Luminant Energy Company LLC v. Public Utility Commission of Texas

TEXAS COURT OF APPEALS, THIRD DISTRICT, AT AUSTIN

NO. 03-21-00098-CV

Luminant Energy Company LLC, Appellant

v.

Public Utility Commission of Texas, Appellee

DIRECT APPEAL FROM THE PUBLIC UTILITY COMMISSION OF TEXAS
PROJECT NO. 51617

OPINION

In this direct appeal, we consider a challenge to the validity of a pair of related

Orders issued by the Public Utility Commission (PUC, Commission) on February 15 and 16 of

2021, respectively, governing scarcity pricing in the wholesale electricity market during Winter

Storm Uri. See Order Directing ERCOT to Take Action and Granting Exception to Commission

Rules, PUC Project No. 51617 (Feb. 15, 2021); Second Order Directing ERCOT to Take Action

and Granting Exception to Commission Rules, PUC Project No. 51617 (Feb. 16, 2021). Appellant

Luminant Energy Company LLC (Luminant) and aligned intervenors 1 (collectively, Appellants)

1 Appellant-Intervenors are Constellation NewEnergy, Inc.; Exelon Generation Company,
LLC; Logan’s Gap Wind LLC; Pattern Energy Group LP; Pattern Gulf Wind, LLC; Pattern
Panhandle Wind, LLC; Pattern Panhandle Wind 2 LLC; RWE Renewables Americas LLC; Texpo
Power LP; and TX Hereford Wind, LLC. Appellee-Intervenors are Calpine Corporation; Talen
Energy Corporation; and TexGen Power, LLC. Intervenor DGSP2 LLC aligns with different
parties on different issues.
contend that the subject Orders (1) constitute de facto competition rules under Chapter 39 of the

Texas Utilities Code, (2) were adopted in violation of the rulemaking provisions of the

Administrative Procedure Act (APA), and (3) exceed the Commission’s statutory authority. We

agree with Appellants’ first and third points and therefore do not reach the second. We reverse the

Commission’s Orders and remand for further proceedings consistent with our ruling.

I.
BACKGROUND

The generation and sale of electric power is subject to a number of unique physical

and engineering constraints that make it unlike other goods and services. Although battery

technology is constantly improving, electricity remains difficult to store at scale, meaning that

most electricity generation must occur concurrently with consumption. See, e.g., TXU Generation

Co., v. Public Util. Comm’n of Tex., 165 S.W.3d 821, 827 (Tex. App.—Austin 2005, pet. denied)

(discussing background of electricity market in Texas). Moreover, for technical reasons,

generation and consumption must at all times be maintained in near-perfect balance at an

equilibrium point of 60 Hertz, or else a total grid collapse, together with serious damage to grid

equipment, could result. Id. at 828. Moreover, constraints on transmission can result in grid

congestion, and power generated in one geographic region may not be available to consumers in

another. Id. at 827.

In part for the foregoing reasons, electric power in Texas as elsewhere was

historically thought of as a “natural monopoly,” with economies of scope and scale that rendered

it relatively insusceptible to efficient delivery through market competition. Id. (citing Reliant

Energy, Inc. v. Public Util. Comm’n of Tex., 101 S.W.3d 129, 133 (Tex. App.—Austin 2003),

rev’d in part by CenterPoint Energy, Inc. v. Public Util. Comm’n of Tex., 143 S.W.3d 81

2
(Tex. 2004)). Through much of the Twentieth Century, electric utilities were authorized by law

to operate in tightly regulated, vertically integrated monopolies throughout the state. Monopoly

power meant that, in any given geographical area, only one electric utility was authorized to

provide electricity to retail customers. Vertical integration meant that, for any given customer, a

single entity controlled all three of the principal components of electricity delivery: “generation

of power; transmission of that power on high-voltage lines over long distances; and distribution of

power over shorter distances to the ultimate consumer.” TXU Generation, 165 S.W.3d at 827

(quoting City Pub. Serv. Bd. of San Antonio v. Public Util. Comm’n of Tex., 9 S.W.3d 868, 870

(Tex. App.—Austin 2000), aff’d, 53 S.W.3d 310, 312 (Tex. 2001)). Tight regulation meant that,

for any utility, the rates it could charge for electricity were set by the Commission.

The regulated monopoly model provided reliable electric power at a price to the

consumer that actually fell year over year until the 1970s. Beginning in or around 1973, however,

the OPEC oil embargo and the resultant energy crisis—which came at a time when a substantial

portion of electricity generation in the U.S. was fueled by petroleum—produced dramatic price

increases and gave rise to political pressure from consumers for some kind of reform in order to

control costs. See Seth Blumsack, Measuring the Benefits and Costs of Regional Electric Grid

Integration, 28 Energy L. J. 147, 149 (2007). In response, Congress and several states enacted

measures to encourage competition in the electric power market through deregulation

and restructuring. See, e.g., Public Utility Regulatory Policies Act of 1978, Pub. L. 95–617,

92 Stat. 3117 (allowing electricity production by independent producers in federally regulated

markets). That restructuring has taken a variety of forms in different regions, but essential

components typically include (1) the vertical disintegration of electric utilities into separate

generation, transmission, and distribution utility entities; (2) retail competition at the generation

3
level, with customers able to choose from among competing providers selling at rates based on

market prices rather than regulated costs; (3) the use of spot markets for energy and, in some cases,

ancillary services such as capacity and reserves; and (4) the management of the transmission and

distribution grid on a regional scale by an independent entity rather than the transmission owners.

Blumsack, supra at 152.

The latter feature was necessary to ensure that competing generators received

nondiscriminatory access to transmission infrastructure in order to reach their customers. Put

another way, for generation to be deregulated and open to competition, it was necessary for

transmission to remain tightly regulated and transparent, lest transmission owners grant

preferential access to favored generators. Toward that end, the Federal Energy Regulatory

Commission (FERC) in 1996 issued Orders 888 and 889 to require transmission utilities to give

generators nondiscriminatory access under an open-access tariff and to publish real-time available

capacity. See Order No. 888, Promoting Wholesale Competition Through Open Access

Nondiscriminatory Transmission Services by Public Utilities; Recovery of Stranded Costs by

Public Utilities and Transmitting Utilities, 61 Fed. Reg. 705-02 (1996) (codified at 18 C.F.R.

pt. 35); Order No. 889, Open Access Same-Time Information System (formerly Real-Time

Information Networks) and Standards of Conduct, 61 Fed. Reg. 21,737-01 (1996) (codified at

18 C.F.R. pt. 37). These orders, together with Order 2000, introduced the closely related concepts

of Independent System Operator (ISO) and Regional Transmission Organization (RTO).

Order No. 2000, Regional Transmission Organizations, 65 Fed. Reg. 45,854-01 (2000) (codified

at 18 C.F.R. pt. 35). Though distinct in terms of governance structure and congestion

protocols, ISOs and RTOs are sufficiently similar in function that the terms are often used

interchangeably. Blumsack, supra at 147 n. 1. Either designation, RTO or ISO, refers to the

4
independent, not-for-profit organization that manages the joint transmission assets of a number of

regionally interconnected transmission utilities in a deregulated region. Such entities have no

assets of their own and take no position in the wholesale market, but are responsible for managing

bulk transmission, ensuring equal access to the grid by customers and suppliers, dispatching

reserve generation as needed to balance supply and demand, and often making a market for

wholesale electric generation and ancillary services. Id. at 147. Today, seven such regional

entities are operating in the United States, covering approximately half of the states and two-thirds

of total annual demand.

Texas’s foray into deregulation began in 1999, when the Legislature enacted Senate

Bill 7, amending the Public Utility Regulatory Act (PURA) to deregulate Texas’ electricity market

and require that wholesale and retail rates generally be set by competition rather than regulation.

See Act of May 27, 1999, 76th Leg., R.S., ch. 405, 1999 Tex. Gen. Laws 2543, 2543–2625

(codified at Tex. Util. Code §§ 39.001–.910.) Texas is distinct in that, by sheer size, its grid is one

of the nation’s three major regional interconnections, and yet, because it is entirely within the state,

it is not subject to FERC regulation. Nevertheless, deregulation in Texas has largely conformed

to the pattern seen elsewhere in the country. With SB 7, the Legislature required monopolies in

Texas to unbundle their generation, transmission, and distribution units and form separate entities.

See Tex. Util. Code § 39.051(b). The newly deregulated market was placed under the regulatory

jurisdiction of the PUC which, in turn, certified the Electric Reliability Council of Texas (ERCOT)

as the ISO for a sprawling, self-contained grid that covers approximately 70 percent of the state

geographically and serves approximately 90 percent of the state’s electricity customers. Texas

Com. Energy v. TXU Energy, Inc., No. CA No. C-03-249, 2004 WL 1777597, at *1–2 (S.D. Tex.

June 24, 2004), aff’d, 413 F.3d 503 (5th Cir. 2005).

5
 As the ISO, ERCOT is responsible for ensuring (inter alia) system reliability,

nondiscriminatory access to transmission and distribution systems, and clearance of all market

transactions in the region served by the grid. Importantly, ERCOT acts as the central counterparty

for all transactions it settles and is deemed to be the sole buyer to each seller (typically a generator),

and the sole seller to each buyer (typically a retailer) of all energy and related services. ERCOT

Nodal Protocols § 1.2(4). From a central control room that has been likened to an air traffic control

center, ERCOT monitors the flow of energy from more than 680 generation units to more than

26 million customers over more than 46,500 miles of transmission lines, all while balancing

generation with load to maintain a system frequency of 60 Hertz. Extreme weather condition

preparedness/power outages: Hearing Before the S. Comm. On Bus. & Com. (Senate B&C Hr’g),

87th Leg., R.S. (Tex. 2021) (testimony of Bill Magness, President and CEO, ERCOT); see also

Bill Magness, Review of February 2021 Extreme Cold Weather Event—ERCOT Presentation at 4

(Feb. 24, 2021). In addition to PURA and the substantive regulations adopted by the PUC pursuant

to its rulemaking authority, ERCOT’s board is also authorized to adopt rules or so-called

“protocols” that provide a framework for the administration of the electricity market, subject to

Commission oversight and review. BP Chemicals, Inc. v. AEP Tex. Cent. Co., 198 S.W.3d 449,

452 (Tex. App.—Corpus Christi–Edinburg 2006, no pet.); Tex. Util. Code § 39.151(d), (g).

Scarcity pricing

By statute, the PUC is required to “adopt and enforce rules relating to the reliability

of the regional electrical network” or “delegate to an independent organization responsibilities for

adopting or enforcing such rules.” Tex. Util. Code § 39.151(d). In 2006, pursuant to that statutory

directive, the Commission by rule established the general outline of a scarcity pricing mechanism

6
(SPM) for use during periods of high demand. 16 Tex. Admin. Code § 25.505(g) (effective

May 30, 2019, to July 13, 2021). The rule required that ERCOT “use a stakeholder process to

develop and implement rules” to add detail and specificity to the general mechanism.

Id. § 25.505(h) (effective May 30, 2019, to July 13, 2021). Pursuant to an earlier version of that

mandate, ERCOT in 2014 revised its protocols to include a complex mathematical formula and to

enumerate several variables to determine when and how scarcity pricing takes effect. See

ERCOT Nodal Protocols § 6.5.7.3.1, Determination of Real Time On-Line Reliability Deployment

Price Adder.

Scarcity pricing is limited by rule to a system-wide offer cap (frequently

abbreviated “HCAP”) of $9,000 per megawatt hour (MWh). 16 Tex. Admin. Code

§ 25.505(g)(6)(B) (effective May 30, 2019, to July 13, 2021). To put that figure in perspective, a

typical market clearing price for electricity in Texas can be as little as $30/MWh. The reason for

the orders-of-magnitude delta has to do with the way the state’s wholesale market is structured.

Texas is what is known as an “energy-only” market, meaning that generators are

compensated only for energy actually sold, as contrasted with a so-called “capacity” market, in

which generators are compensated for making capacity available even if it is not ultimately used.

A “base-load” generation utility (that is, one intended to operate continuously for long periods

under normal market conditions) may incur significant initial costs investing in capacity, while

expecting to recoup those costs through daily operations by offering at or below the market

clearing price on most days. A so-called “peaker” generator, however, may invest a comparable

amount of capital in generating capacity that it expects to use only during a few peak hours on

summer afternoons. To defray the high costs of developing and maintaining such seldom-used

capacity, it is necessary for such generators to receive a very high rate of return per megawatt hour

7
on the energy they do sell. Accordingly, the scarcity pricing mechanism is intended to accomplish

two goals. First, by offering windfall prices to peak generators through a number of scarcity price

“adders,” it creates incentives for any idle generation capacity to come online when demand (or

“load”) threatens to exceed supply. Second, by imposing sticker-shock costs on consumption, it

encourages conservation by any institutional consumers that may have some elasticity of demand,

thus hopefully easing system load.

Winter Storm Uri

Texas typically experiences peak electricity demand during its long hot summers

as consumers rely on electric power to run their air conditioners. In 2021, however, the winter

weather event that has come to be known colloquially as Winter Storm Uri subjected the state’s

electrical grid to historically unprecedented stress.

The “perfect storm” of factors that would ultimately coalesce to produce Uri first

came to meteorologists’ attention in late December and early January, with a rapid rise in

temperatures by as much as 90 degrees in the Earth’s stratosphere over the North Pole. Senate

B&C Hr’g, supra at 08:28–54:30 (testimony of Bob Rose, Meteorologist, Lower Colorado River

Authority). Known simply as a “sudden stratospheric warming event,” the phenomenon is not

terribly unusual, occurring on average approximately once in four years, but, when combined with

other factors, it can produce disruptions in the so-called “polar vortex,” allowing super-chilled

arctic air to flow south and at ground-level altitudes. Id. By late January, a pair of high-pressure

ridges had begun to form near Greenland and the Gulf of Alaska, creating conditions favorable to

a pronounced disruption of the jet stream that would allow bitterly cold Siberian air to spill over

the pole and down across North America as far south as Texas. Id.

8
 The looming emergency did not go unnoticed by ERCOT. As early as November

of 2020, the entity’s meteorologist issued a winter outlook for market participants warning of “very

good” chances for an extreme cold weather event during the 2020-21 winter season. Magness,

supra at 9. On February 3, 2021, ERCOT warned market participants and the public of the coldest

weather of the year, with additional advisories, operating condition notices, and press releases

issued on February 10, 11, 13, and 14. Id. In anticipation of possible weather-related outages,

pre-event precautions included canceling planned maintenance outages for more than 1,600

transmission devices, waiving COVID restrictions, activating additional on-site and remote

engineering and support staff, and requesting emissions waivers from federal and state regulators.

Id. at 8.

The Arctic air first began to arrive in Texas on Friday, February 12, 2021, with up

to half-inch accumulations of ice observed on some surfaces. Senate B&C Hr’g at 9:20 (Rose

testimony). Temperatures remained low into the weekend, hovering in the 20s and 30s, but dipped

significantly on Sunday, February 14, as Siberian air flooded into Texas and plunged temperatures

into the teens. Id. On Monday, February 15, the statewide average temperature in Texas was

11 degrees Fahrenheit. Id. Several cities posted record or near-record lows, with the Dallas–Fort

Worth area logging its third lowest temperature ever observed at 2 degrees below zero, Austin

logging its fifth lowest at 7 degrees, and Corpus Christi logging its all-time lowest observed

temperature at 17 degrees. Id. Moreover, the duration of the freeze shattered records, with cities

across north Texas experiencing more than 200 consecutive hours of temperatures below freezing,

and parts of central Texas seeing more than 160 hours. Id. Temperatures in Austin remained

below freezing for 144 consecutive hours, the longest period on record. Id. Unlike comparable

cold-weather events in 1989 and 2011, the extreme cold combined with heavy precipitation to

9
produce record or near-record snowfall in parts of the state, with Dallas receiving 4–5 inches of

accumulation over the weekend, Midland 5–6 inches, Austin between 6 and 8 inches (its fourth

largest two-day total on record), and Del Rio nearly a foot of snow, an all-time record. Id. For

temperatures alone, Uri entered the history books among the four or five coldest weather events

since reliable recordkeeping began in 1895. Id. When duration and precipitation are taken into

account, Uri may have been the most severe winter weather event in the recorded history of Texas.

At 8:30 a.m. on the morning of Sunday, February 14, 2021, ERCOT issued a

conservation alert and media appeal. Magness, supra at 11. By mid-morning, it was already

becoming clear to ERCOT’s leadership that offers for so-called “ancillary services,” i.e., additional

generation necessary to make up a shortfall in scheduled base generation relative to demand, would

fall short of the anticipated load, and a watch to this effect was issued at 10:54 a.m. that morning.

Id. Based on projected demand, it was anticipated that ERCOT would probably need to order

transmission entities to “shed load” in order to maintain the 60-Hertz balance with generation, but

it was expected that these measures would take the form of brief rotating outages that would not

leave any customer without power for longer than a few hours. Senate B&C Hr’g 1:06:27

(Magness testimony). Despite the fact that demand was projected to break an all-time record,

exceeding even the highest summer demand on record, it appeared to ERCOT personnel as though

the system had enough generation online to get through the ensuing days with minimal disruption

to service. Id. As night fell, however, all of that began to change.

At 9:58 p.m., ERCOT issued a reserve capacity watch. Magness, supra at 11.

At 11:17 p.m., ERCOT deployed responsive reserve. Id. By 11:32 p.m., ERCOT issued another

advisory, announcing a physical response reserve of less than 3,000 megawatts remaining. Id.

By 12:15 a.m. on the morning of the 15th, reserves had fallen below 2,300 megawatts, sending

10
ERCOT into Energy Emergency Alert (EEA) Level 1. Id. By 1:07 a.m., reserves had fallen below

1,750, triggering EEA2 and prompting deployment of load resources. Id. At 1:20 a.m., ERCOT

entered EEA3—its highest alert level—and began ordering that load be shed. Id.

As the storm continued to move southward through Texas in the early morning

hours of the 15th, it froze poorly winterized natural gas wellheads and gathering lines, nearly

halving the supply of the state’s predominant fuel for electric generation at a time when the gas

was already in high demand for direct-burn home heating in many areas. As more and more

generation units went offline for want of fuel, the plummeting electricity supply rapidly

overwhelmed ERCOT’s ability to manage scarcity and began to threaten the stability of the grid

itself. As of 1:23 a.m., a combined capacity of 35,343 megawatts of generation was reported

offline due to weather-related causes, and the system frequency fell to 59.9 Hertz, prompting

ERCOT to order 1,000 megawatts of load shed to stabilize it. Id. at 12. Just a few minutes later,

however, at 1:26 a.m., three additional outages of 1,481 megawatts, 248 megawatts, and

329 megawatts occurred in rapid succession, causing the system frequency to fall to under

59.8 Hertz by 1:43 a.m. Id. This prompted another 1,000 megawatts of load shed, but the grid

continued hemorrhaging generation, with outages of 606 megawatts, 688 megawatts, and

511 megawatts, respectively, between 1:48 a.m. and 1:53 a.m. Id.

These additional failures drove the grid past a critical threshold, as the system

frequency dipped below 59.4 Hertz. Id.; see also Senate B&C Hr’g at 1:16:30 (Magness

testimony). At that level there existed an imminent risk of oscillations that could damage grid

components, perhaps permanently. As a rule, the system could remain below 59.4 for no more

than nine minutes. The grid frequency that night was below 59.4 for a total of four minutes and

three seconds, during which time two more rounds of load shed were ordered, of 3,000 megawatts

11
and 3,500 megawatts, respectively. Magness, supra at 12. The frequency bottomed out at

59.302 Hertz before beginning to recover. Id. By 2:03 a.m. it reached 60 Hertz again, but not

before additional generation outages of 843 megawatts and 841 megawatts forced another 2,000

megawatts of load shed, bringing the total load shed that night to a staggering 10,500 megawatts.

The storm, at its worst, had taken out 48.6 percent of the generation available to

ERCOT to manage the system. Id. at 13; see also Senate B&C Hr’g at 1:01:103 (Magness

testimony). Some 4.5 million households and businesses lost power and, because of the volume

of load shed, these outages could not easily be rotated without threatening critical infrastructure

(e.g., hospitals). For that reason, many of the homes affected were left without power for days at

a time during some of the coldest days ever recorded in the state’s history. Bad as this outcome

was, ERCOT’s then-CEO later testified that it could have been much worse. The system had come

perilously near collapse and, had it been allowed to fail, the result would have been a weeks-long,

statewide blackout. The worst potential outcome had been averted.

Also closely monitoring the situation that weekend were the then-current members

of the Commission, all of whom have since resigned. During the course of the foregoing events,

the previously established SPM (scarcity pricing mechanism) remained in effect but resulted in

market clearing prices of just over $1,200/MWh. Hearing to examine billing errors resulting from

Winter Storm Uri and explore the means available for correcting those errors: Hearing Before

the S. Comm. on Juris., 87th Leg., R.S. (Tex. 2021) at 43:42—52 (Mar. 11, 2021) (testimony of

Independent Market Monitor (IMM) director Carrie Bivens). In consultation with ERCOT

personnel, then-PUC Chair DeAnn Walker concluded that the SPM had malfunctioned. Senate

B&C Hr’g at 6:13:29 (testimony of DeAnn Walker, Chair, PUC). Chair Walker reasoned that the

SPM was intended to move prices in an inverse correlation with reserve capacity and that, at load

12
shed, maximum demand had by definition been reached, such that the maximum offer price should

be in effect as well. Id. Moreover, after speaking with ERCOT’s leadership, she had a working

hypothesis to explain exactly why the SPM had failed. Specifically, her understanding was that,

as the grid shed load pursuant to ERCOT’s emergency orders, the SMP algorithm erroneously

disregarded the lost load altogether for purposes of computing price.

The significance of this inference should not be understated. Clearing prices act as

price signals, letting market participants know when additional generation is needed. According

to Chair Walker’s interpretation, ERCOT’s computer system was, in effect, sending the market a

false signal—signaling that additional generation was not needed, or at least not needed as urgently

as was actually the case—and thereby failing to adjust incentives to answer the exigency presented

by the storm.

On the afternoon of February 15, 2021, the Commission held a brief open meeting

in which the commissioners discussed the SPM and issued the first of the two Orders that are the

subject of this litigation. Finding that the electricity market was clearing as low as $1,200 and that

this outcome was “inconsistent with the fundamental design of ERCOT,” the Commission directed

ERCOT “to ensure that firm load that is being shed in EEA3 is accounted for in ERCOT’s scarcity

pricing signals.” Order Directing ERCOT to Take Action and Granting Exception to Commission

Rules, PUC Project No. 51617 at 1-2 (Feb. 15, 2021) (First Order). The First Order stated, “If

customer load is being shed, scarcity is at its maximum, and the market price for the energy needed

to serve that load should also be at its highest.” Id. at 1. By its terms, the First Order applied only

to the then-in-effect EEA event.

Later that same day, ERCOT issued a market notice reflecting its understanding

that the PUC’s “directive [was] based on the [PUC’s] observation in the order that energy prices

13
of less than $9,000/MWh during load shed conditions are ‘inconsistent with the fundamental

design of the ERCOT market.’” ERCOT, M-C021521-01 Emergency Order of the Public Utility

Commission Affecting ERCOT Market Prices (Feb. 15, 2021). ERCOT then adjusted its price

algorithm to cause the market to clear at the $9,000/MWh cap.

On the afternoon of the 16th, the Commission held a second meeting to revisit the

First Order, ultimately issuing a second order that was substantially identical to the first, except

that it rescinded language in the First Order that would have required retroactive repricing. Second

Order Directing ERCOT to Take Action and Granting Exception to Commission Rules, PUC

Project No. 51617 (Feb. 16, 2021) (Second Order). The Second Order again directed ERCOT that,

if there was load shed, the market price for the energy needed to serve that load should also be “at

its highest.” Id. at 1. ERCOT has since issued settlement statements to market participants

reflecting the $9,000/MWh clearing price.

II.
PROCEDURAL HISTORY

Appellant Luminant, a market participant subject to ERCOT protocols, filed its

notice of direct appeal in this Court on March 2, 2021, challenging the first and second Orders as

competition rules under Section 39.001(e) of the Utilities Code. Several parties intervened on

either side. Generally, Appellants proceed on the theory that the subject Orders were procedurally

invalid, beyond the PUC’s statutory authority to adopt, or both. Luminant has also filed

administrative challenges to settlement invoices issued by ERCOT for transactions during the

storm, asserting that, because the subject Orders are invalid, the price of electricity in the settlement

invoices is also incorrect. See Luminant v. Public Util. Comm’n, No. 03-21-00126-CV (Tex.

App.—Austin).

14
 In their responses, the Commission and its aligned intervenors argue that (1) the

subject Orders are not rules; (2) if they are rules, they were adopted in substantial compliance with

emergency rulemaking provisions of the APA; (3) the PUC was expressly authorized by statute to

correct the allegedly erroneous pricing by the SPM; and (4) for a variety of reasons, this Court

lacks subject-matter jurisdiction. As jurisdiction is a threshold issue, we address it first, before

turning to the merits.

III.
DISCUSSION

Subject-Matter Jurisdiction.

The PUC challenges this Court’s subject-matter jurisdiction in five separate but

sometimes overlapping arguments. Each is addressed in turn.

Appellants’ Claims are Not Moot.

The Commission argues first that, because the Orders by their own terms applied

only during the then-in-effect EEA3 event and consequent load shed, the Orders have now expired,

and any controversy between the parties is now moot. We disagree.

Mootness doctrine generally requires that a case or controversy exist at all stages

of a proceeding and that, if events resolve the dispute, the case should be dismissed as moot.

See, e.g., Matthews v. Kountze Indep. Sch. Dist., 484 S.W.3d 416, 418 (Tex. 2016). Events that

may render a case moot include the death of a party, the settlement of the claim, the repeal of a

challenged statute—essentially any change in the facts that ends the controversy. See

Electric Reliability Council of Tex., Inc. v. Panda Power Generation Infrastructure Fund, LLC,

619 S.W.3d 628, 634–35 (Tex. 2021). The doctrine has a constitutional dimension, in that it

15
prevents courts from rendering advisory opinions. Wilson v. Community Health Choice Tex., Inc.,

607 S.W.3d 843, 849 (Tex. App.—Austin 2020, pet. denied) (citing Matthews, 484 S.W.3d at 418).

Here, the controverted issue is the lawfulness of the Orders that purported to price

electricity at its statutory cap of $9,000/MWh. If the Orders are held to have been lawful, so too

(barring other grounds for challenge) will the invoices or settlement statements issued during the

period in which the Orders were putatively in effect. If the Orders are held unlawful, Appellants

(barring other factors not considered here) may succeed in their administrative challenges and

secure adjustments to the invoiced amounts. The live controversy, then, is whether Appellants are

presently entitled to the difference between the invoiced prices and the much lower clearing prices

that they argue would have been charged in the absence of the allegedly invalid Orders. To grant

or deny such material relief would not be to issue an advisory opinion.

The Commission cites a number of cases for the proposition that, generally, “where

an order expires by its own terms … all issues relating to the validity of the order become moot.”

Rodriquez v. Texas Dep’t of Pub. Safety, 533 S.W.2d 849, 851 (Tex. App.—Tyler 1976, no writ).

Upon review, however, we find the cited authorities consistent with our understanding. In

Rodriquez, for example, the plaintiff appealed from a trial-court judgment affirming a 12-month

administrative suspension of his certificate of appointment as a motor vehicle inspector and of his

garage’s license as an official inspection station. Id. at 850. By the time the case reached the Tyler

court, however, the suspension had expired. Id. On those facts, the court held the case to be moot.

Id. at 850–51. In so holding, the court noted that “courts have sometimes entertained attacks on

orders that have expired by their own terms because the person against whom such order was made

might suffer some subsequent detriment if the legality of the order were not determined.” Id. at

851. Rodriquez, however, sought “no relief other than reversal of the order of suspension”—he

16
did not, for example, claim to be entitled to monetary damages resulting from the suspension. Id.

at 851. As such, an order of reversal could have had no prospective effect on any party’s rights or

duties. Here, in contrast, Appellants are separately disputing the settlement statements through

ERCOT’s administrative ADR process, alleging unlawful overpricing of thousands of dollars per

megawatt hour; our decision in this appeal may have very real material consequences for all

involved. The “subsequent detriment” Appellants will experience if the Orders are allowed to

stand will be the obligation to honor the allegedly unlawful price.

In Texas Department of Agriculture v. Wild Boar Meats, LLC, the Department of

Agriculture adopted an emergency rule to allow the temporary registration of a certain chemical

pesticide for use in feral-hog abatement. No. 03-17-00514-CV, 2018 WL 3748677, at *1 (Tex.

App.—Austin Aug. 8, 2018, pet. denied) (mem. op.). Wild Boar Meats, a company that purchased

feral hogs and hog carcasses from local hunters for use in food products, believed the pesticide

presented safety risks to humans and sued the Department challenging the validity of the rule. Id.

While the case was pending, the sole manufacturer to have requested registration under the rule

asked the Department to cancel its request citing negative publicity, and the Department notified

the Secretary of State that it was withdrawing the rule. Id. at *2. In reversing the trial court’s

denial of the Department’s motion to dismiss on mootness grounds, this Court noted that the rule

was withdrawn several months before the hearing on the motion to dismiss and that nothing in the

record suggested any pesticide had ever been used under the rule, concluding that “any declaration

regarding its validity would have no legal effect” and that the claim was therefore moot. Id. at *3.

The present case is distinguishable. Whereas the declaration sought in Wild Boar would not have

halted any actual or threatened use of the pesticide under the former rule, Appellants in this case

seek to avoid the financial burden they will experience if required to honor the allegedly inflated

17
electricity pricing in their pending administrative challenges to the allegedly invalid rule at

issue here.

In Railroad Commission of Texas v. Oil Production Maintenance, Inc., the owners

and operators of a gas well sued the Railroad Commission to set aside and enjoin enforcement of

an order placing the well under a six-month restriction limiting the daily volume of gas that could

be produced from it. 319 S.W.2d 822, 823 (Tex. App.—Austin 1958 writ ref’d n.r.e.). The trial

court held in favor of the plaintiffs and granted the injunction but, by the time the case came before

this Court on appeal, the six-month interval had run, and the order had expired of its own terms.

Id. Accordingly, we set aside the judgment below and dismissed the case as moot. Id. at 824. As

in the other cases cited by the Commission—and unlike the present case—the plaintiff sought no

greater relief than what had already been brought about by the expiration of the challenged rule.

The same distinction applies to the cases Calpine cites in its brief. In State v. City

of Austin, the State sought to enjoin local orders intended to limit the spread of COVID-19 by

suspending restaurants’ dine-in food and beverage service each night from 10:30 p.m. to 6:30 a.m.

over the holiday weekend beginning December 31, 2020, and ending January 3, 2021.

No. 03-20-00619-CV, 2021 WL 1313349, at *1 (Tex. App.—Austin Apr. 8, 2021, no pet.) (mem.

op.). The orders were in force for a total of 7.5 hours before being temporarily enjoined and then

expiring by their own terms. Id. at *4. Despite the expiration, the State nonetheless proceeded

with its suit. In dismissing, this Court noted that “nothing in the parties’ briefing or the record

indicates that the local officials actually enforced the local orders during the single 7.5-hour period

that they were in effect.” Id. at *5. We held that “[a]bsent any enforcement efforts,” the State’s

18
requested relief would have had no legal effect. Id. at *5 (emphasis added). 2 Impliedly, the

outcome may have been different had the plaintiff been a business owner who sought relief from

a fine issued under the complained-of order. Cf. Lowenberg, v. City of Dallas, 261 S.W.3d 54, 59

(Tex. 2008) (suit for refund of allegedly unlawful municipal fire registration on ground that it was

improper occupation tax not mooted by voluntary payment or subsequent repeal of ordinance).

Perhaps the most factually similar case the Commission cites is the second of two

cases styled El Paso County Hospital District v. Texas Health & Human Services Commission,

cited for the proposition that, “[e]ven if ERCOT undertook to reconstruct alternative pricing, those

calculations would not make previously issued statements for EEA3-operating days any less

valid.” See 400 S.W.3d 72, 82 (Tex. 2013) (El Paso II). In El Paso II, a number of hospitals

alleged that the Texas Health and Human Services Commission (HHSC) had used an invalid rule

to calculate reimbursement rates for Medicaid services, resulting in alleged underpayments to the

hospitals dating back years. Moreover, the hospitals in that case sought not only the prospective

invalidation of the rule but the retroactive recalculation and recovery of past payments. The

Commission’s reliance on El Paso II is problematic, however, because, in that case, the Texas

Supreme Court did not reach the retroactivity issue for which the Commission now cites it. Id.

at 83 (referring to El Paso Hosp. Dist. v. Texas Health & Human Servs. Comm’n, 247 S.W.3d 709

(Tex. 2008) (El Paso I) and explaining that “[w]e did not decide [in El Paso I] whether the

2 See also Methodist Hosps. of Dallas v. Texas Workers’ Compensation Comm’n,
874 S.W.2d 144, 147 (Tex. App.—Austin 1994, no writ) (dismissing as moot challenge to expired
rule when no other relief sought and no suggestion that ruling could result in the recovery of
previously incurred costs or relief from previously assessed but unlawful fees or expenses); Young
Trucking, Inc. v. Railroad Commission of Texas 781 S.W.2d 719, 720 (Tex. App.—Austin 1989,
no writ) (dismissing as moot challenge to lapsed one-year administrative suspension of the
plaintiff’s specialized motor carrier certificate when no prospective relief sought).
19
hospitals could reopen past agency proceedings or obtain relief for past years”). Moreover, much

of the discussion of the availability of retroactive relief in that case concerned a close reading of

administrative rules specifically applicable to HHSC and not relevant here. Finally, the relief

sought here is not impermissibly retroactive because Appellants’ administrative challenges remain

pending. Thus, nothing in El Paso II requires us to deem the previously issued Orders for EEA3

a fait accompli, immune from review or revision.

Here, Appellants do not merely seek a sterile declaration of the invalidity of a now-

defunct rule. Rather, they dispute the enforceability in still-pending administrative challenges of

settlement statements issued pursuant to, and after the expiration of the allegedly invalid rule.

The Commission’s “Voidability” Argument Does not Defeat Jurisdiction.

The Commission also contends that, even if we find the Orders not to have been

adopted in substantial compliance with the Act, we cannot grant the relief Appellants seek because

a rule is valid until adjudicated void. Section 2001.035 of the Texas Government Code provides

that “[a] rule is voidable unless a state agency adopts it in substantial compliance with Sections

2001.0225 through 2001.034.” Tex. Gov’t Code § 2001.035 (emphasis added). The Commission

cites a number of cases for the proposition that a voidable rule—as opposed to a rule that is void

ab initio—is valid until adjudicated void. See Meeker v. Tarrant Cnty. Coll. Dist., 317 S.W.3d 754,

761 (Tex. App.—Fort Worth 2010, pet. denied). Thus, the Commission reasons, “the validity of

any actions ERCOT took in response to the … Orders in the interim (prior to a judgment in this

case) is not contingent on previously-unadjudicated questions about APA compliance.” See

Brazzel v. Murray, 481 S.W.2d 801, 803 (Tex. 1972). “At best,” the Commission concludes,

20
“Appellants’ APA challenge can invalidate only future efforts to include firm-load shed in

calculating scarcity pricing signals.” We disagree.

The Commission makes much of the distinction between an act that is void ab initio,

which is a legal nullity, and an act that is merely voidable, and which therefore remains in effect

until adjudicated void. The Commission would have us conclude that, upon holding a voidable

rule to be void, a court is powerless to grant an aggrieved party any relief that would disturb the

outcomes of actions previously undertaken pursuant to the then-voidable, now-void rule. We find

that proposition poorly supported by the cases the Commission cites.

Two such cases arose under the Texas Open Meetings Act, which provides that

“[a]n action taken by a governmental body in violation of this chapter is voidable.” Tex. Gov’t

Code § 551.141. In the first of these, Meeker challenged the validity of successive employment

contracts between the Tarrant County College District and its chancellor, Dr. de la Garza, arguing

that deliberations on the contracts violated the Open Meetings Act because the agenda notices for

three meetings were improper, making two resultant contracts void. Meeker, 317 S.W.3d at 758.

Meeker sought to enjoin the college from paying de la Garza’s salary or otherwise recognizing

him as chancellor. Id. Finally, Meeker sought attorney’s fees. Id. at 760. The college prevailed

on cross-motions for summary judgment, and Meeker appealed. Id. at 758. While the appeal was

pending, the college entered into an amended contract with de la Garza under which he would step

down as chancellor and receive severance consideration for release of claims. Id. The college

then moved to dismiss the pending appeal as moot. Id. In dismissing the case, the Fort Worth

court held that, because all of Meeker’s requests for injunctive relief related to de la Garza’s service

as chancellor, the cause was moot because de la Garza was no longer the chancellor. Id. at 760

(citing Day v. First City Nat’l Bank of Hous., 654 S.W.2d 794, 795 (Tex. App.—Houston [14th

21
Dist.] 1983, no writ) (“case is moot when the actions sought to be enjoined have been fully

performed”)). The court similarly held that the final contract rendered moot the requests for

declarations that the meeting notices violated the Open Meetings Act and the two prior

employment contracts were void because the requested declarations could not affect a live

controversy. Id. at 761. Finally, the court found the request for attorney’s fees to be wholly

dependent on the claim for injunctive relief, and therefore also moot. Id. at 762.

In Meeker, as in the cases discussed above involving expiration, the dispositive

issue was that the plaintiff sought prospective relief against actions that were no longer in prospect.

An injunction is inherently forward-looking, intended to prohibit present or future conduct. With

de la Garza having stepped down as chancellor, an injunction purporting to prohibit his recognition

in that office would have been pointless. His contracts having been superseded, any declaration

would have been “ineffectual for want of a subject matter upon which it could operate.” Methodist

Hosps., 874 S.W.2d at 147. Whereas, in the earlier cases, the allegedly valid rule had expired, in

Meeker, the allegedly invalid contract had been superseded by a subsequent contract, but in neither

scenario was the court confronted—as we are here—with the issue of whether an aggrieved

plaintiff can recover damages (or obtain relief from obligations) incurred by reason of actions

undertaken during the operation of the allegedly invalid rule or contract. True, the Meeker court

noted that “[a]ctions in violation of [the Open Meetings Act] are voidable, not void, and remain

valid until adjudicated and declared void,” apparently finding significance in the fact that “the

[first] Contract had not been adjudicated void before the [second] Contract superseded it, and the

[second] Contract had not been adjudicated void before the [final] Contract replaced it.” Meeker,

317 S.W.3d at 761. While that language may be read to support the Commission’s position, we

fail to see how it was essential to the outcome in Meeker itself. That is, we do not see how Meeker

22
would have had any greater or lesser entitlement to the prospective relief he sought if the

employment contacts had been void ab initio. In that scenario, in our view, he still would have

been improperly seeking to enjoin conduct completed in the past—the epitome of mootness.

The second Open Meetings Act case is more distinct factually and, therefore, less

instructive. In July of 2006, the City of Dallas executed the Love Field Agreement, formally

binding itself to the terms of an earlier provisional agreement allegedly negotiated in violation of

the Open Meetings Act. Love Terminal Partners v. City of Dallas, 256 S.W.3d 893, 896 (Tex.

App.—Dallas 2008, no pet.). The federal government later enacted the Wright Amendment

Reform Act of 2006, incorporating many provisions of the Love Field Agreement. Id. The

Plaintiffs (LTP) sued the city and others seeking declaratory and injunctive relief concerning Open

Meetings Act violations in adopting the agreement. Id. at 895–96. In affirming the trial court’s

dismissal of the case as moot, the Dallas court held that the Reform Act’s incorporation of the

contract made Dallas’ obligations a matter of federal law, rendering moot any challenge to the

Love Field Agreement regarding such obligations. Id. at 897.

Like Meeker, the LTP case discusses the void/voidable distinction in terms

superficially favorable to the Commission but ultimately of little relevance to the instant case. LTP

had argued that, because the City’s conduct leading up to the agreement violated the Open

Meetings Act, the resulting contract was void and, therefore, the Reform Act did not moot their

claims. Id. Admittedly, language in the opinion can be read as accepting LTP’s premise, if

ultimately rejecting its conclusion. Although the court recites the rule that “if conduct is merely

voidable, the act is valid until adjudicated and declared void,” id., it is unclear how the case could

have come out differently had the agreement been void ab initio rather than merely voidable

because the agreement’s terms had been incorporated into federal law at the time of the litigation.

23
What gave the agreement its continuing force, then, was not the (allegedly defective) process by

which it was originally adopted but its incorporation into federal law. We do not understand our

sister court to imply that, had it found the agreement void, it would have been free to enjoin Dallas’

compliance with the federal Reform Act because, unbeknownst to the Congress that passed the

bill and the President who signed it, they were incorporating a mere nullity into federal law.

Rather, the LTP court probably meant nothing more than that the incorporation of the agreement

into federal law rendered it too late for an adjudication of its voidability to have practical

consequences for the parties, as they would be bound by the federal law in any event. This

illustrates why the case is of limited use in resolving the present question. Whereas the operation

of the Reform Act would have caused any intervention by the LTP court to be of no practical

consequence, no comparable circumstance constrains our decision in the instant case. What

Meeker and LTP have in common, then, is that, despite including language that, admittedly, can

be read as supporting the Commission’s interpretation, the facts of each case disclose a wholly

independent reason for reaching the result: the prospective nature of the relief sought in Meeker,

and the intervention of the federal government in LTP.

Of course, not every judgment or order overturning some putative official act

necessarily operates to unwind every collateral consequence of the defunct act. The Commission

cites Brazzel v. Murray, a 1972 case in which the Texas Supreme Court held that a judgment

voidable for purely procedural reasons has practical effects until voided, unlike a void judgment

that can have no effect from its inception. 481 S.W.2d 801, 803 (Tex. 1972). In that case, Brazzel

obtained judgment against Murray at trial, but the Waco court reversed and remanded with costs

assessed against Brazzel. Id. at 802. Under then-current procedural rules, the clerk of the appellate

court was directed not to deliver the mandate of the court nor to certify the proceedings to the trial

24
court before the payment of costs, but the clerk did so in error. Id. Murray filed a motion to recall

the mandate in the appellate court, which was granted, and later filed a motion to dismiss in the

trial court, relying on a then-current rule providing that, if no mandate had issued within a year of

final judgment, the trial court should dismiss the case from its docket. Id. Murray reasoned that,

because the mandate had been issued in violation of the rule, the time during which the erroneously

issued mandate had been in putative effect counted toward the one-year period prescribed by the

rule. Finding that a year had passed after judgment without issuance of a mandate, the trial court

dismissed the case, and the Corpus Christi court affirmed, holding that the erroneously issued

mandate had been “void ab initio.” Id.

The Supreme Court disagreed. The Court quoted liberally from the early case of

Murchison v. White for the proposition that “[a] void act is one entirely null within itself, not

binding on either party, and which is not susceptible of ratification or confirmation” such that “[its]

nullity cannot be waived,” whereas “[a] voidable act is one which is not absolutely void within

itself, but which is binding until disaffirmed, and which may be made finally valid by failure within

the proper time to have it annulled, or by subsequent ratification or confirmation.” Brazzel,

481 S.W.2d at 803 (quoting Murchison v. White, 54 Tex. 78, 81 (1880) (internal quotation marks

omitted)). Concluding that “judgments which are rendered without observance of statutory

requirements which are purely procedural are not void, however irregular or erroneous they may

be,” the Court held that “[f]rom … the date of the original mandate, until … the date the original

mandate was recalled, a mandate was on file with the trial court in this cause” and that, although

“[it] was an unauthorized mandate, [it] was valid until recalled.” As such, “[at] no time [had] there

been a period of twelve months in which no mandate has been filed with the trial court, as required

by [then-existing rules], as a basis for dismissal.” Brazzel, 481 S.W.2d at 803.

25
 But while Brazzel involves an official action that, once invalidated, was not entirely

annulled, it does not stand for the proposition that Appellants here must live with the allegedly

invalid electricity pricing at issue in this case without an opportunity for judicial review. In

Brazzel, the primary defect in the mandate did not concern any relief to which Murray may have

been entitled. True, Murray claimed a procedural right to challenge the validity of a mandate

issued despite nonpayment of costs, but Murray did not claim the right to recover any costs

incurred as a result of the mandate. The case therefore cannot be read to bar such recovery where

such costs are at issue—nor to establish that the availability of such recovery would depend on

whether the order was merely void rather than void ab initio. In any event, Appellants do not ask

us to hold that the disputed Orders in this case were never issued. Rather, Appellants seek only a

declaration as to the invalidity of the Orders, such that the pricing thereunder may not be applied

in pending administrative proceedings. Such a declaration would not be inconsistent with Brazzel.

In summary, none of the cases cited to us by the Commission or aligned intervenors

compels support for the Commission’s position that courts lack jurisdiction to remedy actions

undertaken pursuant to a voidable order before it is adjudicated as void. Our own review of the

case law on this question inclines us to the opposite view.

The cases on voidability, including those cited by the Commission, trace back to

early cases addressing the voidability of private contracts. See, e.g., Cummings v. Powell, 8 Tex. 80

(1852) (surveying then-current authorities on voidability of contracts with minors). Perhaps the

most familiar context in which the void/voidability distinction arises is the voidability of a contact

with a minor. The hornbook law is that a contract made with a minor is voidable at the minor’s

election. Dairyland Cnty. Mut. Ins. v. Roman, 498 S.W.2d 154, 158 (Tex. 1973); Prudential Bldg.

& Loan Ass’n v. Shaw, 26 S.W.2d 168, 171 (Tex. Comm’n App. 1930, judgm’t adopted).

26
Importantly, a contract so voided is deemed to have been void for both parties from the beginning.

Kargar v. Sorrentino, 788 S.W.2d 189, 191 (Tex. App.—Houston [14th Dist.] 1990, no writ). The

effect, therefore, is essentially identical to the remedy of rescission, i.e., to put the parties in their

respective positions status quo ante, as though the contract had never formed. Accordingly, a

minor who voids a contract on the ground of incapacity may recover the full amount of any

consideration the minor has paid, but also must restore any consideration the minor has received.

Shaw, 26 S.W.2d at 171; James v. Barnett, 404 S.W.2d 886, 888 (Tex. App.—Dallas 1966, writ

ref’d n.r.e.); see also Hogue v. Wilkinson, 291 S.W.2d 750, 753 (Tex. App.—Texarkana 1956,

no writ).

The latter two points are crucial. The minor, upon attaining the age of majority,

has the option to ratify and enforce the contract, enjoying the benefit of the bargain. That is the

very definition of a voidable contract. See CONTRACT, Black’s Law Dictionary (11th ed. 2019)

(defining “voidable contract” as “[a] contract that can be affirmed or rejected at the option of one

of the parties”); Murchison, 54 Tex. at 81 (“[a] voidable act is one … which may be made finally

valid . . . by subsequent ratification or confirmation”). And yet, in the contract context, we find

nothing resembling the rule against retroactive relief urged by the Commission. On the contrary,

not only may the voiding party recover consideration already paid under the contract, if any, the

voiding party must tender any consideration previously received in order to be restored to status

quo ante. By parity of reasoning, it is unclear why a party aggrieved by the voidable act of a state

agency should not be able to recover a remedial sum necessary to make it whole once the act is

finally adjudicated void, or, as here, to avoid an obligation predicated on the void act.

We are mindful that we have found no precedent that clearly and unambiguously

disposes of the question before us. It is, moreover, entirely reasonable to suppose that some

27
allegedly invalid act by an agency may eventually become impracticable to remedy due to

developments over the passage of time. In the above-cited case of Murchison v. White, an early

case involving a challenge to the judgment of a probate court, the Texas Supreme Court held that

“[a] voidable act is one which is not absolutely void within itself, but which is binding until

disaffirmed, and which may be made finally valid by failure within the proper time to have it

annulled, or by subsequent ratification or confirmation.” 54 Tex. at 81 (emphasis added). The

court did not specify the proper time in which to have an act annulled, but Appellants here filed

their direct appeal within 15 days after the First Order was issued on February 15, 2021. Section

39.001(f) of the Utilities Code, requires that validity challenges be brought “not later than the 15th

day after the date on which the rule as adopted is published in the Texas Register.” It is undisputed

that the Order challenged in this case was not filed in the Texas Register, so the event from which

the statutory time limit is to run has not yet occurred. We think it consonant with the spirit and

intent of the statute, however, to hold Appellants’ challenge, filed 15 days after the Order date, to

be timely for purposes of the statutory provision. As such, we further conclude that, under

Murchison, Appellants have acted “within the proper time” to have the Orders annulled.

Appellants’ Alleged Injury is Redressable.

The Commission next argues that this Court lacks jurisdiction because Appellants’

claims are not redressable. We disagree.

The Commission cites cases reciting the general standard that we lack jurisdiction

over cases in which the alleged injury is not likely to be remedied by the requested relief. Referring

to its own prior argument that Appellants can, at most, obtain a declaration concerning future

application of the Orders but cannot disturb transactions already settled, the Commission reasons

28
that, because reversing the Orders “will not result in the return of any money that Appellants have

already paid ERCOT pursuant to settlement invoices,” such reversal will not remedy Appellants’

injury, and our jurisdiction to order such reversal fails for want of redressability. The Commission

begs the question. It imports its own previous conclusion from its mootness/voidness argument

and uses it as a premise in its redressability argument. Having rejected that conclusion in the

former context, however, we likewise reject its application as a premise here. We see no reason

why the resolution of challenges to settlement invoices in ADR proceedings, whether due and

payable or already paid under timely preserved protest, cannot and should not be governed by this

Court’s determination regarding appropriate pricing.

Citing a pair of statutory provisions, the Commission contends that “Appellants

cannot show that any further proceedings on remand to reconsider or further justify the substance

of the orders . . . will lead to application of a different rule or otherwise change the outcome of any

pending administrative disputes.” See Tex. Util. Code § 39.001(f); Tex. Gov’t Code § 2001.040.

Section 39.001(f) of the Utilities Code provides that a court “shall . . . if appropriate on reversal,

remand [a rule held to be invalid] to the commission for further proceedings.” Government Code

Section 2001.040 requires that, on remanding certain rules to a state agency, a court “shall provide

a reasonable time for the agency to either revise or readopt the rule through established procedure.”

The Commission argues in essence that, for purposes of redressability analysis, these provisions

render the question whether Appellants’ requested relief will remedy the alleged injury merely

speculative, because the Commission may on remand simply ratify the previously adopted pricing

Order. We disagree.

First, we do not share the Commission’s view that the cited provisions limit our

power to order an appropriate remedy. Taken in context, Section 39.001(f) of the Utilities Code

29
makes clear that we may affirm, reverse, or remand a rule. Reversal or remand of the rule could

change the result of the pending administrative proceedings concerning amounts Appellants owe.

Second, Section 2001.040 of the Government Code uses permissive language—a

court “may remand” a rule—but does not expressly prohibit a court from reversing outright. To

the extent any prohibition is thought to be implied by that provision, we think it controlled by the

more specific authority found in the Utilities Code that permits remand “if appropriate” upon

reversal, suggesting that reversal without remand is the default. See Tex. Util. Code § 39.001(f);

see also id. § 39.001(e) (providing that “Judicial review of competition rules adopted by the

commission shall be conducted under Chapter 2001, Government Code, except as otherwise

provided by this chapter” (emphasis added)).

Third, Section 2001.040 is limited by its own terms to remands based on a finding

that an agency has not substantially complied with one or more procedural requirements of

Sections 2001.0225 through 2001.034. Appellants’ challenge is not limited to alleged lack of

compliance with those provisions, however, but includes others, such as the failure to file with the

Secretary of State under Section 2001.036. Thus, even the Commission’s interpretation of Section

2001.040 would not deprive us of jurisdiction for want of redressability.

The Commission’s final argument on redressability in turn consists of three

essential components: (1) that ERCOT, rather than the Commission, is the appropriate entity to

initiate a market-wide price correction, and ERCOT is not a party; (2) that Section 39.001(f),

Utilities Code, does not authorize us to order the PUC to order ERCOT to undertake such repricing;

and (3) that ERCOT’s authority to act is too limited to redress Appellants’ injury and, in any case,

that it is too late under the Nodal Protocols for ERCOT to act. We disagree.

30
 First, we are not persuaded that ERCOT must be brought before us as a party in

order for us to grant the requested relief in light of the Commission’s “complete authority” over

ERCOT under Section 39.151(d), Utilities Code, together with ERCOT’s power to determine

market clearing prices unless “otherwise directed by the commission” under Title 16, Texas

Administrative Code Section 25.501(a), both of which provisions were expressly referenced in the

Order that is the subject of this suit.

Second, we do not find the limitation on our authority the Commission argues may

be found in Section 39.001(f) of the Utilities Code. That provision directs a court of appeals

considering a challenged rule “to render judgment affirming the rule or reversing and, if

appropriate on reversal, remanding the rule to the commission for further proceedings, consistent

with the court’s opinion and judgment.” Tex. Util. Code § 39.001(f) (emphasis added).

Third, the Commission overstates the effects of the Nodal Protocols. The

Commission acknowledges that those rules “provide ERCOT with some discretion after settlement

point prices have become final to determine that prices qualify for a correction and then seek

ERCOT Board review of such prices.” Per the Nodal Protocols, however, ERCOT would have

had to notice participants during the month after the subject Order was issued and, the Commission

argues, “[t]hat ship has sailed.” Thus, “even if this Court were to order the PUC to issue an order

directing ERCOT to resettle the market,” the Commission reasons, “Appellants have not identified

any means under the Nodal Protocols by which ERCOT can now make any market-wide

corrections to pricing for EEA3-operating days.” This argument lacks merit. The fact that ERCOT

may be time-barred under the protocols from effecting a price correction on its own initiative has

few if any implications for the question of whether it may do so pursuant to an order of this Court.

While we can imagine scenarios in which limitations on an entity’s powers to act may operate as

31
practical limitations on what a court may order it to do, this is not such a case. Where, as here,

administrative disputes of the settlement statements were timely initiated and stayed pending

decision of this timely-filed direct appeal, we discern no reason why the notice provisions that

would ordinarily apply to ERCOT acting alone should prevent us from hearing Appellants’

challenge and, if appropriate, granting relief.

The Orders are Rules.

In its fourth challenge to this Court’s jurisdiction, the Commission argues that the

Orders were not competition rules within the meaning of Section 39.001(e), Utilities Code, and

thus are not subject to the direct-appeal process provided by that statute. We agree that the statute

is jurisdictional; as such, unless we find the subject Orders to be rules, we must dismiss. Appellants

propose that the APA’s definition of “rule” should govern here. The Commission contests the

applicability of that definition, but also argues that, in any event, the Orders do not fall within its

ambit. We will begin our analysis by assuming the statutory definition applies.

The APA defines “rule” as “a state agency statement of general applicability that:

(i) implements, interprets, or prescribes law or policy; or (ii) describes the procedure or practice

requirements of a state agency,” including “the amendment or repeal of a prior rule”, but not

including “a statement regarding only the internal management or organization of a state agency

and not affecting private rights or procedures.” Tex. Gov’t Code § 2001.003(6). The Commission

argues, first, that the Orders cannot be rules because they are not generally applicable but are mere

statements regarding internal management or organization and not affecting private rights or

procedures. We have previously observed that the distinction between a rule and an internally

directed agency statement may sometimes be “elusive” but have held the “core concept” to be that

32
“the agency statement must in itself have a binding effect on private parties.” Slay v. Texas

Comm’n on Envtl. Quality, 351 S.W.3d 532, 546 (Tex. App.—Austin 2011, pet. denied) (citing

Texas Dep’t of Pub. Safety v. Salazar, 304 S.W.3d 896, 905 (Tex. App.—Austin 2009, no pet.)).

The Commission reasons in essence that, because the Orders at issue here were directed to ERCOT

rather than to market participants, the Orders could not in themselves bind private parties and thus

cannot be rules within the meaning of the APA. We disagree.

The Commission’s position here suffers from two significant defects. First, there

is no dispute that the Orders at issue were intended to affect the rights of private parties and in fact

did so. The Commission’s hypothetical scenario, in which ERCOT fails to comply with the Order,

may be an interesting thought experiment, but it does not refute the conclusion that the purpose

and effect of the subject Orders was to cause market participants to pay (or receive) $9,000/MWh

during the EEA3 event. For this reason, we are also unpersuaded by the Commission’s proposed

distinction between this case and the cases relied on by Appellants, involving “self-executing”

rules rather than orders directed at (or through) ERCOT. The distinction arises from an artifact of

the unique regulatory structure governing electric power in Texas, in which ERCOT acts as

something akin to a subagency between the PUC and market participants. ERCOT is a separate

entity, but one over which the PUC has “complete authority.” Tex. Util. Code § 39.151(d). Where,

as here, ERCOT acts as the Commission’s agent, its acts are imputed to its principal. Put another

way, we fail to see legal or factual support for the Commission’s proposed distinction that would

insulate it from review in countless cases in which it acts through ERCOT.

Second, the Commission’s argument proves too much. For, if being directed solely

to ERCOT rather than to market participants sufficed to negate a statement’s status as a rule, any

number of rules found in Title 16, Texas Administrative Code, Chapter 25 (“Substantive Rules

33
Applicable to Electric Service Providers”) would likewise not be rules. Indeed, Section 25.505 of

that chapter, the subject of the Orders at issue in this case, consists almost entirely of mandatory

directives to ERCOT, and was adopted entirely through the formal APA rulemaking process. We

think no one would seriously dispute that that section is a rule. Section 25.505 is general, in the

sense that it affects the entire unregulated market in this state, or 26 million customers, and there

can be little doubt that it “(i) implements, interprets, or prescribes law or policy” within the

meaning of Section 2001.003(6)(A) of the Government Code. And, by definition, “the amendment

… of a prior rule” is itself a rule. Tex. Gov’t Code § 2001.003(6)(B). Thus, if the subject Order

had the effect of amending a rule, then the Order by definition must also be a rule. Citing an

outcome that was believed to be “contrary to the purpose of the rule,” the Commission in its Order

expressly granted an “exception” to Section 25.505’s application, the basis for which is neither

found in nor logically derived from the text of the rule itself. Accordingly, the effect of the Order

on the cited rule fits the ordinary meaning of “amendment,” and the Order, however packaged,

therefore fits the statutory definition of a “rule.”

None of the cases the Commission cites compel the contrary conclusion. At issue

in Slay, for example, was whether an internal memorandum detailing the Texas Commission on

Environmental Quality’s (TCEQ) penalty policy constituted a rule for purposes of establishing

jurisdiction under a statutory waiver of governmental immunity. Slay, 351 S.W.3d at 532. Critical

to our finding that the trial court did not abuse its discretion in answering that question in the

negative was the fact that, although TCEQ staff was required to follow the penalty policy’s

methodology in determining penalty recommendations, the commissioners were not bound to do

so when acting on those recommendations but retained discretion to depart from the methodology

when imposing penalties. Id. at 546. This is a significant distinction. In the present case, the

34
Commission issued the directive to ERCOT with the expectation that it be followed. Though

ERCOT might have declined to follow the Commission’s Order, but it had no discretion to do so

and, in any event, it followed the Commission’s directive in fact.

Similarly unavailing is Vista Community Medical Center, another case

distinguishing internally directed statements from those of general applicability. Texas Mut. Ins.

v. Vista Cmty. Med. Ctr., LLP, 275 S.W.3d 538 (Tex. App.—Austin 2008, pet. denied). In Vista,

we reversed the trial court’s determination that a one-page staff report, delivered by the director of

medical review to the agency commissioners of the Texas Department of Insurance Division of

Workers Compensation at an open meeting and arguing for one of two conflicting interpretations

of an existing agency rule, was itself a rule that had to be adopted in conformance with APA

rulemaking proposals. Id. at 545–46. We found that the staff report was not a “statement by a

state agency” within the meaning of the APA’s definition of a rule. Id. at 555. We noted that the

report “was a one-page document prepared by the director of the Medical Review Division … that

was intended to address an internal agency matter—namely, the inconsistent application” of a

particular rule. Id. The present case could scarcely be more different. Here, the subject Orders

were voted on and issued by the Commission. There is therefore no dispute that they constitute a

statement by a state agency.

We went further in Vista, noting that, even if the report had been an agency

statement, it lacked other characteristics of a rule. We noted that “not every administrative

pronouncement is a rule within the meaning of the APA.” Id. (citing Texas Educ. Agency

v. Leeper, 893 S.W.2d 432, 443 (Tex. 1994); Brinkley v. Texas Lottery Comm’n, 986 S.W.2d 764,

769 (Tex. App.—Austin 1999, no pet.)). Acknowledging that “agencies routinely issue letters,

guidelines, and reports, and occasionally file briefs in court proceedings, any of which might

35
contain statements that intrinsically implement, interpret, or prescribe law, policy, or procedure or

practice requirements,” we reasoned that subjecting every such statement to rulemaking

requirements would make it impossible for an agency to carry out its legislative functions. Id.

(quoting Brinkley, 986 S.W.2d at 769). In determining whether an agency’s interpretation of its

own rule was also a rule, following the Supreme Court’s holding in El Paso I, we concluded that

an interpretation that contradicted or otherwise amended an existing rule would be a new rule, but

found no such amendment in the case then before us. Id. at 555-56 (citing El Paso I, 247 S.W.3d

at 714–15). The report “was presented to the Division at the January 2005 open meeting, but the

Division simply thanked the director for the report and took no official action.” Id. Again, the

present case could not be more different. Far from an internal discussion of opinion, the subject

Order was just that—an order, intended to secure compliance by ERCOT and bring about a desired

outcome. And, as already noted above, the Order at issue in this case had the effect of creating an

extratextual exception to a previous rule, and therefore did not merely interpret but also amended

that rule.

Appellants Challenge the Validity of the Orders.

The Commission further argues that we lack jurisdiction because Appellants

challenge the application, rather than the validity of the subject Orders. We disagree. It is true

that the PUC gave incomplete guidance to ERCOT, with the understanding that the latter entity

would take appropriate measures to achieve the desired result. But from the face of the pleadings,

it is apparent that most if not all Appellants clearly challenge not only the measures but the

guidance itself.

36
 In the subject Order, the Commission stated the problem it confronted as follows:

ERCOT has informed the Commission that energy prices across the system
are clearing at less than $9,000, which is the current system-wide offer cap
pursuant to 16 TAC § 25.505(g)(6)(B). The Commission believes this
outcome is inconsistent with the fundamental design of the ERCOT market.
Energy prices should reflect scarcity of the supply. If customer load is being
shed, scarcity is at its maximum, and the market price for the energy needed
to serve that load should also be at its highest.

Having thus summarized the problem it sought to address, the Commission continued,

“[a]ccordingly, the Commission directs ERCOT to ensure that firm load that is being shed in EEA3

is accounted for in ERCOT’s scarcity pricing signals.” Id.

We read the foregoing Order to require—quite expressly—that ERCOT take steps

necessary to ensure that energy prices would clear at $9,000 during the duration of the EEA3 event.

We understand Appellants to complain that ERCOT in fact did so. We do not hear Appellants to

challenge only the manner in which ERCOT achieved that result; nor has any party urged an

interpretation of the Order in support of an argument that, for example, a rogue ERCOT made the

complained-of pricing changes without authorization from the PUC. Moreover, Appellants

explicitly invoke the language of a substantive and procedural validity challenge. In sum, we see

nothing in the record or the briefing before us to indicate that this case involves a challenge to

application rather than validity.

For the reasons set forth above, we overrule the Commission’s challenges to our

jurisdiction and proceed to consider the merits of the Appeal.

Merits

Appellants bring two challenges to the validity of the subject Orders in this case.

First, Appellants argue that the Orders constitute competition rules and that the Commission in

37
adopting those rules failed to comply with APA’s regular or emergency rulemaking procedures.

Second, as an independent ground for relief, Appellants contend that the substance of the Orders

themselves exceeded the Commission’s statutory authority. Because we agree with Appellants as

to the second ground, we do not address the first.

Luminant’s substantive challenge to the Commission’s statutory authority in turn

rests on two sub-arguments: first, that the subject Orders “contravene[ ] statutory language and

run[ ] counter to statutory objectives that electricity prices should be determined by the normal

forces of competition,” and, second, that the Orders were not properly limited in duration. Because

we agree with the first of these arguments, we do not reach the second.

State administrative agencies have only those powers that the legislature expressly

confers upon them and those implied powers that are reasonably necessary to carry out their

express functions or duties. See Public Util. Comm’n v. City Pub. Serv. Bd., 53 S.W.3d 310, 315

(Tex. 2001). Absent specific or implied statutory authority, an agency rule is invalid. Id.

Furthermore, an agency may not exercise what is effectively a new power based on a claim that

the exercise is expedient for administrative purposes. Id. To establish a rule’s facial invalidity, a

challenger must show that the rule (1) contravenes specific statutory language; (2) is counter to

the statute’s general objectives; or (3) imposes additional burdens, conditions, or restrictions in

excess of or inconsistent with the relevant statutory provisions. Texas Ass’n of Acupuncture &

Oriental Med. v. Texas Bd. of Chiropractic Exam’rs, 524 S.W.3d 734, 739 (Tex. App.—Austin

2017, no pet).

Determining the validity of a rule requires us to construe the statutory language

from which the agency’s authority purportedly flows. Texas Orthopaedic Ass’n v. Texas State Bd.

of Podiatric Med. Exam’rs, 254 S.W.3d 714, 719–20 (Tex. App.—Austin 2008, pet. denied). In

38
construing a statute, our primary objective is to give effect to the legislature’s intent. Liberty Mut.

Ins. v. Adcock, 412 S.W.3d 492, 494 (Tex. 2013). “The ‘surest guide to what lawmakers intended’

is the enacted language of a statute.” Youngkin v. Hines, 546 S.W.3d 675, 680 (Tex. 2018) (quoting

Entergy Gulf States, Inc. v. Summers, 282 S.W.3d 433, 463 (Tex. 2009)). We defer to the agency’s

construction of a statute “only when the statutory language is ambiguous.” Southwest Royalties,

Inc. v. Hegar, 500 S.W.3d 400, 405 (Tex. 2016). If the statute is unambiguous, agency

deference “has no place.” TracFone Wireless, Inc. v. Commission on State Emergency Commc’ns,

397 S.W.3d 173, 182 (Tex. 2013).

Two Utilities Code provisions are key here. The first provision is the

policy-and-purpose provision set forth at the beginning Chapter 39, which includes the

Legislature’s express finding that “the production and sale of electricity is not a monopoly

warranting regulation of rates, operations, and services and that the public interest in competitive

electric markets requires that … electric services and their prices should be determined by

customer choices and the normal forces of competition.” Id. § 39.001(a). That section further

provides that “[r]egulatory authorities … shall authorize or order competitive rather than

regulatory methods to achieve the goals of this chapter to the greatest extent feasible and shall

adopt rules and issue orders that are both practical and limited so as to impose the least impact on

competition.” Id. § 39.001(d).

The second provision is Section 39.151(d), Utilities Code, which the Commission

expressly invoked in the Orders. Importantly, the Orders expressly addressed the scarcity pricing

mechanism that by its own terms was adopted pursuant to Section 39.151(d). See Tex. Admin.

Code § 25.505(g)(6)(B) (eff. May 30, 2019, to July 13, 2021). Section 39.151 chiefly concerns

the creation and management of ISOs but also provides, in relevant part, that “[t]he commission

39
shall adopt and enforce rules relating to the reliability of the regional electrical network and

accounting for the production and delivery of electricity among generators and all other market

participants, or may delegate to an independent organization responsibilities for establishing

or enforcing such rules.” Tex. Util. Code § 39.151(d). “Independent organization” is defined

as a region’s ISO, which, in this case, is ERCOT. Id. § 39.151(b); Texas Com. Energy,

2004 WL 1777597, at *2. An adjacent provision requires such entities to “ensure the reliability

and adequacy of the regional electrical network.” Id. § 39.151(a)(2).

A threshold question is whether the authority granted by Section 39.151 qualifies,

or is qualified by, the limitations imposed by 39.001. Put another way, the question is whether, or

to what extent, Section 39.151’s directive to ensure system reliability provides an exception to

Section 39.001’s general preference for reliance on competition rather than regulation to set prices.

This appears to be a case of first impression. We must interpret statutory text in light of the statute

as a whole. Davis v. Morath, 624 S.W.3d 215, 222 (Tex. 2021). We must, if possible, give effect

to all included words without treating any language as surplusage. Hlavinka v. HSC Pipeline

P’ship, LLC, 650 S.W.3d 483, 491 (Tex. 2022), reh’g denied (Sept. 2, 2022); see also Tex. Gov’t

Code § 311.021 (providing that “[i]n enacting a statute, it is presumed that … the entire statute is

intended to be effective”). Here, Section 39.001 makes clear that “competitive rather than

regulatory methods” are preferred “to the greatest extent feasible,” and with “the least impact on

competition.” Id. § 39.001(d). In contrast, Section 39.151 is silent as to whether “regulatory”

rather than “competitive” methods may be adopted to ensure grid reliability. In the absence of

specific guidance from Section 39.151, the Commission’s actions must be subject to the constraint

provided by the text of Section 39.001. Fidelity to the whole-text canon, as codified in our Code

Construction Act, requires that we give effect to the phrase “greatest extent feasible.” To find that

40
the subject Orders were compliant with this statutory constraint, then, we must find that the Orders

could not have used “competitive rather than regulatory methods” to any greater extent than they

did as issued. This we cannot do; the very notion is belied by the operation of the SPM as it existed

immediately before the issuance of the subject Orders, and by the fact that, operating

independently, this SMP did not result in “HCAP” pricing.

In extreme circumstances under extraordinary pressure, the Commission exceeded

its power by eliminating competition entirely. The previously adopted rule set the system-wide

offer cap at $9,000/MWh. 16 Tex. Admin. Code § 25.505(g)(6)(B). The typical market clearing

price in Texas can be $30/MWh. During Winter Storm Uri, the price had risen to just above

$1,200, but that rise was not having the effect the Commission desired. The Orders, which we

determined above are rules, instructed that if customer load is being shed, scarcity is at its

maximum, and “the market price for the energy needed to serve that load should also be at its

highest.” First Order, at 1; Second Order, at 1. Knowing that the First Order caused ERCOT to

set the market clearing price at the market cap of $9,000/MWh, the Commission issued the Second

Order with the identical language directing that the market price for energy be at its highest while

there was load shed. For four days under the Orders, the minimum price was the same as the

maximum price by operation of executive fiat. While the breadth of the Commission’s discretion

largely resists sharp delineation, the Legislature clearly stated that the Commission’s rules must

be “limited so as to impose the least impact on competition.” See Tex. Util. Code § 39.001(d)

(emphasis added). Instead, the Orders had the maximum impact on competition conceivable by

setting a single price for power and directing ERCOT to take all necessary steps to ensure the

market cleared at that single price. While the extraordinary circumstances of Winter Storm Uri

may have required extraordinary modifications to the SPM to send appropriate pricing signals to

41
prompt the necessary market response, the Commission here exceeded the Legislature’s limits on

its power. Setting a single price at the rule-based maximum price violated the Legislature’s

requirement in the Utilities Code Section 39.001(d) that the Commission use competitive methods

to the greatest extent feasible and impose the least impact on competition.

IV.
CONCLUSION

For the foregoing reasons, we reverse the Commission’s Orders of February 15

and 16, 2021, and remand this case for further proceedings consistent with our ruling.

_________________________________________
Edward Smith, Justice

Before Chief Justice Byrne, Justices Goodwin and Smith;
Justice Goodwin not participating

Reversed and Remanded

Filed: March 17, 2023

42