1991) (finding that the probability of a particular conspiracy being caught is between 13% to 17% in a given year); John M. Connor & Robert H. Lande, Cartels as Rational Business Strategy: Crime Pays, 34 Cardozo L. Rev. 427, 435–42 (2012) (describing the debate generally between optimal “law and economics” deterrence approaches targeting corporations and individual sanctions); John M. Connor, Price-Fixing Overcharges: Legal and Economic Evidence, 22 Res. L. & Econ. 59, 107 (2007) (“Despite the evident increases in cartel detection rates and the size of monetary fines and penalties in the past decade, a good case can be made that current global anticartel regimes are under-deterring.”); Joseph E. Harrington, Jr. & Yanhao Wei, What Can the Duration of Discovered Cartels Tell Us About the Duration of All Cartels?, 127 Econ. J. 1977, 1977–79, 2003 (2017) (describing generally the theoretical issues with estimating the duration of cartels from observed behavior of discovered cartels and effects on claims about deterrence); Louis Kaplow, An Economic Approach to Price Fixing, 77 Antitrust L.J. 343, 446 (2011) (“Because of difficulties with detection, achieving effective deterrence is not easy, and empirical evidence suggests that deterrence is inadequate even in the United States, which is regarded to have the toughest enforcement in terms of both overall (public and private) effort and the level of sanctions.”); Gregory J. Werden, Sanctioning Cartel Activity: Let the Punishment Fit the Crime, 5 Eur. Comp. J. 19, 29–33, 36 (2009) (arguing that deterrence requires individual sanctions, and more than monetary sanctions); see also Hovenkamp, Federal Antitrust Policy, supra note 5, §§ 4.1–.7, 17.1–.7.

End-user purchasers who are not in a position to pass on anything should be awarded overcharge damages for the full overcharge they paid, for that measures the injury that they have sustained. All intermediaries beginning with the direct purchaser should be awarded damages for lost profits, which represents injuries from both absorbed overcharges and the loss of sales that always accompany collusion. None of these recover­ies is “duplicative,” and none requires complex apportioning of passed on damages. More importantly, this methodology is driven by the facts of each case and brings antitrust damages measurement more in line with dam­ages formulation across the full range of commercial harms.