Reset A A Font size: Print

United States Court of Appeals, Fourth Circuit.

David HOLBROOK, Plaintiff-Appellant, v. TENNESSEE VALLEY AUTHORITY; BVU Authority, Defendants-Appellees.

No. 21-1415

Decided: September 07, 2022

Before RICHARDSON, RUSHING, and HEYTENS, Circuit Judges. ARGUED: Martin Bienstock, BIENSTOCK PLLC, Washington, D.C., for Appellant. Maria Victoria Gillen, TENNESSEE VALLEY AUTHORITY, Knoxville, Tennessee; Cameron Scott Bell, PENN, STUART & ESKRIDGE, Abingdon, Virginia, for Appellees. ON BRIEF: Mark T. Hurt, Abingdon, Virginia, for Appellant. Karissa H. Range, PENN, STUART & ESKRIDGE, Abingdon, Virginia, for Appellee BVU Authority. David D. Ayliffe, Director of Litigation, TVA GENERAL COUNSEL'S OFFICE, Knoxville, Tennessee, for Appellee Tennessee Valley Authority.

The Tennessee Valley Authority sells its power to the BVU Authority in Virginia, one of its many customers. The BVU Authority in turn sells its power to local consumers who need electricity. Among those local consumers is David Holbrook, and Holbrook thinks he has been paying too much for power. He believes that the TVA has a statutory duty to use the fruits of its sales to large industrial buyers to subsidize consumers' electricity consumption. He bases this view largely on § 11 of the Tennessee Valley Authority Act of 1933, Pub. L. No. 73-17, § 11, 48 Stat. 58, 64–65 (codified at 16 U.S.C. § 831j). That provision gives the TVA two goals: The first goal is to operate “primarily ․ for the benefit of the people ․ particularly the domestic and rural consumers to whom the power can economically be made available,” and a “secondary” goal is to use sales to industry “principally to secure a sufficiently high load factor and revenue returns which will permit domestic and rural use at the lowest possible rates.” Holbrook reads that language as a command to the TVA to subsidize consumers from the pockets of industry. Under that reading, Holbrook believes that a string of TVA rate changes, shifting costs from industry to consumers, were illegal. So he sued BVU Authority and TVA under three theories, which all more or less amount to claims that the TVA failed to live up to its statutory duties under § 11. The district court dismissed all three claims because TVA's ratemaking authority is committed to agency discretion and thus unreviewable. We affirm.

I. Background

A. The Tennessee Valley Authority

“Congress created the TVA—a ‘wholly owned public corporation of the United States’—in the throes of the Great Depression to promote the Tennessee Valley's economic development.” Thacker v. TVA, ––– U.S. ––––, 139 S. Ct. 1435, 1439, 203 L.Ed.2d 668 (2019) (quoting TVA v. Hill, 437 U.S. 153, 157, 98 S.Ct. 2279, 57 L.Ed.2d 117 (1978)).1 Congress created the TVA through the Tennessee Valley Authority Act of 1933.2 And Congress gave the TVA a lot to do. It was tasked with improving the navigability of the Tennessee River, preparing for flood control along the river, reforesting parts of the Valley, and putting other areas to more productive use. TVA Act, §§ 1, 22–23. It was even told to act “in the interest of the national defense” by operating a fertilizer plant in Muscle Shoals, Alabama. §§ 1, 5, 11.3

What matters for our purposes is TVA's role in producing electricity. To this day, the TVA is a major purveyor of electricity to the Tennessee Valley.4 It sells electricity wholesale to federal agencies, industrial users, localities, local governments, and co-ops who do the retail selling to consumers (or use the power themselves).

B. Holbrook's Claims and TVA Rate-Setting

David Holbrook is one of those consumers. He lives in Bristol, Virginia, and buys his power from the BVU Authority, a power company created by the State of Virginia. The TVA sells power to the BVU Authority who serves local customers. Holbrook's suit here focuses on the rates that TVA requires BVU Authority, as a wholesale buyer, to charge Holbrook, as a retail power consumer.

When the TVA sets rates, it aims to meet a slew of policy objectives that Congress laid out in the Act. To start, § 10 of the Act instructs the TVA to give preferences to “States, counties, municipalities, and cooperative organizations of citizens or farmers, not organized or doing business for profit, but primarily for the purpose of supplying electricity to its own citizens or members.”

Section 11 of the Act is of particular importance to this suit:

It is hereby declared to be the policy of the Government so far as practical to distribute and sell the surplus power generated at Muscle Shoals equitably among the States, counties, and municipalities within transmission distance. This policy is further declared to be that the projects herein provided for shall be considered primarily as for the benefit of the people of the section as a whole and particularly the domestic and rural consumers to whom the power can economically be made available, and accordingly that sale to and use by industry shall be a secondary purpose, to be utilized principally to secure a sufficiently high load factor and revenue returns which will permit domestic and rural use at the lowest possible rates and in such manner as to encourage increased domestic and rural use of electricity.

Holbrook reads this piece of the statute as a directive to use industry sales to subsidize “domestic and rural consumers.” See § 11.

Holbrook's claims concern several revenue-neutral rate changes implemented through BVU Authority contracts under a new TVA policy called the “Strategic Pricing Plan.” J.A. 107. To understand that claim, we need to understand how the TVA sets its rates. To grossly simplify, when the TVA sets rates, it takes two distinct steps. First, it determines the amount of revenue it needs to collect from all its sales. Then, it determines how to assign prices to various customers to meet that overall goal. When the TVA changes that total amount it needs, that's called a “rate adjustment.” In contrast, a “rate change” occurs when the TVA rearranges or otherwise changes the various rates among customers while keeping its overall revenue goal the same. So rate changes must be revenue neutral.

The TVA sets its rates largely through entering power contracts with local power companies, who then resell the power to different classes of customers at different rates as required by the power contracts with TVA.

Back to the Strategic Pricing Plan. In 2010, the TVA began putting the Plan, which focused on “TVA's long-term pricing,” into motion in power contracts with local power companies. J.A. 249. The Plan aimed to achieve fairness in pricing and increase competitiveness by charging customers based on their proportion of total cost of service. The TVA aimed to change rates so “that revenue [would] be recovered in proportion to costs by customer class.” J.A. 187. Because supplying power to industry is cheaper, TVA sought to create new benefits and discounts for industrial consumers, things like manufacturing credits and high-volume discounts. TVA hoped the Plan would better transform the TVA's savings from efficient, industrial-energy use into savings for the industrial consumers who made those efficiencies possible. But Holbrook alleges that all those changes to benefit industrial customers unjustifiably shifted costs onto consumers.

To illustrate the effect of these changes, Holbrook says that in 2015, compared to its rivals, TVA had industrial rates in the cheapest quartile and consumer rates around the median, which already suggests that consumers were not top priority. But then the TVA added a “General Manufacturing Credit,” which reduced industry rates by an average of 6.4% and increased consumer rates by 1.1%. By 2016, Holbrook says this shifted nearly half a billion dollars in costs from industry to consumers, and that has only increased since then.

Holbrook's power supplier BVU entered contracts with TVA over the past decade, carrying out the rate structure envisioned by the Strategic Pricing Plan. Based on his theory that TVA was illegally requiring his electric company to charge him too much for power, Holbrook brought this Complaint on behalf of “[a]ll persons who within the statute of limitations period were domestic customers of TVA” and “[a]ll persons who within the statute of limitations period were domestic customers of BVU and who received energy pursuant to a contract between the TVA and the BVU.” J.A. 119. The Amended Complaint focuses on three claims: a third-party breach of contract claim against both TVA and BVU, an Administrative Procedure Act claim under 5 U.S.C. § 702 against TVA alone, and an “Unlawful Exaction” claim against both TVA and BVU, which alleges that TVA caused Plaintiffs to pay more for electricity than “could lawfully be charged under the TVA Act.” J.A. 121–22.

While we do not have the exact contracts used during the relevant period since the district court dismissed the case by granting a motion to dismiss, we do have an example of a similar contract from 2006. Holbrook alleges that this contract is representative of the contracts they have used since. The 2006 contract reiterates many of the TVA's goals and purposes as laid out in the TVA Act. It may be more accurate to say that the statutory language is copied into the contract. And importantly, the contract includes the § 11 language that “power shall be primarily for the benefit of the people of the section as a whole and particularly the domestic and rural consumers, to whom it is desired to make power available at the lowest possible rates.” J.A. 10.

The district court relied on precedent from courts in the Sixth Circuit to hold that TVA's ratemaking authority is “committed to agency discretion by law” under 5 U.S.C. § 701(a)(2). And because § 701(a)(2) precludes judicial review over issues committed to agency discretion, the district court dismissed Holbrook's APA claims. From there, the court dismissed the contract and exactions claims for the same reason, finding that the contract and exaction claims were essentially the same claim as the APA challenge and were therefore barred by § 701(a)(2) just the same. See Holbrook v. TVA, 527 F. Supp. 3d 853, 858 (W.D. Va. 2021).

II. Discussion

We review the dismissal of these claims de novo. Demetres v. E.W. Constr., Inc., 776 F.3d 271, 272 (4th Cir. 2015). We take each claim—the APA claim, the breach-of-contract claim, and the unlawful-exaction claim—in turn.

A. APA Claim

Under the APA, “[a] person suffering legal wrong because of agency action ․ is entitled to judicial review thereof.” 5 U.S.C. § 702. That language sets up a “basic presumption of judicial review” of agency action. See Abbott Lab'ys v. Gardner, 387 U.S. 136, 140, 87 S.Ct. 1507, 18 L.Ed.2d 681 (1967). But the APA's text lays out two exceptions to that basic principle: first, where “statutes preclude judicial review,” § 701(a)(1), and second, where “agency action is committed to agency discretion by law,” § 701(a)(2). Only the second exception might apply here, so we must figure out whether TVA ratemaking is “committed to agency discretion by law.”5

Courts have dealt with two initial puzzles about what it means under the APA for something to be “committed to agency discretion by law.” The first puzzle is how to differentiate the two exceptions to judicial review. At a glance, it's hard to see the difference between a statute that precludes judicial review and law that commits decisions to agency discretion (thereby precluding judicial review). Yet the Supreme Court has given us some guidance. The Court tells us the § 701(a)(1) exception for statutes precluding judicial review “applies when Congress has expressed [its] intent” and the § 701(a)(2) standard for agency discretion applies when there is “no meaningful standard against which to judge the agency's exercise of discretion.” Heckler v. Chaney, 470 U.S. 821, 830, 105 S.Ct. 1649, 84 L.Ed.2d 714 (1985). So the first exception is for explicit statutory limitations on review, and the second exception—the one at issue—is for implicit limitations on review. See Dep't of Just., Attorney General's Manual on the Administrative Procedure Act 94 (1947) (“A statute may in terms preclude, or be interpreted as intended to preclude, judicial review altogether.”).6

The second puzzle arises from the seeming tension between the second exception, § 701(a)(2), and § 706, the APA's provision defining the scope of agency review. Under § 706(2)(A), courts are instructed to “hold unlawful and set aside agency action, findings, and conclusions found to be ․ an abuse of discretion.” One might wonder how courts can set aside something as abuse of discretion when discretionary questions committed to the agency by law are insulated from judicial review in the first place. The Supreme Court's solution to this puzzle has been to focus on the suitability of the agency action for judicial review—“if no judicially manageable standards are available for judging how and when an agency should exercise its discretion, then it is impossible to evaluate agency action for ‘abuse of discretion.’ ” Chaney, 470 U.S. at 830, 105 S.Ct. 1649. If courts can naturally review for an abuse of discretion, they should; if they can't, § 701(a)(2) tells them to steer clear. See Lincoln v. Vigil, 508 U.S. 182, 190–91, 113 S.Ct. 2024, 124 L.Ed.2d 101 (1993). So the main task under § 701(a)(2) is to determine when there are or are not “judicially manageable standards” for judging an agency's exercise of discretion.

Early cases applying this subsection used a “no law to apply” test drawn from the legislative history of the APA. See Chaney, 470 U.S. at 830, 105 S.Ct. 1649 (quoting Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 410, 91 S.Ct. 814, 28 L.Ed.2d 136 (1971)). The test asks whether this is one of “those rare instances where ‘statutes are drawn in such broad terms that in a given case there is no law to apply.’ ” Overton Park, 401 U.S. at 410, 91 S.Ct. 814 (quoting S. Rep. No. 79-752, at 26 (1945)). The problem with that test is that there is nearly always some law to apply—“beginning with the fundamental constraint that the decision must be taken in order to further a public purpose rather than a purely private interest.” Webster v. Doe, 486 U.S. 592, 608, 108 S.Ct. 2047, 100 L.Ed.2d 632 (1988) (Scalia, J., dissenting). Remember § 706 and abuse-of-discretion review as well: Arbitrary-and-capricious review only involves “articulat[ing] a satisfactory explanation for [agency] action including a rational connection between the facts founds and the choice made.” Motor Vehicle Mfrs. Ass'n of U.S., Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43, 103 S.Ct. 2856, 77 L.Ed.2d 443 (1983) (quoting Burlington Truck Lines v. United States, 371 U.S. 156, 168, 83 S.Ct. 239, 9 L.Ed.2d 207 (1962)); see Ronald M. Levin, Understanding Unreviewability in Administrative Law, 74 Minn. L. Rev. 689, 708 (1990) (“Pure abuse of discretion inquiries do not depend on the contents of the statute under which an agency acts.”). We could always apply that legal test by making sure the agency had offered reasoned explanation of its actions.

Because the “no law to apply” test is so difficult to meet, the Supreme Court has often taken a different approach to § 701(a)(2), one that operates more like a common-law analysis than a task of statutory interpretation. The aim of this common-law approach has been to determine categories of administrative action that “courts traditionally have regarded as ‘committed to agency discretion.’ ” Dep't of Com. v. New York, ––– U.S. ––––, 139 S. Ct. 2551, 2568, 204 L.Ed.2d 978 (2019) (quoting Lincoln, 508 U.S. at 192, 113 S.Ct. 2024).7 Once we are in a traditional category, the “presumption of reviewability” under the APA flips, and the agency action becomes “presumptively unreviewable.” Chaney, 470 U.S. at 831–32, 105 S.Ct. 1649. But only presumptively. Even in an area that has been traditionally insulated from review, “Congress may limit an agency's exercise of enforcement power if it wishes, either by setting substantive priorities, or by otherwise circumscribing an agency's power to discriminate among issues or cases it will pursue.” Id. at 833, 105 S.Ct. 1649.

We take all this to create a two-part inquiry. We begin by considering whether TVA ratemaking is the kind of agency action that “has traditionally been ‘committed to agency discretion.’ ” Id. at 832, 105 S.Ct. 1649. We hold that it is. From there, we determine whether the TVA Act intentionally limits agency discretion by setting guidelines or otherwise providing a limit. We hold that it does not. So we affirm the district court's decision that TVA ratemaking is “committed to agency discretion by law.”

1. Traditional Categories Committed to Agency Discretion

No clean rule materializes for determining whether an agency action is the kind of action that has traditionally been committed to agency discretion. But the Supreme Court has looked to a few factors that characterize such action. First, these actions involve “complicated balancing of a number of factors which are peculiarly within [the agency's] expertise,” Lincoln, 508 U.S. at 193, 113 S.Ct. 2024 (quoting Heckler, 470 U.S. at 831, 105 S.Ct. 1649), especially decisions that involve resource allocation and the need for flexibility to “adapt to changing circumstances,” Lincoln, 508 U.S. at 192, 113 S.Ct. 2024. Next, these are areas that often do not involve the use of coercive power, which means they will not trigger the traditional rights-protecting duties of the federal courts. Chaney, 470 U.S. at 832, 105 S.Ct. 1649. And perhaps most importantly, these areas enjoy a tradition of nonreviewability. Id. at 832, 105 S.Ct. 1649. Past practice should guide us. And an unbroken practice of judicial deference that predates the APA is strong evidence of an area where judicial review is inappropriate. See ICC v. Brotherhood of Locomotive Eng'rs, 482 U.S. 270, 282, 107 S.Ct. 2360, 96 L.Ed.2d 222 (1987).8

TVA ratemaking has each of these characteristics. To start, TVA price setting is a balancing act that demands significant expertise and involves complicated, counterfactual questions of resource allocation. As we explain below, the TVA Act tasks the TVA with several goals that necessarily require trade-offs, including a focus on self-sufficiency, equitable service across States, building up capacity, repaying the Treasury, supporting consumers, and more. And as a look through the TVA's 2018 Wholesale Rate Change shows, the practical difficulties of electricity pricing are even more complicated, including additional hurdles like “distributed generation, energy efficiency, technological advances, shifts in customer behavior, and regulatory requirements,” not to mention the interplay between price (which is calculated to the quarter cent) and demand. J.A. 170–71. All that suggests a “complicated balancing of a number of factors which are peculiarly within [the agency's] expertise.” Lincoln, 508 U.S. at 193, 113 S.Ct. 2024 (quoting Chaney, 470 U.S. at 831, 105 S.Ct. 1649).9 Setting a price is complicated, and it is not a task on which judges are traditionally expected to be experts. Indeed, the opposite may be closer to the truth.

And price setting isn't coercive either. It does not exercise power over “an individual's liberty or property rights.” Chaney, 470 U.S. at 832, 105 S.Ct. 1649. Prices are agreed-to, not enforced. Therefore, these issues will rarely implicate the traditional rights-protecting duties of the federal courts. See id. Any argument that Holbrook is somehow forced to buy from BVU because of limited options would stretch the idea of coercive power beyond recognition. When the government disposes of its own property, it will seldom be acting coercively.10

Finally, federal courts in the Tennessee Valley region have a long history of declining to review TVA ratemaking. See, e.g., McCarthy v. Middle Tenn. Elec. Membership Corp., 466 F.3d 399, 407 (6th Cir. 2006); 4-County Elec. Power Ass'n v. TVA, 930 F. Supp. 1132, 1138 (S.D. Miss. 1996); Carborundum Co. v. TVA, 521 F. Supp. 590, 593 (E.D. Tenn. 1981); Mobil Oil v. TVA, 387 F. Supp. 498, 506–07 (N.D. Ala. 1974); see also Matthews v. Town of Greeneville, 932 F.2d 968, 1991 WL 71414, *4 (6th Cir. 1991) (unpublished). This trend reaches back at least 84 years to a case decided just a few years after the TVA Act was passed. Tenn. Elec. Power Co. v. TVA, 21 F. Supp. 947 (E.D. Tenn. 1938), aff'd, 306 U.S. 118, 59 S.Ct. 366, 83 L.Ed. 543 (1939). “The Tennessee Valley Authority Act authorizes the Board of Directors of the Authority to fix the rates at which the electric energy generated at the dams authorized by the Tennessee Valley Authority Act may be sold. The statute vests discretion in the board in fixing such rates, and the exercise of this discretion is not subject to judicial review.” Mobil Oil, 387 F. Supp. at 508 (quoting Tenn. Elec. Power Co., 21 F. Supp. 947 (Conclusion of Law 33) (unpublished)). And Tennessee Electric Power Company was decided eight years before the APA was passed, which makes this tradition a part of the “existing law” that the APA was understood to embrace and “preserve.” Brotherhood of Locomotive Eng'rs, 482 U.S. at 282, 107 S.Ct. 2360; see Chaney, 470 U.S. at 832, 105 S.Ct. 1649 (quoting 5 Davis, Administrative Law § 28:6 (1984), for the proposition that the “APA did not significantly alter the ‘common law’ of judicial review of agency action”). In fact, Holbrook has not provided—nor have we found—any case in which a federal court has subjected the TVA's ratemaking to judicial scrutiny in the way that is requested here.11 Given the TVA is going on a century old, this long tradition counsel against reviewing TVA's rates here.

We also draw more support for this principle by zooming out from TVA ratemaking to electric ratemaking by government-owned utilities generally, see, e.g., V.I. Hotel Ass'n v. V.I. Water & Power Auth., 465 F.2d 1272, 1274–75 (3d Cir. 1972); Elecs. of N.C., Inc. v. Se. Power Admin., 774 F.2d 1262, 1265–67 (4th Cir. 1985); and then zooming out further to government-utility ratemaking generally, see Pan. Canal v. Grace Line, 356 U.S. 309, 318–19, 78 S.Ct. 752, 2 L.Ed.2d 788 (1958); Rural Electrification Admin. v. N. States Power Co., 373 F.2d 686, 700–01 (8th Cir. 1967); Hahn v. Gottlieb, 430 F.2d 1243, 1250–51 (1st Cir. 1970); Langevin v. Chenango Court, Inc., 447 F.2d 296, 304 (2d Cir. 1971); cf. United States v. George S. Bush & Co., 310 U.S. 371, 379, 60 S.Ct. 944, 84 L.Ed. 1259 (1940) (“No one has a legal right to the maintenance of an existing rate or duty.”) (quoting Norwegian Prods. Co. v. United States, 288 U.S. 294, 318, 53 S.Ct. 350, 77 L.Ed. 796 (1933)).12

Holbrook responds with an argument based on the language of the APA. Holbrook argues that (1) under the APA, “the approval or prescription for the future of rates [or] prices [or] costs” is defined as a kind of “rule,” 5 U.S.C. § 551(4); (2) all “rules” are “agency action,” § 551(13); and (3) that all “agency action” is subject to judicial review, § 704. By adding these premises up, Holbrook argues that TVA ratemaking must not be one of the traditional categories we are talking about.

But that argument misses the point. No one has questioned, and we do not deny, that TVA ratemaking is agency action or that the general rule is that agency action is presumptively reviewable. See Abbott Lab'ys, 387 U.S. at 140, 87 S.Ct. 1507. The question here is whether this is the kind of agency action where that presumption is flipped because of § 701(a)(2), which is separate from the analytically antecedent answer that this was an “agency action.” After all, Chaney dealt with “agency decisions to refuse enforcement,” 470 U.S. at 831, 105 S.Ct. 1649, and we know that “failure to act” is defined as “agency action” under § 551(13). But the Court there found refusal to enforce to be “committed to agency discretion” under § 701(a)(2) anyway. To say the TVA's actions were agency action is to make an obvious point about why the APA applies at all; we instead focus on § 701(a)(2)'s specific exception, which immunizes certain “agency action” from judicial review.

In sum, given the complexity of the task, the lack of coercive power, and the long, unrebutted history of courts refusing to review these prices, we hold that TVA ratemaking is a decision committed by tradition to agency discretion. That tradition brings with it a presumption against judicial review, which can still be overcome by showing that Congress intended to cabin the exercise of that traditional discretion. We now turn to that question.

2. Congressional Guidelines or Limits on Traditional Discretion

Congress may overcome the presumption against review by providing “guidelines for the agency to follow in exercising its enforcement powers,” by “setting substantive priorities, or by otherwise circumscribing an agency's power.” Chaney, 470 U.S. at 833, 105 S.Ct. 1649. Because the question is about what Congress did, it amounts to a question of statutory interpretation. See Webster, 486 U.S. at 600, 108 S.Ct. 2047 (“[Section] 701(a)(2) requires careful examination of the statute on which the claim of agency illegality is based”). The only argument that Holbrook makes here is based on the twin goals of TVA Act § 11, but we do not read that provision to provide the kind of clear guidance or instruction that would overcome the presumption against judicial review.

An example will help illustrate the kind of language we are looking for. The Chaney Court uses Dunlop v. Bachowski, 421 U.S. 560, 95 S.Ct. 1851, 44 L.Ed.2d 377 (1975), as a prototype. At issue there was the Secretary of Labor's decision to decline to prosecute a violation of the Labor-Management Reporting and Disclosure Act. As Chaney makes clear, decisions about whether to prosecute are traditionally insulated from judicial review, but in Dunlop, the statute provided the kind of command that overcomes that presumption. The statute said that after a union member files suit in the agency, “[t]he Secretary shall investigate such complaint and, if he finds probable cause to believe that a violation ․ has occurred ․ he shall ․ bring a civil action.” Id. at 563 n.2, 95 S.Ct. 1851 (quoting 29 U.S.C. § 482(b)). Because the statute “quite clearly withdrew discretion from the agency,” judicial review was appropriate. Chaney, 470 U.S. at 834, 105 S.Ct. 1649. So even in the quintessentially insulated prosecutorial-discretion arena, plain statutory commands will provide meaningful standards for judicial review.

Holbrook focuses on § 11 of the TVA Act, arguing that the provision “includes both a statement of purpose and a directive.” Appellant's Op. Br. 5. But § 11 cannot be read as a clear directive and therefore will not overcome the presumption against a review.

Section 11 has two relevant sentences.13 The first sentence reads: “It is declared to be the policy of the Government so far as practical to distribute and sell the surplus power ․ equitably among the States, counties, and municipalities within transmission distance.” The next sentence elaborates on that policy by laying out a primary and a secondary purpose:

This policy is further declared to be that the projects herein provided for shall be considered primarily as for the benefit of the people of the section as a whole and particularly the domestic and rural consumers to whom the power can economically be made available, and accordingly that sale to and use by industry shall be a secondary purpose, to be utilized principally to secure a sufficiently high load factor and revenue returns which will permit domestic and rural use at the lowest possible rates ․

Holbrook argues that this is a command that the TVA use industry sales to subsidize consumer sales. He argues that this provision provides a “directive” to the TVA by giving both a goal—“lowest possible rates” for consumers—and a methodology for achieving that goal—using industry for its load factor and revenue returns—and that taken together this provides enough “law to apply” to invite judicial review. We disagree. Instead, we read this provision as a general policy statement and, in places, as a kind of aspiration about what Congress hopes will be accomplished.

Start with the fuzzy language in the provision: “so far as practical,” “primarily,” “economically,” “sufficiently.” Each of those words suggests room for discretion. And all that discretion adds up. Taken together, the mass of discretionary lingo suggests that, far from being a provision that withdraws discretion, this provision acknowledges and accentuates that discretion. Webster, 486 U.S. at 600, 108 S.Ct. 2047 (“This standard fairly exudes deference”).14

Next, consider a phrase we've already mentioned: “so far as practical.” Notice that the first sentence of the provision suggests that the policy should be carried out only “so far as practical,” and notice further that the second sentence of § 11 begins “This policy is further declared to be ․,” before then discussing consumer and industry sales. Read together, this suggests that both sentences are referencing the same policy, and that the policy should only be pursued “so far as practical.” That is not a directive. Since the Act does not define “practical,” determining the limits of what is and what is not practical must be a matter of discretion.

Finally, turn to the discussion of sales to industry: “[S]ale to and use by industry shall be a secondary purpose, to be utilized principally to secure a sufficiently high load factor and revenue returns which will permit domestic and rural use at the lowest possible rates.” § 11 (emphasis added). Holbrook wants this to mean that sales to industry must be used to subsidize consumers, but we do not agree. The text says that sales to industry are to be used to secure high load factors and strong revenues. And then it says those things “will permit” better treatment for consumers, in the form of “the lowest possible rates.” The “will permit” suggests that this isn't really a command or a “methodology” for achieving a specified “goal” as Holbrook argues. Rather, the text suggests that Congress had an expectation, that by selling to industry, the TVA would get higher load factors, allowing more consistent energy usage, which in turn would bring in revenues to the company, which would help to increase returns to scale, and all of that “will” naturally make sales to consumers easier and cheaper. “Will permit” highlights how this policy is an aspiration not a command. And even if we read that as something more than an aspiration, we would be confronted again by the discretionary phrases “sufficiently high” and “lowest possible” which do little to cabin the agency's actions.

We cannot read § 11 as the kind of guideline or command that would overcome the presumption against judicial review here. Because TVA ratemaking is a category that has traditionally been insulated from judicial review and because Congress has not provided clear limits on the exercise of that discretion, we hold that TVA ratemaking is “committed to agency discretion by law.” So the district court was correct to dismiss the APA claim under § 701(a)(2).

B. The Other Claims

We also affirm the district court's dismissal of Holbrook's two other causes of action, but we must take a moment to explain our different reasoning. The district court held that it lacked jurisdiction over all three legal theories because they were all various flavors of the same claim—challenges to the discretionary ratemaking decisions of the TVA. Holbrook, 527 F. Supp. 3d at 858 (citing McCarthy, 466 F.3d at 407 (“If we were to review the Cooperatives' actions in enforcing the contract, we would still be reviewing the TVA's actions and thus ignoring the APA's prohibition on judicial review.”)). But § 701(a)(2) of the APA says only that “This chapter applies ․ except to the extent that agency action is committed to agency discretion by law.” (Emphasis added.) By focusing on “this chapter” only, Congress did not go so far as to preempt any judicial review anywhere in the vicinity of a discretionary agency judgment. The § 701(a)(2) bar only prevents us from hearing claims under the APA itself.

It's not clear why this should prevent courts from reviewing other claims—even other claims that look a lot like the APA claims they tag along with—so long as those claims themselves are otherwise cognizable.15 So even though we will not generally review TVA rates, once the TVA signs a contract, it's possible we may have—as Holbrook claims here—contract “law to apply.” See Portland Gen. Elec. Co. v. Bonneville Power Admin., 501 F.3d 1009, 1032 (9th Cir. 2007). Or if there is a claim that the TVA's discretionary ratemaking was racially discriminatory, we might have federal civil-rights “law to apply.” Cf. Angelex Ltd. v. United States, 723 F.3d 500, 508 (4th Cir. 2013).

So while we agree with the district court that the APA claim was properly dismissed under § 701(a)(2), we do not agree that strips us of jurisdiction to consider any other claims that happen to look like the APA challenge to TVA ratemaking. We must consider the other claims separately. In the end, however, this quibble won't help Holbrook; we will affirm those dismissals individually instead of en mass. See Willner v. Dimon, 849 F.3d 93, 103 (4th Cir. 2017) (“We may affirm on any grounds supported by the record, notwithstanding the reasoning of the district court.” (cleaned up)).

First to the contract claim. As a member of a class of intended beneficiaries of the contract between TVA and BVU Authority, Holbrook argues that TVA has violated the contract provision which includes the same statutory duties to provide the lowest feasible rates.16 While not barred by § 701(a)(2), this attempt fares no better than the APA claim. The Supreme Court has held that, where a statute provides no cause of action and where an agency signs a contract including those statutory obligations as boilerplate, a breach of contract suit alleging violation of those statutory-but-also-contractual provisions “is in essence a suit to enforce the statute itself.” Astra USA, Inc. v. Santa Clara Cnty., 563 U.S. 110, 118, 131 S.Ct. 1342, 179 L.Ed.2d 457 (2011). A nonparty to the contract cannot evade Congress's decision to not include a statutory cause of action by casting themselves as a third-party beneficiary of the identically worded contract. Id. In this case, the self-styled contract action is, instead, an attempted cause of action under § 11 of the TVA Act—which has no private cause of action. “The statutory and contractual obligations, in short, are one and the same.” Id. We therefore affirm the dismissal of the contract claim.

Finally, to the exactions claim.17 An illegal exaction claim is ultimately a claim under the Due Process Clause of the Fifth Amendment that there was a deprivation of property without due process of law. See Norman v. United States, 429 F.3d 1081, 1095 (Fed. Cir. 2005). In layman's terms, an exaction claim is when the government takes your things without justification. So an exactions claim asks “for recovery of monies that the government has required to be paid contrary to law.” Elec. Welfare Tr. Fund v. United States, 907 F.3d 165, 170 (4th Cir. 2018) (quoting Aerolineas Argentinas v. United States, 77 F.3d 1564, 1572 (Fed. Cir. 1996)). Holbrook cannot make out an exaction here because “a reasonable user fee is not a taking if it is imposed for the reimbursement of the cost of government services.” United States v. Sperry Corp., 493 U.S. 52, 63, 110 S.Ct. 387, 107 L.Ed.2d 290 (1989); see Koontz v. St. Johns River Water Mgmt. Dist., 570 U.S. 595, 615, 133 S.Ct. 2586, 186 L.Ed.2d 697 (2013). In short, a voluntary payment for services rendered is not an exaction, illegal or otherwise. And even if it were an exaction, Holbrook's payments were not illegal because the TVA Act cannot be read to require subsidies to consumers as we have already said. So again, we affirm the dismissal of Holbrook's exactions claim.

* * *

Section 11 of the TVA Act lays out broad policies and goals that operate more like aspirations than commands. It does not support any of the claims that Holbrook offers against TVA or BVU Authority. TVA ratemaking is a presumptively unreviewable category of agency action under 701(a)(2), and the policy-laden language of § 11 does not provide any guidelines or limits to overcome that presumption. Because the TVA-BVU contract simply repeats the vague statutory language, Holbrook's contract claim is really a statutory claim in disguise, and § 11 of the TVA Act does not provide a private cause of action. And finally, because Holbrook voluntarily bought power from BVU Authority, nothing was exacted or taken from him at all. So the district court is



1.   The TVA is a unique federal agency. See North Carolina ex rel. Cooper v. TVA, 515 F.3d 344, 349 (4th Cir. 2008). For it is also a corporation and can sue and be sued in its corporate name. TVA Act, § 4(b). Consequently, it has been exempted from many provisions that govern other agencies. The TVA is explicitly exempt from suit in the Court of Federal Claims. 28 U.S.C. § 1491(c). That's why we, and not the Federal Circuit, are hearing this case. The TVA is also exempt from many rules that govern other federal agencies, including civil service laws, TVA Act, § 3, purchasing requirements, § 9, and systematic review under 5 U.S.C. § 305(a)(2), (b) (“[E]ach agency shall review systematically the operations of each of its activities, functions, or organization units, on a continuing basis.”). But the provision on judicial review of agency action in the Administrative Procedure Act does not exempt the TVA. 5 U.S.C. § 704. And § 701 of the APA lays out the meaning of “agency” for the purposes of agency review, and the TVA is not listed among the types of entities exempt from review. § 701(b).

2.   We adopt the practice of the parties who refer to the TVA's provisions by their original section titles in the Public Law. The most important provision for our purposes is § 11, which is codified at 16 U.S.C. § 831j.

3.   The Muscle Shoals plant produced nitrates — a primary ingredient in both fertilizer and explosives. § 5(g). So in peacetime, those nitrates were used to make fertilizer, but they could be directed to explosives manufacturing during wartime.

4.   A note on why the Fourth Circuit is hearing this case. The Tennessee Valley is the drainage basin of the Tennessee River, which is—you guessed it—mostly in the state of Tennessee. Tennessee, of course, sits in our sister circuit, the Sixth. But the River and the Valley and therefore the TVA dip into other states in the Sixth Circuit (Kentucky), the Fifth Circuit (Mississippi), the Eleventh Circuit (Alabama and Georgia), and parts of the Fourth Circuit (North Carolina and Virginia). This case comes to us from the Western District of Virginia.

5.   One influential administrative law scholar had this to say about the task of figuring out what § 701(a)(2) means: “I don't see how anybody can find the meaning of those words. The words seem to contradict themselves; they don't make any sense; if they do, what might the sense be? Nobody can extract from the words an answer to this simple question: When discretionary power is conferred by statute on an agency, when, if ever, may a court review for abuse of discretion?” Present at the Creation: Regulatory Reform Before 1946, 38 Admin. L. Rev. 507, 519 (1986) (remarks of Kenneth Culp Davis). Well, that's our task.

6.   The Attorney General's Manual on the APA has often been relied on by the Supreme Court in interpreting the Act. See Vt. Yankee Nuclear Power Corp. v. Nat. Res. Def. Council, Inc., 435 U.S. 519, 546, 98 S.Ct. 1197, 55 L.Ed.2d 460 (1978). Because the Department of Justice played an unusually important role in the drafting of the APA, id., the Court has been willing to cite this contemporaneous guide. See Kisor v. Wilkie, ––– U.S. ––––, 139 S. Ct. 2400, 2419, 204 L.Ed.2d 841 (2019). Even Justice Scalia, ever the opponent of legislative history, found that the Attorney General's Manual is often a “persuasive” source in expounding the meaning of the APA given its unique status. See Norton v. S. Utah Wilderness All., 542 U.S. 55, 63–64, 124 S.Ct. 2373, 159 L.Ed.2d 137 (2004). While of course it cannot overcome the plain text of the Act, we consider it.

7.   This switch from the “no law to apply” test to a more common-law approach was implied in Chaney but got a more extensive treatment in a dissent by Justice Scalia in Webster, 486 U.S. at 606–21, 108 S.Ct. 2047. There, Justice Scalia argued that § 701(a)(2) cannot be limited to situations when there is no law to apply for the reasons we've just pointed out. He argued instead, as the Court had hinted at in Chaney, that § 701(a)(2) was meant to refer to a preexisting body of common law of judicial review of agency action that the APA was meant to embrace and carry on. Id. at 609, 108 S.Ct. 2047 (“The intended result of § 701(a) is to restate the existing law as to the area of reviewable agency action.” (quoting AG's Manual, supra, at 94)). In essence, “committed to agency discretion by law” means action that is “of the sort that is traditionally unreviewable,” which includes looking at past practice, practical consequences, and whether the decision involves inherently discretionary judgment calls. Id. at 609, 108 S.Ct. 2047.This understanding of § 701(a)(2) was adopted by a majority of the Court in Lincoln, 508 U.S. at 191, 113 S.Ct. 2024 (citing Webster, 486 U.S. at 609, 108 S.Ct. 2047 (Scalia, J., dissenting)), where the Court found that “the allocation of funds from a lump-sum appropriation is another administrative decision traditionally regarded as committed to agency discretion.” Id. at 192, 113 S.Ct. 2024. And this approach to § 701(a)(2) has often been reiterated in the years since—though not often used to find new categories. See, e.g., Dep't of Com., 139 S. Ct. at 2568 (“We have generally limited the exception to certain categories of administrative decisions that courts traditionally have regarded as committed to agency discretion.” (cleaned up)); Weyerhaeuser Co. v. U.S. Fish & Wildlife Serv., ––– U.S. ––––, 139 S. Ct. 361, 370, 202 L.Ed.2d 269 (2018) (“The few cases in which we have applied the § 701(a)(2) exception involved agency decisions that courts have traditionally regarded as unreviewable.” (citing Lincoln, 508 U.S. at 191, 113 S.Ct. 2024)).

8.   The Attorney General's Manual on the APA offers a few examples of unreviewable action under the “committed to agency discretion” provision. AG's Manual, supra, at 94. One such example is United States v. George S. Bush & Co., 310 U.S. 371, 60 S.Ct. 944, 84 L.Ed. 1259 (1940), which dealt with the President's powers to adjust tariffs under the Tariff Act of 1930. Id. at 375–76, 60 S.Ct. 944. In finding that “[n]o question of law is raised when the exercise of [the President's] discretion is challenged,” the Court appealed in part to a long history of discretionary tariff adjustments. Id. at 379–80, 60 S.Ct. 944 (citing Norwegian Nitrogen Prods. v. United States, 288 U.S. 294, 308, 53 S.Ct. 350, 77 L.Ed. 796 (1933)). That history was central to the holding that tariff adjustments are not subject to review.

9.   Holbrook suggests that TVA's competing goals can be harmonized. Remember that rate setting involves rate adjustments (the grand total) and rate changes (shifting around who owes what). So at step one, TVA might say, “We need $X billion to meet all costs.” Then it would look to the likely sales and set prices for each kind of sale to try to add up to that $X billion figure. Holbrook argues that at that second step you could tend toward higher prices for industry.But his argument ignores the feedback loop between those two steps. Supply and demand are dynamic processes, and the cost of producing power for different customers can vary. So making power more expensive for some group means they will buy less of it, which changes TVA's costs and requires a different revenue goal, meaning you have to readjust step two, and so on and so on. Finding the right equilibrium can be complicated.

10.   In its discussion of coercion, Chaney focuses in part on the agency's refusal to act, as distinct from affirmative agency action. In those circumstances, the agency “generally does not exercise its coercive power over an individual's liberty or property rights” because the agency has done nothing. 470 U.S. at 832, 105 S.Ct. 1649. And on the flip side, where there is action, there is usually a “focus for judicial review.” Id. But Chaney's focus was on coercion, not inaction—inaction was useful to prove coercion, not the other way around. While rate-setting is action of a kind—and while we concede that action is more likely to be reviewable than inaction and more likely to be coercive—rate-setting still does not influence any coercive power over liberty or property rights, which we take to be the more important takeaway from Chaney's coercion discussion.

11.   Holbrook mentions that some TVA action has been found reviewable at the Supreme Court, citing Hardin v. Ky. Utils. Co., 390 U.S. 1, 88 S.Ct. 651, 19 L.Ed.2d 787 (1968) (holding that a TVA decision to provide power to a certain area was subject to review under § 15(d) of the TVA Act); cf. id. at 13, 88 S.Ct. 651 (Harlan, J., dissenting). But that case was about where the TVA could operate—i.e, whether it could expand into two towns near the Tennessee-Kentucky border—a meaningfully different question from how they set their prices. And the TVA Act contains a clear rule about the corporation's geographic scope, the sort which permits judicial review. Section 15d(a) of the TVA Act—added in twenty-five years after the TVA Act's initial passage—says, “Unless otherwise specifically authorized by Act of Congress the [TVA] shall make no contracts for the sale or delivery of power which would have the effect of making the Corporation or its distributors, directly or indirectly, a source of power supply outside the area for which the [TVA] or its distributors were the primary source of power supply on July 1, 1957.” The case dealt with determining the boundaries of that forbidden “area.” Hardin, 390 U.S. at 8, 88 S.Ct. 651. The Court did not defer, but determined for itself whether the towns were in that area. But this says nothing about a tradition of reviewing TVA rates.

12.   This is a different context than government regulation of private utility companies' ratemaking, where courts often intervened. See, e.g., Missouri ex rel. S.W. Bell Tel. Co. v. Pub. Serv. Comm'n, 262 U.S. 276, 287, 43 S.Ct. 544, 67 L.Ed. 981 (1923) (collecting cases). Here the government is setting the price for something it is itself selling.

13.   The third sentence of § 11 is important but irrelevant here. It deals with fertilizer.

14.   Holbrook points out that this passage includes “shall” twice, and that this word choice suggests a directive. “Shall” does often mandate behavior. But here the shalls are attached to broad policy goals. “The agency shall have this goal” is not a command in the same way as “the agency shall prosecute if they have probable cause.”

15.   The Attorney General's Manual on the APA provides more support. See AG's Manual, supra, at 95. It imagines a time when a civil servant challenges their discretionary firing. In that case, the employee could not demand arbitrary-and-capricious or substantial-evidence review under the APA, but they could still challenge the firing to ensure the Civil Service Act procedures were followed. Id. (citing Levine v. Farley, 107 F.2d 186 (D.C. Cir. 1939))

16.   Even though Holbrook was not a party to the TVA contracts, it is possible he had standing to bring a contract action as an intended beneficiary—assuming that the relevant contract here, which is not in the record, includes the same language as the 2006 contract in the record. TVA contracts are governed by federal law. See Salary Pol'y Empl. Panel v. TVA, 731 F.2d 325, 330 (6th Cir. 1984); Stock Equip. Co. v. TVA, 906 F.2d 583, 585 n.1 (11th Cir. 1990); see also Priebe & Sons v. United States, 332 U.S. 407, 411, 68 S.Ct. 123, 92 L.Ed. 32 (1947). And under federal law, a nonparty to a contract has standing to bring a contract action when the contract “reflects the express or implied intention of the parties to benefit the third party.” Montana v. United States, 124 F.3d 1269, 1273 (Fed. Cir. 1997) (cleaned up). “The intended beneficiary need not be specifically or individually identified in the contract, but must fall within a class clearly intended to be benefited thereby.” Id. Here, the 2006 contract between TVA and BVU Authority expressly states that the contract was “primarily for the benefit of the consumers of electricity.” J.A. 10. So we proceed on the assumption that Holbrook is an intended beneficiary of the contract with BVU.

17.   Illegal exactions claims are often brought in the Court of Federal Claims but claims against the TVA are specifically exempted from that court's jurisdiction. 28 U.S.C. § 1491(c) (“Nothing herein shall be construed to give the United States Court of Federal Claims jurisdiction․ of any action against, or founded on conduct of, the Tennessee Valley Authority, or to amend or modify the provisions of the Tennessee Valley Authority Act of 1933 with respect to actions by or against the Authority.”).

RICHARDSON, Circuit Judge:

Affirmed by published opinion. Judge Richardson wrote the opinion, in which Judge Rushing and Judge Heytens joined.

Copied to clipboard