Amicus Curiae Brief — West Virginia, et al., Petitioners v. Environmental Protection Agency, et al.

No. 20-1530

================================================================================================================

In The

Supreme Court of the United States

---------------------------------♦--------------------------------STATE OF WEST VIRGINIA, et al.,

Petitioners,

v.

UNITED STATES ENVIRONMENTAL

PROTECTION AGENCY, et al.,

Respondents.

---------------------------------♦--------------------------------On Writ Of Certiorari To The

United States Court Of Appeals

For The District Of Columbia Circuit

---------------------------------♦--------------------------------BRIEF OF AMICI CURIAE GRID EXPERTS

BENJAMIN F. HOBBS, BRENDAN KIRBY,

KENNETH J. LUTZ, AND JAMES D. MCCALLEY

IN SUPPORT OF RESPONDENTS

---------------------------------♦--------------------------------Of Counsel:

WILLIAM BOYD

UCLA SCHOOL OF LAW

405 Hilgard Avenue

Los Angeles, CA 90095

(310) 206-5280

boyd@law.ucla.edu

CARA A. HOROWITZ

Counsel of Record

ANDRIA SO

FRANK G. WELLS

ENVIRONMENTAL LAW CLINIC

UCLA SCHOOL OF LAW

405 Hilgard Ave.

Los Angeles, CA 90095

(310) 206-4033

horowitz.elc@law.ucla.edu

Counsel for Amici Curiae

================================================================================================================

COCKLE LEGAL BRIEFS (800) 225-6964

WWW.COCKLELEGALBRIEFS.COM

i

TABLE OF CONTENTS

Page

TABLE OF AUTHORITIES .................................

iii

STATEMENT OF INTEREST .............................

1

SUMMARY OF ARGUMENT ..............................

4

ARGUMENT ........................................................

8

I.

Effective Power-Sector Pollution Controls

Work with the Distinctive Characteristics

of Electricity and the Interconnectedness

of the Regional Grids .................................

8

A. Electricity Is a Uniquely Fungible and

“Real-Time” Good ................................

8

B. Each of the Three Regional Grids Operates as a Single Machine ................. 10

C. Dispatch Governance Frameworks Are

Designed to Facilitate Shifts Among

Generators and Ensure Affordable, Reliable Electricity .................................. 15

II.

Power Companies and Grid Operators

Have Historically Responded to Air Pollution Controls by Shifting to Lower-Emitting Generators.......................................... 16

A. Air Pollution Control Measures Ordinarily Affect Dispatch Order, Causing

Generation Shifting ............................. 17

B. The Clean Power Plan Went Further

by Embedding Generation Shifting in

its Definition of the Best System of

Emission Reduction ............................. 19

ii

TABLE OF CONTENTS—Continued

Page

C. To Reduce Power-Sector CO2 At Lower

Cost, Regulated Entities and States

Should Have The Flexibility to Employ Off-Site Measures ........................ 22

III.

CO2 Pollution Control Measures Are Incorporated Easily Into Power-Sector Operations ......................................................... 25

A. U.S. Power Sector Is Shifting Toward

Cleaner Energy Sources ...................... 25

B. Propping Up Coal Is Unnecessary for

Grid Reliability .................................... 28

CONCLUSION..................................................... 30

iii

TABLE OF AUTHORITIES

Page

CASES

Fed. Power Comm’n v. Fla. Power & Light Co.,

404 U.S. 453 (1972) ...................................................9

Gainesville Util. Dep’t v. Fla. Power Corp., 402

U.S. 515 (1971) ........................................................14

New York v. Fed. Energy Regulatory Comm’n,

535 U.S. 1 (2002) .......................................................9

STATUTES

42 U.S.C. § 7651 ..........................................................18

42 U.S.C. §7651o .........................................................18

FEDERAL REGISTER

80 Fed. Reg. 64,662 (Oct. 23, 2015) ...... 6, 13, 16, 20, 26

83 Fed. Reg. 44,746 (Aug. 31, 2018) ...........................22

84 Fed. Reg. 32,520 (Jul. 8, 2019) ....................... passim

OTHER AUTHORITIES

Analysis Grp., Electric System Reliability and

EPA’s Clean Power Plan: The Case of PJM

(2015) .......................................................................29

Elec. Reliability Council of Tex., 2018 State of

the Grid (2018) ........................................................29

iv

TABLE OF AUTHORITIES—Continued

Page

Emanuele Massetti et al., Oak Ridge Nat’l Lab.,

Environmental Quality and the U.S. Power

Sector: Air Quality, Water Quality, Land Use

and Environmental Justice (2017) .........................18

Energy Info. Admin., U.S. Battery Storage Market Trends (2018) .....................................................10

Environmental Integrity Project, Greenhouse

Gases from Power Plants 2005-2020: Rapid

Decline Exceeded Goals of EPA Clean Power

Plan (2021) ........................................................ 26, 27

EPA, Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2019 (2021) .........................27

EPA, Regulatory Impact Analysis for the Repeal

of the Clean Power Plan, and the Emission

Guidelines for Greenhouse Gas Emissions

from Existing Electric Utility Generating

Units (2019) ....................................................... 22, 26

GE Energy Consulting, PJM Renewable Integration Study: Executive Summary Report

(2014) .......................................................................30

ICF Int’l, Inc., Assessing Effects on the Power

Sector of Greenhouse Gas Emission Standards (Oct. 31, 2018), Docket No. EPA-HQOAR-2017-0355-24419 ...................................... 20, 22

IHS Markit, Ensuring Resilient and Efficient

Electricity Generation (2017) ..................................28

Mass. Inst. of Tech., The Future of the Electric

Grid (2011) ................................................................5

v

TABLE OF AUTHORITIES—Continued

Page

N. Am. Elec. Reliability Corp., 2021 Long-Term

Reliability Assessment (2021) .................................28

Nat’l Renewable Energy Lab., Eastern Renewable Generation Integration Study (2016) ................29

Nat’l Renewable Energy Lab., Inertia and the

Power Grid: A Guide Without the Spin (2020) .......29

Paul Hibbard et al., The Economic Impacts of the

Regional Greenhouse Gas Initiative on Nine

Northeast and Mid-Atlantic States (2018) .............19

Paul L. Joskow, Creating a Smarter U.S. Electricity Grid, 26 J. Econ. Persp. 29 (2012) ................10

Paul L. Joskow et al., The Market for Sulfur Dioxide Emissions, 4 Am. Econ. Rev. 669 (1998)........23

Phillip F. Schewe, The Grid: A Journey Through

the Heart of Our Electrified World (2007) ............ 5, 8

Thomas M. Jackson et al., Evaluating Soft Strategies for Clean-Air Compliance, 6 IEEE, Computer Applications in Power (1993) ........................18

U.S. Dep’t of Energy, North American Electric

Reliability Corporation Interconnections,

https://www.energy.gov/oe/downloads/northamerican-electric-reliability-corporationinterconnections ......................................................12

U.S. Dep’t of Energy, Staff Report to the Secretary on Electricity Markets and Reliability

(2017) .......................................................................28

1

STATEMENT OF INTEREST1

Amici are among the nation’s leading engineers

with expertise in the operation, structure, economics,

and reliability of the U.S. power system. They have expertise in grid structure, operations, economics, and

modernization; integration of renewable energy generation; and power-system reliability and planning.

Amici have a significant interest in the efficient functioning and regulation of the grid. To aid the Court’s

understanding of the technical matters at issue in this

case, this brief clarifies how and why the grids are designed and operated as they are; the implications of the

grids’ unique features for pollution controls; and how

pollution controls generally interact with grid operations and the industry.

Benjamin F. Hobbs is the Theodore M. and Kay

W. Schad Professor in Environmental Management in

the Department of Environmental Health and Engineering at Johns Hopkins University (JHU). He has

a joint appointment in the Department of Applied

Mathematics and Statistics, and codirects the YaleJohns Hopkins Solutions for Energy, Air, Climate &

Health Center. He is also on the Leadership Council of

the JHU O’Connor Sustainable Energy Institute. His

research focuses on electric power and energy market

planning, risk analysis, and environmental and energy

systems analysis and economics. He is Chair of the

1

No counsel for a party authored this brief in whole or in

part, and no person other than amici made a monetary contribution intended to fund the preparation or submission of this brief.

The parties have all consented to the filing of this brief.

2

California Independent System Operator Market Surveillance Committee and a Fellow at the Institute of

Electrical and Electronics Engineers and the Institute

of Operations Research and Management Science. He

was also a consultant to the PJM Independent System

Operator and developed the methodology it uses to create its capacity market demand curve. From 1995 to

2002, he was consultant to the Federal Energy Regulatory Commission’s Office of the Economic Advisor. He

holds a Ph.D. in Civil and Environmental Engineering

from Cornell University.

Brendan Kirby is a private consultant with clients including the Hawaii Public Utilities Commission,

Grid Lab, National Renewable Energy Laboratory, Energy Systems Integration Group, Electric Power Research Institute, American Wind Energy Association,

Oak Ridge National Laboratory, and others. He has

forty-seven years of electric grid experience, and has

published over 180 papers, articles, book chapters, and

reports on power system reliability and integrating

renewable energy generation into the power grid. He

was a member of the North American Electric Reliability Corporation’s Essential Reliability Services Task

Force, and previously served on its Standards Committee. He retired from the Oak Ridge National Laboratory’s Power Systems Research Program. He is a

Licensed Professional Engineer with an M.S. degree in

Electrical Engineering (Power Option) from CarnegieMellon University and a B.S. in Electrical Engineering

from Lehigh University.

3

Kenneth J. Lutz is an Affiliated Professor in the

Department of Electrical and Computer Engineering

at University of Delaware, where he does research and

teaches specially designed courses on the modernization of the electric grid. He has decades of experience

in the regulation of utilities. He founded AMR Strategies, LLC, to help utilities modernize their grids. Previously, he served as an IEEE/American Association

for the Advancement of Science Congressional Fellow

for United States Senator Ron Wyden, where he played

a key role in drafting federal legislation for renewable

energy and energy efficiency. He has a Ph.D. in electrical engineering from the Johns Hopkins University

and a B.E.E. from the University of Delaware.

James D. McCalley is an Anson Marston Distinguished Professor and the London Chaired Professor

of Power System Engineering in the Electrical and

Computer Engineering Department at Iowa State University. He has graduated thirty-five Ph.D. students

under his supervision and is the author of over 280

publications in electric power systems engineering.

His areas of research include: transmission planning,

power-system security, power-system dynamics, wind

energy, long-term investment planning for energy

and transportation systems at the national level, and

power-system decision problems under uncertainty, including those encountered in operations and planning.

Dr. McCalley has been an IEEE Fellow since 2004. He

chaired the IEEE Power and Energy Society’s Subcommittee on Risk, Reliability, and Probability Applications from 2004 to 2006. He has been involved in the

4

International Conference on Probabilistic Methods Applied to Power Systems (PMAPS) since PMAPS-4 in

1994 and served as General Chair of PMAPS-8. Prior

to joining the Iowa State University faculty, from 1985

to 1990, he was a Transmission Planning Engineer

with Pacific Gas and Electric Company in San Francisco, California, and a licensed professional engineer.

He holds Ph.D., M.S., and B.S. degrees in electrical engineering from the Georgia Institute of Technology.

---------------------------------♦---------------------------------

SUMMARY OF ARGUMENT

Effective air pollution controls for the U.S. power

sector work with the interconnected design and operation of the U.S. electric grids. When regulating carbon

dioxide, just as when regulating any other air pollutant in this sector, the U.S. Environmental Protection

Agency (“EPA”) must give appropriate consideration to

grid design and operation. Fundamental characteristics of the electric grids mean that in response to any

air pollution control measure, operators may shift

power generation from higher-emitting to lower-emitting sources, a result known as “generation shifting.”

Generation shifting is an ordinary consequence of

workaday pollution-control rules.

Recognizing this, regulators designing the Clean

Power Plan went a step further: They calibrated that

rule’s emission limitations in reliance on an achievable, and significant, degree of generation shifting. A

key issue in this case is whether the Clean Air Act

5

(“Act”) authorizes regulators to take that approach.

However that legal question is resolved, grid engineers

and operators understand that any pollution control

measure may lead to generation shifts, for reasons described below. Moreover, pollution control measures for

carbon dioxide (“CO2”), as for other air pollutants, are

incorporated easily into grid workings and change

nothing about the fundamentals of the sector.

Engineers have declared the U.S. power system as

the largest, “most complex machine ever made.” Phillip

F. Schewe, The Grid: A Journey Through the Heart of

Our Electrified World 1 (2007); see also Mass. Inst. of

Tech., The Future of the Electric Grid 1 (2011). Every

electric generator in the continental United States is

embedded within one of three regional grids and linked

to other generators and consumers through transmission and distribution lines. Each grid operates as a single

integrated machine. The fundamental purpose of each

machine’s interconnectedness is to allow grid operators

to continuously balance electricity supply and demand in

real time, over vast regions, thus ensuring all consumers

access to affordable and reliable power. This feat is accomplished through orchestrated, second-by-second shifts

among different generators, facilitated by the grids’

physical structure and design and by complex dispatch

software and regional spot electricity markets. The use

of any individual generator is thus dependent on the

performance of other components of the machine.

Amici emphasize three key points:

First, effective power-sector emission controls reflect grid operations, which are defined

6

by fundamental characteristics of electricity

and of the infrastructure and markets that connect and coordinate power generation and demand. The power sector has distinctive operational

features that create both opportunities and challenges

for pollution control, and EPA must account for these

features in determining the Best System of Emission

Reduction (“BSER”) under Section 111(d) of the Act.

For example, a defining feature of the three regional

grids is that each operates as a single, interconnected

machine. Governance frameworks for the dispatch of

electricity are designed to facilitate seamless shifts

among generators to ensure affordable, reliable electricity. For these reasons, the most effective and least

costly CO2 pollution control measures for the power

sector allow for shifting of generation to lower-emitting

generators. The approach taken in the Clean Power

Plan (“CPP”) accomplished this by including shifts

from higher-emitting to lower-emitting generators as

part of its definition of the Best System of Emission

Reduction. See 80 Fed. Reg. 64,662, 64,717 (Oct. 23,

2015), J.A. 273–1445, rescinded by “Repeal of the Clean

Power Plan; Emission Guidelines for Greenhouse Gas

Emissions From Existing Electric Utility Generating

Units; Revisions to Emission Guidelines Implementing

Regulations,” 84 Fed. Reg. 32,520 (Jul. 8, 2019), J.A.

1725–2030 (“Rule” or “ACE Rule”). In other words, EPA

relied on generation shifting in defining the BSER and

in setting the Clean Power Plan’s required degree of

emission-reduction stringency.

7

Second, any air pollution standard applied

to the power sector may induce generationshifting effects, even if it is not designed—as

the Clean Power Plan was—in reliance on such

shifts. All power-sector environmental regulation may

affect operating costs or constrain operation of regulated entities. Given the interconnected nature of the

power grid, this may cause dispatch to shift to units

whose relative costs decrease. Regulators have long

incorporated this feature of grid operations into the

design of pollution controls to minimize costs of compliance. Generation shifting is an ordinary consequence of all power-sector pollution controls. Moreover,

industry experts understand that the easiest, cheapest, and best way to reduce CO2 emissions from coalfired power plants is to shift generation away from

those plants and toward cleaner sources of energy.

Third, pollution control measures blend

seamlessly into power-sector operations and

change nothing about the fundamentals of the

sector. Pollution control measures are business-asusual for this industry and are incorporated easily into

grid workings and dispatch. This is as true for CO2 as

for other air pollutants: Regulating CO2 pollution from

the power sector reduces harms to human health

without significantly affecting grid operations or

risking reliability. Because renewable sources of power

are now cheaper than or cost-competitive with fossil

fuel generation, regulation of power-sector CO2 emissions builds on existing energy-sector trends in a way

8

that reinforces, rather than disrupts, longstanding industry practices.

---------------------------------♦---------------------------------

ARGUMENT

I.

Effective Power-Sector Pollution Controls

Work with the Distinctive Characteristics

of Electricity and the Interconnectedness

of the Regional Grids.

The fungible nature of electricity and the need to

instantaneously and continuously balance supply and

demand in real time have driven the design of the

world’s most “complex machine”—the U.S. power system. Schewe at 1. Every generator in the continental

United States is embedded within one of three regional, interconnected electric grids. To ensure that

consumers receive reliable, affordable power that

meets environmental standards, each grid is designed

and operated specifically to facilitate, within its respective region, shifts among different generators. Shifting

among generators is both unique to the power sector

and an essential, routine feature of grid operations.

Regulators have long harnessed these shifts as an efficient tool to reduce power-sector air pollution.

A. Electricity Is a Uniquely Fungible and

“Real-Time” Good.

Electricity has two fundamental distinguishing

features. First, electricity is fungible. In the continental United States, “any electricity that enters the grid

9

immediately becomes a part of a vast pool of energy

that is constantly moving in interstate commerce.”

New York v. Fed. Energy Regulatory Comm’n, 535 U.S.

1, 7 (2002). Electricity moves across the grid according

to the laws of physics, following the path of least resistance. It cannot be directed (like an e-mail or package) to a particular recipient.

Second-by-second variation in withdrawals of electricity (demand) is balanced by injections of electricity

from generators connected to the grid (supply), by responding to the frequency variation that those imbalances cause. The frequency is analogous to the water

level in a swimming pool fed by many spigots located

around the pool’s edges. When the water level (frequency) increases, the water supply (generation) decreases, and vice versa. All spigots have the same effect

on maintaining a constant water level, independent of

their location around the pool (grid). For example, “If

[someone] in Atlanta on the Georgia system turns on a

light, every generator on Florida’s system almost instantly is caused to produce some quantity of additional electric energy which serves to maintain the

balance in the interconnected system.” Fed. Power

Comm’n v. Fla. Power & Light Co., 404 U.S. 453, 460

(1972) (citation omitted).

Electricity that is added to the grid energizes the

entire grid. Generators do not “generate” electrons and

consumers do not “consume” electrons, as is commonly

believed—electric power is injected into and withdrawn from the grid. An electromagnetic wave, propagated by generators, moves at the speed of light along

10

wires. Electrons in an alternating current network

merely move back and forth at a frequency of sixty

cycles per second. Because all electricity within a

grid is pooled, the electric power added by any single

generator becomes part of this undifferentiated supply.

As with water added to a pool, consumers cannot distinguish coal-generated power from solar-generated

power once it is injected into the grid.

The second distinctive feature of electricity is that

it is only beginning to be able to be stored economically

on a large scale. The present difficulty of storing large

amounts of electricity means generation (supply) and

load (demand) must continuously and precisely be balanced. This makes electricity the ultimate “just-intime” product. See Paul L. Joskow, Creating a Smarter

U.S. Electricity Grid, 26 J. Econ. Persp. 29, 33 (2012);

but see Energy Info. Admin., U.S. Battery Storage Market Trends 4 (2018) (noting rapid advances in energy

storage technology that may someday overcome this

hurdle). As battery technology advances and costs decline, utilities are gaining greater experience with energy storage, enabling higher penetrations of wind and

solar generation and enhancing reliable operations

with reduced fossil-fueled generation.

B. Each of the Three Regional Grids Operates as a Single Machine.

The infrastructure necessary to balance supply

and demand distinguishes the power system from any

other industry or supply chain. Its defining feature is

11

interconnection. Each of the three regional grids, or

“interconnections”—Eastern, Western, and Texas—operates as a single, synchronized machine.2

2

Hawaii and Alaska have their own grids.

12

Figure 1. U.S. Power-System Interconnections3

3

U.S. Dep’t of Energy, North American Electric Reliability

Corporation Interconnections, https://www.energy.gov/oe/downloads/

north-american-electric-reliability-corporation-interconnections.

13

Each of the grids consists of three components essential to delivering reliable and cost-effective power

to consumers: generation, transmission, and distribution. First, a diverse set of generators converts primary

energy (such as coal, sunlight, or wind) into electricity.

Second, within each grid, a giant network of high-voltage transmission lines allows power to flow where it is

needed, sometimes over hundreds or even thousands

of miles. The transmission network is crucial because

many generators are located far from population centers. The transmission network also facilitates system

reliability: If one line goes down, electricity can flow

through alternate routes; when a generator fails, other

generators can pick up the load smoothly without a

power interruption. Third, local substations receive

electricity from high-voltage transmission lines and

lower the voltage for delivery to consumers via local

distribution networks.

Grid interconnectedness is a product of history.

The first power plants constructed in the late 1800s initially served only a small set of local customers.

Backup generators maintained reliability. Local systems gradually consolidated to reduce costs and improve reliability. Consolidation required local systems

to become interconnected through transmission lines.

Networks continued to grow, ultimately giving rise to

the three interconnections. 80 Fed. Reg. at 64,690–92,

J.A. 401–13.

Today, each of the three interconnections is highly

coordinated to maintain reliability. The balancing of

generation and load must be virtually instantaneous

14

across each interconnection, such that the amount of

power dispatched to the grid is identical to the amount

withdrawn for end uses in real time. Like orchestra

conductors signaling entrances and cut-offs, grid operators use automated systems to signal particular generators to dispatch more or less power to the grid as

needed over the course of the day, thus ensuring that

power pooled on the grid rises and falls to meet changing demand.

As components of an integrated machine, each

generator is interdependent with every other generator, and routine operations are coordinated by grid operators. Because the performance and usage of their

units depends on the operation of other units outside

their control, power companies regularly coordinate

with each other to plan new investments, plan unit retirements, and balance their respective systems—for

example, through joint dispatch arrangements (which

pool the generation sources of multiple utilities to reduce operating costs and increase reliability), joint

power-plant ownership agreements, bilateral power

purchase agreements, and short-term balancing transactions. As this Court has recognized, “generating facilities cannot be maintained on the basis of a constant

demand.” Gainesville Util. Dep’t v. Fla. Power Corp.,

402 U.S. 515, 518 (1971). Coordinated planning is critical to ensure there is always adequate generation to

meet expected regional demand, plus additional capacity in case generators fail during times of peak demand. Id.

15

C. Dispatch Governance Frameworks Are

Designed to Facilitate Shifts Among

Generators and Ensure Affordable, Reliable Electricity.

Regional energy governance frameworks keep the

“complex machine” operating reliably. Although governance differs within and across the three interconnections, the standard approach all grid operators use

to dispatch generation is called “Security Constrained

Unit Commitment and Economic Least-Cost Dispatch”

(hereinafter “Constrained Least-Cost Dispatch” for

brevity). As its name implies, Constrained Least-Cost

Dispatch deploys generators with the lowest variable

costs first, as system operational limits allow, until all

demand is satisfied. Constraints that grid operators

routinely consider include transmission limits, generators’ physical constraints, and environmental standards.

In competitive wholesale markets, which govern

about two-thirds of the power sector, federally regulated

entities called Independent System Operators (“ISOs”)

or Regional Transmission Organizations (“RTOs”) use

a series of auctions to match generation and load. Generators bid into a regional market with a price at which

they are willing to sell electricity during specified periods, and the ISO/RTO ranks bids according to Constrained Least-Cost Dispatch principles. In traditional

cost-of-service states outside of ISOs/RTOs, utilities use

generators’ marginal costs, rather than bid prices, to

determine dispatch order. In these ways, Constrained

16

Least-Cost Dispatch principles guide all dispatch planning across the country.

Dispatch and the necessary planning for it occur

on multiple scales—yearly, seasonally, monthly, weekly,

daily, hourly, and five-minute intervals—as grid operators respond to variable supply, demand, and operational constraints by managing shifts among different

generators. In both organized markets and traditional

cost-of-service regimes, renewable energy generators

typically receive dispatch priority because they have

lower variable costs than fossil-fuel-fired generators,

which must purchase fuel. 80 Fed. Reg. at 64,693, J.A.

413–15.

Power companies recognize that their units are

subject to Constrained Least-Cost Dispatch and have

long planned their operations and investments accordingly. Power companies routinely execute contracts to

purchase power from third-party generators; invest in

demand-side energy efficiency programs; invest in battery storage facilities; and, as existing units retire, invest in more efficient and cost-competitive generation

facilities, such as natural gas and renewable sources,

to compete for dispatch priority.

II.

Power Companies and Grid Operators

Have Historically Responded to Air Pollution Controls by Shifting to Lower-Emitting Generators.

Because electricity is a unique good that requires

each regional grid to operate synchronously, the power

17

sector is designed to allow constant, real-time shifting

among generators to maintain a balanced grid. These

distinctive characteristics of the power grid mean that

pollution controls for the power sector can easily lead

to shifts to lower-emitting sources of power.

Among pollution control measures that result in

such generation shifts, an important distinction exists

between those that affect dispatch order only incidentally, by affecting relative costs—which nearly all

pollution control measures do—and the Clean Power

Plan, which determined its degree of emission limitation in reliance on generation shifting, built into its

definition of the Best System of Emission Reduction.

A. Air Pollution Control Measures Ordinarily Affect Dispatch Order, Causing

Generation Shifting.

All power-sector environmental regulations impact dispatch, either by increasing or decreasing the

relative operating costs of affected sources or by constraining their operations. Because grid operators in

both organized markets and traditional cost-of-service

regimes employ Constrained Least-Cost Dispatch

principles, a unit that experiences a cost increase or

operational constraint will tend to operate less often,

while units whose costs decrease will be dispatched

more. Existing pollution regulations already affect

the dispatch competitiveness of fossil-fuel-fired power

plants. Under Constrained Least-Cost Dispatch, fuel

costs and other costs are treated identically; the

18

cheapest overall generation, once variable costs are accounted for, is used.

Effective air pollution controls account for and utilize this tendency toward generation shifting. Congress, EPA, and state regulators have long recognized

that a systemwide approach to reducing pollution

works most efficiently with grid operations. They have

accordingly harnessed shifts among generators as an

economical tool to reduce harmful air emissions.

One highly-regarded example of a pollution control program that resulted in generation shifting is the

Clean Air Act’s Acid Rain Program, which set a nationwide cap on sulfur dioxide emissions from fossil-fuelfired generators and required affected generators to

hold a tradable allowance for each ton of sulfur dioxide

emitted. 42 U.S.C. §§ 7651–7651o; see also Emanuele

Massetti et al., Oak Ridge Nat’l Lab., Environmental

Quality and the U.S. Power Sector: Air Quality, Water

Quality, Land Use and Environmental Justice 20

(2017). The allowance requirement increased the costs

of regulated units, which decreased the dispatch competitiveness of those units and led some to reduce their

generation. That, in turn, led grid operators to dispatch

cheaper, less-polluting generators to meet consumer

demand. Industry quickly recognized that incorporating allowance costs into dispatch planning was costeffective and did not disrupt power reliability or normal grid operations. See Thomas M. Jackson et al.,

Evaluating Soft Strategies for Clean-Air Compliance,

6 IEEE, Computer Applications in Power 46 (1993).

19

Industry was thus empowered to achieve the program’s emissions targets, in part, through generation

shifting.

The Regional Greenhouse Gas Initiative (“RGGI”)

provides an example of these dynamics at work to control carbon dioxide. RGGI is a cap-and-trade program

for power-sector CO2 pollution in eleven northeast and

mid-Atlantic states. The participating states span

three ISOs/RTOs, all of which have been able to integrate the price of carbon allowances into their dispatch

methods with ease. Affected sources simply incorporate the cost of carbon allowances into their auction

bids. This generally prompts grid operators to deploy

lower-cost sources, such as renewable sources, first.

See, e.g., Paul Hibbard et al., The Economic Impacts of

the Regional Greenhouse Gas Initiative on Nine Northeast and Mid-Atlantic States 6 (2018).

In these varied ways, air pollution regulators have

long crafted emission programs that leverage the integrated operations of the grid to lower compliance costs,

and that induce a degree of generation shifting in response to pollution control measures.

B. The Clean Power Plan Went Further by

Embedding Generation Shifting in its

Definition of the Best System of Emission Reduction.

While successful pollution control programs have

often induced generation shifting to reduce emissions

from regulated sources, the CPP took the next logical

20

step: It identified the emissions reductions that could

be achieved by regulated sources in each state based,

in part, on gains from generation shifting. It did this

by defining the Best System of Emission Reduction to

include reductions in coal generation and increasing

natural gas and renewable energy generation, relying

on the interconnected workings of the grids. See 80

Fed. Reg. 64,662, 64,717, J.A. 273, 529–30. In other

words, it built an assumption of generation-shifting

into its calculation of achievable—and required—

gains.

The result was a flexible, cost-effective regulation

with lower emission limits than might otherwise have

been set. The approach embraced by the CPP leverages

the grids’ interconnected, synchronous operation to allow for meaningful—and very cost-effective—cuts in

emissions from the sources it regulated.4 See ICF Int’l,

Inc., Assessing Effects on the Power Sector of Greenhouse Gas Emission Standards at 3, 5, 7 (Oct. 31, 2018)

(showing that a regulatory design relying on generation shifting, emissions trading, and reduced utilization would cause an additional 18–27% reduction in

CO2 emissions by 2030, at low per-ton abatement

cost). By recognizing that the grid operates as a single

4

This is not to say that the CPP mandated specific forms of

compliance. On the contrary, the rule was designed to promote

state flexibility in meeting its standards. Compliance options to

meet state emission-reduction targets were plentiful and were

well-matched to grid operations. Regulated sources could reduce

emissions through a mix of generation shifting, reduced utilization, emissions trading, heat rate improvements, and other

measures. See 80 Fed. Reg. at 64,666–67, J.A. 297–300.

21

integrated machine and by encouraging generation

shifting rather than costly source-specific reductions,

the CPP capitalized upon the unique features of the

grid to reduce emissions.

By contrast, the ACE Rule excluded emissionreduction measures that take advantage of grid operations and interconnectedness, such as generation

shifting. In defining the Best System of Emission Reduction in the Rule, EPA looked only to certain

measures that can be “applied at and to” individual

coal-fired units and settled on a definition that included only certain changes to the physical equipment

and processes of generators, specifically, heat-rate improvements at coal-fired power plants. Rule at 32,532,

32,536, J.A. 1787, 1802–03. The Rule excluded many of

the pollution-control measures that experts know to

be the most effective at reducing emissions from coal

unit operations, rejecting, for example, natural-gas

co-firing, decreased utilization of the highest-emitting

units, and generation shifting. Id. at 32,532, 32,543,

J.A. 1787, 1836–37. In this way, the Rule failed to enable the emissions reductions possible by leveraging

grid interconnectivity against CO2 pollution.

Moreover, the ACE Rule adopted only those

measures that reduce an individual facility’s rate of

emissions—that is, its emission of CO2 per unit of electricity produced—and failed to credit measures that

would reduce a facility’s total emissions without necessarily affecting rate, such as reduced utilization. Rule

at 32,555, J.A. 1890–92. Notably, a facility that improves its rate of emissions can still increase its total

22

emissions in aggregate, simply by operating more. Id.

at 32,542–43, J.A. 1833–36.

The consequence of the overly-narrow approach

adopted in the ACE Rule—which excluded even some

measures that can be accomplished on-site, like reduced utilization—was a rule that barely moved the

needle on CO2 pollution and that left health benefits

worth billions of dollars unrealized. See EPA, Regulatory Impact Analysis ES-6 (2019) (“Final RIA”) (projecting that the ACE Rule would result in less than a

percentage point in additional CO2 emissions reductions by 2030); see ICF Int’l, Inc. at 6 fig. 6 (showing

that a CPP-style regulatory approach would achieve

an order of magnitude greater emissions reductions

than the reductions projected under the ACE rule); see

83 Fed. Reg. 44,746, 44,790 tbl. 14 (Aug. 31, 2018) (conceding that repealing the CPP in favor of the ACE Rule

would significantly increase co-pollutants and inflict

several billion dollars in health damages).

C. To Reduce Power-Sector CO2 At Lower

Cost, Regulated Entities and States

Should Have The Flexibility to Employ

Off-Site Measures.

Because the power sector responds to pollution

controls with dispatch shifts regardless of rule design

whenever those controls alter the relative costs of

sources (as they almost always do), excluding generation shifting from power-sector regulation does not

mean that shifting will not occur; it simply won’t be

23

captured and used by regulators to craft a cheaper,

more effective rule. For these reasons, a “best” system

of emission reduction for this sector will often be one

that allows for measures such as generation shifting.

But even if this Court concludes that the Act does

not permit EPA to rely on generation shifting in defining the BSER here—an outcome we would strongly

disfavor—it should segregate that determination from

questions of allowable compliance with CO2 pollutioncontrol measures. To reduce CO2 from the power sector,

it would be nonsensical to limit power-plant operators

to using only site-constrained approaches, when grids

operate as integrated machines.5 Excluding generation

shifting from CO2 pollution control regimes would result in more expensive, less effective, and less flexible

regulation.

A hypothetical illustrates why. Consider a coalfired power plant (“Plant-A”) that is subject to a CO2

pollution control measure and that installs solar panels as part of its facility. By generating power with both

5

Such an approach would be reminiscent of a previously rejected approach to controlling sulfur dioxide from power plants.

In the debates over the 1990 Clean Air Act amendments, some

had suggested that only site-specific “scrubbers” be used to control SO2, in lieu of the Acid Rain Program’s more flexible approach

that allows for substituting lower-emitting units for higher-emitting units. History has since shown that the more flexible approach is a superior way to control pollution without endangering

reliability. See Paul L. Joskow et al., The Market for Sulfur Dioxide Emissions, 4 Am. Econ. Rev. 669, 683 (1998). The rejected sitespecific approach would have been significantly less effective and

more expensive. See id. at 669–70.

24

its solar panels and coal-fired boiler, Plant-A can lower

its CO2 emissions rate (emissions per megawatt-hour).

Plant-A can continue to produce the same amount of

power by shifting some of its generation from coal to

solar, thereby reducing the numerator of its emissions

rate. Or, Plant-A can increase its annual output by

adding solar to its coal generation, thereby increasing

the emissions-rate denominator. In either case, PlantA has installed a compliance mechanism that “can be

applied at and to a stationary source (i.e., as opposed

to off-site measures)” and “lead[s] to continuous emission reductions.” Rule at 32,534, J.A. 1796.

Now, imagine that Plant-A instead installs solar

panels on a field located next to its coal unit. The emissions rate result is the same. Likewise, the same emissions rate would result from solar panels instead

installed several miles away. Regardless of where the

solar panels are located, Plant-A would rely on the

same regional network of transmission lines to pool

power generated by the solar panels on the grid. From

the perspective of regulators, consumers, grid operators, and EPA, it is irrelevant whether the solar panels

that reduce Plant-A’s emission rate are located on

Plant-A’s rooftop or in the next state over. From the

perspective of Plant-A’s owner, it is far more desirable

to install solar panels in the most cost-effective location, whether or not that location is within the plant.

In promulgating the ACE Rule, EPA took account

of none of these possible approaches, the consequence

of which was to leave low-hanging emission reduction fruit unharvested. Furthermore, none of these

25

approaches was allowable as a compliance method under the Rule (save perhaps the on-site panels, about

which the Rule is ambiguous). Rule at 32,555, J.A.

1890–92. The effect of this was to reduce industry’s

flexibility in choosing low-cost emission reduction

strategies.

No coal-fired unit operates by itself. Each is a piece

of a power plant that, in turn, is part of the grid. It

would be unreasonable and unduly costly to exclude

generation shifting from CO2 pollution-control strategies when regulated units, like all generators, are

part of one big machine that delivers undifferentiated

power to a unitary grid.

III. CO2 Pollution Control Measures Are Incorporated Easily Into Power-Sector Operations.

Regulating CO2 pollution from the power sector alleviates harms to human health without harming grid

operations or reliability. Such regulation builds on existing trends and reinforces, rather than disrupts,

longstanding industry practices.

A. U.S. Power Sector Is Shifting Toward

Cleaner Energy Sources.

The U.S. power sector is shifting from coal-fired

plants to lower- and zero-carbon sources like natural

gas, wind, and solar. Successful regulation and market

forces have driven large reductions in power-sector

CO2 emissions. In 2017 and 2018, the U.S. power sector

26

emitted 30% less CO2 compared to 2005 levels. Final

RIA at 2–14.

In particular, the power sector has shifted generation away from coal and will continue to do so. By 2025,

the average age of coal-fired units is projected to be 49

years old, with 20% of units older than 60—well beyond their expected operating life of 40 years. See 80

Fed. Reg. at 64,694, 64,872, J.A. 352, 1248; see also Rule

at 32,548 n.215, J.A. 1859 n.215. As coal plants age,

they become more expensive compared to newer units.

Final RIA at 2–7. Natural gas prices, meanwhile, are

low because of abundant supply, while renewable energy costs have plunged because of improving technology and government policy incentives. Id. at 2–11. The

falling price of renewable energy has been particularly

dramatic: The cost of building and operating new solar

and wind projects is now cheaper than the cost of continuing to operate many coal-fired units. Id. These

market forces are pushing coal-fired plants offline, replacing them with cleaner energy resources. See id. at

2–7. Increasingly, as battery costs decline, utilities are

deploying batteries in combination with wind and solar generation, which can provide so-called firm power

to balance the grid and replace the need for certain

gas-fired plants.

Strikingly, the Clean Power Plan’s goal of reducing

carbon dioxide emissions to 32% below 2005 levels by

2030 had already been exceeded by the end of 2019—

notwithstanding that the CPP never went into effect.

See Environmental Integrity Project, Greenhouse Gases

from Power Plants 2005-2020: Rapid Decline Exceeded

27

Goals of EPA Clean Power Plan (2021). Much of this

decrease is attributable to the country’s shift to renewable energy and natural gas rather than coal as a

source of electricity. See id. at 3. That the sector has

moved as quickly as it has toward low-emission and

renewable energy sources, even without the CPP’s implementation, shows the modest influence of CO2 pollution control measures in comparison to other sectoral

forces at play.

Yet despite progress made to date, the U.S. power

sector remains a significant source of CO2 emissions

endangering public health and welfare. In 2018, it

emitted more than a quarter of total annual U.S. greenhouse gas emissions. See EPA, Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2019 ES-5 fig.

ES-1, 2–3 tbl. 2–1 (2021). And progress limiting U.S.

power-sector emissions may be slowing or halting: Annual emissions in 2018 increased by 1.2%. EPA, Inventory of U.S. Greenhouse Gas Emissions at ES-7, tbl. ES2. It is therefore sensible and consistent with industry

trends and operations for EPA to aim to reduce CO2

pollutants from regulated sources by building on the

last decade’s power-sector shifts. The easiest, cheapest,

and least disruptive method for reducing emissions

from coal-fired power plants is to continue the shift in

generation away from those plants toward cleaner

sources.

28

B. Propping Up Coal Is Unnecessary for

Grid Reliability.

In particular, generation shifting away from coal

would not harm grid reliability. Despite a large number of coal retirements in recent years, grid-wide indicators for reliability have been “adequate for all

interconnections and generally trending in a positive

direction.” N. Am. Elec. Reliability Corp., 2021 LongTerm Reliability Assessment 36 (2021). Even in those

limited areas where anticipated capacity reserves

begin to fall after 2024, see id. at 12, the changing energy mix requires new and flexible grid operation

strategies to promote reliability and meet operational

needs, not artificial lifelines for coal-fired resources.

See id. at 9–10, 22.

The Department of Energy has found that, despite

coal retirements, “markets have achieved reliable

wholesale electricity delivery.” U.S. Dep’t of Energy,

Staff Report to the Secretary on Electricity Markets and

Reliability 10 (2017) (“Staff Report”). Independent

market analysis has also found that “the diverse US

power supply portfolio has proven resilient to significant deviations from normal operating conditions.”

IHS Markit, Ensuring Resilient and Efficient Electricity Generation 4 (2017). Coal-fired power, furthermore,

is no cure-all for reliability concerns. For example, the

2014 Polar Vortex froze coal piles solid, leaving many

coal plants inoperable during a surge in energy demand. Staff Report at 98.

29

In fact, renewable sources can help improve reliability in some circumstances. Wind and solar can provide stability to the system by quickly detecting

frequency deviations and responding to system imbalances faster than conventional generators, decreasing

the need for inertia, the tendency for conventional generators and motors to continue spinning during power

failure. See, e.g., Nat’l Renewable Energy Lab., Inertia

and the Power Grid: A Guide Without the Spin v-vi

(2020). For instance, wind generation was key in maintaining service in the northeast and mid-Atlantic during the 2014 Polar Vortex, when demand spiked to one

of the highest winter peaks in regional history. Analysis Grp., Electric System Reliability and EPA’s Clean

Power Plan: The Case of PJM 3, 12 (2015).

It is true that the availability of renewable energy is more variable than other types of generation,

leading system operators to maintain generation reserves that provide back-up when renewable energy is

unavailable. The U.S. power sector has successfully

managed large amounts of renewable power in this

manner, and technical studies have concluded the sector is capable of integrating even more without significant reliability impacts. See, e.g., Nat’l Renewable

Energy Lab., Eastern Renewable Generation Integration Study xvii (2016) (concluding that the U.S. Eastern Interconnection can accommodate upwards of 30%

wind and solar photovoltaic generation); Elec. Reliability Council of Tex., 2018 State of the Grid 2, 4 (2018)

(reporting Texas’s electricity grid was “operating effectively and efficiently” with about 19% energy provided

30

by wind sources); GE Energy Consulting, PJM Renewable Integration Study: Executive Summary Report 6–

7 (2014) (finding that the RTO PJM could operate with

up to 30% of generation from wind and solar with no

significant harm to reliability). By contrast, we know

of no good evidence to support the idea that propping

up coal generation is necessary for grid reliability.

---------------------------------♦---------------------------------

CONCLUSION

The judgment of the U.S. Court of Appeals for the

District of Columbia Circuit should be affirmed.

Respectfully submitted,

Of Counsel:

WILLIAM BOYD

UCLA SCHOOL OF LAW

405 Hilgard Avenue

Los Angeles, CA 90095

(310) 206-5280

boyd@law.ucla.edu

CARA A. HOROWITZ

Counsel of Record

ANDRIA SO

FRANK G. WELLS

ENVIRONMENTAL LAW CLINIC

UCLA SCHOOL OF LAW

405 Hilgard Ave.

Los Angeles, CA 90095

(310) 206-4033

horowitz.elc@law.ucla.edu

Counsel for Amici Curiae

January 25, 2022