Competitive Enterprise Institute v. FCC

970 F.3d 372

• Court: Court of Appeals for the D.C. Circuit
• Decided: August 14, 2020
• Docket: 18-1281
• Status: Published

Opinion

United States Court of Appeals FOR THE DISTRICT OF COLUMBIA CIRCUIT

Argued September 5, 2019 Decided August 14, 2020

No. 18-1281

COMPETITIVE ENTERPRISE INSTITUTE, ET AL., APPELLANTS

v.

FEDERAL COMMUNICATIONS COMMISSION, APPELLEE

On Appeal from Orders of the Federal Communications Commission

Melissa Holyoak argued the cause for appellants. With her on the briefs were Theodore H. Frank and Sam Kazman.

Thaila Sundaresan, Counsel, Federal Communications Commission, argued the cause for appellee. With her on the brief were Thomas M. Johnson Jr., General Counsel, David M. Gossett, Deputy General Counsel, and Richard K. Welch, Deputy Associate General Counsel. Jacob M. Lewis, Associate General Counsel, entered an appearance.

Before: HENDERSON and KATSAS, Circuit Judges, and SENTELLE, Senior Circuit Judge.

Opinion for the Court filed by Circuit Judge KATSAS. 2 Dissenting opinion filed by Senior Circuit Judge SENTELLE.

KATSAS, Circuit Judge: This appeal involves conditions imposed by the Federal Communications Commission on a merger of three cable companies. The conditions regulate in detail how the merged entity, which we call New Charter, may provide cable broadband Internet service. Among other things, the conditions (1) prohibit New Charter from charging programming suppliers for access to its broadband subscribers, (2) prohibit New Charter from charging broadband subscribers based on how much data they use, (3) require New Charter to provide steeply discounted broadband service to needy subscribers, and (4) require New Charter to substantially expand its cable infrastructure for broadband service.

The appellants include three of New Charter’s customers, whose bills for cable broadband Internet service increased shortly after the merger. They contend that the conditions caused this injury, which would likely be redressed by an order setting the conditions aside. We hold that these appellants have standing to challenge the first and third conditions, which we vacate given the FCC’s refusal to defend on the merits.

I

A

The Communications Act of 1934, Pub. L. No. 73-416, 48 Stat. 1064, empowers the FCC to regulate communications by wire or radio. Title I of the Act, 47 U.S.C. §§ 151–163, gives the FCC ancillary regulatory authority over any “information service.” See id. § 153(24). Title II of the Act, id. §§ 201–276, subjects most providers of a “telecommunications service” to regulation as common carriers. See id. § 153(11), (50)–(53). 3 Title III of the Act, id. §§ 301–399b, provides for regulation of wireless radio communications.

Within Title II, section 214(a) prohibits common carriers from constructing, operating, or acquiring any new or extended communications line without first obtaining from the FCC “a certificate that the present or future public convenience and necessity require or will require the construction, or operation, or construction and operation, of such additional or extended line.” 47 U.S.C. § 214(a). Section 214(c) provides that the FCC “may attach to the issuance of the certificate such terms and conditions as in its judgment the public convenience and necessity may require.” Id. § 214(c).

Title III creates a licensing scheme for wireless radio communications. Section 301 requires a station license to make radio transmissions. See 47 U.S.C. § 301. Section 307(a) requires the FCC to grant any applicant such a license if the “public convenience, interest, or necessity will be served thereby.” Id. § 307(a). Section 308 sets forth citizenship, character, fitness, and technical requirements for holding a station license. Id. § 308. Section 310(d) prohibits transferring any license without an FCC finding that “the public interest, convenience, and necessity will be served” by the transfer, and it requires transfer applications to be adjudicated as if the transferee “were making application under section 308.” Id. § 310(d).

The disputed merger conditions involve cable broadband Internet service, which gives subscribers the ability to send and receive data over the Internet. The FCC has shifted positions on whether this is an “information service” subject to regulation under Title I or a “telecommunications service” subject to common-carrier regulation under Title II. In a 2002 rulemaking, the agency concluded that cable broadband 4 Internet service is not subject to regulation under Title II, In re Inquiry Concerning High-Speed Access to the Internet over Cable & Other Facilities, 17 FCC Rcd. 4798, 4802, ¶ 7 (Mar. 15, 2002) (2002 Title II Order), and the Supreme Court upheld that position, Nat’l Cable & Telecomm. Ass’n v. Brand X Internet Servs., 545 U.S. 967 (2005). In 2015, the FCC reinterpreted Title II to encompass cable broadband Internet service, but it decided to forbear from requiring a section 214(a) certificate to provide that service. In re Protecting and Promoting the Open Internet, 30 FCC Rcd. 5601, 5610, 5848– 49, ¶¶ 29, 511 (Apr. 3, 2015) (2015 Title II Order). In 2018, the FCC again reversed course and concluded that cable broadband Internet service is not subject to regulation under Title II. In re Restoring Internet Freedom, 33 FCC Rcd. 311, 312, ¶ 2 (Jan. 4, 2018) (2018 Title II Order). Following Brand X, this Court upheld that interpretation. Mozilla Corp. v. FCC, 940 F.3d 1, 18–35 (D.C. Cir. 2019) (per curiam).

The FCC takes an expansive view of its authority to review license transfers incident to the merger of telecommunications companies. In particular, the agency thinks itself empowered to consider not only whether the “construction” or “operation” of specific cable lines would be in the public interest at the time of a merger, 47 U.S.C. § 214(a), and not only whether the merged entity satisfies the requirements for holding radio licenses “under section 308,” id. § 310(d), but also whether the merger itself would be in the public interest. See, e.g., In re Applications of Charter Commc’ns, Inc., Time Warner Cable, Inc., and Advance/Newhouse P’ship, 31 FCC Rcd. 6327, 6336–39 (May 10, 2016) (New Charter Order). The FCC thus duplicates the analysis of the Department of Justice in its review of possible anticompetitive effects. See id. at 6337–38. But the Commission further considers “diversity, localism, [and] other public interest considerations” besides antitrust ones. Id. at 6338. It thus seeks to impose conditions that 5 “confirm specific benefits or remedy harms likely to arise from transactions” under consideration. Id. at 6339. These conditions often regulate the terms of providing cable broadband Internet service, even though cable companies have never had to secure certificates under section 214(a) or licenses under section 301 in order to provide that service. Unlike the Justice Department, the Commission can effectively block mergers without going to court, simply by withholding approval of the transfer of these licenses.

B

New Charter, officially Charter Communications, Inc., was formed by the merger of Charter Communications, Inc., Time Warner Cable Inc., and Bright House Networks, LLC. Each of the merging companies had been engaged in various communications businesses, including the provision of cable broadband Internet service. Before the merger, Charter provided this service to some five million subscribers; Time Warner, to twelve million; and Bright House, to two million. New Charter—with about nineteen million subscribers— would become one of the largest cable broadband Internet providers in the United States.

To consummate the merger, the three companies applied to the FCC under sections 214(a) and 310(d) for permission to transfer their various cable and radio licenses to New Charter. These included certificates under section 214(a), cable television relay service licenses, and various wireless licenses. The FCC invited public comments on the application, and responses poured in. The index of filings in the agency docket spans 47 pages, not counting almost 170,000 comments made directly on the agency’s website through an online form.

The FCC ultimately approved the transfer of the licenses. The agency concluded that the “public interest benefits” of the 6 merger would outweigh its “public interest harms,” but only with six elaborate conditions. New Charter Order, 31 FCC Rcd. at 6530. The conditions are set forth in an appendix comprising 24 pages of fine print, including provisions for compliance, reporting, enforcement, and penalties for violations. See id. at 6539–62.

Four of the conditions, which address the provision of cable broadband Internet service, are at issue here. First, for seven years, New Charter cannot charge programming suppliers for access to its network of Internet subscribers. New Charter Order, 31 FCC Rcd. at 6540–42. Second, for seven years, New Charter may neither charge Internet subscribers based on actual data usage nor impose data usage caps. Id. at 6543–44. Third, New Charter must provide Internet service at steeply discounted prices to at least 525,000 low-income households within four years. Id. at 6547–49. Fourth, New Charter must build out its cable infrastructure to offer Internet service “to at least 2 million additional mass market customer locations” within five years. Id. at 6544–47. The FCC reasoned that the first two conditions were necessary to mitigate potential anti-competitive effects on suppliers of programs that consumers may watch on the Internet. Id. at 6362–91. The last two conditions, according to the Commission, would increase the merger’s public-interest benefits. Id. at 6504–07, 6528–40.

Two commissioners dissented. Commissioner Pai voted to block the merger. He argued that the FCC had “turned the transaction into a vehicle for advancing its ambitious agenda to micromanage the Internet economy.” New Charter Order, 31 FCC Rcd. at 6666. He criticized each of the conditions as either “radical,” addressed to issues unrelated to the merger, or otherwise arbitrary. See id. at 6666–68. He also objected to the FCC’s process for crafting the conditions, which he said 7 had involved a politicized, closed-door negotiation between the applicants and the Office of the Chairman. Id. at 6669. Commissioner O’Rielly voted to approve the merger but dissented from the conditions. He argued that sections 214(a) and 310(d) authorize only a narrow review focused on the transfer of individual licenses, not a review of entire mergers. Id. at 6671–72. Alternatively, he argued that any merger conditions must address problems caused by the merger itself, rather than pre-existing or independent problems. Id. at 6672– 74. And because the requirements for a low-income program and expanded infrastructure were neither “license-specific” nor “transaction-specific,” id. at 6672–74, Commissioner O’Rielly concluded that they “reside somewhere in the space between absurdity and corruption.” Id. at 6674.

The merging companies acceded to the conditions and formed New Charter on May 18, 2016.

C

The appellants are four consumers and a consumer- advocacy organization. John France, Daniel Frank, Jean- Claude Gruffat, and Charles Haywood each previously purchased cable broadband Internet service from one of the merging companies, and each continues to subscribe to New Charter. They contend that the merger conditions have caused their Internet bills to rise. Gruffat also serves on the board of the Competitive Enterprise Institute, which describes itself as an organization dedicated to the principles of limited constitutional government and free enterprise. CEI claims associational standing based on Gruffat’s board membership.

CEI filed a comment supporting the merger with the FCC. It urged the Commission to ensure that any conditions “are relevant to the particular [license] transfers at issue—not the merger as a whole.” Competitive Enterprise Institute, 8 Comment on Proposed New Charter Merger at 6 (Oct. 13, 2015). CEI further argued that the FCC cannot “impose conditions to remedy pre-existing harms or harms that are unrelated to the transaction.” Id. at 8–9. The individual subscribers did not file comments at that time.

After the FCC approved the merger, the individual subscribers and CEI jointly filed a petition for reconsideration, which sought removal of the four conditions. After the agency failed to act on the petition within the statutory 90-day deadline, see 47 U.S.C. § 405(a), the petitioners sought mandamus to compel it to act, In re Competitive Enter. Inst., No. 17-1261 (D.C. Cir. Dec. 12, 2017). One week before oral argument in this Court, the FCC finally denied reconsideration, thus mooting the mandamus action. After waiting for two years to issue an order, the agency offered only four pages of reasoning. It concluded that the petitioners were procedurally barred from challenging the conditions and lacked standing to do so under FCC rules. In re Applications of Charter Commc’ns, Inc., Time Warner Cable Inc., and Advance/Newhouse P’ship, 2018 WL 4347182 (Sept. 10, 2018).

On appeal to this Court, the four subscribers and CEI now seek review of both the New Charter Order and the order denying their petition for reconsideration.

II

We begin, as we must, with questions of our own jurisdiction. See Steel Co. v. Citizens for a Better Env’t, 523 U.S. 83, 93–96 (1998). We hold that the appellants have properly invoked our statutory jurisdiction and that three of 9 them have Article III standing to challenge two of the four disputed conditions.

The Communications Act permits appeals to this Court by any person “aggrieved” or “adversely affected” by an FCC order falling into any of five categories, 47 U.S.C. § 402(b)(6), including orders denying an application to transfer an “instrument of authorization,” id. § 402(b)(3). As explained above, the New Charter Order denied the unencumbered transfer of both section 214 certificates and radio station licenses. A party is “aggrieved” under section 402(b)(6) “if it satisfies both the constitutional and prudential requirements for standing.” New World Radio, Inc. v. FCC, 294 F.3d 164, 169 (D.C. Cir. 2002). Prudential standing is now understood as a question of “who may invoke the cause of action” at issue. Lexmark Int’l, Inc. v. Static Control Components, Inc., 572 U.S. 118, 130 (2014). That is a forfeitable issue, see id. at 128 n.4; Am. Inst. of Certified Pub. Accountants v. IRS, 804 F.3d 1193, 1199 (D.C. Cir 2015), which the FCC has forfeited by not contesting it in this case. Thus, if the appellants satisfy the constitutional standing requirements of Article III, they may seek review of the New Charter Order under section 402(b)(6).

The FCC contends that the individual appellants forfeited any right to seek review of the New Charter Order by not filing comments in the initial agency proceeding. The FCC reasons in two steps: first, they could not seek reconsideration because they failed to file comments earlier in the proceeding; and second, because they could not properly seek reconsideration before the agency, they cannot seek judicial review. We reject the second point and thus need not reach the first.

The Communications Act provides that “[t]he filing of a petition for reconsideration shall not be a condition precedent 10 to judicial review of any [FCC] order … except where the party seeking such review (1) was not a party to the proceedings resulting in such order …, or (2) relies on questions of fact or law upon which the Commission, or designated authority within the Commission, has been afforded no opportunity to pass.” 47 U.S.C. § 405(a). We have held that even a non-party to FCC proceedings may seek judicial review if the Commission had an “opportunity to pass” on its claims. Office of Commc’n of United Church of Christ v. FCC, 779 F.2d 702, 706–07 (D.C. Cir. 1985). And we have adhered to this precedent despite criticism that it is inconsistent with the statute’s plain language. WSB, Inc. v. FCC, 85 F.3d 695, 698 n.7 (D.C. Cir. 1996).

We have further held that the FCC can have an “opportunity to pass” on a question even if the party seeking judicial review never raised it with the agency. Time Warner Entm’t Co., L.P. v. FCC, 144 F.3d 75, 79 (D.C. Cir. 1998). The FCC has such an opportunity when another party raises the issue, Bartholdi Cable Co. v. FCC, 114 F.3d 274, 280 (D.C. Cir. 1997); when a dissenting Commissioner raises the issue, ICO Glob. Commc’ns (Holdings) Ltd. v. FCC, 428 F.3d 264, 269 (D.C. Cir. 2005); or when the agency addresses it anyway, EchoStar Satellite L.L.C. v. FCC, 704 F.3d 992, 996 (D.C. Cir. 2013).

The New Charter subscribers easily clear this hurdle. To begin, CEI filed comments arguing that the FCC could consider only whether the transfer of individual licenses was in the public interest and could not consider issues unrelated to the merger itself. Likewise, the dissents of Commissioners Pai and O’Rielly made the same global objections and further criticized each of the conditions at issue. And the agency itself undertook to justify each of those conditions. The FCC not only had an “opportunity to pass” on the conditions, but did pass on each 11 one at length. We therefore reject the FCC’s argument that section 405(a) bars the individual appellants from seeking review of the New Charter Order.1

We now turn to constitutional standing, which is necessary to establish a “case or controversy” within the meaning of Article III. Spokeo, Inc. v. Robins, 136 S. Ct. 1540, 1547 (2016). For Article III standing, the appellants “must have (1) suffered an injury in fact, (2) that is fairly traceable to the challenged conduct of the [appellee], and (3) that is likely to be redressed by a favorable judicial decision.” Id. An injury in fact is “an invasion of a legally protected interest that is concrete and particularized and actual or imminent, not conjectural or hypothetical.” Id. at 1548 (quotation marks omitted). The party invoking federal jurisdiction must prove each of these elements. Lujan v. Defs. of Wildlife, 504 U.S. 555, 561 (1992).

Where traceability and redressability depend on the conduct of a third party not before the court, “standing is not precluded, but it is ordinarily substantially more difficult to establish.” Defs. of Wildlife, 504 U.S. at 562 (quotation marks omitted). The party invoking our jurisdiction must show that the third party will act “in such manner as to produce causation

1 While this point is not critical for the analysis that follows, we note that the appellants also may seek review of the order denying their petition for reconsideration. Section 405(b)(2) of the Communications Act provides that FCC orders denying reconsideration “may be appealed under section 402(a),” which, for orders not covered by section 402(b), permits judicial review through the Hobbs Act, 28 U.S.C. § 2342. The appellants properly invoked these provisions to seek review of the order denying reconsideration. 12 and permit redressability of injury.” Id. A permissible theory of standing “does not rest on mere speculation about the decisions of third parties; it relies instead on the predictable effect of Government action on the decisions of third parties.” Dep’t of Commerce v. New York, 139 S. Ct. 2551, 2566 (2019).

In many cases, we have found standing where third-party conduct has been adequately proven. To pick just a few examples: In CEI v. NHTSA, 901 F.2d 107 (D.C. Cir. 1990), we held that a consumer organization had standing to challenge fuel-efficiency regulations based on evidence that non-party manufacturers, if given the choice, would be “substantially likely to respond to market forces” by producing larger vehicles desired by its members. Id. at 117. In Tozzi v. HHS, 271 F.3d 301 (D.C. Cir. 2001), we held that a manufacturer had standing to challenge an agency decision classifying a chemical in its product as a known carcinogen, based on evidence that third parties would be more likely to buy the product without the classification. Id. at 307–11. In Teton Historic Aviation Foundation v. DoD, 785 F.3d 719 (2015) (per curiam), we held that an organization seeking to buy aircraft parts had standing to challenge a Department of Defense policy limiting their sale. Despite the absence of any legal compulsion to sell, we credited evidence that the Department likely would sell through a specific contractor, who in turn likely would auction the parts to the public. See id. at 727–28 (“We have previously found standing in cases where a third party would very likely alter its behavior based on our decision, even if not bound by it.”). And in Energy Future Coalition v. EPA, 793 F.3d 141 (D.C. Cir. 2015), we held that biofuel producers had standing to challenge a rule prohibiting non-party manufacturers from using biofuel in emissions testing, because there was “substantial reason to think that at least some vehicle manufacturers would use” biofuel if that option were legally permitted. Id. at 144. 13 In considering the likely reaction of third parties, we may consider a variety of evidence, including “the agency’s own factfinding,” CEI, 901 F.2d at 114; affidavits submitted by the parties, Sierra Club v. EPA, 292 F.3d 895, 898–99 (D.C. Cir. 2002); evidence in the administrative record, id. at 900–01; arguments “firmly rooted in the basic laws of economics,” United Transp. Union v. ICC, 891 F.2d 908, 912 n.7 (D.C. Cir. 1989); and conclusions in other agency orders and rulemakings, CEI, 901 F.2d at 115–17.

As “standing is not dispensed in gross,” the appellants here must separately prove standing “for each claim” that they seek to press. Davis v. FEC, 554 U.S. 724, 734 (2008) (cleaned up). We must therefore separately assess their standing to challenge each of the disputed conditions.

The five appellants raise interrelated theories of standing. France, Frank, and Haywood argue that the merger conditions caused higher prices for the Internet service that they buy from New Charter. In support of that claim, each of them offered evidence that New Charter raised their Internet charges shortly after the merger. Gruffat alleges the same injury, but without evidence of higher bills. Finally, CEI asserts associational standing based on Gruffat’s board membership and individual injuries. CEI and Gruffat thus present less evidence for standing than the other consumers, despite asserting the same theory of harm. Since all appellants raise the same merits arguments and seek the same relief, we may assess standing only for the three individual consumers. See Comcast Corp. v. FCC, 579 F.3d 1, 6 (D.C. Cir. 2009). 14 1

The first challenged condition requires New Charter to “interconnect” its network “on a settlement-free basis” with parties supplying data to its subscribers. New Charter Order, 31 FCC Rcd. at 6559. Some background explains the jargon. As a broadband Internet provider, New Charter connects the personal devices of individual subscribers to the rest of the Internet. To do so, it negotiates agreements governing the exchange of Internet traffic and “the compensation, if any, to be paid by one party to the other.” Id. at 6375. Agreements without payments are called “settlement-free.” Id. Many interconnection agreements are made between broadband Internet providers and “edge providers” such as Netflix—i.e., those who provide content to consumers through the Internet. See Verizon v. FCC, 740 F.3d 623, 629 (D.C. Cir. 2014). Since broadband providers allow edge providers to reach their subscribers, the broadband providers often can extract payments from edge providers. The disputed condition prohibits New Charter from doing so.

The consumers argue that requiring New Charter to use free interconnection agreements—and thus to forgo revenue from edge providers—injures them in two ways: by increasing Internet prices and decreasing Internet quality. The three individual consumers have provided evidence that their cable bills increased after the merger. Increased Internet prices are “certainly an injury-in-fact,” Consumer Fed’n of Am. v. FCC, 348 F.3d 1009, 1012 (D.C. Cir. 2003), and next month’s cable bill is “certainly impending,” Clapper v. Amnesty Int’l USA, 568 U.S. 398, 409 (2013) (cleaned up). In contrast, the allegations about decreased quality do not pass muster. The consumers provide no evidence that quality declined after the merger nor anything else suggesting a significant possibility of future declines. Moreover, they do not spell out how quality 15 might suffer going forward—whether by lower speeds, slower ping, more frequent outages, or some other measure. For these reasons, the alleged quality injury is too abstract.

As for causation and redressability, the consumers propose a relatively simple causal chain. By requiring New Charter to forgo revenue from edge providers, the condition caused New Charter to raise prices on broadband subscribers. And vacating the condition would redress this injury because New Charter likely would respond by raising revenue from edge providers and lowering charges to subscribers.

To begin, the condition plainly caused New Charter to forgo revenue from edge providers. Before the merger, Time Warner, the largest broadband provider among the merging companies, raised substantial revenue from paid interconnection agreements. See New Charter Order, 31 FCC Rcd. at 6377–78, 6585. So did Bright House. Id. at 6377. But the merger condition prohibits New Charter from using those same revenue sources.

It is also clear that the consumers’ bills increased shortly after the merger. Before the merger, France and Haywood subscribed to Bright House’s broadband service, and Frank subscribed to Time Warner’s. Shortly after, New Charter raised their monthly bills: France’s bill increased about 20 percent, from $84 to $101, Haywood’s about 40 percent, from $51 to $71; and Frank’s about 5 percent, from $75.99 to $79.99.

The consumers also marshal evidence connecting the increased prices to the merger condition. Dr. Robert Crandall, a professor in the field of telecommunications economics, explained the connection based on how pricing works in two- sided markets. Without the condition, New Charter “would find it profitable to reduce its subscriber charges somewhat to 16 attract more subscribers and thus greater revenues from interconnection fees.” Appellants’ Add. at 28 (Crandall Decl.). By prohibiting such revenues, the condition removes this incentive to lower consumer prices. In other words, pricing in this market is like a waterbed—push down on one side, and the other side goes up. Crandall’s analysis, which the FCC does not meaningfully contest, is “firmly rooted in the basic laws of economics.” United Transp. Union, 891 F.2d at 912 n.7. And it tracks analyses by the Supreme Court and the FCC itself.

In Ohio v. American Express Co., 138 S. Ct. 2274 (2018), the Supreme Court recently explained pricing dynamics in two- sided markets. “As the name implies … two-sided [markets] offer[] different products or services to two different groups who both depend on the [provider] to intermediate between them.” Id. at 2280. In such markets, “indirect network effects” influence product pricing. Id. In the credit card market, for example, “[a] credit card … is more valuable to cardholders when more merchants accept it, and is more valuable to merchants when more cardholders use it.” Id. at 2281. In two- sided markets, firms “therefore must take these indirect network effects into account before making a change in price on either side.” Id.

The New Charter Order also makes this point. The FCC explained that broadband Internet providers “operate within a two-sided market,” with consumers at one end and edge providers at the other. 31 FCC Rcd. at 6374. Thus, “edge providers value interconnection with [broadband] providers more as the providers service more subscribers.” Id. This conclusion makes sense because New Charter connects subscribers and edge providers, much as American Express connects cardholders and merchants. And as Dr. Crandall explained, lower consumer prices will yield more subscribers, 17 which in turn will yield “greater revenues from interconnection fees.” Appellants’ Add. at 28.

In 2018, the FCC elaborated on this point in lifting a regulatory prohibition on paid interconnection agreements between broadband providers and edge providers. As the Commission explained: “increased prices from edge providers are to a potentially significant extent passed through to end users in the form of lower prices for broadband Internet access service.” 2018 Title II Order, 33 FCC Rcd. at 466. We upheld that analysis in Mozilla Corp. See 940 F.3d at 55–56. And even when the FCC sought to prohibit all paid interconnection agreements in 2015, it recognized the same relationship between edge-provider revenue and consumer Internet prices. See 2015 Title II Order, 30 FCC Rcd. at 5645. For these reasons, France, Frank, and Haywood have shown a substantial likelihood that that the prohibition on paid interconnection agreements caused their cable bills to increase.

The same evidence also proves redressability. Before the merger, the companies raised significant revenue from paid interconnection agreements, and the FCC concluded that New Charter would continue such agreements if allowed to do so. New Charter Order, 31 FCC Rcd. at 6378. Moreover, just as prohibiting paid interconnection agreements would likely cause broadband prices to rise, permitting those agreements would likely cause broadband prices to fall. As noted above, the FCC itself recognized as much in lifting its global ban on paid interconnection agreements. See 2018 Title II Order, 33 FCC Rcd. at 466. And Dr. Crandall confirmed that these economic principles have not changed. Thus, a favorable ruling is likely to redress the consumers’ financial injuries.

The FCC objects that other factors, such as increased servicing costs, might have caused the price increases. But the 18 agency offers only speculation on this point. In any event, the subscribers need not show that prohibiting paid interconnection agreements caused the entirety of the price increases, or even that it caused price increases of some specific amount. For standing purposes, even a small financial injury is enough, see Carpenters Indus. Council v. Zinke, 854 F.3d 1, 5 (D.C. Cir. 2017), and the consumers have shown a substantial likelihood that their bills are higher because of the prohibition on paid interconnection agreements.

The FCC and our dissenting colleague note that New Charter might not lower consumer prices even if we set aside the prohibition on paid interconnection agreements. Post, at 4. That is theoretically possible, but all we require is proof of a substantial likelihood. As explained above, an entire line of cases finds redressability, as well as causation, in comparable circumstances turning on third-party conduct that is voluntary but reasonably predictable. We concluded that an injury was redressable based on the likely choices of the manufacturers in CEI, 901 F.2d at 117, and Energy Future Coalition, 793 F.3d at 144–45; the buyers in Tozzi, 271 F.3d at 307–10; and the independent contractor in Teton, 785 F.3d at 126–30. Given the findings and evidence here, we reach a similar conclusion.

Our dissenting colleague spots us causation on the front end but stresses that “causation does not inevitably imply redressability,” because a “new status quo” may be “held in place by other forces” besides the government action at issue. Renal Physicians Ass’n v. HHS, 489 F.3d 1267, 1278 (D.C. Cir. 2007) (citing Nat’l Wrestling Coaches Ass’n v. Dep’t of Educ., 366 F.3d 930, 939–40 (D.C. Cir. 2004)); see post, at 2–4. We have no quarrel with that general proposition. But here, we can discern no “other forces” that might cause redressability on the back end to diverge from traceability on the front end. To the contrary, so far as the record reflects, the same market forces 19 that caused New Charter’s predecessor companies to secure paid interconnection agreements before the merger, and (according to the FCC) to lower consumer prices as a result, continue to operate in the two-sided market for broadband Internet service.

Our dissenting colleague further argues that Simon v. Eastern Kentucky Welfare Rights Organization, 426 U.S. 26 (1976), and Warth v. Seldin, 422 U.S. 490 (1975), preclude a finding of redressability here. Post, at 5–6. Neither case distinguished between traceability and redressability, and neither cuts against finding both causation requirements in this case. In Eastern Kentucky, the plaintiffs alleged that the Internal Revenue Service, by affording favorable tax treatment to certain hospitals, caused the hospitals not to provide free care to indigent patients. The Court dismissed for lack of standing because the allegations did not support a plausible inference that the hospitals would have chosen to provide free care but for the challenged tax benefit. 426 U.S. at 42–44. In Warth, the plaintiffs alleged that a zoning ordinance caused low-income housing to be unavailable. The Court dismissed for lack of standing because the plaintiffs provided no reason to conclude that developers would have built such housing but for the ordinance. 422 U.S. at 505–06. This case involves very different causal chains, market forces, and evidence. As shown above, Dr. Crandall’s declaration and the FCC’s own analysis establish a reasonable likelihood that the prohibition on paid interconnection agreements by New Charter caused, and still causes, higher prices for its broadband consumers.

The second challenged condition prohibits New Charter from charging subscribers based on how much data they transfer to their devices, whether directly or through data caps. 20 See New Charter Order, 31 FCC Rcd. at 6543–44. This requires the Internet equivalent of an all-you-can-eat buffet, rather than á-la-carte service. The customers persuasively argue that such pricing forces rare Internet users to subsidize frequent ones. Commissioner Pai made this objection in dissent. See id. at 6667 (“The elderly woman on a fixed income who uses the Internet to exchange e-mail messages with her grandchildren must pay more so that an affluent family watching online HD video for many hours each day can pay less.”). And the majority offered no response.

Nonetheless, the consumers have failed to prove causation because there is scant evidence that New Charter would offer usage-based pricing if allowed to do so. Before the merger, its predecessor companies rarely offered it. Charter had specifically rejected it. See New Charter Order, 31 FCC Rcd. at 6368. Time Warner offered one plan with usage-based pricing, but abandoned efforts to expand the practice after “significant public backlash.” Id. at 6363. Bright House never offered it. Id. at 6364. Given the lack of evidence that New Charter’s predecessor companies had offered usage-based pricing before the condition was imposed, or that New Charter would offer usage-based pricing if allowed to do so, the appellants have failed to show traceability or redressability.

The third challenged condition requires New Charter to offer steeply discounted Internet service to qualifying low- income individuals. Within four years, New Charter must enroll at least 525,000 households in a discounted broadband plan featuring 30 megabits-per-second (Mbps) download speed for only $14.99 a month. New Charter Order, 31 FCC Rcd. at 6547–49. For this program, New Charter must also provide, among other things, a free modem, a free “self-installation kit,” 21 free professional installation if self-installation would be too hard, a Wi-Fi router at a price set by the FCC, a dedicated phone number and website, and specially trained customer service representatives. See id. at 6547–48. To comply with these requirements, New Charter has offered a “Spectrum Internet Assist” program since April 2017.

The appellants have standing to challenge this set of conditions as likely causing higher prices for them. For causation and redressability, the appellants highlight Dr. Crandall’s conclusion that the low-income program will “likely” cause higher prices for other consumers. Appellants’ Add. 28. Likewise, Commissioner O’Rielly predicted in dissent that the condition would “result in increases in the cost of cable and broadband service for every current cable subscriber of the three companies,” New Charter Order, 31 FCC Rcd. at 6674, and the majority had no response.

These assessments are “firmly rooted in the basic laws of economics.” United Transp. Union, 891 F.2d at 912 n.7. The condition requires price discrimination—charging some customers less and others more for the same product. As the Supreme Court has long recognized, price discrimination operates “for the benefit of some favored persons at the expense of others.” ICC v. Balt. & Ohio R.R., 145 U.S. 263, 276 (1892). Likewise, the FCC has noted that “[t]he general effect of [price] discrimination is a redistribution of income from the customers discriminated against to the price discriminator or favored customers.” In re AT&T Co. Revisions to Tariff F.C.C. No. 259, Wide Area Telecommc’ns Serv. (WATS), 89 F.C.C.2d 889, 896 (Apr. 16, 1982). The Horizontal Merger Guidelines of the Department of Justice and the Federal Trade Commission explain why the discrimination is likely to inflate prices for the disfavored consumers— because “[a] price increase for targeted customers may be 22 profitable even if a price increase for all customers would not be profitable because too many other customers would substitute away.” Horizontal Merger Guidelines, § 3 (2010). Similarly, a leading commentator has explained that price discrimination allows businesses to “obtain higher rates of return” from the disfavored customers, H. Hovenkamp, Federal Antitrust Policy: The Law of Competition and its Practice, § 14.4 (4th ed. 2011), because charging lower prices to “more price-sensitive customers” allows firms to “avoid[] price reductions across the board,” Menasha Corp. v. News Am. Mktg. In-Store, Inc., 354 F.3d 661, 662 (7th Cir. 2004) (Easterbrook, J.). In sum, the price discrimination mandated by the FCC allows New Charter to increase prices for disfavored customers without having to worry about driving away low-income customers who are more price sensitive. The appellants have proven causation.

What remains is a distinct question of redressability— whether there is a substantial likelihood that New Charter would change course if allowed to do so. We think that there is. To begin, consider the past practices of the merging companies. Before the conditions were imposed, Charter and Time Warner offered no discounted services to low-income customers. Bright House did, but its program was much narrower than the one now mandated by the FCC. See New Charter Order, 31 FCC Rcd. at 6528 n.1482. With those facts in mind, the FCC itself found no reason to think that New Charter would voluntarily offer up what the agency compelled it to provide. Id. at 6529.

Moreover, the terms mandated by the FCC sharply depart from industry pricing. Beyond free installation and hardware, the conditions require New Charter to offer broadband service with 30 Mbps speed for only $14.99 a month. Before the merger, Time Warner charged $54.99 and Bright House 23 charged $74 for similarly fast service. See New Charter Order, 31 FCC Rcd. at 6372 nn.298–99. Moreover, the FCC catalogued prices from at least seven different broadband providers, and none offered service anywhere near as fast as Spectrum Internet Assist at anywhere near the same price. See id. at 6371–74. We thus think it unlikely that New Charter would retain the current program voluntarily.

In arguing to the contrary, the FCC points to Time Warner’s former “Everyday Low Price” plan, which also cost $14.99 per month. See New Charter Order, 31 FCC Rcd. at 6528 n.1482. But Spectrum Internet Assist offers download speeds fifteen times faster, which only tends to confirm that it is the product of agency compulsion. The FCC also notes that New Charter, in its license transfer application, offered to implement a discounted plan for low-income individuals. But given the FCC’s expansive view of its conditioning power, and with a $100 billion merger hanging in the balance, one may wonder “whether voluntary commitments are truly voluntary.” Id. at 6672 (O’Rielly dissent); see also id. at 6669 (Pai dissent). In any event, the mandated conditions went far beyond what New Charter had proposed. See id. at 6529–30. For example, New Charter proposed “build[ing] upon Bright House Networks’ broadband program for low-income consumers,” New Charter Applications, Public Interest Statement, FCC Dkt. No. 15-149, at 20 (filed June 25, 2015), but it was even slower than Time Warner’s Everyday Low Price plan, see New Charter Order, 31 FCC Rcd. at 6528 n.1482.

In sum, the appellants have shown a substantial likelihood that New Charter would narrow the Spectrum Internet Assist program if allowed to do so, which in turn would produce lower prices for subscribers who, like the individual appellants, are on short end of the price discrimination. The appellants have standing to challenge the discounted-services condition. 24 4

The buildout condition requires New Charter to create cable infrastructure necessary to offer broadband service “to at least 2 million additional mass market customer locations” within five years. New Charter Order, 31 FCC Rcd. at 6544, as modified by New Charter Applications, Order on Reconsideration, 32 FCC Rcd. 3238 (2017). This condition nicely illustrates our dissenting colleague’s point that traceability on the front end can sometimes diverge from redressability on the back end. By now, more than four years after the condition was imposed, New Charter already has built much of the required infrastructure, and its sunk costs in doing so cannot be recovered. Regardless of whether New Charter would have undertaken to build this infrastructure voluntarily, the consumers offer no reason to think that New Charter will abandon the project if now allowed to. Likewise, the consumers offer no reason to think that if New Charter were to abandon the project at this late date, thus ensuring a wasted investment, the decision to do so would somehow lower the prices for its broadband customers.

* * * *

The three individual appellants have standing to challenge the interconnection and discounted-services conditions, but not the usage-based pricing and buildout conditions.

III

On the merits, the appellants raise several troubling objections. For one thing, the governing statutes focus on individual licenses, not entire mergers: Section 214(a) authorizes the FCC to consider whether the “construction” or “operation” of a specific communications line is in the public interest at the time of an acquisition, while section 310(d) 25 authorizes it to consider whether a proposed transferee meets the specific criteria for holding a station license under section 308. Moreover, after broadening its focus to the entire merger, the FCC imposed conditions sweeping even beyond that. For example, the agency readily acknowledged that providing discounted service to needy consumers “is not a transaction- specific benefit,” but it nonetheless required New Charter to do so as a condition of approving the merger. New Charter Order, 31 FCC Rcd. at 6529. The Supreme Court has described such non-germane conditions as “an out-and-out plan of extortion.” Nollan v. Cal. Coastal Comm’n, 483 U.S. 825, 837 (1987) (quotation marks omitted). Commissioner O’Rielly made the same point in dissent: “Once delinked from the transaction itself, such conditions reside somewhere in the space between absurdity and corruption.” 31 FCC Rcd. at 6674. The conditions target the provision of broadband Internet service, which is not covered by Title II, much less by section 214(a), under the FCC’s current interpretation of the Communications Act. And to insinuate itself into that cable market, the FCC imposed conditions on the transfer of all licenses held by the appellants, including wireless licenses with no conceivable relevance to it.

We need not resolve these questions, however, for there is a simpler ground of decision. The lawfulness of the interconnection and discounted-services conditions are properly before us, yet the FCC declined to defend them on the merits. The agency’s only explanation for doing so was its view that we cannot reach the merits. Having lost on that question, the FCC has no further line of defense. “Because the Commission chose not to argue the merits in the alternative, we have no choice but to vacate the challenged portions of the order.” Time Warner, 144 F.3d at 82. 26 Two final housekeeping points. First, we set aside only the two conditions properly subject to review, for no party has asked us to set aside other portions of the New Charter Order as inseverable from them. Second, we dismiss as moot the appeal from the denial of reconsideration, for the appellants have now obtained full relief from the only two conditions that they have standing to challenge.

IV

For these reasons, we set aside the interconnection and discounted-services conditions in the New Charter Order, and we dismiss the remaining aspects of the appeal for lack of an appellant with Article III standing.

So ordered. SENTELLE, Senior Circuit Judge, dissenting: I express no opinion about the merits of this case, and I would not reach the merits at all because CEI lacks standing to challenge any of the proposed conditions. I do concur with the majority’s analysis and conclusion that CEI does not have standing to challenge the condition concerning charging subscribers based on data usage or the condition requiring New Charter buildout its cable infrastructure.

The Constitution defines a limited role for the federal courts, namely resolving cases and controversies. U.S. Const., art. III, §2, cl. 1; see, e.g., Chi. & Grand Trunk Ry. Co. v. Wellman, 143 U.S. 339, 345 (1892). Because Article III courts are courts of limited jurisdiction, we must examine our authority to hear a case before we can determine the merits. Steel Co. v. Citizens for a Better Env’t, 523 U.S. 83, 101–02 (1998). Standing is a doctrine that helps us “set[] apart the ‘Cases’ and ‘Controversies’ that are of the justiciable sort referred to in Article III” as opposed to disputes to be handled by the legislature or the executive. Lujan v. Defs. of Wildlife, 504 U.S. 555, 560 (1992). In order to satisfy the “irreducible constitutional minimum of standing,” the plaintiff or the petitioner must establish three essential elements. Id. It must demonstrate that it has suffered a “concrete and particularized” injury that is: 1) “actual or imminent,” id.; 2) caused by, or fairly traceable to, an act that the litigant challenges in the instant litigation, see Allen v. Wright, 468 U.S. 737, 752 (1984); and 3) capable of being redressed by a favorable decision of the court, see Simon v. E. Ky. Welfare Rights Org., 426 U.S. 26, 38 (1976). Standing ensures judicial intervention for only those disputes between adverse parties that are “in a form . . . capable of judicial resolution.” Schlesinger v. Reservists Comm. to Stop the War, 418 U.S. 208, 218 (1974) (quoting Flast v. Cohen, 392 U.S. 83, 101 (1968)).

The majority is perhaps correct that appellants have demonstrated injury to a legally protected interest, specifically 2 the obtaining of internet services at a lower rate, and may even have shown causation, but they have most assuredly not shown that this injury will be redressed by a favorable decision of the court. After the mandate issues in this case the bills for service will not thereby be diminished in any way nor will they ever, absent the volitional act of a third party.

Redressability and causation are often described as “two facets of a single causation requirement,” but, importantly, redressability “examines the causal connection between the alleged injury and the judicial relief requested,” which might not be present even if the injury is fairly traceable to the defendant’s actions. Allen, 468 U.S. at 753 n.19. In other words, “causation does not inevitably imply redressability. There might be some circumstances in which governmental action is a substantial contributing factor in bringing about a specific harm, but the undoing of the governmental action will not undo the harm, because the new status quo is held in place by other forces.” Renal Physicians Ass’n v. U.S. Dep’t of Health & Human Servs., 489 F.3d 1267, 1278 (D.C. Cir. 2007). The proposition that judicial intervention will undo the harm is often less clear in cases where the party inflicting the injury is a third party not before the court. In those circumstances, “much more is needed” to show that a plaintiff’s or petitioner’s injury will be redressed by a court order because the outcome “hinge[s] on the response of the regulated (or regulable) third party to the government action or inaction—and perhaps on the response of others as well.” Lujan, 504 U.S. at 562; see also E. Ky. Welfare, 426 U.S. at 41–43; Nat’l Wrestling Coaches Ass’n v. Dep’t of Educ., 366 F.3d 930, 940–41 (D.C. Cir. 2004). It is vital, then, that plaintiffs harmed by third parties show that that the dispute be one “capable of judicial resolution,” and that it is “likely, as opposed to merely speculative, that the injury will be redressed by a favorable decision.” Bennett v. Spear, 3 520 U.S. 154, 167 (1997); see also E. Ky. Welfare, 426 U.S. at 38.

In National Wrestling Coaches Association v. Department of Education, we articulated two situations where a court order could be said to redress injuries inflicted by third parties, but those are narrow circumstances. The first being when “the intervening choices of third parties are not truly independent of government policy.” Nat’l Wrestling, 366 F.3d at 941. A third party’s actions are not “truly independent” when the conduct would be illegal absent the government’s policy. Id. There are instances when those facts are present, such as when a party is injured by a third party through increased economic competition when that third party would not have been permitted to enter the market in question absent a change in government policy. Inv. Co. Inst. v. Camp, 401 U.S. 617, 620 (1971). Another example was when a candidate for political office sued the federal government when a city prevented him from running in a nonpartisan, general election because the city could not change its election format without approval from the federal government under the Voting Rights Act. LaRoque v. Holder, 650 F.3d 777, 790–91 (D.C. Cir. 2011).

The second is when “the record presented substantial evidence of a causal relationship between the government policy and the third-party conduct, leaving little doubt as to causation and the likelihood of redress.” Nat’l Wrestling, 366 F.3d at 941 (emphasis added). There was such evidence in Tozzi v. United States Department of Health and Human Services because the plaintiff submitted affidavits and other record evidence tying third-party decisions to stop carrying the plaintiff’s product to the government’s change in regulations. 271 F.3d 301, 308–09 (D.C. Cir. 2001). Similarly, in Block v. Meese, the plaintiff was injured by third parties not purchasing his film, but he introduced record evidence, including affidavits 4 from potential customers explaining that they had declined to purchase the film because the Department of Justice had labeled it political propaganda. 793 F.2d 1303, 1308 (D.C. Cir. 1986).

In this case, the injury is the increased price associated with purchasing New Charter’s services. But, as noted above, our decision will not reduce that price. New Charter, the third party whose actions are implicated in this case, may reduce the price voluntarily, but the voluntary actions of a third party are not enough to establish redressability, especially when the plaintiff introduces no evidence to support its allegations about what that third party is likely to do. The only evidence CEI cites in its brief is the idea that “the competitive market would restrict the cost increases to consumers.” Appellant Br. at 41. In support of that argument, CEI relies on statements made by Commissioner O’Reilly. There is no reference to statements made by New Charter or anyone with knowledge about the increase in price. For all we know, New Charter may be satisfied with the present rates or reluctant to rock a boat that is apparently sailing profitably.

The majority roots its conclusion of redressability in its confidence that the “basic laws of economics” will compel New Charter to provide appellants desired relief. Majority Op. at 16–18. I am not so sanguine. In the nontheoretical world, New Charter, which would apparently have standing to do so, did not bring this action. Even when CEI brought the action, New Charter made no move to intervene or even file an amicus brief on behalf of CEI. All this leads me to the conclusion that the majority’s finding of third-party redressability rests on speculation. This, in my view, leads to the broader conclusion that the use of the adverb “likely” in previous discussions of redressability refers to the likely result of the court’s judgment, not the likely volitional act of third parties not before the court. 5 This reading is not only consistent with the Article III requirement of case or controversy but also provides a clear delineation of jurisdiction not requiring speculation as to the “likely” responses to the court’s judgment of independent actors.

Significantly, neither of the cases cited by the majority in its discussion of “basic laws of economics” uses that term to support a finding of third-party-based redressability. In United Transportation Union v. ICC, our use of the term was part of a discussion that led to a finding of no standing. 891 F.2d 908, 912 n.7, 913–15 (D.C. Cir. 1989). In the Supreme Court’s discussion of economic principles in Ohio v. American Express Co., the Court was considering a question that had nothing to do with standing. __ U.S. __, 138 S. Ct. 2274, 2281–82 (2018). Under the majority’s approach, instead of being in a situation similar to that of Block or Tozzi, where we could say that the court order would likely redress the injury, we are left to guess what a large corporation, which just underwent major restructuring, would do. We are in much the same situation outlined in Simon v. Eastern Kentucky Welfare Rights Organization. In that case, the plaintiffs were injured when third-party hospitals denied service but argued that a change to the IRS’s policy, to make it more specific and restrictive, would discourage those hospitals from denying services. 426 U.S. at 30–34, 42. The Supreme Court ultimately determined it was “speculative whether the desired exercise of the court’s remedial powers in this suit would result in the availability to respondents of such services” because hospitals might continue to deny service to patients who could not pay for other economic reasons. Id. at 42–43. Just so in this case, we are left to speculate whether the desired exercise of our remedial power would in itself result in the availability of less expensive services. 6 The Supreme Court has also relied on the same reasoning regarding redressability when the question was about the outcome if the Court lifted a restrictive measure, rather than imposing one. In Warth v. Seldin, the city adopted a zoning ordinance that dictated “lot size, setback, floor area, and habitable space” and enforced it against developers that plaintiffs alleged “had the consequence of precluding the construction of housing suitable to their needs at prices they might be able to afford.” 422 U.S. 490, 495, 504 (1975). While the Supreme Court recognized there can be standing “[w]hen a governmental prohibition or restriction imposed on one party causes specific harm to a third party,” in Warth, it was likely that the plaintiffs’ inability to find affordable housing was “the consequence of the economics of the area housing market” and not the zoning ordinance. Id. at 505–06. Moreover, the plaintiffs relied “on little more than the remote possibility, unsubstantiated by allegations of fact, that their situation might have been better had respondents acted otherwise, and might improve were the court to afford relief.” Id. at 507. All told, there was not enough for the Court to find that the plaintiffs had standing. Id. at 508. CEI presents similarly unsubstantiated allegations about the likely effect of a court order in this case.

The majority’s citations support the proposition that probability about the actions of a third party are enough to demonstrate causation for standing purposes. But causation and redressability are not the same inquiry. In this case there is insufficient evidence to show that the injury to the consumer- appellants would be redressed if this court were to order the vacation of the conditions imposed by the government on New Charter. It may be that New Charter would take actions beneficial to the appellants, but it is not the case that this court can redress their injuries. I respectfully dissent.