Energy Mich., Inc. v. Mich. Pub. Serv. Comm'n

126 F.4th 476

• Court: Court of Appeals for the Sixth Circuit
• Decided: January 16, 2025
• Docket: 23-1324
• Status: Published

Opinion

RECOMMENDED FOR PUBLICATION Pursuant to Sixth Circuit I.O.P. 32.1(b) File Name: 25a0011p.06

UNITED STATES COURT OF APPEALS FOR THE SIXTH CIRCUIT

┐ ENERGY MICHIGAN, INC.; ASSOCIATION OF BUSINESSES │ ADVOCATING TARIFF EQUITY (ABATE), │ Plaintiffs-Appellants/Cross-Appellees (23-1280/1323/1324), │ │ > Nos. 23-1280/1323/1324 v. │ │ MICHIGAN PUBLIC SERVICE COMMISSION, │ Defendant, │ │ │ DANIEL C. SCRIPPS; ALESSANDRA CARREON; KATHERINE L. │ PERETICK, │ Defendants-Appellees/Cross-Appellants (23-1280/1324), │ │ CONSUMERS ENERGY COMPANY, │ │ Intervenor-Appellee/Cross-Appellant (23-1280/1323). ┘

Appeal from the United States District Court for the Eastern District of Michigan at Detroit. No. 2:20-cv-12521—David M. Lawson, District Judge.

Argued: December 7, 2023

Decided and Filed: January 16, 2025

Before: BOGGS, SUHRHEINRICH, and READLER, Circuit Judges. _________________

COUNSEL

ARGUED: Brion B. Doyle, VARNUM LLP, Grand Rapids, Michigan, for Energy Michigan, Inc. Zachary C. Larsen, CLARK HILL PLC, Lansing, Michigan, for ABATE. Nicholas Q. Taylor, OFFICE OF THE MICHIGAN ATTORNEY GENERAL, Lansing, Michigan, for Michigan Public Service Commissioners. Spencer A. Sattler, CONSUMERS ENERGY COMPANY, Jackson, Michigan, for Consumers Energy Company. ON BRIEF: Brion B. Doyle, VARNUM LLP, Grand Rapids, Michigan, Zachary C. Larsen, CLARK HILL PLC, Lansing, Michigan, for Energy Michigan, Inc. and ABATE. Nicholas Q. Taylor, Steven D. Hughey, OFFICE OF THE MICHIGAN ATTORNEY GENERAL, Lansing, Michigan, for Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 2 Mich. Public Serv. Comm’n, et al.

Michigan Public Service Commissioners. Spencer A. Sattler, Kelly M. Hall, CONSUMERS ENERGY COMPANY, Jackson, Michigan, for Consumers Energy Company.

READLER, J., delivered the opinion of the court in which SUHRHEINRICH, J., concurred. BOGGS, J. (pp. 34–43), delivered a separate dissenting opinion. _________________

OPINION _________________

CHAD A. READLER, Circuit Judge. This appeal does not lack for challenging features. The factual backdrop is the complex market for electricity generation, transmission, and distribution in the United States. And the chief legal doctrine at play, the so-called dormant or negative Commerce Clause, has been unflatteringly described as a “quagmire,” Nw. States Portland Cement Co. v. Minnesota, 358 U.S. 450, 458 (1959), “hopelessly confused,” Kassel v. Consol. Freightways Corp. of Del., 450 U.S. 662, 706 (1981) (Rehnquist, J., dissenting), and “inherently unpredictable,” Am. Trucking Ass’ns, Inc. v. Smith, 496 U.S. 167, 203 (1990) (Scalia, J., concurring in the judgment).

But in practice, today’s case turns on some relatively basic questions. Can the State of Michigan require someone selling a product in Michigan to procure that product from the state? Or, phrased in the language of the coin’s other side, can Michigan bar in-state retailers from obtaining their merchandise from outside the state? On these issues, negative Commerce Clause jurisprudence is straightforward. Whether the product at issue is milk, see Dean Milk Co. v. City of Madison, 340 U.S. 349, 352 (1951), or coal-based electricity, see Wyoming v. Oklahoma, 502 U.S. 437, 440 (1992), the Commerce Clause prohibits such state restrictions unless they clear strict scrutiny’s high bar, see Maine v. Taylor, 477 U.S. 131, 138 (1986).

At issue here are Michigan electricity market regulations that expressly restrict where Michigan’s electricity retailers may procure their capacity. Accordingly, that regulatory regime must be evaluated through the lens of strict scrutiny. To allow the district court to engage in that analysis with the benefit of our views here, we reverse and remand. Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 3 Mich. Public Serv. Comm’n, et al.

I.

A. The Market. At the heart of this appeal is a basic but ubiquitous product, electricity. Described in its simplest terms, electricity is the flow of electrons resulting from the conversion of other forms of energy, including coal, natural gas, the sun, uranium, water, and wind. See FERC, Staff Report, Energy Primer: A Handbook for Energy Market Basics 33 (Jan. 2024) (hereinafter “FERC Primer”). Ubiquity often signals demand. That is the case for the electricity market. The United States spends roughly $419 billion annually on electricity, accounting for more than one percent of America’s gross domestic product. See U.S. Energy Info. Admin., Inflation-Adjusted U.S. Energy Spending Increased by 25% in 2021 (Aug. 3, 2023), https://perma.cc/F2ZG-JWXL. But electricity is unlike many goods bought and sold in our economy. Demand is neither constant nor especially predictable. At any given moment, in fact, it can be seemingly unquenchable. See Elec. Power Supply Ass’n v. FERC, 89 F.4th 546, 550 (6th Cir. 2023).

Meeting market demand thus requires more than a mere “flick of a switch.” Id. At least three events play a role in assuring necessary supply: “electricity generation; high voltage, long- distance power transmission . . . ; and . . . lower voltage, local distribution of electricity from the transmission facilities to end users.” Niagara Mohawk Power Corp. v. FERC, 452 F.3d 822, 824 (D.C. Cir. 2006); FERC Primer 44. Each function requires considerable upfront costs. See FERC Primer at 34. Historically, a self-contained utility provided all electricity generation, transmission, and distribution services. Id. Traditionally, utility suppliers would, at great expense, retain excess capacity in reserve to ensure reliable electric service. Elec. Power Supply Ass’n, 89 F.4th at 550 (noting risks in investing in electric capital infrastructure that could “sit unused” when demand dips). Over time, however, the market has adapted to meet demand in a more economically efficient manner. See FERC Primer 34.

For one, electricity markets today no longer are dominated by vertically integrated monopolies, but instead consist of many players involved in generation, transmission, or distribution services. Id. at 35–37. These entities have coordinated their efforts so that additional capacity can be more easily procured from a neighbor. Elec. Power Supply Ass’n, 89 Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 4 Mich. Public Serv. Comm’n, et al.

F.4th at 550. Through interconnected transmission lines, FERC Primer 34, a generator takes the electricity it has created and then “mix[es] [it] with power from other plants on its way” to the end user to meet immediate demand in a cost-effective way. Elec. Power Supply Ass’n, 89 F.4th at 550. The modern electrical market thus involves “electricity flow[ing] . . . through an interconnected ‘grid’ of near-nationwide scope.” FERC v. Elec. Power Supply Ass’n, 577 U.S. 260, 267 (2016).

In this modern distribution scheme, various players buy and sell electricity from each other in a wholesale market, while suppliers then deliver electricity purchased at wholesale to retail consumers. See generally Elec. Power Supply Ass’n, 89 F.4th at 550; Hughes v. Talen Energy Mktg., LLC, 578 U.S. 150, 154–55 (2016). For suppliers in particular, the market is dynamic. To meet constantly changing consumer need, suppliers forecast demand and regularly purchase in advance—sometimes years in advance—a commitment from a generator to provide electricity at a future point. See FERC Primer 40–41; Hughes, 578 U.S. at 155.

B. The Regulatory Scheme. Both federal and state regulators oversee this complex market. The Federal Power Act (FPA) empowers the Federal Energy Regulatory Commission (FERC) with exclusive authority over the sale of electric energy at wholesale—that is, the purchase of electric energy for resale. See 16 U.S.C. § 824(b)(1), (d); Hughes, 578 U.S. at 154. Exercising that authority, FERC issued orders encouraging nonprofit entities to “manage wholesale markets on a regional basis.” Elec. Power Supply Ass’n, 577 U.S. at 267. Two entities heeded that call: regional transmission organizations (RTOs), and independent system operators (ISOs). RTOs and ISOs, while differing in their governance structure and management protocols, function in largely indistinguishable ways. See Pub. Util. Dist. No. 1, v. FERC, 272 F.3d 607, 611–12 & n.3 (D.C. Cir. 2001); FERC Primer 37. As reflected by the map below, RTOs and ISOs serve roughly two-thirds of the country’s electric load, with the Southeast, Southwest and Northwest regions still generally operated by vertically integrated utilities. FERC Primer 60–65; see also infra Figure 1. Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 5 Mich. Public Serv. Comm’n, et al.

Figure 1. North American Regional Transmission Organization and Independent System Operator Regions. See FERC, RTOs and ISOs, (Jan. 17, 2024) https://perma.cc/97S8-ZLUZ.

In addition to administering a part of the grid and affording access to transmission lines, RTOs and ISOs conduct competitive auctions to set wholesale prices for electricity. Elec. Power Supply Ass’n, 577 U.S. at 267–68. FERC “extensively regulates” RTOs and ISOs to ensure their auctions “efficiently balance[] supply and demand” to produce a “just and reasonable clearing price.” Hughes, 578 U.S. at 157.

At the same time, the FPA leaves it to the states to regulate “any other sale” of electricity. See 16 U.S.C. § 824(b). That means states have authority over generation, intrastate transmission, local distribution, and wholly intrastate sales of electricity, including retail sales (i.e., sales directly to users). See § 824(b)(1); Hughes, 578 U.S. at 154; Elec. Power Supply Ass’n, 577 U.S. at 267. State utility commissions tend to oversee those activities. Elec. Power Supply Ass’n, 577 U.S. at 267.

While the federal and state roles in this sphere are distinct, their concerns are “inextricably linked.” Id. at 265. That leads to areas of mutual interest. One is ensuring that suppliers of electricity have sufficient capacity, leaving them the ability to provide electricity on demand. In re Reliability Plans of Elec. Utils. for 2017–2021, 949 N.W.2d 73, 79 (Mich. 2020); Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 6 Mich. Public Serv. Comm’n, et al.

Conn. Dep’t of Pub. Util. Control v. FERC, 569 F.3d 477, 479 (D.C. Cir. 2009). How the respective federal and state electricity regulations play out in Michigan’s market is at the heart of today’s case.

1. Start with the federal role. The RTO Midcontinent Independent System Operator (MISO) covers parts of 15 states, including almost all of the Michigan wholesale electricity market. See supra Figure 1. (The southwest corner of Michigan is under the RTO PJM Interconnection. Id.) For resource capacity purposes, MISO has divided itself into ten local resource zones. See infra Figure 2; Midcontinent Indep. Sys. Operator, Inc., 165 FERC ¶ 61,067, para. 2 (Oct. 31, 2018) (hereinafter “2018 MISO Order”). Zone 7 is located within the lower peninsula; Zone 2 covers the overwhelming bulk of the upper peninsula.

Figure 2. MISO, Planning Year 2023–2024 Loss of Load Expectation Study Report 7, (May 1, 2023), https://perma.cc/XAD3-6B9Y.

Note, that for certain zones, not at issue in this appeal (e.g., Zone 1), the Resource Zone extends beyond the MISO’s boundaries.

Within this framework, MISO imposes resource capacity requirements on any entity that provides electricity to an end user in a MISO zone—in FPA parlance, a “Load Serving Entity” or “LSE.” 2018 MISO Order, 165 FERC para. 2; Hughes, 578 U.S. at 155. This regulatory regime is complex, but its details are largely unimportant here, save for MISO’s Local Clearing Requirement (LCR). The LCR requires that, annually, a certain amount of capacity be Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 7 Mich. Public Serv. Comm’n, et al.

physically located in a zone. 2018 MISO Order, 165 FERC para. 4. Because electricity dissipates over long distances, the LCR, by ensuring access to local energy resources, promotes grid reliability. See, e.g., Midwest Indep. Transmission Sys. Operator, Inc., 126 FERC ¶ 61,144, paras. 43–47 (Feb. 19, 2009). The LCR, we note, focuses on total in-zone capacity. So under MISO’s rules, an LSE can acquire all of its electricity outside of the local resource zone without penalty so long as the total in-zone capacity from all LSEs is sufficient—a market condition that has been satisfied to date.

2. Now turn to Michigan. Before discussing the Wolverine State’s rules on resource adequacy, it is worth stepping back to consider broadly how the state regulates the non- wholesale electricity market. The Michigan Public Service Commission (MPSC), an agency created just before World War II, see Mich. Comp. Laws Ch. 460 (Act 3 of 1939), oversees Michigan’s retail electricity market. Reliability Plans, 949 N.W.2d at 90. Echoing national trends, Michigan’s electricity market has taken on a less anticompetitive bent during aspects of MPSC’s existence. See Hughes, 578 U.S. at 154 (discussing the movement toward divestment of energy monopolies in the late twentieth century); New York v. FERC, 535 U.S. 1, 7–8 (2002). Traditionally, vertically integrated monopolies (i.e., utility companies) controlled the generation, transmission, and distribution of electricity in Michigan. See Reliability Plans, 949 N.W.2d at 78. During these periods, the MPSC imposed price controls and other regulations on the utilities to limit abuse of their market power.

In 2000, Michigan opted to restructure its electric power industry. See Mich. Comp. Laws § 460.10 (Act 141). The generation and supply of electricity was opened to competitive suppliers, dubbed alternative energy suppliers or AESs. Customers now had options. They could continue to buy electricity from their incumbent electric utility. But they could also do so from one of the new players. See Commission History, MPSC, https://perma.cc/5X94-7TFJ. And while electricity distribution remained a regulated monopoly, Act 141 provided for the incumbent utilities to divest their transmission assets. Id.; Mich. Comp. Laws § 460.10w(1). What resulted was a hybrid system. Utilities continued to operate regulated generation and distribution services. But AESs now could provide electricity to customers at a market rate by Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 8 Mich. Public Serv. Comm’n, et al.

buying the electricity at wholesale and paying the utilities for use of their existing distribution system to furnish electricity to end users.

In 2008, Michigan backed away from the deregulation movement by limiting what AESs could contribute to the grid. From that point on, an incumbent utility’s distribution of electricity from an AES was capped at ten percent of the utility’s average weather adjusted retail sales for the preceding year. See Mich. Comp. Laws § 460.10a(1)(a) (Act 286). If the ten percent cap was met, a customer seeking to obtain electricity from an AES was placed in a queue and was otherwise limited to obtaining electricity from the utility itself. See Electric Customer Choice, MPSC, https://perma.cc/6662-EUCS. At present, “[d]ue to the limit on participation,” no AESs are enrolling new customers, and utilities control 90% of the marketplace. Id.

Against this backdrop, consider Michigan’s approach to regulating for resource adequacy. See Mich. Comp. Laws § 460.6w (Act 341). Through Act 341, the Michigan legislature directed the MPSC to require all LSEs (i.e., both utilities and AESs) to “demonstrate” that they “own[]” or have “contractual rights” sufficient to meet certain capacity obligations going forward. Id. § 460.6w(8). If an LSE fails to do so, it must buy capacity from either the incumbent utility or an AES at a set capacity charge. Id. § 460.6w(7)–(8). As a practical reality, because utilities dominate the local capacity market due largely to incumbency, utilities in effect are positioned to be the supplier of last resort should an AES fail to meet its capacity obligations. See In re Implementing Section 6w of 2016 PA 341 for Cloverland Elec. Coop., 942 N.W.2d 38, 43 (Mich. Ct. App. 2019).

Act 341 paved the way for the MPSC to set operating requirements for LSEs. Standards such as what capacity each LSE must demonstrate, and what cost LSEs will incur for failing to meet those minimum capacity requirements, were answered by the MPSC through a series of orders, which together established the Individual Local Clearing Requirement (ILCR). The ILCR echoes the MISO LCR in that it contemplates that some electric capacity be derived locally for reliability purposes. And the ILCR largely piggybacks on the MISO LCR in that it is based on MISO’s total capacity forecasts and MISO’s local resource zones. See In re the Investigation, on the Comm’n’s Own Motion, into the Elec. Supply Reliability Plans of Mich.’s Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 9 Mich. Public Serv. Comm’n, et al.

Elec. Utils. for the Years 2017 Through 2021, No. U-18197, 2017 WL 4155229, at *32 (MPSC Sept. 15, 2017).

But the ILCR differs from the MISO rules in a few key respects. For starters, Michigan requires LSEs to plan not just for the year ahead, but for four years into the future. See id. at *32. The ILCR also requires each LSE to produce or purchase a certain amount of locally generated energy. Id. This approach differs from MISO’s capacity requirements, which, recall, operate in the aggregate and allow an LSE to avoid acquiring any in-zone capacity without penalty so long as other in-zone generators supply sufficient capacity. In turn, if an LSE cannot satisfy its individual capacity obligations to the MPSC, its customers must buy capacity from an incumbent utility at a set price.

The MPSC did not apply the ILCR’s local generation requirement uniformly. Instead of applying the mandate throughout Michigan, the MPSC set Zone 2’s (the upper peninsula’s) local capacity metrics at zero percent of its total needed capacity. In re, on the Comm’n’s Own Motion, to Open a Contested Case Proc., No. U-18444, 2018 WL 3302792 (MPSC June 28, 2018). So the ILCR, in its current form, is only operative with respect to Zone 7—an area entirely within the lower peninsula. Id. And within Zone 7, the MPSC adopted an “incremental need approach,” wherein each LSE would initially need to have 2.7% of the amount of electrical capacity necessary to serve peak customer demand located in the zone, with that percentage increasing over time. Id.

What does this all mean for an LSE serving the Michigan retail market? Save for a voluntary stay of its orders during the pendency of this litigation, the MPSC’s orders require each LSE serving Zone 7 to increase gradually the amount of its electrical capacity generated locally in that zone, and, in turn, to guarantee that capacity for four years. In other words, every LSE (whether a utility or AES) needs to procure some amount of its total capacity from within the confines of Michigan’s lower peninsula.

C. This Litigation. Energy Michigan and the Association of Businesses Advocating Tariff Equity (ABATE) sued the MPSC and its individual commissioners, challenging the ILCR on dormant Commerce Clause grounds. Energy Michigan represents AESs, including entities Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 10 Mich. Public Serv. Comm’n, et al.

with “substantial capacity resources” throughout the country. ABATE is an association of industrial and manufacturing entities that purchase their electricity from AESs.

The case kept the district court occupied. An early motion resulted in MPSC’s dismissal on Eleventh Amendment grounds, leaving its individual members as defendants. In the same order, the district court rejected the argument that the FPA authorized the ILCR. Competing motions for summary judgment followed, but they did not resolve the case. The district court denied summary judgment to defendants on the view that the ILCR did not discriminate against similarly situated entities. At the same time, the district court rejected plaintiffs’ argument that the ILCR facially discriminates against interstate commerce, and determined that there were several fact disputes remaining as to whether the ILCR otherwise ran afoul of the dormant Commerce Clause. The district court then held a three-day bench trial, during which the parties amassed a record totaling nearly 20,000 pages. In the end, the court concluded that the ILCR did not violate the Commerce Clause.

Appeals and cross appeals followed. At bottom, each appeal centers on whether Michigan’s ILCR offends the dormant Commerce Clause. We review questions of law de novo, and review for clear error any questions of fact found by the district court. See Monasky v. Taglieri, 140 S. Ct. 719, 730 (2020); S.C. v. Metro. Gov’t, 86 F.4th 707, 714 (6th Cir. 2023).

II.

A. To appreciate the history behind the key legal issue in this case, turn back the clock to the years leading up to the Constitution’s ratification. In that era, where the colonies (and, later, states) had plenary power to regulate and even prohibit the movement of goods across their borders, regional economic rivalries were not uncommon. See Gibbons v. Ogden, 22 U.S. (9 Wheat.) 1, 224 (1824) (Johnson, J., concurring). States enacted their own tariffs on both foreign and interstate commerce, even going so far as to embargo certain products from neighboring states to protect domestic markets. See Brannon P. Denning, Confederation-Era Discrimination Against Interstate Commerce and the Legitimacy of the Dormant Commerce Clause Doctrine, 94 Ky. L.J. 37, 59–66 (2006). Those measures were met with retaliatory ones by other states. Id. at 62–63. The ensuing infighting impaired the fledgling nation’s economy and general stability. Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 11 Mich. Public Serv. Comm’n, et al.

Michael J. Klarman, The Framers’ Coup 23 (2016); see also James Madison, Vices of the Political System of the United States (Apr. 1787), reprinted in 2 The Writings of James Madison 361, 362–63 (Gaillard Hunt ed., 1901). The national government, meanwhile, could do little to intervene, as it lacked power under the Articles of Confederation to regulate commerce among the states. See Camps Newfound/Owatonna, Inc. v. Town of Harrison, 520 U.S. 564, 571 (1997).

The Constitution was a direct reaction to the chaos wrought by these protectionist measures. Tenn. Wine & Spirits Retailers Ass’n v. Thomas, 139 S. Ct. 2449, 2460 (2019) (“[R]emoving state trade barriers was a principal reason for the adoption of the Constitution.”); see also Dep’t of Revenue of Ky. v. Davis, 553 U.S. 328, 363–64 (2008) (Kennedy, J., dissenting) (recounting pre-ratification history). The document’s text and structure reflect as much. It empowers Congress “[t]o regulate Commerce with foreign Nations, and among the several States, and with the Indian Tribes.” U.S. CONST. art. I, § 8, cl. 3. Juxtaposed with this grant of authority are restrictions on the states, such as the general prohibition on imposing “Imposts or Duties on Imports or Exports,” see id. art. 1, § 10, cl. 2, a similar prohibition on laying of any “Duty of Tonnage,” id. art. 1, § 10, cl. 3, and the affirmation that citizens of each state are entitled to the “Privileges and Immunities of Citizens in the several States,” id. art. IV, § 2 cl. 1. These provisions underscore the “very structure” of the Constitution, which was “framed upon the theory that the peoples of the several states must sink or swim together.” See Nat’l Pork Producers Council v. Ross, 143 S. Ct. 1142, 1153 (2023) (cleaned up).

From this foundation, the Supreme Court would find a constitutional prohibition on states impermissibly interfering with interstate commerce. See, e.g., Baldwin v. G.A.F. Seelig, Inc., 294 U.S. 511, 522–23 (1935). Where precisely is that prohibition grounded in the Constitution? The Supreme Court generally reads Article I, Section 8’s affirmative grant of authority to Congress to regulate interstate commerce as likewise implicitly including a negative component forbidding certain state regulations, even in the absence of congressional legislation. Nat’l Pork Producers Council, 143 S. Ct. at 1152–53. But that is not a universal view. See Tenn. Wine & Spirits Retailers Ass’n, 139 S. Ct. at 2460 (cataloging Justices Scalia’s, Thomas’s, and Gorsuch’s views). In any event, the core quibble with much of this jurisprudence is not with its underlying principles, but instead its source for those principles. Nat’l Pork Producers Council, 143 S. Ct. Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 12 Mich. Public Serv. Comm’n, et al.

at 1152–53. All seem to agree that at the “very core” of the case law is an antidiscrimination principle, one that even the dormant Commerce Clause’s fiercest critics would concede is grounded in the original public meaning of the Constitution. Id. And a law that violates the principle is “per se invalid,” “save in a narrow class of cases” in which the state can show, “under rigorous scrutiny, that it has no other means to advance a legitimate local interest.” C & A Carbone, Inc. v. Town of Clarkstown, 511 U.S. 383, 391 (1994) (citation omitted).

B. What, then, is the antidiscrimination principle? At a broad level, the rule prohibits states from differentiating between “in-state and out-of-state economic interests” to benefit the former and burden the latter. Granholm v. Heald, 544 U.S. 460, 472 (2005) (citation omitted). So a state cannot “impose commercial barriers or discriminate against an article of commerce by reason of its origin or destination out of State.” C & A Carbone, 511 U.S. at 391. The “clearest example of such legislation is a law that overtly blocks the flow of interstate commerce at a State’s borders.” City of Philadelphia v. New Jersey, 437 U.S. 617, 624 (1978). The antidiscrimination principle is sometimes justified by the belief that a state will not police itself from imposing a discriminatory burden on those outside its borders, demanding judicial intervention to enforce the constitutional norm. See S.C. State Highway Dep’t. v. Barnwell Bros., 303 U.S. 177, 185 n.2 (1938); S. Pac. Co. v. Ariz. ex rel. Sullivan, 325 U.S. 761, 767 n.2 (1945) (collecting cases).

From these foundations, we recognize three ways a state law can violate the antidiscrimination rule: facially, purposefully, or in practical effect. E. Ky. Res. v. Fiscal Ct. of Magoffin Cnty., 127 F.3d 532, 540 (6th Cir. 1997). Of these, only the first and third are at issue here.

Start with facial discrimination. It occurs when a law “expressly” differentiates to favor in-state “commerce or entities” at the expense of out-of-state comparators. Truesdell v. Friedlander, 80 F.4th 762, 769 (6th Cir. 2023). Think of laws whose benefits or burdens are explicitly “territorially based”—that is, those concerned with city, county, or state lines. Maharg, Inc. v. Van Wert Solid Waste Mgmt. Dist., 249 F.3d 544, 551 (6th Cir. 2001); see also Interstate Towing Ass’n v. City of Cincinnati, 6 F.3d 1154, 1162 (6th Cir. 1993). Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 13 Mich. Public Serv. Comm’n, et al.

Contrast that with discrimination occurring in practice. The Commerce Clause forbids discrimination whether “forthright or ingenious.” Best & Co. v. Maxwell, 311 U.S. 454, 455 (1940). In limited ways, then, we can consider a law’s “effects” when looking at whether it is discriminatory. Effects-based discrimination concerns seemingly neutral state laws that necessarily create new market conditions that depend on geography. See Truesdell, 80 F.4th at 771–73; see, e.g., Foresight Coal Sales, LLC v. Chandler, 60 F.4th 288, 298 (6th Cir. 2023) (recognizing effects-based discrimination in a Kentucky law that “artificially discounted” the price of coal from states (including Kentucky) that imposed severance taxes, which necessarily burdened states without such taxes). We sometimes look more broadly at whether a law happens to impose more burdens on out-of-state entities than in-state counterparts. See Truesdell, 80 F.4th at 771; Garber v. Menendez, 888 F.3d 839, 843 (6th Cir. 2018). That function generally is the job of Pike balancing. See Pike v. Bruce Church, Inc., 397 U.S. 137, 142 (1970); Nat’l Pork Producers Council, 143 S. Ct. at 1164 n.4 (explaining that Pike balancing seeks to “smoke out purposeful discrimination” “as illuminated by those laws’ practical effects”). The parties, however, do not ask us to engage on that question.

Two other asides about the antidiscrimination principle deserve mention. First, while our focus is on discrimination between in-state and out-of-state “commerce or entities,” Truesdell, 80 F.4th at 769, that a law may also discriminate against in-state commerce is immaterial. See Dean Milk Co., 340 U.S. at 354 n.4. A state thus cannot “avoid the strictures of the Commerce Clause by curtailing the movement of articles of commerce” through part of the state, rather than all of it. Fort Gratiot Sanitary Landfill, Inc. v. Mich. Dep’t of Nat. Res., 504 U.S. 353, 361 (1992). Second, the “magnitude and scope of the discrimination have no bearing on the determinative question whether discrimination has occurred.” Assoc. Indus. of Mo. v. Lohman, 511 U.S. 641, 650 (1994). In other words, a state’s partial import ban, for purposes of the threshold discrimination inquiry, should be treated the same as a total ban, see Fort Gratiot Sanitary Landfill, Inc., 504 U.S. at 363, leaving questions of magnitude and scope for consideration at strict scrutiny. Waste Mgmt., Inc. v. Metro. Gov’t of Nashville & Davidson Cnty., 130 F.3d 731, 736 (6th Cir. 1997). Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 14 Mich. Public Serv. Comm’n, et al.

C. Measured against these principles, the ILCR is facially discriminatory. Whether a capacity requirement “count[s]” toward the ILCR hinges on the “zonal location” of the LSE’s resources. And that “zonal location,” in turn, depends on whether the resource is in a particular MISO Zone. Relevant here is a zone encompassing a subdivision of the State of Michigan. See Fort Gratiot Sanitary Landfill, Inc., 504 U.S. at 363 (recognizing facial discrimination based on a subdivision of a state). By requiring electricity to be generated from that zone, the ILCR relies on almost a “near perfect proxy” for the State’s lower peninsula. Foresight Coal, 60 F.4th at 297 (noting facial discrimination when a state law contains a “near perfect proxy” for geography). That leaves a law that is explicitly “territorially based,” requiring those that sell electricity in Michigan’s lower peninsula to procure a certain percentage of its electrical capacity from that region. Cf. Maharg, 249 F.3d at 551. Ignoring, as we must, the extent of the discrimination and the reasons for such discrimination at the threshold step of determining whether a discriminatory practice has in fact occurred, see Assoc. Indus. of Mo., 511 U.S. at 650, we can conceptualize the ILCR in the same way we would think of a law that wholly prohibits the procurement of out-of- state electrical capacity. That leaves perhaps the clearest example of a dormant Commerce Clause violation: “a law that overtly blocks the flow of interstate commerce at a State’s borders.” City of Philadelphia, 437 U.S. at 624.

1. History uniformly supports this conclusion. Lacking a clear textual anchor for the antidiscrimination principle at issue here, see Nat’l Pork Producers Council, 143 S. Ct. at 1152– 53, we start by considering founding-era precedent undergirding the principle. See N.Y. State Rifle & Pistol Ass’n, Inc. v. Bruen, 597 U.S. 1, 26–27 (2022) (considering first the text before turning to historical examples to construe the Constitution); United States v. Rahimi, 144 S. Ct. 1889, 1912 (2024) (Kavanaugh, J., concurring) (“History can supply evidence of the original meaning of vague text.”); Robert H. Bork, Neutral Principles and Some First Amendment Problems, 47 Ind. L. J. 1, 8 (1971) (“Where constitutional materials do not clearly specify the value to be preferred, . . . [t]he judge must stick close to the text and the history . . . .”). In this vein, experiences under the Articles of Confederation are a powerful tool in attempting to “discern the original meaning of the Constitution,” see New York v. United States, 505 U.S. 144, 163 (1992), or perhaps more precisely, “what the Constitution does not mean,” Rahimi, 144 S. Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 15 Mich. Public Serv. Comm’n, et al.

Ct. at 1914 (Kavanaugh, J., concurring). See also Haaland v. Brackeen, 143 S. Ct. 1609, 1664 (2023) (Thomas, J., dissenting); Bruen, 597 U.S. at 34 (observing that “when it comes to interpreting the Constitution, not all history is created equal,” and recognizing the primacy of historical evidence near the time of enactment). Especially so, it has been said, as those experiences inform the meaning of the Commerce Clause. See Tenn. Wine & Spirits Retailers Ass’n, 139 S. Ct. at 2461 (“[O]ur cases have long emphasized the connection between the trade barriers that prompted the call for a new Constitution and our dormant Commerce Clause jurisprudence”); see also Gregory E. Maggs, A Concise Guide to the Articles of Confederation as a Source for Determining the Original Meaning of the Constitution, 85 Geo. Wash. L. Rev. 397, 427 (2017).

Turning to that history, recall that the antidiscrimination principle is understood as a response to protectionist state law measures that proliferated during the pre-ratification period. See Tenn. Wine & Spirits Retailers Ass’n, 139 S. Ct. at 2460. One such law is a telling analog to the ILCR. Before the advent of electricity, New Yorkers in the 1780s regularly used firewood supplied from Connecticut as their primary source of home energy. See 1 John Bach McMaster, A History of the People of the United States from the Revolution to the Civil War, at 404 (3d ed. 1883). Over time, this arrangement was thought to be “ruinous” to New York’s domestic industry, prompting the state to incentivize New Yorkers to buy their firewood closer to home. See John Fiske, The Critical Period of American History: 1783–1789, at 146 (3d ed. 1899). New York imposed new taxes on imports from their neighbors in the Nutmeg State, meaning that “not a cart-load of Connecticut firewood could be delivered at the back-door of a country-house in Beekman Street [in Manhattan] until it should have paid a heavy duty.” Fiske, supra at 147; McMaster, supra at 404–05. Retaliatory measures ensued, leading to “chronic quarrels [that] were destroying . . . trade” between rival states. See Indep. Warehouses, Inc. v. Scheele, 331 U.S. 70, 94 (1947) (Jackson, J., dissenting). The later-ratified Constitution, Justice Robert Jackson observed, aimed “to free trade from local burdens and controls” like the 1787 New York firewood tax. Id. (discussing the same). Save for the fact that New York only taxed its out-of- state source of energy, rather than wholly barring it from entering the state, Michigan’s approach Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 16 Mich. Public Serv. Comm’n, et al.

to restricting the procurement of out-of-state energy echoes the experience from the Empire State nearly two-and-a-half centuries earlier.

2. Turn next to legal precedent. See Gamble v. United States, 587 U.S. 678, 685–86 (2019) (considering legal precedent after examining text and history); Rahimi, 144 S. Ct. at 1920 (Kavanaugh, J., concurring) (recognizing that “text, as well as pre-ratification and post- ratification history, may appropriately function as a gravitational pull” when interpreting existing precedent). There is no shortage of cases applying strict scrutiny to state laws that ban or restrict interstate transactions in favor of local ones. See, e.g., Heald, 544 U.S. at 493 (invalidating state law allowing only local wineries to ship alcohol directly to consumers); C & A Carbone, Inc., 511 U.S. at 394 (striking down local waste processing requirements); Taylor, 477 U.S. at 137–38 (subjecting a state law that blocked “all inward shipments of live baitfish” to strict scrutiny); City of Philadelphia, 437 U.S. at 629 (invalidating state law prohibiting the importation of waste into a state); Dean Milk Co., 340 U.S. at 354 (holding that a law requiring that any pasteurized milk sold in Madison, Wisconsin, be processed within five miles of the city’s center “plainly discriminates against interstate commerce”). That is no surprise. After all, it is black letter law that “[s]tate and local governments may not use their regulatory power to favor local enterprise by prohibiting patronage of out-of-state competitors or their facilities.” C & A Carbone, Inc., 511 U.S. at 394. And it is hard to think of a more apt description of what the ILCR aims to do— using the power of the state to favor local energy production at the expense of out-of-state production.

We are not the first court to apply these principles in the realm of electricity markets. Consider Wyoming v. Oklahoma, where the Supreme Court addressed whether an Oklahoma energy law violated the dormant Commerce Clause. At the time, the Sooner State required that all in-state electric utilities using coal-fired electric generating plants burn a mixture that contained at least ten percent Oklahoma-mined coal. Wyoming, 502 U.S. at 440, 442–44. By so doing, the Supreme Court concluded, Oklahoma impermissibly discriminated against out-of-state coal. In “expressly reserv[ing] a segment” of its market for Oklahoma-mined coal, the Sooner State’s law discriminated both on its face and in practical effect against interstate commerce. Id. at 455. Accordingly, the Supreme Court subjected the law to strict scrutiny, which Oklahoma Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 17 Mich. Public Serv. Comm’n, et al.

could not satisfy. Id. at 454, 456–57. Deeming the entire dispute “not a close case,” id. at 455 n.12, the Supreme Court invalidated the state law. Id. at 461. That was so even though the law’s purported effect was minimal, setting aside only a “small portion” of the Oklahoma market. Id. at 455. The scope of Oklahoma’s discrimination was of “no relevance,” the Supreme Court explained, “to the determination whether a State has discriminated against interstate commerce.” Id.

Thirty years later, the discriminatory nature of Michigan’s electricity distribution regime would no doubt make even Oklahoma’s regulators blush. Consider the following. Instead of regulating only a subset of electrical suppliers, the ILCR sweeps up all LSEs into its net. And rather than imposing a buy-local requirement for only one source of energy, the ILCR functionally mandates that all entities that supply any retail electricity in the lower peninsula of Michigan buy some percentage of their electrical capacity locally or have their own local generation. What was not a close case three decades ago is miles from what is constitutionally permissible today. Just as Oklahoma could not reserve a segment of its coal market for Oklahoma-mined coal to the exclusion of coal mined elsewhere, Michigan cannot reserve a segment of its electricity market for Michigan electricity to the exclusion of that generated in other states without being subject to strict scrutiny.

D. Defendants’ responses do not move the needle. They begin by arguing that the ILCR does not facially discriminate because its language does not mention any state boundaries. As a result, they say, the ILCR requires the “additional step” of considering whether the LSE has demonstrated access to a resource within a “particular [MISO] zone.” But that is not much of a step. MISO zones are geographic regions. And the relevant MISO zone corresponds with the borders of Michigan’s lower peninsula. So the ILCR’s facial discrimination is obvious. If the underlying resource is within the borders of the lower peninsula, it counts toward the ILCR. Otherwise, it does not. In any event, the Constitution “deals with substance, not shadows.” Students for Fair Admissions, Inc. v. President & Fellows of Harvard Coll., 143 S. Ct. 2141, 2176 (2023) (citation omitted). So we need not quibble with whether such a step puts the ILCR under the rubric of facial discrimination or kicks it into the effects territory. See Foresight Coal, Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 18 Mich. Public Serv. Comm’n, et al.

60 F.4th at 297. Either way, because the law discriminates between articles in interstate commerce (here, electrical capacity), strict scrutiny applies.

Pressing ahead, defendants maintain that because the Michigan rules impose identical burdens to “both incumbent utilities and the [AESs],” the latter of which may be based outside of Michigan, no constitutional infirmity exists. Comm’rs Second Br. at 58. True, as defendants suggest, a state can violate the Commerce Clause by imposing different burdens on in-state and out-of-state entities. Truesdell, 80 F.4th at 769. And perhaps plaintiffs have not shown sufficient discrimination between utilities and AESs. But the discrimination at issue here is not isolated to the Michigan rules’ effects on those that supply electricity at retail. Instead, it is primarily aimed at an article in commerce—electrical capacity. And the discriminatory policies result in differential treatment between electricity generated in-state and that derived out-of-state.

Nor do we have any preservation concerns. Plaintiffs pressed this ground of discrimination—that the ILCR discriminates between in-state and out-of-state energy—in the district court. They likewise argued the point defendants refute here—that the ILCR discriminates in practical effect against AESs in favor of utilities. On appeal, plaintiffs again make both arguments. And they only need prevail on one. So even if defendants are correct in their assessment of any purported discrimination between utilities and AESs in commerce, that does not lessen plaintiffs’ theory of discrimination tied to the procurement of the underlying commodity, electricity. Said differently, cases upholding laws that facially impose uniform burdens on certain in-state and out-of-state entities on the same underlying product do little to respond to the chief claim of facial discrimination here. See, e.g., Am. Beverage Ass’n v. Snyder, 735 F.3d 362, 371 (6th Cir. 2013) (holding that a Michigan law that requires all beverage manufactures to impose a unique mark designation on all beverages is a facially neutral law).

Next, defendants and the dissenting opinion attempt to distinguish Wyoming on the basis that the underlying purpose of the ILCR is to promote resource adequacy, not protect domestic industry. Divining a law’s purpose, however, is tricky business. Indeed, it is the rare law from which one can deduce a single underlying purpose. See Henson v. Santander Consumer USA Inc., 582 U.S. 79, 89 (2017) (recognizing that no law “yet known” pursues one stated purpose at Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 19 Mich. Public Serv. Comm’n, et al.

all costs (citation omitted)); Truesdell, 80 F.4th at 773 (questioning whether a “single ‘intent’ behind legislation” can be determined for purposes of Commerce Clause scrutiny). It is thus perhaps no surprise that plaintiffs have a different understanding of the law, namely, that the ILCR was intended to drive AESs out of the marketplace. See Oral Arg. at 10:30 (appellant arguing that the ILCR is intended to “chase[]” AESs out of the market). In the end, any assessment of purpose is irrelevant. When considering whether a state law violates the Commerce Clause’s antidiscrimination principle, inquiry into purpose is distinct from inquiry into the law’s text. See Truesdell, 80 F.4th at 769. Even the most benign purpose, for instance, cannot save a facially discriminatory law from strict scrutiny. Defenders of Oklahoma’s locally mined-coal requirement made a nearly identical argument—that the regulation there was needed to ensure a reliable supply of energy in Oklahoma. 502 U.S. at 456–57 (addressing Oklahoma’s argument that the coal rule serves to “lessens the State’s reliance on a single source of coal delivered over a single rail line”). At most, “justifications” for a discriminatory law are a topic for strict scrutiny, not for the threshold question whether the law is discriminatory in the first place. Id. That is true whether those justifications address coal usage in Oklahoma or electrical supply in Michigan.

It makes no difference that the ILCR does not limit an LSE’s ability to import electricity into Michigan. Here, we note a distinction between electricity and electric capacity. Electricity is the end product, whereas electric capacity refers to the ability to produce that product when necessary—functionally an option contract to purchase electricity. See Conn. Dep’t of Pub. Util. Control, 569 F.3d at 479; Entergy Ark., LLC v. FERC, 109 F.4th 583, 587–88 (D.C. Cir. 2024) (describing capacity as a “commitment[] from a generator to produce set amounts of electricity in the future” (alteration in original) (quotations omitted)). The ILCR regulates electrical capacity; it requires LSEs to have a set percentage of electrical capacity based in Michigan. That rule facially restricts how an article of commerce, specifically, electricity as procured through the “call option[s]” that generators sell, can be obtained based on geography. Conn. Dep’t of Pub. Util. Control, 569 F.3d at 479; see also C & A Carbone, 511 U.S. at 391 (recognizing that unlawful discrimination under the Commerce Clause can extend beyond discriminating against an underlying product to the means associated with processing that product). Because the Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 20 Mich. Public Serv. Comm’n, et al.

amount of discrimination is irrelevant to the question of whether discrimination exists, we gauge the ILCR the same way we would a requirement that a far larger amount of electrical capacity— say 100% or 200% of peak demand—be procured from Michigan, or even an entire ban on electricity supply derived outside the state’s borders. See Fort Gratiot Sanitary Landfill, Inc., 504 U.S. at 363; Oral Arg. at 19:00 (intervenor-appellant conceding that “if the rule was . . . the only energy you can supply had to be . . . produced in Zone 7” the analysis “doesn’t change whatsoever”).

III.

Two remaining contentions deserve separate attention. One is that the ILCR is lawful under what defendants view to be an exception to the dormant Commerce Clause enunciated in General Motors Corp. v. Tracy, 519 U.S. 278 (1997). Two, that Congress authorized the ILCR in the FPA. We turn to those arguments now.

A. General Motors v. Tracy. Some background on Tracy is necessary. The dispute there concerned Ohio’s natural gas market as it existed in the latter part of the twentieth century. Like the electrical market, our nation’s natural gas market traditionally had an organically monopolistic structure, with large utilities controlling the local distribution of natural gas. Id. at 283. Federal deregulation in the late 1970s and early 1980s, however, opened access to interstate natural gas pipelines, allowing producers and independent marketers of natural gas to sell to those who purchased gas from the interstate market. Id. at 282–84. Ohio’s intrastate pipelines, however, remained under the control of local utilities. Id. at 284. To gain access to the competitive interstate market, large industrial end users in Ohio began constructing their own pipelines in the Buckeye State. Id. With Ohio natural gas utilities at risk of losing these customers, Ohio “took steps . . . to keep some income from large industrial customers within the utility system” by allowing industrial users in Ohio to buy natural gas from the public utilities or independent marketers in Ohio and pay fees to the utilities to use their intrastate pipelines. Id. The result was two markets operating side by side: a new competitive market for large industrial concerns along with a residual captive market for residential users who could purchase natural gas from the utilities only. Id. at 293–94. Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 21 Mich. Public Serv. Comm’n, et al.

Part and parcel of this regulatory scheme, Ohio law imposed general sales and use taxes on all natural gas purchases but exempted public utilities from those taxes. Id. at 281–82. After Ohio applied a use tax to GM’s purchases of natural gas from independent marketers, GM sought an exemption in state court. Id. at 285. The auto giant argued in part that the tax exemption violated the Commerce Clause as a facially discriminatory tax in that it granted a tax exemption to sales from the utilities (all of which were located in Ohio) while imposing a tax on other natural gas sales. Id. at 288. When the Ohio courts disagreed with GM, the company pursued its claims before the United States Supreme Court.

Yet it fared no better. To start, the Supreme Court rejected GM’s theory on a “threshold” ground. “[A]ny notion of discrimination” under the dormant Commerce Clause, the Supreme Court explained, presupposes a “comparison of substantially similar entities,” that is, entities that compete against each other in a single market. Id. at 298, 300. Ohio’s natural gas market, however, consisted of two distinct (but related) markets, with utilities’ profits in the noncaptive market helping subsidize the captive one. Id. at 301–02. With this understanding in mind, it was left to the Supreme Court to decide whether to “accord controlling significance” to the “noncaptive” market in which utilities and marketers compete, or, as Ohio urged, the “noncompetitive, captive market in which the local utilities alone operate[.]” Id. at 303–04.

Ohio’s esteemed advocacy prevailed. Tracy opted to give “greater weight to the captive market,” and thus “to treat marketers and [utilities] as dissimilar for present purposes” for three largely pragmatic reasons. Id. at 304. One, invalidating the tax exemption could “imperil” the captive market by reducing a competitive advantage in the noncaptive market—namely, eating into the customer base of large industrial entities that help reduce costs for individual consumers. Id. at 304–07. Second, judges lack expertise in predicting the effects of judicial intervention on the utilities’ capacity to serve the captive market. Id. at 304, 308–09. And third, Congress was best situated to strike the appropriate balance in this setting. Id. at 304, 309–10. In the end, far from establishing an exception to the dormant Commerce Clause, Tracy simply clarified what qualifies as discrimination in the unique regulatory setting in which that case arose. Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 22 Mich. Public Serv. Comm’n, et al.

1. Noting some factual similarities between Tracy and this litigation, defendants and the dissenting opinion maintain that the cases should have a similar outcome. To be sure, surface- level comparisons can be made. Chief among them, both cases involve hybrid energy markets: one part noncompetitive, consisting of only publicly regulated utilities serving residential customers; the other part competitive with regard to a small segment of large companies.

Yet critical differences remain. Most notable is the nature of the discrimination at play. In Tracy, the Supreme Court considered whether Ohio’s tax law discriminated between in-state and out-of-state retailers, in particular, by treating a utility differently from a non-utility with respect to each entities’ sales and use taxes. Here, on the other hand, the ILCR’s facial discrimination concerns the geographic origins of a product purchased at wholesale, as Michigan law explicitly favors local electrical capacity. In so doing, the ILCR expressly differentiates on a geographic basis where an LSE can procure electrical capacity. If a Michigan LSE is seeking to satisfy the ILCR’s local generation requirements, contracting with an Indiana wind farm or a major Ohio electrical utility is of no use. Under the ILCR, remember, the LSE (and functionally the Indiana and Ohio firms) would be penalized for doing so. Any differences between an AES and a Michigan utility, therefore, are immaterial. The ILCR’s preference for in-state electrical capacity harms a range of stakeholders in the energy supply chain seeking to sell electricity into the Michigan grid.

That factual distinction makes all the difference. Much of defendants’ Tracy argument is premised upon the idea that AESs and utilities are different creatures. The dissenting opinion likewise would apply Tracy as if the only discrimination afoot here is between Michigan utilities and AESs. See Dissenting Op. at 34–40. But the ILCR’s discrimination is not aimed—at least facially—at AESs; indeed, the ILCR by its terms treats both utilities and AESs identically, as both are LSEs. Instead, the ILCR’s wrath is turned on out-of-state electrical capacity—and, as a result, those entities with “capacity resources” outside of Michigan, including some plaintiffs here. There is no distinction between the electrical capacity these entities offer at wholesale to AESs relative to what in-state utilities offer, save for geography. Indeed, LSEs regularly buy and sell capacity at wholesale, as their products are interchangeable on a national grid at that part of the stream of commerce. See Elec. Power Supply Ass’n, 89 F.4th at 550. Tracy simply demands Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 23 Mich. Public Serv. Comm’n, et al.

that the “objects of the disparate treatment” be similarly situated before a law may be deemed to have run afoul of the dormant Commerce Clause. Camps Newfound, 520 U.S. at 601 (Scalia, J., dissenting). Here, the most obvious objects of the ILCR’s discrimination, in-state and out-of- state electrical capacity, so qualify.

After dismissing the argument that the Ohio tax exemption was discriminatory based on how the law treated out-of-state marketers vis-à-vis in-state utilities, the Supreme Court separately considered whether the exemption was discriminatory because it favored natural gas purchases from in-state utilities as opposed to similar purchases from out-of-state utilities. Tracy, 519 U.S. at 310. Tracy rejected this secondary argument. But not, it bears emphasizing, by utilizing the just-discussed similarly situated analysis. It did so instead on the grounds that the Ohio tax exemption was likely to be extended to out-of-state utilities. Id. at 311.

This appeal presents the same secondary issue as in Tracy: does the ILCR facially discriminate between electrical generation from similarly situated entities, for example, in-state and out-of-state utilities? And here, unlike in Tracy, there is no denying that discrimination is afoot. The ILCR facially treats out-of-state generation differently than in-state generation.

2. What should we make of Tracy’s recognition of the public’s need for dependable energy and the importance of the captive market in Ohio? At times, to be sure, Tracy speaks warmly of the need for state regulation of energy markets and the desire for courts to avoid interfering with state protectionism of public utilities. See id. at 304–10; Camps Newfound, 520 U.S. at 602 (Scalia, J., dissenting) (“Tracy paints a compelling image of people shivering in their homes in the dead of winter without the assured service that competition-sheltered public utilities provide.”). Whatever value one draws from those observations, they do not change our conclusion here. Keep in mind the context in which Tracy arose. Against the backdrop of the preexisting natural monopoly, Ohio regulated a narrow, noncaptive natural gas market for large businesses to help prop up the public utilities in the captive market. Tracy, 519 U.S. at 282–84. In the absence of a noncaptive market, utilities would lack an “adequate customer base” to continue to serve Ohio’s captive natural gas market. Id. at 309. Customers in that market were captured economically in that they depended on a stable rate and supply and had no economic Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 24 Mich. Public Serv. Comm’n, et al.

ability to participate in the noncaptive market. Id. at 301–02. In the end, Tracy simply preserved Ohio’s ability to retain its natural monopoly to ensure residential customers continued to obtain needed services.

Michigan’s regulatory history tells a very different tale. The state’s captive market did not result directly from a natural monopoly, as Michigan ended its natural electric monopoly in 2000 through the creation of its choice program. See 2000 Mich. Legis. Serv. P.A. 141 (S.B. 937) (codified at Mich. Comp. Laws § 460.10). Only later did Michigan opt artificially to create a captive market and a noncaptive market with the ten percent choice cap rule. See 2008 Mich. Legis. Serv. P.A. 286 (H.B. 5524) (codified at Mich. Comp. Laws § 460.10a(1)). At the time, “utility customers vigorously resisted” the imposition of the choice cap. See Lisa Babcock and Rodger Kershner, Changes in the Law Governing Public Utilities, Mich. Bar. J., Jan. 2011, at 37, 39, https://perma.cc/ZZV7-D2RH. Nor did all customers support Act 341 and the resulting ILCR, which was pushed by the utilities themselves. To this day, there is a long waiting list to obtain electricity from an AES in Michigan, as electricity prices have increased since the end of full deregulation. See Electric Customer Choice, MPSC, https://perma.cc/6662-EUCS; A Policy Guide to Energy Choice in Michigan, Mackinac Ctr. for Pub. Pol’y, https://perma.cc/26FX- WL7E.

These distinct contexts deprive Tracy of any vitality here. Unlike the Buckeye State’s retail natural gas customers, the Wolverine State’s residential electricity customers are not economically captive to their utility. Quite the opposite, in fact. Rather, Michiganders are living with an artificially created market due to rent seeking by the utilities. See Comm’rs’ Second Br. at 34 (“The Tracy court makes clear that the two gas markets were divided on purely economic grounds . . . . Michigan law, on the other hand, expressly limits the electric load that [AESs] may serve.”). Relatedly, unlike the tax exemption at play in Tracy, there is no evidence that Michigan utilities rely on the ILCR’s effects on the noncaptive market to subsidize or prop up the captive one. See 519 U.S. at 309. The ILCR may help answer reliability concerns for the energy market as a whole, although that is not a certainty, as defendants acknowledge. See Comm’rs’ Br. at 39–40 & n.9 (recognizing that the interconnected nature of the grid coupled with MISO procedures for service interruptions mean that service issues would not be uniquely Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 25 Mich. Public Serv. Comm’n, et al.

experienced by retail customers). Either way, a state’s generic interest in local energy reliability does not exempt it from Commerce Clause scrutiny. Wyoming, 502 U.S. at 456. In sum, Tracy may justify a state’s efforts to retain a natural monopoly for economically dependent retail consumers. But it does not bless a state’s efforts to aid an artificial monopoly, as defendants would have us hold.

3. At bottom, defendants press for an aggressive, policy driven rule divorced from the factual setting of Tracy. If a state has a captive market overseen by a public utility (e.g., the state authorizes only one entity to provide a service and regulates the rate that entity charges, etc.), then the state, they contend, can otherwise discriminate against interstate commerce with respect to a parallel noncaptive market in which the utility also operates.

Perhaps there is a policy argument for such a public utility exception to the dormant Commerce Clause. But a legal one? Not in the original public meaning of the Constitution. As discussed, the founding generation saw no distinction between state laws discriminating in the energy arena as opposed to any other means of trade. See supra 15–16. And there is nothing to suggest that the Framers, well-familiar with the concept of public service companies, implicitly sought to exempt state laws that favored such entities from the Constitution’s antidiscrimination principle. See Biden v. Knight First Amend. Inst. at Columbia Univ., 141 S. Ct. 1220, 1222 (2021) (Thomas, J., concurring) (“[O]ur legal system and its British predecessor have long subjected certain businesses . . . to special regulations, including a general requirement to serve all comers.” (citation omitted)). When given a choice between an expansive reading of a precedent on policy grounds and the original public meaning of the Constitution, we should pick the latter. See Johnson v. Bauman, 27 F.4th 384, 394 (6th Cir. 2022); Rahimi, 144 S. Ct. at 1920 (Kavanaugh, J., concurring).

Nor does a public utility or energy exception to the dormant Commerce Clause derive from Tracy itself. At least twice, the opinion recognizes that no such exception exists. See 519 U.S. at 291 n.8 (“[U]tilities should not be insulated from our contemporary dormant Commerce Clause jurisprudence . . . .”); id. at 307 n.15 (“[I]f a state discriminates against out-of-state interests by drawing geographical distinctions between entities that are otherwise similarly Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 26 Mich. Public Serv. Comm’n, et al.

situated, such facial discrimination will be subject to a high level of judicial scrutiny even if it is directed toward a legitimate health and safety goal.”).

Looking to other cases only makes matters worse. None from the Supreme Court endorse defendants’ broad reading of Tracy. Its dormant Commerce Clause precedents, which Tracy did not purport to touch, have long been applied to regulations concerning in-state utilities. See, e.g., Wyoming, 502 U.S. at 461 (invalidating an Oklahoma law that required in-state utilities to supply ten percent of their needs for fuel from Oklahoma coal); New England Power Co. v. New Hampshire, 455 U.S. 331, 339 (1982) (invaliding a New Hampshire law prohibiting a utility “from selling its hydroelectric energy outside the State”).

As for the circuit courts, we have applied Tracy narrowly for the more modest proposition that Commerce Clause discrimination presupposes discrimination between two similar entities or articles of commerce. See, e.g., LensCrafters, Inc. v. Robinson, 403 F.3d 798, 804 (6th Cir. 2005) (citing Tracy to recognize that a law that in effect harmed optical companies to favor optometrists did not violate the Commerce Clause because its benefits and burdens were meted out to non-similarly situated entities); Paul’s Indus. Garage, Inc. v. Goodhue County, 35 F.4th 1097, 1100 (8th Cir. 2022) (listing various examples of Tracy’s application to dissimilar entities—vacation homes v. primary residences, humane societies v. for-profit breeders, brick- and-mortar stores v. online counterparts); Allco Fin. Ltd. v. Klee, 861 F.3d 82, 105 (2d Cir. 2017) (citing Tracy and concluding that a law distinguishing between two distinct inventions of state law—Georgia renewable energy credits used for Georgia’s renewable portfolio standard and credits issued under Connecticut law—did not discriminate). And our most recent pronouncement on Tracy cited it for the rule opposite one defendants press here. As we explained, “‘ordinary’ . . . negative Commerce Clause” principles apply “to all energy regulations.” Truesdell, 80 F.4th at 780. This view echoes that of the Fifth Circuit, which held that Tracy cannot be read to immunize public electric utilities from ordinary Commerce Clause jurisprudence. NextEra Energy Cap. Holdings, Inc. v. Lake, 48 F.4th 306, 318–20, 325 (5th Cir. 2022) (“What is true for alcohol and milk under the dormant Commerce Clause must be true for electricity transmission.”). What is more, both the Trump and Biden Departments of Justice seem to agree that Tracy should be read in line with the “case specific factors” at play with the Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 27 Mich. Public Serv. Comm’n, et al.

Ohio sale tax exemption, and thus should not be understood to create a broad “public utility” exception to the dormant Commerce Clause. See Brief of the United States of America as Amicus Curiae at 10–11, LSP Transmission Holdings, LLC v. Sieben, 954 F.3d 1018 (8th Cir. 2020) (18-2559), 2018 WL 5318514; Brief of the United States of America as Amicus Curiae at 14, Lake v. NextEra Energy Cap. Holdings, Inc., 144 S. Ct. 485 (22-601), 2023 WL 7002451.

Limiting Tracy to its unique factual setting makes good sense. The alternative would exempt a major sector of the U.S. economy from the Commerce Clause. And it would license blatant economic protectionism when a state favors a utility in a noncaptive market. As the Fifth Circuit persuasively described things, “provid[ing] Commerce Clause immunity to any law that grants a preference to a company that has at least one foot in a captive market” would allow states to “grant in-state utilities the exclusive right to operate coal mines in the state (or, for that matter, the exclusive right to sell ice cream in the state).” NextEra Energy, 48 F.4th at 320. That a state might structure its market to enact discriminatory laws to support its utilities is not purely fanciful. Remember Michigan’s approach here. The state artificially created a captive market, going so far as to place utilities in the position of being the provider of last resort should any competitor fail to meet its capacity obligations. See Cloverland Elec. Coop., 942 N.W.2d at 43. Now, defendants and the dissenting opinion rely on those features of state law to argue that it is exempt from ordinary dormant Commerce Clause principles. The Constitution stands in their way.

4. Turning to the dissenting opinion, it offers two primary critiques. The first centers on our reading of Wyoming and Tracy. Starting with Wyoming, the dissenting opinion characterizes the express geographic discrimination at play here as both necessary and “different in kind” than the discrimination at play there. See Dissenting Op. at 40–41 (maintaining that any discriminatory aspects of Michigan’s regulatory scheme are lessened because the ILCR is built on MISO Zones, whose connection to state borders is “incidental” and based on “technical judgments about grid reliability”); id. at 42 (distinguishing Wyoming because the reliability arguments here are “different in kind from Oklahoma’s”). But however significant one might deem the “justifications” for a nonetheless facially discriminatory law, we view those justifications, as did the Supreme Court in Wyoming and elsewhere, through the lens of strict Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 28 Mich. Public Serv. Comm’n, et al.

scrutiny, not as a part and parcel of a threshold question concerning the proper level of scrutiny. 502 U.S. at 456–57; Assoc. Indus. of Mo., 511 U.S. at 650. And we see Tracy as largely inapplicable here for reasons discussed above—that is, the ILCR facially discriminates with respect to the wholesale market, as opposed to the retail side (the only issue at play in Tracy), and because Michigan’s approach to energy regulation differs materially from the Ohio scheme at issue in Tracy. See supra at 22–25. On the latter front, the dissenting opinion seemingly misunderstands our position. We do not critique Michigan’s regulatory choices. See Dissenting Op. at 39 n.8. Instead, we view the fact that Michigan created a captive market as distinctive from Ohio’s experience and thus instructive as to Tracy’s reach. See supra at 23–25, 27. After all, if Tracy were read as also honoring Michigan’s approach, states could simply create captive markets, insulate market participants from all interstate competition, and immunize them from any Commerce Clause scrutiny. See NextEra Energy, 48 F.4th at 320.

More broadly, the dissenting opinion faults us for failing to “give full weight to the judgment of state and local regulators on a matter of state and local concern.” Dissenting Op. at 39. We appreciate this general sentiment. The Constitution, lest we never forget, envisions states as separate sovereigns who are generally afforded discretion to enact a wide range of policy choices, judgments we must respect. See New State Ice Co. v. Liebmann, 285 U.S. 262, 311 (1932) (Brandeis, J., dissenting); L.W. ex rel. Williams v. Skrmetti, 83 F.4th 460, 487 (6th Cir. 2023); cf. Russo v. City of Cincinnati, 953 F.2d 1036, 1050 (6th Cir. 1992) (Suhrheinrich, J., concurring in part and dissenting in part) (“[W]hen a federal court reviews municipal or state executive conduct or policy . . . it must be very careful not to violate principles of federalism . . . [and] remain ever-mindful of its limited competence” in reviewing such laws); Daunt v. Benson, 999 F.3d 299, 326–27 (6th Cir. 2021) (Readler, J., concurring in the judgment) (criticizing placing “a judge’s inherent policy preferences front-and-center” to invalidate state laws). Yet it is beyond dispute that our precedent, to say nothing of the Constitution and our founding history, contemplates a less passive role for the judiciary when a state interferes with interstate commerce. See Tenn. Wine & Spirits Retailers Ass’n, 139 S. Ct. at 2460. Especially so in the face of express discrimination. See Nat’l Pork Producers Council, 143 S. Ct. at 1152–53. In the end, energy regulation, which is not excepted from this constitutional history, is simply not a Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 29 Mich. Public Serv. Comm’n, et al.

matter of exclusive “state and local concern.” See supra at 10–12 (discussing the historical foundation for the anti-discrimination principle); id. at 14–17 (applying founding-era precedent supporting the anti-discrimination rule as applied to energy regulation, before turning to relevant judicial precedent).

This understanding leads us to reject the dissenting opinion’s novel approach to Commerce Clause challenges to state energy regulations: divining whether the underlying purpose of the state law undermines “a national market for competition,” and, if so, weighing any associated concerns against the benefits the law bestows upon the “vital” electricity market. See Dissenting Op. at 43. Whatever the perceived merits of such a grand balancing test, see Antonin Scalia, The Rule of Law as a Law of Rules, 56 U. Chi. L. Rev. 1175, 1180, 1187 (1989), both the Constitution and our precedent place Congress—not this Court—as the primary body that may ascertain the benefits and burdens of a facially discriminatory law. See Bendix Autolite Corp. v. Midwesco Enters., Inc., 486 U.S. 888, 897–98 (1988) (Scalia, J., concurring in the judgment); Merrion v. Jicarilla Apache Tribe, 455 U.S. 130, 154 (1982) (acknowledging Congress’s role in limiting judicial review of facially discriminatory laws “[o]nce Congress acts” and “has struck the balance it deems appropriate” as to a state’s role in burdening commerce). When a state expressly discriminates against interstate commerce, including in the energy sector, our role is simply to consider the state’s interests through the lens of strict scrutiny.

What is more, as a practical matter, the dissenting opinion’s approach seems to collapse upon itself. After evaluating the Michigan market, the dissenting opinion would uphold Michigan’s regulatory scheme by giving “full weight to the judgment of state and local regulators.” See Dissenting Op. 36–39. But that evaluation is premised entirely on the trial court’s findings, which followed a trial that entertained the parties’ competing views on the ILCR’s benefits. See id. at 39–40 (quoting at length from the district court’s post-trial opinion). Yet the dissenting opinion, remember, does not think that a trial should ever have come to pass, believing the district court “erred” in not granting defendants’ summarfpay judgment motions. See id. at 34. And those pretrial Tracy arguments, it bears emphasizing, never delved into the policy considerations underlying Michigan’s regulatory approach; they instead were premised on the mere existence of a captive retail energy market. Only by reverse engineering the district Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 30 Mich. Public Serv. Comm’n, et al.

court proceedings and invoking an argument defendants never made in district court can the dissenting opinion now give “full weight” to the views of local regulators expressed at trial. See id. at 39.

Finally, we likewise reject the suggestion that our opinion is simply a veiled criticism of how Michigan “chooses to organize its regulatory environment.” See id. at 39 n.8. Our analysis of the underlying constitutional history and precedent belies that assertion. See Rahimi, 144 S. Ct. at 1920 (Kavanaugh, J., concurring) (“[R]eliance on history is more consistent with the properly neutral judicial role than an approach where judges subtly (or not so subtly) impose their own policy views on the American people.”). So too should notions of judicial modesty. Like all jurists, we admittedly are poorly equipped to offer an informed view of energy regulation, a deeply complex subject. See Am. Elec. Power Co. v. Connecticut, 564 U.S. 410, 427–28 (2011) (explaining that “[f]ederal judges lack the scientific, economic, and technological resources” to evaluate fully the “competing” environmental and economic interests implicating federal energy policy); Elec. Power Supply Ass’n, 577 U.S. at 295 (recognizing the judiciary’s “limited role” in assessing electricity regulation).

B. The Federal Power Act. That leaves one remaining argument from defendants: that the FPA authorizes the ILCR, functionally allowing Michigan to discriminate against interstate commerce. This argument invokes a peculiarity of dormant Commerce Clause jurisprudence. Ordinarily, Congress cannot license the states to violate the Constitution. See Tyler Pipe Indus., Inc. v. Wash. State Dep’t of Revenue, 483 U.S. 232, 263 n.4 (Scalia, J., concurring in part and dissenting in part). That said, Congress can authorize state or local laws that the negative Commerce Clause would otherwise prohibit. Prudential Ins. v. Benjamin, 328 U.S. 408, 418–27 (1946). This seemingly unusual practice stems in part from the text of Article I, which authorizes Congress to “regulate Commerce,” see U.S. CONST. art. I, Sec. 8, cl. 3. That provision has been understood to afford Congress the power to prohibit or restrict commerce across the nation. Lottery Case, 188 U.S. 321, 328 (1903). Congressional intervention, it is said, alleviates political process concerns driving the antidiscrimination principle. As Congress reflects “all segments of the country,” when the national legislature acts to permit discriminatory state conduct, the logic goes, there is “significantly less danger” that the action is meant merely Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 31 Mich. Public Serv. Comm’n, et al.

to benefit one state exploiting another. S.-Cent. Timber Dev., Inc. v. Wunnicke, 467 U.S. 82, 92 (1984). But to ensure that there has been “such a collective decision,” the Supreme Court imposes a clear statement rule in this context. Id. Congress must “manifest its unambiguous intent” before a federal law is read to allow a state to “discriminat[e] against interstate commerce.” See Nw. Airlines, Inc. v. County of Kent, 510 U.S. 355, 373 n.19 (1994) (citation omitted).

What provision of the FPA amounts to a clear statement authorizing Michigan to discriminate against out-of-state energy through the ILCR? Defendants point to § 201(b)(1), which removes from federal jurisdiction (and preserves for the states) “facilities used for the generation of electric energy.” 16 U.S.C. § 824(b)(1). Fairly read, the provision does recognize state authority over local generation, which presumably extends to regulating for resource adequacy. But it is difficult to see how this provision authorizes, let alone unambiguously so, Michigan to use its authority over local energy generation to discriminate against interstate commerce through the ILCR. True, § 201 provides a “clear and specific grant of jurisdiction to FERC over interstate transmissions.” See New York, 535 U.S. at 22 (quotations omitted). But that tells us little about whether the same provision is clear enough to immunize a state from Commerce Clause scrutiny.

In the end, defendants run headlong into a familiar foe: Wyoming. Recall that after agreeing that Oklahoma’s ten percent ban on outside coal violated the Commerce Clause, the Supreme Court considered whether the FPA’s reserving to the states the authority to regulate local retail electric rates (also found in § 201(b)(1)) authorized the discrimination. 502 U.S. at 458. That provision, Wyoming recognized, simply left “standing” a state’s rate-regulating authority. Id. (citing New England Power Co., 455 U.S. at 341). Accordingly, Congress in the FPA did not clearly and unambiguously “permit the discrimination against interstate commerce occurring” because of the Oklahoma coal law. Id. While defendants highlight a different part of the FPA’s savings provision, one that preserves a state’s authority over generation, the logic of Wyoming applies with equal force here—recognition of a state’s general authority to regulate does not amount to a clear and unambiguous statement immunizing the state from Commerce Clause scrutiny when it acts under that general authority. Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 32 Mich. Public Serv. Comm’n, et al.

IV.

Given our view that the ILCR facially discriminates against interstate commerce and that the FPA does not immunize the ILCR, strict scrutiny governs. That imposing standard requires defendants to show that the ILCR “advances a legitimate local purpose that cannot be adequately served by reasonable nondiscriminatory alternatives.” New Energy Co. of Ind. v. Limbach, 486 U.S. 269, 278 (1988); Maine, 477 U.S. at 138. The district court, we note, engaged in the strict scrutiny calculus as an alternative holding to its determination that the ILCR was nondiscriminatory. In concluding that the ILCR would survive strict scrutiny, the district court considered two purposes of the law—ensuring grid reliability and doing so in an equitable manner (framed as “all customers bear[ing] the cost of providing reliability through local resources”)—against four alternatives to the ILCR offered by plaintiffs. In the end, the district court concluded that “[n]one” of plaintiffs’ proposed “alternatives” would satisfy the purposes served by the ILCR.

We see at least two problems with this approach. First is the purported “legitimate local purpose[s]” advanced by the law. See New Energy Co., 486 U.S. at 278. Ensuring a reliable energy supply is an understandable issue of local interest. Cf. Hughes v. Oklahoma, 441 U.S. 322, 337 (1979) (recognizing state interest in health and safety of its citizens). But, for purposes of the dormant Commerce Clause, an interest in achieving reliability in an equitable manner— that is, a manner that ensures everyone procures energy locally—is not. See Chem. Waste Mgmt., Inc. v. Hunt, 504 U.S. 334, 344 (1992) (“The burden is on the State to show that the discrimination is demonstrably justified by a valid factor unrelated to economic protectionism . . . .”) (cleaned up). After all, this formulation simply repackages a per se violation of the dormant Commerce Clause as an interest in discriminating against those retailers who do not procure their goods locally. Any other understanding would seemingly nullify all dormant Commerce Clause scrutiny. Oklahoma, for example, could have justified its ten percent coal law in Wyoming simply by arguing that utilities using out-of-state coal needed to contribute equitably to local coal usage. Where equity becomes a proxy for discrimination against interstate commerce, that purported local interest should not be part of the strict scrutiny analysis. Or. Waste Sys., Inc. v. Dep’t of Env’t Quality of State of Or., 511 U.S. 93, 106 (1994) Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 33 Mich. Public Serv. Comm’n, et al.

(rejecting similar cost-spreading rationale); see also Foresight Coal, 60 F.4th at 304 (recognizing that “level[ing] the playing field” is not a legitimate local concern).

With equity removed from the equation, the strict scrutiny inquiry becomes whether the state demonstrated that the ILCR is the only means of achieving its goal of securing a reliable energy supply. The district court, however, never reached that question. Instead, it looked at plaintiffs’ alternatives to the ILCR to see how they stacked up against the ILCR vis-à-vis the interests of the state. But the burden was on defendants to make this showing. See C & A Carbone, Inc., 511 U.S. at 392; Maine, 477 U.S. at 138. And more than simply demonstrating that the ILCR is desirable against the backdrop of the status quo or in the abstract, defendants need to prove that the ILCR is superior in achieving its goals relative to all other alternatives that do not expressly discriminate based on geography. Defendants might contend that allowing LSEs to procure electrical capacity from northern Indiana or Ohio (as opposed to similar capacity from the upper reaches of the lower peninsula) would fail to achieve the state’s interests. Yet that is no easy task. Remember, state laws that discriminate explicitly against interstate commerce are “almost always invalid.” Garber, 888 F.3d at 843. Either way, as the district court never engaged on the issue, we leave it to that court to resolve this narrow question in the first instance. See Taylor v. City of Saginaw, 11 F.4th 483, 489 (6th Cir. 2021) (recognizing that we are “a court of review, not first view”).

V.

We reverse the judgment of the district court and remand for proceedings consistent with this opinion. Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 34 Mich. Public Serv. Comm’n, et al.

_________________

DISSENT _________________

BOGGS, Circuit Judge, dissenting. This case lies at the intersection of two highly complicated fields: energy regulation on one axis and the Commerce Clause on the other. Although the majority offers an earnest attempt at navigating these difficulties, it ultimately reaches the wrong conclusion. In my view, this case clearly falls beyond the scope of the Commerce Clause and, under General Motors Corporation v. Tracy, 519 U.S. 278 (1997), is exempt from constitutional scrutiny on the basis that the Michigan Public Service Commission’s (MPSC’s)1 orders impermissibly discriminate against interstate commerce.

When “allegedly competing entities provide different products,” the dormant Commerce Clause applies only if “the companies are indeed similarly situated.” Id. at 299. The Commissioners argue that public utilities and alternative electric suppliers (AESs) serving retail customers in Michigan are not similarly situated entities, and thus do not meet the threshold requirement of a dormant Commerce Clause claim, because public utilities (1) serve residential customers; (2) have an obligation to serve such customers; and (3) are heavily regulated by the Commission in how they may earn a profit. By contrast, the record illustrates that “[t]ypical choice participants are large industrial manufacturers and mid-size commercial customers.” AESs, unlike public utilities, have no obligation to serve any customer and are unregulated when it comes to their rates. The district court, however, concluded that public utilities and AESs are similarly situated because AESs “provide the same commodity in the same markets” as other load-serving entities. The district court erred—public utilities and AESs are not similarly situated under the Court’s Commerce Clause jurisprudence.

The Commerce Clause’s “fundamental objective” is “preserving a national market for competition.” Ibid.; see Wyoming v. Oklahoma, 502 U.S. 437, 469 (1992) (Scalia, J., dissenting)

1 The Michigan legislature delegated regulatory authority to the MPSC to “regulate all rates, fares, fees, charges, services, rules, conditions of service, and all other matters pertaining to the formation, operation, or direction of public utilities.” Mich. Comp. Laws § 460.6(1). Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 35 Mich. Public Serv. Comm’n, et al.

(“Our negative Commerce Clause jurisprudence grew out of the notion that the Constitution implicitly established a national free market . . . .”). Prohibiting state regulation of different products that target distinct markets, with or without the allegedly discriminatory regulation, does not advance this fundamental objective. Tracy, 519 U.S. at 299. That is, the dormant Commerce Clause is implicated only when eliminating the state law at issue would improve competition in a market. Id. at 303–04.

Energy Michigan and ABATE argue that utilities and AESs are similarly situated because “[e]lectricity sold in the interstate wholesale market is comparable wherever a watt is generated.” “A watt is a watt, and electricity is fungible,” they argue. But this is the exact argument that the Court rejected in Tracy—there, the Court made clear that two entities that provide ostensibly the same end product are not necessarily similarly situated under the Commerce Clause.2 See id. at 302–03.

Tracy involved Ohio’s regulation of its retail natural-gas market. The state imposed general sales and use taxes on natural-gas purchasers from all sellers except regulated public utilities. Id. at 281–82. When Ohio partially deregulated its retail natural-gas market, “marketers”—essentially the natural-gas equivalent of Michigan’s AESs3—began to provide “unbundled” natural-gas service, typically to larger industrial customers. Id. at 284, 301–02. The regulated utilities continued to provide natural-gas service that was “bundled” with distribution service and state-mandated rights and obligations, which was usually the only viable natural-gas service for household and small retail customers. Id. at 297–98. Marketers, who tended to purchase out-of-state natural gas, were taxed more than utilities because the Ohio Supreme Court held that marketers did not qualify for the tax exemption given to regulated public utilities. Id. at 285. The Commerce Clause challenge followed.

2 This court rejected a somewhat similar argument in LensCrafters, Inc. v. Robinson, 403 F.3d 798 (6th Cir. 2005). There, this court found that retail optical stores and licensed optometrists are not similarly situated under Tracy “because they provide different services to the market.” Id. at 804. While both retail optical stores and licensed optometrists sell eyeglasses, “licensed optometrists are healthcare providers and, as such, have unique responsibilities and obligations to their patients that are not shared by optometric stores.” Ibid. 3 Independent natural-gas marketers purchase gas from producers, pay interstate pipelines for common- carriage services, and sell the gas to retail customers. Tracy, 519 U.S. at 284. Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 36 Mich. Public Serv. Comm’n, et al.

The Tracy Court determined that marketers and utilities were not similarly situated, even though both types of entities provided some type of natural-gas service to some customers directly. Id. at 310. Focusing on the Commerce Clause’s core objective of protecting competition in interstate markets, the Court first noted that the “bundled product” of the regulated utilities “reflects the demand of a market neither susceptible to competition by the interstate [marketers] nor likely to be served except by the regulated natural monopolies that have historically supplied its needs.” Id. at 303. In that situation, “the dormant Commerce Clause has no job to do.” Ibid. But there was a different market where utilities and marketers did compete: a “noncaptive” market of customers with sufficient natural-gas needs to justify trading the protections of state regulation for the lower prices of unbundled service. Id. at 302– 03. Nonetheless, the Court refused to treat utilities and marketers as similarly situated, even for only this small noncaptive market. Ibid.

Here, public utilities and AESs interact in fragmented markets like those discussed in Tracy. The public utilities here provide electricity service to a large captive market. That is, because Michigan law caps AESs’ market share at ten percent of the retail market,4 the remaining 90 percent is effectively captive to the utilities. See Mich. Comp. Laws § 460.10a(1). The only market in which AESs and utilities compete is the noncaptive market of the ten percent—effectively large industrial customers—that choose to purchase electricity from AESs. See ibid. Courts must consider the entire relationship between allegedly competing entities, rather than the competitive markets in isolation. See Tracy, 519 U.S. at 297–98. In these circumstances, the small noncaptive market is thus not enough to render the entities comparable under the Commerce Clause. That should be the end of this case.

Any doubts about this outcome should be assuaged after examining the policy considerations involved, which are like the policy considerations emphasized by the Tracy Court. See id. at 304–10. By way of background, the “individual local clearing requirement” (LCR) challenged here essentially prescribes the percentage of capacity that suppliers—load serving

4 Public Act 286 caps AESs’ market share by mandating that “no more than 10% of any utility’s average retail sales are supplied with electricity from an alternative electric supplier.” See In re Reliability Plans of Elec. Utilities for 2017-2021, 949 N.W.2d 73, 78 (Mich. 2020) (citing Mich. Comp. Laws § 460.10a(1)(a)). Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 37 Mich. Public Serv. Comm’n, et al.

entities (LSEs) including both utilities and AESs—must obtain from a specific geographic area to reduce blackout risk. This case’s dormant Commerce Clause challenge concerns the LCR’s application to Zone 7,5 a zone located entirely in Michigan’s lower peninsula but not completely coterminous with the peninsula.6

While the MPSC’s LCR differs in a few respects from the MISO rules, as outlined by the majority and court below, the LCR did not redefine Zone 7’s operative boundaries. The boundaries of Zone 7, like the boundaries of all of the Local Resource Zones (LRZs), are defined by MISO to “reflect the need for an adequate amount of Planning Resources to be in the appropriate physical locations within the MISO Region to reliably meet Demand and [Loss of Load Expectation] requirements.” MISO, Business Practices Manual No. 011: Resource Adequacy, at 79 (2023). More precisely, the geographic boundaries of each zone are set based upon analysis that considers: “(1) the electrical boundaries of Local Balancing Authorities; (2) state boundaries; (3) the relative strength of transmission interconnections between Local Balancing Authorities; (4) the results of previous [Loss of Load Expectation] studies; (5) the relative size of LRZs; and (6) market seams compatibility.” Ibid.

Bear in mind that MISO is not a Michigan agency or arm of the state in any sense. Rather, MISO is an independent “quasi-autonomous nongovernmental organization” regulated by the Federal Energy Regulatory Commission (FERC). See Regional Transmission Organization Backgrounders, Sustainable FERC Project, https://perma.cc/JER9-X7CH; see also Participation in Midcontinent Independent System Operator (MISO) Processes, An Introductory

5 MISO establishes a Planning Reserve Margin Requirement (PRMR) for each electricity provider. See Energy Michigan, Inc. v. Scripps, 658 F. Supp. 3d 511, 516 (E.D. Mich. 2023). “Under this planning requirement, electricity providers must ensure that a certain amount of electricity is available to meet its customers’ demands for the upcoming year. To meet part of that requirement, electricity providers also must demonstrate that they will generate enough capacity locally (the LCR).” Ibid. (internal citations omitted). The LCR “percentages were set on August 1, 2018, at 2.7% of the LSEs’ PRMR for 2022/2023 and 5.3% of the LSEs’ PRMR for 2023/2024.” These percentages affect the amount of electrical capacity that must be located in the zone. On September 13, 2018, the MPSC issued a stay of the June 18, 2018, order, which the commission has voluntarily continued during this litigation. 6 The southwest corner of Michigan is in PJM Interconnection territory, while Michigan’s upper peninsula is in MISO Zone 2. See In re the Investigation, on the Comm’n’s Own Motion, into the Elec. Supply Reliability Plans of Mich.’s Elec. Utils. for the Years 2017 Through 2021, No. U-18197, 2017 WL 4155229, at *5 n.5 (MPSC Sept. 15, 2017). Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 38 Mich. Public Serv. Comm’n, et al.

Guide to Participation in Midcontinent Independent System Operator (MISO) Processes, FERC, https://perma.cc/Q8UH-BJ2C.

While the majority gives a mild nod to “the public’s need for dependable energy,” it gives short shrift to that interest “under the banner of the dormant Commerce Clause.” United Haulers Ass’n, Inc. v. Oneida-Herkimer Solid Waste Mgmt. Auth., 550 U.S. 330, 347 (2007). “The size of the captive market, its noncompetitive character, [and] the values served by its traditional regulation” counsel against overriding the MPSC’s judgments and strongly suggest that eliminating the LCR will jeopardize utilities’ capacity to serve the large captive market. See Tracy, 519 U.S. at 307.

Thus, to some degree, this case involves a judgment about the proper trade-off between maintaining the reliability of household electric service and the prerogatives of a home-state regulator on the one hand and amplifying competition on the other.7 Should this court insert itself into a complex web of regulation in hopes of increasing competition within a sliver of the market? This court should err on the side of refusing to do so because the Commerce Clause “does not elevate free trade above all other values,” Maine v. Taylor, 477 U.S. 131, 151 (1986), and because this court is operating in a space that is traditionally the province of the states. See Arkansas Elec. Coop. Corp. v. Arkansas Pub. Serv. Comm’n, 461 U.S. 375, 377 (1983) (“[T]he regulation of utilities is one of the most important of the functions traditionally associated with the police power of the States.”); see also Metro. Life Ins. v. Massachusetts, 471 U.S. 724, 756, (1985) (“The States traditionally have had great latitude under their police powers to legislate as to the protection of the lives, limbs, health, comfort, and quiet of all persons.” (internal quotation marks and citation omitted)); United Haulers, 550 U.S. at 344 (“We should be particularly hesitant to interfere with the Counties’ efforts under the guise of the Commerce Clause because waste disposal is both typically and traditionally a local government function.” (internal quotation marks and citation omitted)); S. Pac. Co. v. Arizona ex rel. Sullivan, 325 U.S. 761, 770 (1945) (holding that states have a “wide scope for the regulation of matters of local state

7 Because customers of alternative suppliers tend to be large industrial purchasers, the captive 90 percent are mostly household customers for whom grid reliability is arguably most important. Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 39 Mich. Public Serv. Comm’n, et al.

concern, even though [these laws] in some measure affect[ ] commerce, provided [they] do[ ] not materially restrict the free flow of commerce across state lines, or interfere with it in matters with respect to which uniformity of regulation is of predominant national concern”).

Declining to give full weight to the judgment of state and local regulators on a matter of state and local concern is a fraught exercise, particularly considering the intricate area of energy regulation at play here and the small, non-captive market in which the majority hopes to amplify competition.8 Eliminating the local clearing requirement “might well intensify competition” in the noncaptive market, but “the importance of traditional regulated service to the captive market makes a powerful case against any judicial treatment that might jeopardize [the utilities’] continuing capacity to serve the captive market.” See Tracy, 519 U.S. at 302–04. That is, as suggested above and as found by the district court, eliminating the local clearing requirement would likely jeopardize the public utility’s ability to serve the market of household and small retail consumers that is not subject to competition from the AESs and whose customers are owed nothing from those suppliers. See Energy Michigan, 658 F. Supp. 3d at 533 (observing that Energy Michigan and ABATE “baldly contend that the defendants failed to demonstrate that the individual LCR provides a reliability benefit despite nearly every witness testifying to the contrary”).

Geographic proximity to generation improves grid reliability, and without the requirement to secure in-state capacity, Michigan would be at risk of falling short of federal reliability standards. As the district court noted, “[o]ne physical reality incorporated into planning decisions is that electrical energy degrades when it is transmitted over long distances due to energy losses that naturally occur over transmission facilities and ‘transmission constraints,’ terms that refer to the current-carrying capability of the facilities in the transmission

8 As Tracy reminds us, courts of review are “institutionally unsuited to gather the facts upon which economic predictions can be made, and professionally untrained to make them.” See Tracy, 519 U.S. at 308. Indeed, a critique of how Michigan chooses to organize its regulatory environment seems to underlie the majority opinion. See Maj. Op. at 24 (stating that “Michiganders are living with an artificially created market due to rent seeking by the utilities”); id. at 25 (casting the LCR as supporting a “generic interest in local energy reliability”). One must keep in mind, however, that “[t]he dormant Commerce Clause is not a roving license for federal courts to decide what activities are appropriate for state and local government to undertake, and what activities must be the province of private market competition.” United Haulers, 550 U.S. at 343. Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 40 Mich. Public Serv. Comm’n, et al.

system.” Id. at 516. Again, even with the emergence of AESs, Michigan utilities must serve the captive market defined by Mich. Comp. Laws § 460.10a(1), while AESs have neither the obligation nor ability to serve that market.9 Because AESs’ market share is capped by statute at ten percent of the retail market, they cannot pick up the slack if utilities are unable to provide reliable service to the remaining 90 percent of the market.

As suggested by the MPSC and as recognized by the court below, Zone 7 is in a precarious position because of its unique circumstances: Zone 7’s local capacity requirement is relatively high due to “the age and reliability of resources within the zone, the geographic nature of the zone (a peninsular state with limited interconnection), and the amount of available transmission import capacity.” See id. at 517, 525 (recounting testimony that Zone 7 has a particular need for resources to be located within the zone “due to constraints within the transmission system”). The district court found that “Zone 7 presently faces a loss-of-load expectation of nearly twice the excepted [sic] risk standard.” Id. at 536.

The majority relies heavily on Wyoming v. Oklahoma, 502 U.S. 437 (1992). While this court need not grapple with Wyoming because Tracy dictates the outcome of this case, Wyoming is distinguishable. There, Wyoming challenged an Oklahoma law that required Oklahoma coal- fired electric generating plants producing power for sale in Oklahoma to burn a mixture of coal containing at least 10% Oklahoma-mined coal. Id. at 440. Prior to Oklahoma’s enactment of the law, several Oklahoma utilities purchased nearly all of their coal from Wyoming sources. Id. at 444–45. The Court held that Oklahoma’s legislation violated the dormant Commerce Clause because the legislation “expressly reserve[d] a segment of the Oklahoma coal market for Oklahoma-mined coal, to the exclusion of coal mined in other States.” Id. at 455. The Court found that “[s]uch a preference for coal from domestic sources cannot be characterized as anything other than protectionist and discriminatory.” Ibid. Indeed, the protectionism of Oklahoma’s legislation was almost called out by name in the Oklahoma Legislature’s recitals and resolutions, which, among other things, emphasized that requiring plants to burn a blend of

9 Likewise, without an individual requirement to maintain locally generated capacity, “AESs lack any incentive to own local generation or add capacity in Zone 7.” See Energy Michigan, 658 F. Supp. 3d at 533 (internal quotation marks and citation omitted). Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 41 Mich. Public Serv. Comm’n, et al.

Oklahoma-mined coal “would assure at least a portion of the ratepayer dollars remain[] in Oklahoma and enhanc[e] the economy of the State of Oklahoma.” Id. at 443 (internal citation omitted).

Concerns about economic protectionism are at the core of the Court’s dormant Commerce Clause jurisprudence. See, e.g., Nat’l Pork Producers Council v. Ross, 598 U.S. 356, 369 (2023). While those concerns animated Wyoming, they are not present here. First, note that the Oklahoma Act at issue in Wyoming, on its face, required the purchase of coal based on its origin within the state of Oklahoma. 502 U.S. at 456. Here, while Zone 7 is, in fact, entirely within Michigan’s lower peninsula, it is not coterminous with the state (or the lower peninsula), and, more to the point, a multitude of factors beyond state borders were considered in MISO’s creation of Zone 7.10 As stated above, those factors included, among other things, “the relative strength of transmission interconnections between Local Balancing Authorities” and “the results of previous [Loss of Load Expectation] studies.” See Business Practices Manual No. 011, at 79. Insofar as Zone 7’s geographic location is in large part a product of technical judgments about grid reliability and the engineering reality that a reliable grid must have some amount of electricity generated close to where it is consumed, its connection to state borders is incidental

10 The majority finds that the MPSC’s orders facially discriminate on the basis that the orders “result in differential treatment between electricity generated in-state and that derived out-of-state.” The district court found that the orders do not facially discriminate because the orders do not distinguish between entities based on their in- state or out-of-state status; that is, the orders do not explicitly target out-of-state actors. See Energy Michigan, 658 F. Supp. 3d at 532 (“It is not evident, however, that local economic actors are favored by the individual LCR. To the contrary, witnesses offered extensive testimony at trial establishing that the requirement imposes the same burdens on utilities that it imposes on AESs.”). As the MPSC presses in its brief, and as credited by the district court, the LCR applies equally to in-state and out-of-state entities, as well as to both incumbent utilities and AESs. Tracy itself concerned facial discrimination. 519 U.S. at 291; see Camps Newfound/Owatonna, Inc. v. Town of Harrison, 520 U.S. 564, 582 n.16 (1997) (noting that Tracy “premised its holding that the statute at issue was not facially discriminatory on the view that [marketers and utilities] were principally competing in different markets”). But because Tracy’s threshold “similarly situated” determination dictates the outcome of this case, it is unnecessary to opine on whether the order here is indeed facially discriminatory. Tracy paints the “similarly situated” requirement using broad strokes. See 519 U.S. at 298–99 (“Conceptually, of course, any notion of discrimination assumes a comparison of substantially similar entities.” (emphasis added)). And the Supreme Court has repeatedly declined to distinguish under Tracy between facial and non-facial discrimination. See Dep’t of Revenue of Ky. v. Davis, 553 U.S. 328, 342–43 (2008) (describing Tracy’s substantially-similar-entity requirement as a “fundamental element of dormant Commerce Clause jurisprudence”); United Haulers, 550 U.S. at 342 (applying Tracy to uphold an ordinance that was not facially discriminatory). Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 42 Mich. Public Serv. Comm’n, et al.

and far afield from the ostensibly deliberate act of economic favoritism found in Oklahoma’s Act.

Second, and perhaps more importantly, the Court appreciated that nothing in Wyoming’s record could plausibly explain the Oklahoma Act’s preference for coal from domestic sources, aside from economic protectionism. The majority likens the MPCS’s power-grid-reliability argument to Oklahoma’s unsuccessful and “briefly” made argument that the regulation at issue in Wyoming was needed to lessen Oklahoma’s “reliance on a single source of coal delivered over a single rail line.” 502 U.S. at 456. However, there was nothing about Oklahoma coal qua Oklahoma coal that made the legislation necessary for energy reliability. Oklahoma was content with up to 90% of its coal coming from Wyoming, showing the single-source argument is best understood as a thinly veiled (and speculative)11 explanation for why the state felt it important to engage in economic protectionism; it aspired to strengthen its own coal industry by insulating it from competition.

And finally, electricity transmission is fundamentally different from the transportation of coal. Oklahoma could have purchased coal from other sources and stockpiled it. See id. at 457. Electricity cannot be stockpiled like coal can; electricity is instantaneous and grid reliability depends on the dispatchability of electricity—for instance, a state’s local control and accessibility to electricity sources are crucial on a hot day when a disproportionately high number of residents turn on their air conditioners at the same time. See In re Reliability Plans of Elec. Utilities, 949 N.W.2d at 77. MPCS’s reliability arguments are thus different in kind from Oklahoma’s passing mention of reliability concerns.12

11 In its reply brief, Oklahoma stated that its Act “prevents the State from becoming solely reliant on a fuel supply far removed from the State of Oklahoma. Certainly the State of Oklahoma does not have to wait for a coal strike or a rail strike before anticipating this obvious potential problem and insisting on the development of local coal supplies.” Reply Brief for the State of Oklahoma at 9, Wyoming v. Oklahoma, 502 U.S. 437 (1992) (No. 112), 1989 WL 1642575, at *9. 12 See Energy Michigan, 658 F. Supp. 3d at 533–34 (finding that “[t]he plaintiffs’ witnesses did not refute [the] reliability benefit [of the LCR]” and recounting one expert’s statement that “without the MPSC’s individual local clearing requirement, ‘over 1,400 MW of capacity could be excluded from the State’s long-term planning requirement of being sited within Michigan,’ translating to ‘over 300,000 customers in Michigan’ being served by resources ‘potentially located nowhere close to where it’s needed to be delivered”’). Nos. 23-1280/1323/1324 Energy Mich., Inc., et al. v. Page 43 Mich. Public Serv. Comm’n, et al.

At bottom, eliminating the local clearing requirement would do nothing to further the Commerce Clause’s “fundamental objective of preserving a national market for competition,” Tracy, 519 U.S. at 299, and it would undermine the reliability of the state’s grid. The majority of Michigan’s retail electricity market remains in the hands of the public utilities, who have an unshakable obligation to serve that vital market. The district court should have determined that, under Tracy, the local clearing requirement is not subject to review under the dormant Commerce Clause.13 Accordingly, I dissent.

13 To be sure, nothing in this opinion should be read as a suggestion that utilities are universally “immune from [ ] ordinary Commerce Clause jurisprudence.” Tracy, 519 U.S. at 291 n.8. Tracy rejected that contention, while still carving out the “similarly situated” requirement as a limiting principle. See id.