Brief of ICLE and Scholars of Law and Economics to the 9th Circuit in SIS v Intuitive Surgical
INTEREST OF AMICI CURIAE[1]
The International Center for Law & Economics (“ICLE”) is a nonprofit, non-partisan global research and policy center aimed at building the intellectual foundations for sensible, economically grounded policy. ICLE promotes the use of law and economics methodologies and economic learning to inform policy debates and has longstanding expertise evaluating antitrust law and policy. Amici also include 19 scholars of antitrust law and economics (“Scholars of Law and Economics”) at leading universities and research institutions across the United States. Their names, titles, and academic affiliations are listed in the Appendix. All possess relevant expertise and experience in antitrust law and economics.
ICLE and the Scholars of Law and Economics have an interest in ensuring that antitrust law promotes the public interest by remaining grounded in sensible rules informed by sound economic analysis. That includes advising against decisions that rest chiefly on doctrinal abstractions that are poorly fit to the substantive competitive factors at issue in a given market and that fail to consider the benefits of tying or bundling, as well as the costs, in context. We advise, in particular, against both per se and quasi-per se approaches to tying that lack an adequate economic foundation.
SUMMARY OF ARGUMENT
Intuitive Surgical, Inc. placed downstream restrictions on the use of its da Vinci surgical systems, Class II medical devices regulated by the Food and Drug Administration (FDA) and granted clearance under the Agency’s 510(k) clearance process. These restrictions ensure that Intuitive’s customers—hospitals—may use Intuitive’s systems only with products and services that have been authorized by Intuitive. Surgical Instrument Service Company, Inc. alleges these restrictions are anticompetitive; specifically, that restrictions on the “EndoWrist” attachments Intuitive designed for the da Vinci system—also FDA-regulated Class II medical devices—cannot be justified by regulatory requirements. 6-ER-1256–57. In short: Because of Intuitive’s product design and downstream agreements with hospital customers, SIS alleges that Intuitive refuses to deal with customers on terms that would benefit SIS.
As a matter of law and economics, there are two principal problems with the arguments advanced by SIS and several of its amici.
First, SIS’s case hinges on the existence of a single-brand aftermarket of EndoWrist surgical devices. SIS and its amici argue that the trial court erred in its application of the “Epic factors”—a doctrinal expression of the economic screen when a firm with a large installed base of consumers in a foremarket seeks to exploit that installed base to anticompetitive effect in an aftermarket. But the test articulated in Epic Games, Inc. v. Apple, Inc., 67 F.4th 946 (9th Cir. 2023), derived from the Supreme Court’s decision in Eastman Kodak Co. v. Image Technical Services, Inc., 504 U.S. 451 (1992), properly applied here. The point of Epic’s test is to screen whether consumers “knowingly and voluntarily” make choices when purchasing in the foremarket, which effectively constrains the aftermarket. Newcal Indus., Inc. v. Ikon Office Sols., 513 F.3d 1038, 1048–49 (9th Cir. 2008). That test does not disappear merely because a plaintiff relabels the foremarket or alleges a high market share there.
Indeed, SIS would turn Kodak’s holding on its head. The Supreme Court held that Kodak could possibly be liable for illegal tying even if it did not have market power in the tying market (the “foremarket” here). That is, a defendant’s lack of market power in the tying market does not render tying per se lawful. Kodak thus explains the circumstances in which an otherwise unlikely prospect of anticompetitive tying might be realized. It does not follow—by law, logic, or economics—that the factors are irrelevant when a plaintiff alleges market power in the tying market. SIS’s proposed shortcut is not the holding in Kodak, and it should be rejected as legally and economically unsound.
Second, this case—which SIS styles as principally about tying—is at bottom about Intuitive’s refusal to deal on terms agreeable to the plaintiff. It therefore should be informed by the principle that “the Sherman Act ‘does not restrict the long recognized right of [a] trader or manufacturer engaged in an entirely private business, freely to exercise his own independent discretion as to parties with whom he will deal.’” Verizon Commc’ns Inc. v. Law Offs. of Curtis v. Trinko, LLP, 540 U.S. 398, 408 (2004) (quotation omitted). Moreover, as Trinko cautions, courts should be especially loath to impose a duty to deal in regulated-product markets, where the relevant regulatory scheme restricts the terms of dealing. Id. at 411; see also Credit Suisse Securities (USA) LLC v. Billing, 551 U.S. 264, 270–71 (2007).
ARGUMENT
I. The Epic Factors Apply to SIS’s Single-Brand Aftermarket Claims.
SIS’s antitrust allegations depend entirely on the existence of a single-brand aftermarket for Intuitive’s EndoWrists. Courts are rightly skeptical of single-brand aftermarkets, which require unusual economic conditions. Importantly, SIS is incorrect that market power in the foremarket alone is legally or economically sufficient to support this theory. Under this Circuit’s precedent, plaintiffs must apply the Epic factors to determine whether consumers are truly “locked in” if they propose a single-brand aftermarket. To hold otherwise, as SIS invites, would dangerously expand antitrust liability by conflating brand power with market power and undermining economic principles that safeguard competitive system design.
A. SIS’s Allegations Hinge on the Existence of a Single-Brand Aftermarket.
Market definition is based on “the theory of harm at issue.” Teradata Corp. v. SAP SE, 124 F.4th 555, 570 (9th Cir. 2024). On appeal, SIS and amici focus on tying in a single-brand aftermarket. Tying law starts from a simple baseline. A plaintiff must show (1) distinct tying and tied products in distinct markets, (2) market power in the tying market, and (3) the existence of a tie that harms competition in the tied market. Jefferson Parish Hospital District No. 2 v. Hyde, 466 U.S. 2 (1984). Consistent with decades of economic learning that ties are ubiquitous and do not typically threaten competition, such claims are most often evaluated under the rule of reason—with careful definition and analysis of the relevant market; see, e.g., Epic, 67 F.4th at 976; Illinois Tool Works v. Independent Ink, 547 U.S. 28, 46 (2006); Jefferson Parish, 466 U.S. at 29.1F[2]
SIS’s tying theory proceeds as follows. The alleged tying market comprises minimally invasive soft tissue (MIST) surgical robots, in which Intuitive’s innovation in developing the da Vinci system has given it a near-total monopoly. 3-ER-478. The alleged tied market consists of EndoWrists, attachments used with the robot arms of the da Vinci system for surgical procedures. These attachments have a use counter that expires after a set number of procedures based on FDA clearance. 1-SER-39–41. Remanufactured EndoWrist instruments can be used with the da Vinci systems provided the third-party remanufacturer received clearance from the FDA (or provided clinical proof supporting their proposed modifications). Intuitive otherwise restricts the use of unapproved attachments.
SIS’s claim thus involves an alleged tie between an asserted foremarket and aftermarket. In a foremarket, a consumer purchases a product; an aftermarket comprises goods whose value depends in part on the foremarket purchase, but which, for tying purposes, also constitutes an economically distinct market. Epic, 67 F.4th at 976. Tying a foremarket and aftermarket can be illegal when the defendant has market power in the foremarket. But analysis of market share in the foremarket alone is not enough to determine anticompetitive foreclosure in the aftermarket. If a firm had market power in the foremarket, but consumers were not locked into that same firm’s products in the aftermarket, there would be no anticompetitive harm in the aftermarket. Epic, 67 F.4th at 976–97. See also Carl Shapiro & David J. Teece, Systems Competition & Aftermarkets: An Economic Analysis of Kodak, 39 Antitrust Bull. 135, 143–44 (1994) (“If consumers’ brand-switching costs are uniformly low, there can be no monopoly power in aftermarkets.”). Even where some lock-in exists, “an equipment vendor with 100% of its aftermarket cannot set aftermarket prices with impunity” because consumer responses constrain pricing power. Id.
Accurate market definition is paramount for rule of reason claims like the one here. See Gregory J. Werden, Why (Ever) Define Markets? An Answer to Professor Kaplow, 78 Antitrust L.J. 729, 730–35 (2013). One kind of market—a single-brand aftermarket, where the market is limited to one brand of products—is “highly disfavored” and “extremely rare.” Epic Games, Inc. v. Apple Inc., 559 F. Supp. 3d 898, 1021 (N.D. Cal. 2021) (quotations omitted), aff’d in part and rev’d in part, 67 F.4th 946 (9th Cir. 2023); see also Reilly v. Apple, 578 F. Supp. 3d 1098, 1107 (N.D. Cal. 2022) (“[f]rom nearly the inception of modern antitrust law, the Supreme Court has expressed skepticism of single-brand markets.”); Apple, Inc. v. Psystar Corp., 586 F. Supp. 2d 1190, 1198 (N.D. Cal. 2008) (“[s]ingle-brand markets are, at a minimum, extremely rare”); Carl Shapiro, Aftermarkets and Consumer Welfare: Making Sense of Kodak, 63 Antitrust L.J. 483, 484 (1995) (noting that single-brand aftermarkets could “markedly expand[] the number of companies that may be regarded as monopolists,” and cautioning that that approach “holds considerable dangers of restraining the behavior of firms that possess no genuine monopoly power”).
In an attempt to all but guarantee Intuitive’s market power—and ability to foreclose competition—in the aftermarket, SIS seeks to define such a single-brand aftermarket for Intuitive’s EndoWrist attachments. 5-ER-1165.2F[3] Indeed, SIS provides no basis for proving its claims except through the definition of a single-brand aftermarket. SIS’s case thus hinges on proving that this case exemplifies a rare, exceptional instance in which a single-brand aftermarket is proper. See Coronavirus Reporter v. Apple, Inc., 85 F.4th 948, 955 (9th Cir. 2023).
B. Single-Brand Aftermarkets Require the Use of the Epic Factors.
SIS argues that it can assume a single-brand aftermarket without proof of the lock-in factors because it alleges Intuitive has market power in the foremarket. That is wrong as a matter of both law and economics. To prove a single-brand aftermarket, a plaintiff must show (1) the challenged aftermarket restrictions are “not generally known” in the foremarket; (2) “significant” information costs prevent accurate life-cycle pricing; (3) “significant” monetary or non-monetary switching costs exist; and (4) general market-definition principles do not undermine the proposed single-brand market. Epic, 67 F.4th at 977. These requirements, which are not driven by whether a defendant has power in a foremarket, were applied by this Court in Epic in response to a plaintiff’s alleged single-brand aftermarket. Id.; see also Coronavirus Reporter, 85 F.4th at 956. The District Court was right to say they apply here too.
1. To understand this Court’s rule for single-brand aftermarkets, it is important to understand the Supreme Court’s decision in Kodak. There, the Court held that a single-brand aftermarket can constitute a relevant antitrust market if the “commercial realities” faced by durable equipment purchasers make service and parts for that brand non-interchangeable. Kodak, 504 U.S. at 481–82. The aftermarket in Kodak arose after Kodak changed course by restricting sales of replacement parts to buyers who used Kodak service (or self-serviced) and otherwise cutting off provision of parts. The Court held that evidence of “lock-in,” high information and switching costs, and price discrimination supported an inference of Kodak’s power in the aftermarket even if the equipment foremarket was competitive. Id. at 477–78. But, crucially, the Court did not adopt any presumption that power (or lack thereof) in the foremarket determines the competitiveness of the aftermarket. It rejected Kodak’s bid for a legal presumption against aftermarket power and emphasized the need for a fact-intensive inquiry. Id. at 478–79.
In economic terms, Kodak addresses the real but unlikely risk of installed-base opportunism, where a firm with a large installed base of consumers in a foremarket seeks to exploit that installed base in an aftermarket. See Shapiro, supra, at 510–11 (concluding durable consumer injury from monopolized aftermarkets is typically small); Benjamin Klein, Market Power in Antitrust: Economic Analysis after Kodak, 3 Sup. Ct. Econ. Rev. 43, 49–51 (1993) (discussing installed?base “hold?up” and its limits).
Kodak crystallizes conditions under which that risk may be realized: when (1) customers are “locked in” by high switching costs; and either (2) information problems—high information costs, information asymmetries, or deception of naive consumers—prevent foreseeable life?cycle pricing; or (3) the seller makes an unanticipated policy change that defeats buyers’ ex ante contractual safeguards. Shapiro & Teece, supra, at 150–52. Under those conditions, increasing aftermarket prices may be profitable, even if it damages the company’s reputation, because the short?run gains a company earns from its installed base can exceed long?run losses in new?equipment sales.
This Court’s Epic factors are the doctrinal expression of the economic screen for such a rare situation. See Shapiro, supra, at 488–92; Severin Borenstein, Jeffrey MacKie-Mason, & Janet Netz, Antitrust Policy in Aftermarkets, 63 Antitrust L.J. 455, 462–66 (1995). This test recognizes that, unless consumers lack information about aftermarket restrictions that let a manufacturer gouge already?locked?in users, competition will effectively discipline any single?brand aftermarket. Epic, 67 F.4th at 979.
Antitrust law does not police disclosed, competitively priced systems such as da Vinci. Kodak allows a single?brand aftermarket theory of harm only when plaintiffs show that high switching costs and lack of information break the equipment?aftermarket link and enable profitable hold?up. Kodak, 504 U.S. at 475–76. The Ninth Circuit’s application of these factors in Epic exemplifies the reluctance to label any firm with market power in a foremarket a monopolist in a single-brand aftermarket—an important and prudent limitation, consistent with economics, that SIS asks the Court to jettison.
2. Both SIS and amici assert that the factors drawn out of Kodak are inapplicable because Kodak applies only when there is “meaningful interbrand competition upstream.” Dkt 28.1 at 27. Economically, abstract power in one market does not establish the capacity to raise market price or profitably hold up locked?in buyers. And power in the foremarket alone does not mean there is no competition in that market. Newcal, 513 F.3d at 1048–49. What matters is the mechanism that severs foremarket discipline: unexpected policy changes plus costly switching, coupled with inadequate contractual or reputational constraints. SIS’s argument is fundamentally misguided as an economic matter, as this Court’s precedent recognizes.
To start, SIS misreads Kodak. In Kodak, the Court rejected the argument that robust competition in the foremarket always forecloses aftermarket power as a matter of law. It thus refused to recognize foremarket competition as a categorical defense to an aftermarket foreclosure theory of harm. See Kodak, 504 U.S. at 471. Foremarket competitiveness was discussed in responding to an argument made by Kodak; the Court did not hold that the presence or absence of competition in the foremarket had any bearing on the application of the lock-in factors. Id. at 467–70.
Crucially, the Court in Kodak did not hold that a lack of competition in the foremarket forecloses aftermarket competitiveness as a matter of law—nor is that claim a corollary of the holding in Kodak. Nowhere in Kodak did the court consider that a plaintiff does not need to prove “lock-in” to establish a single-brand aftermarket so long as the plaintiff can prove the defendant has market power in the foremarket.
While power in an aftermarket can go hand in hand with power in the foremarket, market power in each remains a “highly fact-dependent” determination after Kodak. Shapiro & Teece, supra, at 137. In technical terms, foremarket power alone does not enable profitable aftermarket “double?monopoly” pricing. A firm already charging a monopoly price for the system cannot generally profit by adding a second monopoly in a complementary input market, especially when the products are used in fixed (or similar) proportions. At most, tying enables short?run price discrimination, which can even expand output. See Keith N. Hylton & Michael Salinger, Tying Law and Policy: A Decision-Theoretic Approach, 69 Antitrust L.J. 469, 513 (2001). In short, Kodak is a special case: it does not authorize a shortcut for plaintiffs to define single-brand aftermarkets whenever they can allege a modicum of foremarket power.
Rather, this Court’s insistence on proof of lock?in before allowing single?brand aftermarkets to stand is well-grounded in economics: without lock?in, aftermarket leveraging is either unprofitable or benign. In Epic, the Court affirmed Epic’s failure to prove the prerequisites to a single-brand market—including that consumers were unaware of restrictions in the aftermarket or significant switching costs (regardless of Apple’s share of the alleged foremarket). Epic, 67 F.4th 946, 979. So too in Coronavirus Reporter, where the plaintiffs alleged a foremarket in which Apple held an asserted share of 80%. Coronavirus Reporter v. Apple Inc., 2021 WL 5936910, at *3 (N.D. Cal. Nov. 30, 2021); Coronavirus Reporter, 85 F. 4th at 956.
Moreover, this Court has repeatedly made clear that courts must assess ordinary principles of substitution in the aftermarket, whether or not a plaintiff can prove consumers are locked into a foremarket. Newcal, 513 F.3d at 1046–47, 1051; Epic, 67 F.4th at 970, 977. Epic confirms that demonstrating the lock-in mechanics is a necessary predicate to treating a single-brand marketplace as a cognizable market, but it is not dispositive. Plaintiffs who clear that threshold still must prove that the aftermarket is properly defined and that those mechanics caused harm to aftermarket competition. Epic, 67 F.4th at 977 (“the aftermarkets inquiry does not end as soon as a plaintiff checks the Kodak-based boxes related to consumer knowledge, information costs, and switching costs.”). If demonstrating lock-in—itself a significant showing—is insufficient to obviate the need for proving actual competitive harm in the aftermarket, then, a fortiori, mere foremarket market power cannot suffice to skip that analysis as SIS maintains.
Nothing in Kodak, Epic, or Coronavirus Reporter supports SIS’s economically counterintuitive argument that a single-brand aftermarket can be established by presumption from the fact that a defendant has alleged power in a foremarket. Without that, SIS offers no reason to constrict the tied market to Intuitive’s product alone. SIS did not allege—nor could they—that hospitals purchasing MIST surgical systems were naïve buyers who were unaware of the requirements for those particular tools.
SIS’s alternative formulation—which would limit Epic’s lock?in screen to cases with interbrand competition upstream—fares no better. It is fundamentally vague in a way that confounds administration: It does not explain how far upstream the interbrand competition must be; by what metric, or according to what standard, it is to be judged; or who bears the burden. It is also wrong as a matter of economics. The point of Epic’s screen is to test whether foremarket competition failed to constrain the aftermarket because the normal mechanism linking system-level rivalry to aftermarket pricing broke down. That test does not disappear merely because a plaintiff re?labels the foremarket or alleges a high share there. The Court should affirm the District Court’s correct application of the Kodak factors under this Court’s precedent.
3. As amici, the FTC argues that “[w]hen a company has market power in the foremarket, consumers lack the ability to use their foremarket purchase decisions to discipline the company’s conduct in an aftermarket—regardless of factors such as information costs and switching costs.” FTC Br. at 15. The amici Antitrust Law Professors make a similar case, saying “it makes no sense” to combine lock-in and foremarket-monopoly theories. Antitrust Law Professors Br. at 13. Those assertions confuse rivalry with discipline and misstate prevailing economic theory.
Assuming arguendo market power in the foremarket, a monopolist sets the price for its system of products to maximize profit. Starting from the one?monopoly?profit principle, a firm that possesses monopoly power in one product market generally cannot increase total profits by adding a second monopoly markup on a complementary product—especially when used in relatively fixed proportions (as is the case with Intuitive’s da Vinci and EndoWrists). By the same token, it is generally impossible to increase that price by imposing contractual terms that restrain conduct in the aftermarket, where those constraints are known at the time of contract formation. “If the tie and increased price of [the aftermarket product] were, instead, anticipated, the tie is just the form in which [the defendant] is collecting the total price.” Klein, supra, at 52. Unless the monopolist can unexpectedly change terms or exploit unanticipated changes in relevant circumstances to raise total profits, a second markup serves only to suppress usage and reduce system demand. See Shapiro & Teece, supra, at 142–43. Thus, under standard economic conditions, foremarket power does not imply any economic incentive or ability to harm aftermarket competition.
To be sure, economic theory identifies limited circumstances where this baseline may not hold, but those exceptions reinforce—not eliminate—the need for Epic’s lock-in analysis. Even when one-monopoly-profit constraints relax, welfare effects remain ambiguous. Compare Michael D. Whinston, Tying, Foreclosure, and Exclusion, 80 Am. Econ. Rev. 837 (1990) (arguing tying can exclude rivals and harm competition) with Dennis W. Carlton & Michael Waldman, Competition, Monopoly, and Aftermarkets, 26 J.L. Econ. & Org. 54, 55 (2010) (showing that aftermarket monopolies can correct inefficiencies, improving social and consumer welfare). The relationship between foremarket power and aftermarket efficiency is therefore context-dependent. That is precisely why the Epic factors are necessary: they identify the main conditions under which the one-monopoly-profit principle breaks down and aftermarket exploitation becomes feasible. Absent the special (and uncommon) features described in Kodak, Epic, and other similar cases, a company with alleged foremarket power is unlikely to be able or willing to anticompetitively monopolize an alleged aftermarket.
Moreover, it is well-established that when buyers choose among rival systems, aftermarket prices are disciplined at the time of purchase by interbrand competition and life?cycle pricing. Absent unusual information problems or switching costs, any potential overcharge in the aftermarket is rebated back as an equipment discount. This is the classic “razors and blades” logic: A company cannot overcharge for blades, even if it gives away the razors. Shapiro, supra, at 494, 505. This principle is important in cases, like this one, where a seller provides the system as a whole, and consumers need parts in relatively fixed proportions. In such cases, what matters is the total cost of using the system. If system rules (such as approved parts, required calibrations, maintenance schedules, or regulatory requirements) are disclosed ex ante and consumers can effectively comparison?shop on total cost, then competition at the system level disciplines the cost of the whole package.
Competition over the “system” neutralizes aftermarket markups in the form of lower equipment prices. Shapiro, supra, at 505. Any risk of significant, durable consumer harm within the aftermarket is rare. A hospital that buys one da Vinci robot will use a predictable number of instruments. Interbrand competition at the time of the system sale disciplines the whole package through life?cycle pricing; aftermarket margins will be rebated into lower equipment prices, because buyers compare the total expected cost over the product’s life—purchase price plus parts and service. Shapiro & Teece, supra, at 143 n.15 (“if consumers assess the total systems price and engage in ‘accurate life cycle pricing,’ then service market prices will affect equipment demand and a separate market for parts and/or service is unlikely.”). A firm that plans to charge more later must cut the price of the first good to win the sale, so any extra profit in the aftermarket is returned to consumers through a lower initial price. This remains true when the firm has foremarket power. Even monopolists cannot charge infinite prices, and a firm with market power is still constrained by what consumers are willing to pay over the entire life-cycle of the product.
The FTC conflates brand power with antitrust power. Under the FTC’s interpretation, courts would find a single?brand aftermarket whenever the foremarket is concentrated and the aftermarket lacks “reasonable substitutes.” FTC Br. at 14. That collapses market definition into an ipse dixit and invites what Professor Klein warned against: equating market power with a firm’s ability to raise prices and retain some number of customers who have already bought into the system. Klein, supra, at 71–85.
Nearly every brand has some loyalty such that certain customers would pay more; adopting this standard would mean almost every firm could be labeled as having market power and would perversely penalize companies that have forged strong brand trust with customers. Klein, supra, at 78. Indeed, a firm might have some degree of market power and yet face meaningful competitive constraints, both in the form of viable alternatives in the market and potential entrants.
Here, for example, hospitals and surgeons may look to alternative MIST systems or non-robotic laparoscopic surgical techniques; product development from leading device firms, such as Medtronic, constrains pricing and output in the relevant markets. See generally, e.g., Louis Kaplow, The Accuracy of Traditional Market Power Analysis and a Direct Adjustment Alternative, 95 Harv. L. Rev. 1817, 1845–46 (1982). The correct test asks whether the conduct enables a company to profitably raise market-wide prices or reduce market-wide output, and that requires analyzing Epic’s factors that SIS and its amici would skip. See Klein, supra, at 76.
The request from SIS and amici to forgo any showing of the Epic factors (while alleging a single-brand aftermarket) is incompatible with Jefferson Parish’s rejection of “abstract” power in a market. Jefferson Parish, 466 U.S. at 26. It is also incompatible with Kodak’s emphasis on switching? and information?costs. Kodak, 504 U.S. at 475–76. And it runs afoul of the Ninth Circuit’s emphasis on applying market-definition principles, Newcal, 513 F.3d at 1051, and precedent applying Epic even when plaintiffs alleged large foremarket shares. Coronavirus Reporter, 85 F.4th at 956–57. The law on single-brand aftermarkets requires a showing of consumer lock-in because economics demands it: otherwise, competitive, disclosed system design would be penalized and conversion of contract disputes into treble?damages antitrust would become routine.
* * *
In the end, Kodak is best read as an exception keyed to a specific economic risk: installed?base opportunism. But Kodak?type concerns thus arise only when a supplier springs unexpected aftermarket policy changes on an installed base. When restrictions are announced prospectively and buyers price them in, there is no competition problem for courts to solve. Newcal, 513 F.3d at 1048 (antitrust laws are unconcerned with “market power that arises solely from contractual rights that consumers knowingly and voluntarily gave to the defendant”). Hold?up problems are ubiquitous but belong to contract and reputation, not antitrust, absent the narrow circumstances identified in cases like Kodak involving unanticipated policy changes that let a firm tax non?redeployable, brand?specific investments. Klein, supra, at 51, 56–57.
By contrast, relaxing consideration of the Epic factors would fling open the door to single?brand aftermarkets and invert tying law. Plaintiffs could define the tied market as comprising only “Brand X” and argue that a modest showing of upstream power eliminates the need to prove lock?in or competitive effects in the tied market. Virtually any contractual restriction imposed by a firm with upstream market power would become presumptively illegal, converting ordinary contract terms into treble-damages antitrust violations. Plaintiffs would define the foremarket narrowly enough to allege market power, define the tied aftermarket as a single brand, and market power in the latter would become automatic—allowing almost any contractual restriction to be condemned as a “tie.” Indeed, they could do so even where there are aftermarket alternatives. That would fundamentally distort antitrust law in a way that harms competition and consumers alike.
II. SIS Improperly Seeks to Impose a Duty to Deal on Defendants and Burdensome Roles on Courts.
Although SIS tries to frame its case as one challenging a tie, antitrust law is not beholden to “wooden application” of any particular label or characterization. United States v. Microsoft Corp., 253 F.3d 34, 94, 95 (D.C. Cir. 2001) (en banc) (per curiam). Substance, not form, matters. As Intuitive explains in its brief, this is not a case involving a true tie: Intuitive did not coerce purchasers to buy only Intuitive’s aftermarket products. See Intuitive Br. 50-54. What SIS in fact challenges is Intuitive’s terms of dealing—namely, the fact that aftermarket sellers must receive FDA clearance for their products. Intuitive’s refusal to deal with SIS stems from SIS’s lack of clearance; other aftermarket sellers who have sought and received clearance from the FDA (or have clinical proof of safety and efficacy) deal freely with Intuitive. SIS seeks to impose on Intuitive a duty to deal: an obligation to change the terms of Intuitive’s dealing, which are derived from FDA regulatory requirements, to allow a particular competitor to sell products on its own, preferred terms.
SIS must thus travel “the hard road of refusal to deal doctrine.” Novell, Inc. v. Microsoft Corp., 731 F.3d 1064, 1078 (10th Cir. 2013). In Trinko, the Supreme Court held that regulatory requirements—providing local telephone carriers access to network elements there, or receiving 510(k) clearance from the FDA for surgical tools here—did not impose additional antitrust duties to assist potential competitors. Trinko, 540 U.S. at 406–07. Restrictions that arise from regulation, particularly where the products at issue are “not otherwise marketed or available to the public” and were created or shaped by the regulatory regime, do not, without more, give rise to a duty to deal. Id. at 410–12. In this case, the terms of entry are restricted by the federal Food Drug & Cosmetic Act, its implementing regulations, and the judgments of an expert health and safety regulator, the FDA. See generally 21 U.S.C. § 331; 21 U.S.C. § 360c; and 21 U.S.C. § 360j. Ultimately, “[t]he FD&C Act mandates that all devices have a reasonable assurance of safety and effectiveness and gives FDA the authority to establish regulatory controls to provide such.” U.S. Food & Drug Admin., FDA Report on the Quality, Safety, and Effectiveness of Servicing of Medical Devices 4 (2018). With respect to the products and services at issue, external regulatory requirements restrain entry.
Economists recognize that claims like SIS’s are fraught with the risk of chilling competitively benign conduct. See, e.g., Dennis Carlton, A General Analysis of Exclusionary Conduct and Refusals to Deal: Why Aspen and Kodak are Misguided, 68 Antitrust L.J. 659, 673 (2001). That is plainly the case here, where the challenged term of dealing is one that promotes the safety and efficacy of the product at issue. The ultimate goal of the antitrust laws is to maximize consumer welfare—meaning intervention should occur only if it improves the overall state of competition, which includes consideration of product quality and safety. Intuitive’s alleged restraints here serve an obvious procompetitive function of compliance with regulatory requirements in the health sector. Notably, other third parties complied with these regulatory requirements and received approval from Intuitive to market remanufactured EndoWrists. 3-SER-548–56, 5-SER-1126–30. In such circumstances, a defendant should not have the burden to explain particular industry practices as justifications for its actions. See Carlton, supra, at 680.
What is more, the Supreme Court expanded on the intersection between regulation and competition in Credit Suisse: where a robust regulatory scheme governs the challenged conduct, courts should not use antitrust law to impose additional obligations—such as a duty to deal—if doing so would be “clearly incompatible” with that scheme. Credit Suisse, 551 U.S. at 265. The inquiry is whether there is a “clear repugnancy” between the antitrust claim and the regulatory program. Four considerations guide that determination: (1) the regulator has authority over the conduct; (2) the regulator actively exercises that authority; (3) applying antitrust law risks creating conflicting “guidance, requirements, duties, privileges, or standards of conduct”; and (4) the conduct lies within the heart of the regulated field. Id. at 271–76. When these conditions are met, the antitrust laws cannot be used to manufacture a duty to deal. Id. at 275–76, 284–85.
The conduct SIS challenges falls squarely within this sort of regulatory space. Medical devices and their servicing and repair are regulated by the FDA under the federal Food, Drug, and Cosmetic Act (FD&C Act). See U.S. Food & Drug Admin., supra, at 4 (noting that “proper servicing is critical to the ongoing safety and effectiveness of many devices, particularly those used on numerous patients over long periods of time; poor quality servicing may lead to poor device performance, malfunction, and adverse events. . . . FDA believes it has statutory authority to regulate device servicing”). The parties and the trial court all recognize that Intuitive’s da Vinci systems and its EndoWrist attachments are medical devices, subject to FDA implementing regulations, and regulated as Class II medical devices. The parties and the trial court also agree that the appellant’s medical device products—labeled “repaired” EndoWrists by SIS and “unauthorized” and “remanufactured” EndoWrists by Intuitive—have received neither pre-market approval nor 510(k) clearance by the FDA. The FDA actively regulates medical products such as the EndoWrists (as evidenced by their involvement in granting “clearance” to certain modified EndoWrists). Entry into Class II medical device product markets is regulated by the FDA, an expert federal health and safety agency. Requiring Intuitive to remedy its allegedly anticompetitive conduct by approving and subsequently servicing any EndoWrist would run up against the need to obtain FDA pre-marketing approval or 510(k) clearance for class II medical devices. SIS thus asks the Court to ignore existing precedent on the duty to deal in a regulated space.3F[4]
To ignore this precedent—and the important considerations underlying it—would require the court to fashion a complicated regulatory remedy. Intuitive does not have FDA clearance to market EndoWrist attachments under the terms that SIS would prefer. A judicial remedy mandating that access would have to grapple with questions like the following: Should SIS be obligated to seek FDA clearance and engage in whatever further research or product development might be required by the FDA? At what cost to Intuitive and its customers? And what if the FDA were to refuse to grant clearance for changed device specifications? Would the court’s order require Intuitive to service all Da Vinci machines, including remanufactured EndoWrists, nonetheless? What would be reasonable rates to charge for service? How should manufacturers price components they are ordered to make available to third parties?
If a firm is made to service third-party equipment or provide parts to third-parties, then the firm would also have an incentive to set prices which allow them to recoup the lost opportunity cost of servicing or providing replacement parts to consumers. Shapiro, supra, at 502. Such questions are all the more difficult where products in both alleged markets—the foremarket and the aftermarket—are innovative ones, developed at substantial cost, with investments in both initial product development and subsequent refinements at risk, contingent on both regulatory clearance and market developments. See Trinko, 540 U.S. at 407 (mere monopoly power—including power acquired through regulation—will not be found unlawful to “safeguard [that] incentive to innovate”).
Courts have historically been averse to acting as “central planners” and regulating the prices that businesses choose to set. Trinko, 540 U.S. at 408 (courts are “ill suited to act as central planners, identifying the proper price”); Pac. Bell. Tel. Co. v. linkLine Commc’ns, Inc., 555 U.S. 438, 453 (2009) (antitrust courts in particular “normally avoid direct price administration”); Nat’l Collegiate Athletic Ass’n v. Alston, 594 U.S. 69, 103, (2021) (courts should only issue decrees they can “adequately or reasonably supervise”) (quoting Trinko, 540 U.S. at 415). And a hard rule mandating that firms sell parts to or service replacement parts from third parties at the same rate they do for end users would interfere with charging a profit-maximizing price that accounts for lost margins taken by those third parties. Shapiro, supra, at 502–03.
CONCLUSION
Kodak recognized that the risk of anti-competitive installed base opportunism may persist in the absence of foremarket power. But the converse is not true; the risk does not become automatic when there is (or is alleged to be) market or monopoly power in the foremarket. To sidestep the considerations embodied in Kodak—as this Court has applied from Epic to Coronavirus Reporter—would open the door wide to rightly disfavored single-brand aftermarkets. It would, moreover, ignore the serious and general concerns of this Court, and the Supreme Court, about judicially imposed duties to deal, imposing a regulatory or central planning role on the courts. The District Court properly applied the Epic factors to this case’s invocation of a single-brand aftermarket, relying on this Court’s prior decisions on the very issue. That decision was grounded in clear precedent, and it should be affirmed.
[1] Pursuant to Rule 37.6, amici affirm that no counsel for a party authored this brief in whole or in part, and that no person other than amici or their counsel contributed money to fund preparing or submitting this brief. The parties have consented to this brief’s filing. 9th Cir. R. 29-2(a).
[2] SIS does not argue that the per se rule should apply (although its asserted framework comes dangerously close). A downstream contractual restriction on regulated surgical devices (characterized by SIS as Intuitive refusing to deal on its desired terms) is not conduct that can be condemned as anticompetitive without assessing factual claims about safety and regulation. See Broad. Music, Inc. v. Columbia Broad. Sys., Inc., 441 U.S. 1, 19–20 (1979) (confining the per se rule to practices that “would always or almost always tend to restrict competition and decrease output”).
[3] Even though Plaintiff seeks to define a single-brand aftermarket, it tacitly recognizes that Intuitive faces competition within its artificially narrow alleged market because other companies have received clearance and approval to remanufacture and market EndoWrists. 3-SER-548–56, 5-SER-1126–30.
[4] Trinko does not require a defendant to provide access to a competitor if it isn’t already providing access elsewhere. See MetroNet Servs. Corp. v. Qwest Corp., 383 F.3d 1124, 1132 (9th Cir. 2004).