The Roberts Court’s most recent antitrust decision, American Needle, Inc. v. National Football League,45 resulted in the first Supreme Court judgment in favor of an antitrust plaintiff since 1992. Given that the Court cut back on the scope of an antitrust immunity, thereby permitting more antitrust actions to proceed to discovery, the decision might at first seem inconsistent with decision theory’s focus on the limits of antitrust. But, as Judd Stone and Joshua Wright have recently explained, American Needle actually moves antitrust in a direction consistent with decision theory’s instruction to minimize the sum of decision and error costs.46  
At issue in American Needle was whether a vote by the members ofthe National Football League (NFL) to authorize an action by a corporate entity they created and controlled, the National Football League Properties (NFLP), could constitute a contract, combination, or conspiracy for purposes of Sherman Act Section 1. In 1963, the members of the NFL established NFLP to develop, license, and market their intellectual property. For almost four decades, NFLP granted nonexclusive licenses, permitting multiple manufacturers and vendors to produce and sell team-branded apparel. In 2000, the teams voted to authorize NFLP to grant exclusive licenses, and NFLP granted Reebok International Ltd. a ten-year exclusive license to produce and sell trademarked headwear for the NFL teams. NFLP then declined to renew plaintiff American Needle’s nonexclusive license. American Needle sued, claiming that the agreements between the NFL, its teams, NFLP, and Reebok violated Sections 1 and 2 of the Sherman Act.
In defending against American Needle’s Section 1 claim, the defendants asserted that they were incapable of conspiring “because they are a single economic enterprise, at least with respect to the conduct challenged.” The district court agreed, holding that the NFL, NFLP, and respective NFL teams qualified as a “single entity” and therefore could not conspire in violation of Section 1. The US Court ofAppeals for the Seventh Circuit affirmed, carefully limiting its holding to whether the defendants were acting as a single entity with respect to the particular conduct at issue, the licensing of teams’ intellectual property. Both American Needle and, somewhat surprisingly, the defendants then petitioned the Supreme Court for writ of certiorari, the former on grounds that the defendants were capable of conspiring with respect to the challenged conduct and the latter on grounds that the Seventh Circuit should have held more broadly that the NFL (and other sports leagues) act as a single entity generally, not just with respect to some of their conduct. Disregarding the Solicitor General’s advice to deny certiorari, the Supreme Court accepted the appeal, characterizing the issue before it as “whether the NFL respondents are capable of engaging in a ‘contract, combination ... , or conspiracy’ as defined by § 1 of the Sherman Act, 15 U.S.C. § 1, or ... whether the alleged activity by the NFL respondents ‘must be viewed as that of a single enterprise for purpose of § 1.’ ”
American Needle therefore afforded the Court an opportunity to reconsider the contours of the “intraenterprise immunity” doctrine. That doctrine recognizes that even the obviously unilateral conduct of individual business firms often involves some literal agreements (e.g., between agents of the firm) but generally should not be considered concerted conduct for purposes of Sherman Act Section 1. The point of the intraenterprise immunity doctrine is to insulate from liability those literal agreements (such as understandings between a parent corporation and its wholly owned subsidiary) that cannot really reduce competition by removing independent centers of decision making from the economy and thereby potentially consolidating market power.
The Supreme Court most fully articulated the intraenterprise immunity doctrine in Copperweld Corp. v. Independence Tube Corp.,47 in which the Court had to decide whether a parent corporation and its wholly owned subsidiary were capable of “conspiring” for purposes of Section 1. In answering that question in the negative, the Court began by noting two fundamental and distinct screens inherent within the Sherman Act’s structure: (1) Section 1 prohibits only “concerted” conduct but does not require that the defendant(s) possess market power; and (2) Section 2 reaches “unilateral” conduct but generally requires actual market power or a dangerous probability of attaining it. Because Section 1 lacks Section 2’s market power screen, the Court reasoned, it is important to honor its concerted conduct screen by finding a Section 1 violation only when a literal combination has “deprive[d] the marketplace of independent centers of decisionmaking” by joining two entities that would otherwise be expected to pursue their own, perhaps divergent, interests. When it comes to a parent corporation and the wholly owned subsidiary that it fully controls and whose gain and loss it captures in full, the Court reasoned, divergent interests are impossible. Thus, the Court concluded, a literal combination between a parent corporation and its wholly owned subsidiary cannot constitute a contract, combination, or conspiracy for purposes of Section 1.
Copperweld was a bit of a mixed bag. The Court’s desire to eliminate liability under Section 1 for literal combinations such as the one at issue in the case was certainly laudable. When a literal combination involves no actual or threatened market power and does not deprive the market of any independent center of economic decision making, it is highly unlikely to harm consumers; instead, the literal combination has probably been effected because it is efficient. Assigning liability for such combinations would thwart efficient relationships without providing any benefit for consumers. A Copperweld doctrine that allows early termination of conspiracy claims premised on harmless intraenterprise combinations thus seems desirable from a decision-theoretic perspective.
In terms of its direction on how to identify literal combinations that should be immune from Section 1 scrutiny, though, Copperweld was a mess. In concluding that Copperweld and its wholly owned subsidiary were incapable of conspiring, the Court reasoned that:
a parent and its wholly owned subsidiary have a complete unity of interest. Their objectives are common, not disparate; their general corporate actions are guided or determined not by two separate corporate consciousnesses, but one. 
While this is all true, the Court created some confusion by simultaneously emphasizing the “unity of interest” of a parent and its wholly owned subsidiary and the fact that the two entities are subject to common control. Implementation difficulties were bound to arise because unity of interest and common control need not follow each other. Firms with unified interests may lack common control, and commonly controlled business divisions may diverge in their interests.
In applying Copperweld, lower courts generally latched onto the “unity of interests” language, looking to see if the parties to the purported agreement face any divergence in their incentives. This proved problematic. For one thing, focusing on whether the combining units share a unity of interests led to significant divergence in outcomes. Some courts construed unity of interests broadly, holding, for example, that pure sister corporations (wholly owned subsidiaries of a common parent) merit Copperweld immunity, that a franchisor and its franchisees could be a single entity, that separately owned franchisees may constitute a single entity, and that one firm’s ownership of a bare majority of the other’s stock creates a single entity. Other courts were disinclined to find a unity of interests, ruling, for example, that sibling corporations sharing a common parent are not a single entity and that a parent and subsidiary corporation are not a single entity if more than a de minimis percentage of the subsidiary’s stock (less than 10 percent or so) is owned by someone other than the parent.
In addition to creating implementation difficulties, the focus on whether business units share a unity of interests ultimately seems inapposite to whether they are, in reality, a single economic entity. There are often incentive conflicts among agents within what is obviously a single firm, and within obvious cartels there is frequently no divergence in interests. A more economically sensible approach would endeavor to immunize from antitrust liability those literal combinations involving parties who are subject to common control. Such an approach would correspond to the economic understanding of the firm, which consists of a profit-seeking association in which resources are allocated according to managerial fiat in order to reduce transaction costs, avoid hold-up problems resulting from asset-specific investments, and create incentives for performance.
That, however, is not the tack the Court took in American Needle. Instead, it disregarded control questions and focused exclusively on whether the defendants possessed a complete unity of interests. In holding that the NFL members could conspire in jointly authorizing NFLP to grant exclusive licenses, the Court emphasized that “[although [they] have common interests such as promoting the NFL brand, they are still separate, profit-maximizing entities, and their interests in licensing team trademarks are not necessarily aligned.” The Court then observed that while “[c]ommon interests in the NFL partially unite the economic interests of the parent firms, ... the teams still have distinct, potentially competing interests.”
The Court was not persuaded that because a joint venture is necessary to produce NFL football, promotion of the jointly produced product (including the licensing of intellectual property) should be deemed unilateral conduct of the single joint venture. It stated that “[t]he justification for cooperation is not relevant to whether that cooperation is concerted or independent action,” and it emphasized that “necessity of cooperation” does not necessarily “transform concerted action into independent action.” Rather, the need for cooperation is relevant to (1) whether concerted conduct is evaluated under the rule of reason; and (2) how that conduct fares under the rule. Indeed, the Court emphasized that although the NFL members’ joint conduct would not be exempt from liability under the intraenterprise immunity doctrine, it might still pass muster under a rule of reason analysis. In sum, the Court seemed to reason that only a complete unity of interests will invoke the intraenterprise immunity doctrine and that the need for joint conduct to produce a product is not enough to render that conduct unilateral, but that the rule of reason may acquit joint actions that appear to be output enhancing.
How could this analysis, which weakens the degree to which the intraenterprise immunity screen may weed out meritless antitrust actions and permits more claims to proceed to discovery, comport with decision theory and a sensitivity to the limits of antitrust? Consideration of the justices’ questions at oral argument suggests an answer to that question. As Stone and Wright observe, much of the discussion at oral argument centered on the relative costs and benefits of intraenterprise immunity and rule of reason adjudication as alternative means of screening out meritless antitrust conspiracy claims; the Court “to its credit was very much focused on the ‘compared to what?’ question,” apparently seeking to construct screening mechanisms in a manner that would minimize administrative and error costs. In the end, the Court surmised that Copperweld, which had generated tremendous confusion among the lower courts and had led to extensive and costly disputes over singleentity status, was not a very cost-effective screening mechanism.
There are, however, alternative methods for screening out antitrust conspiracy claims involving related entities whose combination would not seem to threaten consumer harm. The most obvious one, the focus of the justices’ questioning, is the rule of reason. Since the time Copperweld was decided, courts and commentators have provided greater “structure” to the rule of reason as applied to joint ventures, making the rule easier to administer, more predictable, and less prone to generate errors (and thus error costs).
Of course, rule of reason adjudication occurs after costly discovery, so the rule may not provide the optimal device for screening out patently meritless conspiracy claims. There is, though, another screen. As Stone and Wright emphasize, Twombly’s requirement that antitrust conspiracy plaintiffs plead a “plausible” claim, including a plausible theory of anticompetitive harm, provides an additional mechanism for screening out meritless conspiracy actions. They explain that “Twombly dismissals indeed satisfy both components of a workable substitute for Copperweld immunity—they both allow for an early [prediscovery] dismissal of marginal antitrust cases and force antitrust plaintiffs to articulate theories of anticompetitive harm grounded in economics.” Thus, the advent of a structured, more predictable, “cheaper” rule of reason, coupled with more stringent pleading standards requiring plaintiffs to set forth a “plausible” theory of anticompetitive harm, enabled the Court to jettison another costly screening mechanism. When American Needle is read, not in isolation but in light of the Court’s entire Section 1 jurisprudence, it appears to be consistent with an effort to minimize the sum of decision and error costs related to antitrust adjudication.
-  560 U.S. 183 (2010).
-  See Judd E. Stone & Joshua D. Wright, Antitrust Formalism Is Dead! Long Live AntitrustFormalism!—Some Implications of American Needle v. NFL, 2009-2010 Cato Supreme Ct. Rev. 369.
-  467 U.S. 752 (1984).
-  Copperweld, 467 U.S. at 771.
-  Stone & Wright, supra note 44, at 379-80.
-  American Needle, 560 U.S. at 198 (emphasis added).
-  Id. (emphasis in original).
-  Id. at 199.
-  Stone & Wright, supra note 44, at 392.
-  See Bell Atlantic Corp. v. Twombly, 550 U.S. 544, 570 (2007) (to survive motion to dismiss,antitrust plaintiff must plead “enough facts to state a claim to relief that is plausible on its face”).
-  Stone & Wright, supra note 44, at 403.