Loading case…
United States Court of Appeals for the Fifth Circuit
Appeal from the United States District Court for the Southern District of Texas USDC No. 4:19-CV-4508
Before King, Smith, and Ramirez, Circuit Judges .
King, Circuit Judge :
A conservation district in Montgomery County, Texas, mandated a 30% reduction in groundwater usage by large volume groundwater users. To attain collective compliance, the San Jacinto River Authority (the 'River Authority') executed individual contracts with 80 utilities, including Quadvest, L.P. ('Quadvest'). Believing those contracts to be unlawful restraints of trade in violation of the Sherman Act, Quadvest brought this suit. After a ten-day bench trial, the district court sided with the River Authority. We AFFIRM .
The River Authority is “a political subdivision of the State of Texas, created in 1937 by the Texas Legislature to conserve, control, and utilize the storm and flood waters of the San Jacinto River and its tributary streams.” Quadvest, L.P. v. San Jacinto River Auth. , 7 F.4th 337, 340 (5th Cir. 2021). To those ends, it enjoys broad powers under its enabling statute, including to:
Id. at 349–51.
Page 3What it cannot do, however, is collect taxes. Instead, its major source of funding is from contract revenue services ( i.e. , rates it sets and collects).
To an extent, the River Authority is also a market player. It “provides groundwater services or wastewater services . . . in The Woodlands, based on an arrangement dating back to the 1970s.' But that is the only place it provides such services in, and it has not tried to grow or expand these services.
Quadvest is a family-operated, investor-owned utility that operates in Montgomery and surrounding counties to provide water and wastewater services. It relies solely on groundwater to provide those services.
When it entered into the contract at issue with the River Authority, it operated only in the retail market, selling to end-users of water. And as a retailer, it competes for contracts with developers for the right to sell retail water to customers in a particular geographic area. Once it secures a retail contract, it must obtain a permit and receive approval of its rates from the Texas Public Utility Commission.
A few years after executing the contract, Quadvest expanded into the wholesale market. The wholesale market is not regulated like the retail market. Wholesalers compete to contract with retailers or to provide supplemental water to those who need it. And the rates need not be approved by the Public Utility Commission and are set by contract.
Much of this case revolves around the difference between surface water and groundwater. To the consumer, the two are indistinguishable. But surface water is more expensive because of the infrastructure needed to produce, treat, and deliver it. For example, before the regulations giving rise Page 4to this suit took effect, the River Authority’s cost to produce surface water in Montgomery County was about $7.00 per thousand gallons, while Quadvest's cost to produce groundwater was about $1.69 per thousand gallons.
That cost differential meant groundwater was much more popular. So since the 1990s, the Texas Legislature has expressed concerns that existing supplies of groundwater may be insufficient to meet rising demands and thus be depleted.
To address that concern, in 2001, the Texas Legislature created the Lone Star Groundwater Conservation District (the “Conservation District”) and tasked it with managing groundwater in Montgomery County. The Conservation District is governed by a nine-member board of directors, and from the Conservation District’s creation until 2017, each director was appointed by relevant government stakeholders, such as County Commissioners Court, mayors, municipal utility districts, and the River Authority. After 2017, the board members were elected.
What the Conservation District saw was projected groundwater overuse. It concluded that the county's aquifers were 'recharging' at 64,000 acre-feet per year. But by 2010, groundwater demand was estimated to exceed 70,000 acre-feet per year, and by 2020, 82,000 acre-feet per year. The Conservation District thus sought to limit groundwater withdrawals to the recharge rate of 64,000 acre-feet per year.
So came the groundwater reduction rule (the 'Rule'). The Rule mandated a 30% reduction in groundwater usage by 'Large Volume Groundwater Users” (“LVGUs”) by 2015. An LVGU is any user pumping more than 10 million gallons of groundwater per year and includes Quadvest and the River Authority. Importantly, the Rule permitted collective compliance. That is, two or more LVGUs could join a reduction plan, under Page 5which some users would over-convert to surface water or alternative sources and others would under-convert; if in the aggregate, the collective achieved the 30% reduction, every member of the collective would be deemed to have complied, even if individual members may not have.
The River Authority proposed just the thing. The River Authority’s joint groundwater reduction plan (“Joint GRP”) proposed converting two of the largest groundwater producers in Montgomery County (The Woodlands and the City of Conroe) to reduce the burden of compliance on smaller entities in the plan and to leverage economies of scale. It would achieve that conversion by tapping into Lake Conroe, in which the River Authority holds a one-third interest of the annual permitted yield and the City of Houston holds the other two-thirds. All told, approximately 80 water utilities, supplying 80% of the water in Montgomery County, joined the Joint GRP. The Conservation District ultimately approved the Joint GRP.
To join the Joint GRP, each participating LVGU ('Participant') executed an individual contract (“GRP Contract”) with the River Authority. Quadvest did so on July 1, 2010, 'believe[ing] that it was 'buying compliance' with' the Rule. And in a letter to its customers, Quadvest described the Joint GRP as “the most economical way to reduce groundwater pumpage” and achieve compliance—“the best option available.” Indeed, compliance was crucial for Quadvest’s survival. Because the Conservation District imposed escalating penalties—starting at $500 per day in 2008 and increasing to $10,000 per day by 2011—Quadvest was facing up to $150,000 in fines per day for its 15 groundwater permits, more than its daily revenue. And noncompliance could result in permit revocations altogether, taking it out of the market.
Page 6The GRP Contracts-which are substantially identical among all Participants—essentially sold the River Authority’s compliance services in exchange for fees. For its part, the River Authority agreed to provide the following services:
And for its part, each Participant 'agree[d] to pay [the River Authority] certain fees, rates, and charges pursuant to the terms and provisions of th[e] Contract.”
Two main issues confronted the Joint GRP-cost and a free-rider problem. Paying Houston for its two-thirds interest in Lake Conroe water and building the necessary infrastructure to treat and transport the surface water were by no means going to be cheap. In fact, the total cost of this proposed venture was estimated to be approximately $480,000,000. The Texas Page 7Legislature, however, declined to let the River Authority impose fees directly on the LVGUs.
And like other collective actions, this one posed a free-rider problem. While some would over-transition to more expensive surface water, others would continue pumping cheaper groundwater while getting credit for compliance. Worse still, there would be no incentive for anyone to convert to surface water at all.
Two features of the GRP Contracts, central to this appeal, sought to address these issues. Quadvest calls them the “cost equalization” provision and “mandatory connection” provision. We explain them in turn.
The GRP Contract establishes two parallel fees: a pumpage fee assessed on each Participant’s groundwater production and a delivery fee charged for treated surface water delivered by the River Authority to the Participant. Each fee is a function of (1) the quantity of water pumped or delivered and (2) a “prevailing rate” for pumpage or delivery set by the River Authority.
In setting the prevailing rate, the River Authority is constrained by certain guiding principles. For the groundwater pumpage prevailing rate, it must be set so that, 'as nearly as practicable in [the River Authority's] reasonable determination[,] (i) the Participants are neither benefitted nor penalized for utilizing groundwater from Wells, and (ii) reasonable allowance is made for the Participants’ costs of operating and maintaining their Wells.” And the pumpage prevailing rate 'shall be equal and uniform among all classes of Participants that pump groundwater.” Similarly, the surface-water delivery prevailing rate must be set so that 'the Participants are neither benefitted nor penalized for being required to take Water from the Project Page 8under the GRP.” And both rates must be set “such that no special advantage or disadvantage is realized by any of the Participants.”
The fees have two purposes. First, they ensure that the costs to produce groundwater and surface water are equal so that those who overconvert to surface water for the compliance of the collective are not penalized. On the flipside, too, they ensure that those who remain on groundwater do not get to free-ride on the costly conversion of others.
Second, they help the River Authority repay its debt. The River Authority initially self-funded the venture by issuing more than $530,000,000 in bonds for the construction of the surface water treatment plant and delivery infrastructure. And these fees—the sole source of money for the River Authority under the GRP Contracts—would be used to pay off those bonds. Crucially, under the GRP Contract, the River Authority did not earn a profit.
The GRP Contract also provides that the River Authority ' shall decide when, if ever, Participant must connect to the Project' and take surface water. And once the River Authority mandates the connection, the Participant must take treated surface water in quantity determined by the River Authority based on “the capacity of the Project and the requirements of the GRP.” The River Authority may also increase this mandatory take.
This provision ensures that some participants actually convert to surface water. Without it, the Joint GRP could not compel any member to convert, and it could not achieve the 30% reduction mandated by the Rule.
In 2015, the necessary infrastructure was completed, and the River Authority began delivering surface water from Lake Conroe. Since then to Page 9now, it has continued to do so to 17 LVGUs, but it has never required Quadvest to take surface water.
Even so, Quadvest took aim at the Rule undergirding the GRP Contracts. In August 2015, Quadvest sued the Conservation District in state court, challenging the validity of the Rule. While that suit was pending, Quadvest overhauled the Conservation District’s board of directors through a series of political maneuvers and advocacy. And in 2019, the new board approved a settlement agreement with Quadvest and agreed to a final judgment stating that it was “without legal authority” to limit pumping and that the pumping limits “are, and have been, unlawful, void, and unenforceable.” Now, the Conservation District provides a guideline that no more than 97,000 acre-feet per year of groundwater should be pumped, but no regulations currently enforce a reduction in groundwater pumping.
With the underlying regulation gone, Quadvest brought this challenge to the GRP Contracts.1 Quadvest , 7 F.4th at 339-40. The River Authority moved to dismiss, asserting statute-of-limitations, laches, and state-actionimmunity defenses and arguing that the plaintiffs failed to state a claim. Id. at 342. The district court denied the motion, prompting an interlocutory appeal on the state-action immunity issue. Id. This court affirmed, holding that the River Authority is “not entitled to state-action immunity at this stage.” Id. at 348.
Page 10After remand, the case went to trial in January 2024 on four Sherman Act claims: (1) § 1 price-fixing; (2) § 1 market-allocation; (3) § 1 tying; and (4) § 2 attempted-monopolization claims. Crucially, 'Quadvest d[id] not challenge any other Joint GRP contracts between [the River Authority] and other [LVGUs] even though they have essentially identical terms.” Following a bench trial, the district court issued a 106-page Findings of Fact and Conclusions of Law, concluding that “Quadvest failed to prove any of its claims,” and a final judgment in the River Authority’s favor. Quadvest timely appealed, but only the price-fixing and market-allocation claims.
“The standard of review for a bench trial is well established: findings of fact are reviewed for clear error and legal issues are reviewed de novo .” Guzman v. Hacienda Records & Recording Studio, Inc. , 808 F.3d 1031, 1036 (5th Cir. 2015).
On appeal, Quadvest presents seven points of error. They essentially boil down to three: (1) the district court erred in concluding that Quadvest failed to establish an antitrust injury; (2) the GRP Contract is a per se illegal horizontal price-fixing or market allocation agreement in violation of § 1 of the Sherman Act; and (3) even if the GRP Contract is not per se illegal, it fails the rule of reason.
Even if the district court erred in finding no antitrust injury, however, Quadvest failed to show that the GRP Contract is a violation of the Sherman Act as a per se illegal arrangement or under the rule of reason. That is enough for us to affirm, so we do not reach antitrust injury. See BRFHH Shreveport, LLC v. Willis-Knighton Med. Ctr. , 49 F.4th 520, 525 (5th Cir. 2022); see also McCormack v. NCAA , 845 F.2d 1338, 1343 (5th Cir. 1988) ('Whether a Page 11plaintiff has antitrust standing does not raise a question of jurisdiction on which we are required without exception to satisfy ourselves.”).
“Section 1 of the Sherman Act proscribes ‘[e]very contract, combination . . . or conspiracy [] in restraint of trade or commerce . . . .' Dillard v. Merrill Lynch, Pierce, Fenner & Smith, Inc. , 961 F.2d 1148, 1158 (5th Cir. 1992) (quoting 15 U.S.C. § 1). “In order to state a claim for a violation of Section 1, a plaintiff must allege (1) the existence of a [contract, combination,] or conspiracy (2) affecting interstate commerce (3) that imposes an unreasonable restraint of trade.” Id. (citation omitted).
We first address (1) the interstate-commerce nexus, then (2) whether there is a contract, combination, or conspiracy for § 1 to apply, and, if so, (3) whether it imposes an unreasonable restraint of trade.
“[T]he reach of the [Sherman A]ct is coextensive with the reach of congressional power under the Commerce Clause of the Constitution.” Gulf Coast Hotel-Motel Ass’n v. Miss. Gulf Coast Golf Course Ass’n , 658 F.3d 500, 504 (5th Cir. 2011). That is to say, it is expansive. See id. at 505 (collecting cases). Indeed, “subject matter jurisdiction based on the Commerce Clause has ‘long been interpreted to extend beyond activities actually in interstate commerce to reach other activities that, while wholly local in nature, nevertheless substantially affect interstate commerce.’” Id. (quoting Cowan v. Corley , 814 F.2d 223, 226 (5th Cir. 1987)). At issue here is whether the Page 12relevant activity substantially affects interstate commerce, even if it is not in it.2
The district court found that “Quadvest has not met its burden to establish that the alleged Sherman Act violation-[the River Authority's] GRP Contract with Quadvest-affects interstate commerce.' But that examines the incorrect conduct; it is not the violation that must affect interstate commerce, but the defendant's commercial activity. As the Supreme Court explained, a Sherman Act plaintiff “need not make the more particularized showing of an effect on interstate commerce caused by the alleged conspiracy . . . or by those other aspects of respondents’ activity that are alleged to be unlawful.” McLain v. Real Estate Bd. of New Orleans, Inc. , 444 U.S. 232, 242–43 (1980). Rather, it must do no more than “demonstrate a substantial effect on interstate commerce generated by respondents' [commercial] activity.' Id. at 242. 'If establishing jurisdiction required a showing that the unlawful conduct itself has an effect on interstate commerce, jurisdiction would be defeated by a demonstration that the Page 13alleged restraint failed to have its intended anticompetitive effect. That is not the rule of our cases.” Id.
Because it relied on an erroneous view of the law, the district court made no factual findings related to the interstate-commerce effects of the River Authority’s commercial activities. Ordinarily, that may be grounds for a remand. See Pullman-Standard v. Swint , 456 U.S. 273, 291 (1982). But “when the record permits only one resolution of the factual issue after the correct law is applied, remand is unnecessary.” Aransas Project v. Shaw , 775 F.3d 641, 658 (5th Cir. 2014). And here, the record shows ample interstatecommerce connection, even under the narrow standard the district court applied.
Uncontroverted record evidence shows that vast sums of money crossed state lines in furtherance of the Joint GRP and in the River Authority’s name. Recall that the River Authority issued $530,000,000 in bonds. The depository bank for the bond funds was the Bank of New York Trust Company, which, as its name suggests, is not a Texas entity. The Bank of New York Trust Company, in turn, disbursed the funds to various other entities, including to a Massachusetts entity and a Texas entity. And the funds were eventually used to build infrastructure in Texas. We see no difficulty in concluding that hundreds of millions of dollars crossing state lines have a substantial effect on interstate commerce. See McLain , 444 U.S. at 245.
Having ascertained Sherman Act jurisdiction, we turn to whether § 1 of the Sherman Act is triggered at all. “The Sherman Act contains a ‘basic distinction between concerted and independent action.’” Copperweld Corp. v. Indep. Tube Corp. , 467 U.S. 752, 767 (1984). 'The conduct of a single firm”— i.e., independent action—“is governed by § 2 alone and is unlawful Page 14only when it threatens actual monopolization.” Id. For instance, predatory pricing—below-cost prices that drive rivals out of the market and allow the monopolist to raise its prices later and recoup its losses-is unilateral conduct scrutinized under § 2. Pac. Bell Tel. Co. v. linkLine Commc’ns, Inc. , 555 U.S. 438, 448 (2009). “Section 1 of the Sherman Act, in contrast, reaches unreasonable restraints of trade effected by a ‘contract, combination . . . or conspiracy' between separate entities.” Copperweld , 467 U.S. at 768. “It does not reach conduct that is ‘wholly unilateral.’” Id.
The district court concluded as to the price-fixing claim that the River Authority’s “offered pricing is ‘unilateral conduct’ that section 1 does not reach.” But that conclusion misses the mark because the alleged illegal pricefixing is the fees in the contract , not the offered pricing. A formed contract—a step beyond a mere offer—requires two or more parties, and no one disputes that the GRP Contract at issue is between Quadvest and the River Authority—“between separate entities.” See Copperweld , 467 U.S. at 768.
To be sure, two parties can enter an agreement to form a joint venture that operates as a single entity. See Texaco Inc. v. Dagher , 547 U.S. 1, 4 (2006). But the district court never found that the Joint GRP is a joint venture, and the River Authority conceded below that it 'has never argued that its contract with Quadvest created a single entity.” Moreover, a finding that the Joint GRP was a joint venture would be hard to square with the district court's finding that 'Quadvest is [the River Authority's] customer,' not partner.
The district court thus erred in concluding that the challenged conduct is ''unilateral conduct' that section 1 does not reach.' It is a concerted action between the River Authority and Quadvest, so § 1 applies.
Finally, and perhaps most importantly, Quadvest must show that the GRP Contract was an unreasonable restraint of trade. Antitrust analysis under § 1 of the Sherman Act generally falls into two categories. “In the first category are agreements whose nature and necessary effect are so plainly anticompetitive that no elaborate study of the industry is needed to establish their illegality-they are 'illegal per se .'' Nat’l Soc’y of Prof’l Eng’rs v. United States , 435 U.S. 679, 692 (1978). 'In the second category are agreements whose competitive effect can only be evaluated by analyzing the facts peculiar to the business, the history of the restraint, and the reasons why it was imposed,'-commonly known as the 'rule of reason.' Id. Because the per se label almost always sounds the death knell for the challenged scheme, courts “have expressed reluctance to adopt per se rules.” State Oil Co. v. Khan , 522 U.S. 3, 10 (1997).
Quadvest claims that the GRP Contract is per se illegal and, even if it is not, fails the rule of reason. To the contrary, we conclude that the GRP Contract is not subject to per se condemnation because it is neither a horizontal agreement nor an agreement to fix prices or allocate markets. And we further conclude that the GRP Contract survives the rule of reason.
a
'Typically[,] only 'horizontal' restraints-restraints 'imposed by agreement between competitors’—qualify as unreasonable per se .' Ohio v. Am. Express Co. , 585 U.S. 529, 540 (2018) (quoting Bus. Elecs. Corp. v. Sharp Elecs. Corp. , 485 U.S. 717, 730 (1988)). The Supreme Court has recognized two such horizontal restraints subject to the per se rule: price fixing and market allocation. Texaco , 547 U.S. at 5 (price fixing); Palmer v. BRG of Ga., Inc. , 498 U.S. 46, 49 (1990) (market allocation). Quadvest claims that the GRP Contract is both. It is neither.
Page 16“Horizontal agreements are . . . agreements between competitors.” New Orleans Ass'n of Cemetery Tour Guides & Cos. v. New Orleans Archdiocesan Cemeteries , 56 F.4th 1026, 1035 (5th Cir. 2023). 'Vertical agreements are those between entities at different levels of distribution.' Id. The district court held that “Quadvest is [the River Authority’s] customer,” not competitor, making the agreement vertical. The River Authority similarly argues that the GRP Contract is an “agreement between buyer and seller to achieve regulatory compliance.' Quadvest, on the other hand, argues that the River Authority and Quadvest 'are not part of any distribution chain-they are both water wholesalers. Because [they] both operate on the same level of production, their agreement is not vertical.”
The agreement is vertical for two reasons. First , regardless of whether Quadvest and the River Authority compete in the wholesale market generally, the specific transaction at issue is vertical. “[N]ot every agreement between horizontally related firms is necessarily a horizontal agreement.” Phillip E. Areeda & Herbert Hovenkamp , Antitrust Law: An Analysis of Antitrust Principles and Their Application ¶ 1901b (5th ed. 2026). For example:
Chrysler, currently having excess capacity in windshield wiper production facilities, might sell a wiper blade factory to Ford, who currently does not make such blades. Although Chrysler and Ford are rivals in the sale of automobiles, and presumably in most aspects of their automobile-manufacturing business, this particular transaction is actually vertical, in the sense that Chrysler is acting as a supplier of an input into the production of Ford automobiles.
Id. And here, even if Quadvest and the River Authority are wholesale competitors, the River Authority “is acting as a supplier of an input into the Page 17production of' Quadvest's groundwater-services to obtain regulatory compliance.3
Second , more fundamentally, the River Authority and Quadvest were not competitors at the time the agreement was made. “[I]t is a basic antitrust principle that the impact of an agreement on competition is assessed as of ‘the time it was adopted.’” Impax Lab’ys, Inc. v. Fed. Trade Comm’n , 994 F.3d 484, 496 (5th Cir. 2021) (quoting Polk Bros. v. Forest City Enters. , 776 F.2d 185, 189 (7th Cir. 1985) ( Easterbrook, J. )); see also Areeda & Hovenkamp , supra , ¶ 1901b (explaining that rivalry must exist 'at the time the agreement is made”). Here, the district court specifically found— and Quadvest does not dispute—that “[w]hen it signed the GRP Contract, Quadvest did not have any wholesale water customers.” And it would not have one until 2012.
In fact, even now, Quadvest and the River Authority do not compete. The River Authority provides groundwater and wastewater services only in The Woodlands, based on an arrangement dating back to the 1970s. Quadvest does not point to any part of the record or the district court’s factual findings that show that it too operates in that market or competes for the same customers. Nor has Quadvest shown that the River Authority is encroaching on Quadvest’s market. Indeed, the district court found that “Quadvest did not establish that [the River Authority] tries to sell its wholesale treated surface water on the free market or has any plans to do so.” And while, since 2012, Quadvest’s wholesale customers included home builders, real estate Page 18developments, and new municipal utility districts, the River Authority has not contracted with any such entities in the meantime. In fact, the River Authority has no sales team at all. Moreover, while certain Joint GRP Participants are drawing water from the River Authority, the district court found that “Quadvest has not tried to compete with [the River Authority] to deliver water to any GRP Participants.' Quadvest does not contend otherwise.
In short, Quadvest has not entered and does not intend to enter the River Authority’s part of the market, and the River Authority has not entered and does not intend to enter Quadvest’s. They are not and have never been competitors.
According to Quadvest, that matters little because they “compete by seeking to provide the same services.” To start, that begs the question by asserting that they “compete” in any capacity, contrary to the district court’s unchallenged findings of fact. And, again, at the time of the agreement, Quadvest and the River Authority did not seek to provide the same services because Quadvest had not yet entered the wholesale market. Moreover, mere provision of the same services—without more-is insufficient. The purported competitors must do so in the same market to the same pool of customers. Again, the district court found that Quadvest and the River Authority do not do so.
Tellingly, the three cases Quadvest relies on do not support its proffered proposition. Take Arizona v. Maricopa County Medical Society , 457 U.S. 332 (1982). There, the Supreme Court entertained an antitrust challenge to two nonprofit foundations formed “to provide the community with a competitive alternative to existing health insurance plans.' Id. at 33940. To that end, the foundations established a schedule of maximum fees the participating doctors agreed to accept. Id. at 339. In holding that the fee Page 19schedule was per se illegal price fixing, the Court wrote, '[t]he agreement under attack is an agreement among hundreds of competing doctors concerning the price at which each will offer his own services to a substantial number of consumers.” Id. at 356–57.
The issue for Quadvest is that Maricopa 's holding was built on the factual predicate that the parties to the agreement were 'competing doctors.” Id. at 357. In fact, the foundations themselves conceded that “the agreements at issue are horizontal.” Id. at 342. Here, of course, the River Authority does not concede so, and Quadvest had the burden to establish that factual predicate at trial but failed to carry it.
Similar problem for Quadvest's reliance on North Texas Specialty Physicians . North Texas Specialty Physicians ('NTSP') was an organization of physicians and physician groups that 'negotiate[d] contracts between these groups and ‘payors’ such as insurance companies.” N. Tex. Specialty Physicians v. F.T.C. , 528 F.3d 346, 352 (5th Cir. 2008). The organization’s eight-member elected board of directors would disseminate the payors' offers to physicians, and if a majority of the physicians voted to accept the offer, the organization would proceed to negotiating a contract. Id. at 353.
But there too, the record before this court showed that the physicians were competitors. As we explained, “[t]he ALJ found, and NTSP does not dispute, that in Tarrant County NTSP specialists were a large percentage of the practitioners within a specialty, for example 80 percent in pulmonary disease, 59 percent in cardiovascular disease, and 69 percent in urology. Many NTSP physicians compete with one another.' Id. Indeed, “[t]he FTC found that NTSP is controlled by competing physicians,” a factual finding we deemed conclusive because it was supported by evidence. Id. at 354, 356; see also 15 U.S.C. § 45(c).
Page 20So again, North Texas Specialty Physicians never purported to hold that mere possibility of competition stemming from the provision of the same services suffices for § 1 liability. Instead, like the Court in Maricopa , this court built its holding on the factual predicate before it—that the signatories to an agreement were in fact competitors. We have no such predicate before us.
The third case is even less applicable. Quoting American Needle, Inc. v. National Football League , Quadvest tells us “[a]bsence of actual competition may simply be a manifestation of the anticompetitive agreement itself.” See 560 U.S. 183, 198 (2010). But that plucks the quote out of its context. At issue in American Needle was whether the conduct of a marketing entity formed by professional football teams in the National Football League ('NFL') should be viewed as a concerted action of those teams or a unilateral action of the entity. Id. at 189. In arguing that it was a unilateral action, the NFL pointed out that its teams “have for some time marketed their trademarks jointly.” Id. at 198. The Court rejected that argument, and it was in support of an unremarkable proposition-that a history of anticompetitive behavior does not sanitize it-that the Court said the absence of actual competition may be because of the anticompetitive agreement itself. Id.
Here, however, there is no history of competition before the allegedly anticompetitive agreement. Nor does Quadvest argue that it was bound by another anticompetitive agreement before the GRP Contracts that inhibited its competition with the River Authority. American Needle is thus inapposite.
In a similar vein, Quadvest argues that mere potential competitor status is sufficient to form a horizontal agreement. To be sure, in United States v. Topco Associates, Inc. , the Supreme Court held as a per se illegal horizontal agreement an agreement among small and medium sized regional supermarket chains to divide up markets to sell private-label products. 405Page 21 U.S. 596, 598, 601-02, 608 (1972). And it did so even though '[t]he defendants in Topco had never competed in the same market, but had simply agreed to allocate markets.' Palmer , 498 U.S. at 49.
But two points undermine Quadvest’s reliance on Topco . First, the defendants there never disputed that the market is allocated among wouldbe competitors, “accept[ing] as true most of the Government’s allegations regarding territorial division and restrictions on wholesaling.' Topco , 405 U.S. at 604. The only issue before the Court was whether the defendants’ proffered justification for the horizontal agreement-that the territorial divisions are necessary to compete with larger chains—saves the agreement from the per se rule. Id. at 605.
Second, nothing in the record shows that Quadvest was even a potential competitor to the River Authority. We reiterate that, at the time of the GRP Contract, Quadvest was not a participant in the wholesale market. And the record does not show that it even attempted to enter that market but could not. In fact, it took two years after the GRP Contract to secure its first wholesale contract.
That failure of proof is compounded by the district court’s finding that “[t]ransporting water over large distances is costly, with water transmission lines costing on average about $7.2 million per mile to construct.” “Based on Quadvest’s total net income over the decade between 2011 and 2020, Quadvest could afford to build only about two and a half miles’ worth of instructor in the same decade.” Consequently, “[g]roundwater usage is very localized, with groundwater being able to be transported only a few miles in any direction.” The costly infrastructure therefore casts further doubt on even the potential competitor status of Quadvest to the River Authority.
Lastly, Quadvest urges us to follow a 76-year-old case from the Fourth Circuit. In Pennsylvania Water & Power Co. v. Consolidated Gas, Electric LightPage 22 & Power Co. , two utilities (Penn Water and Consolidated) were held to have entered into a per se unlawful horizontal market-allocation agreement. 184 F.2d 552, 558 (4th Cir. 1950). Per Quadvest, the Fourth Circuit held so even though the parties were mere potential competitors. But that potentialcompetitor status was by virtue of their existing infrastructure, which Quadvest lacks. As the Fourth Circuit explained, “[i]f it were not for the agreement between the parties which [was] the subject of th[at] suit, the parties would be potential competitors in the generation and sale of electric energy through their present facilities or other facilities that might be constructed .” Id. at 556 (emphasis added).
Indeed, the court took pains to explain the extent of the existing infrastructure that would have enabled the two utilities to compete. Penn Water had lines that “connect[ed] directly” to others, allowing it to draw from them “if it were not for its contract with Consolidated”; Consolidated had “four large steam generating plants” and “distribution facilities in and around” the geographic market, with “extensive transmission lines”; and both had charter rights for the purchase and sale of electricity in the geographic market. Id. at 555-56. It was by virtue of such existing infrastructure and modest extensions thereof that these utilities were “potential competitors.” Id. at 556. The same cannot be said with respect to Quadvest, whose infrastructure was limited and capital insufficient to greatly expand it.
Therefore, the GRP Contract is a vertical agreement.
Even if the GRP Contract is a horizontal agreement, it is not a pricefixing agreement.
“The price-fixing within the scope of the per se prohibition of § 1 . . . is an agreement to fix the price to be charged in transactions with third Page 23parties, not between the contracting parties themselves.” United States v. All Star Indus. , 962 F.2d 465, 470 n.10 (5th Cir. 1992). But nothing in the GRP Contract sets wholesale water prices or affects Quadvest’s ability to negotiate them with potential customers. In fact, in 2021, Quadvest entered a wholesale water contract in which it expressly agreed not to pass along the pumpage fees it was charged under the GRP Contract.
Quadvest counters that a “combination formed for the purpose and with the effect of raising, depressing, fixing, pegging, or stabilizing the price of a commodity in interstate or foreign commerce is illegal per se .' See United States v. Socony-Vacuum Oil Co. , 310 U.S. 150, 223 (1940) (emphasis added). The purpose and effect of the cost equalization provision, says Quadvest, were to raise the cost of groundwater and lower the cost of surface water to create “price neutrality” in the market.
While Quadvest is correct that the GRP Contract's demonstrable purpose was to raise and lower the respective production costs of the different waters, it is an unsupported logical leap to conclude that the purpose—not merely the effect—was to manipulate the market price . Start with two givens. First, the GRP Contract, by design, affects the production cost. The River Authority concedes as much, explaining that the cost equalization is necessary to ensure that “participants who received surface water were not penalized relative to those who continued pumping groundwater” “[b]ecause surface water is more expensive to produce than groundwater.” Second, the GRP Contract had an effect on the market price of water.
Quadvest merges the two givens. It claims that because there was a purpose of affecting the cost, and the effect of affecting the market price , it means there was the purpose of affecting the market price . It makes that logical leap by characterizing the pumpage fee as “an agreed cost floor set by the Page 24participants, not the free market,” because the GRP Participants “have to pass on” those fees.
But the record contradicts that characterization. No provision of the GRP Contract sets a “cost floor” for the water. Again, nothing in the GRP Contract sets or affects Quadvest’s ability to negotiate wholesale water prices with potential customers. Nor does Quadvest “ have to pass on those fees”; it in fact entered into a wholesale agreement specifically agreeing not to pass on those pumpage fees. Quadvest’s contention may make sense if the alleged market and the relevant commodity was the act of production itself, not water; the GRP Contract indisputably sets an agreed-upon cost floor on production. But it does not set the floor on the downstream wholesale or retail market.
Instead, the Joint GRP was agnostic as to the market price. As the district court found, the River Authority 'does not know what prices Quadvest charges its customers, does not monitor those prices, has never sought to determine what those prices are, and Quadvest has never approached [the River Authority] to approve the prices that Quadvest charges its customers.” Thus, the GRP Contract does not evince a purpose of manipulating the market price and is not a price-fixing agreement.
Next, the market-allocation claim. Before wading into the merits, we first explain our disagreement with the district court’s conclusion that the market-allocation claim is moot. Turning to the merits, we hold that the GRP Contract is not a market-allocation agreement.
“The Constitution’s case-or-controversy limitation on federal judicial authority, Art. III, § 2, underpins . . . [the] mootness jurisprudence.” Friends of the Earth, Inc. v. Laidlaw Envt’l Servs. (TOC), Inc. , 528 U.S. 167, 180 (2000). If there is no more controversy between the litigants, “they no Page 25longer qualify as adverse parties with sufficient legal interests to maintain the litigation,' depriving this court of the 'power to entertain the case.' Sossamon v. Lone Star State of Tex. , 560 F.3d 316, 324 (5th Cir. 2009).
But an exception exists: “[T]he voluntary cessation of a complainedof activity by a defendant ordinarily does not moot a case.' Id. The standard for a party to sidestep this so-called 'voluntary cessation doctrine' is 'stringent: A case might become moot if subsequent events make it absolutely clear that the allegedly wrongful behavior could not reasonably be expected to recur.' Laidlaw , 528 U.S. at 189. This court, however, has effectively carved out an exception to the exception. 'Although Laidlaw establishes . . . a heavy burden . . . government actors in their sovereign capacity and in the exercise of their official duties are accorded a presumption of good faith because they are public servants, not self-interested private parties.” Sossamon , 560 F.3d at 324.
Relying on that “lighter burden” enjoyed by government actors, id. , the district court found mootness because the River Authority 'has represented to the [c]ourt multiple times that it will not (and cannot) force Quadvest to connect to the surface water treatment system now that the Conservation District has rescinded its groundwater pumping limits.”
That was in error. Quadvest’s claim is not that the River Authority may force Quadvest to connect to the surface water treatment system. It is that the River Authority has forced other GRP Participants to do so, allocating to itself a part of the wholesale market and depriving Quadvest of potential customers. And 'in evaluating mootness, we must assume that Plaintiff[] will prevail on the merits of [its] claims.' Sandpiper Residents Ass’n v. U.S. Dep’t of Hous. & Urb. Dev. , 106 F.4th 1134, 1141 (D.C. Cir. 2024). So whether the River Authority will not or cannot force Quadvest to connect has no bearing on the existence of the claim that the River Authority has forced Page 26others to connect. This is especially true when the district court found that the River Authority’s delivery of treated surface water has not changed since 2016,' suggesting that some entities remained locked into the River Authority’s supply and out of Quadvest’s reach. The case is therefore not moot.
On the merits, however, Quadvest fails. “One of the classic examples of a per se violation of § 1 is an agreement between competitors at the same level of the market structure to allocate territories in order to minimize competition.” Topco , 405 U.S. at 608. It includes agreements “either not to compete with one another in the market, or to divide customers or potential customers between them.” Optronic Techs., Inc. v. Ningbo Sunny Elec. Co., Ltd. , 20 F.4th 466, 481 (9th Cir. 2021).
Quadvest claims the mandatory connection provision does “exactly that'-'The provisions give [the River Authority] the power to decide which participants are required to purchase its surface water. [The River Authority] exercised that right, making the largest groundwater users its customers.” The River Authority takes a narrower view of the contract. By its reading of the GRP Contract, Quadvest agreed to “allocate” itself to the River Authority as a possible customer of surface water, and that is not improper market allocation.
The River Authority has the better view. Quadvest challenges its own GRP Contract and no one else's. And the subject of the Quadvest-River Authority GRP Contract's mandatory connection provision is not the entirety of the Joint GRP or even some of its members—it is only Quadvest. The GRP Contract says the River Authority 'shall decide when, if ever, Participant must connect to the Project.' And '[i]f Participant is required to connect to the Project . . . then Participant shall take Water . . . in accordance with the requirements of this Contract and when directed by written notice Page 27from” the River Authority. “Participant,” in turn, is defined as none other than “QUADVEST, L.P., a Texas limited partnership.” So contrary to the Quadvest’s position, the plain language of the GRP Contract allows the River Authority to require Quadvest—and only Quadvest—to take water. It does not hold captive Quadvest’s potential customers.
To be sure, the GRP Contract also sometimes refers to “Participants” in the plural. 'Participants,' though, is a separately defined term that “includes [the River Authority], Participant, and any other Regulated User that enters into and remains subject to a written agreement with [the River Authority] in a form substantially similar to this Contract.” The fact that the challenged GRP Contract specifically sets out a separate definition of the plural “Participants” to include other members of the Joint GRP, yet the mandatory connection provision only refers to the singular “Participant,” cuts against Quadvest. And of course, however often the contract refers to third parties, the River Authority cannot require non-parties to a contract to undertake an obligation under that contract.
In sum, far from allocating the market, the sole contract at issue merely allows the River Authority to sell Quadvest surface water. That does not amount to a market-allocation agreement. The GRP Contract is therefore not subject to the per se rule. 4
Without the per se rule, Quadvest is left with the rule of reason. Under the rule of reason, an agreement’s “competitive effect can only be evaluated by analyzing the facts peculiar to the business, the history of the restraint, and the reasons why it was imposed.” Nat’l Soc’y of Prof’l Eng’rs , 435 U.S. at 692.
“[T]o determine whether a restraint violates the rule of reason, . . . a three-step, burden-shifting framework applies.' Am. Express , 585 U.S. at 541. “Under this framework, the plaintiff has the initial burden to prove that the challenged restraint has a substantial anticompetitive effect that harms consumers in the relevant market.” Id. 'If the plaintiff carries its burden, then the burden shifts to the defendant to show a procompetitive rationale for the restraint.” Id. “If the defendant makes this showing, then the burden shifts back to the plaintiff to demonstrate that the procompetitive efficiencies could be reasonably achieved through less anticompetitive means.' Id. at 542.
Quadvest falters out of the gate. It cannot carry its initial burden because it failed to define the relevant market at trial and forfeited any arguments to the contrary on appeal. We do not reach the other prongs.
Quadvest can carry its initial burden directly or indirectly. Id. 'Direct evidence of anticompetitive effects would be proof of actual detrimental effects on competition, such as reduced output, increased prices, or decreased quality in the relevant market.” Id. 'Indirect evidence would be proof of market power plus some evidence that the challenged restraint harms competition.' Id. Both paths require Quadvest to “define the relevant market” because “courts usually cannot properly apply the rule of reason without an accurate definition of the relevant market.' Id. at 542-43. 'Without a definition of [the] market there is no way to measure [the Page 29defendant’s] ability to lessen or destroy competition.” Id. at 543 (quoting Walker Process Equip., Inc. v. Food Mach. & Chem. Corp. , 382 U.S. 172, 177 (1965)).
The relevant market has two components: a product market and a geographic market. Hornsby Oil Co., Inc. v. Champion Spark Plug Co., Inc. , 714 F.2d 1384, 1393 (5th Cir. 1983). “[T]he relevant product market is composed of products that have reasonable interchangeability,” id. , that is, '[g]oods that consumers view as substitutes for other goods,' R.D. Imports Ryno Indus. Inc. v. Mazda Distrib. (Gulf), Inc. , 807 F.2d 1222, 1225 (5th Cir. 1987). The relevant geographic market is “the area of effective competition,” considering “economic and physical barriers to expansion as transportation costs, delivery limitations and customer convenience and preference.' Hornsby Oil , 714 F.2d at 1394-95.
The district court found that Quadvest did not prove either the relevant product market or the relevant geographic market. As to the product market, the court found, Quadvest excluded from the proposed product market several commodities—such as water from the Catahoula aquifer and reused/reclaimed wastewater effluent—even though Quadvest’s own expert acknowledged that Joint GRP Participants use those commodities. And “a proposed relevant market that clearly does not encompass all interchangeable substitute products” is “legally insufficient.” Apani Sw., Inc. v. Coca-Cola Enter., Inc. , 300 F.3d 620, 628 (5th Cir. 2002). Moreover, the district court found the product market evidence “not clear, credible or reliable” because Quadvest pointed only to Joint GRP Participants, whose substitution choices were constrained not by market forces but by their contractual obligations under their GRP Contracts.
As to the geographic market, the district court rejected Quadvest’s proffered countywide market. It found that Quadvest’s own evidence and Page 30expert demonstrated that high transportation costs make it economically infeasible to transport water more than a few miles, making the proposed geographic market of Montgomery County “far too large.” And Quadvest did not prove that every wholesale water customer in every part of Montgomery County can turn to the River Authority for water since the River Authority does not operate in most of the county. The court concluded, the River Authority “has no realistic possibility of operating in most, let alone all, of Montgomery County.'
To all this, Quadvest says nothing. Its opening brief focuses only on the next question—whether there are substantial anticompetitive effects of the GRP Contract. And its reply brief attempts to shed the burden altogether by arguing that no proof of market power is required where “the very purpose and effect of a horizontal agreement” is to make the prices unresponsive to a competitive market. Even at oral argument, Quadvest’s counsel reprised the same burden-shedding argument, contending that Quadvest need not define the market because the restraint is per se illegal.
As explained above, however, the GRP Contract is neither horizontal nor price fixing, and thus, is not per se unlawful. That means Quadvest faces the burden of proving the relevant market. And by not challenge the district court's adverse fact-finding on that point-reviewed for clear errorQuadvest has forfeited any arguments to the contrary and failed to carry that burden. See Vernon Smith v. Sch. Bd. of Concordia Par. , 88 F.4th 588, 594 (5th Cir. 2023). Consequently, it has not proven that the GRP Contract is an unlawful restraint of trade under the rule of reason.
For the foregoing reasons, we AFFIRM the judgment of the district court.
1 Initially, Quadvest was joined by another investor-owned water utility, Woodland Oaks Utility. See Quadvest , 7 F.4th at 339. But before trial, Woodland Oaks Utility voluntarily dismissed its claims. From trial through appeal, Quadvest was and remains the sole plaintiff.
2 The River Authority suggests that the interstate-commerce requirement is a merits issue, not a jurisdictional one. As its argument goes, the Supreme Court in Arbaugh v. Y&H Corp. announced a 'bright line' rule that 'when Congress does not rank a statutory limitation on coverage as jurisdictional, courts should treat the restriction as nonjurisdictional in character.” 546 U.S. 500, 516 (2006). And the Sherman Act does not expressly say the interstate-commerce requirement is jurisdictional. But see Gulf Coast , 658 F.3d at 503–08 (postArbaugh , characterizing the interstate-commerce requirement as one of “subject matter jurisdiction”).
That distinction may matter if there were no such interstate-commerce nexus because then the district court’s final judgment on the merits would be invalid. See In re Majestic Energy Corp. , 835 F.2d 87, 89 (5th Cir. 1998) (“Where a federal court rules in a matter over which it does not have jurisdiction, its decisions, opinions and orders are without effect.”). But because we hold that there is sufficient interstate-commerce nexus, we need not conclusively resolve whether this requirement is jurisdictional.
3 At oral argument, Quadvest's counsel argued that the River Authority cannot act as a supplier of an input by selling compliance because it “do[es]n’t have compliance to sell.” We agree, but that misses the point. The River Authority is not selling compliance per se , but services to obtain it. In that way, it is akin to an accountant who, while not having tax-code compliance to sell, can offer services to achieve that compliance.
4 Quadvest also challenges the district court's alternative conclusion that, even if the GRP Contract is a horizontal restraint on competition, it benefits from the “ancillary restraints” doctrine. Because the ancillary restraints doctrine is an exception to the per se rule, see Aya Healthcare Servs., Inc. v. AMN Healthcare, Inc. , 9 F.4th 1102, 1109 (9th Cir. 2021), and because we hold that the GRP Contract is not subject to the per se rule, we do not reach whether the ancillary restraints doctrine applies here.