1. business and industrial
2. energy
3. natural gas

Comments of the American Public Power
Association (APPA) on EPA’s Section
111(d) Proposed Rule for Carbon Dioxide
Emissions from Existing EGUs EPA-HQOAR-2013-0602
December 1, 2014
Submitted by:
James J. Nipper
Senior Vice President, Regulatory Affairs and
Communications
American Public Power Association
2451 Crystal Drive, Suite 1000
Arlington, VA 22202-4804
202-467-2931
[email protected]
Desmarie M. Waterhouse
Director of Government Relations and Counsel
American Public Power Association
2451 Crystal Drive, Suite 1000
Arlington, VA 22202-4804
202-467-2930
[email protected]
Theresa Pugh
Director of Environmental Services
American Public Power Association
2451 Crystal Drive, Suite 1000
Arlington, VA 22202-4804
202-467-2943
[email protected]
Alex Hoffman
Energy and Environmental Services Manager
American Public Power Association
2451 Crystal Drive, Suite 1000
Arlington, VA 22202-4804
202-467-2956
[email protected]
Contents
I. Executive Summary................................................................................................................. 7
II.
Introduction........................................................................................................................ 10
III.
The Proposed Rule Denies States the Primacy and Implementation Authority Accorded by
Congress........................................................................................................................................ 13
A.
EPA’s Proposed Rule Unlawfully Constrains State Primacy Under Section 111(d)..... 14
B.
EPA’s Discussion of State Implementation Issues Reveals Significant Problems with
the Proposed Rule...................................................................................................................... 16
1. The Proposed Rule Does Not Provide States with “Flexibility.” ................................... 16
2. The Proposed Rule Encroaches on Areas of Exclusive State and Local Government
Authority................................................................................................................................ 17
3. The Option of a Rate-Based or Mass-Based Goal ......................................................... 18
4. The Four Types of State Plans ....................................................................................... 19
5. Timing ............................................................................................................................ 20
6. Criteria to Approve State Plans ...................................................................................... 20
7. Components of Approvable Plans.................................................................................. 21
8. Deadlines and Process for State Plan Submittal............................................................. 21
9. EPA’s “Key Considerations” That States Must Address in Developing Their Plans .... 22
IV. The Proposed Rule Conflicts with Federal, State, and Local Utility Laws and Disregards
How Electric Markets Work. ........................................................................................................ 24
A.
The Proposed Rule Conflicts with the Federal Power Act’s Division of Regulatory
Authority between Federal, State, and Local Governments...................................................... 25
1. The Proposed Rule Unlawfully Usurps State Authority Preserved by the FPA and the
Tenth Amendment. ................................................................................................................ 25
2. The Proposed Rule Unlawfully Usurps FERC’s Regulatory Authority Under the FPA. ..
........................................................................................................................................ 28
B.
The Proposed Rule Relies on a Flawed Understanding of Regulated Wholesale
Electricity Markets and the Bulk Power System. ...................................................................... 33
1. EPA Has Failed to Address Reliability Issues. .............................................................. 34
2. EPA Misunderstands the Role of the States in Regulating Dispatch in RTO Regions. . 34
3. EPA Has Failed to Take into Account Impediments to Deploying the New Energy
Infrastructure That the Proposed Rule Would Necessitate.................................................... 35
4. EPA Has Not Accounted for Recent Developments Regarding the Participation of
Demand-Side Resources in the Electricity Markets. ............................................................. 36
5. The Proposed Rule Is Based on Fundamental Misunderstandings of RTO Capacity
Markets and Their Potential to Facilitate EPA’s Goals......................................................... 37
V.
New Source Review (NSR) Issues .................................................................................... 38
1
VI.
The Proposed Rule Contains Many Inequities and Is Unfair in Many Key Respects. ...... 39
A.
B.
C.
D.
E.
State Goal Computation ................................................................................................. 40
Early Action Credit ........................................................................................................ 40
Transmission Lines and Natural Gas Pipelines .............................................................. 41
Interaction with Other Clean Air Act Rules................................................................... 41
Public Health Benefits.................................................................................................... 42
VII. EPA’s Premise That a Significant Portion of the CO 2 Reductions the Proposed Rule Seeks
to Achieve Can Be Done Through Fuel Switching from Coal to Natural Gas Is Based on
Questionable Assumptions Regarding Natural Gas Supply, Price, and Infrastructure Availability.
............................................................................................................................................ 42
A.
EPA Assertions About Natural Gas Supply Fail to Adequately Account for the
Difficulty of Projecting Unconventional (Shale) Natural Gas Supplies as Well as Other Factors
That Could Impact Supply. ....................................................................................................... 43
1. Shale Gas Reserves Are More Difficult to Project Than Conventional Gas Reserves. . 44
2. EPA Has Failed to Take into Account the Varying Accuracy of EIA Projections. ....... 45
3. EPA Has Failed to Consider the Impact of Liquefied Natural Gas (LNG) Exports and
Increased Manufacturing and Transportation Demand on Supply. ....................................... 46
4. The Proposed Rule Also Fails to Take into Account the Use of Canadian Natural Gas
by U.S. Electric Utilities and How Market Conditions in Canada Could Impact Supply and
Prices in the U.S. ................................................................................................................... 51
B.
EPA’s Assumption That Natural Gas Prices Will Remain Relatively Flat Through 2030
Fails to Take into Account the Historic Volatility of Natural Gas, the Impact on Price from
Future Regulations on Upstream Production, or How Future Increased Demand Will Put
Upward Pressure on Prices........................................................................................................ 52
1. Historically, Natural Gas Prices Have Been Volatile. ................................................... 53
2. The Proposed Rule Does Not Take into Account the Potential Impact on Price of Future
Upstream Regulations............................................................................................................ 57
3. The Proposed Rule Does Not Take into Account the Potential Impact on Price of
Increased Non-Electric Utility Demand for Natural Gas. ..................................................... 58
C.
EPA’s Assertions Regarding the Adequacy of Existing Natural Gas Infrastructure and
the Ability to Expand It to Facilitate Fuel Switching Fails to Take into Account Impediments
to Infrastructure Development and the Lack of Sufficient Storage. ......................................... 58
1. The Proposed Rule Presumes That Significant Pipeline Expansion Is Possible, but Does
Not Take into Account Impediments to Pipeline Construction and Expansion That May
Impact the Large-Scale Fuel-Switching to Natural Gas to Reduce CO 2 Emissions. ............ 59
2. The Proposed Rule Fails to Examine Whether There Is Sufficient Natural Gas Storage
Needed to Support Large-Scale Fuel Switching to Natural Gas for Electric Generation. .... 61
2
VIII.
Gas-Electric Industry Coordination Issues Pose Barriers to the Rapid Increase in the
Use of Natural Gas for Electric Generation. ................................................................................. 69
IX. The Proposed Rule Fails to Take into Consideration Other Federal Environmental
Regulations That Will Impact the Ability of the States to Require Large-Scale Fuel Switching
from Coal to Natural Gas to Achieve Their CO 2 Reduction Goals. ............................................. 72
A.
EPA Did Not Consider That New NGCC Generation Must Meet Existing and Revised
NAAQS. .................................................................................................................................... 73
B.
The Proposed Rule Does Not Take into Consideration Non-Clean Air Act Regulations
That Will Impact the Ability of States to Require Large-Scale Fuel Switching from Coal to
Natural Gas to Achieve Their CO2 Reduction Goals. ............................................................... 74
X.
EPA Should Withdraw and Re-Propose the Rule.............................................................. 74
XI. If EPA Will Not Withdraw the Proposed Rule, APPA Recommends Several
Modifications That Would Improve Its Workability. ................................................................... 75
XII. EPA’s Selection of 2012 as the Baseline Is Inappropriate; States Should Be Allowed
Flexibility in Establishing a Representative Baseline................................................................... 76
XIII.
The Baseline and BSER Computations Should Allow Full Credit for Early Action. .... 77
XIV.
The Assumptions EPA Made in the Building Blocks Are Flawed and Do Not Provide
the Flexibility States Need to Meet Their CO 2 Reduction Goals. ................................................ 79
A.
Building Block 1—Heat Rate Improvements ................................................................ 80
1. EPA Overlooks the Significance of NSR Issues in Building Block 1 “EGU Efficiency
Improvements.” ..................................................................................................................... 80
2. EPA’s Analysis of Historical Data from Coal-Fired Units Fails to Provide Any Support
for Its Claim that Heat Rate Improvements of 4 to 6 Percent Are Achievable. .................... 87
B.
Building Block 2—Redispatch of Natural Gas Units .................................................... 92
1. EPA Improperly Calculated Capacity Factor and Number of Hours in Building Block
2—Existing Natural Gas Combined-Cycle Generation—and Should Correct Its
Calculations. .......................................................................................................................... 92
2. EPA Should Adjust Its Calculated Building Block 2 Targets Where the Integrated
Planning Model (IPM) Does Not Assume Removal of Coal Will Occur. ............................ 94
3. In Building Block 2, EPA Double Counted Some Units in Both the Existing NGCC
Capacity and the “Under Construction” Capacity. EPA Should Remove Those Units from
Its Goal Calculations. ............................................................................................................ 97
4. EPA’s Assumption That Each State’s Entire Fleet of Existing NGCC Units Can Match
the Operational Level of Its Top 10 Percent of Units Is Unsupported and Should Be
Corrected. .............................................................................................................................. 99
5. EPA Unreasonably Applied the Building Blocks to Non-Affected Subpart KKKK
Units..................................................................................................................................... 102
6. EPA Correctly Excluded Natural Gas Conversion and Co-Firing from BSER. .......... 104
3
C.
Building Block 3 - Renewable and Other Non-CO2 Emitting Generation................... 104
1. EPA’s Approach on Building Block 3 Fails to Take into Account the States’ Historical
Renewable Generation Mix and How an Individual State’s Source Mix Compares to the
Other States in EPA’s Designated Regions. ........................................................................ 104
2. EPA’s Application of an RPS from a State with a Rapidly Increasing Renewable
Energy Source to a State in Which Its Primary Renewable Energy Source Has Remained
Almost Flat Can Result in a Significant Overestimation of Renewable Generation Capability
in the Latter State................................................................................................................. 108
3. EPA Should Clarify Its Stance on Biomass Fuel. ........................................................ 109
4. There Are Significant Additional Costs and Constraints Not Factored into the EPA’s
Analysis of Building Block 3. ............................................................................................. 111
5. In Building Block 3, EPA Has Erred by Including Nuclear Capacity in Its State Goals. ..
...................................................................................................................................... 114
6. To Determine Lowest Cost BSER on a State-by-State Basis, EPA Should Modify Its
Determination of BSER to Include Additional Time and Consideration of Relevant Costs. ...
...................................................................................................................................... 119
7. The State Renewable Energy Generation Targets Are Unreasonably Aggressive and Do
Not Take into Account Factors Affecting the Actual Renewable Energy Growth Potential in
Each State. ........................................................................................................................... 120
8. APPA Agrees with EPA’s Assessment that Hydro Power Is Not a Universal Resource
and Should Be Excluded from EPA's Method for Quantifying Renewable Energy
Generation Potential. ........................................................................................................... 123
9. The Alternative Renewable Energy Approach Is Unworkable. ................................... 124
D.
Building Block 4 – Energy Efficiency ......................................................................... 126
1. The Load Growth Analysis in Building Block 4 Is Insufficient to Properly Account for
Potential Fluctuations. ......................................................................................................... 126
2. Environmentally-Friendly Electric Technologies That May Contribute to Positive Load
Growth ................................................................................................................................. 127
3. EPA Did Not Properly Account for the Decreasing Return on Investment in Energy
Efficiency in Its Development of Building Block 4 and Should Adjust Its Effi ciency
Requirement Downward...................................................................................................... 130
4. EPA Needs to Reconsider the Proposed Best Practices Level of Performance to Be Less
Stringent and Reflective of a Feasible Level for All States. ............................................... 141
5. APPA Supports EPA’s Assumptions Between the Years 2012 and 2017 in the Best
Practices Scenario................................................................................................................ 144
6. EPA Also Needs to Account for Differences in Reported and Projected Energy
Efficiency Savings Versus Actual Savings as Well as Acknowledge the Potential
Consequences if Projected Savings Are Not Met................................................................ 145
7. EPA Should Provide Additional Guidance in Multiple Areas. .................................... 147
4
8. EPA Should Provide Relief for New Electricity Use Driven Solely by Compliance
Requirements from Other EPA Rules. ................................................................................ 148
XV. States Need More Time to Prepare, Submit and Obtain EPA Approval for Their Plans. 149
XVI.
EPA Provides Too Little Guidance on Establishing Multi-State Plans and Interstate
Trading and Cooperation. ........................................................................................................... 151
XVII. EPA Should Eliminate the Interim Reduction Goal and Allow States to Determine Their
Own Glide Path........................................................................................................................... 152
XVIII. States Should Be Allowed an Opportunity to Adjust Their Final Reduction Goals, the
Year That the Goals Are to Be Achieved, and/or the Glide Path Based on Materially Changed
Circumstances. ............................................................................................................................ 156
XIX.
EPA Should Allow Additional Compliance Flexibility. .............................................. 157
XX. The Cost of Electricity to Consumers Has Been Increasing and Will Increase Even More
Under This Proposal ................................................................................................................... 157
A.
EPA’s Regulatory Impact Analysis Is Flawed. ............................................................ 158
B.
Electricity Prices Continue to Rise Generally.............................................................. 159
C.
Costs in Regions with RTO Markets............................................................................ 161
D.
Retail Electricity Prices Are Rising at a Faster Rate in States Within RTO Markets. . 163
E.
January 2014 Polar Vortex Gas and Electric Price Spikes........................................... 164
F.
Increases in Electricity Prices Disproportionately Impact Low and Fixed Income
Consumers............................................................................................................................... 166
G.
Increases and Volatility in the Cost of Natural Gas Flow Directly and Automatically to
Consumers............................................................................................................................... 166
H.
Remaining Useful Life of the Facility.......................................................................... 167
I.
Stranded and Replacement Costs ................................................................................. 171
J.
The Proposed Rule Will Impact Electricity Rates, Pushing Them Higher Than They Are
Today....................................................................................................................................... 174
K.
Potential Impacts on Credit Ratings Could Raise Borrowing Costs for Public Power
Utilities. ................................................................................................................................... 177
XXI.
The Proposal Raises Concerns About Reliability. ....................................................... 179
A.
B.
NERC Initial Reliability Review.................................................................................. 181
Transmission Planning ................................................................................................. 182
XXII. APPA Supports the Concept of a Reliability Safety Valve ......................................... 186
XXIII. The RTOs/ISOs Should Not Be Given Any New Market-Related Role in Implementing
the Final Rule.............................................................................................................................. 188
A.
B.
XXIV.
Regions
Overview of RTOs/ISOs .............................................................................................. 188
Impediments to the Use of RTO-Operated Markets for CO2 Reduction Strategies ..... 189
The Difficulties Facing Proposals for Environmental Dispatch or Redispatch in RTO
.................................................................................................................................. 191
5
A.
B.
1.
Summary of RTO Market-Based CO2 Reduction Proposals ....................................... 191
Critiques of RTO Market CO2 Reduction Proposals ................................................... 195
Prices and Costs Are Not Always Aligned. ................................................................. 196
C.
The Regional Greenhouse Gas Initiative Is Not an RTO-Operated Program, but a
Voluntary Program. ................................................................................................................. 200
D.
Summary of APPA Position on Environmental Dispatch............................................ 201
XXV. RTO-Operated Mandatory Capacity Markets Pose Significant Barriers to New
Generation Resource Development and Thus to Implementation of the Proposed Rule. .......... 202
A.
Background on RTO-Operated Capacity Markets ....................................................... 202
B.
RTO-Operated Mandatory Capacity Markets Have Not Been Effective in Leading to
the Construction of New, More Efficient Resources at a Reasonable Cost to Consumers..... 203
C.
Recent Mandatory Capacity Market Developments Create Direct Impediments to New
Resources. ............................................................................................................................... 204
XXVI.
Public Power’s “One Unit” Utility Members ........................................................... 207
XXVII.
The Final EPA Rule Should Respect the Importance of U.S.-Canadian Electricity
Generation Resources for Both Countries. ................................................................................. 210
XXVIII. Carbon Capture and Sequestration on Existing Power Plants .................................. 211
XXIX.
The NSPS Process for Existing Plants Does Not Require Automatic Revisions Every
Eight Years.................................................................................................................................. 212
XXX. Miscellaneous Issues. ................................................................................................... 212
XXXI.
Potential Constitutional Issues Raised by Proposed Rule. ....................................... 213
A.
The Proposed Rule’s Allocation of Renewable Energy Credits Potentially Raises
Constitutional Concerns Under the Fifth Amendment Takings Clause. ................................. 214
B.
The Proposed Rule’s Allocation of Renewable Energy Credits Potentially Raises
Constitutional Concerns Under the Article 1 Contracts Clause. ............................................. 216
XXXII.
XXXIII.
Conclusion ................................................................................................................ 217
Attachments .............................................................................................................. 218
6
I.
Executive Summary
The American Public Power Association (APPA) submits these comments to the Environmental
Protection Agency (EPA or the Agency) on the Proposed Rule (or Proposal) under section
111(d) of the Clean Air Act (CAA)1 to reduce emissions of carbon dioxide (CO2) from fossil
fuel-fired electric generating units (EGUs).2 EPA’s stated goal is to reduce CO2 emissions by 30
percent in 2030 from 2005 levels. APPA and its members believe the Proposed Rule aims to do
too much too quickly. As a result, it will create economic inefficiency; impose inequitably
distributed additional costs on consumers; threaten the reliability of the electricity system; and
force a risky over-reliance on a single fuel—natural gas—to generate electricity.
APPA agrees that the electricity sector needs to reduce CO2 emissions to address the adverse
impacts of climate change. APPA greatly prefers congressional action to address the issue, given
the inherent limitations of the current Clean Air Act, the fact that this issue needs to be addressed
on an economy-wide basis, and the ubiquitous nature of CO2 and other greenhouse gas (GHG)
emissions. At the same time, APPA recognizes that congressional action is unlikely in the
foreseeable future and that the President has directed EPA to issue a final rule to reduce CO 2
emissions in June 2015 under its existing authority. Thus, APPA’s comments emphasize a
number of recommendations to improve the Proposed Rule that, if incorporated, would make it
more workable for industry and more affordable for consumers, while still allowing substantial
progress towards the Agency’s ultimate goal.
The electric utility industry generally, and public power utilities in particular, have already made
good progress in reducing CO2 emissions. In 2012, the industry’s CO 2 emissions were at their
lowest level since 1994. Between 2007 and 2012, those emissions fell by 12 percent, though
recently there has been a slight increase. The overall decrease that has occurred is mainly the
result of investments in renewable energy (RE) and energy efficiency (EE), an increase in the use
of natural gas to generate electricity, and the retirements of coal-fired generation units. Public
power utilities are consistently recognized as leaders in renewable energy and energy efficiency .
These utilities are also making new investments in nuclear and hydro energy, key non -emitting
sources of baseload generation. All indications are that these CO2-reducing activities would
continue and increase, even in the absence of new EPA regulation.
1
42 U.S.C. § 7411 (2012).
Carbon Pollution Emission Guidelines for Existing Stationary Sources: Electric Utility Generating Units, 79 Fed.
Reg. 34,830 (June 18, 2014) (Proposed Rule or Proposal).
2
7
APPA has multiple concerns with the Proposed Rule. Its requirements go beyond what is legally
permissible under Section 111(d) and conflict substantially with the authority of other federal,
state, and local governmental entities. The Proposal envisions compliance measures far beyond
those that can be implemented at the affected sources of emissions, creating uncertain and legally
untested compliance obligations for non-utility entities and the potential for enforcement actions
against them.
EPA asserts that, while electricity costs will rise due to compliance with the Proposal, consumers
will see only “negligible” increases in their actual bills after 2020, and could see decrease in the
long term as a result of the expected energy efficiency gains. APPA is highly skeptical of that
assertion. APPA believes the Agency has relied too heavily on optimistic assumptions on a
number of key elements, such as the price of natural gas; the ability of utilities and system
operators to dispatch natural-gas units at significantly higher capacity factors; the availability in
some states of viable, economic, renewable energy resource; and the rate at which new energy
efficiency programs can be implemented. APPA also believes that EPA has underestimated or
ignored other critical factors, such as the likelihood of stranded costs and economic value due to
the forced early retirement of many coal-fired units, the availability of natural gas infrastructure
necessary to support its projections of natural gas use, and the barriers to new resource
development posed by the mandatory capacity markets in the eastern regional transmission
organizations (RTOs). Thus, APPA believes it is more likely that costs and consumers’ bills will
increase for years to come unless EPA modifies its Proposed Rule as recommended in these
comments. For EPA to assert that the electric utility industry can achieve a 30 percent reduction
in CO2 emissions and also lower consumers’ electricity bills by 2030 recalls the adage that “if it
sounds too good to be true, it probably isn’t true.”
APPA is also concerned about the Proposed Rule’s potential negative impact on electric service
reliability. It essentially requires a rapid transition in the composition of the nation’s electricity
generating fleet and end-use efficiencies that, if not implemented precisely as envisioned, can
create gaps between supply and demand and other reliability problems. APPA notes the strong
comments and recommendations on this issue of the Southwest Power Pool (SPP), the North
American Electric Reliability Corporation (NERC), and other entities responsible for ensuring
the reliability of the system, as well as individual APPA members.
Other concerns noted in these comments and by APPA members include:



The lack of sufficient credit for investments and other actions to reduce CO2 emissions
taken before 2012.
The use of a single year (2012) as the baseline.
The lack of sufficient time for states to develop and gain approval for their compliance
plans.
8


The imposition of an interim goal starting in 2020 that comes a mere two years after
approval of state plans and that, for many states, constitutes the majority of their final
reduction requirement due in 2030.
The inappropriate treatment of new nuclear units currently under construction in both the
calculation of the relevant states’ required reduction and those units’ use for compliance.
For all these reasons, APPA believes EPA should withdraw and re-propose its Proposal. If EPA
moves forward with this Proposal, however, then APPA strongly recommends certain changes
that, taken together, would improve its workability and affordability, while still continuing to
reduce CO2 emissions.
APPA’s recommendations generally are intended to address our overarching concern noted
earlier that the Proposal tries to do too much too fast. Our recommendations also provide states
and utilities with the flexibility to address their individual circumstances and to accommodate
unanticipated and/or uncontrollable events. Lastly, the recommendations incorporate into the
Proposal a greater level of state authority and discretion that hews much more closely to the
model of cooperative federalism Congress intended when it enacted Title I of the Clean Air Act.
APPA urges EPA to modify the Proposed Rule to:











Allow states to choose a baseline that accurately reflects their unique circumstances.
Provide full credit for investments already made that reduce or offset CO2 emissions.
Fix the errors and revise the assumptions in the computations of the four building blocks
to reflect what the states can realistically accomplish and ensure more equity among the
states.
Provide a streamlined process for new source review determinations and stipulate that an
EGU’s energy efficiency upgrade under a state compliance plan should be considered
greenhouse gas Best Available Control Technology (BACT) for Prevention of Serious
Deterioration (PSD) determinations.
Remove nuclear units under construction from the relevant state baselines.
Allow all generating resources that emit no CO2 to be used for compliance.
Provide states with more time to develop state compliance plans.
Provide more guidance on the development of multi-state plans and interstate
agreements.
Eliminate the interim reduction requirement and allow states to determine their own
emission reduction trajectory (glide path) to reach their final reduction goal.
Allow a state’s final reduction goal, the year to achieve that goal, and/or the glide path to
be adjusted if a state can demonstrate that circumstances have materially changed.
Include and allow mechanisms to ensure that entities with a compliance obligation under
a state plan have the maximum degree of flexibility to comply at reasonable cost,
9

including through reduction or avoidance measures from non-electricity portions of the
broader energy sector.
Provide for the establishment of a reliability “safety valve” to ensure that compliance
with mandated emission reduction requirements does not inadvertently impair system
reliability or conflict with NERC standards.
APPA very much appreciates the Agency’s decision to extend the comment deadline to allow a
fuller opportunity to analyze the myriad details of the Proposal. APPA also appreciates the
positive and constructive attitude that the Agency and its staff have displayed during the
extended comment period on the Proposed Rule, especially their willingness to liste n to APPA’s
and its members’ concerns. APPA stands ready to continue to work with the Agency after the
close of the comment period to help craft a Final Rule that further reduces CO2 emissions, while
assuring electric system reliability, keeping associated cost increases to reasonable levels, and
avoiding the stranding of significant utility assets.
II.
Introduction
APPA is the national service organization representing the interests of not-for-profit, publicly
owned electric utilities throughout the United States. More than 2,000 public power utilities
provide over 15 percent of all kilowatt-hour sales of electricity to consumers and do business in
every state except Hawaii. All APPA utility members are Load Serving Entities (LSEs), with the
primary goal of providing customers in the communities they serve with reliable electric power
and energy at the lowest reasonable cost, consistent with good environmental stewardship. This
orientation aligns the interests of APPA utility members with the long-term interests of the
residents and businesses in their communities. Collectively, public power utilities serve more
than 47 million customers.
The Proposed Rule, which seeks to address climate change concerns, would have a tremendous
impact on APPA’s members and their communities. It would establish CO2 emission guidelines
for existing fossil fuel-fired electric generating units (EGUs) under Section 111(d) of the CAA
and require the states to submit plans to EPA for complying with those guidelines. APPA
prefers congressional action to address climate change. However, in the absence of legislation,
APPA wants to work with EPA to improve any final rule. Thus, these comments include several
recommendations that would improve the Proposal.
The Proposed Rule is simply unworkable. The building blocks in the Proposal are unrealistic,
and few states will be able to meet the required emissions reductions by the interim and final
deadlines. Further, the Proposal fails to provide public power utilities with full credit for taking
early action to reduce their CO2 emissions by adding non-emitting or lower-emitting energy
resources and adopting energy efficiency measures. This failure is doubly unfair because public
10
power utilities were encouraged by local, state, and other federal policies to reduce CO2
emissions through these methods. EPA describes the Proposed Rule as flexible and states that
the deadline is 2030. In reality, however, the enforceable interim goal that must be met on an
average basis starting in 2020 contains such a steep reduction for most states that the final 2030
deadline is not the crux of the problem. The fundamental issue is that the interim goal timeframe
is too steep and comes much too fast. States (and public power utilities) need a longer glide
path. These comments address the glide path and issues with the building blocks in Sections VI,
VII, XIV, XV, XVII, and XVIII.
The Proposed Rule deviates sharply from EPA’s past methods of regulation under Section 111,
which have addressed a specific pollutant with a specific control technology. Of the nearly 100
New Source Performance Standard (NSPS) and emission guidelines that EPA has promulgated
and revised since 1970, every single standard of performance has been based on a “system of
emission reduction” that is incorporated into the design or operation of individual sources. By
contrast, the Proposal’s “system” has states changing their statewide power-plant fleets, altering
how these fleets operate, and reducing their citizens’ consumption of electricity through a set of
“options” that really do not present any true choice. The stringency of the Proposed Rule’s goals
means that in most states, all four building blocks must be used in order for the state to achieve
the required emission reductions.
Further, the building blocks, in combination with the interim and final enforceable goals, mean
that many public power utilities would find it difficult—if not impossible—to keep their coalfired power plants operational. Worse, public power utilities may sustain significant financial
harm because their fossil-fired generation fleets contain units with remaining useful lives of five,
ten, or twenty years (or even more years for units that have just recently retrofitted emission
controls to meet the latest EPA rules, such as the Mercury and Air Toxics Standards (MATS)),
that are cut short by the constraints of the Proposed Rule on the states—and indirectly on the
utilities. Public power utilities will have stranded and lost opportunity costs if their existing coal
and natural gas-fired power plants are prematurely shut down. Some public power utilities might
have remaining debt service on these generation units, but the problem is bigger than that,
because the remaining useful life of a power plant may continue even after that debt is retired.
APPA’s comments address these concerns for public power utilities in more detail in Section
XX(H&K), XXVI, and XXXI(A&B).
APPA is a member of the Utility Air Regulatory Group (UARG) and the National Climate
Coalition. APPA endorses the legal and technical commentary, questions, and critiques offered
by UARG on the Proposed Rule and incorporates those comments by reference. While UARG’s
comments focus on broad issues relevant to the entire electric utility industry, APPA’s comments
address legal issues of particular concern to public power utilities. APPA urges EPA also to
carefully consider the comments submitted by individual APPA members, as those comments
11
contain detailed information and examples on issues presented by the Proposal, as well as
recommendations for improvements. In addition, APPA endorses the comments of the National
Climate Coalition. In particular, APPA asserts that the Proposed Rule is inconsistent with the
requirements of the CAA because it would violate the Act’s clear division of responsibility for
regulatory decision making between the federal government and the states, eliminating the broad
discretion Congress granted to the states when it enacted section 111(d) of the Act. Instead, the
Proposal seeks to assign all of that discretion to the Agency itself.
Similarly, the Proposed Rule shows little regard for the complex division of federal, state, and
local authority with respect to the governance and regulation of the electricity industry. The
Federal Power Act (FPA) assigns federal regulatory authority in this area to the Federal Energy
Regulatory Commission (FERC). The authority of FERC, however, is strictly limited to
transmission and wholesale sales of electricity in interstate commerce. States retain broad and
exclusive regulatory authority over retail sales and service, local distribution service, and the
need for and siting of generation facilities.3 States exercise that authority through comprehensive
direct regulation of utility companies’ facilities, services, and rates—and by providing for the
creation of state or local public power utilities and electric cooperatives that are largely selfgoverned and answerable to the customers in the communities they serve. The FPA preserves
state sovereignty and local control of public power utilities by excluding states and their
subdivisions and agencies from FERC’s plenary regulatory authority. EPA, although it has no
authority to address any of these matters concerning the governance and regulation of the electric
utility industry and its internal operations, would assume broad, overarching control of the
industry with the finalization of the Proposed Rule.
It is also clear that compliance with EPA’s Proposed Rule would expose utilities to substantial
risk under the CAA’s New Source Review (NSR) program. EPA attempts to ignore this problem
rather than address it directly. Indeed, there are numerous other examples in the Proposed Rule
where EPA proposes to adopt mistaken or dubious policy positions or factual assumptions that
result in the Proposal being unrealistic, technically infeasible, and arbitrary and capricious. That
the Proposed Rule, if implemented, would have such enormous impact—economically, and with
respect to federalism principles—makes these flaws all the more problematic.
EPA has only just recently been reminded that it cannot simply seize massive new regulatory
authority for itself. In Utility Air Regulatory Group v. EPA,4 the Supreme Court invalidated
another of EPA’s regulations aimed at limiting the emission of CO2 because EPA overstepped
3
4
See, e.g., New York v. FERC, 535 U.S. 1 (2002).
134 S. Ct. 2427 (2014) (“UARG v. EPA”)
12
the authority granted to it in the CAA. In doing so, the Court provided valuable lessons that EPA
should heed in these proceedings. The Supreme Court makes clear that regulation of greenhouse
gases, including CO2, cannot be “‘extreme,’ ‘counterintuitive,’ or contrary to ‘common sense.’” 5
Regulations often fall into those impermissible categories, the Court explained, when an agency
interprets a statute in a way that “would … bring about an enormous and transformative
expansion in [its] regulatory authority without clear congressional authorization.” 6 The Court
further cautioned that “[w]hen an agency claims to discover in a long-extant statute an
unheralded power to regulate ‘a significant portion of the American economy,’ … we typically
greet its announcement with a measure of skepticism.” 7
The Proposed Rule envisions an enormous and transformative expansion of the Agency’s
authority, and the Clean Air Act evinces no congressional intent supporting EPA’s action, much
less the clear authorization the Supreme Court requires. In the Proposal, the Agency would
become the primary regulator of electric power within the United States, including regulating (i)
the dispatch of electric generation, (ii) the amount of renewable power to be built, and (iii)
requiring customers to limit their electricity consumption. This broad assertion of new-found
regulatory authority is exactly what the Supreme Court has held EPA cannot do absent
unequivocal statutory authority, which EPA does not have in this instance. The Proposal is thus
flawed legally and as a matter of policy and should be withdrawn.
However, APPA is keenly aware that EPA intends to issue a final rule under Section 111(d) in
June 2015 pursuant to a specific directive from the President, and that any final rule must be
based on the Proposed Rule. Thus, as mentioned previously, APPA’s primary intent in these
comments is to recommend changes to the Proposed Rule to improve its workability and
affordability for the consumer/owners of public power utilities. Those specific recommendations
are summarized in Section XI and delineated more fully in subsequent sections of these
comments.
III.
The Proposed Rule Denies States the Primacy and Implementation
Authority Accorded by Congress.
The Proposed Rule disrupts the congressionally established and judicially recognized division of
responsibility for implementing existing source standards under Section 111(d) of the Clean Air
Act. The Proposed Rule conflicts with the Act by reducing states to secondary partners with
5
Id. at 2441 (quoting Massachusetts v. EPA, 549 U.S. 497, 531 (2007)).
Id. at 2432 (citing FDA v. Brown & Williamson Tobacco Corp., 529 U.S. 120, 160 (2000)).
7
Id. at 2444 (quoting Brown & Williamson, 529 U.S. at 159).
6
13
EPA, when, in fact, Congress intended states to take the lead. This flaw is evident in the general
approach EPA proposes to take: first, in establishing the CO 2 emission rate limits applicable to
each state; and second, in the specific details of state plan implementation discussed throughout
the Proposal. Each of these issues is addressed below.
A.
EPA’s Proposed Rule Unlawfully Constrains State Primacy Under
Section 111(d).
EPA’s authority under Section 111(d) is limited. EPA is empowered to “establish a procedure
similar to that provided by [section 110 of the Act 8] under which each State shall submit to
[EPA] a plan which … establishes standards of performance” for existing sources within the
state (emphasis added).9 EPA can set substantive standards of performance only when a state
fails to submit a “satisfactory” plan.10 The Act gives states, on the other hand, broad discretion
to develop plans to implement section 111(d) standards of performance subject to a general
requirement that the state’s exercise of discretion be “satisfactory.”11
For a plan to be “satisfactory,” it must include performance standards that are consistent with the
definition of “standard of performance” in Section 111(a)(1), Id. § 7411(a)(1), and it must
“provide[ ] for the implementation and enforcement” of the standards, Id. U.S.C.
§§ 7411(d)(1)(B). Moreover, EPA must permit the state in applying a standard “to take into
consideration, among other factors, the remaining useful life of the existing source to which the
standard applies.”12 Apart from those statutory requirements, states have significant discretion to
develop their plans, including discretion to adopt state plans that differ from EPA’s emission
guidelines.13
EPA predicates the Proposed Rule on the Agency’s purported authority to impose “binding” CO2
emission rate goals on each state.14 But that assertion of authority ignores the structure and
context of Section 111(d) and EPA’s implementing regulations. Those regulations direct EPA to
8
42 U.S.C. § 7410
42 U.S.C. § 7411(d)(1).
10
Id. § 7411(d)(2)(A).
11
Id.
12
Id. APPA’s comments on remaining useful life here and in other sections of these comments address economic
and practical operational issues. APPA also incorporates by reference the detailed comments provided by UARG on
remaining useful life issues as identified from the Proposed Rule and the Oct. 28 Notice of Data Availability,
Carbon Pollution Emission Guidelines for Existing Stationary Sources: Electric Utility Generating Units, 79 Fed.
Reg. 64,543 (Oct. 30, 2014) (NODA).
13
See 40 C.F.R. § 60.24(f) (2014).
14
See, e.g., 79 Fed. Reg. at 34,844, 34,892.
9
14
publish emission guidelines and contemplate that state-developed emission standards included in
state plans will generally “be no less stringent than” EPA’s emission guidelines.15 The
regulations make clear, however, that in contrast to the “emission standards” that states adopt,
EPA’s emission guidelines are not “legally enforceable.”16 Instead, EPA’s emission guidelines
are “only…criteria for judging the adequacy of State plans.”17 Thus, EPA cannot prescribe to
states a standard of performance. The binding state emission rates in the Proposal do precisely
that.
Further, the Act and EPA’s regulations provide states with considerable flexibility to deviate
from EPA’s emission guidelines in their plans. The regulations provide that states may apply
“less stringent emission standards or longer compliance schedules” to individual facilities or
classes of facilities if adopting the standards reflecting the emission guidelines would be
unreasonably costly, physically impossible, or for other reasons.18 EPA must permit states to
take into consideration the “remaining useful life of the existing source,”19 and in doing so,
permit states to grant individual sources or classes of sources longer periods of time to comply,
or to apply less stringent standards than set forth in EPA’s emission guidelines. EPA’s
regulations also require that its emission guidelines address subcategories of “different sizes,
types, and classes” of existing sources where factors like “costs of control, physical limitations,
[or] geographical location” warrant the application of different guidelines.20 The Proposed Rule
fails to do any of these things.
Case law interpreting section 110 of the Act, which governs the requirements of state
implementation plans (SIPs) to attain and maintain national ambient air quality standards
(NAAQS), helps illustrate why the Proposed Rule exceeds the Agency’s authority. Courts have
explained that states are “given wide discretion in formulating” their SIPs under section 110,
Union Elec. Co. v. EPA, 427 U.S. 246, 250 (1976), and made clear that EPA’s role is “confine[d]
… to the ministerial function of reviewing [SIPs] for consistency with the Act’s requirements,”
Luminant Generation Co. v. EPA, 675 F.3d 917 (5th Cir. 2012). EPA, on the other hand has “no
authority to question the wisdom of a State’s choices of emission limitations if they are part of a
plan which satisfies the standards of § 110(a)(2).” Train v. NRDC, 421 U.S. 60, 79 (1975).
Section 110 places more constraints on state discretion than does Section 111(d). Yet not even
15
40 C.F.R. § 60.24(c).
State Plans for the Control of Certain Pollutants from Existing Facilities, 40 Fed. Reg. 53,340, 53,341 (Nov. 17,
1975).
17
Id. at 53,343.
18
40 C.F.R. § 60.24(f)(1)-(3).
19
42 U.S.C. § 7411(d)(1)(B)
20
40 C.F.R. § 60.22(b)(5).
16
15
Section 110 would allow EPA to establish state-specific CO2 emission goals as part of a SIP.
The Agency certainly cannot establish such binding emission requirements for the states under
Section 111(d). The Proposed Rule is plainly inconsistent with the Act.
The Proposed Rule does not address any of these issues in a meaningful way. EPA merely
points to the “flexibility” the Proposal would afford states. Regardless of the flexibility states
would or would not have—and as explained elsewhere in these comments any flexibility
afforded by the Proposal is very limited—EPA does not have the authority to disregard the Clean
Air Act’s basic structure and eliminate the primacy and the discretion Congress afforded to the
states in implementing any binding emission standards under Section 111(d).
B.
EPA’s Discussion of State Implementation Issues Reveals Significant
Problems with the Proposed Rule.
The Proposal leaves states in a difficult position. While touting the flexibility available to states,
EPA provides only a bare bones description of how states might develop “satisfactory” plans to
implement Section 111(d) requirements for existing EGUs. This approach, which leaves states
without sufficient guidance on many practical matters, also obscures the fact that the P roposed
Rule would, in reality, improperly constrain state discretion under Section 111(d). Each of the
state plan development and implementation issues EPA identifies is discussed below.
1.
The Proposed Rule Does Not Provide States with “Flexibility.”
EPA repeatedly claims that it is offering states flexibility to design their Section 111(d)
programs. By establishing state CO 2 emission goals based on aggressive assumptions about
what each of the four building blocks can achieve (and by attempting to elimi nate state authority
to revise those goals), EPA already has made the most significant policy decisions by itself.
EPA claims that states retain the flexibility not to implement each of the building blocks in the
manner EPA assumed when developing state goals.21 In reality, however, it would be difficult
for a state to achieve the emission targets set by EPA without implementing all of the building
blocks. Accordingly, states do not have the option of not implementing a building block or
substantially deviating from EPA’s assumptions. Moreover, even if states could manage to get
more emission reductions from one building block to offset fewer reductions from another, that
would be considerably less flexibility than states must be afforded under Section 111(d).
21
79 Fed. Reg. at 34,926.
16
2.
The Proposed Rule Encroaches on Areas of Exclusive State and
Local Government Authority.
Under state laws, most, if not all, of the emission-reducing measures contemplated in building
blocks 2, 3, and 4 fall under the exclusive authority of state utility regulators or the governing
boards of local public power utilities or rural electric cooperatives established pursuant to state
law, not state environmental agencies or EPA. In the absence of a clear congressional
authorization, which does not exist here, EPA cannot infringe upon traditional state sovereign
functions. EPA even concedes in the Proposed Rule that including measures that are “the
exclusive preserve of the state” in a plan that becomes federally enforceable might be unlawful. 22
EPA suggests these problems might be avoided if plans simply do not include such measures and
states instead rely on them (and their emission reductions) as complementary programs. 23 This
suggestion ignores the fact that states would still need to make significant changes in policies
over which EPA has no regulatory authority in order to comply with the Proposed Rule.
The Proposed Rule intrudes upon the local control of public power utilities. Most state laws
allow for autonomous self-governance of their state and municipal public power utilities, most
often by a board of directors or commissioners, which may be appointed or elected . The
governing boards of public power utilities determine the composition of the utilities’ generation
resource portfolios (the development and retirement of specific generation units) and how their
generation fleets are operated and dispatched to meet the requirements of their customers. State
utility commissions typically have limited or no regulatory authority over public power utilities .
In some instances, the state utility commission approval of new generation units may be required
through the issuance of a certificate of public convenience and necessity. But state utility
commissions typically do not otherwise regulate public power utilities in their states.
The Proposal assigns implementation of the portfolio-based emission-reducing measures in
building blocks 2, 3, and 4 to the states. EPA does not specify whether the state utility
commission or the state environmental regulatory agency is charged with implementing building
blocks 2, 3, and 4. However, state environmental agencies generally do not have authority under
state law to implement these building blocks—or have any experience in integrated resource
planning—and thus they may be ill-suited to the task in comparison to the state utility
commission. But the Proposal fails to appreciate that state utility commissions can implement
building blocks 2, 3, and 4 only for the utilities they regulate under state law, and nearly all state
utility commissions lack the authority to implement these building blocks with respect to their
22
23
Id. at 34,902.
Id.
17
state’s public power utilities. Thus, EPA oversimplifies the difficulty of implementation of the
Proposal’s portfolio-based building blocks under existing state laws.
Moreover, EPA is mistaken, and oversteps its authority, if it believes that states can simply
revise their laws to give their utility commissions (or environmental agencies) additional
authority over public power utilities for the sole purpose of ensuring compliance with EPA’s
Proposed Rule by all utilities in the state. Because building blocks 2, 3, and 4 require changes in
core utility operations, state commission oversight of public power utilities for purposes of
compliance with the Proposed Rule would be tantamount to plenary state commission (or
agency) regulation of public power utilities. This would fundamentally reorder state choices on
the governance of public power utilities and undermine the local control that has been the
bedrock principle of the nation’s public power utilities for over a century. APPA and its
members would oppose such policy initiatives. Indeed, in keeping with this principle, some state
constitutions forbid the state’s public utility commission from asserting any authority over
municipal utilities or municipal joint action agencies.24 In these states, implementation of the
Proposed Rule cannot be done by special legislation and would require amendment of the state
constitution, which is impracticable under short deadlines for submission of a state
implementation plan under the Proposed Rule.
3.
The Option of a Rate-Based or Mass-Based Goal
Under the Proposed Rule, states must select either a rate-based or mass-based emission goal.
EPA claims that this mandate provides flexibility to the states. 25 This is not the case given that
EPA requires that any mass-based goal must be equivalent to the rates the Agency would
prescribe.26 States were originally invited to calculate their own mass-based goals, and EPA
provided some guidance in the Proposal on how it believes those calculations should be made.
On October 28, 2014, EPA issued additional information on converting state emission goals
from rate to mass.27 In response to concerns from numerous states, however, about the
complications of calculating the mass-based goal and proposed regulatory language suggesting
EPA believes it could reject a state plan that calculates a mass-based goal in a manner with
which EPA does not agree, EPA released a technical support document (TSD) that provided each
state’s mass-based goal less than 30 days before the deadline for these comments on November
24
See, e.g. Logan City v. PSC, 271 P.2d 961 (1929) (Utah).
See 79 Fed. Reg. at 34,897.
26
Id. at 34,892 (mass-based goal must “achiev the same degree of emission limitation” as EPA’s rate-based goal).
27
http://www2.epa.gov/carbon-pollution-standards/clean-power-plan-proposed-rule-translation-state-specific-ratebased-co2
25
18
6, 2014.28 EPA should not reject any state plan that proposes a mass-based goal calculation that
addresses the source category even if it does not achieve the same “emissions levels” as would
be calculated under a rate-based goal. The two methods of achieving emissions goals are
entirely different, and because of the significant discretion afforded to states under Section
111(d), states should be permitted to design or translate a mass-based goal in any reasonable
manner that they see fit. This should include giving states the discretion to develop both a ratebased or mass-based approach for the power plants or utilities in their state. This discretion
would allow states to take into consideration the individual circumstances of the utilities and
their resources in that state. It should not matter whether different approaches are applied to
utilities in the state as long as states can demonstrate that the emissions achieved by their state
plans would be equal to or less than a rate-based or mass-based approach set at the state level.
4.
The Four Types of State Plans
The Proposed Rule acknowledges that “section 111(d) gives states the primary responsibility for
designing their own state plans for submission to the EPA.” 29 The Proposal and EPA’s State
Plan Considerations Technical Support Document (“State Plan TSD”) identify four types of
plans that EPA believes will be approvable: (1) rate-based CO2 emission limits applied to
affected EGUs; (2) mass-based CO2 emission limits applied to affected EGUs; (3) a state-driven
portfolio approach; and (4) a utility-driven portfolio approach. 30 Because states have the primary
role in designing plans, EPA cannot disapprove a plan simply if it does not conform to the plan
types it has identified. EPA acknowledges that there are material differences between state plans
submitted under Sections 110 and 111.31 The courts have consistently recognized a state’s
primary discretion in setting the pace of implementation under the Act.32.33
28
79 Fed. Reg. 67,406 (Nov. 13, 2014) (notice).
79 Fed. Reg. at 34,901.
30
State Plan TSD at 5; see also 79 Fed. Reg. at 34,901-02.
31
79 Fed. Reg. at 34,834.
32
See, e.g., Bethlehem Steel Corp. v. Gorsuch, 742 F.2d 1028, 1036 (7th Cir. 1984)(“Congress has given the states
the initiate and a broad responsibility regarding the means to achieve those ends through state implementation plans
and timetables for compliance…The Clean Air Act is an experiment in federalism, and the EPA may not run
roughshod over the procedural prerogatives that the Act as reserved to the states…”) See also CAA §110(a)(2), 42
U.S.C. § 7410(a)(2) (“Each implementation plan shall-(A) include enforceable emission limitations and other
control measures, means or techniques as well as schedules and timetables for compliance, as may be necessary or
appropriate to meet the applicable requirements of this Act…”).
33
The issues pertaining to the types of state plans, timing of the plans and flexibility of the plans are also implicit in
the Agency’s call for comments in the NODA. APPA’s comments on natural gas infrastructure for building block 2
also respond to the NODA. These are found in APPA’s comments in Section VI and Section XIV.
29
19
The Proposed Rule also acknowledges that the plans EPA envisions would differ significantly
from other types of Clean Air Act implementation plans and plans submitted under Section
111(d). The Proposal, however, glosses over the unprecedented and arguably unlawful elements
of each plan type, including measures that would be enforceable against entities other than
affected EGUs and treating demand-side energy efficiency (EE) and renewable energy (RE)
requirements as “standards of performance” for the affected EGUs. EPA must provide a more
thorough analysis of these issues rather than simply leaving states to grapple with them.
5.
Timing
The Proposed Rule states that its interim goals, which must be achieved on average over 2020 to
2029, and its final goals, which must be achieved by 2030, provide states with flexibility to
design plans over the long term. 34 But because the interim goals are so stringent, many states
will have to take significant actions by 2020 in order to comply.
The Proposed Rule also requires significant effort by the states in preparing their plans, including
the requirement to include achievement “demonstrations” that use utility-scale capacity
expansion and dispatch planning models to show the state can meet the interim and final goals.35
This requirement exceeds the statutory standard that a state plan be “satisfactory.” 36 Moreover,
if EPA is going to require this type of effort from the states to develop the plans, then it must
give far more time to the states than the Proposal contemplates.
6.
Criteria to Approve State Plans
The Proposed Rule includes “four general plan approvability criteria,” explaining that, at a
minimum, a state plan must meet these criteria to be “satisfactory” under section 111(d)(2)(A).37
The approvability criteria are: (1) Enforceable Measures; (2) Emission Performance; (3)
Quantifiable and Verifiable Emission Performance; and (4) Reporting and Corrective Actions. 38
EPA cannot bootstrap the statute’s single “satisfactory” criterion into something far more
demanding.
In addition, there are practical issues with each of these factors that EPA must further address
before states can reasonably be expected to submit plans that might reflect the criteria EPA has
34
79 Fed. Reg. at 34,904-05.
Id. at 34,904; Plan TSD at 28-31.
36
42 U.S.C. § 7411(d)(2)(A).
37
79 Fed. Reg. at 34,900, 34,909.
38
Id. at 34,909-11.
35
20
identified. For example, with regard to enforceable measures, the Proposed Rule notes that there
are serious questions about whether plans will be able to ensure enforceability given the nature
of the Proposal.39 As to emission performance, the Proposed Rule states that plans must be at
least as stringent as the proposed goals, but again, there are serious legal issues with EPA’s
proposal to impose binding goals for the states. There are similar problems with the remaining
criteria, and EPA should recognize that it is for states to determine how to address these issues in
their plans.
7.
Components of Approvable Plans
In addition to the four criteria, the Proposed Rule identifies twelve components that must be
included in a plan in order for it to be approved by EPA: 40
1.
2.
3.
4.
5.
6.
Identification of affected entities;
Description of plan approach and geographic scope;
Identification of state emission performance level;
Demonstration that plan is projected to achieve emission performance level;
Identification of emissions standards;
Demonstration that each emissions standard is quantifiable, non-duplicative, permanent,
verifiable, and enforceable;
7. Identification of monitoring, reporting, and recordkeeping requirements;
8. Description of state reporting;
9. Identification of milestones;
10. Identification of backstop measures;
11. Certification of hearing on state plan; and
12. Supporting material.
The Proposed Rule is flawed because Congress afforded the states primary authority over the
contents of their Section 111(d) plans.
8.
Deadlines and Process for State Plan Submittal
Section 111(d) by its terms authorizes EPA to “establish a procedure” for state submission of
implementation plans.41 The procedure EPA has proposed here, however, has multiple
problems. First, the Proposed Rule would allow only 13 months for initial plan submittal, with
39
79 Fed. Reg. at 34,909.
Id. at 34,852, 34,911-14.
41
42 U.S.C. § 7411(d)(1).
40
21
extensions of one year or two years possible depending on whether states are submitting single state or multi-state plans.42 Although at first blush this time limit may seem reasonable, as EPA
acknowledges, the plans the Agency seeks here are unlike any regulatory plan states have
experience preparing.43 The impacts of the Proposed Rule and the complexity of the issues it
raises require much more time for plan development.
The Proposed Rule also notes that states may need to modify their plans over time, and EPA
proposes that modification will generally be allowed provided the revision “does not result in
reducing the required emission performance for affected EGUs specified in the original approved
plan.”44 EPA cites no authority for imposing this limitation. If the modified plan continues to be
“satisfactory,” then state modification of the plan should be permissible.
EPA does ask whether it would be helpful if the Agency developed a template plan that states
could use.45 Such a plan would be helpful, but EPA cannot require states to follow the template
in order to have their plans approved.
9.
EPA’s “Key Considerations” That States Must Address in
Developing Their Plans
EPA closes its discussion of state plan and implementation issues with a discussion of what it
calls “key considerations” for states as they develop their plans. APPA discusses each of these
considerations below.
Affected Entities Other Than Affected EGUs. EPA acknowledges that placing enforceable
goals on non-EGUs (i.e., the portfolio approach) may be “challenging” and tasks states with
working out these problems.46 EPA cannot impose goals if it does not know if they are lawful,
and it cannot circumvent that by asking the states to figure it out.
Treatment of Existing State Programs. EPA states early emission goals (those that took place
before the date of the Proposal), are reflected in the baseline for each state but are not otherwise
credited. EPA should improve its treatment of existing programs and early reductions by
expanding the ways states can apply those goals to meeting their goals. One way EPA could do
that is by adopting alternative approaches described in the Proposed Rule for taking existing EE
42
79 Fed. Reg. at 34,915.
Id.
44
Id. at 34,917.
45
Id.
46
Id.
43
22
programs into account. Specifically, EPA should change the cutoff for the date by which actions
to put EE into place may be taken to at least 2005. EPA should also allow emission reductions
that occur prior to the initial performance period—at least as early as 2005—to count toward
meeting state goals.47
Incorporating RE and Demand-Side EE Measures Under a Rate-Based Approach. EPA asks
how states should credit or adjust CO2 emission rates to take RE and EE into account.48 EPA
should acknowledge that states have broad discretion over such decisions.
Quantification, Monitoring, and Verification of RE and Demand-Side EE Measures. EPA
seeks comment on emission monitoring and verification guidance it intends to prepare.49 That
guidance should ensure that states have broad discretion over how to address such matters in
their plans.
Projecting Emission Performance. Again, this is an area where states have discretion. At a
minimum, as discussed above, if EPA insists on states providing detailed modeling with their
plans, then more time must be provided to ensure the states can meet their deadlines.
Potential Emission Reduction Measures Not Used to Set Proposed Goals. According to EPA,
“[s]tates may include measures in their plans beyond those that the EPA included in its
determination of the [best system of emission reduction (BSER)].”50 States should be free to
include any measures they deem appropriate.
Consideration of a Facility’s “Remaining Useful Life” in Applying Standards of Performance.
Section 111(d) expressly requires EPA to permit states “to take into consideration, among other
factors, the remaining useful life of the existing source to which such standard applies.”51 The
Proposed Rule, however, eliminates state authority to consider remaining useful life in deciding
whether to apply an EPA emission guideline to a source or class of sources and instead allows
states only the supposed flexibility to adjust requirements applicable to specific EGUs, making
some less stringent and others more stringent than might otherwise be the case.52 This is not
consistent with Section 111(d), which does not require that if a state applies a less stringent
standard to a source based on a factor, such as remaining useful life, the state must then extract
47
See Id. at 34,919 (describing EPA’s proposed and alternative provisions for counting emission goals).
Id.
49
Id. at 34,920.
50
Id. at 34,923.
51
42 U.S.C. § 7411(d)(1)(B).
52
79 Fed. Reg. at 34,925.
48
23
further emissions reductions from another source to make up for the less stringent standard. See
Section XX(H&I) for further explanation on remaining useful life of a plant.
Emissions Averaging and Trading. The proposed emission guideline includes EPA’s legal
rationale for why emissions averaging and trading are allowable under Section 111(d) of the
Clean Air Act. APPA agrees that averaging and trading are permissible under Section 111(d).
However, averaging and trading do not justify a more stringent determination by EPA of BSER.
Multi-State Plan Considerations. The Proposed Rule sets forth considerations for multi-state
plans. Of particular significance, EPA seeks comment on joint demonstration of emission
performance in multi-state plans and proposes two alternative approaches to doing that. Under
the first option, the weighted average emission rate goal for a group of participating states is
computed using each state’s emission rate goal from the emission guidelines and the quantity of
electricity generation by affected EGUs in each of those states during the 2012 base year that
EPA used in calculating the state-specific goals. Different levels would be computed for the
interim and final goals.
Under the second option, the weighted average emission rate goal for a group of participating
states is computed using each state-specific emission rate goal and the quantity of projected
electricity generation by affected EGUs in each state. The calculation would be performed for
the 2020 through 2029 period to produce a multi-state interim goal, and for 2030, to produce a
multi-state final goal.53 States should have discretion to decide which approach makes the most
sense for their multi-state jurisdiction. EPA has no basis for disapproving plans that adopt either
of these approaches, or any other reasonable approach, for that matter.
IV.
The Proposed Rule Conflicts with Federal, State, and Local Utility
Laws and Disregards How Electric Markets Work.
The Proposal is contrary to a host of local, state, and federal laws governing the electric utility
industry—including laws in 47 states providing for the creation and governance of state and
municipal public power utilities—by seeking to regulate both the broad matters that Congress
preserved as the province of exclusive state or local regulation and the specific matters Congress
assigned to FERC. EPA’s intrusion into the regulation of these areas of utility governance and
operations, over which it has no jurisdiction and no expertise, has resulted in a Proposal that will
53
79 Fed. Reg. at 34,911-12.
24
cause serious harm to public power utilities, wholesale electric markets, and electric reliability, if
finalized and implemented.
A.
The Proposed Rule Conflicts with the Federal Power Act’s Division of
Regulatory Authority between Federal, State, and Local Governments.
The facilities, services, operations, rates, and governance of the electric utility industry is subject
to regulation at all three levels of government—federal, state, and local. Part II of the Federal
Power Act (FPA)54 establishes a strict division between federal and state roles regarding the
generation, transmission, distribution, and sale of electricity. Section 201(b) of the FPA
recognizes and preserves the states’ traditional authority over electric generation facilities, local
distribution, retail sales of electricity, and intrastate transmission.55 Section 201(a) limits federal
authority to “the transmission of electric energy in interstate commerce and the sale of such
energy at wholesale in interstate commerce.” 56 Section 201(a) also expressly states that federal
authority “extend[s] only to those matters which are not subject to regulation by the States.”57
The provisions of the FPA also generally do not apply to the U.S. and federal agencies (including
federal power marketing administrations), states and municipalities and their agencies (including
state and local public power utilities), and most rural electric cooperatives.58 EPA’s Proposed
Rule impermissibly interjects EPA into all of these federal, state, and local regulatory spheres.
Accordingly, the Proposed Rule is invalid and must be withdrawn.
1.
The Proposed Rule Unlawfully Usurps State Authority Preserved
by the FPA and the Tenth Amendment.
The FPA draws clear lines between state and federal jurisdiction over electricity markets and
generation facilities.59 The Supreme Court has observed that the FPA’s “legislative history is
replete with statements describing Congress’ intent to preserve state jurisdiction over local
54
16 U.S.C. §§ 824–824w (2012)
Id. § 824(b).
56
Id. § 824(a).
57
Id. See Fed.Power Comm’n v. S. Cal. Edison Co., 376 U.S. 205, 218 (1964) (“FPC v. SCE”).
58
16 U.S.C. § 824(f). While the FPA exempts APPA’s members (like other federal and state entities) from FERC’s
jurisdiction under most provisions of the FPA, 16 U.S.C. § 824(f), APPA’s members are generally subject to
pervasive oversight by state, municipal, or other local bodies. Accordingly, to the extent that the Proposed Rule
impacts the price of wholesale sales or transmission by APPA’s members, it is invading the regulatory sphere that
the FPA reserves for state regulation and not the sphere reserved for FERC. This does not alter the fact that the
Proposed Rule improperly invades both state and FERC jurisdictional spheres under the FPA.
59
FPC v. SCE, 376 U.S. at 215 (“Congress meant to draw a bright line easily ascertained, between state and federal
jurisdiction…”).
55
25
facilities.”60 The Court further noted that FERC itself “has recognized that the States retain
significant control over local matters even when retail transmissions are unbundled.”61 As for
state and local public power utilities, the situation is even clearer—Congress expressly stated that
FERC’s general FPA authority does not apply to states and subdivisions of states, unless the FPA
provision expressly so provides.62
Importantly, the Supreme Court has also emphasized that federal regulation by agencies other
than FERC may not invade the domain that the FPA has preserved for the states. In Pacific Gas
& Electric Co. v. State Energy Resources Conservation & Development Commission,63 it
concluded that federal regulation of the safety of nuclear power plants did not override state
authority over “the regulation of electricity production….” The Court explained that states have
“traditional authority over the need for additional generating capacity, the type of generating
facilities to be licensed, land use, ratemaking, and the like.”64 The Court noted that the only
exception to state regulation of these activities was FERC’s authority under the FPA over
interstate wholesale sales and transmission by FERC-jurisdictional utilities.65
Decisions by the U.S. Court of Appeals for the D.C. Circuit likewise confirm that federal
regulation cannot directly or indirectly intrude into the sphere that the FPA reserves for state
regulation. See, e.g., Electric Power Supply Ass’n v. FERC, 753 F.3d 216, 224 (D.C. Cir. 2014)
(“the Federal Power Act unambiguously restricts FERC from regulating the retail market”)
(“EPSA v. FERC”); Duke Power Co. v. FPC, 401 F.2d 930, 935 (D.C. Cir. 1968) (explaining
that the “major emphasis” of the FPA “is upon federal regulation of those aspects of the industry
which—for reasons either legal or practical—are beyond the pale of effective state supervision”);
see also EPSA v. FERC, 753 F.3d at 221 (“noting FERC cannot ‘do indirectly what it could not
do directly’”) (quoting Altamont Gas Transmission Co. v. FERC, 92 F.3d 1239, 1248 (D.C. Cir.
1996)).
60
New York v. FERC, 535 U.S. at 22-23.
New York v. FERC, 535 U.S. 1, 24 (2002) (citing Order No. 888, at 31,782, n.543 “(‘Among other things,
Congress left to the States authority to regulate generation and transmission siting’); Order No. 888, at 31,782, n.544
(‘This Final Rule will not affect or encroach upon state authority in such traditional areas as the authority over local
service issues, including reliability of local service; administration of integrated resource planning and utility buyside and demand-side decisions, including DSM [demand-side management]; authority over utility generation and
resource portfolios; and authority to impose non-bypassable distribution or retail stranded cost charges’)”).
62
16 U.S.C. § 824(f).
63
461 U.S. 190 (1983)
64
Id. at 212.
65
Id. at 205.
61
26
The FPA’s preservation of state regulatory authority of the electric utility industry is consistent
with the Tenth Amendment to the United States Constitution, which declares that “[t]he powers
not delegated to the United States by the Constitution, nor prohibited by it to the States, are
reserved to the States respectively, or to the people.” Before Part II of the FPA was enacted in
1935, states pervasively regulated utilities within their borders based on their general police
powers. Congress adopted the 1935 amendments to the FPA only after the Supreme Court held
that states could not regulate interstate sales of electricity under the Commerce Clause in
Commission of Rhode Island v. Attleboro Steam & Electric Co.66 FERC was given authority
over interstate transmission and interstate wholesale sales of electric energy solely to close this
“Attleboro gap.”67 In so doing, however, Congress chose to confine FERC’s sphere of excusive
regulation to these matters68 and expressly preserved traditional state regulatory authority over
all other sales—including retail sales—and over generation and local distribution facilities.69
Thus, Congress intended to “tak[e] no authority from State Commissions.”70 Moreover,
Congress specifically preserved the autonomy of state and local public power utilities from
plenary FERC regulation.71
EPA’s Proposed Rule ignores the careful preservation of state and local authority embodied in
the FPA and associated constitutional requirements and attempts to claim state authority over
generation matters for EPA. As stated by FERC Commissioner Tony Clark in recent
congressional testimony, the Proposed Rule would “dramatically alter” the “traditional lines of
authority by creating a new paradigm of oversight of net carbon emission from a state” and
potentially result in states “ceding ultimate authority of the regulation of their state’s public
utilities and energy development to the EPA.”72
In enacting Section 111(d) of the CAA, Congress could not possibly have intended to empower
the EPA to take on such a monumental task. As the Supreme Court recently stated, EPA cannot
reasonably interpret a statute to authorize “an enormous and transformative expansion in EPA’s
regulatory authority without clear congressional authorization.”73 EPA can have no greater
66
273 U.S. 83, 90 (1927).
See New England Power Co. v. New Hampshire, 455 U.S. 331, 340 (1982).
68
See FPC v. SCE, 376 U.S. at 215,
69
16 U.S.C. § 824(b).
70
New England Power, 455 U.S. at 341 (quoting H.R. Rep. No. 1318, 74th Cong.,1st Sess. 8 (1935)) (emphasis
omitted).
71
See 16 U.S.C. § 824(f).
72
Written Testimony of Commissioner Tony Clark, Before the Committee on Energy and Commerce Subcommittee
on Energy and Power, United States House of Representatives, Hearing on FERC Perspective: Questions
Concerning EPA’s Proposed Clean Power Plan and other Grid Reliability Challenges at 5 (July 29, 2014) (“Clark
Testimony”).
73
UARG v. EPA, 573 U.S. at 2432.
67
27
authority to address matters Congress has expressly reserved to state regulation than FERC has
under the FPA. Accordingly, the Proposed Rule is unlawful and must be withdrawn and reproposed.
2.
The Proposed Rule Unlawfully Usurps FERC’s Regulatory
Authority Under the FPA.
Although the FPA preserves state and local control over generation resource facilities, their
development and procurement, generation resource adequacy, and generation resource portfolio
diversity, a long line of cases makes clear that FERC has exclusive jurisdiction over interstate
transmission and wholesale sales of electricity. 74 Because the Proposed Rule also would
interfere with areas that fall squarely within FERC’s FPA purview, as described below, it must
be withdrawn and re-proposed.
a.
The Proposed Rule Supplants FERC’s Authority Under
Sections 205 and 206 of the FPA.
Sections 205 and 206 of the FPA govern FERC’s authority to regulate interstate transmission and
wholesale sale of electric energy, and the measures called for by the Proposed Rule would step
directly into these areas, in violation of the FPA. 75 Under Section 205(b), jurisdictional public
utilities must file with FERC “all rates and charges for any transmission or sale subject to the
jurisdiction of the Commission, and the classifications, practices, and regulations affecting such
rate and charges, together with all contracts which in any manner affect or relate to such rates,
charges, classifications, and services.” 76 Under section 205(a), FERC may accept filed rates,
rules, and practices if it determines that they are “just and reasonable.” 77 Courts have affirmed
that the section 205 filing requirement applies broadly to all classifications, practices, and
procedures that “significantly affect” the price or non-price terms and conditions of FERCjurisdictional services.78
74
See, e.g., New York v. FERC, 535 U.S. at 6–7; Nantahala Power & Light Co. v. Thornburg, 476 U.S. 953, 966
(1986) (quoting FPC v. SCE, 376 U.S. at 215-16 (“Congress meant to draw a bright line easily ascertained, between
state and federal jurisdiction….This was done in the [FPA] by making [FERC] jurisdiction plenary and extending it
to all wholesale sales in interstate commerce except those which Congress has made explicitly subject to regulation
by the States.”) (internal quotation marks omitted)).
75
16 U.S.C. §§ 824d, 824e.
76
Id. § 824d(b).
77
Id. § 824d(a).
78
See, e.g., City of Cleveland v. FERC, 773 F.2d 1368, 1376 (D.C. Cir. 1985) (“[T]here is an infinitude of practices
affecting rates and service. The statutory directive must reasonably be read to require the recitation of only those
practices that affect rates and service significantly….”).
28
Section 206 of the FPA requires FERC to revise any filed “rate, charge, or classification,” or any
“rule, regulation, practice, or contract” affecting them, if FERC determines that they are “unjust,
unreasonable, unduly discriminatory, or preferential.”79 In recent decades, FERC, with varying
degrees of success, has generally attempted to rely on open-access, non-discriminatory interstate
transmission service and competitive market forces to ensure that wholesale prices are “just and
reasonable.”80
Implementation of building block 2 would infringe upon the state and local control of generation
resources that the FPA reserves to the states, as explained above. But it also would improperly
impinge on FERC’s section 205 and 206 authority to regulate FERC-jurisdictional public
utilities. Building block 2 requires “environmental dispatch” of generation from higher emitting
sources to favor “EGUs with expanded low- or zero-carbon generation.”81 The Proposed Rule
states that EPA has determined that environmental dispatch is feasible and that its costs are
reasonable.82 But, as stated above, under sections 205 and 206 of the FPA, FERC must
determine if practices that significantly affect the price of its jurisdictional wholesale and
transmission services are just and reasonable. EPA has no role to play and cannot create one for
itself.
To ensure non-discriminatory transmission service and to separate the operation of the
transmission grid from the economic interests of generators, FERC has also encouraged the
creation of Independent System Operators and Regional Transmission Organizations
(collectively “RTOs”).83 These entities, which do not own or operate any generation facilities,
operate the transmission facilities of transmission-owning utilities and “provide open access to
the regional transmission system to all electricity generators at rates established in a single,
unbundled, grid-wide tariff that applies to all eligible users in a non-discriminatory manner.”84
These RTOs are FERC-jurisdictional public utilities, subject to comprehensive FERC regulation
under the FPA.
Implementation of the Proposed Rule’s building block 2 would impermissibly interfere with
FERC-approved generation dispatch and redispatch rules in these RTO regions. In these regions,
which encompass nearly two thirds of the load in the U.S., the dispatch of generation resources is
79
16 U.S.C. § 824e(a).
See Morgan Stanley Capital Group v. Public Util. Dist. No. 1, 544 U.S. 527, 535-38 (2008).
81
79 Fed. Reg. at 34,836.
82
Id. at 34,865 (“We view these estimated costs as reasonable and therefore as supporting the use of a 70 percent
utilization rate target.”).
83
See Midwest ISO Transmission Owners v. FERC, 373 F.3d 1361, 1364 (D.C. Cir. 2004).
84
Id. at 1364 (internal quotations omitted).
80
29
largely governed by RTO operating rules and RTO tariffs that are on file with FERC as required
by section 205 of the FPA and subject to modification by FERC under section 206 of the FPA.85
The following map depicts the territories administered by RTOs. Of these, the New England
ISO (ISO-NE), New York ISO (NYISO), PJM ISO (PJM), Mid-Continent ISO (MISO),
Southwest Power Pool (SPP), and California ISO (CAISO) are all FERC-jurisdictional RTOs.
The Electric Reliability Council of Texas (ERCOT) employs dispatch systems like those used in
the other RTOs but is exempt from FERC regulation under section 205 of the FPA.
Figure 1: Regional Transmission Organizations (RTO)/Independent System Operators
(ISO)
See http://www.ferc.gov/industries/electric/indus-act/rto.asp.
EPA has no authority to modify RTO tariffs or generation-dispatch rules.86 But that is precisely
what the Proposed Rule attempts to do through building block 2. Indeed, FERC Commissioner
Philip Moeller has stated that FERC-jurisdictional RTO markets “would need to be
85
See, e.g., Security Constrained Economic Dispatch: Definition, Practices, Issues, and Recommendations: A
Report to Congress Regarding the Recommendations of Regional Joint Boards for the Study of Economic Dispatch
Pursuant to Section 223 of the Federal Power Act, FERC (July 31, 2006) (describing existing dispatch regimes,
their emphasis on security-constrained economic dispatch, and the role of RTOs in administering them).
86
See, e.g., Atlantic City Electric Co. v. FERC, 295 F.3d 1 (D.C. Cir. 2002).
30
fundamentally altered and redesigned to implement EPA’s proposal to accommodate
environmental dispatch.”87 These are regulatory determinations that only FERC can make under
the FPA with respect to FERC-jurisdictional RTOs.
Similar limits on EPA authority over generation dispatch by FERC-jurisdictional public utilities
apply in areas of the nation that do not have RTOs. In those regions, FERC-jurisdictional public
utilities provide open-access, non-discriminatory transmission service under tariffs that are
modeled closely on FERC’s pro forma open-access transmission tariff (OATT), and many nonFERC-jurisdictional federal power marketing administrations, state and local public power
utilities, and rural electric cooperatives provide transmission service under comparable
“reciprocity” tariffs or through other arrangements.88 In these non-RTO regions, FERCjurisdictional public utilities sell wholesale power under tariffs and contracts subject to exclusive
FERC regulation.89 These tariffs and contracts may govern the dispatch of EGUs. In any event,
these tariffs and contracts can be amended only by the public utilities or by FERC under sections
205 and 206 of the FPA—not by EPA or state regulators.90 These public utilities’ retail sales,
local distribution facilities and services, and generation facilities remain subject to
comprehensive state regulation, as already noted. The generation portfolios of these public
utilities, and the dispatch of the units in these portfolios to provide retail service, are subject to
comprehensive oversight by state utility regulatory authorities—not by state environmental
officials and not by EPA.
In the case of public power utilities, EPA’s lack of authority over their operations is even clearer.
The generation dispatch of public power utilities is not subject to plenary FERC regulation91 and,
in most states, is not subject to oversight by a state regulator. The makeup of public power
generation portfolios may be subject to state regulation in the issuing of certificates of public
convenience and necessity for new generating facilities, and in some states, by renewable
portfolio standards, but public power utilities are not generally subject to plenary state
regulation; indeed, some state constitutions forbid such regulation, as already noted. Consistent
87
Commissioner Philip Moeller’s Answers to Preliminary Questions for the Federal Energy Regulatory
Commission, House Energy and Commerce Committee, Subcommittee on Energy & Power (July 29, 2014).
88
See Preventing Undue Discrimination and Preference in Transmission Service, Order No. 890, FERC Stats. &
Regs. ¶ 31,241, order on reh’g, Order No. 890-A, FERC Stats. & Regs. ¶ 31,261 (2007), order on reh’g, Order No.
890-B, 123 FERC ¶ 61,299 (2008), order on reh’g, Order No. 890-C, 126 FERC ¶ 61,228, order on clarification,
Order No. 890-D, 129 FERC ¶ 61,126 (2009) (establishing the current version of the pro forma OATT used by most
FERC-jurisdictional utilities that are not participating in RTOs).
89
See, e.g., Nantahala, 476 U.S. at 966; FPC v. SCE, 376 U.S. at 215.
90
See Miss. Power & Light Co. v. Mississippi, 487 U.S. 354, 369-74 (1988); Ark. La. Gas Co. v. Hall, 453 U.S.
571, 577-78 (1981).
91
See 16 U.S.C. § 824(f),
31
with the animating principle of public power—local control—the operation of the EGUs of a
public power utility are assigned under state and local laws to the sound business judgment of the
public power utility, subject to the authority of the utility’s governing officials or board, who
may be elected or appointed (e.g., a mayor, city manager, or utility board or commission).
Ultimately the managers and governing officials of a public power utility must answer to the
customers in the local community.
Finally, EPA attempts to justify its intrusion into these areas by noting that the RTOs’ existing
“least-cost” economic dispatch rules can account for existing environmental regulations by
permitting sellers to include environmental costs in their offers. 92 This is no defense. As noted
above, EPA has proposed a radical transformation of electric dispatch rules that is nothing like
current RTO rules, and, regardless, EPA has no authority to take such action.
b.
The Proposed Rule Supplants FERC’s Authority Under
Section 202(a) of the FPA.
Section 202(a) of the FPA authorizes FERC to “divide the country into regional districts for the
voluntary interconnection and coordination of facilities for the generation, transmission, and sale
of electric energy” in order to ensure “an abundant supply of electric energy” with “the greatest
possible economy” and with regard to the “proper conservation and utilization of natural
resources.”93 The Proposed Rule would override FERC’s authority under section 202(a) by
promoting multi-state plans to implement EPA’s environmental dispatch policies in regions
effectively established by EPA based on other criteria.
Moreover, FERC’s section 202(a) authority to “divide the country into regional districts” is for
the “voluntary … coordination” of generation facilities in utility operations.94 To the extent the
Proposed Rule requires the regional, coordinated operation of FERC-regulated generation
facilities, EPA would be supplanting Congress’ judgment that these matters be left to
administration by FERC.
92
79 Fed. Reg. at 34,862.
16 U.S.C. § 824a.
94
See, e.g., Atlantic City Elec., 295 F.3d at 12 (citing Duke Power Co., 401 F.2d at 943). Cf. S. Car. Pub. Serv.
Auth. v. FERC, Nos. 12-1232 et al., slip op. at 25–31 (D.C. Cir. Aug. 14, 2014) (upholding FERC construction of
section 202(a) to include coordinated operations but not planning).
93
32
c.
The Proposed Rule Supplants FERC’s Authority Under
Section 215 of the FPA.
Section 215 of the FPA authorizes FERC to certify an electric reliability organization to develop
and enforce mandatory reliability standards for the nation’s bulk-power system, and further
authorizes FERC to approve and enforce those reliability standards.95 Pursuant to this authority,
FERC has certified the NERC as the electric reliability organization. NERC and its various
regional reliability entities now administer a comprehensive set of mandatory reliability
standards subject to FERC’s oversight. The Proposed Rule acknowledges that reliability is an
issue of concern, but ultimately rests on a conclusion that it provides sufficient flexibility to
avoid reliability concerns. 96
EPA has not adequately consulted with FERC or NERC regarding its conclusion on the potential
reliability implications of the Proposed Rule or other issues related to NERC’s reliability
standards approved by FERC. There is every indication, however, that the massive changes in
generation dispatch that the Proposed Rule would require will lead to conflicts between EPA’s
policies and NERC’s reliability standards. Moreover, by disregarding the FERC-approved
reliability standards and by making its own reliability-related determinations, EPA contravenes
Section 215 of the FPA by supplanting FERC as the ultimate authority over the reliability of the
bulk-power system. See Section XXI for more discussion of APPA’s concerns regarding
impacts on reliability.
B.
The Proposed Rule Relies on a Flawed Understanding of Regulated
Wholesale Electricity Markets and the Bulk Power System.
The Proposal reflects fundamental misunderstandings of FERC-regulated wholesale electricity
markets in RTO regions and the bulk power system. For that reason, it is arbitrary and
capricious and must be withdrawn and re-proposed. See, e.g., Motor Vehicle Mfrs. Ass’n v. State
Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983) (agency action will be upheld only if it
“articulates a satisfactory explanation for its action including a ‘rational connection between the
facts found and the choice made’”) (quoting Burlington Truck Lines, Inc. v. United States, 371
U.S. 156, 168 (1962)).
95
96
16 U.S.C. § 824o
79 Fed. Reg. at 34,836.
33
1.
EPA Has Failed to Address Reliability Issues.
As noted above, the Proposed Rule rests on a finding that it will not significantly impact electric
reliability. That finding, however, involved no substantive coordination with FERC, the agency
with statutory responsibility for establishing reliability standards. EPA’s failure to adequately
consult with, and defer to, FERC concerning issues that fall squarely within its expertise is
arbitrary and capricious.
2.
EPA Misunderstands the Role of the States in Regulating Dispatch
in RTO Regions.
As discussed above, the Proposed Rule assumes that states can regulate dispatch and require
environmental dispatch to implement the measures contemplated in building block 2. In RTO
regions, however, such matters are managed by RTOs pursuant to FERC-approved rules or
directly through FERC-approved tariffs. The Proposal is arbitrary and capricious because, in
failing to recognize FERC’s role and its statutory obligations when regulating dispatch, it would
create serious regulatory conflicts. As explained by FERC Commissioner Tony Clark:
[E]ven if all states in a region band together under the regional grid
operator, any changes in the wholesale markets must necessarily be
vetted and approved by FERC. The Commission would be
charged with the awkward task of evaluating fundamental
wholesale market design changes driven by environmental
priorities approved by EPA. Yet FERC is an economic and
reliability regulator. Any decisions made by FERC must be rooted
not in the Clean Air Act, but in our “just and reasonable” and not
“unduly discriminatory or preferential” rate standard in the Federal
Power Act. FERC’s ability to alter or reject an RTO-proposed
compliance mechanism would present a conflict with EPA’s
evaluation of the compliance plans. Absent Congress stepping in
and clearly defining FERC authority and EPA authority, it is hard
not to envision a future jurisdictional train wreck. 97
Similarly, if states do not coordinate their policies, then “regional grid operators will be faced
with an increasingly complex task of implementing multiple compliance mechanisms into what
97
See Clark Testimony at 7.
34
was once an efficiently dispatched regional electric grid.” 98 FERC Commissioner Philip Moeller
has also explained that because electricity markets are actually “interstate in nature,” the
Proposed Rule’s “state-by-state approach results in an enforcement regime that would be
awkward at best, and potentially very inefficient and expensive.” 99
EPA’s failure to consider these complications has resulted in an unworkable Proposed Rule. It is
therefore arbitrary and capricious.
3.
EPA Has Failed to Take into Account Impediments to Deploying
the New Energy Infrastructure That the Proposed Rule Would
Necessitate.
EPA believes that “system operators typically have flexibility to choose among multiple EGUs
when selecting where to obtain the next [megawatt hours (MWh)] of generation needed” and that
electricity is “fungible.” 100 It also believes that the natural gas transmission system is capable of
supporting the increase to 70 percent utilization of natural gas combined cycle (NGCC) units
envisioned by the Agency to result from implementation of the Proposed Rule. 101 These
assumptions are not warranted.
FERC commissioners have testified to Congress that EPA’s assumptions are likely unrealistic
and that adequate infrastructure would not likely be in place in time to ensure compliance with
the Proposed Rule. As Commissioner Moeller has explained, natural gas infrastructure
constraints will likely exist because it is difficult to finance construction of new pipelines before
new markets are established.102 EPA has also given too little consideration to constraints
imposed by the nature of the electric transmission grid. As Commissioner Moeller informed
Congress:
As we have seen with the implementation of EPA’s mercury rule
(MATS), load pockets matter because the laws of physics trump
written words. Although a specific generating plant may not
contribute significant power to the grid, its other output such as
98
Id. at 6.
Written Testimony of Commissioner Philip D. Moeller, Before the Committee on Energy and Commerce
Subcommittee on Energy and Power United States House of Representatives, Hearing on FERC Perspective:
Questions Concerning EPA’s Proposed Clean Power Plan and other Grid Reliability Challenges at 3 (July 29,
2014) (“Moeller Testimony”).
100
79 Fed. Reg. at 34,880.
101
Id. at 34,863-64.
102
See Moeller Testimony at 3-4.
99
35
voltage support or “inertia” qualities may contribute significantly
to grid stability. Moreover, the details of how reserve margins are
calculated can have a significant impact on the ability of excess
capacity in one load pocket to transfer power to another load
pocket that is short. These challenges can be addressed but it takes
engineering expertise, especially when designing optimal
infrastructure improvements.103
Similarly, EPA has failed to consider transmission grid “integration issues” (e.g., voltage control,
natural gas backup power, etc.) that would have to be addressed in order to accommodate the
substantial influx of renewable resources that the Proposed Rule contemplates.104 All of these
serious shortcomings go to the heart of the Proposed Rule and render it arbitrary and capricious.
4.
EPA Has Not Accounted for Recent Developments Regarding the
Participation of Demand-Side Resources in the Electricity
Markets.
The D.C. Circuit’s recent decision in EPSA v. FERC has significant ramifications for the
Proposed Rule that EPA has not taken into account. Building block 4 of the Proposal involves
“[r]educing emissions from affected EGUs in the amount that results from the use of demandside energy efficiency that reduces the amount of generation required.” 105 In EPSA, however,
the court ruled that FERC’s wholesale jurisdiction does not extend to the regulation of putative
“wholesale sales” by demand-response resources because that would constitute “direct
regulation” of state-jurisdictional retail markets. 106 EPSA v. FERC thus holds that demandresponse resources may not be permitted to participate as sellers of wholesale energy in FERCjurisdictional, RTO-administered markets. Some have argued that this holding would also apply
to demand-response resources participating as sellers in RTO capacity markets.
At the same time, the EPSA decision does not disturb state-administered demand-response
programs and does not foreclose the recognition in wholesale markets of such demand-response
programs in wholesale demand. The court of appeals has denied rehearing of its decision, but
has stayed the effectiveness of its decision to allow FERC time to seek U.S. Supreme Court
review. Thus, no one predict the final outcome of the case at this juncture. APPA believes that
the EPSA decision respects state and local authority over demand-response resources and will, in
103
Id. at 7.
Id. at 1, 7.
105
79 Fed. Reg. at 34,836.
106
EPSA, 753 F.3d at 222, 224.
104
36
the long run, better enable the development of demand response in wholesale markets. Thus, the
EPSA decision is not an impediment to EPA’s long-term goals. But since the decision has
vacated FERC’s attempt to supplant state and local authority over retail demand response, to the
extent EPA predicated the Proposed Rule on the existence of FERC’s now-vacated demandresponse regime, EPA may need to reevaluate its assessment of this building block and
appropriately adjust each state’s near-term emission goals.
5.
The Proposed Rule Is Based on Fundamental Misunderstandings
of RTO Capacity Markets and Their Potential to Facilitate EPA’s
Goals.
The Proposed Rule relies in part on the RTO capacity markets to serve the policies that the
Proposed Rule seeks to implement. EPA misunderstands, however, how those markets function.
For instance, EPA asserts that in states where RTOs manage markets, the RTOs “administer[]
auctions for forward capacity” in which generators “all compete to provide potential resources
for meeting the projected demand for electricity services.” 107 But only two of the five RTOs
administer such auctions.
Further, RTO capacity markets are extremely controversial and subject to many pending legal
disputes. Many FERC stakeholders, including APPA, have argued for years that RTO capacity
markets have failed to serve their intended basic function, i.e., to procure capacity needed for
reliability in an economically efficient way, and thus should be substantially modified, or
eliminated—not expanded in scope or introduced to new regions. A major open FERC docket is
currently considering the core design and purpose of RTO capacity markets. 108 Individual RTOs
are likewise continuously making alterations that seek to correct, in APPA’s view
unsuccessfully, identified flaws in existing capacity markets. 109 Consequently, it is not reasoned
decision-making for EPA simply to assume that RTO capacity markets can take on the new
function of advancing the Proposed Rule’s objectives. 110 More discussion of these issues is
contained in Section XX of these comments.
107
79 Fed. Reg. at 34,881.
See Centralized Capacity Market Design Elements, Docket No. AD13-7-000.
109
See, e.g., PJM Capacity Performance Proposal, PJM Staff Proposal (Aug. 20, 2014) (preliminary proposal to
establish new categories of capacity products to address defects in existing PJM capacity market designs subject to
eventual filing with, and review by, FERC under the FPA).
110
See Markets Matter: Expect a Bumpy Ride on the Road to Reduced CO2 Emissions, Navigant Consulting (May
2014) (analysis, funded by APPA, describing long-running RTO capacity market problems including the ways in
which existing RTO capacity market structures impede the development of renewable energy and energy efficiency
resources) (Attachment 3 to these Comments).
108
37
EPA needs to withdraw the Proposed Rule, fully engage with FERC regarding these issues, and
revise its Proposed Rule to reflect the actual market conditions under which any Section 111(d)
regulations would be implemented.
V.
New Source Review (NSR) Issues
Building block 1 of the Proposed Rule consists of measures to reduce CO 2 emissions through
heat rate improvements at affected EGUs. EPA explains that these heat rate improvements can
be achieved by “installing and using equipment upgrades … such as extensive overhaul or
upgrade of major equipment (turbine or boiler) or replacing existing components with improved
versions.”111
EPA has targeted these sorts of projects as triggers for compliance with the Clean Air Act’s NSR
requirements. Indeed, EPA has previously determined that a project’s potential to improve an
EGU’s efficiency strongly supports a finding that the project is not excluded from NSR
requirements as “routine maintenance, repair or replacement” (RMRR).112 EPA’s policy
position is incorrect as a matter of law, 113 but it has nevertheless led to hundreds of NSR
enforcement actions and citizen suits targeting the very sort of projects that the Proposed Rule
would now seek to require EGUs to undertake. See Comments of the Utility Air Regulatory
Group (listing projects of the same type identified in the Sargent & Lundy report and the
Technical Support Document for GHG Abatement Measures that EPA or citizens have claimed
have violated NSR requirements).
Despite its own stated policy positions and recent history, EPA claims in the Proposed Rule that
there will be “few instances” where “an NSR permit would be required” as a result of
implementing building block 1 measures.114 But EGUs cannot rely on this vague statement.
Indeed, EPA has previously given utilities such assurances with respect to NSR only to reverse
111
Technical Support Document for GHG Abatement Measures at 2-16; see also Sargent & Lundy LLC, COALFIRED POWER PLANT HEAT RATE REDUCTIONS at 2-1 to 5-4 (Jan. 22, 2009) (cited by EPA and identifying projects
to improve heat rate).
112
See Letter from Francis X. Lyons, Reg’l Adm’r, EPA, to Henry Nickel at 2-5 (May 23, 2000), available at
www.epa.gov/ttn/nsr/gen/letterf3.pdf.
113
See, e.g., Nat’l Parks Conservation Ass’n v. TVA, No. 3:01-cv-71, 2010 WL 1291335 (E.D. Tenn. Mar. 31, 2010)
(finding economizer and superheater replacements to be RMRR); Pennsylvania DEP v. Allegheny Energy, Inc., No.
05-885, 2014 WL 494574 (W.D. Pa. Feb. 6, 2014) (finding superheater, lower slope panel and reheater
replacements RMRR). But see United States v. Louisiana Generating LLC, No. 09-100, 2012 WL 4107129 (M.D.
La. Sept. 19, 2012) (finding reheater replacements not to be RMRR).
114
79 Fed. Reg. at 34,859.
38
course and seek penalties for alleged NSR violations years later. 115 Regardless of EPA’s current
policy position, it is apparent that environmental organizations and others will target EGUs
undertaking heat rate improvement projects as a result of section 111(d). Even if those suits lack
merit—and APPA believes that they undoubtedly will—they will take years and consume
enormous resources to litigate.116 As not-for-profit, government-owned entities, public power
utilities cannot risk such expensive and time-consuming litigation.
Despite these significant risks, EPA has failed to provide a clear statement and proposed
regulatory text ensuring that any project necessary to implement building block 1 will not trigger
NSR. EPA’s only proposed solution is that states attempt to fix this problem of EPA’s creation.
EPA first suggests that states somehow attempt to balance any increased utilization of more
efficient units by adjusting demand-side management and renewable energy requirements. The
Agency provides no indication, however, of how states reasonably could be expected to achieve
this. Moreover, whether such an approach could offer sufficient protection from liability is
questionable given that EPA has maintained that it can pursue alleged NSR violations even when
actual emissions decrease and challenge a projection of post-project emissions performed by a
utility. EPA’s second suggestion is to impose synthetic minor limits on all coal-fired sources.
But this approach would eliminate almost all flexibility that states otherwise might have to
implement section 111(d) requirements.
VI.
The Proposed Rule Contains Many Inequities and Is Unfair in Many
Key Respects.
The Proposed Rule is unlike nearly every other significant EPA regulatory action. Apart from
the proposed procedures for the submission of state plans, most of the Proposal is a description
of the policy decisions underlying the proposed state CO2 emission goals. The policy decisions
and the analysis supporting them, however, are deeply flawed and result in a Proposal that places
enormous burdens on the electric utility industry. EPA should withdraw and re-propose the
Proposed Rule for these reasons alone.
115
See, e.g., United States v. Ala. Power Co., 681 F. Supp. 2d 1292, 1310 (N.D. Ala. 2008) (EPA “could not tell
Congress it envisioned very few future WEPCO-type enforcement actions on the one hand, and then argue in
subsequent enforcement actions that the utility industry was unreasonable in relying on those, or similar, EPA
statements.”).
116
See, e.g., Pennsylvania v. Allegheny Energy, Inc., No. 05-885 (W.D. Pa.) (NSR citizen suit that took
approximately nine years to litigate, resulting in dismissal of all claims).
39
A.
State Goal Computation
EPA applied the four building blocks to each state’s 2012 electric generation to determine state
goals for CO2 reductions.117 EPA’s methodology for applying the building blocks is difficult to
interpret. But it is apparent that, in many instances, EPA has simply assumed that individual
states will be able to implement the building blocks, without closely examining whether this is
feasible in each case. This in itself is unreasonable and inadequate. EPA also claims that it is
imposing “reasonable … rather than the maximum” levels for each building block and that states
can compensate for not using one building block by increasing the use of another. 118 But
because EPA’s one-size-fits-all approach does not take into account the actual limitations on the
states’ abilities to implement the building blocks, this purported “flexibility” is in many cases
simply nonexistent.
EPA’s assumptions underlying the state targets are also questionable. In particular, EPA has
included NGCC units that are under construction in its applications of building block 2. 119 Those
units, however, are “new” sources under Section 111 and cannot be regulated under the existing
source rules.120 Moreover, in setting state goals based on units that may never be completed,
EPA distorts its assessment of achievable emission reductions.
EPA has further distorted its assessment of achievable state goals by inappropriately including
units smaller than the Proposal’s 73 MW applicability threshold in its calculations of the state
goals. EPA has provided no justification for its decision to include these units in setting state
goals.
B.
Early Action Credit
When determining BSER, it is EPA’s responsibility to identify what emission reduction systems
exist and how much they reduce air pollution in practice. This allows EPA to identify potential
emission limits for the purpose of evaluating each limit in conjunction with its costs and benefits.
EPA may not prescribe a particular technological system that must be used to comply with a
New Source Performance Standards (NSPS).121 However, by departing from the NSPS power
plant “fence line” in determining the “best” modeling technique used to identify reductions for
CO2, EPA’s approach was flawed in that it arbitrarily stretched an identified “best” measure to
117
See 79 Fed. Reg. at 34,863.
Id.
119
Id. at 34,877.
120
CAA § 111(a)(2).
121
42 U.S. Code § 7411, CAA § 111(b)(5).
118
40
match with states that had already spent effort to appropriately reduce CO 2 emissions. In
glossing over prior actions taken to reduce CO 2, EPA effectively ignored the situational increase
in compliance costs with the Proposal and set each building block measure at a level that is too
stringent to reflect practical reality. Because EPA does not fully consider early action, the
Proposal imposes undue additional compliance costs on states and entities. For further
examples, please see section “XIII. The Baseline and BSER Computations Should Allow Full
Credit for Early Action.”
C.
Transmission Lines and Natural Gas Pipelines
EPA’s Proposed Rule for existing sources cannot be implemented without significant new
generation from natural gas and renewable energy sources. Increased utilization of such sources
will require major changes to the nation’s natural gas pipeline and electric transmission
infrastructure as discussed in greater detail in Sections VII, VIII, and IX. Making these
infrastructure improvements will require lengthy permitting processes involving local, state, and
federal agencies. Those permitting actions, moreover, are all likely to be targets of litigation,
further slowing the path to compliance with any Section 111(d) requirements.
D.
Interaction with Other Clean Air Act Rules
EPA has not adequately considered how implementation of the Proposed Rule’s requirements
will impact sources that are already complying with other federal requirements, including EPA
rules. For instance, EPA projects that that the Proposed Rule will require Arizona to shut down
all coal-fired generation by 2020 to achieve the state’s interim target and the final target in 2030.
Pursuant to EPA’s own regional haze rule for three facilities in that state, however, utilities are
currently in the process of investing hundreds of millions of dollars to install controls that limit
emissions of nitrogen oxides (NOx) at units that would have to cease operation less than three
years after those projects must be completed.
In addition, the Proposed Rule will require ramping up generation from facilities that are located
in current nonattainment areas, which would violate NAAQS requirements. As EPA revises the
NAAQS, the number of nonattainment areas will grow, causing even greater problems.
41
E.
Public Health Benefits
EPA claims the Proposed Rule will lead to billions of dollars in health co-benefits due to
reductions in pollutants other than CO 2, namely ozone and fine particulate matter. 122 These
pollutants are independently regulated under the NAAQS program. Accordingly, to the extent
EPA relies on emission reductions that will otherwise be required by compliance with the
NAAQS, those benefits cannot be attributed to the Proposed Rule. Indeed, EPA itself
acknowledges that it may be double-counting benefits.123
The NAAQS, moreover, reflect the level of air pollution that EPA itself has determined is
requisite to protect the public health and welfare. 124 If EPA believes the NAAQS do not
sufficiently protect public health, the proper way to address that problem is through the NAAQS
themselves—not a “back door” amendment through new CO 2 regulations.
VII.
EPA’s Premise That a Significant Portion of the CO2 Reductions the
Proposed Rule Seeks to Achieve Can Be Done Through Fuel
Switching from Coal to Natural Gas Is Based on Questionable
Assumptions Regarding Natural Gas Supply, Price, and
Infrastructure Availability.
The Proposed Rule states that it would reduce nationwide CO 2 emissions from the power sector
by approximately 30 percent from 2005 levels by 2030. A significant portion of this reduction
would be achieved through fuel switching from coal to natural gas, which when burned to
generate electricity emits about half the CO 2. This is a matter of high irony given that the large
amount of coal generation in existence today is a direct result of action by the federal
government.
Passage of the Powerplant and Industrial Fuel Use Act in 1978 by Congress restricted the
construction of natural gas-fueled power plants.125 Electric utilities were not allowed to
construct new natural gas plants until passage of the Natural Gas Utilization Act of 1987, which
repealed the provisions of the Powerplant and Industrial Fuel Use Act.126 Federal policies such
122
See 79 Fed. Reg. at 34,939, Table 16.
Regulatory Impact Analysis at 4-15 (estimated benefits of the Proposal “may account for the same air quality
improvements as estimated in the illustrative NAAQS [regulatory impact analyses]”).
124
CAA § 109(b).
125
See http://www.eoearth.org/view/article/155329/
126
Id.
123
42
as these and others promoted coal as a fuel source for electric generation. By 1987, 56.9 percent
of all electric generation was from coal. 127 It should be noted that this Emily Litella “never
mind” episode foisted on consumers and the electric power industry was the product of natural
gas supply forecasts by the natural gas industry and the federal government.
In the Proposal, EPA assumes relative ease to the power sector and low cost to consumers in
essentially requiring a massive switch from coal to gas. This assumption by EPA is problematic
given it fails to take into account a host of issues that are very likely to impact the supply and
price of natural gas, as well as impediments to the expansion of natural gas infrastructure needed
to facilitate fuel switching. For example, the agency assumes relatively flat long-term natural
gas prices and ample supply, but does not take into account how increased demand for natural
gas by domestic manufacturers, the transportation sector, and international markets is likely to
reduce supply and increase prices. The agency also assumes that natural gas infrastructure can
be expanded to meet the significant increased demand of electric utilities in a relatively short
amount of time. EPA needs to reexamine these assumptions and adjust its timetables for
reducing CO2 emissions from fuel switching.
A.
EPA Assertions About Natural Gas Supply Fail to Adequately Account
for the Difficulty of Projecting Unconventional (Shale) Natural Gas
Supplies as Well as Other Factors That Could Impact Supply.
APPA has filed extensive comments in earlier proceedings at EPA regarding its concerns about
fuel switching. These concerns were largely based on a study commissioned by APPA from the
Aspen Environmental Group in 2010 entitled, Implications of Greater Reliance on Natural Gas
for Electricity Generation (APPA Natural Gas Report).128 However, it does not appear that EPA
has taken into consideration many of the concerns APPA raised, and those concerns not only
remain, but are heightened by the rapidity with which the Proposed Rule assumes increases in
electricity generation from natural gas. While the data in the APPA Natural Gas Report has not
been updated, it is still relevant in showing the potential impacts of a significant increase in
natural gas use by the electric utility industry. This increase in demand could put upward
pressure on prices even if natural gas supply were to remain constant. EPA’s Proposed Rule
assumes that prices will remain relatively flat, but nothing in the Proposal or Technical
127
See EIA Electricity Net Generation: Electric Power Sector, 1949-2007.
APPA’s report was prepared by Aspen Environmental Group
http://www.publicpower.org/files/PDFs/ImplicationsOfGreaterRelianceOnNGforElectricityGeneration.pdf, EPAHQ-OAR-2011-0660 (http://www.publicpower.org/files/PDFs/APPA-NSPS-Comments-WithAttachmentsFinal.pdf), and EPA-HQ-OAR-2013-0495
(http://www.publicpower.org/files/PDFs/APPA2014NewPlantGHGNSPSCommentsWithAttachments.pdf).
128
43
Supporting Documents addresses the question of whether shale gas production can actually occur
at the levels needed to keep prices flat.
1.
Shale Gas Reserves Are More Difficult to Project Than
Conventional Gas Reserves.
EIA’s Annual Energy Outlook (AEO) 2013 Early Release projects U.S. natural gas production to
increase from 23 trillion cubic feet in 2011 to 33.1 trillion cubic feet in 2040, a 44 percent
increase.129 Almost all of this projected increase in domestic natural gas production is from shale
gas production, which is expected to grow from 7.8 trillion cubic feet in 2011 to 16.7 trillion
cubic feet in 2040.130 Due to the fact that shale gas reserves are more difficult to project than
conventional gas reserves, EIA has had to revise its shale gas numbers downwards several times
in the last five years. Nothing in the Proposed Rule or Technical Support Documents
acknowledges these downward projections or what that could mean for long-term supply.
Without new natural gas production from shale gas (and oil) formations, it would be extremely
difficult to achieve the CO2 emissions reductions required by the Proposal. Natural gas
production from conventional supplies (found in reservoirs) has decreased over the last thirty
years and is expected to continue to do so. Various sources have made widely divergent
predictions. Thus, at best, it remains an open question just how much shale gas exists in the
U.S., and more fundamentally, what the production rate will be. While technological
innovations and efficiencies have made directional drilling far more effective, the brittleness of
rocks in shale formations makes it much more difficult to predict their long-term output.
As the figure below illustrates, from 1975 to 2010, all conventional gas wells tended to have
lower production after 10, 20, or 30 years. Available shale gas production records from a few
states (predominantly the Texas Railroad Commission 131) show shale gas wells tend to deplete
significantly after about three years.
129
http://www.eia.gov/forecasts/aeo/pdf/0383(2013).pdf
Id.
131
http://webapps.rrc.state.tx.us/PDQ/home.do and http://www.rrc.state.tx.us/oil-gas/research-and-statistics/
130
44
Figure 2: U.S. Dry Natural Gas Production 132
(trillion cubic feet)
Source: U.S. Energy Information Administration, Annual Energy Outlook 2013 Early Release
Insufficient information exists thus far to predict whether the dry shale natural gas plays in the
Utica, Marcellus, and other formations will be as lucrative as those in the Williston Basin and the
Eagle Ford Shale formation. The availability and timeliness of drilling reports varies among
states, further complicating the ability to predict supply. Moreover, EPA staff may not be
familiar with all the nuances of drilling reports or the fact that many EIA reports comingle liquid
natural gas or wet and dry data when reporting on annual drilling and production records. Such a
commingling can inadvertently provide an overly optimistic picture of shale gas production.
2.
EPA Has Failed to Take into Account the Varying Accuracy of
EIA Projections.
Neither the Proposed Rule nor its Technical Support Documents address concerns about the
historic variations of EIA projections of natural gas supply. Estimates regarding production from
shale gas formations have varied widely, and as mentioned earlier, had to be adjusted downward
several times due to a variety of factors, including poor methodology and inaccurate or untimely
132
http://www.eia.gov/energy_in_brief/article/about_shale_gas.cfm
45
information, among others. EPA appears to take it on faith that EIA’s current long-term
estimates of low-cost, plentiful domestic natural gas supply are accurate. APPA urges EPA to
examine these projections carefully and to incorporate analyses from other respected sources to
arrive at a more balanced assumption about future supply.
According to EIA, the Marcellus formation, which stretches from New York to West Virginia,
produced about 15.6 billion cubic feet of natural gas per day in August 2014, about 38 percent of
total U.S. natural gas production for the month. 133 However, records are minimal for dry gas
production from the Marcellus and Utica formations. Given how little information actually
exists about production, continuing public concern about fracking, and other issues such as water
and land use, it seems premature to assume these formations will be able to produce the gas
needed for large-scale and long-term fuel switching.
3.
EPA Has Failed to Consider the Impact of Liquefied Natural Gas
(LNG) Exports and Increased Manufacturing and Transportation
Demand on Supply.
The Proposed Rule and accompanying Technical Support Documents also fail to take into
consideration the impact exports of LNG and non-electric utility demand will have on natural gas
supply and prices. Given the recent applications to the Department of Energy (DOE) for
approval to construct new LNG export terminals (see Figure 3), U.S. producers will begin
exporting natural gas to European and Asian markets where demand for gas is also increasing
and prices are higher than in the U.S. Domestic demand for natural gas is also projected to
increase for use by manufacturers as feedstock and by the transportation sector as a fuel
source.134 However, EPA apparently has not examined any of these issues that could impact the
supply available for electric utilities or the price they would pay for the gas. The agency needs
to look at these issues and accordingly revise its assumptions about the timing and feasibility of
pervasive fuel switching from coal to natural gas to reduce CO 2 emissions from power plants.
a.
Liquefied Natural Gas
Not long ago, the U.S. was an importer of LNG. However, shale gas production has increased
domestic supply significantly and reduced prices. These reduced prices, along with international
demand for natural gas, have led to recent efforts by domestic producers to export their gas as
133
The Fuel Fix, Aug. 27, 2014http://fuelfix.com/blog/2014/08/27/plenty-of-pluck-left-in-the-marcellus-report-says/
See Financial Times article, U.S. Sees $90bn Boost from Shale Gas Boom,” December 14, 2012, available at
http://www.ft.com/cms/s/0/4b3f6280-4609-11e2-ae8d-00144feabdc0.html#axzz3ES0K0bQY. Also see copy of the
Wyden-Dow $80 Billion list is inserted into Attachments Section.
134
46
LNG. In addition, European concerns about reliance on Russian gas have led to calls by some
for the U.S. to export gas to the continent. While it is still unclear how much domestic natural
gas may be exported, it is very likely that exports would put upward pressure on domestic prices,
which would impact electricity rates. EPA needs to factor this real possibility into its
assumptions about how much fuel switching from coal to natural gas can occur to reduce CO2
emissions from the utility sector.
Domestic natural gas producers have a lot of incentive to export LNG. As of fall 2014,
international natural gas is trading at about $16 Mcf. In the U.S., it is trading at $4.00 Mcf.135
Thus far, more than 25 applications for LNG export terminals have been filed with DOE, and
several have been approved. Further review is required by FERC or the State Department. It is
unclear how many of these facilities, once approved, will be constructed or how much natural
gas will be exported. But given the great demand for natural gas in Europe and Asia, where
prices are much higher than they are in the U.S., it is very likely the U.S. will start exporting
LNG, which will impact domestic prices, as well as the long-term supply of natural gas for use
by the utility, manufacturing, and other sectors.
Most LNG facility applications at FERC would require substantial construction time. However,
the Sabine-Golden Pass project136 that was approved in late September 2014 neither requires
much time nor needs much new construction. The project is currently an LNG import terminal,
whose owners, ExxonMobil and the Qatar Government, plan to change the directional flow to
enable the export of an amount of gas that is approximately three percent of all U.S. natural gas
production per day. Unlike other LNG terminal projects pending FERC approval that also need
state and local permitting approval, the Sabine-Golden Pass project does not need such approvals
and thus will begin exports in the near future.
In addition, EPA has not looked at the impact Canadian LNG exports could have on U.S. natural
gas prices and supply. The U.S. is the largest user of Canadian gas and many utilities use it for
generation,137 including a number of public power utilities located on the west coast, in the
Northwest, and along the Iron Range of the U.S. If Canada exports some of its natural gas that it
currently supplies to the U.S., that will also put upward pressure on gas prices. U.S. shale gas
can be exported through Canadian LNG terminals to other international markets.138 According
to a Liquefied Natural Gas Ltd. (LNGL) announcement made on August 28, 2014, a site in
135
http://www.rollcall.com/news/us_natural_gas_exports_could_change_market-236112-1.html
http://www.nytimes.com/2014/09/30/business/energy-environment/a-u-turn-for-a-terminal-built-in-texas-toimport-natural-gas.html?_r=0
136
137
138
http://www.eia.gov/countries/cab.cfm?fips=ca
http://www.forbes.com/sites/judeclemente/2014/08/30/49/ and
47
Richmond County, Nova Scotia (Canada) that was originally designed to be an LNG import
facility will now be an export facility, which is expected to be operational by 2022. LNGL plans
to export Canadian natural gas along with U.S. Marcellus shale gas through Nova Scotia to
international markets.139
It is also worth noting that Asia petrochemicals manufacturers want to take advantage of U.S.
natural gas supplies and plan to build tanks and retool plants to store and process liquefied
petroleum gas (LPG) derived from shale gas. LPG would replace costlier naphtha, a byproduct
from oil and gas wells, as a raw material for the manufacturing of chemical products and plastics.
Samsung Total Petrochemical, LG Chem, and Royal Vopak are among a number of companies
in Asia expanding import terminals or retrofitting plants over the next one to two years as they
buy more LPG.140 According to an August 22, 2014, Reuters story, Asian imports of LPG will
reduce U.S. domestic supply of shale gas.141
The EPA Proposal and Technical Supporting Documents do not take into consideration the
potential impacts of LNG exports on domestic prices or the subsequent potential impact on
domestic electricity prices. EPA should examine these issues closely and adjust the relevant
assumptions in building block 2 and state reduction requirements accordingly.
139
http://marcellusdrilling.com/2014/07/new-lng-plant-in-nova-scotia-will-use-marcellus-gas/
http://www.reuters.com/article/2014/08/21/us-asia-lpg-idUSKBN0GL2AD20140821, Aug. 22, 2014
141
Id.
140
48
Figure 3: Lower 48 Proved Nonproducing Reserves
Since 2000, 168.7% Increase
Source: http://www.ferc.gov/industries/gas/indus-act/lng/lng-approved.pdf
b.
Manufacturing Demand
The Proposed Rule also does not factor in announcements by the manufacturing sector of more
than $90 billion in new investments in the U.S. that would use shale gas as a feedstock to
produce a wide variety of chemicals, fertilizers, specialty tubing, specialty steel, plastics and
other commodities.142 Greater use of natural gas as a feedstock and the increasing number of
power plants that will use natural gas will very likely put upward pressures on the price of
natural gas as demand increases.143 In July 2014, the University of Michigan studied this issue
and issued a report entitled, Shale Gas: A Game Changer for U.S. Manufacturing. It asserted
that “lower feedstock and energy costs could help U.S. manufacturers reduce natural gas
expenses by as much as $12 billion annually through 2025, creating one million new
manufacturing jobs” and noted that “In February 2014, the American Chemistry Council (ACC)
142
See Financial Times article, U.S. Sees $90bn Boost from Shale Gas Boom,” Dec. 14, 2012, available at
http://www.ft.com/cms/s/0/4b3f6280-4609-11e2-ae8d-00144feabdc0.html#axzz3ES0K0bQY. Also see copy of the
Wyden-Dow $80 Billion list is inserted into Attachments Section
143
http://www.houstonchronicle.com/business/energy/article/LyondellBasell-adds-to-Gulf-Coast-boom-with-plans5711990.php
49
reported 148 chemical and plastics projects totaling $100 billion in potential new investment in
the U.S.” 144
While the report focused on how “the shale gas boom can be used to the best advantage of U.S.
manufacturing,” it also discussed challenges that had to be considered, including infrastructure
issues, upward pressure on supply and prices from the exporting of LNG, and use of natural gas
by the electric utility industry. It also cited an EIA projection that “use of natural gas for electric
power generation is expected to eclipse industrial use by 2040 when, EIA predicts, industrial
usage will slow in response to rising price.” 145 It is troubling that entities examining the U.S.
manufacturing renaissance have examined the future impact of electric utility use of natural gas
on supply and prices, but EPA has not examined the combined impact of greater use of natural
gas for electric generation and manufacturing on supply and prices. The Agency needs to do so
and adjust its assumptions accordingly.
The U.S. Industrial Energy Consumers of American (IECA) has also studied natural gas issues
related its members’ consumption and the potential impact electric utility fuel switching could
have on natural gas prices. Included below is a graphic that IECA included in a 2014 letter to
FERC that discusses the group’s concerns about LNG.
c.
Transportation Demand
As APPA pointed out in its May 9, 2014, comments on the proposed NSPS for new fossil fuelfired power plants, the manufacturing and electric utility sectors are not the only parties seeking
natural gas or compressed natural gas (CNG) for use. Many transportation companies have
announced plans to move to CNG. Some of these cite the lower price of natural gas, while
others cite meeting other environmental regulations. Regardless of the motivation, many parties
are expected to use U.S. shale gas for transportation purposes.
While natural gas exports via LNG for international transportation demand may be at least five
or ten years away, it is also relevant for consideration under the Proposed Rule. Energy
consultants Wood Mackenzie say global gas demand in the transport sector could grow from
under 5 billion cubic meters (bcm) in 2012 to over 160 bcm by 2030, which would be equivalent
to two-years-worth of current (2014) British gas demand (approximately 3 million barrels of
oil).146 With international CNG running at about double the price of U.S. natural gas in Europe
144
Shale Gas: A Game-Changer for U.S. Manufacturing, prepared by the University of Michigan, July 2014,
available at http://energy.umich.edu/sites/default/files/PDF%20Shale%20Gas%20FINAL%20web%20version.pdf.
145
Id. at 24.
146
http://www.reuters.com/article/2014/02/04/gas-transport-idUSL5N0L51QV20140204
50
($10 Mcf) and almost four times the price of U.S. natural gas in Asian import markets
(approximately $20 Mcf), the export of CNG for transportation markets in not implausible.147
4.
The Proposed Rule Also Fails to Take into Account the Use of
Canadian Natural Gas by U.S. Electric Utilities and How Market
Conditions in Canada Could Impact Supply and Prices in the U.S.
The Proposed Rule does not take into account the fact that some natural gas supplies used by
electric utilities and the manufacturing sector come from Canada, nor the variability in that
supply. Between 2002 and 2008, Canadian gas producers saw a tremendous uptick in drilling in
response to U.S. import pressures (See below). That largely fell off in late 2009 because of the
impact of the economic recession of 2007-2009. Canadian natural gas producers have also seen
some decline in production from wells. For example, wells located in the Horseshoe Canyon
formation in Alberta have seen a decline in production despite the fact that prices increased in
U.S.–Canadian natural gas transactions. 148 As the table below indicates, natural gas production in
Canada has been falling for a decade.
Table 1: Canadian Daily NG Production 2001 to 2012 149
Year
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
20122
Average daily production in bcf/d1
16.56
16.69
16.12
16.14
16.55
16.60
16.17
15.27
14.22
13.96
14.05
13.00
% Change from previous year
NA
+0.1%
-2.8%
0%
-2.5%
0
-2.65%
-5.6%
-6.9%
-1.8%
+.01%
-7.5%
Source: National Resources Canada
1
Billion cubic feet per day
2
Author’s estimate
147
Id. and http://www.aogr.com/web-exclusives/exclusive-story/market-realities-to-determine-ngv-growth-rate
Bill Powers, Cold, Hungry, and in the Dark (New Society, 2013), 168.
149
Natural Resources Canada, “Energy Sector: Energy Sources: Canadian Natural Gas: Monthly Market Update:
Historical Data, “January 2001 – April 2012, date modified 4/13/2011, nrcan.gc.ca/energy/sources/naturalgas/monthly-market-update/1173.
148
51
However, while Canadian natural gas production has decreased, prices have remained low,
which has attracted some new industrial manufacturing to Canada, although not at the scale
expected in the U.S.150 Low gas prices have also made Canadian natural gas attractive to U.S.
users, including electric utilities. Canada has plans to build pipelines to supply the U.S. with
natural gas. However, there are questions about whether the U.S. will approve the new interstate
pipelines needed to move Canadian gas from the border to where it is needed within the U.S.
Canada will look to other markets if the U.S. fails to build such pipelines, such as those in Asia.
The Proposed Rule does not consider how increased demand for Canadian natural gas could put
upward pressure on the prices paid by U.S. electric utilities nor does it consider the impact
reduced gas from Canada could have on the ability of utilities to use more gas for electric
generation to reduce their CO2 emissions. (See Section XXVII related to exports of electricity to
the U.S. from Canada and whether that hydro or nuclear-based generation of electricity would
qualify under the NSPS for states to meet their interim or 2030 goals).
B.
EPA’s Assumption That Natural Gas Prices Will Remain Relatively Flat
Through 2030 Fails to Take into Account the Historic Volatility of
Natural Gas, the Impact on Price from Future Regulations on Upstream
Production, or How Future Increased Demand Will Put Upward Pressure
on Prices.
In the Proposed Rule, EPA states its belief that natural gas prices will remain relatively flat
through 2030. That is a problematic assumption. APPA believes there is considerable
uncertainty on that point, and EPA should reconsider its assumption to recognize that
uncertainty. There is a direct correlation between supply and demand and price. When demand
increases, if there are constraints on supply, prices will increase accordingly. The section above
discusses the uncertainty about the supply of natural gas, even though EPA assumes a plentiful
supply for many years to come. In addition, the agency seems to not recognize that historically,
natural gas prices have been volatile due to a variety of factors including federal policies, shortterm imbalances between supply and demand, and the availability of necessary infrastructure,
among other things. EPA also fails to take into account the potential impact on natural gas prices
of future regulation of the upstream production of gas, or increased demand for gas by domestic
manufacturers and the transportation sector. Any or all of these factors could very likely put
upward pressure on natural gas prices. EPA needs to take all of these issues into account, adjust
its assumptions, and then modify the states’ emission reduction goals accordingly.
150
http://www.conferenceboard.ca/e-library/abstract.aspx?did=5251
52
1.
Historically, Natural Gas Prices Have Been Volatile.
Historically, the percentage of natural gas used for electric generation has been small. In 1949,
about 13 percent of electricity was generated from natural gas. 151 The amount steadily increased
to a then high of approximately 24 percent in 1970 and decreased below 10 percent in 1990 152 as
a result of federal policies that curtailed natural gas deliveries due to high demand and low
supplies as discussed above. Figure 4 charts the percentage of total power generation share in
the U.S. by fuel type from 1949 to 2013.
Figure 4: Net U.S. Power Generation Share by Source, 1949-2012153
In addition to federal policies impacting the use of natural gas for electric generation, price
volatility also affected its use. A July 15, 2010, Navigant Consulting paper for the Task Force on
Natural Gas Market Stability entitled, Price Instability in the U.S. Natural Gas Industry
Historical Perspective and Overview, defines volatility as “sustained, unpredictable price
movements that frustrate the economics of high-load factor use of natural gas in industrial,
chemical, and power-generation applications (on the upside), or frustrate the organized, sustained
151
See National Renewable Energy Laboratory Renewable Energy Project Finance article by Jeffrey Logan, Feb. 26,
2013, citing EIA Feb. 25, 2013, Electric Power Monthly at https://financere.nrel.gov/finance/content/us-powersector-undergoes-dramatic-shift-generation-mix
152
Id.
153
Id. citing Energy Information Administration, Feb. 25, 2013, Electric Power Monthly, available at
http://www.eia.gov/electricity/monthly/
53
growth of deliverability from domestic onshore unconventional resources.”154 The Navigant
paper includes three EIA charts that show electric utility use of natural gas for generation
increasing by 61 percent between 1990-2000, a period defined by relatively stable prices and
ample supply.155 The Navigant paper then contrasts that period with 2000-2010, which it calls
“Crisis, Volatility, Growth, and New Natural Gas Abundance.” Between 2000 and 2010, price
movements on the Henry Hub spot market and wellhead prices tracked fairly closely.156 See
Figure 5, Figure 6, and Figure 7.
Figure 5: Henry Hub Natural Gas Spot Price157
154
See p. 11 at
http://bipartisanpolicy.org/sites/default/files/Introduction%20to%20North%20American%20Natural%20Gas%20Ma
rkets_0.pdf
155
Id. at 21.
156
Id. at 25.
157
http://www.eia.gov/dnav/ng/hist/rngwhhdd.htm
54
Figure 6: U.S. Natural Gas Wellhead Price158
Figure 7: Henry Hub Spot Prices and U.S. Natural Gas Wellhead Price Overlaid
158
See EIA website at http://www.eia.gov/dnav/ng/hist/n9190us3m.htm
55
The Navigant paper notes there were three major drivers of large price movements during the
decade—the California energy crisis, Hurricane Katrina, and price runs with oil. Commonalities
shared by the three drivers include growth of natural gas use for electric generation, insufficient
natural gas storage, and the lack of long-term contracting for natural gas supply. The report
concludes that “the vitality of and responsiveness of the supply-demand balance is the most
important factor determining whether price volatility in either direction will occur.” While
current natural gas supply is ample and prices are relatively low, APPA is concerned this may
not remain the case prospectively. The agency needs to ensure that its analysis of long-term
natural gas prices and the corresponding impact on future electricity prices takes into account
historic natural gas volatility and the potential for future volatility.
And while the shale gas boom has resulted in a sharp decrease in prices in recent years, gas
prices are still volatile, as evidenced in early 2014 during the two polar vortexes. This volatility
has impacted not only the prices utilities pay for natural gas, but also the wholesale price of
electricity. EIA reported on January 21, 2014 that “day-ahead, on-peak power prices at the
Massachusetts Hub went slightly above $200 per megawatt hour (MWh) during a brief cold spell
in mid-December 2013 and up to $237.75/MWh during the early January freeze.”159 EIA
attributed these prices to “corresponding movements in natural gas prices as the demand for
natural gas for both power and heating led to full use of natural gas pipelines in [New England]
and a scarcity of supply.160 Prices at the Algonquin Citygate trading point in Massachusetts…
were up to $38.09/MMBtu in early January.”161 They are usually around $3-6 during
unconstrained periods.162 In New York City, “spot natural gas prices reached as high as
$47.80/MMBtu, higher than New England—likely because New England was able to meet part
of its natural gas demand with imported supplies of liquefied natural gas[] and Canadian offshore
natural gas production. Power prices hit $233.59/MWh on January 8.”163 It is very likely further
price spikes will occur as the demand for natural gas increases for electric generation.
On November 6, 2014, the Wall Street Journal pointed out that a colder December weather
pattern appears to already have influenced natural gas prices earlier than normal.164 While no one
can predict what the winter 2015 natural gas prices will be, the price of natural gas has increased
159
EIA Today, January 21, 2014, available at
http://www.eia.gov/todayinenergy/detail.cfm?id=14671#tabs_SpotPriceSlider- 1
160
Id.
161
Id.
162
Id.
163
Id.
164
http://online.wsj.com/articles/natural-gas-rallies-on-cold-weather-forecast-1415290331
56
significantly since the Proposed Rule was announced in June 2014 from about $3.50 to $4.04
Mcf. Some variability in price is expected in anticipation of colder weather and higher
residential use.
2.
The Proposed Rule Does Not Take into Account the Potential
Impact on Price of Future Upstream Regulations.
Another issue that could potentially impact the price of natural gas that EPA failed to account for
in the Proposed Rule is future local, state, and federal regulations on the upstream production of
natural gas. Such regulations increase the costs of doing business and would be passed on to
natural gas users through higher prices. Potential regulations include those to capture methane
from upstream oil and gas production, controls for capturing other volatile organic compounds
(VOC) from upstream natural gas production,165 NOx regulations for new nonattainment areas 166
where natural gas production did not exist before, and future treatment requirements for fracking
or production water.167 In August 2014, EPA announced efforts to regulate methane emissions
from natural gas and oil drilling, production, and gas processing that could result in increased
natural gas prices over time. Further, DOE is looking into the possibility of regulating methane
from natural gas pipelines.168 It is unclear whether methane regulations in the upstream sectors
producing and transmitting natural gas will be issued under the CAA (as VOCs or GHGs) or
some other authority. Regardless of how such emissions are regulated, it is very likely the costs
of complying with them will be borne by the consumers of natural gas, including electric
utilities.
In addition, the U.S. Department of Transportation may develop new regulations to improve
safety for pipelines, specialty tank cars and other rail equipment used to transport natural gas.169
EPA needs to look at all of these issues and factor them into the final rule’s presumptions about
the future of natural gas and the effects on electricity rates.
165
http://www.epa.gov/ttn/atw/oilgas/oilgaspg.html
http://www.epa.gov/ttn/naaqs/standards/ozone/s_o3_index.html
167
http://www2.epa.gov/regulatory-information-sector/oil-and-gas-extraction-sector-naics-211
168
http://energy.gov/articles/factsheet-initiative-help-modernize-natural-gas-transmission-and-distribution
169
http://www.phmsa.dot.gov/pipeline/regs
166
57
3.
The Proposed Rule Does Not Take into Account the Potential
Impact on Price of Increased Non-Electric Utility Demand for
Natural Gas.
As described earlier in this section, EPA has failed to consider the impacts of LNG exports and
increased non-electric utility demand on natural gas supply. The Agency also failed to look at
the effects of these phenomena on natural gas prices. Also, increased demand for natural gas by
manufacturers for feedstock and the transportation sector will likely lead to increased natural gas
prices. With all of the expected growth in demand for natural gas, EPA should re-examine its
assumption that prices will remain relatively flat and not impact electricity prices.
C.
EPA’s Assertions Regarding the Adequacy of Existing Natural Gas
Infrastructure and the Ability to Expand It to Facilitate Fuel Switching
Fails to Take into Account Impediments to Infrastructure Development
and the Lack of Sufficient Storage.
According to the EIA, currently there are approximately 300,000 miles of interstate and
intrastate natural gas pipeline capacity in the U.S. 170 EPA asserts that pipeline capacity can be
added to enable the large-scale fuel switching the Proposed Rule envisions is needed to reduce
CO2 emissions. It states that “over a longer time period, much more significant pipeline
expansion is possible.”171 Unfortunately, EPA’s analysis fails to examine whether this
significant expansion can happen by 2020 or even 2030 given the impediments the pipeline
industry faces in constructing and expanding pipelines, barriers posed by certain restructured
wholesale electricity markets, and differences in the business models for the electricity and gas
pipeline industries. According to comments filed with the U.S. Fish and Wildlife Service in
October 2014, the Interstate Natural Gas Association of America (INGAA) anticipates that the
pipeline industry would need to build approximately 2,000 miles of pipeline each year or a total
of 300,000 miles of pipelines (interstate and intrastate) between 2014-2035 to meet anticipated
natural gas demand.172 .
The Proposed Rule also does not examine whether sufficient natural gas storage exists to support
large-scale fuel switching. Many states do not have adequate geologic formations to store gas
that is needed by electric utilities for generation as baseload power, to back up intermittent
renewables, or provide peaking power. The Agency needs to look at these issues and adjust its
170
http://www.eia.gov/pub/oil_gas/natural_gas/analysis_publications/ngpipeline/intrastate.html
79 Fed. Reg. at 34,864
172
See INGAA comments filed regarding Endangered Species Act listings determinations, but not specifically for
EPA’s ESPS/NSPS for power sector http://www.ingaa.org/File.aspx?id=22680
171
58
assumptions about the timetables by which states and the electric utility industry can meet the
requirements of building block 2.
1.
The Proposed Rule Presumes That Significant Pipeline Expansion
Is Possible, but Does Not Take into Account Impediments to
Pipeline Construction and Expansion That May Impact the LargeScale Fuel-Switching to Natural Gas to Reduce CO2 Emissions.
As APPA pointed out in its 2012 and 2014 comments to EPA on the proposed NSPS for new
fossil fuel-fired power plants, an enormous amount of natural gas pipeline and storage
infrastructure will be needed to enable fuel switching from coal to natural gas in many states.
While some of the information in APPA’s 2010 Natural Gas Study is out of date and does not
reflect new pipeline projects pending approval before FERC in Georgia, Florida, Pennsylvania,
and New England,173it is still the case that very few pipeline investments have been made by
major pipeline companies for downstream customers, such as power plants. A March 2014 study
by ICF International for the INGAA Foundation 174 shows that most of the oil and gas pipeline
investments made in the last four years were made to move the upstream production of natural
gas to gas processing centers or oil to refineries. Very little pipeline capacity has been built to
address electric utilities’ current or future demand for natural gas.
EPA’s Proposed Rule assumes that sufficient natural gas infrastructure will be in place by 20 20
for states to reduce CO2 emissions through the dispatch (or redispatch) of NGCC units running at
up to a 70 percent capacity factor. 175 The U.S will need thousands of miles of pipelines to meet
the demand for natural gas by the electric utility sector once the Proposed Rule is finalized.176 It
will take time for these projects to be sited, permitted, financed, and constructed. Unfortunately,
the Proposed Rule does not take into account how much time will actually be needed to build
this infrastructure, nor does it look at potential impediments to the siting and permitting of
natural gas pipelines. 177
173
http://www.eenews.net/energywire/2014/11/13/stories/1060008811
North America Midstream Infrastructure through 2035: Capitalizing on Our Energy Abundance,
http://www.ingaa.org/Foundation/Foundation-Reports/2035Report.aspx and INGAA’s related Building Interstate
Natural Gas Transmission Pipelines: A Primer http://www.ingaa.org/File.aspx?id=19618
175
APPA’s comments will address the weaknesses in the EPA assumptions about the 70 percent capacity factor in
Section XIV(B) on page 92.
176
http://naruc.org/Grants/Documents/Final-ICF-Project-Report071213.pdf
177
Please see the report by INGAA on impediments to pipeline construction available at
http://www.ingaa.org/Topics/Pipelines101.aspx
174
59
For example, FERC must review and approve all projects to construct new interstate pipelines or
expand existing ones before they can be built. During the FERC approval process, an
environmental impact statement (EIS) must be prepared, as required by the National
Environmental Policy Act (NEPA). FERC must also verify that applicants have secured permits
from local, state, and federal agencies before construction can begin.178 Conflicts can arise
during the review process between the various permitting agencies involved in the approval
process. These conflicts slowdown the approval process, and in some cases, prevent the project
from going forward. Many times the delays that occur from these conflicts on projects that
ultimately do proceed result in increased construction costs that are eventually borne by natural
gas end users. In addition, contentious permitting issues can lead to avoidance of certain areas of
the country for future permitting, “constraining the ability of supply to reach markets.”179
Market-related issues can also present impediments to new gas pipeline development. RTOoperated capacity markets in the eastern part of the U.S. create obstacles to the construction of
new pipelines needed to supply sufficient gas to electric utilities (as well as obstacles to the
construction of renewable generation). This has been particularly problematic in New England,
which generates over half of its electricity from natural gas and where the lack of sufficient
pipeline capacity is widely expected to have significant, adverse reliability and cost impacts.
(See Section VII for more discussion of this capacity market issue.) Concerns about overreliance on natural gas and the need for sufficient pipeline capacity to meet demand by the New
England states led to an attempted effort to develop a new business model for obtaining needed
natural gas pipeline and transmission infrastructure.
In December 2013, the Governors of the New England states issued a joint statement expressing
their commitment to “continue to work together, in coordination with ISO-NE, and through the
New England States Committee on Electricity (NESCOE), to advance a regional energy
infrastructure initiative that diversifies [its] energy supply portfolio while ensuring that the
benefits and costs of transmission and pipeline investments are shared appropriately among the
New England States.”180 To further the Governors’ stated goals, NESCOE launched the Energy
Infrastructure Initiative (“Initiative”) with two primary purposes; to increase pipeline capacity in
178
See INGAA Foundation report, Avoiding and Resolving Intergovernmental Conflicts with Interstate Natural Gas
Facility Siting, Construction, and Maintenance, available at http://www.ingaa.org/Foundation/FoundationReports/Studies/FoundationReports/52.aspx
179
Id.
180
“New England Governors’ Commitment To Regional Cooperation On Energy Infrastructure Issues,” Dec. 5,
2013, available at: http://www.nescoe.com/uploads/New_England_Governors_Statement-Energy_12-5-13_final.pdf
60
the region and to expand electric transmission to facilitate the delivery of electricity from “low to
no-carbon resources.”181
NESCOE recognized that the natural gas and electricity markets are not producing needed
pipeline construction. Natural gas pipelines require long-term contracts to provide a steady
stream of revenue, and in contrast the natural-gas-fired merchant generators operate in a shortterm restructured market with volatile revenue streams. As a result, these generators are not
willing to commit to a long-term contract. Pipeline projects in New England have had no
electric power generators purchasing firm gas transportation. The absence of generator
subscription caused Spectra Energy’s Algonquin Incremental Market (AIM) pipeline project to
be downsized from the planned 500 mmcf/day to 342 mmcf/day. 182
However, the NESCOE Initiative is currently on hold as a result of the Massachusetts
legislature’s adjournment without passage of a bill to provide for state long-term contracting for
electricity from resources with zero or low carbon dioxide emissions. Although this
development primarily affected the electric transmission component of the Initiative, the natural
gas infrastructure efforts were also stalled. 183
Natural gas is not the only fuel for electricity generation that faces obstacles with respect to its
transportation from its source to the point of use. Continued monopoly abuses by the railroads,
for example, pose serious issues and increased costs for those shipping coal for electricity
generation. However, given the EPA’s heavy reliance on increased use of natural gas to
implement the Proposal, EPA needs to revise its assumptions, related calculations, and timelines
to reflect the reality of the significant amount of natural gas pipeline capacity that will be needed.
2.
The Proposed Rule Fails to Examine Whether There Is Sufficient
Natural Gas Storage Needed to Support Large-Scale Fuel
Switching to Natural Gas for Electric Generation.
EPA’s Proposed Rule does not appear to address any issues related to natural gas transport to the
power sector. This is disappointing since APPA has met with EPA, filed comments on these
issues, and given presentations on natural gas storage, transport, and the concept of “rapid turn”
181
“Update on the New England Governors' Proposal to Invest in Strategic Infrastructure and Address Price
Disparities,” NESCOE presentation to the US DOE Electricity Advisory Committee, Sept. 25, 2014, at 12, available
at:
http://www.nescoe.com/uploads/DOE-EAC-Gas-ElectricPanel_Sep2014.pdf,
61
or “rapid use” several times since 2009.184 UARG, the National Rural Electric Cooperative
Association (NRECA), the Industrial Energy Users of America, and others also filed comments
on these issues between 2011and 2014.
While natural gas pipelines can typically be permitted, financed, and constructed in four to eight
years, construction of natural gas storage facilities is far more difficult because of the limited
geology for natural gas storage. Most parts of the U.S. lack the appropriate geology to store
natural gas in the subsurface. Natural gas transmission pipelines cannot store significant
quantities of gas. At best, they can store a marginal amount through “linepacking.”185
The question of sufficient storage is often not considered because many existing natural gas-fired
power plants are located in regions of the U.S. with suitable geology to store sufficient amounts
of natural gas for the power and industrial-manufacturing sectors without any glitches in supply
and delivery, or they are part of a larger, diversified fleet where they provide back-up power for
renewables or peak power. In a few states, semi-depleted natural gas fields offer excellent
locations for storage of natural gas, but they are not suitable for natural gas re-injection on a
“multi-turn” or “rapid-use” basis needed by the power sector.186 Factories do not need the ability
to send back natural gas as they very rarely re-inject natural gas into storage fields on a much
smaller volume. Typically, they use the gas they purchase in anticipation of their daily
manufacturing need.187 Utilities, however, encounter weather events that alter how much gas
they may need and have to resell it since they do not have sufficient localized storage.
Underground natural gas storage will be necessary to allow the electric utility industry to
generate more power from natural gas-fired power plants. “Storage is very useful in providing
flexibility to support gas burns by electric generators. Natural gas storage lets a power plant
operator:
 ramp up or ramp down operations quickly (especially for intermittent
renewables);
184
EPA-HQ-OAR-2011-0660, EPA-HQ-OAR-2013-0495,
http://www.publicpower.org/files/PDFs/ImplicationsOfGreaterRelianceOnNGforElectricityGeneration.pdf, and
EPA-HQ-OAR-2011-0090
185
“Linepack is an extra amount of gas in a pipeline or distribution line relative to maximum anticipated load….It
can be thought of as a system’s first form of temporary gas storage.” See APPA Natural Gas Study at p. 59.
186
“Single-turn storage is used to inject and withdraw generally once per year; multi-turns storage allows several
cycles of injections and withdrawals over the course of the year.” See APPA Natural Gas Study at p. 61, available
at: http://www.publicpower.org/files/PDFs/ImplicationsOfGreaterRelianceOnNGforElectricityGeneration.pdf and
http://www.publicpower.org/PDFs/AttachB_Aspen_GasStorage2012.pdf
187
In the rare instances, manufacturers need to sell off already purchased natural gas due to an unexpected problem
in the manufacturing process or maintenance issue.
62



manage its imbalances;
potentially hold less firm pipeline capacity; and
maintain reliability.”188
Table 2Table 2 and Figure 8 (pages 64 and 65) help illustrate the importance of natural gas
storage and how important storage will be to the attainment of the Proposed Rule’s interim and
final goals.
EIA reported that as of 2013, roughly 4.6 Tcf of underground storage working gas capacity,
representing roughly 410 underground gas storage fields, were in operation.189 This is an
increase of approximately 0.6 Tcf, or 15 percent relative to the 4 Tcf cited in APPA’s Natural
Gas Study in 2010.190 But these underground storage fields are not equal. For example, most are
not configured to provide rapid withdrawals for power plants.191 In addition, many are not
located close enough to power plants to support the rapid change in linepack needed when a
power plant fires or otherwise accommodates multiple turns often needed to support power plant
operations. Figure 8 on page 65 shows the location of natural gas storage facilities and their
proximity to existing coal-fired power plants. Of the fields currently in operation, 90 percent are
“reservoir or aquifer storage where gas is generally injected during the summer months and
withdrawn during winter to serve seasonal demand (with certain exceptions).”192 The other 10
percent are high-deliverability salt cavern facilities.193 “Most, but not all, of the multi-turn, salt
cavern storage is located along the Gulf Coast.”194 Additionally, most of the new storage added
since the 2010 APPA Gas Study was released is “producing area storage” located in the states of
Mississippi, Louisiana, and Alabama.
Table 2 summarizes the key characteristics of U.S. natural gas storage.
188
See APPA Natural Gas Study at p. 57.
Id. at p. 57. “Terminology: cushion (sometimes called pad or base gas) is gas intended to stay in the formation to
maintain field pressures sufficient to achieve desired withdrawal levels; working gas is the amount of gas that can be
injected into or withdrawn from the field and still be able to fill it in a given period of time; withdrawal capability is
the amount of gas that can be withdrawn in a given period, usually a day; the withdrawal capability is higher when
the field is filled with more gas and achieves maximum field pressures; injection capability is the amount of gas that
can be injected in a given period, usually a day; the injection capability falls at the field is filled and operating
pressures rise.”
190
This excludes the roughly 100 small LNG needle peaking operations that exist for meeting peak day demands in
specific locations where there is no underground gas storage and pipeline capacity into the region is lower than peak
day demand.
191
Semi-depleted gas fields are also being used for storage of water, condensate, or other petroleum products, which
would make them unavailable for electric utility use. The EIA data reported above excludes underground storage of
those products.
192
Id.
193
Id.
194
Id.
189
63
Table 2: Storage Summary
Type
Characteristic
Owner
User
Purpose
Price
Sites
Working Gas (Bcf)
Maximum Daily
Withdrawal (MMcf/d)
Reservoir/Aquifer195
Single Turn
LDC or Pipeline
LDC
Seasonal Demand196
Cost of Service
379
4,293
82,271
Salt Formation
Multi-Turn
Independent
Gas Marketers
Arbitrage or Daily Peak
Option Value
40
456
32,158
Source: Aspen Compilation of EIA Data, 2014
195
Also sometimes called “traditional” storage.
Some reservoirs can be configured for multi-turn high-deliverability storage. The Lodi and Wild Goose facilities
in northern California are examples of reservoirs that provide multi-turn storage and that serve primarily the price
arbitrage market. We have shown them as reservoir storage nonetheless because that is how they are characterized
in the EIA data.
196
64
Figure 8: Geographic Distribution of Underground Gas Storage Facilities and Coal-Fired Power Plants197
197
A prior version of this appeared in the APPA Natural Gas Study at p. 58. The red dots represent coal-fired power plants. The green circles represent
traditional, reservoir-based underground gas storage, while the gold circles represent salt formation storage facilities.
65
As the 2010 APPA Natural Gas Study notes:
Most existing natural gas storage was built before the trend towards more
use of natural gas to generate electricity and tends to be located where a
pipeline or a local distributor chose to build based on the accident of
geology and the economics of revenue recovery.198 Pipelines that are not
connected to storage tend to impose stricter balancing rules. Stricter
balancing rules make it harder or more costly for electric generators to
operate.199
Fundamentally, natural gas storage balances demand against production.
We use it particularly to allow producers to operate their wells on a
relatively levelized basis: when demand is lower than production, the
excess gas goes into storage. We then withdraw it when demand exceeds
production. Figure 9 provides an illustration of this concept. Storage also
allows local demand to be met with gas stored near the load center, thus
reducing the need to size trunkline transmission capacity into the load
center large enough to meet peak day demand. Some types of storage also
lend themselves to either medium-term price arbitrage or to intraday
peaking service.200 Another use of storage is to remedy imbalances
between deliveries into a pipeline against the quantity of gas burned by
end-users. As explained elsewhere in this paper, those two quantities
often differ. They often differ by fairly large amounts for electric power
plants. When they differ by more than the pipeline operator can address
198
New storage being added today tends to be added by local distributors who can make incremental investments to
existing facilities or by independent merchant storage providers who charge market-based rates for storage service
assuming several cycles, or turns, of gas are made through the inventory space.
199
See p. 59.
200
“It used to be common for summer natural gas prices to be lower than winter prices; thus, LDCs would purchase
gas under levelized take contracts and store the excess gas until winter. Even recognizing the carrying cost on the
inventory, consumers routinely benefited from these transactions. With the kind of price volatility that exists today,
however, that can result in winter prices being lower than summer prices, these seasonal transactions cannot be sure
to provide price benefits. Instead, we see marketers using short-term storage to capture the intrinsic and extrinsic, or
real option value of storage. The intrinsic value is based on the cost of spot gas today versus today’s forward value.
One can buy gas, inject into storage, and assure a profit spread by locking in today the sale upon withdrawal. The
extrinsic value is based on changes one might realize due to the future movement of prices until the gas is
withdrawn, e.g., actual spot prices being higher or lower than the purchase price or the locked-in forward price on
the day of purchase.” 2010 APPA Natural Gas Study at p. 59.
66
with linepack or offsets by other shippers, the excess or shortage of gas
must be addressed with gas in or out of storage .201
Linepack cannot be used to store enough natural gas for major power plants. It can only be used
for marginal hedging against unexpected weather events and often will provide up no more than
12 hours of supply, assuming that only one or two utilities are doing this on an entire pipeline.
Figure 9: How Storage Balances Seasonal Demand with Monthly Production202
201
Id.
“Note that the injections do not exactly equal the withdrawals; this difference ends up as left-over gas inventory
at the start of the next annual cycle. This outcome is not that uncommon.” Id. at 60.
202
67
“Electricity generators can benefit from multi-turn or rapid storage by subscribing, for
example, to enough storage to meet their total gas requirements for the five days that might be
the maximum likely interruption in the case of a gas curtailment. The generator could use the
space to inject or withdraw gas to cover its daily imbalances or for price arbitrage in the
meantime. Multi-turn storage costs more, but most of the costs are fixed, so the more times
the generator cycles gas in and out, the lower the amortized cost per MMBtu” - Aspen
Figure 8 from the APPA Natural Gas Study shows that storage facilities are not evenly
distributed, either geographically or relative to demand around the country. 203 Most stored gas is
used relatively close to the region in which it is located, and geology does not always provide
sites where storage is needed. Certain pipelines and regional markets do not currently have
access to any underground gas storage. Figure 8 plainly shows that Nevada, Idaho, and Arizona
have none. The Central Plains states and Missouri have virtually none. The entire East Coast
has none other than far upstream in western New York, western Pennsylvania, and West
Virginia.204 To the extent that there is a small amount of storage in the Southeast, Figure 8
shows that none of it is in the eastern coastal states.205
It is also worth noting that NERC reliability standards require electric generators to be able to
generate within 10 to 15 minutes of receiving notice to do so (depending upon NERC region).
This requirement means that the natural gas to run those generators needs to be relatively close
by.
In summary, the Proposed Rule fails to look at key issues related to natural gas storage. EPA
needs to examine these issues and revise its assumptions regarding the timetable by which
electric utilities could convert from coal-fired generation to natural gas-fired generation to reduce
their CO2 emissions. It is likely that the success of many state compliance plans will depend
heavily on the ability to build new natural gas infrastructure in a specific timeframe. The
inability to do so, based on factors beyond their control, is a prime example of why states should
be allowed the opportunity at any time to request that EPA consider changes to their CO2
emission goals and/or an extension of the final compliance date as discussed in Section XVIII.
203
Aspen Environmental Update, Nov. 2014
The LNG terminals located at Elba Island, Georgia; Cove Point, Maryland; and Everett, Massachusetts might
provide some important natural gas storage peaking benefits, but they cannot provide all of the storage needs of
utilities in those areas.
205
“The states without storage now are largely without the depleted reservoirs or salt formations that can
economically be turned into storage.” Id. at 61.
204
68
VIII.
Gas-Electric Industry Coordination Issues Pose Barriers to the
Rapid Increase in the Use of Natural Gas for Electric Generation.
APPA is concerned that a prolonged dash to gas will lead to over-reliance on one fuel type and
have adverse consequences for the balance and diversity of the power sector and the economy.
There are numerous barriers to fuel diversity within the electric generation fleet, including the
Proposed Rule. The increased use of natural gas to generate electricity will put stress on the
natural gas system, which is presently designed to meet peak winter heating needs by requiring
increasingly larger supply and flow rate to power plants. This increased reliance on natural gas
has already contributed to rapid price spikes in the cost of gas, which translates into much higher
wholesale electricity prices. Given the Proposed Rule’s key assumption that large-scale fuel
switching can occur to help achieve state CO 2 reduction goals, it is concerning that EPA has
failed to look at the impediments to greater use of natural gas for electric generation or consider
the implications of over-reliance on natural gas for generation.
There are additional concerns surrounding the synchronization of electricity and natural gas
markets as supplies of power and natural gas are secured on a different time basis. This
disconnect may prevent facilities committed to provide electric power from securing the gas
supplies they need to operate, or require them to pay higher prices for longer time periods. This
issue is further complicated by the interdependent nature of the natural gas and electric
generation industries. As more power generation comes from gas, the impact of simultaneous
peak electricity demand and peak consumer heating demands converge, creating a scenario
where gas deliverability capability can become a bottleneck. This is particularly true in the
winter when shorter days and colder temperatures increase demands for heating and lighting.
While adequate supplies of natural gas exist, delivering at the rate needed during peaks could be
constrained. Additional coordination between these two industries is needed, in addition to fuel
diversity, which will reduce the interdependence risk.
Since a series of events in 2011 and 2012 cast a light on mismatches in the operating cycles
between the natural gas and electricity generation markets and pipeline capacity shortages,
FERC has led an effort to encourage stakeholders and numerous collaborative bodies to identify
regional issues and propose possible solutions. A number of RTOs have established task forces
on electric and gas coordination, looking at information sharing, operations coordination, and
process improvements.
In November 2013, FERC issued an order allowing the voluntary sharing of non-public
operational information between interstate gas pipelines and electric transmission providers to
promote grid reliability and operational planning. While FERC was able to take steps quickly to
address communications between the two industries, the issues of insufficient gas infrastructure
69
and the nonalignment of gas and electricity markets have proven to be thornier for the
Commission to deal with as the U.S. gas boom whets the appetite for more gas-fired power.
Extreme spikes in spot gas and electricity prices, accompanied by widespread unplanned outages
of gas-fired generation facilities, experienced in the eastern United States, and especially in New
England, during the polar vortex of January 2014 brought gas-electric interdependency to the
forefront once again. Though likely planned before that event, FERC’s proposed changes to
pipeline nominations and scheduling procedures soon followed.
In March 2014, FERC issued a Notice of Proposed Rulemaking (NOPR),206 which has spawned
an extraordinary effort by the North American Energy Standards Board (NAESB) and natural
gas and electric industry participants to develop a consensus on standards for coordinating the
scheduling processes of the natural gas and electricity markets. FERC took the unprecedented
step of delaying the filing of comments in its NOPR docket proposing changes to the gas
pipeline nomination and scheduling timetable to provide NAESB the opportunity to develop a
counterproposal broadly supported by the industry. The NOPR required NAESB to submit any
counterproposal developed by the industry by September 29, 2014.
The NOPR points out that, under the current scheduling timelines in the gas and electric
industries, natural gas-fired generators in RTO markets must purchase and nominate gas before
they know whether they will be called upon in the Day-Ahead energy markets.207 In areas
outside of RTO markets, the NOPR indicates that gas-fired generators may benefit from
additional nomination opportunities in order to reflect weather conditions or other operating
needs.208 In response, FERC proposed to move the start of the Gas Day from 9:00 a.m. Central
Clock Time (CCT) to 4:00 a.m. CCT, to institute four intra-day nomination opportunities for
pipeline shippers, and to maintain the no-bump rule for the final intra-day cycle – 8:00 a.m. CCT
(bump), 10:30 a.m. CCT (bump), 4:00 p.m. CCT (bump), and 7:00 p.m. CCT (no-bump).
In response to the NOPR, NAESB convened several meetings of its Gas Electric Harmonization
Task Force (GEH Task Force) after FERC issued the NOPR. The NAESB Board imposed
super-majority voting requirements on the GEH Task Force. No proposal would pass unless it
obtained at least 2/3 of the votes from the Wholesale Electric Quadrant (WEQ) and the
Wholesale Gas Quadrant (WGQ). In addition, proposals would need at least 4 percent of the
votes from each segment of the WEQ and WGQ to pass. The GEH Task Force, which consisted
206
Coordination of the Scheduling Processes of Interstate Natural Gas Pipelines and Public Utilities, Docket No.
RM14-2-000 (March 20, 2014) (NOPR).
207
NOPR at P 29.
208
Id. at P 30.
70
of well over 400 participants, solicited proposals from any interested entity. Thirteen companies
or groups submitted proposals for consideration by the GEH Task Force. As the various
proposals were discussed, participants were able to synthesize the most controversial issues for
consideration:



What time should the Gas Day start?
How many intra-day nomination cycles should be adopted?
Should the final intra-day cycle be subject to the no-bump rule?
After substantial discussion and a number of straw polls, the GEH Task Force took binding votes
at a two-day meeting held June 2-3, 2014. Neither the 4:00 a.m. CCT nor the 9:00 a.m. CCT
obtained super-majority support. The 9:00 a.m. CCT Gas Day, however, received slightly more
support than the 4:00 a.m. proposal. As a result, NAESB will not propose an alternative to the
NOPR’s proposal to commence the Gas Day at 4:00 a.m. CCT.
In contrast to the deep disagreements among GEH Task Force participants regarding the start of
the Gas Day, the Task Force seemed to reach consensus on the number and timing of intra-day
nomination cycles. With little disagreement, the GEH Task Force agreed that timely
nominations should be due at 1:00 p.m. CCT prior to the Gas Day; evening nominations should
be due at 6:00 p.m.; the first intra-day nominations should be due at 10:00 a.m. on the Gas Day;
the second intra-day nominations should be due at 2:30 p.m.; and the third and final intra-day
nominations should be due at 7:00 p.m. The first and second intra-day nomination cycles would
be bumpable under the GEH Task Force proposal, while the no-bump rule would apply to the
third intra-day nomination cycle.
When the GEH Task Force considered this timeline without a proposed start to the Gas Day, the
group was unable to muster super-majority support, as many participants expressed concern
about supporting an incomplete proposal. Nonetheless, the NAESB Board voted to direct the
WGQ to draft standards adopting this timeline without a proposed Gas Day start. If the WGQ is
successful and its standards are adopted by the NAESB Board and its membership, NAESB will
submit those standards—without a proposed start to the Gas Day—to FERC as an alternative to
the nomination and scheduling timeline proposed in the NOPR.
With respect to the applicability of the no-bump rule, a group of natural gas customers in the
desert Southwest (Desert Southwest Group) claimed that, unlike the concerns of gas-fired
generators in the Northeast, gas-fired generators in the Southwest face issues in the late
afternoon if solar facilities are unable to deliver the expected levels of energy. The Desert
Southwest Group holds firm pipeline capacity, but notes that its members are sometimes
precluded from accessing that capacity late in the Gas Day due to restricted intra-day nomination
opportunities and the no-bump rule.
71
The remaining participants agreed that additional intra-day nomination cycles should be adopted,
but overwhelmingly rejected any attempts to make the final cycle subject to bumping. Many
noted that the Desert Southwest Group’s issues seemed regional in nature and could possibly be
resolved by discussions with the interstate pipelines serving the region.
Comments on the NOPR were due on November 28, 2014. If NAESB elects to support the
pipeline scheduling and nomination timeline, minus a proposed start to the Gas Day, interested
parties may simultaneously submit comments on that proposal. The GEH Task Force process
demonstrates that the energy industry’s views on the NOPR vary widely by region and by
segment, which may make it difficult for FERC to issue a final rule that is not subject to timeconsuming rehearing and appellate processes. Moreover, any additional steps FERC may take to
bridge the gas-electric divide will come with costs for consumers. So FERC needs to work with
the gas and electric industries to ensure those costs are manageable.
As noted elsewhere in these comments, the EPA Proposal relies heavily on the ability to rapidly
increase the amount of natural gas used to generate electricity. However, the gas/electric
coordination issues raised in this section pose significant barriers to such an increase. In
addition, there remains considerable uncertainty as to whether, and if so, how, FERC may
continue to address these issues. Thus, EPA should re-examine and revise the assumptions it has
made with respect to related provisions in the Proposed Rule, particularly in its construction of
building block 2.
IX.
The Proposed Rule Fails to Take into Consideration Other Federal
Environmental Regulations That Will Impact the Ability of the
States to Require Large-Scale Fuel Switching from Coal to Natural
Gas to Achieve Their CO2 Reduction Goals.
The assumptions EPA makes in the Proposed Rule about the ability of the utility industry to fuel
switch from coal to natural gas fail to take into consideration other environmental regulations
that impact the construction of natural gas infrastructure or the operation of NGCC units. While
EPA does not administer all of the relevant environmental regulations it should have considered
in this rulemaking, it does administer some of them, such as NAAQS. These regulations are
likely to increase the difficulty states face in achieving their goals. For additional information on
permitting requirements please see Section IX(A).
72
A.
EPA Did Not Consider That New NGCC Generation Must Meet Existing
and Revised NAAQS.
NOx emissions from all fossil fuel generation are regulated as precursors to ozone and fine
particulate patter (PM2.5) under the NAAQS program. The level of stringency of NOx emission
limits may depend on whether the generation is located in an attainment area or a nonattainment
area. Existing NGCC units are usually licensed to operate at full load continuously. However,
given what is essentially a requirement under the Proposal to re-order unit dispatch, a significant
change in ozone standards could result in a request to conduct additional Air Dispersion
Modeling to address the higher NOx resulting from related operational changes such as ramping.
It is difficult to anticipate the impact of a revised ozone standard on existing coal or natural-gas
fired units without running air dispersion modeling for each location. Moreover, results will
vary depending on topography and proximity to other sources such as factories and roads.
However, EPA did not appear to take into account possible increases in NOx emissions that
might occur.
The Proposed Rule’s compliance timeline for meeting the interim and final goals does not
provide the utility sector or states with sufficient time to go through all of the steps required to
conduct the new Air Dispersion Modeling runs needed to verify attainment of existing NAAQS.
EPA apparently presumes that all or most existing NGCC units that would need to be
redispatched to a higher capacity factor to meet building block 2 would immediately pass the Air
Dispersion Modeling tests. Likewise, the Agency didn’t take into account the impact of
complying with revised NAAQS under a redispatch scenario. EPA did not even consider the
time needed to acquire revised Prevention of Significant Deterioration (PSD) permits for existing
units when it set the interim and final compliance dates in the Proposed Rule. Had EPA done so,
it would have realized that more time will likely be needed to achieve the reduction goals the
agency set for each state.
In addition, EPA finalized a new NAAQS for nitrogen dioxide (NO 2) that set a 1-hour standard
for ambient NO2 at 100 parts per billion.209 NGCCs can comply with this standard, but because
background ambient NO2 levels will be determined by monitors placed near highways, there is a
concern that they will be found in non-compliance because the monitors cannot separate the
NGCC unit emissions from the emissions attributable to motor vehicles. This could result in
urban areas falling into nonattainment, which would trigger more stringent permitting
209
See Docket ID: EPA-HQ-OAR-2006-922. INGAA filed written comments that the one-hour standard “could
require emissions levels beyond the capability of current control technologies.”
73
requirements for NGCCs. EPA’s list of counties that would be in nonattainment 210 is
conservative as the NAAQS rule requires new dispersion monitors be placed in more counties
where monitors are currently not located. The monitors could be placed near highways and
existing factories where NOx emissions would be higher. If an existing NGCC unit is sited in a
newly designated NOx nonattainment area, emission offsets and/or use of some sort of control
equipment not currently envisioned, or a higher stack, may be required. Those costs or possible
ozone precursor limitations of replacing existing coal plants with new NGCC units are not
considered in the Proposal. Also, EPA’s state plan requirements do not give states adequate time
to conduct PSD determinations using Dispersion Modeling since the state plans will presume
that any existing coal plant with co-firing, or any gas plant needing any PSD modeling approval,
would be completed before 2020. See Sections IX(A), XV, and XVI addressing state plans.
B.
The Proposed Rule Does Not Take into Consideration Non-Clean Air Act
Regulations That Will Impact the Ability of States to Require LargeScale Fuel Switching from Coal to Natural Gas to Achieve Their CO2
Reduction Goals.
When developing the basis for its BSER determination, EPA did not look at any non-Clean Air
Act regulations that will impact the ability of utilities to fuel switch at the large scale needed to
comply with the state goals in the Proposed Rule. For example, EPA did not consider the role of
Endangered Species Act (ESA) regulations on the natural gas pipeline permitting process or new
Department of Transportation regulations that require the retrofitting of existing natural gas
pipelines to improve their safety. EPA also did not take into account its own pending regulations
to eliminate polychlorinated biphenyls (PCBs) along pipeline segments211 with new non-PCB
materials. It is fair to assume that the application of such regulations will add time and costs to
the development of the infrastructure required to support the new gas generation. EPA should
have factored this into its analysis.
X.
EPA Should Withdraw and Re-Propose the Rule.
In these comments, APPA provides detailed information and analysis on the concerns we have
with the Proposed Rule. These include that the Proposal: (1) goes far beyond EPA’s statutory
authority under Section 111(d); (2) conflicts significantly with other federal, state, and local
authorities; (3) relies on questionable data and assumptions with respect to the long term
availability and price of natural gas; (4) makes unrealistic assumptions about the availability of
210
211
www.epa.gov/air/nitrogenoxides/actions.html
http://srrttf.org/wp-content/uploads/2012/08/EPAs-PCB-Use-Reassessment-08-24-12-2012.pdf
74
the infrastructure necessary to support the assumed massive increase in the use of natural gasfired generation; (5) would increase substantially the cost of electricity to consumers; (6)
imposes unduly stringent reduction requirements on many states by using unrealistic and overly
aggressive assumptions in the building block calculations; and (7) does not provide states with
sufficient time or flexibility to develop, implement, or modify, when appropriate, their plans for
implementation.
For all these reasons and others discussed in these comments, APPA urges EPA to withdraw and
re-propose the rule.
XI.
If EPA Will Not Withdraw the Proposed Rule, APPA Recommends
Several Modifications That Would Improve Its Workability.
In the event that EPA does not withdraw and re-propose the Proposed Rule, APPA strongly
recommends that EPA modify the Proposed Rule as follows:











Allow states to choose a baseline that accurately reflects their unique circumstances.
Provide full credit for investments already made that reduce or offset CO2 emissions.
Fix the errors and revise the assumptions in the computations of the four building blocks
in a manner that reflects what can realistically be accomplished and ensures greater
equity among the states.
Provide a streamlined process for NSR determinations and stipulate that an EGU’s
energy efficiency upgrade under a state compliance plan should be considered GHG Best
Available Control Technology (BACT) for PSD determinations.
Remove under-construction nuclear units from the relevant state baselines.
Allow all non CO2-emitting generation resources to be used for compliance.
Provide states with more time to develop state compliance plans.
Provide more guidance on the development of multi-state plans and interstate
agreements.
Eliminate the interim reduction goal and allow states to determine the emission reduction
trajectory (glide path) to reach their final reduction goals.
Allow a state’s final reduction goal, the year to achieve that goal, and/or the glide path to
be adjusted based on the discovery of materially changed circumstances, with the burden
of so demonstrating placed on the state.
Include and allow mechanisms to ensure that potentially regulated entities ha ve the
maximum degree of flexibility to comply with state plans at reasonable cost, including
additional reductions or avoidance measures from the energy sector.
75

Provide for the establishment of a reliability “safety valve” to ensure that compliance
with mandated emission reduction goals do not inadvertently impair system reliability or
conflict with NERC standards.
These recommendations are discussed in much more detail in the following sections.
XII.
EPA’s Selection of 2012 as the Baseline Is Inappropriate; States
Should Be Allowed Flexibility in Establishing a Representative
Baseline.
Choosing any single year for the baseline is problematic because it is inconsistent with the way
the industry operates. Various anomalies and unanticipated events can occur in any single year
that render that year unrepresentative of the performance of generating units within a state, and
often within a single utility. EGUs require regular maintenance and must be taken out of service
for extended periods of time to complete maintenance and upgrades. For example, Laramie
River Station (LRS) in Wyoming, which is comprised of three generating units, is on a threeyear maintenance schedule, with each unit being shut down for maintenance on a rotating basis.
Many utilities in the industry adhere to a three-year maintenance schedule. However, regardless
of the specific schedule utilities follow, the selection of a single year as a baseline of
measurement is unlikely to reflect the actual capability of the generating fleet.
Moreover, not all power plants were operating in 2012. In Minnesota, Sherburne County
Generating Station (Sherco) Unit 3 was offline for all of 2012 due to an unplanned outage,
making the starting point of Minnesota’s goal calculation atypical. This should be corrected, as
Minnesota will continue to rely on this generating unit (approximately 900 MW) for the duration
of the planning period, and likely beyond. Other units were running for only part of the year due
to other types of unplanned outages, such as those resulting from storms or floods. In addition,
changes in electricity demand, such as the loss of an industrial or large commercial customer,
can result in reduced output from a generating unit, even though it is still operational.
As recognized in EPA’s NODA,212 variations in weather have a significant impact on not only
the demand for electricity, but also the type of electricity generated in a given year. For
example, the amount of fossil generation in the Upper Great Plains region is significantly
affected by the amount of hydropower generation: during high water years, there is more hydro
212
http://www2.epa.gov/carbon-pollution-standards/clean-power-plan-proposed-rule-technical-documents#NODA
and http://www2.epa.gov/carbon-pollution-standards/clean-power-plan-proposed-rule-translation-state-specific-ratebased-co2
76
and less fossil generation, and likewise in low water years, there is less hydro and more fossil
generation. Averaging the baseline will tend to reduce the impact of weather anomalies as well
as unit outages.
Therefore, instead of prescribing 2012 as the baseline, the final rule should allow each state to
establish a baseline that the state believes equitably represents its circumstances, utility
generation, and related emissions. For example, EPA should allow a state to choose any three
years within the most recent ten-year period for which emissions data is available and then
average those three years to establish a baseline for its emission reductions and compliance plan.
Variations of this approach are obviously available as well. The key element is to provide some
flexibility so that the final emission goal is equitable.
XIII.
The Baseline and BSER Computations Should Allow Full Credit for
Early Action.
In the Proposed Rule, EPA did not properly take into account efforts by states and their utilities
to reduce emissions that occurred prior to 2012. By ignoring effects from efforts already
undertaken by electric utilities and states when it performed its analysis of what is re asonably
achievable to reduce CO2, EPA erred in setting BSER. This results in a BSER that effectively
penalizes a state for early actions that reduced CO 2 emissions. Early action typically involves
establishing energy efficiency programs that are most cost effective, as well as employing
renewable energy technologies that produce energy at the least cost to the consumer. This means
that early action states will find that both renewable energy and energy efficiency options to
meet building block 3 and 4 are more cost-limited than for states which did not aggressively
pursue early action to reduce CO 2 emissions. The result is a higher long-run cost of compliance
and fewer alternative choices for states that chose to reduce CO 2 prior to the EPA Proposal.
While actions, such as increasing energy efficiency, can improve environmental outcomes, there
is an optimal investment point over which further investment in a particular measure becomes
uneconomical, even in the context of compliance. Because states with utilities that have
undertaken renewable energy and energy efficiency programs are potentially forced to spend
more than other entities that may be regulated under this rule, it is appropriate for EPA to
designate some mechanism to provide credit for early action in establishing state emission
reduction goals. This concept is also appropriately mentioned in EPA’s NODA. To address the
inequities created by the lack of credit for early action in the Proposal, APPA suggests that if a
state or entity can show verifiably that it has reduced its emissions via some measure, it should
receive credit under the building block computation of the state’s emission reduction goal.
Emissions goal credit could be accounted for in state plans and EPA goal calculations by using
units of MWh for each state representing all actions taken by the state since 2005 that reduce
77
CO2. This unit of MWh would be subtracted from the denominator of the state rate calculation
or total mass reduction requirement incrementally starting in 2017 and depreciated at a rate of 10
percent over the following 10 years. For example, a state has entities that have verifiably
reduced CO2 emissions relative to business as usual by 10 MWh via a solar power system and
energy efficiency measures since 2005. The state should be allowed to calculate these measures
into its state goals by adding the 10 MWh to its goal/target denominator starting in 2017 and
decreasing that 10 MWh amount by 10 percent each year. This method would also help improve
the glide path issues noted by EPA in its NODA.213
Consider the following examples of early action from public power utilities:






In Wisconsin, the early addition of renewable resources significantly in excess of state
regulatory requirements amounting to 16.04 percent renewables in 2013 versus the
Wisconsin RPS requirement of 6.24 percent.
A 2010 steam turbine retrofit at Boswell Unit 4 in Minnesota, which increased plant
output by approximately 8.5 percent with no increase in emissions. In the case of EPA’s
proposal and goal calculation methodology, the state is penalized twice for the project–it
lowers Minnesota’s 2012 baseline emission rate and removes a potential compliance
measure.
2011 agreement to purchase approximately 162 MW resulting from an extended power
uprate of NextEra Energy Resources’ Point Beach Nuclear Plant. The uprate has the
effect of increasing the at-risk nuclear component in Wisconsin’s goal-setting calculation,
making Wisconsin’s goal more stringent.
In Minnesota, energy efficiency programs typically have an 11to12 year weighted useful
life. Between 2005 and 2012, state efficiency programs implemented by the Southern
Minnesota Municipal Power Agency have reduced over 900,000 tons of CO2. The state
of Minnesota should be allowed to take credit for that early action.
In Missouri, City Utilities of Springfield installed dense pack turbines to improve unit
efficiency in 2010, a project that was touted by EPA in the TSD to this rule. This early
action removed the lowest cost opportunities from the power plant and makes
significantly less additional efficiency available.
In recent years, the owners of LRS have taken significant steps to improve its heat rate,
including the following projects, along with the noted improvement in heat rate: turbine
upgrades, 200 Btu/kWh, hydrojet installation to clean the boiler walls, 40 Btu/kWh,
213
http://www2.epa.gov/carbon-pollution-standards/clean-power-plan-proposed-rule-technical-documents#NODA
and http://www2.epa.gov/carbon-pollution-standards/clean-power-plan-proposed-rule-translation-state-specific-ratebased-co2
78
installation of an intelligent soot blowing program, 20 Btu/kWh, and CO
monitors/combustion optimizer, 25 Btu/kWh. These upgrades, along with a rigorous
maintenance program, have helped LRS achieve and maintain a 3 percent efficiency
improvement. This improvement should be accounted for in the target requirement using
the method proposed above to reduce the overall compliance burden required by the state.
To prevent the punishing effect of increasing marginal compliance costs unduly, EPA should
allow full credit for early action.
XIV.
The Assumptions EPA Made in the Building Blocks Are Flawed and
Do Not Provide the Flexibility States Need to Meet Their CO2
Reduction Goals.
The Proposed Rule states that while it “lays out state-specific CO2 goals that each state is
required to meet, it does not prescribe how a state should meet its goal.” 214 It further states that
“each state will have the flexibility to design a program to meet its goal in a manner that reflects
its particular circumstances and energy and environmental policy objectives.” 215 APPA
appreciates the stated intent of EPA to provide states with flexibility to meet their goals, but
upon careful review believes the assumptions underlying the building blocks, which are used to
calculate those goals, are flawed and actually limit state flexibility. In addition, several of the
building blocks appear to work at cross-purposes with one another and further limit state
flexibility to achieve CO2 reduction goals. Below is a discussion of some of the key flaws APPA
has identified in each building block.
EPA’s approach to calculating each state’s interim and final emission goals is flawed. The
Agency’s goal calculation methodology contains numerous errors, inconsistencies, and
oversights that substantially and adversely impact each state’s goals. These errors do not affect
just a handful of states; they undermine the proposed goals for every state. To remedy this, EPA
must withdraw its Proposed Rule and re-propose a revised set of emission goals for public
comment. If EPA proceeds with this rulemaking, it cannot and should not finalize the Proposal
in its current form. The goal calculation methodology is so problematic and full of errors that
any rule, if corrected and finalized, would not be a logical outgrowth of the Proposal. Therefore,
EPA should make all necessary corrections, withdraw and re-propose this rule.
214
215
Page 16 prepublication version of proposed rule.
Id. at 16-17.
79
Figure 10: BSER Building Blocks
Source: Latham & Watkins 2014, the cost estimates are EPA’s and not endorsed by either APPA
or Latham & Watkins
A.
Building Block 1—Heat Rate Improvements
1.
EPA Overlooks the Significance of NSR Issues in Building Block 1
“EGU Efficiency Improvements.”
Building block 1 of EPA’s Proposed Rule consists of measures to reduce CO2 emissions from
coal-fired EGUs by improving heat rate, reducing the amount of fuel needed to produce
electricity.216 The Agency cites a 2009 Sargent & Lundy Report that identifies “equipment
upgrades at a facility that could provide total heat rate improvements in a range of approximately
4 to 12 percent.”217 The projects identified by Sargent & Lundy 218 are also included in a TSD
for this rulemaking, and they include upgrading soot blowers, boiler feed pumps, economizers,
turbines, boilers, air heaters, feed water heaters, condensers, forced draft (FD) and induced draft
216
See 79 Fed. Reg. 34,928.
79 Fed. Reg. 34859
218
See Sargent & Lundy LLC, Coal-Fired Power Plant Heat Rate Reductions at 2-1 to 5-4 (Jan. 22, 2009)
217
80
(ID) fans, pulverizers, condensate pumps, flue gas conditioning systems, selective catalytic
reduction systems, ash handling systems, neural network optimization systems, electrostatic
precipitators, and system controls.219 The chart below (Table 3) lists various efficiency
improvements in the TSD for this rulemaking and alleged NSR compliance issues raised by EPA
and citizens in EGU NSR enforcement actions.
All but one of the upgrades has been targeted by EPA and citizens in CAA NSR enforcement
actions. The significance of PSD permitting is two-fold: (1) “preconstruction review” requires
14-24 months and air quality dispersion modeling and NAAQS and increment analysis, as well
as the application of BACT (not “existing source NSPS”), and (2) in order to legally avoid NSR
review, the operator must permit emission limits on the unit’s production.220
The chart below lists various efficiency improvements in the TSD for the Proposal and NSR
compliance issues raised by EPA and citizens in EGU NSR enforcement actions.
219
U.S. EPA, Technical Support Document for Carbon Pollution Guidelines for Existing Power Plants: Emission
Guidelines for Greenhouse Gas Emissions from Existing Sources, GHG Abatement Measures, at 2-1-16 (June 10,
2014) (same) (“GHG Abatement Measures TSD”).
220
APPA cites further comments in UARG regarding EPA’s omission in the rulemaking of these NSR impediments
and inherent legal inconsistency of BACT with NSPS.
81
Table 3: Technology Assessment Modified and Existing Sources
Efficiency Improvements for Existing Coal-Fired EGUs and ICI Boilers
Efficiency
Improvement
Technology
Replace Turbine
Blading and/or
Rotor
Description
Combustion
Control
Optimization
Automated adjustment of
coal and air flow to
optimize steam
production.
Cooling System
Heat Loss
Recovery
Recovery of a portion of
the heat loss from cooling
water exiting the steam
condenser
Reported
Efficiency
Increase221
0.84-2.6%
EGU watt-peak
(Wp) 28
APPA IDENTIFIED EPA ALLEGED
COMPLIANCE ISSUE
0.15-0.84%
EGU Wp-28
EPA and Sierra Club NSR complaints
assert that any change to an analogue
system to adjust air flow into a boiler
optimizes steam production, enabling the
unit to burn more fuel, triggering NSR
because they are routine changes.
Replacement and rebuilds of condensers
are sometimes identified by EPA as alleged
NSR violations because of their capital
cost, but they have not to our knowledge
been identified as an alleged NSR
violation.
Instrumentation
adds up to ~3%
ICI Wp- 8
0.2 to 1 %
EGU Wp 28
221
In one of the most notorious of the CAA
NSR cases alleging PSD/NSR violations
for replacement and upgrades in high
pressure section of two steam turbines
involved retrofit of a Detroit Edison
turbine with a General Electric (GE) dense
pack (rotor and blades). Starting with the
initial complaints against Cinergy and
Tennessee Valley Authority (TVA) in
1999, replacement turbine blades are
alleged to violate NSR.222
Citations are to EPA’s White Paper titled “Available and Emerging Technologies for Reducing Greenhouse Gas
Emissions from Coal-Fired Electric Generating Units” (Oct. 2010),
http://www.epa.gov/nsr/ghgdocs/electricgeneration.pdf (“EGU WP”) and “Available and Emerging Technologies
for Reducing Greenhouse Gas Emissions from Industrial Commercial, and Institutional Boilers (Oct. 2010)”
http://www.epa.gov/nsr/ghgdocs/iciboilers.pdf (“ICI WP”).
222
Letter from Francis X. Lyons, Reg’l Adm’r, EPA, to Henry Nickel at 2, 3 (May 23, 2000) (“Detroit Edison
Determination”), available at www.epa.gov/ttn/nsr/gen/letterf3.pdf. See Also EPA legal analysis of NSR impacts at:
http://www.sagady.com/stuff/EPAMonroePlantBrief.pdf
82
Efficiency
Improvement
Technology
Replace/
Upgrade Burners
Flue Gas Heat
Recovery
Description
Older, wrongly sized, or
mechanically deteriorated
burners are typically
inefficient. Inoperable
dampers, broken registers,
or clogged nozzles will
render an otherwise good
burner into a poor
performer. These
inefficiencies result in
incomplete combustion
(high carbon monoxide
(CO) emissions and
unburned carbon) and the
need for high excess air.
It may be possible to
recover heat lost when
flue gas exits the boiler by
installation of a condenser
exchanger to heat preheat
boiler feedwater.
Reported
Efficiency
Increase221
Up to 4-5%
ICI Wp—8
0.3 to 1.5%
EGU Wp 28
83
APPA IDENTIFIED EPA ALLEGED
COMPLIANCE ISSUE
From the initial November 3, 1999, NSR
utility enforcement initiative action filed
by U.S. Attorney General Janet Reno
against eight investor-owned utilities
(IOUs) and TVA-to-the present NSR
actions against utilities and nearly every
other industry sector in the U.S., EPA has
asserted that addition and/or replacement
and/or or upgrading burners in any type of
boiler resulted in a significant emissions
increase from the unit requiring PSD
review.
http://www2.epa.gov/enforcement/coalfired-power-plant-enforcement Eleven
burner upgrades and replacements were
alleged to have violated NSR in the first
nine utility enforcement cases. Note also
that even installation of low-NOx burners,
a pollution prevention device, is not
exempt from NSR if it increasing the
efficiency of a boiler See New York v. EPA,
433 F.2d 3 (D.C. Cir. 2005), cert denied.
EPA, Sierra Club, and Wild Earth
Guardians have routinely identified as
NSR violations the installation of
condensers to recover flue gas as a
physical change to the combustion unit that
increases the capacity of a boiler because it
significantly increases SO2, NOx, and CO
emissions.
Efficiency
Improvement
Technology
Air Preheaters and
Reheaters
Economizers—
replacements or retubing
Feedwater Heaters
and Improvements
Description
For most fossil fuel-fired
heating equipment,
energy efficiency can be
increased by using waste
heat gas recovery systems
to capture and use some
of the heat in the flue gas.
Heat recovery equipment
includes various type of
heat exchangers
(economizers and air
heaters), typically located
after the gases have
passed through the steam
generating sections of the
boiler.
An economizer preheats
condensed feed water
recycled back to the boiler
tubes to the boiler
enabling the boiler to heat
water more efficiently
increasing output,
including air emissions as
a result of increased
output.
Using other heat sources
for the feedwater heater
avoids the need to extract
steam from the turbine
allowing the steam to be
used for electric power
generation
Reported
Efficiency
Increase221
~1% per 40
degree temp.
decrease up to ~
4%
ICI Wp 9
APPA IDENTIFIED EPA ALLEGED
COMPLIANCE ISSUE
40 degree
decrease in flue
gas temp = ~1%
ICI Wp 8-9
The cost of a new economizer can exceed
the cost of $2.3 million for a large boiler
according to EPA’s EGU and ICI
whitepapers. Replacement of portions of
an economizer tubing, typically exceeds
$100,000 and is not considered routine
maintenance because they are neither
frequent or “like kind replacements,”
typically involving installation of higher
grade chemical resistant coating.
Economizer replacement and economizer
tubing replacement is cited in nearly every
complaint EPA or any citizen has ever
filed against an electric utility. NSR
violations for economizer installations,
repairs and replacements were alleged 37
times in the first nine NSR enforcement
cases brought by EPA.
Increases the output of the steam cycle in
turn increasing the output from the boiler
and EPA has alleged that feedwater heater
replacement and new installation is not
done frequently, and boosts output or
restores lost capacity which triggers NSR
review.
EGU Wp p. 34
discusses but does
not specify
efficiency
improvements.
84
Heat recovery also includes installation of
air preheaters, as well as changes in design
of preheaters, including baskets in a
preheater. EPA and Sierra Club assert they
trigger NSR because they increase or
recover lost capacity from a boiler
increasing NSR-regulated pollutants.
Efficiency
Improvement
Technology
Cooling System
Heat Loss
Recovery
Flue Gas Heat
Recovery
Low-Rank Coal
Drying
Sootblower
Optimization
Description
Recovery of a portion of
the heat loss from cooling
water exiting the steam
condenser
Recovery of the heat lost
when flue gas is sprayed
with flue gas
desulfurization (FGD)
reagent slurry and cools
Drying of subbituminous
and lignite coals using
waste heat from flue gas
and/or cooling water
systems
Intermittent injection of
high velocity gets of
steam or air to clean coal
ash deposits from boiler
tube surfaces to maintain
adequate heat transfer
Reported
Efficiency
Increase221
0.2 to 1 %
EGU Wp 28
0.3 to 1.5%
EGU Wp 28
0.1 to 0.65%
EGU Wp 28
0.1 to 0.65%
EGU Wp 28
85
APPA IDENTIFIED EPA ALLEGED
COMPLIANCE ISSUE
Recover a portion of the heat loss from the
warm cooling water exiting the steam
condenser prior to its circulation thorough
a cooling tower or discharge to a water
body. The identified technologies include
replacing the cooling tower fill (heat
transfer surface) and tuning the cooling
tower and condenser. Replacement and
rebuilds of condensers are generally
always identified by EPA as alleged NSR
violations. Replacing the cooling tower fill
has not to our knowledge been identified as
an alleged NSR violation.
Recovering lost energy in the flue gas to
preheat boiler feedwater via use of a
condensing heat exchanger has been
alleged by EPA to violate NSR because it
increases potential output from the boiler.
“Low-rank coal drying” has not nominally
been alleged as an efficiency improvement
that triggers NSR in enforcement actions.
However, general modifications of coal
handling systems to dry coal to make it
easier to handle, convey, and pulverize –
improving the overall efficiency have
generically been described in a number of
EPA NSR complaints and EPA CAA
requests for information utilized by the
Agency in preparation of NSR
enforcement actions.
Sootblowers intermittently inject high
velocity jets of steam or air to clean coal
ash deposits from boiler tube surfaces in
order to maintain adequate heat transfer.
The identified technologies include
intelligent or neural-network sootblowing
(i.e., sootblowing in response to real-time
conditions in the boiler) and detonation
sootblowing. Sootblowing has been
alleged by the Sierra Club to violate state
and federal opacity standards, as well as
NSR for particulate.
Efficiency
Improvement
Technology
Steam Turbine
Design
Description
Maintain mechanical and
physical condition of
steam turbine use of
efficiently designed
turbine blades and stead
seals
Reported
Efficiency
Increase221
0.84-2.6%
EGU Wp-28
APPA IDENTIFIED EPA ALLEGED
COMPLIANCE ISSUE
In the Notice of rulemaking, EPA recognizes the potential NSR consequences of implementing
building block 1 and states that it expects there will be “few instances” where “an NSR permit
would be required.”223 But clearly these views are not shared by EPA’s enforcement arm, as is
evident from the Detroit Edison determination and the hundreds of projects targeted in the
enforcement initiative since then. Based on the history of EPA’s NSR enforcement initiative,
EPA’s assurance that there will be “few instances” in which building block 1 projects would
trigger NSR is not supported by the preponderance of past actions by the agency.
Even if EPA believes there will be “few instances” where an NSR permit would be required,
there is no suggestion that all states or citizens share that belief. Citizen plaintiffs have been just
as active as EPA in litigating NSR suits over the past 15 years. Even when those citizen suits
lack merit, they often delay the implementation of efficiency improvement projects, take several
years to litigate, are very expensive, and drain scarce resources of the parties and courts. This is
an artifact of EPA’s Proposal that adds considerable risk and expense.
EPA’s failure to account for the potential cost of NSR—and NSR uncertainty—for building
block 1 projects in the Proposed Rule is arbitrary and capricious. As a result, utilities
implementing building block 1 that do not project an increase in emissions as a result of an
equipment upgrade will face the ongoing threat of NSR litigation years after their projects. In an
enforcement action filed just last year, for example, EPA sued Oklahoma Gas & Electric
Company for allegedly violating NSR even though emissions have decreased since the projects
were undertaken.224 In this suit, EPA brought these claims following the very same types of
upgrades it is now recommending states to implement under building block 1: the replacement of
economizers and turbine blades and the addition of heat transfer surface in boilers. EPA should
eliminate the threat of protracted NSR litigation and provide a clear statement that any
upgrades necessary to implement building block 1 do not trigger NSR.
223
224
79 Fed. Reg. 34,859.
United States v. OG&E, No. 5:13-cv-00690 (W.D. Okla.).
86
Because EPA’s justification of the state emission goals relies on the ability of sources to
implement efficiency improvement measures, and because EPA has failed to propose any
credible regulatory provisions to otherwise address this issue, EPA has failed to demonstrate the
achievability of its goals as required by section 111. APPA asks EPA to review the UARG
comments for more information.
2.
EPA’s Analysis of Historical Data from Coal-Fired Units Fails to
Provide Any Support for Its Claim that Heat Rate Improvements
of 4 to 6 Percent Are Achievable.
EPA is proposing to find that overall heat rate improvements of 6 percent (or 4 percent under the
alternate goals EPA solicits comment on) are achievable at existing coal-fired EGUs under
building block 1.225 The Agency based its estimates of achievable heat rate improvements
primarily on measures described in a 2009 report by Sargent & Lundy, along with its own
limited analysis of historical generating data. 226 EPA’s estimates are arbitrary and capricious,
and they demonstrate the Agency’s poor understanding of the nature, cost, and availability of
heat rate improvement measures. EPA has failed to consider several critical factors that will
make it impossible for affected EGUs to reach this goal individually or on average across any
state.
APPA also notes that forcing changes in the operation of coal plants will create stranded costs.
For example, at LRS in Wyoming, because there is no practical way to reduce the heat rate under
building block 1, the only option remaining at the source is a drastic one: the owners are faced
with the requirement to run the plant less or shut down one or more units. The first option,
running the plant less, will reduce the efficiency of the plant (and the heat rate, contrary to EPA’s
objective) and cause an increase in the operating costs—wholesale cost increases that will have
to be passed on to their customer-owners. It is unlikely, however, that simply running the plant
less will be sufficient to meet the CO 2 emission reduction required of LRS, whether under the
rate-based or mass-based approach. If, instead, the owners are forced to shut down a unit
prematurely, it will cause significant stranded costs and will run afoul of the directive of 111(d)
to take into account remaining useful life.227 In EPA’s NODA,228 it asks if it should consider
225
79 Fed. Reg. at 34,861.
Id. at 34,859; GHG Abatement Measures TSD.
227
CAA 111(d)(1)(B)
228
http://www2.epa.gov/carbon-pollution-standards/clean-power-plan-proposed-rule-translation-state-specific-ratebased-co2
226
87
such issues in adjusting timeline and compliance requirements.229 At a minimum, EPA should
allow all stranded costs to be considered in adjusting timelines and targets accordingly to allow
recovery of costs.
As a threshold legal and regulatory issue, many of the operating practices and equipment
upgrades that are the basis of the Agency’s assertion that 4 to 6 percent heat rate improvements
are achievable are not included in a Subpart Da affected facility. For this reason, EPA lacks
authority to regulate such equipment under Section 111(d). Subpart Da is the codification of the
NSPS for electric utility steam generating units, and these pieces of complicated equipment,
logically enough, generate steam. The “affected facility” is “each electric utility steam
generating unit.”230 Equipment within the boiler island is the Subpart Da affected facility, while
equipment beyond the boiler island is not.
APPA agrees with UARG’s comments that the following equipment is specifically beyond the
purview of this rule: steam turbines, water purification equipment, water-supply systems, air
cleaning and cooling apparatuses, condensers, main exhaust and main steam piping, water
screens, motors, and moisture separators for turbine steam. EPA’s attempt to require
improvements in these components’ efficiency for building block 1 is contrary to law.
Even if EPA had authority to regulate these pieces of equipment under Section 111, its technical
conclusions are erroneous. As discussed by UARG’s consultants J. Edward Cichanowicz and
Michael Hein, the Agency does not account for the fact that the efficiency benefits associated
with the measures identified in the Sargent & Lundy report are highly variable by unit, are not
cumulative, and are only temporary.231 Large benefits from steam turbine upgrades (the highestpayoff measure) are only possible for units that are already severely degraded; for most units, the
available payoffs would lie at the low end of the possible range. Many of the available heat rate
improvements for which the efficiency benefits outweigh the costs have already been
implemented by EGU owners for economic reasons and would not yield substantial benefits if
they were implemented again.
Many of the available actions to improve heat rate do not provide cumulative benefits and thus
cannot be added together to estimate the potential efficiency gains at coal-fired EGUs. In
particular, measures that increase heat removal from the boiler, such as economizer
modifications, improved air heater performance, and low temperature heat recovery, do not
229
http://www2.epa.gov/carbon-pollution-standards/clean-power-plan-proposed-rule-technical-documents#NODA
40 C.F.R. § 60.40Da.
231
J. Edward Cichanowicz & Michael C. Hein, “Evaluation of Heat Rate Improving Techniques for Coal-Fired
Utility Boilers as a Response to Section 111(d) Mandates” (Sept. 12, 2014).
230
88
provide additive efficiency benefits because any heat that is recovered by an individual project
cannot be recovered a second time.
EPA ignores in its building block 1 analysis the ways in which EGUs’ heat rates (and thus, CO2
emissions) are negatively impacted by operating load and by auxiliary power requirements
needed to run associated equipment and emission controls. In particular, the Agency has
overlooked the negative impact on coal-fired EGU efficiency of an EGUs’ obligations under the
remaining components of its proposed BSER and under other CAA programs. For example,
based on regional haze rule requirements, the operators of LRS are installing a selective catalytic
reduction system. The fans associated with this system are expected to increase station power
requirements by 22.5MW, or 7.5 MW per unit. Based on generating capacities of 570MW per
unit, this power increase means regional haze rule related compliance controls contribute to a 1.3
percent reduction of plant efficiency at full power and a 1.7 percent reduction in efficiency at 75
percent capacity. EPA should adjust its modeling assumptions and allow states to adjust their
targets based on compliance obligations that power plants have under other CAA programs.
Because EPA failed to consider the effect of looming or recently completed emission control
projects that adversely affect EGU heat rate, affected EGUs will be forced to overcome the
energy penalties associated with these controls and then achieve an additional 4 to 6 percent
improvement in heat rate, on average, in order to comply with the Proposed Rule. In addition to
these fundamental oversights, the Agency has failed to provide any reasonable analysis of
whether its heat rate improvement targets are achievable. For these reasons, the “assessment of
heat rate improvement potential” in the GHG Abatement Measures TSD does not support EPA’s
claim that state-wide heat rate improvements of 4 to 6 percent are achievable.232 Again, because
plants will have to install control equipment for both this rule and other rules, and EPA ignored
the energy penalty associated with such equipment, the Agency’s “assessment of heat rate
improvement potential” is arbitrary and capricious. 233
Under building block 2 of the Proposed Rule, EPA is requiring states to redispatch generation
from coal-fired EGUs to NGCC units, which will result in more coal-fired EGUs being
dispatched as load-following units rather than providing baseload at high capacity factors. As
EPA itself acknowledges, EGUs have higher heat rates when operating as load-following units
and during periods of startup and shutdown.234 The Agency should adjust its assumptions and
232
See GHG Abatement Measures TSD at 2-30 to 2-34.
Id.
234
GHG Abatement Measures TSD at 2-5, 2-21
233
89
allow states to modify their goals to account for reduced heat rate that occurs due to loadfollowing, variable renewable energy resources.
APPA agrees with UARG that EPA’s “model” assessing potential heat rate improvements via
“best practices” is arbitrary and capricious. In this “model,” the Agency grouped hourly heat
input data into “bins” representing ranges of load and ambient temperature, then performed crude
calculations to reduce the variation among observations in each bin by arbitrary increments of
10, 20, 30, 40, and 50 percent. EPA used these adjusted values to calculate an adjusted heat rate
for each unit, then compared the study population’s average adjusted heat rate under each
scenario to the population’s actual average heat rate in order to determine the “improvement”
under each scenario, with a roughly 30 percent reduction in each unit’s variation corresponding
to a 4 percent improvement in the overall study population’s heat rate.
APPA agrees with UARG that this conclusion is arbitrary and unfounded. Absent from EPA’s
analysis is any discussion linking the 4 percent “best practices” estimate to any identifiable,
technically feasible heat rate improvement measures. At most, EPA’s discussion shows that a 4
percent reduction in overall heat rate is mathematically feasible if sources can find a way to
reduce variation in hourly heat input. The Agency makes no attempt to show whether the
measures it identifies as “best practices” are technically capable of reducing heat input variation
at individual units at all, let alone reducing variation to a sufficient degree to reach the target
efficiency level.
On a more fundamental level, EPA’s analysis incorrectly assumes that the existence of variation
in heat rate at individual units means there is “significant variation in the operation of EGUs,”
indicating that “significant potential for heat rate improvement is available through the
application of best practices.”235 In fact, heat rate variability at individual units is driven by the
design, duty cycle, fuel type, size, cooling conditions, and location of each unit, and it cannot be
ameliorated by changes in operating practices. EPA did not assess differences in variability
based on these factors or control for these factors in its achievability analysis.
EPA’s assessment of available heat rate improvements through “equipment upgrades” is
similarly arbitrary. The Agency set this component of the target by identifying the four most
costly heat rate improvement methods listed in the Sargent & Lundy report, based on average
estimated $/kW costs, then simply adding together the average estimated Btu/kWh
improvements for each. Based on this calculation, EPA concluded that the four specified
equipment upgrades could provide a 4 percent heat rate improvement if all were applied on an
235
GHG Abatement Measures TSD at 2-30
90
EGU that has not already made them, but “conservatively” reduced the target to 2 percent
because “some units may have applied at least some of the upgrades.”236
This methodology erroneously assumes that the heat rate improvements from these upgrades are
cumulative and that they provide consistent long-term benefits. In reality, the combined payoff
from these four upgrades will be less than the sum of each measure’s payoff would be when
applied alone to an EGU, and these heat rate benefits will begin to degrade immediately once the
unit returns to service. EPA cannot require affected EGUs to duplicate emission reductions they
have achieved using measures they have already taken.
Some of the units that are in service currently were not running at or near full capacity in 2012
and therefore have resulted in an incorrect baseline for the purposes of calculating goals. For
example, Sherco Unit 3 in Minnesota was not running in 2012 due to issues related to a turbine
upgrade. EPA’s NODA appears to address this question, but if it promulgates a final rule that
relies solely on 2012 data, the Agency should clarify how efficiency can be improved on a unit
that was not running at the time of its analysis and hence not included in the baseline. EPA
should also address units that are operating at maximum efficiency. For example, Iatan 2, Plum
Point, and Prairie State units all came online in the last three years and all have the most efficient
technology. There is no improvement technically available for these units regardless of cost and
that was not factored into EPA’s baseline calculations. The Agency should adjust its
assumptions to take into account units that do not have any efficiency improvement technically
available.
EPA has failed to satisfy its obligation to demonstrate that its proposed Building Block 1 target
of 4 to 6 percent improvement in heat rate for coal-fired EGUs statewide is achievable “under the
range of relevant conditions which may affect the emissions to be regulated.”237 Therefore, EPA
should assess what efficiency upgrades have already been implemented by affected coalfired EGUs in order to determine what level of additional heat rate improvements is
achievable. It should also assess other factors that will determine what level of heat rate
improvements is achievable for individual coal-fired EGUs across the source category as a
whole, such as whether the upgrades are compatible with the design and hardware present
at a unit and whether the upgrades may present operational reliability issues at a
particular unit.
236
237
79 Fed. Reg. at 34,860.
Nat’l Lime Ass’n v. EPA, 627 F.2d at 433.
91
An even bigger concern is that building block 1 seems fundamentally at odds with building block
2. Why would a utility make heat rate improvements at a coal-fired power plant that would cost
hundreds of thousands or millions of dollars only to not be able to run that unit as often, if at all,
because of compliance with block 2, where natural gas units would be dispatched first? How
would a utility finance such a project if its creditors can have no assurance that the plant would
be operated sufficiently to pay back the debt service? Even if heat rate improvements were
made, decreased plant utilization (lower load operation and increased cycling) would degrade the
plant’s heat rate; partially or fully offsetting the effect of the heat rate improvement projects.
The inherent conflict of building blocks 1 and 2 call into question the assumptions underlying
those building blocks, as well as EPA’s understanding of how utilities operate. The Agency
must address these issues and amend building block 1 to enable realistic heat rate improvements
at power plants.
B.
Building Block 2—Redispatch of Natural Gas Units
1.
EPA Improperly Calculated Capacity Factor and Number of
Hours in Building Block 2—Existing Natural Gas CombinedCycle Generation—and Should Correct Its Calculations.
In computing the level of existing NGCC generation that will be redispatched to displace both
coal- and oil- & gas-fired (O/G) steam generation, EPA erred in using nameplate capacity and
not summer net capacity to apply its NGCC post-redispatch capacity factors. Using nameplate
capacity results in EPA adding an erroneously high level of gas-fired generation into a state’s
goal calculation, thereby excessively reducing the amount of coal-fired generation included in
determining a state’s goal above and beyond what can reasonably be considered BSER. Summer
is when NGCC units are most often dispatched at the highest capacity factors, reflecting when
demand for electricity is greatest in most areas. The Agency properly used 2012 summer net
capacity from each state’s nuclear fleet in determining each state’s nuclear capacity “at risk .”238
This method should have been applied for NGCC units and EPA erred in not doing so.
EPA used a leap year to calculate the hours that must be run to achieve a 70 percent capacity
factor, another erroneous assumption. The number of hours used to determine what megawatt
hours are run by the fleet at 70 percent capacity factor should be set using either a typical year or
the average of hours per year over four years (8,760 or 8,766). Despite the theory EPA uses to
take the generation in 2012 and divide it by the hours in 2012 (because there is only one year of
238
See table 4.10 in EPA’s Greenhouse Gas Abatement TSD
92
data and some plants may not have been functioning normally), the extra 24 hours of data could
influence a state’s requirement by up to .002 percent compared to a normal year, which in some
states could amount to several hundred thousand MWh. For example, in Texas, the amount of
NGCC generation calculated at 70 percent capacity factor over 8,784 hours is 230,875,142 MWh
while at 8,760 it could be 201,331,956 MWh. The only way for EPA to solve this problem is to
allow states to use multiple, representative data years. The Agency appears to attempt to address
this in the NODA and APPA encourages it to allow states to use multiple years in setting
baseline requirements, as discussed in Section XII, as a solution for this issue.
Table 4 below illustrates the difference in generation output using a 70 percent annual capacity
factor with nameplate capacity versus summer net capacity for eight states. These states were
used as examples, but many states are impacted significantly by EPA’s invalid assumptions. The
difference in generation represents the excessive level of NGCC generation that EPA has
assumed into its building block 2 for those seven states by using nameplate capacity instead of
summer net capacity. For a state, such as Minnesota, to be required to meet a NGCC capacity
factor greater than 90 percent based on summer net capacity is both unequitable and unrealistic.
The Agency should adjust the target appropriately to reflect summer net capacity for all states as
illustrated in the table.
Table 4: Capacity Factors for Eight State Examples and Corrected Numbers for EPA from
EIA Generator Y2012
Corrected
NGCC
Calculation
at 70%
Capacity
Factor using
Summer
Net
Generation
(MWh)
Hours that
Will Need
to Be Run
Based On
Net
Summer
Capacity of
Generating
Fleet if EPA
Does Not
Revise its
Calculation
Real
Capacity
Factor
that EPA
Applied
by Not
Using
Net
Summer
Capacity
for
NGCC
Corrected
Maximum
Capacity
Factor EPA
Should
Have
Applied If
Using
Nameplate
Capacity
Corrected
2030
Goal in
lb/MWh
State
NGCC
Nameplate
Capacity
(MW)
NGCC
Summer
Net
Capacity
(MW)
EPA's NGCC
Calculation
at 70%
Capacity
Factor
(MWh)
AL
10,333
9,278
63,535,550
57,048,566
6,848
78%
63%
1,124.00
AR
5,588
4,660
34,359,494
28,653,408
7,353
84%
58%
1,039.87
FL
29,485
23,784
181,297,368
146,243,059
7,602
87%
56%
912.98
GA
8,354.9
7,956
51,372,609
48,919,853
6,439
73%
67%
860.18
NC
4,709
4,075
28,954,699
25,056,360
7,086
81%
53%
1,063.40
OK
8,035
7,512
49,405,608
46,189,786
6,559
75%
65%
934.76
TX
37,548
32,833
230,875,142
201,883,550
7,013
80%
61%
863.22
MN
2808.5
2120.8
17,268,905
13,040,375
8,120
92%
54%
992.71
93
Net summer capacity is the maximum output, commonly expressed in megawatts (MW), that
generating equipment can supply to system load, as demonstrated by a multi -hour test, at the
time of summer peak demand (period of June 1 through September 30). This represents the
reality of the electric system. As Table 4 shows, the capacity factors that would actually be
required to reach these generation levels when applied to summer net capacities in these seven
states are extremely high.
In addition, duct burners may be operated when determining summer and winter peak ratings,
and the amount of heat input from duct burners can be significant. EPA should use summer
capacities without duct burner operation for purposes of goal setting as duct burners decrease the
efficiency of a plant and therefore do not fit the Agency’s ideal NGCC operation
model. Modeling goals without duct burners would help to insure that duct burners do not need
to be used excessively or exclusively to meet the MWh or mass goals.
EPA’s methodology for building block 2 includes excessive NGCC generation into a state’s goal
calculation due to its use of nameplate capacity, thereby excessively reducing the amount of
coal-fired generation included in determining a state’s goal. EPA should correct for this
erroneous calculation and remove this additional gas-fired generation from the
redispatched NGCC generation element of the target in order to make its Proposal more
workable.
2.
EPA Should Adjust Its Calculated Building Block 2 Targets
Where the Integrated Planning Model (IPM) Does Not Assume
Removal of Coal Will Occur.
In EPA’s goal setting methodology for Building Block 2 under Option 1, it essentially retires all
coal capacity in 11 states and replaces that generation with existing NGCC. These states are:
Alaska, Arizona, California, Connecticut, Massachusetts, Mississippi, Nevada, New Hampshire,
New Jersey, Oregon, and Washington. Two of these states (Oregon and Washington) have
announced the retirements of their entire coal-fired capacity as a result of various state
agreements. However, the remaining nine states have only announced the retirement of some of
their existing coal capacity and have plans to keep their remaining coal fleet operating into the
future. EPA’s calculated requirements show a disconnect between the operational reality of the
U.S. electric system and the Agency’s assumptions.
EPA’s parsed IPM modeling files projecting compliance with the Option 1 state goals in 2025
under a state-based approach indicate that coal-fired capacity is required to meet demand in
Arizona, Massachusetts, Mississippi, Nevada, and New Jersey. This indicates a disconnect
between EPA’s assumptions related to NGCC redispatch in building block 2 and their modeling
of state compliance with Option 1 goals. For example, EPA’s building block 2 calculations
displace (retire) all of Arizona’s existing coal-fired capacity; however, IPM’s 2025 results for
94
Option 1 under a state approach indicate that Springerville Units 1 through 4 will be needed to
generate 5,653 gigawatt hours (GWh) of electricity, which is equivalent to 23.2 percent of the
state’s 2012 coal-fired generation. Assuming a state can shift all of its coal-fired generation to
its NGCC units without taking into account the reliability needs of specific electricity markets,
EPA shows a lack of understanding of how electric utilities plan and operate to meet electric
demand and ensure reliability of the grid. EPA should correct the Proposal to not indicate coal
fleet retirement where the IPM does not indicate retirement.
More importantly, the use of average annual capacity factor for purposes of redispatch instead of
considering such factors as peak demand in a particular state is a faulty approach. In meeting
peak demand, many states have all their existing NGCC resources in use, along with all other
resources, leaving no available NGCC generation available to replace existing coal and oil and
gas (O/G) steam generation during these periods. This is further complicated by states in which
there are significant amounts of merchant gas generation, which are used to meet peak demand
not only in their home states, but also in neighboring states. Therefore, it is inaccurate for EPA
to assume that all merchant generating capacity will be available to displace coal-fired
generation from within a particular state.
An analysis done by the Salt River Project (SRP) on this particular issue, which was submitted to
the state of Arizona, illustrates the various faults in EPA’s methodology for building block 2.239
On August 8, 2012, SRP reached a peak hourly load of 6,663 MW, in which all of its generation
resources (including coal and O/G steam units) were being utilized at full capacity, as shown in
the figure below.
Figure 11: SRP Resources Needed to Meet Peak Demand
239
See Salt River Project, Building Block #2 Impacts on the Emission Rate Goals for Arizona Under EPA’s Clean
Power Plan Proposal, Aug. 2014.
95
Even then, SRP was forced to purchase power to meet its peak demand and its mandated reserve
requirement. With coal and O/G steam units prohibited from operating for purposes of meeting
the state’s CO2 emission goal, as EPA’s Interim and Final Goals for Arizona require, SRP would
have been unable to meet its peak electricity demand without purchasing very substantial
additional amounts of electricity on the short-term market. This highlights another hurdle that
EPA failed to consider: the 5,000 MW of merchant NGCC in the state.
The state’s various LSEs purchase electricity from merchant generators on either long- or shortterm agreements, but have no control on how these units are dispatched. SRP found that all
merchant plants in Arizona operate at or near full output during the peak months, not only to
meet the demand in Arizona, but also in neighboring states. Consequently, Arizona utilities
cannot rely on merchant generation to meet long-term demand requirements during peak summer
months, and both coal and O/G steam capacity provide a vital resource to meet demand and
ensure reliability.
In justifying the need for revised building block 2 calculations based on power system dynamics,
fuel switching infrastructure time frames, and reliability issues, several factors can complicate
the construction of additional NGCC capacity:




Lengthy time lines in siting and permitting new energy infrastructure, which can
take up to 10 years or more if it is on federal lands.
The regulatory complexities of siting a new energy facility in non-attainment
areas, which would require emission offsets.
Recent modeling has indicated that if all existing Arizona coal capacity is retired
in 2020 it would affect both the reliability and load serving capability of the
state’s transmission system.
EPA has simplistically assumed uniformity across RTOs, without considering
intra-company, intra-regional dispatch that does not compromise a company and
RTO’s ability to meet its load.
EPA should work with states to adjust building block 2 targets based on these factors or any
other factors deemed reasonable by a state.
For example, in South Dakota, there is one NGCC—Deer Creek station (owned by Basin
Electric Power Cooperative)—which commenced commercial operation in August 2012 and had
a 1 percent annual capacity factor. In EPA’s goal setting methodology for building block 2, Deer
Creek is ramped up to a 70 percent capacity factor, requiring South Dakota’s lone coal-fired unit
(Big Stone) to reduce operations to an annual capacity factor of 23 percent. However, EPA fails
to consider that the Deer Creek and Big Stone plants have different owners and are located in
96
different RTOs (Big Stone in the Midcontinent Independent System Operator and Deer Creek in
the Southwest Power Pool, beginning in 2015) and serve different loads.
The inclusion of both newer coal and NGCC capacity that entered operation in 2012 also biases
EPA’s displacement of coal-fired generation with existing and under construction NGCC
generation. Many of these newer units had very low capacity factors in 2012, resulting in either
lower amounts of coal-fired generation being reported in the 2012 baseline or an artificially high
amount of NGCC capacity being available to be ramped up to displace coal capacity. EPA
should correct its interim and target CO 2 goals by several percent to accommodate this
assumption.
3.
In Building Block 2, EPA Double Counted Some Units in Both the
Existing NGCC Capacity and the “Under Construction” Capacity.
EPA Should Remove Those Units from Its Goal Calculations.
In building block 2, EPA not only includes existing NGCC capacity, but NGCC capacity under
construction, as elements in setting a state goal. For example in Florida, the Proposed Rule lists
29,485 MW of existing NGCC capacity in 2012; however, this is incorrect. In 2012, Florida had
28,067 MW of existing NGCC capacity. The difference is that EPA included the Cape
Canaveral NGCC (1,295 MW), which did not begin commercial operation until April 2013, and
the Orlando Cogen, a 122.4 MW compressed storage facility. In addition, the Agency also
included the Hansel combined cycle facility (55 MW) that was retired in October 2012. The
Cape Canaveral NGCC facility should be shifted to units under construction, and based upon
EPA’s building block 2, only 15 percent of that capacity should be assigned to goal
development.
In South Dakota, Deer Creek Station, the only NGCC plant in the state, was modeled by EPA as
operating at a 1 percent capacity factor during 2012. However, Deer Creek should have been
considered “under construction” during 2012 because it did not go into commercial operation
until late in 2012 and had only 190 total run hours for the year. While firing of the Deer Creek
unit may have begun in April, it was not commercially operated for the bulk of the year. The 1
percent capacity factor is clearly unrepresentative, and South Dakota was the only state that had
a less than 10 percent NGCC capacity factor applied in EPA’s building block 2 calculation.240
EPA’s proposed methodology, as currently applied to South Dakota, produces flawed targets that
the Agency should adjust. For example, applying building block 2 in South Dakota under the
240
TSD: Goal Computation, Data File: Goal Computation – Appendix 1 and 2 (XLS).
97
proposed methodology results in Big Stone Plant, the state’s only coal-fired EGU, having to
operate at a 23 percent capacity factor, forcing it to be offline at least half of the year. This also
assumes that it is technically feasible for South Dakota to redispatch resources under building
block 2. This assumption is incorrect because Deer Creek Station and the Big Stone Plant
operate in different RTOs. EPA should recalculate its proposed targets with these plants in their
proper categories.
In the Goal Computation TSD, EPA indicated that all NGCC units under construction that were
included in building block 2 were obtained from the National Electric Energy Data System
(NEEDS) v.5.13—they are listed in the table below. In addition to those under construction
NGCC units in NEEDS, the Agency identified three other NGCC plants and one Integrated
Gasification Combined Cycle (IGCC) plant that were under construction and would likely fit the
rulemaking’s definition of “existing” unit. These were the Dominion Brunswick plant in
Virginia (1,358 MW), the Cheyenne Generating Station in Wyoming (220 MW), the Cane Run
plant in Kentucky (640 MW), and the Kemper IGCC plant in Mississippi (582 MW).
Table 5: Units under Construction Shows Double Counting of Existing Units
NEEDS NGCC UNDER CONSTRUCTION
State Name
California
California
California
California
Colorado
Florida
Mississippi
North Carolina
Ohio
Ohio
Ohio
Virginia
Virginia
Virginia
Virginia
Plant Name
WEC_CALN_CA_Combined Cycle
WEC_LADW_CA_Combined Cycle
WECC_IID_CA_Combined Cycle
WECC_SF_CA_Combined Cycle
WECC_CO_CO_Combined Cycle
FRCC_FL_Combined Cycle
S_SOU_MS_Combined Cycle
S_VACA_NC_Combined Cycle
Dresden Energy Facility
Dresden Energy Facility
Dresden Energy Facility
CPV Warren, LLC
CPV Warren, LLC
CPV Warren, LLC
CPV Warren, LLC
UniqueID_Final
83770_C_1
83778_C_1
83802_C_1
83835_C_1
83792_C_1
83609_C_1
83743_C_1
83745_C_1
55350_G_1
55350_G_2
55350_G_3
55939_G_CT01
55939_G_CT02
55939_G_ST01
55939_G_ST02
PlantType
Capacity (MW)
On Line Year
Combined Cycle
441.2
2015
Combined Cycle
560
2015
Combined Cycle
94
2015
Combined Cycle
760
2015
Combined Cycle
200
2015
Combined Cycle
1157
2015
Combined Cycle
150
2015
Combined Cycle
2249
2015
Combined Cycle
158
2013
Combined Cycle
158
2013
Combined Cycle
223
2013
Combined Cycle
180
2015
Combined Cycle
180
2015
Combined Cycle
105
2015
Combined Cycle
105
2015
However, examining this list along with the supplemental units reveals some major errors. First
the Dresden Energy Facility is an existing unit and can be found in EPA’s Plant Level TSD
spreadsheet of plants operating in 2012. Therefore EPA is double counting this facility. Both
the CPV Warren facility in Virginia and Cheyenne facility in Wyoming may have incorrect
capacity levels assigned to them. According to EIA Form-860, CPV Warren has an expected
nameplate capacity of 1,329 MW compared to the NEEDS capacity of 572 MW. The Agency
also reports the combined-cycle capacity at Cheyenne to be 220 MW. However, EIA reports the
facility is to have only 100 MW of combined cycle capacity and four gas turbines that total 120
98
MW.241 Also, in Florida, the new construction capacity is incorrect and should include Cape
Canaveral (1,210 MW) and Riviera NGCC (1,212 MW), which was expected to enter operation
in June 2014. In addition, the fact that many of the units identified as “under construction” in
California, Colorado, Florida, Mississippi, and North Carolina cannot be identified by name
suggests that development of those units is uncertain, and that it would be speculative for EPA to
rely on those units becoming available for redispatch.
According to NEEDS, North Carolina has 2,249 MW under construction and EPA used this
value in its building block 2. However, in evaluating EIA Form-860 data for the state, the only
facility that could be found that either was under construction or entered operation after 2012
was the L.V. Sutton NGCC (622 MW), which entered operation in November 2013. This leaves
1,627 MW of unidentifiable generating capacity “under construction” that the Agency has
included in calculating North Carolina’s state goals. In the alternative, this excess “under
construction” capacity may be the result of EPA double counting the existing Dan River and Lee
Combined Cycle plants (nameplate capacity of 1,759 MW or summer net capacity of 1,540 MW)
in its building block 2.
If the Florida and North Carolina goal computations are corrected to remove the errors discussed
above, specifically removing Cape Canaveral and Orlando Cogen from Florida’s existing
capacity, and removing 1,627 MW of unknown capacity from North Carolina’s capacity under
construction, both states’ 2030 final goals and 2020-2029 interim goals would change. EPA
should make under construction/operating status error corrections for all states to avoid
double counting generation.
4.
EPA’s Assumption That Each State’s Entire Fleet of Existing
NGCC Units Can Match the Operational Level of Its Top 10
Percent of Units Is Unsupported and Should Be Corrected.
EPA did not undertake any assessment of the differences between high- and low-capacity factor
NGCC units that may have led a small subset of those units to operate above a 70 percent
capacity factor in 2012. EPA acknowledged that units operating above 70 percent on an annual
basis were “largely dispatched to provide baseload power,” and that units operating above 70
percent on a seasonal basis typically “were idled or operated at lower capacity factors” during
periods of lower demand. But the Agency did not examine whether the NGCC units providing
baseload power have different characteristics from the other existing NGCC units that are
expected to provide generation for redispatch, or whether units that were idled during periods of
241
Energy Information Administration, Electric Power Monthly, February 2014.
99
relatively low demand did so because of economic, technical, or regulatory constraints on their
operations. Instead, EPA assumed that all NGCC units are identical. This is plainly
unreasonable.
APPA agrees with UARG that in order to demonstrate that a standard is achievable, EPA must
“establish that the test data relied on by the agency are representative of potential industry-wide
performance, given the range of variables that affect the achievability of the standard.”242 This
determination cannot be based on “mere speculation or conjecture.”243 EPA has failed to
establish that the 10 percent of existing NGCC units operating at 70 percent capacity factor or
higher are representative of the remainder of NGCC units in the source category. Indeed, many
of these high-utilization units are likely not representative of the source category, given that EPA
excluded a number of them from its calculations of each state’s existing NGCC capacity in the
Goal Computation TSD.
In reality, many existing NGCC units face constraints that will prevent them from increasing
their utilization to a 70 percent capacity factor. Some units may be located in areas that are
designated as in non-attainment for a NAAQS, and as a result would likely have operating
permits imposing mass limits on CO2 or NOx emissions that would effectively establish a cap on
those units’ operations. Other units were financed, designed, and maintained for the specific
purpose of operating in cycling duty rather than as baseload. Many of these units would not be
able to achieve the target utilization rate without significant upgrades and testing to ensure that
they are technically capable of operating near full load on a continuous annual basis. In addition,
their permitted emission limits may not allow them to operate at a 70 percent capacity factor.
APPA agrees with UARG that EPA does not even acknowledge, let alone adequately address,
the constraints preventing existing EGUs from operating at a 70 percent or higher capacity
factor. The Agency completely ignores potential permit limits on NGCC unit operation and
dismisses infrastructure concerns. EPA’s response proposing an allowance for “emission rate
averaging across multiple units” within a state in the proposed emission guidelines does not
suffice as demonstration that a 70 percent overall capacity factor is achievable.244
EPA wrongly relies on trends in hourly capacity factors to claim that a 70 percent capacity factor
is an achievable goal. According to the Agency, the nationwide NGCC capacity factor during
peak hours of the day averages 11 percentage points higher than the overall annual average,
242
Sierra Club v. EPA, 657 F.2d 298, 377 (D.C. Cir. 1981)
Lignite Energy Council v. EPA, 198 F.3d 930, 934 (D.C. Cir. 1999) (per curiam).
244
GHG Abatement Measures TSD at 3-15.
243
100
suggesting that the current system is able to support national average capacity factors “in the mid
to high 50s for NGCC for peak.” EPA does not explain why it believes it is reasonable to expect
that the current system can accommodate an additional 10-15 percentage points in order to reach
a 70 percent average capacity factor over all hours of the day.
When considering BSER for building block 2, EPA did not adequately consider the need for
significant infrastructure improvements, such as transmission lines and natural gas pipelines.
Both of these forms of infrastructure require many layers of permitting and years of regulatory
approvals, often at federal, state, and local levels. The Proposed Rule is unrealistic in assuming
that this infrastructure will be in place by 2020 for states to begin implementing all four building
blocks.
A large number of permits, consultations, and approvals are needed from multiple government
bodies to get a new transmission line permitted and constructed. The timeline for a transmission
project depends on real estate availability (negotiating rights-of-way or exercising eminent
domain authority), environmental permitting requirements, public opposition, and regulatory
approval. A relatively simple project that will not traverse an environmentally sensitive area,
require the exercise of eminent domain, or involve significant public opposition can take up to
three years prior to construction. More complicated projects that will traverse federal lands,
environmentally sensitive areas, or will generate public opposition may take as much as 10 years
to complete.
Among the many permits that may be required for a new transmission line or natural gas pipeline
are the following: an Environmental Assessment or Environmental Impact Statement (required
under both federal and state law, in some cases), if the project involves significant state or
federal government action of any kind; a Section 404 permit from the Army Corps of Engineers
if dredge or fill material will be placed in “waters of the United States;” Section 7 consultation
with the U.S. Fish and Wildlife Service under the ESA if the project has the potential to impact
threatened or endangered species; a Special Use Authorization under the National Forest
Management Act if the project will traverse federal lands managed by the U.S. Forest Service; a
right-of-way grant under the Federal Land Policy and Management Act from the U.S.
Department of Interior Bureau of Land Management (BLM) if the project traverses federal lands
managed by BLM; a state water quality permit (if required by a state water quality statute); fish,
game, and other wildlife related permits, if the project will divert natural flow of water bodies or
otherwise affect fish and game; Section 106 National Historic Preservation Act consultation if
101
the project might impact cultural or historic resources; a right-of-way lease agreement; and an air
quality permit if disturbed acreage will exceed certain thresholds. 245
EPA has failed to demonstrate that its building block 2 target of redispatching generation from
existing coal- and O/G-fired steam EGUs to existing NGCC units up to an overall NGCC
capacity factor of 70 percent is achievable. It also has failed to assess whether the subset of
NGCC units currently operating at 70 percent capacity factor represents the remainder of
existing NGCC units, and did not properly address economic, technical, regulatory, or
infrastructure constraints preventing some units from operating at the target level. EPA should
correct its state goals to reflect these facts.
5.
EPA Unreasonably Applied the Building Blocks to Non-Affected
Subpart KKKK Units.
EPA’s goal calculation methodology is defective because it applies the measures identified as
BSER to sources that do not qualify as “affected EGUs” for the purposes of the Proposed Rule.
In other words, the proposed state goals assume that states will regulate sources that they are
prohibited from regulating under Section 111(d). This error affects state goals because it
overstates the number of sources that are available for implementing the BSER building
blocks—particularly building block 2, which is based on shifting generation from higher- to
lower-emitting affected EGUs.
Under section 111(d), state plans may establish standards of performance only for “any existing
source…to which a standard of performance under this section would apply if such existing
source were a new source.” 246 In this case, the Proposed Rule may be used only to establish
standards of performance for existing EGUs that otherwise meet the eligibility criteria for EPA’s
proposed NSPS for GHG emissions from new EGUs. 247 Therefore, EPA’s Proposed Rule may
deal only with regulation of existing Subpart KKKK and/or Subpart GG stationary combustion
turbines that meet these same criteria.
APPA agrees with UARG that EPA disregarded these applicability criteria and applied the
BSER building blocks to include ineligible Subpart KKKK units when determining each state’s
obligations. In particular, the Agency made no effort to exclude from the Proposed Rule NGCC
245
California Public Utilities Commission, Federal, State, and Local Permitting Processes Likely to be Required for
Electric Transmission Projects (June 2009), http://www.cpuc.ca.gov/NR/ rdonlyres/D896C1EA-BD35-4BC8-83C832BAE959BF/0/GenericTransmissionLinePermit.pdf
246
CAA § 111(d)(1)(A)(ii).
247
See 79 Fed. Reg. 1430.
102
units that were not “constructed for the purpose of supplying, and suppl[y], one-third or more of
[their] potential electric output and more than 219,000 MWh net-electrical output to a utility
distribution system.”248 Instead of using available data to determine which units met this onethird sales exclusion, EPA pooled excluded units together with affected EGUs and applied the
building blocks to them. Data available in the docket suggest that a substantial number of
NGCC units used in the goal calculation would qualify for this exclusion, although it is currently
difficult to develop a precise estimate of the number of units that should be excluded because the
one-third sales exclusion is based on a three-year rolling average and the docket contains only
NGCC generation data for 2012. For example, of the 464 plants that EPA examined to
determine what capacity factor is achievable for existing NGCC units, 162 had an annual plantlevel capacity factor in 2012 that was less than 33 percent, with some plants operating as low as
zero percent.249 The Agency should examine historical data for existing EGUs in order to
determine which units would be exempt from regulation under the one-third sales
exclusion before calculating each state’s goal.
Because states cannot impose standards of performance on these units, the additional burden
associated with these units will ultimately fall on affected EGUs that do meet EPA’s
applicability criteria. This is particularly important in implementing building block 2, where
EPA’s purported “system of emission reduction” requires transferring energy generation from
coal- and O/G-fired steam EGUs to NGCC units with available generating capacity. Including
NGCC units that meet the one-third sales exclusion in the goal calculation artificially
inflates the amount of NGCC generating capacity that is available for redispatch, which
thus inflates the amount by which coal-fired units must reduce generating under building
block 2. Once this inflated redispatch is incorporated into a state’s goal, affected NGCC units
will be forced to operate at capacity factors significantly above 70 percent to accommodate the
expected generation that states cannot require from non-affected EGUs. This is plainly arbitrary
and capricious.
The proposed building block 2 state goals are fundamentally flawed due to EPA’s failure to
apply its definition of BSER only to the source category subject to regulation. Because this
defect pervades EPA’s entire methodology for calculating state emission goals, EPA should
recalculate and propose corrected building block numbers.
248
249
79 Fed. Reg. at 1511, Proposed 40 C.F.R. § 60.5509(a).
2012 NGCC Plant Capacity Factor (“2012 NGCC Spreadsheet”) Doc. No. EPA-HQ-OAR-2013-0602-0250.
103
6.
EPA Correctly Excluded Natural Gas Conversion and Co-Firing
from BSER.
In the Proposed Rule, EPA states that it “does not propose to consider [natural gas conversion or
co-firing at coal-fired utility boilers] part of the best system of emission reduction adequately
demonstrated for existing EGUs.” 250 This conclusion is due largely to the high costs of
implementing such a conversion. However, the Proposal “solicit[s] comment on whether natural
gas co-firing or conversion should be part of the BSER.” 251
APPA agrees that natural gas conversion and co-firing are not BSER for coal-fired EGUs and
should not be included in the Proposed Rule. Although sources should have the option to
voluntarily use these measures to comply with CO 2 emission standards, natural gas conversion
and co-firing are extremely costly and are appropriate only for certain EGUs based on sitespecific factors. APPA agrees with UARG that natural gas conversion and co-firing are too
expensive to include as BSER.
C.
Building Block 3 - Renewable and Other Non-CO2 Emitting Generation
1.
EPA’s Approach on Building Block 3 Fails to Take into Account
the States’ Historical Renewable Generation Mix and How an
Individual State’s Source Mix Compares to the Other States in
EPA’s Designated Regions.
Public power has a long and proud history of supporting, developing, and deploying renewable
energy resources and distributed generation (DG). While state renewable portfolio standards
(RPS) generally do not apply to public power out of respect for their local governance, public
power utilities typically invest in as much, if not more, renewable generation as utilities subject
to the RPS in response to customer and community values. Many public power utilities are
recognized leaders in both more traditional, as well as more innovative renewable generation.
DG sources, such as rooftop and community solar, have an important role to play in the
country’s energy mix. However, DG must be implemented in a safe, reliable, and cost-effective
manner. Although public power has been a proponent of many DG initiatives, APPA believes
DG installations are inherently situational and need to be modeled at the local distribution system
level. Because benefits and costs to the distribution system vary greatly with penetration of DG,
such as photovoltaic generation (PV), cost-based principles are critical. Public power utilities’
250
251
79 Fed. Reg. 34,875
79 Fed. Reg. 34,876.
104
efforts to cost-effectively integrate PV and other resources to benefit the grid, and all utility
customers, can also help ensure the long-term viability of the electric grid that our economy and
society depend upon.
Rates paid to DG customers should be consistent with conventional methods used by utilities to
value energy resources and compensate comparable utility-scale resources. Under that
framework, utility-sponsored community solar is likely to yield greater net benefits than rooftop
solar. For example, in some PV deployment models, especially those that seek to monetize
wealth transfers from taxpayers and other utility customers, the value to some individual
consumers may be greater, but inequities will be introduced and overall costs to the community
will be higher.
In developing DG resources, the role of government is not to pick winners and losers, but to
develop decisional tools and trusted data. To avoid this consequence arising from its rule, EPA
needs to provide improved policy resources, and develop objective case study data on crossresource optimization. This type of optimization promises to allow communities to make
objective and informed tradeoffs resulting in DG projects that are most beneficial to the
community.
Utilities have decades of experience safely and reliably integrating technologies, such as PV, into
the electric system. For more than 34 years, APPA’s Demonstration of Energy & Efficiency
Developments (DEED) program has funded advanced research on DG, including many PVrelated projects. At the pace of community and customer desire, public power utilities install
everything from wind turbines to smart grid technologies. Locally owned and controlled utilities
know how to facilitate new, more efficient, and cleaner ways of meeting customer demand. In
fact, supporting innovation is necessary for electric supply diversification, which helps the utility
minimize sharp cost increases, or supply disruptions, that can come from overreliance on one
generating technology. EPA’s Proposal oversteps all of these community-based initiatives in
favor of a brute force modeling approach.
In the Proposed Rule, the Agency makes a projection regarding additional renewables that can be
added to the electric system to decrease future utilization of fossil-fuel fired generation. APPA
agrees that this approach is more workable, cost-efficient, realistic and practical than the
alternative of subtracting from baseline generation as raised in the NODA. However, there are
still a number of assumptions embodied in this method that ignore the on-the-ground realities
faced by many operators, based on the principal argument of “because someone else did it, you
can.” To the contrary, each utility has a unique situation that must be assessed before
determining to what degree renewable energy can be deployed.
105
EPA’s approach presents numerous fatal flaws and biases that significantly overestimate the
amount of renewable generation that is feasible for each state. For example, the assumptions and
methodology employed by the Agency to predict renewable generation levels in building block 3
result in a 2029 estimate that ranges between 36 and 47 percent higher than the EIA’s AEO
2013, or the IPM model results. This is a significant overestimation by EPA and should be
reflected in its average annual growth assumption in renewable generation, which is projected at
7.1 percent between 2020 and 2029, as compared with both AEO and IPM’s projections of a
more modest 1.5 percent and 1.1 percent for the same time period as can be seen in the table
below.
Table 6: Comparison of National Renewable Generation Forecasts (GWh)
2012
EPA Option 1 RE Goals
AEO 2013
IPM - State Option 1
217,868
218,333
218,333
2018
2020
2025
241,924 281,295 407,197
308,994 336,126 367,416
305,359 322,657 345,943
Source: APPA
2029
522,723
385,433
356,063
Annual
Growth
2020 to
2029
7.1%
1.5%
1.1%
Comparing some EPA specific state estimates with a comparable regional estimate from AEO
2013 shows the same overestimation in state renewable generation. For example, the projected
2029 renewable generation levels in the Proposal’s state goal calculations for Alabama and
Georgia are almost 42 percent higher than AEO 2013 levels, even though SERC Reliability
Corporation—Southeastern (SERC-SE) includes a larger geographical area. AEO SERC-SE
renewable generation levels are projected to increase only 2 percent annually between 2020 and
2029, whereas the generation levels used in building block 3 for EPA’s state goal computation
are designed to increase between 2020 and 2029 by an astronomical 11.4 percent per year.
EPA overestimates renewable generation by using a broad selection of renewable energy targets
for specific regions, without any detailed examination of each state. EPA should provide
corrected calculations based on each state’s situational factors in consultation with the
state.
Some regions’ renewable energy targets were based solely on a single specific state’s RPS. As
can be seen in Figure 12 and Table 13 below, the characteristics of each state’s renewable
generation mix vary significantly.
106
Figure 12: Relative Proportion of Renewables by State Used in EPA's Model for Building
Block 3 by Fuel Type from EIA Data
Figure 13: Relative Proportion of Renewables by State Used in EPA's Model for Building
Block 3 by Fuel Type from EIA Data
107
For example, EPA set the regional renewable generation target for the South Central Region
solely based on the 2020 RPS goal of one state: Kansas. This results in erroneous state targets
for other states in the region because Kansas’s 20 percent RPS target is dominated by a single
fast-growing renewable source (wind) and is applied to states within the region with an entirely
different renewable generation mix and in different RTOs. The table below illustrates how the
Agency misused the Kansas RPS standard by applying it to a completely different state like
Arkansas. The table below also indicates the renewable energy source that comprises the major
share of each state’s total renewable generation in 2012. 252
Table 7: Percentage of Renewable Energy by Type by State for Arkansas and Kansas
State
Percentage
geothermal
0.00%
0.00%
AR
KS
Percentage
Other
Biomass
2.03%
0.81%
Percentage
PV &
TPV
0.00%
0.00%
Percentage
Wind
0.00%
99.19%
Percentage
Wood &
Wood
Derived
97.97%
0.00%
Total %
100.00%
100.00%
Source: APPA
Industrial combined heat and power (CHP) facilities in Arkansas burning wood or wood-derived
fuels comprised over 97 percent of the state’s total renewable generation between 2008 and
2012. Industrial CHP sources in Arkansas experienced 1.6 percent average annual growth
between 2008 and 2012. However, EPA applied the Kansas RPS, in which wind comprised
almost 100 percent of state’s total renewable generation, at a 24.2 percent average annual growth
between 2008 and 2012. These extreme disparities show that the Agency’s division of states into
regions is arbitrary and capricious. EPA should work with states to determine their building
block 3 BSER-related targets individually, and if needed, validate them based on a selection of
one or more states they determine are similar in order to avoid such inappropriate comparisons.
2.
252
EPA’s Application of an RPS from a State with a Rapidly
Increasing Renewable Energy Source to a State in Which Its
Primary Renewable Energy Source Has Remained Almost Flat
Can Result in a Significant Overestimation of Renewable
Generation Capability in the Latter State.
Calculated from EIA Form 923 2012.
108
EPA should take into account different renewable generation mixes within states by looking at
state historic renewable energy (RE) growth and acknowledging the growth limitations for
various types of RE, such as biomass and wood and wood-derived fuels. As pointed out in a
letter to EPA from the Arkansas Attorney General, the state has limited wind potential. Nearly
all of the non-hydro renewable energy that Arkansas consumes originates in nearby states .253
Therefore, EPA’s goal setting methodology related to building block 3 would require states like
Arkansas to invest in renewable energy resources within the state no matter how inefficient and
costly they might be. This is a very expensive proposed method for compliance, and unless
adjusted to reflect state circumstances, does not constitute BSER.
In the Southeast Region, the Agency computed a regional annual growth rate of approximately
13 percent, with a 10 percent renewable target. Based upon these assumptions, Georgia would
be able to achieve its target of 12,230 GWh of renewable generation by 2027 and Alabama
would not achieve its renewable target before 2030. However, both Alabama and Georgia’s
renewable generation is dominated by industrial CHP sources burning wood or wood-derived
products, such as pulp and paper mills (in excess of 90 percent of each states’ total renewable
generation). Between 2008 and 2012, these Industrial CHP sources grew at an annual rate of 2.1
percent annually in Georgia, while in Alabama these same types of facilities declined by 5.1
percent annually. It is highly unlikely these types of industrial CHP facilities will add the levels
of renewable generation predicted by EPA in building block 3.
For Georgia to achieve its state target, other non-hydro sources (e.g., wind, solar) would have to
increase their output at an astounding 44.5 percent per year. It is clear that the lack of any
detailed evaluation of a state’s renewable energy mix has resulted in a significant overestimation
of the renewable generation that is included in each state’s building block 3 projections. EPA
must correct these overestimations for its method to remain valid.
3.
EPA Should Clarify Its Stance on Biomass Fuel.
In building block 3, EPA assumes zero CO2 emissions from all renewable sources. However,
depending on pending regulations, this may not be entirely correct. The 2012 renewable
generation levels used by the Agency as the basis for its future renewable generation inputs to
building block 3 include various forms of biomass. Since EPA has not determined yet whether,
or to what extent, such sources will be considered CO2-neutral for compliance purposes, citing
its not-yet-finalized Biogenic CO2 Accounting Framework, this assumption imposes significant
253
See letter from Arkansas Attorney General Dustin McDaniel to Avi S. Garbow, General Counsel, U.S.
Environmental Protection Agency, August 4, 2014.
109
compliance risk and uncertainty.254 These biomass CO2-emitting fuels include black liquor,
landfill gas, municipal solid waste (MSW), sludge, and wood wastes, and are common across the
country.255 Because there is a pending regulatory outcome, EPA has not properly accounted for
additional CO2 emissions from biomass sources in building block 3. Again because the Agency
has failed to examine each state’s renewable generation mix when calculating requirements, it
should recalculate its targets based on corrected renewable sources figures, including any
determination it makes on biomass.
As noted in Figure 12, biomass and wood-derived electric generation is practically the only
option in many states for zero CO2 emission baseload generation. In its proposal, EPA
acknowledged that it could not establish regulations for biogenic emissions until it completed a
Biogenic Emissions Framework (BAF), which it would incorporate. EPA should expedite the
adoption of the BAF and incorporate the findings of biogenic fuel carbon neutrality into its
proposal in a consistent fashion across all its NSPS rulemakings.
The state emission limiting goals proposed by EPA for renewable energy in building block 3 are
extremely optimistic. It should amend its definition of affected facilities to fully account for
CO2-neutral biomass fuel streams in all of itsCO2- related NSPS rulemakings. In doing this the
EPA should make clear that to the greatest extent possible that specific biomass fuels should be
defined as carbon neutral to facilitate the use of this CO2-neutral renewable resource.256 By
designating biomass fuel streams categorically identified as clean cellulosic biomass in EPA's
Non-Hazardous Secondary Material (NHSM) rule as CO2-neutral, the Agency could create more
regulatory certainty on this issue. In addition, unless EPA makes a blanket determination that
state-eligible biomass and biomass included in the building block 3 calculation automatically
qualifies as CO2-neutral, its building block 3 calculations are arbitrary, invalid, and need to be
recalculated. If the Agency adjusts its stance on biomass fuels away from a blanket
determination, states should be able to adjust their targets accordingly.
254
79 Fed. Reg. 34,924-5
See Energy Information Administration, Monthly Energy Review, August 2014.
256
EPA's recently Adopted Non-Hazardous Secondary Materials (NHSM) Regulation (79 FR 21006) establishes a
category of nonhazardous secondary materials that are considered fuels. The biogenic fuels identified in the NHSM
regulation were considered carbon neutral in the agency's draft Biogenic Framework.
255
110
4.
There Are Significant Additional Costs and Constraints Not
Factored into the EPA’s Analysis of Building Block 3.
In 2014, the American Bird Conservancy (ABC) announced it was suing the Interior Department
for finalizing a rule in December 2013 that would allow wind farms to take257 eagles for up to 30
years. ABC asserts the new take rule violates existing federal laws and points to the fact that it
was adopted in the absence of any National Environmental Policy Act (NEPA) document or any
consultation under the Endangered Species Act. ABC says it supports green energy, including
wind power, but the Interior Department’s rule goes too far and allows wind power producers to
ignore basic environmental protections and analysis.
As ABC correctly points out:
As the Supreme Court has explained, ‘NEPA’s core focus [is] on
improving agency decision-making,’ Dep’t of Transp. v. Pub.
Citizen, 541 U.S. 752, 769 n.2 (2004), and specifically ensuring
that agencies take a ‘hard look’ at potential environmental impacts
and environmentally enhancing alternatives ‘as part of the
agency’s process of deciding whether to pursue a particular federal
action.’ Baltimore Gas and Elec. Co. v. Nat. Res. Def. Council,
462 U.S. 87, 100 (1983); see also Robertson v. Methow Valley
Citizens Council, 490 U.S. 332, 349 (1989) (NEPA ‘ensures that
the agency, in reaching its decision, will have available, and will
carefully consider, detailed information concerning significant
environmental impacts’). 258
The Proposed Rule takes no consideration of the need for utilities to conduct full NEPA studies
or the protections that law requires. For example, the Proposed Rule does not consider how
much more difficult it is to build in Wyoming because of protections surrounding the sage
grouse. The impact of the Migratory Bird Treaty Act and American Bald and Gold Eagle
Protection Act should also have been factored into EPA’s assumptions about how much wind
generation could be added to reduce CO2 emissions.
In addition to siting and permitting issues related to wildlife, there are numerous local constraints
on wind development that were ignored in EPA’s building block 3 assumptions. First, there are
257
The Endangered Species Act, 16 U.S.C. §1532 (19) states “The term ‘take’ means to harass, harm, pursue, hunt,
shoot, wound, kill, trap, capture, or collect, or attempt to engage in any such conduct.”
258
http://www.abcbirds.org/PDFs/ABCNoticeFinal.pdf
111
feasibility limitations to locating wind to utilize optimal wind currents. There are also often local
limitations on wind generation. For example, according to the Birmingham News, on August 20,
2014, 259 a wind farm developer dropped plans for two Alabama wind farms after the state
passed a wind energy bill that limited the noise that wind farms can produce. The state law also
requires wind projects to have a larger setback from nearby properties. For a broader discussion
on possible permitting requirements see Section XIV(B4).
There is considerable debate surrounding many of the key elements of renewable energy. In the
case of rooftop PV, for example, the net value of solar, including the basic benefit-cost profile,
the nature and magnitude of subsidies, impacts on electric rates, and degree of cost-shifting
among a utility’s retail customers that comes with any PV installation are all things that need to
be considered. For utilities considering the implications of PV net metering programs and RPS
requirements, it can be difficult to reconcile all of the conflicting claims and counter claims.
When it comes to estimates of costs and benefits for the net value of solar, EPA needs to
objectively illuminate the most critical factors in net value assessments and accurately display
the discrepancies between them, not simply wave off legitimate criticisms by pointing at another
building block. In fact, the Agency’s assumptions of increasing renewables under building block
3 do not allow for an alternative in many cases, especially if building blocks 1 and 2 are not
corrected.
EPA should recognize that utilities (and states) are in the best position to optimize the integration
of renewable DG and maximize the flow of benefits to all customers and resulting increased
costs. Forcing rapid development of renewable generating resources will result in a sub-optimal
power cost curve for consumers. Many attributes of a DG installation affect the ability of those
systems to reduce integration costs or turn them into grid benefits and cost savings. Such
attributes include location, size, and dispatchability. To achieve maximum benefit from DG,
coordinated deployment is critical. Unfortunately, EPA did not asses how building block 3 could
create additional issues and costs for consumers, or how states can make optimal environmental
dispatch decisions that properly reflect cost and power quality concerns.
Many renewable generation projects currently benefit from various types of “societal” subsidies.
These include federal and state tax credits, grants, renewable energy credits (RECs), and local
property tax relief. In addition, ratemaking mechanisms such as net metering can lead to de facto
subsidization in the form of cross-customer cost-shifting. On the other hand, larger scale, utility-
259
Birmingham News, Aug. 20, 2014. http://www.al.com/news/annistongadsden/index.ssf/2014/08/alabama_regs_too_strict_for_tu.html#incart_river
112
owned projects are effectively subsidized by all customers through higher utility rates when
project costs exceed the economic value of the output. Currently, these subsidies are crucial for
the development of renewable energy sources. EPA ignores these subsidies and doesn’t
acknowledge in its Proposal how rapid and disproportionately large expenditures on renewable
projects might increase rates and significantly decrease the marginal value of such projects to the
community.
Societal subsidies are generally straightforward, simply offsetting certain costs incurred by those
who receive them. However, the subsidization that results from cross-customer cost shifting and
higher electric rates deserves more explanation. Cost-shifting issues are particularly pronounced
with net-metered projects. The subsidization arises because the generator output displaces utility
production and sales. For example, when the output of the consumer-owned PV system is less
than or equal to the customer’s usage, total utility sales are reduced and total utility revenue
declines by an amount equal to the volumetric ($/kWh or $/kW) rate under which the customer is
taking service, times the volume of solar output. The utility’s total cost will decline, by an
amount equal to the project’s output times the unitized ($/kWh or $/kW) marginal costs that are
avoided as a result of the solar production, which is likely to be less than the amount the
customer avoids paying. So, revenue and cost both decline, but whenever the volumetric electric
rate exceeds the unitized avoided cost, the utility will face a net revenue loss unless it makes up
the shortfall by raising rates and shifting costs to its non-PV customers.
The value of the grid has been recognized by numerous stakeholders and EPA pays it no mind in
its analysis of what is “best” in terms of CO2 emissions. Documents the Agency should review
include:




EPRI Cost of Grid Service: Energy and Capacity Costs (see page 21);260
EEI, The Value of the Grid;261
NARUC resolution262 encouraging state commissions and policymakers to continue to
engage in collaborative dialogue regarding DG policies and regulations;
MIT, The Future of the Electric Grid 263
The grid provides essential services to all customers, including renewable energy customers. All
customers connected to a utility’s distribution system rely on it continually. The distribution
260
http://www.epri.com/abstracts/Pages/ProductAbstract.aspx?ProductId=000000003002002733&Mode=download
http://www.edisonfoundation.net/iei/newsevents/Pages/2013-09-30.aspx
262
http://www.naruc.org/Resolutions/Resolution-Encouraging-State-Commissions-Policymakers-to-Continue-toEngage-in-Collaborative-Dialogue-Regarding-Distributed-Generation-Policies-Regulations1.pdf
263
http://mitei.mit.edu/publications/reports-studies/future-electric-grid
261
113
system provides the services required to manage electrical integrity, including, but not limited to,
frequency and voltage, current, and other aspects of power quality. DG customers use the
distribution infrastructure in more complex ways than other types of customers. DG customers
that remain connected to the grid use grid services to:




Balance supply and demand;
Maintain stable voltage and frequency and high AC wave-form quality;
Resell energy during hours of excess onsite power production; and
Obtain backup service when on-site generation is unavailable.
Since DG customers use the grid in more complex ways than other customers, they are
contributing to additional costs that utilities will incur to transform their distribution systems
from the one-way delivery of power into a far more complex two-way system. These costs can
include investments in new control systems, including IT infrastructure, communications
hardware and software, and protective equipment. They can also include operating procedures
and training to ensure the safety of the workforce and the public. EPA should correct its
building block 3 estimates to reflect the increased costs of integration that come with
increased penetration of renewables and increasing heat rates as more marginal generators
are forced to follow intermittent renewables.
5.
In Building Block 3, EPA Has Erred by Including Nuclear
Capacity in Its State Goals.
EPA has indicated that the two best ways to increase/maintain the amount of nuclear capacity in
the nation’s power supply mix are either to build new nuclear plants or preserve existing nuclear
plants that would otherwise be retired. Because EPA uses additional nuclear generation as a
factor in the denominator of its goal computation formula under building block 3, it is effectively
tightening standards on nuclear states. This penalty decreases the overall calculated target rate
(CO2 lbs/MWh) used to determine the state emission goals, reducing flexibility. Using EPA’s
calculation methodology, a state without existing nuclear capacity will benefit from EPA’s
building block 3 relative to other states, in that its denominator will not include nuclear
generation, resulting in a higher state goal. Therefore, states that have nuclear generating units
with zero CO2 emissions or that are currently developing nuclear units are effectively penalized
and lose any flexibility as to whether to complete or continue to operate those units.
For existing nuclear capacity, EPA assigns a 5.8 percent factor to each state’s 2012 nuclear fleet
to determine the amount of “at risk” nuclear capacity in the state. This “at risk” capacity is then
114
multiplied by a 90 percent capacity factor to yield an “at risk” generation level, which is included
in building block 3. This 5.8 percent figure was computed based upon an AEO 2014 projection,
without evaluating specific nuclear capacity in each state.264 This approach does not in any way
reflect the actual state of existing nuclear capacity in any state, resulting in additional generation
being included in the goal setting equation for some states, if not all. Factors such as license
renewal and expiration dates, expected capital improvements, profitability and other factors
should be considered and an appropriate unit/state specific risk factor be applied.
Possibly the most egregious flaw in EPA’s goal setting methodology pertains to the inclusion of
nuclear capacity under construction. Three states – Georgia, South Carolina and Tennessee -are adversely impacted by EPA’s inclusion of this generation in building block 3. Specifically,
Georgia has an additional 17,345 GWh added into its building block 3 target, while South
Carolina has17,345 GWh added to its target, and Tennessee has 8,846 GWh added to its target.
The utilities constructing these five nuclear facilities likely intend to operate these units, but
various issues beyond their control could affect licensing, start-up, and operations. The inclusion
of under-construction nuclear generation has such a large impact on a state goal that any shortfall
in operations would be very difficult, if not impossible to overcome. For example, including
Georgia’s new nuclear generation in building block 3 lowers the state’s 2030 final goal under
Option 1 by 138 lbs/MWh, from 972 lbs/MWh to 834 lbs/MWh. Consequently, EPA should not
include under-construction nuclear generation in its goal setting methodology for building block
3.
The impact from EPA’s treatment of nuclear under construction and “at risk” in 2030 state goals
is illustrated in the table below for three states.
Table 8: Impact of Nuclear Generation Under Construction and “At Risk” in EPA 2030
State Goals265
2012 Nuclear
Generation Under
Construction And "At
State Risk" (MWh)
10,416,619.07
TN
19,220,561.26
GA
20,340,660.49
SC
EPA 2030
State Goal
(lb/MWh):
1,162.62
833.78
771.75
264
Percentage
CO2 Reduction
From Nuclear
In 2030 Goal
16.46%
15.72%
25.86%
2030 Goal Without Nuclear
Generation Under
Construction And “At Risk”
(lb/MWh):
1,391.71
989.33
1,040.92
See Energy Information Administration, Annual Energy Outlook 2014 and Jeffery Jones and Michael Leff,
Energy Information Administration, Implications of accelerated power plant retirements, April 2014.
265
Source: APPA
115
a.
The Proposed Rule Fails to Properly Address Existing
Nuclear Generation.
Nuclear generation is a valuable non CO2-emitting source of electricity, but is not appropriately
valued and addressed in the Proposed Rule. Rather than allow existing nuclear generation to be
used for compliance with a state’s CO 2 reduction requirement, EPA chose to include existing
nuclear in the goal calculation. Specifically, the Proposal incorporates 5.8 percent of a state’s at risk nuclear capacity in the emission goal. This has the effect of penalizing states with nuclear
generation by making it more difficult for them to comply with their final emission-reduction
goals should that existing nuclear power retire or have to shut down due to unforeseen
circumstances.
The Proposed Rule also treats existing nuclear capacity different than it does other large, existing
sources of zero-CO2 emissions electricity, such as hydropower and renewable energy facilities.
EPA makes assumptions about how much existing nuclear power is at risk due to “continued
economic challenges,”266 yet does not do so for existing hydro and renewable sources that also
face economic and non-economic challenges that could reduce capacity through shutdowns or
retirements. The Proposed Rule provides no rationale for treating existing CO 2-emission free
resources differently. Nor does it acknowledge that states have little control over whether
nuclear capacity is used or preserved. The Nuclear Regulatory Commission (NRC) has sole
authority to determine whether a license extension is granted to an applicant. Why are states
penalized for failing to preserve their existing nuclear resources when they cannot control what
the NRC does?
Any and all proposed EPA rules should support and encourage the continued use of existing
nuclear generation. Support of this zero-emissions resource is essential to maintaining the
nation's clean air generation portfolio. The policy choices EPA made in this rulemaking may
well dictate the fate of many of the nation's nuclear units and consequently the nation's ability to
meet its CO2 reduction objectives. Accordingly, when the operating license of an existing
nuclear generating unit expires, the state goal for the resident nuclear plant should be
subsequently adjusted. EPA should also support the NRC’s current license renewal process to
allow further operating license extension of units with 60-year licenses, to no less than 80
years. Also, it is important that zero CO2-emitting resources be treated equally, regardless of age
or technology, to provide cost-effective CO2 reduction solutions for consumers.
266
79 Fed. Reg. 34830, 34871 (June 18, 2014).
116
Nuclear generation, like all other forms of generation, is subject to equipment failures and other
generation impacting events, including force majeure situations (e.g., fire, floods, etc.). In
addition, nuclear generation output is also limited by periodic refueling outages and regulatory
action than can limit the output of the station. Correspondingly, any regulation that includes
nuclear generation as a zero-emitting source must be designed to accommodate expected year-toyear variations in nuclear energy production.
b.
States with Nuclear Generation Units Currently Under
Construction Are Unfairly Penalized.
The Proposed Rule includes “projected amounts of generation available by completing all
nuclear units under construction…” within building block 3.267 EPA does not define “currently
under construction” or explain why the term is relevant to setting the BSER. It rather identifies
the five nuclear units under construction in Tennessee, Georgia, and South Carolina. The agency
then treats these units as part of BSER because it does not view there is any incremental cost
associated with CO2 emission reductions from completion of these units. No legal or technical
analysis was done to support this conclusion. The result of this erroneous conclusion is to
unfairly and significantly increase the stringency of the goals for these three states for units that
are currently under construction and not scheduled to come on line until well after EPA issues its
final rule.
There is no rational policy basis for EPA to assume that nuclear units under construction will be
built on time, provide CO2 emission reductions at no cost, or that these facilities will operate at a
90 percent capacity factor. As stated by APPA member Santee Cooper in its comments in this
proceeding, “construction of a new nuclear reactor is perhaps the most complex and highly
regulated industrial activity any utility can undertake in the United States.” 268 The NRC
oversees all steps of the construction process and will very likely take even lo nger reviewing the
construction of these new facilities given they are new designs. The Summer, Vogtle, and Watts
Bar units have already experienced delays due to problems with fabrication and the supply of
high-tech equipment.
Furthermore, once these units are constructed, they are required to undergo lengthy testing and
fine-tuning periods before they can run at their designed capacity factors. EPA’s inclusion of
these units into the BSER determination fails to take into account that these facilities are likely to
267
79 Fed. Reg. 34830, 34851 (June 18, 2014).
See page 76 of comments filed by Santee Cooper in docket for Carbon Pollution Emission Guidelines for Existing
Stationary Sources: Electric Utility Generating Units; Proposed Rule; 79 Fed. Reg 34,830 (June 18, 2014); Docket
ID No EPA-HQ-OAT-2013-0602 submitted on December 1, 2014.
268
117
face hurdles during operation that could limit their ability to reduce CO2 emissions on the
schedule the Proposed Rule sets and at the level assumed in the setting of the state goals in South
Carolina, Tennessee, and Georgia. The Proposed Rule provides no rational basis for assuming
that these under construction units, once operational, could achieve a life-time capacity factor of
90 percent. It can be difficult for nuclear units to achieve a high capacity factor given unforeseen
and foreseen shutdowns for refueling, maintenance, and safety considerations. If one of these
new units fails to run at a 90 percent capacity factor, it could result in the state failing to meet its
CO2 emission reduction goal.
Another troubling aspect of the Proposed Rule is how it treats nuclear under construction
differently than renewable resources. It assumes that every state that has committed to building
new nuclear capacity will do so, but it does not assume that states that have committed to
renewable energy policies through state renewable portfolio standards (RPS) will construct all
the RE capacity that is called for under such RPS. The agency provided no rationale for why the
stringency of the goals for South Carolina, Tennessee, and Georgia is higher than for states with
high RE commitments. The inconsistency in treatment is concerning given RPS are mandated at
the state level, but the five under construction projects in three states are not mandated through
any state regulatory requirement. Nor are renewables under construction included in state CO2
emission reduction goals. EPA appears to be favoring renewable energy sources over nuclear
sources, both of which are zero CO2 emitting sources of electricity, with no rationale given for
the disparate treatment.
APPA supports the more detailed joint comments regarding EPA’s Treatment of “Under
Construction Nuclear Units” filed by APPA members Santee Cooper and Municipal Electric
Authority of Georgia and Dalton Utilities, Georgia Power Company, Oglethorpe Power
Corporation, South Carolina Electric and Gas Company, and Tennessee Valley Authority in this
rulemaking,269 as well as the comments filed on September 16, 2014, by the Georgia Department
of Natural Resources.270 EPA should remove these under-construction nuclear units from the
calculation of BSER and those states’ emission goals. In addition, EPA should clearly state that
these units are allowed for compliance under those states’ plans.
269
Joint comments regarding EPA’s Treatment of “Under Construction Nuclear Units” in Carbon Pollution
Emission Guidelines for Existing Stationary Sources: Electric Utility Generating Units; Proposed Rule; 79 Fed. Reg
34,830 (June 18, 2014); Docket ID No EPA-HQ-OAT-2013-0602 submitted by Dalton Utilities, Georgia Power
Company, Municipal Electric Authority of Georgia, Oglethorpe Power Corporation, Santee Cooper, South Carolina
Electric and Gas Company, Southern Company, Southern Nuclear Operating Company, and Tennessee Valley
Authority on November 17, 2014.
270
Letter from Georgia Department of Natural Resources to EPA on the treatment of under construction nuclear
submitted to EPA Docket ID No. EPA-HQ-OAR-2013-0602 on September 16, 2014, available at
www.regulations.gov.
118
6.
To Determine Lowest Cost BSER on a State-by-State Basis, EPA
Should Modify Its Determination of BSER to Include Additional
Time and Consideration of Relevant Costs.
EPA’s proposed renewable energy generation goals are drastically higher than most states can
achieve within EPA’s proposed timeline. A balanced assessment of renewable generation, as a
supply resource that is required to determine the lowest cost BSER, requires an examination of
all the components that are involved in producing, distributing, and consuming electricity. These
components are intertwined and causally linked. No one element can be meaningfully addressed
without touching on the others. In addition, since value-of-service propositions are inherently
uncertain, subjective, and speculative, objectively analyzing the various cost a nd benefit
components is all the more difficult.
For example, there are numerous ways to think about PV project economics. One common
approach is to derive a benefit-cost ratio with the net present value (NPV) of project benefits in
the numerator and the NPV of project costs in the denominator. Using this equation, a benefitcost ratio greater than one indicates the project is economically beneficial because the ratio can
only be greater than one if project benefits exceed the costs. A ratio less than one means that
costs exceed benefits and the project is not economically beneficial. EPA has failed to provide a
methodology that would give states the flexibility needed to do this in the time frame required.
Although the benefit-cost conceptual framework is simple, there is no single, standard modeling
approach that would be accepted by all for this purpose. The results and conclusions can differ
depending on how the analysis is conducted. There are at least three key elements of the
modeling that will affect the results: (1) the structure of the benefit-cost equation in terms of the
variables included and how the terms are arranged; (2) the values assigned to the variables; and
(3) the perspective from which the analysis is conducted.
The following illustrates some key costs that are borne by the utility, but often debated during
cost-benefit analysis. EPA should allow these states to consider these as inputs to their cost
models.


DG does not avoid most utility fixed costs of service. The greatest cost savings from
distributed PV resources are avoided variable fuel costs (provided that distributed PV
generation displaces fossil fuel generation and not other renewables) or electricity
otherwise purchased in wholesale markets.
DG can result in little or no distribution cost savings because many DG systems are not
designed to improve the safety or reliability of the grid, and generally do not provide
backup power in the case of an outage. However, one of the reasons PV is more
119


appealing than other variable resources is the fact that it has a more consistent output
with better coincidence to peak load times.
Investments in the transmission and distribution system might never be completely
avoided, just deferred. To the extent that DG actually produces fixed cost savings,
measuring the part that the utility may avoid or defer is challenging.
Even when the DG system fails to produce energy, the utility remains obligated to
provide uninterrupted service (generation, transmission, and distribution) to the DG
customer.
Some utilities design their rates to accept tradeoffs for policy goals, but overall , their rates must
recover the total cost of service, be defensible, and be equitable to all classes of customers.
Customers pay for the costs of the generation, transmission, and distribution services they
receive from their utility. Customers should not pay for the costs of services provided to other
customers. Payments or credits to customers with PV should be based on directly measurable
cost savings that favorably affect the utilities’ overall cost of service to a customer. Treating all
customers equitably prevents utilities from basing ratemaking on subjective valuations of the
services provided.
Utilities should have an active role in any DG cost-benefit or valuation study. They play an
indispensable part in the integration and interconnection of PV. Utilities also have critical
information on the cost to serve all customers and on the renewable energy project’s contribution
toward that cost. EPA appears to recognize the need for more time to implement its proposal in
its NODA. EPA should absolutely add more time for states and any sources subject to
regulation to provide recommendations for cost-optimal renewable generation deployment.
7.
The State Renewable Energy Generation Targets Are
Unreasonably Aggressive and Do Not Take into Account Factors
Affecting the Actual Renewable Energy Growth Potential in Each
State.
EPA is proposing to find that it is achievable for states to increase in-state generation from
renewable energy sources to an amount that is based on application of a regional annual growth
factor to the state’s 2012 renewable energy generation, gradually approaching a maximum
regional renewable energy generation target.271 The Agency expects states to begin taking
measures to increase renewable energy generation in 2017, three years before the beginning of
271
79 Fed. Reg. at 34,867.
120
the compliance period.272 The regional maximum renewable energy target for each state is based
on the average 2020 primary RPS goal of states within each region that have RPS goals.273
The renewable energy generation goals that EPA has calculated for many states are
unachievable. EPA’s methodology for determining these goals is fatally flawed because it
arbitrarily ties each state’s goal to the RPS goals of other nearby states, which often do not
actually require as much renewable energy generation as they appear to and reflect each state’s
highly unique mix of renewable energy potential. EPA’s failure to account for factors specific to
each state’s potential renewable energy generation leads to several inequitable and unreasonable
results, as described in UARG’s comments.
EPA’s analysis reflects an oversimplified understanding of the nature of state RPS goals. Some
state RPS goals are voluntary, while others only establish requirements for large electric utilities
or apply less stringent “secondary” or “tertiary” RPS goals to smaller utilities, public power
utilities, or cooperative utilities.274 But the Proposed Rule applies the more stringent primary
RPS goals to all in-state generation and treats all goals as mandatory, artificially leading to more
aggressive renewable energy generation targets. In addition, some state RPS goals allow utilities
to comply through mechanisms other than actually developing in-state renewable energy
generation.
For example, North Carolina’s renewable energy and Energy Efficiency Portfolio Standard
(EEPS) goal allows utilities to meet 25 percent of their requirements by reducing energy
consumption through energy efficiency measures, and another 25 percent of their requirements
by purchasing renewable energy certificates (RECs) from out-of-state facilities. Thus, the 10
percent RPS target EPA assumed for North Carolina effectively requires in-state renewable
energy generation to reach only 5 percent by 2020 with remaining 5 percent achievable through
energy efficiency and RECs. Because North Carolina’s RPS alone sets the regional target for the
entire Southeast region, EPA’s target is roughly 100 percent too high for eight states. Other
states, along with North Carolina, have built cost-control mechanisms into their RPS goals to
reduce the requirements for utilities if compliance becomes too burdensome. EPA did not
account for any of these flexible compliance mechanisms when dictating regional renewable
energy generation targets. It should do so.
272
Id.
Id.
274
See GHG Abatement Measures TSD 4-10.
273
121
APPA agrees with UARG that EPA’s methodology is flawed because it requires states to begin
increasing renewable energy generation by 2017, three years before the compliance period
begins for the Proposed Rule. To put this timeline in perspective, initial state plan submissions
are not due to EPA until June 30, 2016, and EPA is allowing itself until June 30, 2017, to take
final action approving or disapproving these plans.275 Thus, EPA’s Proposed Rule would require
states to begin implementing their plans before they have even received plan approval. This
disconnect is even greater in light of the fact that states may delay plan submission until June 30,
2017 (for various reasons), or June 30, 2018 (for multi-state plans), pushing final EPA approval
of these plans to mid-2018 or -2019. In addition, building new renewable energy generation
capacity can often take several years to allow for planning, obtaining funding, seeking regulatory
approval, and construction. Increasing renewable energy capacity will, in many cases, also
require new transmission infrastructure, which can take 8 to10 years to complete. In effect,
states need to have begun efforts to implement building block 3 several years ago. This is
plainly unreasonable.
EPA cannot establish that increases of this magnitude are achievable—particularly because it has
failed to provide any parsed analysis of its IPM runs for 2030 that would allow states to assess
the impact of these required increases. Indeed, EPA has provided data for only four of the 25
IPM runs that it performed as support for the Proposed Rule. The Agency must provide the basis
and purpose for the Proposed Rule under section 307(d)(3) of the CAA. The basis and purpose
must include “a summary of . . . the factual data on which the proposed rule is based; the
methodology used in obtaining the data and in analyzing the data; and the major legal
interpretations and policy considerations underlying the proposed rule.” EPA’s failure to include
the data for 21 of its modeling runs has affected APPA and its members’ ability to comment
meaningfully on the Proposed Rule.
The Agency’s own analysis demonstrates that it has failed to accurately consider factors such as
cost and feasibility. The IPM results suggest that rapid growth in renewable energy generation is
so costly that states are able to achieve only negligible incremental increases in renewable energy
generation above the status quo (at best), and they must instead rely more heavily on other
components of EPA’s selected BSER in order to comply with the Proposed Rule. This, in turn,
will increase the cost of those other building blocks in ways that EPA has failed to analyze in the
Proposed Rule. Therefore, EPA must recalculate its renewable energy generation targets to
consider factors such as cost and feasibility for each state. A failure to do so would
necessitate the Proposed Rule’s withdrawal and re-proposal.
275
79 Fed. Reg. 34,915-16.
122
8.
APPA Agrees with EPA’s Assessment that Hydro Power Is Not a
Universal Resource and Should Be Excluded from EPA's Method
for Quantifying Renewable Energy Generation Potential.
The EPA states that “[h]ydropower generation is excluded … because building the methodology
from a baseline that includes large amounts of existing hydropower generation could distort
regional targets that are later applied to states lacking that existing hydropower capacity .”276
APPA agrees that constructing new hydropower is not feasible in all states. APPA also agrees
that new hydropower, where it can be constructed, should count towards compliance with any
calculated target or goal.
Hydropower output is subject to highly variable and uncontrollable natural factors, such as
annual rainfall and river run requirements, that would make use of MWh produced from hydro
facilities in EPA’s methodology for setting state goals inherently flawed. While APPA
understands there is variation in many renewable resources, the significantly higher degree of
regional variation in average annual rainfall between states makes hydropower unsuitable for use
in EPA’s goal-setting methodology. The difference in average annual variation is significant.
For example, annual average rainfall variation can be different by a factor of more than 10
between the states with less rain and the states with more rain. 277 By contrast, wind speed at 30
meters only varies by a factor of 2.5 across the U.S. 278 APPA agrees with EPA that the
availability of hydropower should not be used in the calculation of any state’s renewable
generation requirement.
APPA does not want hydro power to be “invisible” as states make compliance choices. If a state
deems hydropower generation either “essential” or “at risk” because it is nearing the end of its
design life, facing potential limits on operational flexibility, or scheduled for relicensing before
2030, EPA should allow that state to add that hydro generation in units of estimated MWh into
the denominator of its goal/target compliance calculation in the place of other RE/EE measures.
Where this is done, states should be able to adjust their hydro contribution yearly based on
updated weather information. At a minimum, the rule should do more to ensure that these
hydropower resources are maintained and enhanced in order to continue their contribution to a
lower CO2 emission future. A bad result would be for the Proposal lead to a low-cost, CO2
emission-free resource going offline due to being replaced by, or being bumped by, higher-cost
CO2 emission-free resources or natural gas due to the incentive structures associated with the
BSER.
276
79 Fed. Reg. 34,867.
http://www.wrcc.dri.edu/pcpn/us_precip.gif
278
http://www.nrel.gov/gis/images/30m_US_Wind.jpg
277
123
Also, in cases where a state RPS includes hydro power, it should be allowed to displace
additional renewable requirements, as calculated under building block 3. For example, the 15
percent goal for renewable energy in South Dakota was based on its total 2012 generation of
12,034,206 MWh, which includes 5,980,965 MWh of hydropower—nearly half of the total state
generation. Because existing hydropower is used in making the calculation of renewable energy
targets, it should also be allowed in state plans as an eligible form of renewable energy under
building block 3.
EPA should expressly allow state plans to designate all forms of new and incremental hydro for
compliance, whether owned or contracted for. This should include clearly establishing both
domestic and out-of-nation hydro (such as Canadian hydro) as an eligible renewable resource.
9.
The Alternative Renewable Energy Approach Is Unworkable.
The Proposed Rule sets forth an “Alternative Renewable Energy” approach to calculating the
renewable energy component to support the BSER. 279 This alternative approach relies on a stateby-state assessment of RE technical and market potential. At first glance, it appears to solve
several issues with EPA’s proposed approach that uses RPS-based regional renewable energy
targets. However, applying this alternative RE approach to EPA’s calculation of state-wide
emission goals has a profound impact on the current proposed goals. This is particularly true for
a state which has already reached its regional RE generation target under EPA’s proposed
approach.
For example, South Dakota is in the North Central region that has an average regional RE
generation target of 15 percent under EPA’s proposed approach. South Dakota has already
reached its regional RE target, with 2,915 GWh of RE generation in 2012, and thus its obligation
under the target is capped at its share of the 15 percent regional RE target, 1,819 GWh of RE
generation. Under the alternative RE approach, South Dakota’s obligation under the target is not
capped. Instead, South Dakota’s state-level 2030 generation target, excluding existing
hydropower, is 19,156 GWh of RE generation. This generation target would be incorporated
into the denominator of the state goal calculation in place of the RE generation levels quantified
using EPA’s proposed approach. Including South Dakota’s generation target based on the
alternative approach causes South Dakota’s final rate-based CO2 emission performance goal to
decrease from 741 lbs CO2/MWh under EPA’s proposed approach to 185 lbs CO2/MWh under
the alternative approach. This would dramatically and unsustainably increase the cost curve for
279
79 Fed. Reg. 34869.
124
compliance. In other words, the percent decline in South Dakota’s state emission rate—a state
with one of the lowest CO2 emissions rates in the nation—that the Proposed Rule requires would
increase from 35 percent to 84 percent if EPA were to adopt the alternative approach for
quantifying RE for BSER. This is an extremely expensive compliance alternative.
EPA’s state goal calculation for South Dakota under both EPA’s proposed approach and the
alternative approach are set forth below.
Final State Goal Calculation280
State Emission Rate =
(coal gen. x coal emission rate) + (OG gen. x OG emission rate) + (NGCC gen. x NGCC
emission rate) +“Other” emissions
Coal gen. + OG gen. + NGCC gen. + “Other” gen. + Nuclear gen. uc + ar + RE gen. + EE gen.
Final Proposed State Goal Rate for South Dakota—Proposed RE Approach
((958,046 x 2,130) + (0 x 0)) + (1,992,211 x 1,131) + 0)
= 741 lb/MWh
(958,046 + 0 + 1,992,211 + 0 + 0 + 1,818,850 + 1,028,768)
Final Proposed State Goal Rate for South Dakota—Alternative RE Approach, Excluding
Existing Hydropower
((958,046 x 2,130) + (0 x 0)) + (1,992,211 x 1,131) + 0) = 185 lb/MWh
(958,046 + 0 + 1,992,211 + 0 + 0 + 19,156,000 + 1,028,768)
According to page 8 of the Alternative RE Approach TSD, states would not be required to
achieve the absolute levels of target generation quantified under the alternative approach and
incorporated into the denominator of the state goals. EPA notes that states may consider
including in their state plans compliance measures that do not rely heavily on expanding their RE
capacity. In practice, however, it is difficult to see how certain states will meet their goals if
EPA chooses the alternative RE approach to be used as part of BSER. The multitude of issues
280
These calculations are based on data from the Alternative RE Approach TSD, page 12, and the Goal Computation
TSD containing a Microsoft® Excel attachment of the aggregate state-level data, calculations, and proposed state
emission rate goals.
125
that surround the expansion of RE cannot be adequately taken into account by the use of the
IPM, as relied upon by EPA, to project potential RE generation expansion and those affected by
this alternative approach cannot provide substantial comments on how they might be impacted
because of that.
As acknowledged on page 2 of the alternative RE approach TSD, there are limitations to
technical potential due to grid costs, development costs, resource quality, and uncertainties of
production potential. The expansion of RE is highly dependent on available transmission. Any
approach to quantify RE potential should take into account transmission constraints, engineering
design constraints, such as hosting capacity, cost constraints, and the additional issues mentioned
elsewhere in these comments, surrounding new transmission construction that come from siting,
permitting, environmental impacts, and landowner opposition. For some states, the amount of
renewable growth that EPA expects may well turn out to be unachievable. States should not be
forced to make up the difference elsewhere in their state plans for compliance.
D.
Building Block 4 – Energy Efficiency
APPA considers energy efficiency to be a vital element of our national energy strategy. As such,
APPA maintains a web-based energy efficiency resource database and supports a research and
development program, called DEED, to develop and accelerate deployment of the most practical
and cost-effective efficiency technologies for our members. DEED has funded the development
and implementation of all levels of energy efficient technology for almost 35 years . At times,
the research conducted through DEED projects has been done in collaboration with DOE and
EPA.
Energy efficiency and demand-side management (EE/DSM) measures effectively reduce
electricity consumption by promoting products or programs that support efficiency or
conservation of electricity. APPA supports EE/DSM measures, as well as the combined efforts
of DOE and EPA to encourage these policies and programs as a means to decrease load and/or
abate emissions. However, APPA does not endorse EPA’s proposal to use energy efficiency as a
factor to determine state goals as it currently stands in the Proposed Rule.
In the Proposal, EPA estimates the potential for energy efficiency measures to be implemented in
each state. In this projection, there are a number of assumptions that ignore situational factors
states and utilities may have to face during the actual application of these programs and
measures. Each utility and/or entity involved in energy efficiency for the state has a unique
situation that must be assessed before determining the real potential for demand-side efficiency
improvements.
1.
The Load Growth Analysis in Building Block 4 Is Insufficient to
Properly Account for Potential Fluctuations.
126
EPA included predictive load growth analysis in building block 4, but failed to properly account
for the potential variation in population and load growth that may occur within a state, city or
locality. Also, EPA does not address the situation that may befall states in the case when a
significant shift occurs unexpectedly.
Severe influxes in population can greatly differ from local, state, and regional perspectives . The
potential for spikes or downturns in population will impact a state or locality’s ability to meet the
standards set by EPA in the Proposed Rule due to the omission of appropriate growth analysis in
the calculations. The Agency did include a regional load growth analysis in building block 4,
demand-side energy efficiency, but assumed that states will be able to offset future growth by
implementing and encouraging energy-efficient technologies. However, an aggregate regional
growth rate does not accurately depict the probable and improbable fluctuations that may occur
on a more granular level, such as the county or state level. EPA should also be aware that load
forecasting is limited in its capability to accurately predict future consumption patterns and often
requires adjustments to account for unexpected changes. Therefore, the final rule should
include a mechanism to allow states that undergo severe shifts in population or load growth
affecting their ability to reach their goals to modify such goals. In the case where a
significant shift in population or load does affect a state in achieving its target, EPA should allow
an alternative form of compliance. APPA suggests a compliance method in the form of a fee
reflecting the cost of energy efficiency measures available in that individual state.
While the Proposal Rule would give states the option of using either a mass- or rate-based goal
in their state plans, many states are likely to use a mass-based approach because it simplifies the
compliance process by requiring less taxing calculations and analysis than a rate -based approach.
However, mass-based targets do not account for load growth, which creates an incentive to use
the rate-based method because the rate remains the same regardless of any fluctuations . To
equalize the two options, EPA should consider population changes and load growth when
setting target emission standards.
2.
Environmentally-Friendly Electric Technologies That May
Contribute to Positive Load Growth
The Proposed Rule does not account for the potential for additional load increases as a result of
increasingly prevalent environmentally-friendly electric technologies. The Electric Power
Research Institute’s (EPRI) 2009 PRISM analysis 281 forecasted that use of electric technologies
281
The Power to Reduce CO2 Emissions, 2009 Technical Report. EPRI.
127
in industrial and commercial applications that displace traditional use of primary energy
consumption, such as heat pumps, water heaters, ovens, induction melting, etc., have a potential
to reduce CO2 emissions by 6.5 percent and, according to Table 9, could replace approximately
4.5 percent of direct fossil fuel use by 2030. Therefore, the capacity for abatement of emissions
through alternative electric applications should not be discounted, but rather credited in some
form, to support emissions reductions through different means.
Table 9: 2009 PRISM Analysis Targets
Source: EPRI
A current example in many states is the increasing use of electric vehicles to replace the use of
traditional fossil-fueled energy. According to Table 10 from EPRI’s 2007 PRISM analysis,282
plug-in hybrid electric vehicles are predicted to account for ten percent of new vehicle sales by
2017, and thereafter increase by two percent every year, which shows the current and potential
prominence of this new technology in the market. In EPRI’s 2009 analysis mentioned earlier,
EPRI shows that by 2030, electric vehicles could potentially reduce overall CO2 emissions by 9.3
percent.283 This reduction occurs due to the avoided use of gasoline and diesel fuels. By failing
to credit states to support electric vehicles in the market, however, EPA is discouraging the
production and implementation of electric technologies that could play a significant role in
282
Electricity Technology under a Carbon Future (PRISM Analysis)
http://mydocs.epri.com/docs/public/DiscussionPaper2007.pdf
283
This will almost necessarily increase CO2 from the power sector (which is at odds with the Clean Power Plan),
but that increase will be projected to be offset by the overall CO2 reduction economy-wide.
128
EPA’s overarching goal of CO2 reduction. EPA should modify state targets to reflect the fact
that beneficial electrification will almost certainly decrease economy wide CO2 emissions while
increasing CO2 emissions, though not in a one to one fashion, from the power sector.
129
Table 10: 2007 PRISM Analysis
Source: EPRI, Electricity Technology under a Carbon Future
3.
EPA Did Not Properly Account for the Decreasing Return on
Investment in Energy Efficiency in Its Development of Building
Block 4 and Should Adjust Its Efficiency Requirement Downward
The EPA asserts that states are capable of meeting an energy efficiency standard of 1.5 percent
savings per year after a 0.2 percent ramping period. 284 APPA believes that this implies a
significantly greater expenditure on reducing energy consumption than is modeled. There are a
number of ongoing and past efforts undertaken by utilities that have already “picked the lowest
hanging fruit.” In addition, appliance efficiency standards and building codes have been
improving energy efficiency for decades. In many cases, further increasing expenditures on
efficiency improvements on existing commercial, industrial, and residential structures could lead
to a decreasing return on investment.
Though there are energy consumers that have not replaced their appliances in many years, it is
common knowledge that due to standards and technological advancement, residential and
commercial appliances increase in efficiency over time. Sometimes these efficiency increases
are due to market forces. More often than not, however, efficiency increases are driven by
mandatory codes and standards. For example, DOE develops minimum energy efficiency
standards for most major residential appliances (water heaters, refrigerators, dishwashers, etc.)
284
79 Fed. Reg. at 34,872.
130
and many commercial products (motors, transformers, compressors, etc.). Other commercial
products are required to meet efficiency standards developed by private standards development
organizations (such as the American Society of Heating, Refrigerating and Air Conditioning
Engineers (ASHRAE) and International Code Council). In addition to regulating certain
commercial appliances, building energy codes like ASHRAE Standard 90.1 and the International
Energy Conservation Code have other requirements (for windows, walls, controls, etc.) that limit
the amount of energy commercial and residential buildings can use.
When the concept of minimum energy efficiency standards for appliances and buildings was first
introduced in the 1970s, cost-effective efficiency gains were relatively easy to achieve. But over
time, those gains have become increasingly difficult, and in some cases, add cost to a consumer,
without creating substantial improvement in efficiency.
Appliances and other efficiency elements embodied in codes eventually reach a “maximum
efficiency” level due to constraints on the technology or other constraints, such as thermal limits,
size limits, or other conditions that create a flat line or an asymptote of maximum efficiency.
The same is true for buildings where adding more insulation over the current baseline yields
diminishing returns. Even the most current code often logically stops mandating increases in
efficiency after significantly diminishing returns occur for a particular building type. For
example, a project in Indianapolis, which falls in Climate Zone 5A, a nonresidential, low-sloped
roof assembly exhibiting insulation entirely above deck, is required to have a continuous
insulation value of R-20. Despite the push for higher insulation levels, there is actually a very
good reason why ASHRAE, International Code Council, and others have set the mark at these
specific levels—the diminishing return on investment for further increasing R-values. If the
code were to mandate a doubling of insulation levels from R-20 to R-40, it would effectively
double the price of the project, but only provide a de-minimis amount of heat flow rate
reduction.285
285
http://www.bdmdialog.com/?p=493
131
Figure 14: ASHRAE Diminishing Returns on Insulation Investment for a Non-Residential
Low-Sloped Roof Assembly Building286
As a utility looks at potential efficiency gains from appliance rebates and incentives on a
product-by-product basis, the potential for energy savings, both in terms of percentage and
amount of energy savings, decrease significantly as compared to a “baseline standard” product
that meets the most recent federal appliance efficiency standards or state/local building codes.
This is true whether a program is calculating in-situ savings from replacing older, existing
appliances with new ones (e.g., a 15 to 20 year old refrigerator) or calculating the incremental
savings from purchasing a “high efficiency” product as compared to a “baseline standard”
product that meets the most recent federal appliance efficiency standards or state/local building
codes.
To illustrate: if a utility can successfully incentivize (at some program overhead cost) a consumer
to replace an old refrigerator that uses 700 kWh per year (which itself may have replaced an even
older refrigerator that used 1200 kWh per year) with a new ENERGY STAR model (under a
utility incentive program) that uses 400 kWh per year, and the federal standard is 440 kWh per
year, under a baseline standard calculation, the utility would only get credit for the 40 kWh
savings compared to the federal standard. At the least, EPA should allow state programs to
286
Source: ASHRAE
132
capture the real value of replacement savings rather than the difference between the
current standard and the more efficient model incentivized.
The following five examples illustrate the improvements in efficiency made and highlight the
marginally increasing additional cost of additional efficiency to both consumers and utility:
1) 4-foot linear fluorescent lamps and 2-foot U-shaped lamps
Fluorescent lamps are the most prevalent type of indoor lighting in many commercial buildings
in the U.S. The “workhorse” for many years has been the 4-foot linear lamp housed in a 2’ by 4’
fixture. As shown below, the increase in efficiency (lumens/Watt) was significant from the
1930s through the 1950s, and then from the 1970s through the late 1990s.
Before the Energy Policy Act of 1992 (EPACT 1992), typical efficacies for 4 foot lamps ranged
from 65-85 lumens/Watt (depending on the use of T12 or T8 technologies). EPACT 1992
required minimum color rendering index (CRI) and efficacy for 4-foot medium bi-pin lamps, 2foot “U-tubes” (u-shaped lamps), and 8-foot lamps. The minimum required values as of
November 1, 1995, were:
Table 11: Minimum Required Values as of November 1, 1995
Lamp
4 foot medium bi-pin (all
Wattages)
2 foot U-shaped medium
bi-pin > 35 Watts
2 foot U-shaped medium
bi-pin < 35 Watts
Lumens/Watt
75.0
68.0
Minimum CRI
45 or 69, depending on
the rated lamp Wattage
at least 69 CRI
64.0
at least 45 CRI
Source: APPA
In July 2009, DOE established a final rule for general service fluorescent lamps that took effect
in July 2012. The new requirements as of July 14, 2012, were:
133
Table 12: DOE July 2009 Requirements287
Lamp
Lumens/Watt
4 foot medium bi-pin
4 foot medium bi-pin
89
88
2 foot U-shaped medium
bi-pin
2 foot U-shaped medium
bi-pin
84
81
Color Correlated
Temperature
(< 4,500 K CCT)
(> 4,500 & < 7,000 K
CCT)
(< 4,500 K CCT)
(> 4,500 & < 7,000 K
CCT)
In April 2014, DOE issued a proposed rule that would go into effect in 2017 or 2018 with the
following values:
Table 13: Proposed Rule for 2017 or 2018288
Lamp
Lumens/Watt
4 foot medium bi-pin
4 foot medium bi-pin
92.4
90.6
2 foot U-shaped medium
bi-pin
2 foot U-shaped medium
bi-pin
86.9
84.3
Color Correlated
Temperature
(< 4,500 K CCT)
(> 4,500 & < 7,000 K
CCT)
(< 4,500 K CCT)
(> 4,500 & < 7,000 K
CCT)
The values shown in the April 2014 DOE proposed rule are considered “max tech,” or the most
efficient products available on the market. For these products, the increase in efficiency between
the current (2012) baseline and the “max tech” efficient product ranges from 2.9 to 4.1 percent.
That DOE chose these levels as cost effective and that the increase in efficiency from previous
rulemakings has been decreasing shows a decreasing ability to cost effectively implement
additional energy efficiency measures on fluorescent lights.
287
288
Source: APPA citing DOE
Source: APPA citing DOE
134
Figure 15: Lumens per Watt for 4-Foot Bi-Pin Lamps as Mandated by the Last 3 DOE
Rulemakings – clearly showing decreasing improvement per rulemaking. 289
4 Foot Medium Bi-pin Lamp Lumens Per Watt
100
80
60
40
20
0
1995
2009
2012
The following charts show lighting efficiency changes over time.
Figure 16: Lamp Efficiency in Lumens per Watt by Year 290
289
290
Alex Hofmann – Graph of lumens per watt by regulation from 1995 to 2012 for 4 foot Bi-pin lamp
http://americanhistory.si.edu/lighting/tech/chart.htm
135
2) Centrifugal Chillers
These types of cooling systems are used in large commercial buildings. The values below show
the rise in efficiency standards for centrifugal chillers with a rated cooling capacity of 500 tons.
Table 14: Efficiency Standards for Centrifugal Chillers291
ASHRAE Standard
90.1 Year
1989
1999
2010
Unit Size (Tons)
COP/IPLV Value
> 300
> 300
> 300 ton to < 600 ton
2013
> 400 ton to < 600
5.2 COP / 5.3 IPLV
6.10 COP / 6.40 IPLV
6.10 COP and 6.40 IPLV (Path A) and 5.86
COP and 8.79 IPLV (Path B)
6.28 COP and 7.03 IPLV (Path A) and 6.01
COP and 9.25 IPLV (Path B)
For ASHRAE 90.1-2010, the metric changed, but for the purposes of these comments APPA has
converted the new standard back to the previous units. After a quick look at the data, it can be
seen quite clearly that there is a decreasing rate of return on cost-effective efficiency
improvements mandated in standards. This is due to both thermodynamic and physical
limitations of the technologies being used in chillers and the diminishing return on investment
for a consumer in energy efficiency. This makes a utility’s job designing incentives for energy
efficiency programs much more costly; this increases the price per watt-hour saved.
3) Residential Refrigerator/Freezers
National efficiency standards for residential refrigerators and freezers have been set four
times.292 The chart on the next page shows the current flat line of efficiency improvements:
291
Source: APPA citing DOE
National efficiency standards have been set once by the National Appliance Energy Conservation Act (NAECA)
of 1987 that took effect in 1990, and three times by DOE that took effect in 1993 and 2001, and will take effect in
September 2014.
292
136
Figure 17: Efficiency Improvements and National Standards for Refrigerators
The Energy Efficiency Story: Lower Energy Use/Lower Costs to Consumers
137
Again, this reflects the effectiveness of current standards at pushing appliance technology to its
maximum achievable efficiency. This push means there are very few or very expensive
additional incremental energy efficiency improvements available to utilities from switching out
refrigerators meeting the more recent standards.
4) Residential Dishwashers and Clothes Washers
As with residential refrigerators, standards for dishwashers and clothes washers have been
increased multiple times over the past three decades. In the case of residential clothes washers,
DOE has created “two-step” standards. In 2001, DOE set efficiency standards that took effect in
2004 and higher standards that took effect in 2007. In 2012, DOE established two sets of
increasing clothes washer efficiency standards that will take effect in 2015 and 2018.
The following charts show the impacts of federal efficiency standards:
Figure 18: Efficiency Trends for Clothes Washers293
293
Source: National Association of Home Appliance Manufacturers
138
5) Commercial Building Codes
Below is a slide that shows the impacts of ASHRAE commercial building energy efficiency
codes over the years. A recent Pacific Northwest National Laboratory (PNNL) analysis (entitled
“Building Energy Codes Program: National Benefits Assessment, 1992-2040,” Report #22610)
was published in October 2013 and shows the impact of improved envelope and commercial
equipment efficiency and controls:
Figure 19: Energy Code Levels over Time
EPACT 1992 delegated authority to ASHRAE to develop minimum energy efficiency standards
for commercial construction where states would be required to adopt the ASHRAE standard (or
an equivalent) if DOE determined that newly published ASHRAE standards would save energy
when compared to the previous version of the standard.
ASHRAE published a new standard in 2013. The DOE preliminary determination that was
published in May 2014 stated that the ASHRAE 90.1-2013 building energy efficiency standard
will save about 7.6 percent more energy than a building built to meet the ASHRAE 90.1-2010
139
standard. This will further “raise the floor” as many states (and/or cities and counties) will adopt
and enforce ASHRAE 90.1-2013 by 2016.
There is general agreement within ASHRAE that future efficiency gains for ASHRAE Standard
90.1 will be much more challenging to achieve, as all of the “low-hanging fruit” has been picked
and the standard is already bumping up against technical and economical limits. Due to the
reduced availability of cost-effective energy efficiency gains as energy standards are
approaching “max tech,” and increased energy efficiency standards that will become effective in
the 2015-2018 timeline, EPA should recognize that more efficiency has been taken off the table
than was proposed or modeled and reduce its energy efficiency requirement under building block
4 accordingly.
Below is a list of the residential electric appliance efficiency standards that have increased or
will increase by 2016:


















Incandescent light bulbs (100 Watts to 72 Watts): January 2012
Kitchen Ranges, Ovens, and Microwave Ovens: April 2012
Incandescent Reflector Lamps: July 2012
Electric Boilers: September 2012
Incandescent Light Bulbs (75 Watts to 53 Watts): January 2013
Direct Heating Equipment: April 2013
Pool Heaters: April 2013
Dishwashers: May 2013
Incandescent Light Bulbs (60 W to 43W; 40 W to 29 W): January 2014
Room Air Conditioners: June 2014
Residential Refrigerators/Freezers: September 2014
Residential Clothes Dryers: January 2015
Residential Central Air Conditioners (regional standards): January 2015
Residential Heat Pumps (national standards): January 2015
Residential Clothes Washers: March 2015
Residential Electric Water Heaters: April 2015
External Power Supplies: February 2016
Microwave Oven Standby Power: June 2016
A graph of the historical efficiency gains by appliance shows how the energy use per year for
appliances has been decreasing at a slower rate despite increasingly stringent regulations and
more stringent economic tests, such as tests incorporating the social cost of carbon.
140
Figure 20: Historical Efficiency Gains for Various Appliances
4.
EPA Needs to Reconsider the Proposed Best Practices Level of
Performance to Be Less Stringent and Reflective of a Feasible
Level for All States.
According to EPA, past performance and policy requirements are direct indicators of an
achievable approach. However, this is not the case if the past and future levels for a few states
are used to determine the achievable level for all states. This methodology fails to account for
unique circumstances in each state that may prevent it from employing energy efficiency as
quickly and/or as effectively as other states.
In chapter 5 of the GHG Abatement Measures TSD, the analysis EPA used to calculate the 1.5
percent best practices level of performance is based on reported savings and projected policy
goals for individual states. The Agency uses EIA reported data in the analysis to show that three
out of 50 states have reached 1.5 percent or greater incremental savings in 2012. It is
unreasonable to base the achievable level of savings for all 50 states on the performance of 6
percent of all states. This is another instance, similar to EPA’s approach in building block 3’s
analysis, where the approach of “one-size-fits-all” is not justifiable. APPA understands the
difficulty in setting a unique performance level for individual states, but believes that the
performance level must be achievable for all states, not just a select few.
141
The Agency’s analysis is based on reported savings data from EIA. However, it is important to
note that actual savings are not always reflected appropriately in reported data. In the discussion
paper produced by Resources of the Future, entitled “Energy-Efficiency Program
Evaluations,”294 Figure 21, which includes almost half of the programs included in the analysis,
shows utility reported savings as being substantively higher than the evaluated savings.”
Figure 21: Total Resource Cost (TRC) Test (Program Benefits/ Program Costs)
Source: Resources for the Future, Energy Efficiency Program Evaluations
In the same discussion paper, the analysis provided by RFF found that predicted savings were
also significantly different than evaluated savings. This can be seen by looking at the realization
rate, which is the ex-post estimated savings divided by ex-ante projected savings. By basing the
majority of the proposed plan analysis on reported data, EPA did not provide a cushion for
situations in which the reported savings from energy efficiency measures do not accurately
reflect the actual savings.
On the other hand, the Proposed Rule also uses American Council for an Energy-Efficient
Economy’s (ACEEE) analysis to show that 11 out of 50 states have policies that require the
achievement of a 1.5 percent incremental savings performance level by 2020. EPA states that
the energy efficiency goals were constrained by practical considerations of state EE policy
implementation. However, it is not appropriate to base the feasibility of the goal for all
states on state policies implemented in only a minority of the states.
In the TSD, EPA heavily relied on industry reports to predict the performance level. Of the three
national studies referenced, only one study predicted a 1.5 percent incremental savings level.
294
http://www.rff.org/documents/rff-dp-10-16.pdf
142
The Agency’s use of top-down, policy-based studies to support analysis in the TSD has a serious
technical flaw. EPA should consider using a combination of studies, with a focus on
bottom-up analysis, to make this determination since top-down analysis can fail to
realistically validate the model.295
Table 15: Summary of National EE Potential Studies296
As shown in Table 15, the EPRI analysis projects a 0.6 percent high maximum achievable
potential per year. Although this is considered aggressive according to the 2014 bottom-up
analysis, EPA bases the best case scenario on the results of the top-down approach in the
ACEEE study. As shown in Table 16, the average and median incremental savings from energy
efficiency in 2012 (baseline year), according to EIA reported data, is 0.55 percent and 0.48
percent, respectively. Unlike EPA’s approach, it is reasonable to assume that a regulation would
intend to bring the states in the lowest quartile up to the median or average. As an aggressive
approach, it seems more reasonable than EPA’s performance level to use the third quartile, 0.93
percent, as a bar for states in the lower quartiles to aim for. It is irrational to require all states to
achieve a 1.5 percent level of savings, which has not been attained by even the leading states in
the area of energy efficiency (states in the third quartile).
295
296
http://www.princeton.edu/~achaney/tmve/wiki100k/docs/Top-down_and_bottom-up_design.html
EPA Proposed Rule TSD
143
Table 16: Analysis of Baseline (2012) Energy Efficiency Data from EIA 861 -Survey Results
Source: APPA’s Analysis of 2012 EIA data
The Proposed Rule presented two options, outlining the assumptions used in each option when
calculating the best case scenario for states. APPA does not fully support either option to its
fullest; however, if one option is required, APPA would propose the use of option 2. APPA
sees the 1 percent performance level with a 0.15 percent ramp-up rate in option 2 as a more
reasonable approach than the 1.5 percent performance level with a 0.2 percent ramp-up rate in
option 1. This level of savings is more in tune with the aggressive third quartile approach
discussed earlier. Although the performance level is more justifiable, option 2 is still aggressive
in nature. Under option 1, all states theoretically achieve the 1.5 percent incremental savings
level by 2025, whereas all states reach the 1.0 percent level by 2024. This shows that option 1,
while more reasonable in its approach, is still aggressive. The stringency of the reduction goals
generally necessitates the use of all building blocks; thus any claimed flexibility is illusory. EPA
should re-analyze its methods and assumptions used to calculate the potential for energy
efficiency in each state and take a less aggressive approach.
5.
APPA Supports EPA’s Assumptions Between the Years 2012 and
2017 in the Best Practices Scenario.
In the best practices scenario calculation, APPA supports EPA’s assumption that there are no
annual incremental savings or expiring savings required until the year of 2017. APPA also
supports EPA’s evaluation of expiring savings in the model. We agree with EPA’s assumption
that the states will have no incremental savings from energy efficiency measures until 2017.
This allows states to receive “credit” for energy efficiency savings incurred from 2012 to 2017 to
count towards their goals.
EPA’s proposed approach to expiring savings is reasonable and appropriately reflects a wide
range of products in this analysis. APPA supports the assumption of a linear decline of expiring
savings over 20 years since it provides a steady regression instead of producing spikes and
downturns in savings values for certain years.
144
6.
EPA Also Needs to Account for Differences in Reported and
Projected Energy Efficiency Savings Versus Actual Savings as
Well as Acknowledge the Potential Consequences if Projected
Savings Are Not Met.
Projected savings from energy efficiency measures are to be included in a state plan, but EPA
should acknowledge that projected savings from a demand-side program or measure cannot be
forecasted with exact precision. EPA should provide leniency in the case that energy efficiency
measures do not meet the projected savings. The realization rates shown in Table 17 represent
post estimated savings divided by ex ante projected savings.297 These rates show the significant
difference in projected savings and evaluated savings. In other words, there is inconsistency
between the actual impacts, or real savings, from energy efficiency measures and the forecasts.
This can be a result of the many factors on which effective programs are dependent. Many of
these factors are outside the control of those implementing the technologies.
In the event that a state does not meet its goal due to the difference between projected and
actualized savings from an EE measure, where the cause of the difference is not a result of an
error on the states’ part, EPA should have an alternative means of compliance. As described in
more detail by the comments submitted by National Climate Coalition (NCC) which includes
APPA, there is a precedent for such a fee mitigation instrument. The concept was introduced
during the reauthorization of the Clean Air Act in 1990 and was described in President Bill
Clinton’s July 1997 memorandum to EPA when it revised the NAAQS for ozone and particulate
matter.298 This fee payment has a cap placed on it and should not be used for other state
purposes, such as increasing general revenue.
297
Energy-Efficiency Program Evaluations. http://www.rff.org/documents/rff-dp-10-16.pdf
Presidential Documents, “Memorandum of July 16, 1997, Implementation of Revised Air Quality Standards for
Ozone and Particulate Matter,” 82 Fed. Reg. 38421, 38429 (July 18, 1997).
298
145
Table 17: Summary Conclusions
Source: Energy-Efficiency Program Evaluations - Resources for the Future
The Proposed Rule has developed a “best practices” demand-side energy efficiency scenario
that, according to EPA, estimates the ability of each state to implement policies that increase
investment in energy efficiency technologies. EPA also states that this scenario is intended to
represent a feasible scenario for reductions of emissions from fossil fuel-fired EGUs as a result
of increased energy-efficiency technology use. The Agency claims that the scenario uses a level
of performance demonstrated by many leading states and considers each state’s unique existing
level of performance, while allowing appropriate time for each state to increase from current to
best practices level. While the theoretical concept is beneficial to the environment, the real
application of the scenario will yield different results that may lead to an unattainable
compliance standard.
APPA also notes the lack of guidance in situations where a reduction in utilization is not
achieved, but an approved efficiency measure has been implemented. In this case, CO2
emissions would not be decreased, yet compliance dollars would have been spent,
potentially triggering enforceable action. EPA should provide guidance to states that want
a safety valve or “true-up” mechanism to allow swift resolution of such issues, should they
occur.
146
EPA also needs to provide clear instructions and protection for entities relying on the energy
efficiency measures to be implemented by a contracted third party if that party goes out of
business and does not deliver the CO2 reductions as promised. The Agency should make clear
that the utility is not responsible for the CO 2 reductions required from any unrelated third
party entity that is nominated by the state to implement energy efficiency programs.
7.
EPA Should Provide Additional Guidance in Multiple Areas.
EPA should provide guidance on the issue of interstate entities that would be subject to
duplicative reporting of demand-side energy efficiency efforts. Specifically, it should provide
clarification on how certain interstate scenarios, and other similar situations, would be handled
under the different plan types discussed in the State Plan Considerations TSD. The Agency
should allow flexibility on what can be used for measurement and verification (M&V) at the
state level.
EPA should provide states with indicative approved methodologies to measure and verify energy
savings from energy efficiency projects and programs (both single technology measures and
whole buildings), and a process for states and industry to submit additional methodologies f or
consideration and approval. These should include, but not be limited to:








Uniform methods projects standard;
International Performance Measurement and Verification Protocol;
ASHRAE Guideline 14-2002 Measurement of Energy and Demand Savings;
DOE’s Superior Energy Performance program created an M&V protocol for industry;
Technical Reference Manuals (“deemed savings” charts that provide statistical savings
value for equipment upgrades);
SEE Action Network and regional standards such as those by NEEP and the Northwest’s
Regional Technical Forum;
ISO 50001:2011 Energy management systems; and
Demand response measurement and verification, such as The National Action Plan for
DR M&V element http://emp.lbl.gov/sites/all/files/napdr-measurement-andverification.pdf.
Existing state and utility programs should be recognized if they utilize EPA-approved or
equivalent measurement and verification protocols and standards. The Agency should make it
easy for such programs to scale up if needed and give credit for “early action .” These programs
may include, but not be limited to:
147




Utility DSM – or incremental DSM depending on “credit for early action”;
EERS requirements – or incremental EE additions depending on “credit for early action.”
EERS carve out for ratepayer funded rebates/incentives. Rebate programs;
Other highly efficient Energy-Efficient Appliances and Equipment upgrades not included
in EERS or in states with no EERS; and
Innovative energy savings programs, such as tree planting.
EPA should also provide guidance for smaller entities that will be required under this rule to
implement evaluation, measurement, and valuation (EM&V) protocols for their efficiency
programs. For small utilities,299 this could easily become burdensome and costly since they do
not have the resources (workforce, funding, etc.) to accomplish the same caliber of reporting and
analysis as larger utilities. APPA recommends EPA establish a subcategory group of utilities
that will have more streamlined requirements for their EM&V reporting.
The development of these protocols and methods for measuring and verifying energy savings
will take time and the Agency should explicitly allow additional time, if needed, in state plans
for states to set up the constructs or modify their existing programs to meet any additional
requirements imposed by this rule. This conforms to the timing questions posed in EPA’s
NODA.
8.
EPA Should Provide Relief for New Electricity Use Driven Solely
by Compliance Requirements from Other EPA Rules.
EPA rules periodically come with significant requirements for new consumption of electric
power unrelated to air quality conditions. For instance, EPA recently established new criteria for
lower limits on discharges of ammonia from wastewater treatment plants across the nation in
order to be protective of current and previously occurring communities of mussels. This new
lower limit cannot be achieved by current waste treatment systems commonl y found in smaller
communities around the nation. For instance, in Missouri, lagoon systems which currently
achieve EPA approved limits in over 300 communities will have to be closed down in the future
and replaced with either mechanical plants that operate on electricity around the clock or
distributed as irrigation on surrounding farm land by pumps, also using electricity. This
additional consumption of electric power should not be allowed to count against electric power
savings achieved in other parts of the state.
299
As an example, a small utility could have only a few employees, less than 2,000 customers, and multiple service
responsibilities (e.g. those same employees are also partly responsible for operating the water utility).
148
XV.
States Need More Time to Prepare, Submit and Obtain EPA
Approval for Their Plans.
The Proposed Rule does not allow states sufficient time to prepare compliance plans and submit
them to EPA for approval. As noted in Section III, the Proposal essentially provides states one
year to develop an intrastate plan and two years to develop multi-state plans, with the ability to
request a one-year extension. Given the complexity and pervasiveness of what is required, this is
simply not enough time.
States will need to consult with policy makers, stakeholders, utilities, and other entities with a
potential direct compliance obligation to begin determining the critical elements of the program,
including enforceability. This could be contentious and thus time consuming. Moreover, a
state’s compliance plan is likely to require a variety of changes in state law to be implemented.
Even if agreement for such legislative proposals is achieved in concept, the actual legislative
process itself can be very time consuming and uncertain. Timing issues are only exacerbated in
efforts to develop multi-state plans.300
In its October 2014 comments to EPA, the National Conference of State Legislatures (NCSL)
stated:
However, NCSL believes the 13 months between the expected finalization
of the rule (June 2015) and the deadline for states to submit
implementation plans (June 2016) is not enough time for states to make
any legislative changes that may be needed in order to submit a complete
SIP, given the incompatibility of EPA’s proposed timeframe with state
legislative calendars.
NCSL goes on to point out that some state legislatures will have already adjourned for the year at
the time the final rule is issued, and that four states only hold regular sessions every other year
“putting them at an even further disadvantage.” The NCSL letter acknowledges the possibility
of special legislative sessions to address the state’s compliance plan, but notes that in 16 states,
only the governor can call a special session, and that there are often significant costs to the state
to convene such special sessions. NCSL concludes: “Therefore, the June 2016 timeframe for
states to submit an implementation plan or meet the requirements for an extension…poses
significant challenges.”
300
See comments of the National Conference of State Legislatures, posted Oct 21, in the docket at:
http://www.regulations.gov/#!documentDetail;D=EPA-HQ-OAR-2013-0602-20676
149
An example of such a challenge for a state is in Minnesota, where the existing Integrated
Resource Planning Statute301 would have to be re-written and passed by the legislature, to make
the changes necessary for the state to meet its reduction goals under the Proposal. Likewise, the
Minnesota renewable energy mandate302 and state energy efficiency goal303 would also have to
be revised to reflect the state plan. This assumes there is sufficient political support in the
legislature to do so.
Even if a state has authority to adopt a state plan under its administrative procedures act and
delegated authority, it still must go through a lengthy public process to enact any such rules,
again adding to the delay in the process of even presenting a plan to EPA for approval.
Missouri’s environmental agency, the Department of Natural Resources, which has no legal
authority over the operation of electric utilities, has a statutory timetable that requires as much as
two full years for the implementation of a single rule.
APPA notes that the east coast’s Regional Greenhouse Gas Initiative (RGGI) began with formal
discussions among various states in 2003. The states then announced a memorandum of
understanding in December 2005 and published a model rule in August 2006 setting forth the
recommended regulatory framework for developing regulatory and statutory proposals for the
various state members (see chart RGGI Timeline below). States then began their own in-state
rulemaking processes to adopt a framework for each state to participate in a regional cap and
trade program. They considered and adopted regulatory rules or statutes governing a trading
program, parameters for acceptable emission offset projects, and authorizing and regulating
language for the auctioning of allowances. States completed their regulatory and statutory work
at the end of 2008 and began a regional trading program in January 2009.
Thus, it took a total of five to six years to go from formal planning to actual trading in RGGI,
which is a much simpler program than the regional plans envisioned under EPA’s Proposed
Rule. Similarly, California took several years to develop and adopt its cap and trade program.
301
Minn. Stat. 216B.2422
Minn. Stat. 216B.1691
303
Minn. Stat. 216B.241
302
150
Figure 22: Regional Greenhouse Gas Initiative Timeline
The Proposed Rule should be revised to give additional time to the state planning process. It
should provide states a full five years from the date the rule is finalized to submit their state
plans. In conjunction, EPA should also revise the final compliance deadlines by a corresponding
length of time.
XVI.
EPA Provides Too Little Guidance on Establishing Multi-State Plans
and Interstate Trading and Cooperation.
A number of states have indicated they are interested in joining with other states to develop a
multi-state plan. The primary, though not exclusive, example is the RGGI states. Others states
seem interested in exploring mechanisms that would allow a certain level of cooperation among
the states, such as the possible trading of allowances or credits, but that stop short of full-blown
interstate or regional compliance plans. Or states may simply want to have an agreement among
them with respect to which states will be able to include emissions reductions and/or investments
in compliance measures in their compliance plans. This might be accomplished through
memoranda of understanding or similar agreements.
Multiple issues, questions, and concerns arise in considering such arrangements. One of the
primary issues is how to address compliance in the case of generation resources located in one
state that serve customers in another state or even multiple states. The Utah Associated
Municipal Power Systems’ (UAMPS) ownership stake in the San Juan Generating Station
(SJGS), located in Farmington, New Mexico, serves as an example of this problem. No UAMPS
member that participates in its San Juan Project serves load in New Mexico. UAMPS members
151
who would be interested in using energy efficiency measures adopted in Utah to offset CO 2
emissions from SJGS in New Mexico would be precluded from doing so unless a multi-state
plan is adopted between the two states or some agreement that energy efficiency measures
performed in Utah by UAMPS members can be allocated in New Mexico.
There are related problematic scenarios. Since states are allowed to take varying approaches, it
is possible that State A imposes a CO2 reduction obligation on a utility that is a non-emitting
utility in that state (along with its associated cost to consumers) and that State B where the utility
has generation that emits CO2 takes an approach that only CO2 emitters are obligated to reduce
CO2 (along with that cost). Under this scenario, the utility is paying twice for reducing the same
CO2 emissions, a fact which could be compounded if there are several states served by the utility
that take the same approach as State A. Or, how might an agreement be reached where a state
wants to claim credit for renewable generation built outside its borders based on an assertion that
the renewable project was built pursuant to that state’s renewable policies?
Another question arises where the Proposed Rule treats adjacent states differently with respect to
assumptions and computations in the four building blocks, resulting in inequitable final reduction
goals. This becomes a disincentive for states to work together. For example, why should Iowa,
which has a 16 percent reduction goal, ever partner with Minnesota which has a 41 percent
reduction goal?
The Proposal does not provide states and affected entities with enough guidance on how to
assess or address these situations. Perhaps, for example, EPA should establish a default
adjudication process to resolve interstate disputes with respect to credit for CO 2 reduction
compliance measures and recognizing utility resource investments. Since the Agency seems
generally to prefer multi-state plans over single-state plans, the final rule should provide more
detailed guidance on how to develop such plans.
XVII.
EPA Should Eliminate the Interim Reduction Goal and Allow States
to Determine Their Own Glide Path.
One of the most onerous elements of the Proposal is the interim reduction goal that states must
begin meeting in 2020. In a significant departure from prior Section 111 rules, EPA mandates a
highly prescriptive implementation framework, requiring states to meet aggressive near-term
CO2 goals and demonstrate compliance with those goals over a ten-year averaging period (202029) prior to the final compliance goals in 2030. While EPA stresses flexibility as a key element
of the Proposal, here the Agency has proposed to make mandatory the minimum pace of
implementation. Congress created a framework of cooperative federalism throughout the CAA.
This careful federal-state balance is central to the Act and EPA should respect that balance with
regard to compliance trajectories.
152
The obligations to begin meeting the interim goal start a mere two years after state submittal, and
EPA approval, of a state’s compliance plan (and even sooner if a state obtains an extension).
While it may seem as if this is not an onerous step because the interim goal must be met on
“average” over a ten-year period, the reality is that unless draconian reductions are made by
January 1, 2020, the interim goals cannot be achieved in most states. The interim goal is not
simply a pro-rata based step towards the state’s 2030 final reduction requirement. Often referred
to as a “cliff,” the interim goal requirement is, rather, a drastic reduction constituting a
significant percentage of the final 2030 requirement. As can be seen in Table 18 below, the drop
from the 2012 fossil fuel generation CO2 emissions rate to the interim goal is a significant
percentage of the final goal, and because it must be done so soon after a state plan is finalized,
constitutes an almost impossible burden to meet at all, let alone in a cost-effective or efficient
manner. If not eliminated, or significantly adjusted, “these front-end-loaded” interim goals will
result in huge stranded costs for utilities that will be borne ultimately by consumers.
The 10-year averaging period put forth in EPA’s Proposal between the interim and final goals
does little to relieve the situation. For example, if Minnesota were to delay emissions reductions
even a single year, it would need to maintain an emissions rate below the final requirements for
the remaining 9 years of the compliance period. In its NODA EPA proposed letting states adjust
their timelines and interim requirements to avoid the serious implications from mandatory large
scale fuel switching on system reliability, resiliency, infrastructure, and electricity cost. APPA
agrees that this flexibility is necessary to avoid many of the serious possible consequences of this
Proposal including addressing the minimal relief provided by allowing states to meet their targets
on average over the 10 year period form interim to final calculated goals.
Table 18: Percentage CO2 Reduction by State Interim versus Total
State
2012
Fossil
Rate per
EPA TSD
p25 -26
2020
Interim
Goal
Percentage
reduction
from Fossil
Fuel to
Interim
Goal
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
1,518.46
1,368.31
1,551.23
1,722.36
899.80
1,959.08
844.41
1,228.35
1,198.45
777.69
1,028.47
590.07
1,243.75
661.07
19.1%
12.4%
49.9%
40.3%
34.4%
36.5%
21.7%
153
Final Goal
Total
(2030 and Percentage
thereafter) Reduction
1,059.01
1,003.03
702.07
909.66
536.96
1,107.83
540.26
30.26%
26.70%
54.74%
47.19%
40.32%
43.45%
36.02%
State
2012
Fossil
Rate per
EPA TSD
p25 -26
2020
Interim
Goal
Percentage
reduction
from Fossil
Fuel to
Interim
Goal
Delaware
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
1,255.15
1,238.00
1,598.30
1,783.35
857.99
2,188.78
1,991.00
2,196.76
2,319.54
2,166.33
1,532.85
873.27
2,028.65
1,000.78
1,813.54
2,013.22
1,185.00
2,010.00
2,438.60
2,162.24
1,090.79
1,118.66
1,034.63
1,797.52
1,095.86
1,771.59
2,367.85
1,897.06
1,562.02
1,080.79
1,627.04
917.88
1,790.82
2,255.58
2,015.38
973.01
851.45
966.40
1,458.05
266.49
1,483.48
1,699.00
1,398.37
1,706.61
1,934.04
1,014.78
415.35
1,543.47
738.62
1,310.32
965.39
783.30
1,705.31
2,008.13
1,723.87
753.93
637.44
759.36
1,197.14
729.86
1,181.77
1,852.48
1,570.14
995.60
471.43
1,296.51
866.88
920.90
870.13
1,353.13
22.5%
31.2%
39.5%
18.2%
68.9%
32.2%
14.7%
36.3%
26.4%
10.7%
33.8%
52.4%
23.9%
26.2%
27.7%
52.0%
33.9%
15.2%
17.7%
20.3%
30.9%
43.0%
26.6%
33.4%
33.4%
33.3%
21.8%
17.2%
36.3%
56.4%
20.3%
5.6%
48.6%
61.4%
32.9%
154
Final Goal
Total
(2030 and Percentage
thereafter) Reduction
840.64
740.21
833.78
1,305.54
228.37
1,270.73
1,531.27
1,300.73
1,499.39
1,763.12
883.07
377.58
1,186.71
575.64
1,161.27
872.76
691.77
1,544.41
1,771.27
1,478.94
647.31
486.14
531.09
1,047.62
549.14
992.20
1,783.16
1,338.34
895.30
372.35
1,051.96
782.26
771.75
740.67
1,162.62
33.02%
40.21%
47.83%
26.79%
73.38%
41.94%
23.09%
40.79%
35.36%
18.61%
42.39%
56.76%
41.50%
42.48%
35.97%
56.65%
41.62%
23.16%
27.37%
31.60%
40.66%
56.54%
48.67%
41.72%
49.89%
43.99%
24.69%
29.45%
42.68%
65.55%
35.35%
14.78%
56.91%
67.16%
42.31%
State
2012
Fossil
Rate per
EPA TSD
p25 -26
2020
Interim
Goal
Percentage
reduction
from Fossil
Fuel to
Interim
Goal
Texas
Utah
Virginia
Washington
West Virginia
Wisconsin
Wyoming
1,420.23
1,873.69
1,437.99
1,378.56
2,056.36
1,988.39
2,330.54
930.42
1,446.45
990.95
333.68
1,853.41
1,375.71
1,899.39
34.5%
22.8%
31.1%
75.8%
9.9%
30.8%
18.5%
Final Goal
Total
(2030 and Percentage
thereafter) Reduction
790.82
1,321.73
809.81
214.50
1,619.78
1,202.58
1,714.38
44.32%
29.46%
43.68%
84.44%
21.23%
39.52%
26.44%
Moreover, a federally-imposed interim goal is not necessary. In its public statements, EPA
continuously stresses that the goal of the Proposal is to reduce overall CO2 emissions 30 percent
by 2030.
In addition to eliminating the interim goal, the Agency should allow the states to set their own
“glide path” for compliance with the 2030 final goal. States must be able to determine the
enforceable emission reduction trajectory to reach that goal for several reasons. Many states
have already adopted measures that EPA has included in its BSER determination—RPS and
demand-side management programs, for example. These programs have both rate and reliability
implications that must be taken into account in determining the timing for a state’s
implementation of EPA’s final performance standard. The states are uniquely positioned to
address these considerations and craft an appropriately gradual trajectory.
States are also best positioned to determine resource adequacy implications of implementation
timing. For example, states and regional planning entities are best able to assess existing natural
gas transportation infrastructure (including interstate and distribution pipeline and storage
capacity), as well as natural gas supply and market conditions to determine whether they would
be adequate to support the 70 percent state-average capacity factor ceiling for existing NGCC
capacity targeted by EPA.304
304
FERC Chair Cheryl LaFleur has noted “gas pipeline adequacy should be considered from a regional perspective,
not just a national perspective, due to existing constraints on the system.” Responses of Acting Chairman Cheryl A.
LaFleur to Committee on Energy & Commerce Subcommittee on Energy & Power Preliminary Questions for the
Federal Energy Regulatory Commission at 7. http://www.ferc.gov/CalendarFiles/20140729091732-LaFleur-07-292014.pdf
155
In addition, states can better: 1) examine existing electric transmission infrastructure to
determine whether it would be adequate to support the broad regional or inter-regional
dispatched-based substitution of NGCC generation and increased renewable generation
contemplated by the Proposal; 2) assess potential limitations on a state’s ability to influence
economic dispatch of generation to implement their plan; 3) determine and apply the remaining
useful life of existing coal-fired (and other) generation resources, particularly where affected
sources have made significant investments in such resources to comply with other environmental
regulations (e.g., MATS, Cross State Air Pollution Rule (CSAPR)); 4) plan for any legislation
that may be required; and 5) determine state-specific economic impacts of the interim path and
timing.
Therefore, EPA should allow states to set their own appropriate pace for implementation prior to
the final 2030 performance goals. While states may need to adopt some interim standard to
ensure they are on track to meet the 2030 goals, states should be permitted to adjust the glide
path to 2030, taking into account further analysis of appropriate factors. States are best suited for
this role because they have a unique ability to evaluate reduction opportunities and expected
timelines.
XVIII. States Should Be Allowed an Opportunity to Adjust Their Final
Reduction Goals, the Year That the Goals Are to Be Achieved,
and/or the Glide Path Based on Materially Changed Circumstances.
The Proposed Rule should be modified to expressly permit a state to seek an adjustment of its
2030 emission reduction goal at any time to account for either late discovery of errors in EPA
assumptions or new information that becomes available to the state after the close of this
comment period. As discussed extensively in these comments, the Proposed Rule relies on a
number of highly questionable assumptions. These include assumptions such as those related to
the price and availability of natural gas, reliability, the economics and availability of renewable
energy, the long-term performance of energy efficiency programs, the functioning of wholesale
electricity markets, and the time required to develop state compliance plans. The Proposal also
does not properly recognize or account for other relevant factors, such as shifts in population,
growth in the manufacturing sector, impediments to the development of new transmission
facilities, or the risk associated with relicensing existing nuclear and hydro generating units.
In these comments, APPA provides recommendations to address these assumptions and
oversights, as well as other issues that, if adopted in the final rule, would provide states with
more of the flexibility they need to comply with the final rule. Even with that flexibility,
however, unanticipated events can and will occur that adversely affect a state’s ability to meet its
required reductions.
156
States should be afforded an opportunity to at any time to submit to EPA an amendment to their
state compliance plan that adjusts elements of that plan to equitably account for materially
changed circumstances. This should include the ability to adjust the level of the final reduction
goal, the year in which that reduction is to be achieved, and the glide path to reach that goal.
States bear the burden to adequately support, document, or otherwise justify any proposed
changes to state plans. However, when that justification is adequately provided, EPA should
approve the changes requested by the state.
XIX.
EPA Should Allow Additional Compliance Flexibility.
APPA believes that EPA needs to provide additional flexibility to states to determine or develop
compliance options. For example, EPA should make clear that technologies using fuels, such as
geothermal, landfill methane gas, pumped storage hydropower, dairy digester gas, biogas, and
biomass, are eligible for compliance. Additionally, the rule should make provisions for utilities
to receive credit for improvements to distribution or transmission grids that effectively reduce
CO2 emissions through improvements that reduce line losses. The Agency should also clearly
allow the emission reductions resulting from programs between utilities and their customers,
such as the beneficial use of fly ash from power plants used in the manufacture of cement
(reducing cement plant costs, as well as CO2), to count toward compliance with a state plan.
EPA should allow states to set up alternative compliance systems, subcategories , and alternative
regulatory systems for public power utilities that would face a disproportionate stranded cost or
debt to the local community as a result of this Proposal. It should be left to the states to
determine how best to design, manage, and implement these systems, but the Agency should
clearly recognize this problem. In addition, some states may wish to enact a carbon tax or
similar program to incent the actions necessary to achieve the state’s reduction goal, rather than
through a system of regulations and enforcement. EPA should provide states with this
flexibility.
XX.
The Cost of Electricity to Consumers Has Been Increasing and Will
Increase Even More Under This Proposal
APPA is concerned that the Proposal will result in increases in electricity prices that are likely to
be substantial in a number of states. This will come on top of steadily increasing prices to
consumers that are projected by the federal government to continue upward even without the
implementation of this Proposal. These increases will be exacerbated in some regions due to the
structure and operations of the RTO-administered wholesale markets. And they will have a
disproportionate adverse impact on low- and fixed-income consumers.
157
A June 26, 2014, article in the New York Times reports that in a speech to the League of
Conservation Voters, President Obama said environmental advocates need to acknowledge
Americans’ worries about the economic effects of efforts to combat climate change. 305 The
Times quotes the President as saying:
People don’t like gas prices going up; they are concerned about electricity
prices going up. If we are blithe about saying ‘this is the crisis of our
times,’ but we don’t acknowledge these legitimate concerns – we’ve got to
shape our strategies to address the very real and legitimate concerns of
working families.
A.
EPA’s Regulatory Impact Analysis Is Flawed.
APPA believes that EPA failed to meet the most basic tests of good regulatory analysis regarding
costs in this proposal. Unlike other parts of the Clean Air Act, Section 111(d) provides for cost
considerations (along with others) in the setting of standards. But in the Regulatory Impact
Analysis (RIA), EPA has asserted there will be negligible economic impacts from this rule.
While it acknowledges that electricity prices will rise (Regulatory Impact Analysis at 3-38 to 342), the Agency asserts that from 2025 to 2030, average electricity bills will decline.306 ) EPA
attributes this decline primarily to implementation of energy efficiency measures in each state
commensurate with its calculations under building block 4, coupled with the assumption (or
perhaps just a hope) that the nation will be awash in cheap natural gas for decades. EPA has
been far too optimistic and aggressive in its assumptions on both of these issues, as delineated
more completely in other parts of these comments including Section VII and Section XIV(D).
In addition, the Agency has not adequately considered, or has misjudged, a number of important
factors that will increase electricity prices to consumers as a result of implementing this
proposal. These include differences in wholesale electricity market structures and the cost of
producing, storing, and delivering natural gas as mentioned above, as well as: 1) the potential for
stranded costs arising from the forced retirement of generation units with remaining economical
useful life; 2) the associated potential negative impacts on credit ratings and borrowing costs;
and 3) price volatility in electricity and natural gas markets, such as that delineated in comments
APPA submitted to FERC regarding the Polar Vortex of 2014.307
305
http://www.nytimes.com/2014/06/26/us/politics/obama-warns-climate-campaign-cant-be-deaf-to-economicworries.html?_r=0
306
Regulatory Impact Analysis at 3-43
307
Winter 2013-2014 Operations and Market Performance in Regional Transmission Organizations and
Independent System Operators, FERC, Docket No. AD14-8-000, Mar. 19, 2014.
158
Perhaps most disconcerting is that, while EPA’s Proposed Rule would essentially require a
change in the dispatch from the current fleet of electric generating units to one mostly made up
of natural gas (meeting or approaching a 70 percent capacity factor nationwide) to meet a goal of
an 83 percent reduction in CO2 emissions by 2050, the agency did not conduct a full economic
analysis of these actions beyond the year 2030. This was noted and well stated by the House
Science Committee in its August 13, 2014, letter to Administrator Gina McCarthy:
Finally, EPA’s failure to model impacts between 2030 and 2040 is a
serious analytical shortcoming. The Administration has committed to
reduce emissions by 83% by 2050. As a result, reductions beyond 2030
must be analyzed to understand the implications of the approach. Given
the White House’s promises in this regard, the target reduction for the
power sector for 2040 should be modeled on a trajectory consistent with
the implied 2050 target.
The complete letter from the House Science Committee to Administrator McCarthy is found in
Attachments Section.
EPA’s use of questionable, overly aggressive assumptions on a number of critical elements,
coupled with inadequate or no consideration of other key cost factors, has led the Agency to
project minimal adverse economic impacts that are not realistic. This conclusion—a 30 percent
reduction in CO2 emissions and consumers’ electricity bills remain essentially flat or go down at
the same time—recalls the adage “if it sounds too good to be true, it probably isn’t true.”
B.
Electricity Prices Continue to Rise Generally
There is currently a broad disparity of electricity prices in the United States, as depicted in the
map below. This disparity ranges from a low of 6.35 cents/kwh in Idaho to a high of 24.13
cents/kwh in Hawaii.
159
Figure 23: Map of Residential Average Price
Source: http://www.electricchoice.com/images/map.jpg
The cost of producing and delivering electricity has been increasing, and as noted above, is
projected to continue to do so even if EPA does not implement this Proposal. There are
numerous reasons for this, including increases in the cost of building materials and fuel and the
cost of complying with various local, state, and federal mandates and regulatory requirements.
The chart below from the EIA shows these increases for residential customers. As the chart
further indicates, EIA predicts that costs to residential customers will continue to rise.
Figure 24: U.S. Residential Electricity Price
Source: Energy Information Administration
160
This increase is projected despite the deployment over the past decades of energy efficiency
programs that help offset some of the increase on customers’ bills. For public power utilities, all
of these costs are passed on to consumers in the form of monthly bills.
C.
Costs in Regions with RTO Markets
To determine the costs of implementation of the rule, EPA uses in its RIA, the IPM, developed
by ICF Consulting, Inc. According to EPA, the IPM accounts for different regulatory structures
and “projects changes in regional wholesale power prices and capacity payments related to
imposition of the represented policy that are combined with EIA regional transmission and
distribution costs to calculate changes to regional retail prices.”
As described in the IPM documentation, IPM “models production activity in wholesale electric
markets on the premise that these markets subscribe to all assumptions of perfect competition.
The model does not explicitly capture any market imperfections such as market power,
transaction costs, informational asymmetry or uncertainty.”308
Yet the RTO markets are far from perfectly competitive markets. Prices and actual costs often
diverge and at times, such as during periods of constrained supply, the differential can be
significant. For example, the Market Monitor for PJM found “ in the first six months of 2014,
11.4 percent of units had average dollar markups greater than or equal to $150 [per MWh]” and
concluded that for the first half of 2014, “the behavior of some participants during the high
demand periods in January raises concerns about economic withholding. Given the structure of
the Energy Market, the tighter markets and the change in some participants’ behavior are sources
of concern in the Energy Market.” 309
Concerns about market power are more significant in the RTO-operated capacity markets,
discussed in greater detail in Section XXV. FERC Commissioners Tony Clark and Norman Bay
found that regarding the February 2014 auction for generating capacity in ISO-NE, “there is
evidence suggesting the exercise of market power, and it is uncontroverted that the market
power, if it existed, was not mitigated. In the words of ISO-NE, prices resulted from a ‘noncompetitive auction.’ To the extent any portion of those prices was attributable to an exercise of
308
“Documentation for EPA Base Case v.5.13 Using the Integrated Planning Model, Nov. 2013,” Chapter 2:
Modeling Framework, http://www.epa.gov/powersectormodeling/docs/v513/Chapter_2.pdf
309
Monitoring Analytics, 2014 Quarterly State of the Market Report for PJM: January – June, pages 55 and 59
http://www.monitoringanalytics.com/reports/PJM_State_of_the_Market/2014/2014q2-som-pjm-sec3.pdf
161
market power, the auction will have imposed unwarranted costs upon consumers.”310 Such
concerns about market power are due to the withdrawal immediately prior to the auction of the
Brayton Point generating facility, a 1,544 MW coal plant owned by Energy Capital Partners
(ECP). This withdrawal dramatically lowered supply and increased prices in the auction. ECP’s
affiliate had earlier rejected an offer from ISO-NE to compensate Brayton’s costs of staying in
service because the ISO found that the plant was needed for reliability. The price increase
resulting from the retirement of the plant led ECP’s revenues from other plants it owns to
increase by $77 million.311
In addition to these market imperfections, the costs of implementing the Proposal within the
RTO markets will be exacerbated by frequent changes in market rules—often to the advantage of
the merchant generators (see the RTO governance discussion in Section XXII B). Merchant
generation owners have a financial interest in constraining the supply of resources as a means to
keep prices high. For example, the tightening of Minimum Offer Price Rules (MOPR) and the
removal of the state and self-supply exemptions in the PJM region, were a direct result of a
generator complaint in response to state-initiated efforts to procure new, more efficient naturalgas fired units. (See Capacity Markets fact sheet in Attachment 5) It is likely that as new, low or
non-CO2 emitting resources attempt to enter the markets, there is a potential for additional rule
changes to impede such entry.
As noted in the Navigant Paper on page 16 (see Attachment 3), a result of significant reductions
in CO2 is likely to be that “the effects of energy efficiency improvements on capacity prices will
likely be substantial, depressing prices for many years, and calls for further changes to capacity
markets to reduce this effect might be expected.” To the extent that merchant generators are able
to erect additional barriers to new construction of natural gas generation, renewable resources,
and energy efficiency, so too will the capacity and energy market costs increase beyond what
was modeled by the IPM.
310
Joint statement of Commissioners’ Clark and Bay on ISO-New England’s Forward Capacity Market Case, Sept.
16, 2014, Docket ER14-1409, http://www.ferc.gov/media/statements-speeches/clark/2014/09-16-14clark.asp#.VF0bPsm5QfE
311
For more details on the auction see Joint Motion to Intervene, Protest, and Requests for Evidentiary Hearing,
Investigation and Waiver of Eastern Massachusetts Consumer-Owned Systems, Docket No. ER14-1409-000, April
14, 2014, http://elibrary.ferc.gov/idmws/common/OpenNat.asp?fileID=13514045
162
D.
Retail Electricity Prices Are Rising at a Faster Rate in States Within
RTO Markets.
The discussion in the two preceding sections illustrates in part why wholesale electricity prices
are rising at an even faster rate in regions of the country where the wholesale markets are
administered by RTOs subject to jurisdiction of FERC. With the exception of Montana, Iowa,
Minnesota, Nebraska, North Dakota, and South Dakota, these RTO markets also encompass the
states that chose to restructure their retail electricity markets, and in so doing, generally ceded
authority over their state’s generating facilities to FERC. This lack of direct state authority over
electricity generation substantially dilutes the ability of states, through their public service
commissions, for example, to protect consumers from excessive power prices.
The largest component, by far, of a customer’s electricity bill is the cost of the power itself.
Thus, increases in the wholesale cost of power are reflected in the monthly retail bills paid by
consumers. The chart below shows the national average of retail rates for all states from 1997
(roughly the beginning of electricity market restructuring) and 2013, the most recent year for
which this data is available. The national average is contrasted with the average of states that
have restructured their retail electricity markets and those that have not. The chart shows that,
again, while electricity rates are increasing everywhere, they are rising at a faster rate in states
that have restructured their retail electricity markets and are more dependent on RTO markets for
wholesale power supply.
163
Figure 25: Average Rates: Deregulated vs Regulated States
E.
January 2014 Polar Vortex Gas and Electric Price Spikes
The extreme cold weather events of January 2014 provide an illustration of the potential for
dramatic price increases in the RTO regions during times of scarcity. There were three major
cold events this past January—on January 6-7, January 22, and January 27—and one major event
on February 6. During this time frame, natural gas prices spiked, reaching as high as
$100/MMBtu during some trading periods in the Northeast and Mid-Atlantic regions.
Meanwhile, the high peak demand combined with high levels of generation outages, placed the
eastern RTO regions (PJM, ISO-NE, and NYISO) near their limits of available capacity to meet
system demand. The combination of high gas prices, constrained generation, and PJM’s
implementation of shortage pricing produced average real-time electricity prices that ranged
164
from $300 to $700 per megawatt-hour in PJM, NYISO, ISO-NE, and MISO. As a result of
shortage pricing in PJM, prices reached as high as $2,000 per MWh. 312
While it may appear on the surface that the electricity prices were simply following natural gas
prices, there are indicators that generators built in an extra layer of profits on top of the high
natural gas costs. A primary indicator of the profitability of natural gas plants is “spark spreads,”
which are measures of the differential between the electricity price and the cost of electricity
produced by a natural gas plant given natural gas prices at the time, assuming a heat rate of 7,000
Btu per kilowatt-hour. EIA reported spark spreads on January 6, 2014, of $61.43 per MWh in
ISO-NE at the Massachusetts Hub, $49.06 in New York City and $48.09 in the PJM Western
Hub.313 These spreads ranged from 35 to 39 percent of the day-ahead electricity price, which
was between $125 and $173 per MWh, meaning that over one-third of the price of electricity
was earned in profits by an efficient natural gas plant. Other units may have earned higher or
lower spreads, depending upon their fuel costs and heat rates.
Moreover, a review of earnings reports shows higher levels of profits during the winter for some
merchant generators. For example, PSEG Power reported that its gross margin (revenue net of
expenses from the sale of power) increased by $51 million in PJM and $18 million in ISO-NE
between the first quarters of 2013 and 2014, showing that the company was earning significantly
more from higher electricity prices than incurred in higher fuel costs. 314 Calpine’s commodity
margin in the North (equal to all revenue earned from power sales net of expenses) totaled $261
million in the first quarter of 2014, a 280 percent increase from the first quarter of 2013.315
As mentioned above, also contributing to the high energy prices is the fact that PJM invoked
shortage pricing several times in January, allowing emergency demand response and emergency
imports to set prices at $1,800 per MWh.316 This cap will increase to $2,700 by June 1, 2015,
312
See FERC Staff Presentation, Winter 2013-2014 Operations and Market Performance in RTOs and ISOs, April 1,
2014, http://www.ferc.gov/CalendarFiles/20140402102127-4-1-14-staff-presentationv2.pdf
313
“Today in Energy,” EIA, as posted on January 6, 2014 http://www.eia.gov/todayinenergy/prices.cfm. Prices are
posted on a daily basis.
314
“PSEG Earnings Conference Call, 1st Quarter 2014,” May 1, 2014, at 16,
http://investor.pseg.com/sites/pseg.investorhq.businesswire.com/files/doc_library/file/1Q_2014_Earnings_Slides__May_1_2014.pdf and “PSEG Earnings Conference Call, 1st Quarter 2013,” April 30, 2013 at 15,
http://phx.corporateir.net/External.File?item=UGFyZW50SUQ9MTgyOTUyfENoaWxkSUQ9LTF8VHlwZT0z&t=1
315
Calpine Presentation, “Calpine First Quarter 2014 Investor Update Conference Call, May 1 2014,” at 26,
http://phx.corporate-ir.net/External.File?item=UGFyZW50SUQ9MjI3Nzk3fENoaWxkSUQ9LTF8VHlwZT0z&t=1.
316
“Cold Weather Operations for January 2014, Questions, Comments, and Responses,” PJM Interconnection,
March 6, 2014, http://www.pjm.com/~/media/documents/reports/20140306-january-2014-cold-weatherquestions.ashx.
165
under PJM’s phase-in of shortage pricing,317 meaning that these higher prices will coincide with
the increasing level of coal retirements. As coal plants retire, the probability of electricity price
increases and potential supply disruptions during periods of cold weather and high natural gas
prices is likely to increase.
F.
Increases in Electricity Prices Disproportionately Impact Low and Fixed
Income Consumers.
According to consumer advocacy groups, such as AARP, electricity prices and increases in
prices have a disproportionate impact on low and fixed income consumers. 318 While this is
rather intuitive conceptually, the actual impacts can be dramatic. In the 2011 National Energy
Assistance Survey319 conducted by the National Energy Assistance Directors Association, results
showed, to pay energy bills, 24 percent of Low Income Home Energy Assistance Program
(LIHEAP) recipients went without food, 37 percent went without medical or dental care, and 34
percent did not fill or took less than the full dose of a prescription medicine.
In addition, the LIHEAP Home Energy Notebook for FY 2009 320 reported some cautionary
results from a survey of LIHEAP and low-income households in the Northeast—a region already
heavily dependent on natural gas and straining to keep up. Low-income households in that
region that heat with electricity pay 12.9 to 17.2 percent of their income on energy compared to
non-low income households that pay about 3 percent of their income on energy. 321
G.
Increases and Volatility in the Cost of Natural Gas Flow Directly and
Automatically to Consumers.
Section VII of these comments contains extensive discussion of the potential supply of natural
gas for electricity generation and relevant factors that both influence and create uncertainty about
the price of that gas. A large part of EPA’s assumption that electricity price increases will be
317
See “Shortage Pricing FAQs,” p. 12, PJM Interconnection, July 12, 2012 http://www.pjm.com/~/media/marketsops/energy/shortage-pricing/shortage-pricing-faqs.ashx.
318
Comments of AARP, Public Citizen and the National Consumer Law Center, On Behalf Of Its Low-Income
Clients, Centralized Capacity Markets in Regional Transmission Organizations and Independent System Operators,
Federal Energy Regulatory Commission, Docket No. AD13-7-000,
http://elibrary.ferc.gov/idmws/common/OpenNat.asp?fileID=13434054. See p. 3, where the parties not that the cost
of FERC jurisdictional wholesale purchases “ is flowed through to retail consumers, including AARP members and
low-income households that are face dire choices in order to afford essential energy services.”
319
Available at neada.org/wp-content/uploads/2013/05/NEA_Survey_Nov11.pdf.
320
Available at http://www.acf.hhs.gov/programs/ocs/resource/liheap-2009-home-energy-notebook.
321
Ibid at Tables A-3b and A-4
166
only minimal stem from the Agency’s belief that either sufficient supplies of gas will be
available for the decades it will be needed or that the price of gas will remain at or near current
levels. APPA is not as sanguine. The Agency is making huge assumptions based on selective
data.
Fuels used to generate electricity, especially fossil fuels, are often subject to significant price
volatility. As noted in Section VII, natural gas historically has been highly subject to such
volatility. Moreover, the cost of fuel is one of the largest components of the total cost of fossil
fuel-generated power supply, which in turn is the largest component of a customer’s electricity
bill. Recognizing these factors, it is common and accepted practice among regulators to allow
utilities to pass these fluctuating fuel costs directly on to their customers in the form of a “f uel
adjustment charge.” This mechanism allows the utility to directly recover its costs without
having to obtain specific regulatory approval for what can be substantial increases in a
customer’s bill. These charges are typically adjusted on a quarterly basis to reflect the utility’s
actual fuel costs for the preceding quarter. Thus, even relatively small changes in excess of
EPA’s projections for the price of natural gas will have immediate impacts on consumers that
could eviscerate its conclusion that price increases will be only modest.
H.
Remaining Useful Life of the Facility
It is generally expected that compliance with the emission goals contained in the Proposal will
likely cause the premature retirement of a significant number of U.S. coal-fired generating
plants. By the term premature, APPA means that the plants will be shut down before the end of
their useful lives. Certain stakeholders (e.g., electric utilities and their customers and local
communities) will suffer economic losses as a result. These losses include foregone economic
value because the facility is no longer allowed to operate, as well as paying certain costs now
“stranded” because the facility no longer produces the revenue to cover those costs. The notion
that economic losses will result from premature retirement is inescapable because useful life is
properly thought of in terms of an asset’s ability to yield on-going economic value. Hence,
shutting down an electric generating facility prior to the end of its useful life is sure to result in
losses for someone.
Moreover, as discussed in Section III(B)(9), CAA Section 111(d)(1)(B), permits states to take
into consideration, “among other factors, the remaining useful life of the source” in developing
their state compliance plan. However, in its construction of the building blocks and the resulting
calculation of interim and final emission reduction goals, EPA has substantially encroached on
this state discretion, resulting in losses to the remaining useful life of generation sources and
increased costs to consumers.
167
Like any long-lived asset, electric generating plants require long-term financing. Since public
power utilities are non-profit, their assets are financed primarily with long-term debt, while
investor-owned utilities and merchant suppliers generally finance projects with some
combination of long-term debt and equity. It is sometimes mistakenly thought that an asset’s
useful life is directly related to the debt repayment or servicing schedule associated with its
financing. This in turn might lead some to conclude that retirement of an asset after the debt has
been repaid somehow mitigates the economic losses attendant with retirement. But this is not the
case.
There is no single, unambiguous way to precisely measure the useful life of an asset, but the
underlying notion, as presented in the following definitions and descriptions, is that useful life
represents the period over which an asset is expected to provide value to its owner.
Useful life usually refers to the duration for which the item will be
useful (to the business), and not how long the property will
actually last. Many factors affect a property's useful life, including
the frequency of use, the age when acquired and the repair policy
and environmental conditions of the business. The useful life for
identical types of property will differ from user to user depending
on the above factors, as well as additional factors such as
foreseeable technological improvements, economic changes, and
changes in laws.322
Useful life is “the period of time that the asset can reasonably be expected to operate in the
manner and at the level of efficiency intended,” and “an asset’s useful or productive life is the
period during which the present value of the cash inflows expected to be derived from the asset’s
use (that is, its productive value) exceeds the assets abandonment value.”323 Moreover,
The accounting assumption as to the useful lives of assets should
be based on economic and engineering studies, on experience, and
on any other available information about an asset’s physical and
economic properties.324
322
http://www.investopedia.com/terms/u/usefullife.asp
Clark, John J., Thomas J. Hindelang and Robert E. Pritchard. Capital Budgeting: Planning and Control of Capital
Expenditures. 3d ed. Englewood Cliffs, NJ: Prentice Hall, 1989
324
Leopold Bernstein and John Wild. Analysis of Financial Statements. 5th ed. McGraw-Hill, 1999
323
168
Clearly, useful life is an economic concept with no necessary relationship to the period of debt
service. Since total debt incurred represents a sunk cost that must be paid whether or not the
asset remains useful, the loss of economic value resulting from the premature retirement of an
asset is unaffected by the debt service schedule.325 Rather, useful life pertains to the time
horizon over which an asset is expected to provide economic value to the owner.
In the case of public power electric utilities, the generating assets are, in effect, owned by the
utilities’ customers, with the government entity acting as their agent. While the utility transacts
the purchase and is the legal owner, the economics of ownership, both positive and negative,
redound to the customers. Both the costs and the benefits of asset ownership are conveyed to
customers through their electric rates. For example, if a utility purchases a coal plant, the costs
of ownership (purchase price, financing cost, and on-going operation and maintenance) will be
directly passed through to the electric rates, but if the plant is economic, electric rates will, over
the life of the plant, be lower than they would have been if the utility had met the customer’s ongoing power needs through market purchases or acquisition of alternative resources.
So, customers bear the economic costs of ownership, but also realize the economic benefits. As
long as the going-forward variable costs of producing electricity with the coal plant are less than
the all-in costs (fixed and variable) of viable replacements (market purchases, or resource
acquisitions), the plant will yield economic benefits for customers. If the plant ceases operation
during its economically useful life, customers will suffer an economic loss from electric rates
that will be higher than they would have been it the plant continued to operate. And, as
discussed earlier, the loss borne by customers will be same irrespective of the debt service
schedule.
Under the traditional utility rate making framework, customers also bear the risks associated with
plant ownership. Obviously, at the time a capital investment is made, it is expected that the longterm benefits will exceed the costs of owning and operating the plant. But, there is always much
uncertainty. The benefit streams often extend far into the future, and the final results will be
325
To illustrate, assume an asset that costs $100. The asset lasts five years and produces revenue of $100 per year,
so total, nominal (i.e., undiscounted) revenues, before debt service, amount to $500. Assume that an investor
borrows $100 and then pays it off immediately. His net proceeds will be $400. If he finances the purchase with
debt service of $20 per year, his net proceeds will still be $400. (Note this example abstracts from carrying costs and
discounting, because they would complicate the example with affecting the essential conclusion). If the useful life is
truncated at the end of three years, the investor who paid off the debt immediately would lose $200 of revenues and
his total net proceeds would be $200. When the debt is serviced over five years, the investor must pay off the $100
debt even though the revenue stream is truncated. His total revenues will be $300, his debt service will be $100, his
net proceeds from the project will be $200, and his economic loss will be $200. The economic loss is the same
whether or not the debt is paid off when the asset is retired.
169
affected by a variety of volatile factors. In some cases, the total benefits will exceed total costs
and customers realize net economic gains, while in other cases, benefits will fall short of costs
and customers suffer losses. In either case, artificial interruption of the benefit stream during a
coal plant’s useful life will harm customers by causing losses that would otherwise not have
occurred.
On average, across the sector, approximately 40 percent of total electric output is produced from
coal, with much higher percentages in some situations. This clearly constitutes a significant
investment for these public power utilities and their customers, undertaken in anticipation of
realizing long-term benefits over the useful lives of the coal plants. It also means that coal is an
important resource for public power.326 In many cases, unscheduled retirements over the next
few years, related to the Proposed Rule, will truncate the expected benefit streams while the
plants are still economically useful and thus result in significant economic losses to public power
customers.
As discussed above, the useful lives of individual coal facilities will depend on a variety of
factors, some of which may be unique to specific facilities. Thus, it is difficult to offer a value
that represents the remaining useful life of the combined public power coal fleet, but clearly the
general expectation is that coal plants have very long useful lives. A 2007 survey of 10 state
utility commissions conducted by the Wyoming Office of Consumer Advocate found that
respondents generally expected to see useful lives for coal plants in the range of 40 to 60
years.327 Similarly, data show that the average number of years between the in-service and
expected retirement dates for coal plants owned by public power utilities is 35 years. Thus,
public power electric customers will likely suffer significant economic losses from the premature
retirement of coal facilities if the Proposed Rule is finalized.
These early retirements will lead to higher electric prices related to replacement power costs,
new capital investments, plant decommissioning costs, and possible credit downgrades. As
discussed further in Subsection J below, APPA has conducted some preliminary analysis to
estimate the range of economic losses, and associated rate increases, faced by public power
customers as a result of possible premature retirements of coal plans under the Proposed Rule.
Our analysis indicates that the total, cumulative losses could range from about $10 billion to
326
In some cases the commitment takes the form of long-term purchase power agreements (PPA) as opposed to
actual capital investments, but the PPA’s are often take-or-pay contracts that impose debt like obligations on the
utilities and thus customers.
327
Wyoming Office of Consumer Advocate. Depreciable Life of New Coal Generating Plant. September 2007
170
about $284 billion, with associated electric rate increases ranging from about 4 to 32 percent.328
These values represent an indicative range of possible outcomes for the composite public power
sector.
Many factors will affect the results including: the baseline level of electric rates before the coal
plant retirements; long-term natural gas prices; installed costs of replacement generation
resources (e.g., NGCCs and gas pipeline and storage facilities); variable costs and operating
characteristics of replacement facilities (including energy efficiency and increased utilization of
existing gas plants); the number of coal plants retired and associated electric output; long-term
coal prices; variable costs and operating characteristics of the coal plants subject to retirement;
and the time horizon for the analysis, which reflects the remaining useful lives of the affected
coal plants. Thus, one cannot predict precise outcomes with any degree of confidence. In
APPA’s view, the wide range of possible outcomes is cause for concern. It is not clear what the
final impacts will be, but under some plausible assumptions, they could be quite severe. Thus,
the Proposal should be modified as recommended by APPA in these comments. Those changes
would allow states to truly and accurately incorporate their own determinations of the remaining
useful life of affected facilities into the development and implementation of their compliance
plans.
I.
Stranded and Replacement Costs
As discussed above, compliance with the emission goals contained in the Proposed Rule will
cause the retirement of a significant number of U.S. coal-fired generating plants and—
potentially—other power generating facilities. In so far as these facilities are retired while they
are still able to yield on-going economic value, their retirement will impose an economic cost on
the owner-generator and, in turn, its customers.
The term “stranded cost” was developed in the context of the restructuring of the natural gas
pipeline and then electric utility sector in the 1980s. In the context of electric utility
deregulation, generally the term has been used to refer to a cost that an electric utility is
permitted to recover through its rates, but whose recovery may be impeded or prevented by the
advent of competition in the industry. 329 Amidst this restructuring, FERC, in regulating the
recovery of a stranded cost, further clarified that such a cost must be a “legitimate, prudent, and
328
Dollar values are cumulative present values over the analysis period and rates are levelized over the analysis
period .
329
William J. Baumol and J. Gregory Sidak, Transmission Pricing and Stranded Costs in the Electric Power
Industry, Washington: AEI Press, 1995, 98.
171
verifiable cost.”330 Clearly, an electric utility’s inability to recover the “legitimate, prudent, and
verifiable cost” of generation facility investments which the Proposed Rule would force the
premature retirement of should be considered a stranded cost.
In their individual comments, APPA members will be providing examples of stranded costs
highly likely to result from the Proposed Rule. APPA urges the EPA to examine and consider
these examples carefully. The comments of SRP with respect to the Navajo Generating Station,
for example, are particularly instructive. Another example is LRS in Wyoming, partly owned by
APPA members that serves consumers in several states. If the owners of LRS are forced to retire
a unit or the entire plant, this will result in significant stranded investment. LRS has a gross
book value of about $1.23 billion, and the public power utilities,’ including Missouri River
Energy Services (MRES), share is about $202 million. Currently, LRS is under a mandate from
EPA under the Regional Haze Rule to install Selective Catalytic Reduction (SCRs) on all three
units at LRS. This will come at a cost of $750 million to the project owners. For MRES, its
share of this cost will be approximately $125 million, which will cause an increase in wholesale
rates of 10 percent. These investments must be made by 2019,331 a year before the start of the
interim compliance period under the Proposed Rule. If a unit is forced to retire, $250 million of
consumer-owned investment to meet Regional Haze rules will be stranded, on top of the
economic value of the remaining useful life of the unit. Those costs will no longer be spread out
over the 20 year remaining life of LRS, but must nonetheless be recovered from consumers, a
cost for which they will no longer receive any value.
Again, as noted above, these stranded costs would exist whether the cost of financing a
generating facility has been paid or is still being paid.332 Nonetheless, the existence of stranded
cost is most clearly demonstrated when a facility has been financed with long-term debt,
intended to roughly match the useful life of the facility, that is still being repaid.
As state- or locally-owned, not-for-profit entities, public power utilities are limited in how they
raise funds for long-term infrastructure investments of all sorts, including electric power
generation facilities and facility modifications. They cannot allow partners to “buy” into the
business and cannot issue additional stock to equity shareholders. Also, generally, they do not
amass large cash reserves. Instead, generally public power utilities raise these funds with longterm debt in the form of municipal bonds.
330
18 C.F.R. § 35.26.
This date is subject to change, based on the stay granted by the 10th Circuit in the litigation over the Regional
Haze Federal Implementation Plan. See note 1.
332
By way of analogy, the theft of a car is a loss to the owner whether the car was bought with cash, financed with a
loan that has been paid off, or financed with a loan that is still being repaid.
331
172
Municipal bonds are unique in that they tend to have maturities nearly twice as long as corporate
bonds333 and generally are issued as a series of bonds with varying maturities, rather than a single
maturity. Financing a project with bonds that mature incrementally over a long period of time
allows public power utilities to build projects with capital provided upfront by bond investors,
but repaid over the projects’ useful life by the citizens and customers benefitting from the
project.
Municipal bonds are the largest source of financing for core infrastructure in the U.S.334 and are
the single most important financing tool for public power, given the capital-intensive and longlived nature of assets needed by the electric industry. Each year, on average, public power
utilities make $10 billion in new investments financed with municipal bonds.335 Public power
utilities use municipal bonds to finance investments in power generation (including through
renewable and alternative fuels), transmission, distribution, reliability, demand control,
efficiency, and emissions controls. While the typical power-related bond issue is relatively
small, issuances financing electric generation or transmission projects generally total hundreds of
millions or even billions of dollars and can have maturities as long as 50 years.
For example, in 2007, the Northern Illinois Municipal Power Agency (NIMPA) issued a series of
bonds totaling $318,715,000 with maturities ranging from 6 to 35 years (i.e., maturing from 2013
to 2042) to finance a portion of the Prairie State Project, an approximately 1,600 MW coal-fired
generating station, coal reserves adjacent to the plant site, and coal mining facilities. According
to NIMPA’s auditor, the lives of the bonds do not exceed the project’s useful life.336 As a result,
if the Prairie State Project is shuttered by 2020 because of the Proposed Rule, NIMPA’s
customers will be forced to pay for the cost of financing a new source of power and power
generation, while also repaying the debt associated with the project through 2042.
333
Securities Industry and Financial Marketers Association, US Municipal Issuances,
(http://www.sifma.org/uploadedFiles/Research/Statistics/StatisticsFiles/Municipal-US-Municipal-IssuanceSIFMA.xls?n=04049) (last visited on Aug. 28. 2014) ; Securities Industry and Financial Marketers Association, US
Corporate Bond Issuances (http://www.sifma.org/uploadedFiles/Research/Statistics/StatisticsFiles/CM-US-BondMarket-SIFMA.xls?n=69088) (last visited on Aug. 28. 2014).
334
Cong. Budget Office, J. Comm. on Taxation “Subsidizing Infrastructure Investment with Tax-Preferred Bonds”
(Oct. 2009)(showing that for education, water, and sewer, nearly all capital investments are made by state and local
governments and that for transportation most investments are made by state and local governments).
335
The Bond Buyer & Thomson Reuters “2014 Yearbook” (2014) 150; The Bond Buyer & Thomson Reuters “2009
Yearbook” (2009) 170.
336
Northern Illinois Municipal Power Agency, Audited Financial Statements, 2012 (April 30, 2013)
(http://www.nimpa.us/index.php?option=com_dropbox&view=dropbox&Itemid=64&format=raw&task=download
&mime_type=application%2Fpdf&sub_folder=&file=2012NIMPAAuditedFinancialStatements.pdf) 10.
173
Again, in the likely event that the Prairie State Project’s actual useful life would extend beyond
the term of debt, i.e., 2042, early retirement of the project would impose additional and ongoing
costs on NIMPA customers and other customers of investors in the Prairie State Project.
Another reason EPA should finalize a rule with an explicit opportunity for states to set up
subcategories or alternative regulatory compliance options for public power or municipal
utilities is because municipal bonds tend to have maturities nearly twice as long as
corporate bonds. Thus, the consequences are more significant than for most investor-owned
utilities or merchant power plants. Section 111(d) expressly allows for state consideration of
such factors.
EPA must allow states the flexibility to adjust both the emission reduction goals and
implementation dates to assure that the requirements on the state reflect a BSER that assures the
reliability and security of the electric system while providing reasonably priced electricity to the
consumer.
J.
The Proposed Rule Will Impact Electricity Rates, Pushing Them Higher
Than They Are Today.
The premature shutdown of existing coal plants with a remaining economic life will likely cause
higher electric rates and increased costs for the nation’s electricity consumers than EPA has
projected. Tables 13 and 14 show potential cost increases and rate impacts for customers for the
public power sector, taken as a whole, under varying assumptions for the key factors that will
affect electricity prices when coal plants are prematurely retired. The tables were developed by
APPA using the data and methods described below.
Table 19: NPV Cost Impacts*
Gas Price **
Low
Mid
High
S1
S2
S3
Low
5 Years
$ Billions
$6.99
$8.8
$77.45
Mid
20 Years
$ Billions
$74.89
$93.33
$148.67
High
40 Years
$ Billions
$259.46
$302.07
$4129.88
* Table developed by APPA, see text below and relevant appendix.
** First year gas prices for Low Mid and High cases $6/MMBtu, $8/MMMBtu and $10/MMBtu,
respectively. All prices escalate at 2% per year over respective scenario horizons.
174
Table 20: Levelized Rate Impacts*
Gas Price**
Low
Mid
High
S1
S2
S3
Low
5 Years
%
5.475
6.24%
8.54%
Mid
20 Years
%
13.02%
15.72%
23.71%
High
40 Years
%
36.57%
41.33%
55.4%
* Table developed by APPA. See text below and relevant appendix.
* First year gas prices for Low Mid and High cases $4/MMBtu, $5/MMMBtu and $8/MMBtu,
respectively. All prices escalate at 5% per year over respective scenario horizons. Prices and escalation rate
based on EIA 2014 Annual Energy Outlook
A more detailed explanation of the data and calculations underlying the table follows the
discussion of the results, but a brief overview to guide interpretation is in order. To develop
what we believe are the plausible ranges shown on the tables above, we constructed three
scenarios (S1, S2, and S3) each reflecting a particular combination of key input variables, not
including gas prices, that would yield low, mid, and high cost and rate impacts, respectively. We
then estimated the cost and rate impacts for each scenario under low, mid, and high gas price
assumptions. This creates the 3 by 3(3X3) matrices of possible outcomes shown on the tables.
Each scenario depicts a different time horizon based on different assumptions regarding the
remaining useful lives of coal plants subject to closure.
The cost impacts shown on Table 19, represent the present value, incremental cost borne by
customers from premature plant closures, while the results on Table 20 show the impacts on
electric rates in terms of the percent change in levelized rates resulting from reduced coal
production. The tables show a large range of potential outcomes because there are a number of
factors that can influence the results, all of which are subject to variability and uncertainty. Each
factor can vary, in a correlated or uncorrelated manner, with the others, and this implies a very
large set of possible input combinations, which in turn yields a very large set of possible rate
impacts.
175
As the tables show, the NPV cost impacts range from about $7 billion to almost $430 billion,
with337 associated rate increases of 5.47 to 55.4 percent. This wide range of uncertain outcomes
does not imply that the potential rate and cost impacts are fundamentally unknowable and thus
not helpful for policy guidance. While many outcomes are possible, some are more likely than
others under any specific set of circumstances. By design, the Proposed Rule will give rise to a
variety of different compliance strategies based on disparate circumstances across, and within,
the states. Consideration of the likely impacts in particular situations should inform the
development and implementation of compliance strategies.
There are a number of underlying factors that can influence the results, including, but not limited
to: prices for coal, natural gas, energy efficiency, and renewable resources; remaining useful
lives of the affected coal plants; shares of electric output produced by coal and other resources;
capital cost for replacement resources and associated infrastructure; and composition of the
replacement energy resource portfolio.
It is fairly clear what the directional impact (e.g., higher or lower) on a utility’s total cost of
service, and hence rates, would be for a given change in a particular variable while holding the
other factors constant. For example, all else constant, higher coal prices would imply less of a
cost impact from coal plant retirement than would lower process. As another example, the cost
impacts will be larger for retirement of plants with longer projected economic lives than for
shorter-lived facilities.
We used this reasoning to construct low (S1), mid (S2), and high (S3) scenarios based on the five
crucially important input variables: coal prices, remaining economic life, share of coal in the
supply resource portfolio, amount of new infrastructure need to support new gas plant capacity,
and composition of the replacement portfolio. As noted, we then estimated cost and rate impacts
for each scenario under varying assumptions for natural gas prices. A more detailed description
of the scenario inputs and analytical methods is presented in appendix XX.
This analysis is intended to convey a quantitative sense of how premature retirement of coalfired generators might affect a utility’s retail rates and total cost of service to utility customers
under varying conditions and circumstances. We are not suggesting that any particular outcomes
337
Among other factors affecting the results, the rate impacts are influenced by firm size. This analysis reflects
impacts on the composite public utility sector and it is unlikely that the high and low scenarios would pertain to the
entire public power sector. Thus, it is best to think of the high and low scenarios results in terms of impacts on subsectors, perhaps individual utilities in some cases, within the public power sector. While the levelized rate impacts
shown in Table 14 would pertain to subsectors, the NPV dollar amounts shown in Table 13 would have to be scaled
down when considering subsectors.
176
are acceptable or not, but we are suggesting that analysis of potential cost and rate impacts
should inform policies and implementation strategies. Pursuit of any goal, environmental or
other, will usual involve trade-offs of some kind. The customer cost and rate impacts that might
result from pursuing the goals of the Clean Power Plan (CPP) represent trade-offs that should be
explicitly acknowledged so that policy makers can better balance the interests of affected parties
in their communities.
K.
Potential Impacts on Credit Ratings Could Raise Borrowing Costs for
Public Power Utilities.
EPA rules that result in the premature retirement of existing coal plants can lead to higher utility
rates via direct replacement power costs and through higher borrowing costs if the retirements
have negative impacts on utility credit ratings. Credit ratings can be affected primarily in three
ways. First, the need for replacement power will likely lead to higher levels of debt, and/or other
fixed obligations, thus impairing key credit metrics, especially coverage and liquidity ratios, used
by rating agencies to develop their ratings. Second, rate increases that public power utilities
must implement to recover higher replacement costs can compound the effects of other factors
(e.g., macroeconomics, energy efficiency, net metering, and other distributed generation
initiatives) that lead to reduced sales and higher rates, thus impairing credit ratings by
contributing to the utility “death spiral.” 338 Third, a reduction in fuel diversity (by eliminating
coal as a portfolio option) exposes utilities to greater fuel price risks.
These types of concerns are evident in recent rating agency publications. For example, in its
2014 Outlook for public power utilities, Fitch Ratings offered a stable outlook for public power
utilities in 2014, but also highlighted its longer term concerns related to potential impacts of
environmental rules with the following caution:
Prospective regulations for existing plants are due to be proposed by EPA
in June 2014 could have a more profound impact, particularly on utilities
with a high concentration of coal-fired resources. If emission standards
are applied retroactively, compliance strategies could be extremely costly
or unfeasible, renewing concerns about the premature retirement of
productive generating assets, and significantly higher operating and debt
service costs for replacement capacity. While public power and
cooperative utilities would be expected to recover these higher costs from
338
Essentially, this is a “catch 22” situation whereby the remedy for deteriorating credit metrics, higher rates, might
lead to deteriorating metrics.
177
customers, the financial strain would likely result in weaker financial
metrics and flexibility, and downward rating pressure. 339
In the same report, Fitch expressed additional concern that EPA regulations, which effectively
limit the use of coal-fired resources, will also tend to limit fuel diversity in utility supply
portfolios. Currently, utilities that rely on a variety of fuel-source generation are less subject to
fuel price risk. A spike in the price of one fuel source can be mitigated by the utility by
generating more power from another fuel source. If fuel switching is not practical—for example
when load is close to generating capacity from all fuel sources, a fuel prices spike would still be
mitigated by the fact that not all of the utility’s power is generated from a single fuel source.
Conversely, homogenizing the power generation fleet will leave utilities less able to manage fuel
price risks. In so far as rating agencies believe an affected utility will be unwilling or unable to
pass on the costs of potential price spikes to customers, that utility’s credit rating will be
downgraded.
Similarly, Moody’s Investor Service recently indicated a stable outlook for public power utilities
in 2014, but it also identified potential CO2 regulation as a long-term credit risk.340 In its 2014
Outlook, Moody’s noted that:
An accelerated pace of carbon regulation and advances in technology that
threaten the utility industry’s monopoly on customers are long-term credit
risks.
The Moody’s report went on to say that:
Rate pressure could threaten the willingness of ratepayers to support stable
financial metrics, and this could put downward pressure on the stable
outlook. Stable natural gas prices are a moderating factor.
In addition to the major concerns the rating agencies cite, it is worth noting that various aspects
of centralized, RTO-run capacity markets can make it difficult to develop new capacity
resources, and this can complicate strategies for replacing existing coal-fired resources. These
problems have been observed and capacity markets are undergoing change, so now might not be
the best time to create the need for substantial amounts of new generating capacity.
339
340
Fitch Ratings 2014 Outlook: U.S. Public Power and Electric Cooperative Sector, Dec. 12, 2013.
Moody’s Investor Service, 2014 Outlook-US Public Power Electric Utilities, December 2014.
178
New environmental regulation that leads to premature retirement of existing coal-fired resources,
or significant increases in the costs to operate these facilities, will translate into higher costs to
utility customers through direct expenditures for replacement resources and potentially through
higher borrowing costs. EPA should be mindful of these impacts when formulating rules for
existing resources. Further, the Agency should leave the option to state environmental agencies
to allow for a subcategorization or alternative regulatory option to help public power utilities
from passing through these costs to the community’s consumers.
XXI.
The Proposal Raises Concerns About Reliability.
The reliability of the Nation’s Bulk Power System (BPS) needs to be preserved during the
implementation of the Section 111(d) final rule. APPA is concerned that the Proposed Rule will
not preserve such reliability unless significantly modified. The change in electric generation
resource mix required by the Proposal will need to be modeled so transmission system
modifications can be made to assure reliable operation of the grid. A reliability back-stop
mechanism to delay compliance with the final rule will be essential to preserve BPS reliability
when additional time is needed to complete construction of new electric generating units and
transmission systems.
In 2005, Congress, with the support of the electric utility industry and others, took steps to
strengthen the then-existing voluntary reliability program. The Energy Policy Act of 2005
(EPAct) amended the Federal Power Act by adding section 215, titled “Electric Reliability.”
Section 215 authorized the creation of the “Electric Reliability Organization” to establish and
enforce mandatory bulk-power reliability standards, subject to oversight by FERC. 341 Following
passage of this legislation, FERC certified the North American Electric Reliability Corporation
(NERC) as the “Electric Reliability Organization” under Section 215 of the FPA. 342 NERC
proposed a new set of reliability standards, which was approved by FERC and became
mandatory in 2007.343 In approving the standards, FERC explained that a “Reliability Standard
is a requirement approved by the Commission that is intended to provide for the Reliable
Operation of the Bulk-Power System. Such requirement may pertain to the operation of existing
Bulk-Power System facilities, including cybersecurity protection, or it may pertain to the design
341
See 16 U.S.C. § 824o (2012).
See Order Certifying North American Electric Reliability Corporation as the Electric Reliability Organization and
Ordering Compliance Filing, 116 FERC ¶ 61,062 (July 20, 2006).
343
See Mandatory Reliability Standards for the Bulk-Power System, Order No. 693, 118 FERC ¶ 61,218 (Mar. 16,
2007) (Order No. 693); Order on Rehearing, Order No. 693-A, 120 FERC ¶ 61,053 (July 19, 2007).
342
179
of planned additions or modifications to such facilities to the extent necessary to provide for
reliable operation of the Bulk-Power System.”344
The BPS consists of generating units, transmission lines (generally those 100 kV and above), and
substations and controls. These facilities operate as an interstate grid subject to exclusive FERC
regulation for the purpose of ensuring BPS reliability and do not include facilities used in the
local distribution of electric energy, which remain within state jurisdiction.345
Key drivers for infrastructure investment are system models to assure operating of the BPS
within the mandatory limits required by NERC reliability standards. Since new transmission and
generation projects take many years to plan and construct, NERC, industry, and government
agencies need to work together to assess the long-term reliability of the BPS. The environmental
objectives of the Proposed Rule will need to be incorporated into the utility transmission and
generation planning process to assure continued BPS reliability. NERC points out the need for
detailed analysis of any changes to the BPS in its Initial Reliability Review346 of the Section
111(d) Proposed Rule. NERC states:
The preliminary review of the proposed rule, assumptions, and transition
identified that detailed and thorough analysis will be required to
demonstrate that the proposed rule and assumptions are feasible and can
be resolved consistent with the requirements of BPS reliability. This
assessment provides the foundation for the range of reliability analyses
and evaluations that are required by the ERO, [Regional Transmission
Operators] RTOs, utilities, and federal and state policy makers to
understand the extent of the potential impact. Together, industry
stakeholders and regulators will need to develop an approach that
accommodates the time required for infrastructure deployments, market
enhancements, and reliability needs if the environmental objectives of the
proposed rule are to be achieved. 347
344
See Order No. 693, 118 FERC ¶ 61,218, P 23.
See 16 U.S.C. § 824o (a)(1); 18 C.F.R. § 39.1 (2014) (“Bulk-Power System means facilities and control systems
necessary for operating an interconnected electric energy transmission network (or any portion thereof), and electric
energy from generating facilities needed to maintain transmission system reliability. The term does not include
facilities used in the local distribution of electric energy.”).
345
346
Potential Reliability Impacts of EPA’s Proposed Clean Power Plan, NERC November 2014
http://www.nerc.com/pa/RAPA/ra/Reliability%20Assessments%20DL/Potential_Reliability_Impacts_of_EPA_Proposed_CPP_F
inal.pdf (hereafter “NERC Initial Reliability Review)
347
Id. at 1
180
APPA supports NERC’s initial assessment and encourages EPA to incorporate a detailed and
thorough BPS reliability analysis into the implementation plan of the final rule. The reliability of
the BPS requires careful study and detailed modeling to understand the flow of electricity on the
system. These studies inform the reliable integration of new resources and decommissioning of
old resources. Transmission modeling will assure that all flows of electricity are balanced.
These models will also help system operators respond to unexpected loss of equipment to
prevent a blackout, cascade or uncontrolled separation of equipment.
To assure reliable operation of the BPS during the implementation of the final rule, a reliability
back-stop mechanism needs to be developed and incorporated into the rule. This mechanism
must incorporate grid modeling conducted by qualified personnel who understand BPS
operations. Rule implementation must also allow adequate time for BPS modifications based on
the modeling conclusions.
A.
NERC Initial Reliability Review
NERC’s Initial Reliability Review of the Proposed Rule has identified key areas of concern for
BPS reliability. In this initial review NERC identified four primary impact recommendations: 348




348
Fossil-Fired Retirements and Accelerated Declines in Reserve Margins – The
Regions, ISO/RTOs, and states should perform further analyses to examine potential
resource adequacy concerns.
Transmission Planning and Timing Constraints – EPA and states, along with industry,
should consider the time required to integrate potential transmission enhancements and
additions necessary to address impacts to reliability from the Proposed Rule. EPA and
policy makers should recognize the complexity of the reliability challenges posed by the
rule and ensure it provides sufficient time for the industry to take the steps needed to
significantly change the country’s resource mix and operations without negatively
affecting BPS reliability.
Regional Reliability Assessment of the Proposed CPP – Other ISO/RTOs, states, and
Regions should prepare for the potential impacts to grid reliability, taking into
consideration the time required to plan and build transmission infrastructure.
Reliability Assurance – EPA, FERC, DOE, and state utility regulators should employ
the array of tools and their regulatory authority to develop a reliability assurance
mechanism, such as a “reliability back-stop.” These mechanisms include timing
adjustments and granting extensions where there is a demonstrated reliability need.
Id. at 3, Recommendations to Address Direct Impacts to Resource Adequacy and Electric Infrastructure.
181
APPA believes that NERC’s initial reliability review is valid and raises issues that need to be
addressed in the final rule. NERC plans to produce a Special Reliability Assessment in the first
quarter of 2015 to evaluate the long term reliability impact of the Proposed Rule. APPA
recommends that EPA take into account this baseline technical evaluation of generation and
transmission adequacy prior to developing the final rule and implementation timelines.
B.
Transmission Planning
The final rule must take into account the complexity of reliable BPS operation. A number of
elements need further study to plan for changing loads, to operate within system performance
standards. NERC’s Initial Reliability Review identified a number of issues needing further
study.349
Table 21 in the Initial Reliability Review provides a list of the types of studies and analyses that
must be done to demonstrate reliability, recognizing that the industry does not operate the grid
without a thorough and complete analysis. They include:
Table 21: Study and Assessment Types Needed for a Complete Reliability Evaluation
Local Reliability Assessments
Area/Regional Reliability Assessments










349
Specific generator retirement
studies
Specific generator interconnection
studies
Specific generator operating
parameters
Power flow (thermal, voltage)
Stability and voltage security
Offsite power for nuclear facilities


Resource adequacy
Power flow (regional)
Stability and voltage security (regional)
Gas interdependencies; pipeline
constraints
Operating reserves and ramping
System restoration/blackstart
Id. at 26, Table 4. Study and Assessment Types Needed for a Complete Reliability Evaluation.
182
A. Operating Limits
The Proposed Rule assumes changes in load through energy efficiency and electric generating
resource mix and utilization through increased use of non-emitting and natural gas generation.
These changes need to be planned so system operators can operate within the limits required by
NERC reliability standards. System operators are equipped to handle multiple contingencies and
loss of lines or generation. A sharp decline in generation in a region may lead to operating close
to the limits and give few options for operators to respond in emergencies. The industry will use
transmission modeling to meet minimum system performance criteria required in NERC
Reliability Standard TPL-001-0.1 – System Performance Under Normal Conditions. An excerpt
from the NERC reliability standard states: 350
System simulations and associated assessments are needed periodically to
ensure that reliable systems are developed that meet specified performance
requirements with sufficient lead time, and continue to be modified or
upgraded as necessary to meet present and future system needs.
The [NERC registered] Planning Authority [or Planning Coordinator] and
Transmission Planner shall each demonstrate through a valid assessment
that its portion of the interconnected transmission system is planned such
that, with all transmission facilities in service and with normal (precontingency) operating procedures in effect, the Network can be operated
to supply projected customer demands and projected Firm (non-recallable
reserved) Transmission Services at all Demand levels over the range of
forecast system demands.
These annual models will anticipate negative reliability effects due to changes in generation
resources. The modeling analysis conducted by the Transmission Planners (TPs) and Planning
Coordinators (PCs) will indicate that an electric reliability issue is on the horizon. If any of the
assumptions within the Proposed Rule do not materialize in a timely manner, long term
transmission plans will need to be revised.
APPA recommends using the TPs and PCs to analyze models to gauge the impact of resource
changes during implementation of the final rule. When an issue is raised by a TP or PC over the
350
http://www.nerc.com/_layouts/PrintStandard.aspx?standardnumber=TPL-0010.1&title=System%20Performance%20Under%20Normal%20%28No%20Contingency%29%20Conditions%20%28Category%2
0A%29&jurisdiction=United%20States
183
planning horizon and effective mechanisms to mitigate the reliability issue cannot be identified, a
reliability back-stop mechanism will be needed to preserve BPS reliability.
B. Stability
If the final rule is implemented as proposed, major changes may need to be made to the BPS
transmission system. The current transmission system was designed to transport electric power
from fossil-fired generation to major load centers. Implementing the rule will accelerate
retirements in the fossil-fired generation fleet and require a switch to new gas fired generation
and/or non-emitting energy resources. The realignment of generation resources will need to be
studied for the reliability impact on the system as a whole. One example of impact due to
generation realignment is loss of inertial mass on the system. This mass provides stability to the
BPS during a fault or loss of a major generating unit. Without this mass on the system, faults
may cascade and cause blackouts.
C. Change of Network Flows
The BPS is a network of transmission lines that allow electricity to flow from generating sources
to customer loads even if certain sections of the transmission network are out of service.
However, the realignment of generation resources contemplated in the Proposed Rule may
reverse the flow of electricity on a transmission network. These flows must be studied so they
can be anticipated and managed. An example of reverse flows is highlighted in a study
conducted by the Brattle Group for SRP. This study evaluates the challenges of new flows on
the Western Interconnect with the loss of fossil-fired generation in Arizona.
D. Timing in Building Transmission Facilities
New Extra-High Voltage (EHV) transmission systems will need to be studied and built to
maintain reliability. If the timeline in the Propose Rule is not adjusted, certain regions may see
significant impact on reliable operations. SPP submitted comments on the Proposed Rule that
highlight the reliability issues that will occur if new generation and transmission cannot be built
within the timeframe:
SPP is also concerned with the timing proposed for compliance with the
CPP. Within the SPP region, the timing associated with CPP compliance
is problematic at best. Based on SPP’s review of the proposed CPP, EPA
has considered neither the cost nor the time required to plan and construct
electric transmission facilities. In the SPP region, as much as eight and a
half years to study, plan for, and construct new transmission facilities has
been required. Compliance with the proposed CPP is impossible due to
184
the transmission expansion that will be required and the time it takes to
complete the required transmission expansion. In addition to more time
being needed to develop plans for and construction of necessary
infrastructure, a “reliability safety valve,” as suggested by the ISO/RTO
Council prior to release of the proposed CPP, should be incorporated into
the final rule. Such an approach would require that state plans include a
process to evaluate electric system reliability issues resulting from
implementation of the state plan and require mitigation when needed. 351
NERC’s Reliability Impact Review comes to a similar conclusion: “a construction timeline for a
new high-voltage line can range from 5 to 15 years.”352
Even if the EHV transmission planning and construction processes were shortened to their best
case minimums, utilities will require time to procure the materials to build new transmission
systems such as EHV transformers. Manufacturing of these specialty products can take from one
to five years. Most manufacturers of EHV transformers are located overseas where equipment
and raw materials may be controlled by foreign countries. Instability in countries providing raw
material may also slow the manufacturing process. Lead times are a concern according to
DOE’s June 2012 Study, Large Power Transformers and the US Electric Grid:353
In 2010, the average lead time between a customer’s [Large Power
Transformer] LPT order and the date of delivery ranged from five to 12
months for domestic producers and six to 16 months for producers outside
the United States. However, this lead time could extend beyond 20
months and up to five years in extreme cases if the manufacturer has
difficulties obtaining any key inputs, such as bushings and other key raw
materials, or if considerable new engineering is needed. An industry
source noted that [High Voltage] HV bushings often have a long lead time
extending up to five months. Another industry source added that HV
bushings are usually customized for each power transformer and there are
limited bushing manufacturers in the United States. Manufacturers must
also secure supplies of specific raw materials or otherwise could endure an
extended lead time.
351
Comments of SPP at 8, http://www.spp.org/publications/2014-10-09_SPP%20Comments_EPA-HQ-OAR-20130602.pdf
352
Id. at 20
353
US Department of Energy, Large Power Transformer Study June 2012 at 9,
http://energy.gov/sites/prod/files/Large%20Power%20Transformer%20Study%20-%20June%202012_0.pdf
185
E. Back-Stop Mechanism to Preserve BPS Reliability
As pointed out above, there are numerous factors working against a utility’s ability to meet the
timelines in the Proposed Rule and at the same time meet the mandatory reliability standards
required by NERC. NERC’s Reliability Impact Review also states the strict compliance with the
proposed timelines will impact reliability:
The proposed timeline does not provide enough time to develop sufficient
resources to ensure continued reliable operation of the electric grid by
2020. To attempt to do so would increase the use of controlled load
shedding and potential for wide-scale, uncontrolled outages. 354
APPA supports the concept of reliability assurance raised by NERC’s Initial Assessment Report.
This would include the development of a “reliability assurance mechanism,” such as a reliability
back-stop, or “safety valve” to preserve BPS reliability. The TPs and PCs, as described above,
are best situated to perform the necessary studies and exercise the authority to imple ment
transmission and resource solutions to preserve BPS reliability. These experts could provide an
annual evaluation on the impact of state/regional resource changes. This impact analysis would
inform any decision to delay compliance with the Proposed Rule to preserve reliability. APPA
recommends that EPA work with NERC and the industry to develop and adopt a reliability
assurance mechanism in the final rule.
XXII.
APPA Supports the Concept of a Reliability Safety Valve
The ISO/RTO Council has proposed that EPA include in its final rule a reliability safety valve
(RSV). The RSV would provide for “a reliability review conducted by the relevant system
operator, working with the states and relevant reliability regulators, prior to finalization and
approval of the [state plan]. The review would identify reliability issues and solutions. The RSV
process would then provide for appropriate regulatory review and approval of the reliability
assessment and solution. Next, it would accommodate the reliability solution under the CO2 rule
and/or [state plan] by providing for appropriate compliance and/or enforcement flexibility while
a long-term reliability solution is developed and implemented.” 355
354
Id. at 22.
“EPA CO2 Rule – ISO/RTO Council Reliability Safety Valve and Regional Compliance Measurement and
Proposals,” ISO/RTO Council, January 28, 2014, page 2. (Footnotes removed.)
355
186
The RSV would provide for flexibility in enforcement of the Proposed Rule in the event
reliability became adversely impacted as a result of meeting the state goals within the required
time frame. The ISO/RTO Council notes that if a longer-term solution is needed to ensure
reliability while complying with the Proposed Rule, then an interim plan could be put in place,
such as one where units could remain in operation if needed for reliability until the longer-term
plan is developed.
APPA supports the inclusion of an RSV in the Proposal, although APPA is not necessarily
endorsing the specific ISO/RTO Council RSV proposal or any other specific proposal of the
ISO/RTO Council in these comments. An RSV is essential for two primary reasons. First,
because the Proposed Rule is not unit-specific, various strategies included in state or regional
plans are not known in advance and the reliability impact of the many different components
cannot be predicted until those plans are developed. For example, a particular owner may
propose a compliance plan that entails a significant increased reliance on variable renewable
resources without sufficient flexible ramping capability, along with the retirement of baseload
resources. Unless the owner or other entities within the state or region construct and makes
available sufficient generation units with flexible ramping capacity, reliable operation of the
electric BPS and reliable service to customers could be jeopardized. If that scenario were
projected to occur under a state plan, then reliability solutions would also need to be developed
as part of that plan.
Second, RTO regions with mandatory capacity markets present significant additional
impediments to the construction of new resources, and such resource development may be a
component of state plans. This additional layer of risk and uncertainty introduced by mandatory
RTO capacity markets necessitates reliability safeguards in the event that planned resources are
not built or cannot obtain affordable financing because of such capacity market risks.
In the event the Proposal is finalized as proposed, the incorporation of a reliability review and
development of reliability solutions within state plans will be essential to guard against adverse
impacts on reliability of electric service. Flexibility measures that allow for interim solutions to
be put in place, even if compliance with state goals is not achieved within the regulatory time
frame of the Proposed Rule, will ensure that reliability will be preserved while the goals are
achieved. Incorporating flexibility measure such as an RSV into the Proposal will allow for
more sustainable implementation strategies and is superior to an immediate approach that
restricts the states to shorter-term options that sacrifice reliability in exchange for full
compliance.
187
XXIII. The RTOs/ISOs Should Not Be Given Any New Market-Related Role
in Implementing the Final Rule
A.
Overview of RTOs/ISOs
There are currently six operational RTOs or ISOs (collectively referred to in this section as
“RTOs”) under the jurisdiction of FERC: ISO-NE, NYISO, PJM, MISO, CAISO, and SPP.
ERCOT operates as an ISO solely within the Texas intrastate transmission grid and is therefore
regulated by the Public Utility Commission of Texas and not by FERC.
These RTOs were formed to operate the bulk transmission grids within their respective regions.
The facilities comprising these regional grids are owned by multiple investor -owned, publicpower, and cooperative utilities. In addition, these RTOs operate wholesale electricity markets
that affect the operation and dispatch of existing electric generation units, decisions concerning
the retirement of existing units, and decisions concerning the construction of new units.
Therefore, RTOs may have a role in the implementation of the Proposed Rule. APPA, however,
urges EPA to avoid recommending that states adopt implementation plans that give any new
significant market-related role to the RTOs. APPA’s concerns are outlined in this section, as
well as the following sections on environmental dispatch and the RTO capacity markets.
All RTOs operate markets for wholesale energy and ancillary services, while three of the FERCjurisdictional RTOs also operate markets for capacity. In the RTO-operated wholesale energy
markets, electricity is typically dispatched every five minutes, first in the day-ahead and then in
the real-time market. Generators submit price offers to sell power, and load-serving entities
submit load forecasts, to the RTO. Subject to transmission and other operational constraints, the
RTO commits and dispatches generating units in order of lowest to highest offer to meet the
forecasted load. This procedure is commonly known as Security-Constrained Economic
Dispatch (SCED). (There are many descriptions of SCED that refer to resources being
dispatched in order of least to highest cost, but generators need not offer to sell at their actual
cost of producing the electricity.)
In any event, generator offers are generally subject to an offer cap that is typically $1,000 per
MWh, and in some circumstances are subject to “mitigation” (i.e., reduction) by the RTO
“market monitor” to cost-based levels. RTOs use locational marginal pricing (LMP) in the
energy markets to reflect pricing differentials that occur when transmission congestion prevents
the RTO from dispatching generating units with the lowest-priced offers and the RTO must
therefore dispatch a unit with a higher offer price to serve RTO loads within a constrained zone.
The Proposed Rule contemplates that environmental dispatch or redispatch of electric generating
units would be implemented in the context of these RTO-operated wholesale energy markets.
The difficulty of accomplishing that result is discussed in greater detail in Section XXIV.
188
While the wholesale energy market is where the purchase and sale of electricity occurs from
existing resources in RTO regions, the RTO capacity markets are intended to provide revenue for
the development of new generation resources and to cover the costs of keeping existing resources
ready to supply this electricity when needed. As discussed in Section XXV, the ability of the
RTO capacity markets to accomplish their goals is highly questionable and in some cases these
markets are impeding new resource development.
B.
Impediments to the Use of RTO-Operated Markets for CO2 Reduction
Strategies
There are many real-world features of the RTO wholesale electricity and capacity markets that
would increase costs and create difficulties in the implementation of environmental dispatch or
other CO2 reduction strategies. First, these markets bear little resemblance to truly competitive
markets, as would be necessary for successful implementation of these proposals. This is
discussed in greater detail later in this section.
Second, RTO tariffs and market rules have been in continual flux, creating significant
uncertainty about future market design. For example, there are three open administrative dockets
at FERC that raise fundamental questions about the capacity, energy, and ancillary services
markets and could pave the way for potentially broad changes within these markets. 356 In just
the first half of 2014, PJM alone had six dockets that were either initiated or had orders issued by
FERC that would tweak the already complex capacity market rules. In addition, PJM has just
initiated a never-before-used Enhanced Liaison Committee to seek approval for significant
changes to the capacity markets to ensure that there is a set of generation resources that are
available throughout the year, especially during extreme weather events.
Third, these markets are opaque and have limited data transparency. For example, offers to sell
electricity are published with a three or four month time lag and then are only posted with the
identity of the bidders masked. Such limited data makes it difficult for market participants and
observers to determine the extent of deviations in market behavior from a competitive market.
356
Price Formation in Energy and Ancillary Services Markets Operated by Regional Transmission Organizations
and Independent System Operators, Docket No. AD14-14-000; Winter 2013-2014 Operations and Market
Performance in Regional Transmission Organizations and Independent System Operators, Docket No. AD14-8-000;
and Centralized Capacity Markets in Regional Transmission Organizations and Independent System Operators,
Docket No. AD13-7-000
189
Fourth, the governance structure of the RTOs does not ensure adequate representation of the
views of all entities affected by RTO policies, including the states that would be charged with
implementing the Proposed Rule. Were an RTO to manage a CO2 emissions reduction program
for a state or group of states, it would be difficult to ensure that state goals would be achieved or
that that the public would be adequately protected. Frank A. Felder of Rutgers University sums
up the longstanding critiques of RTO governance as:
Larger entities that can form large voting blocks; smaller entities do not
have the financial resources to participate in stakeholder meetings that
occur almost every business day, so their participation is not meaningful;
even if other stakeholders have interests that overlap in part with those of
consumers, it is unreasonable to expect those entities to adequately
represent consumers; and ISOs are able to take advantage of competing
stakeholder interests to advance their own agendas.357
In fact, RTO governing boards at times have directed RTO management to submit tariff or
market rule changes to FERC without the majority support of the RTO’s membership. For
example, ISO-NE filed with FERC and received approval for a controversial proposal to
establish pay-for-performance incentives and penalties for generators that was supported by just
10 percent of the membership.358 The most egregious recent example of non-representative RTO
governance is the overturning of previously negotiated self-supply and state-sponsored resource
exemptions in the PJM MOPR, discussed in detail in Section XXXI.
Finally, and perhaps most importantly, several states (see page 30) are split by RTO boundaries,
and in fact some individual utilities must operate in more than one RTO. This would put
affected states in the position of having to draft their proposed plan with a view toward
complying with requirements of more than one RTO in order to achieve approval from the EPA.
Additionally, the use of RTOs to manage all or part of the Proposed Rule would place several
states, and perhaps in several utilities, in the untenable position of having to operate with
differing RTO requirements, which might be at odds from time to time.
357
Watching the ISO Watchman, by Frank A. Felder, The Electricity Journal, December 2012, Vol. 25, Issue 10. T.
The term ISO in this article is the same as the use of the term RTO in these comments.
358
NEPOOL Proposed Revisions to Market Rule 1 of the ISO-NE Tariff, Transmittal Letter, Docket No. ER141050, Federal Energy Regulatory Commission, January 17, 2014, page 6 notes that “the ISO-NE Proposal received a
Vote of only 10.28% in favor, with only 5.5 members supporting the ISO-NE Proposal. Available at: 325-cf14696c1b50-499c-93b7-8ba3d57511a3.PDF
190
XXIV.
The Difficulties Facing Proposals for Environmental Dispatch or
Redispatch in RTO Regions
In anticipation of this Proposed Rule, several proposals have been made to achieve CO2 emission
reductions through adjustments to the pricing mechanisms of wholesale electricity markets. A
number of these proposals focus on redispatch of the RTO-operated wholesale energy markets,
described previously in Section XXIII. This approach to reducing CO2 emissions will be
referred to as “environmental dispatch” or “redispatch.”
Two key features of the Proposed Rule bear similarities to the environmental dispatch proposals.
First, building block 2 would increase the dispatch of existing natural gas combined cycle plants ,
combined with a decrease in the dispatch of coal and oil steam units. The Proposed Rule does
not provide a specific mechanism for achieving such redispatch, leaving that to the states, but it
does reference a shift in dispatch achieved through the addition of the cost of CO2 emission
allowances to the variable costs of generation, such as through regional cap-and-trade programs
like RGGI or through absolute limits on CO 2 emissions at higher emitting units.359 The Proposed
Rule also states that if redispatch can be accomplished on a regional level, as opposed to within a
state boundary, the costs of achieving redispatch would be lower.360
Because environmental dispatch appears likely to be considered as part of a CO 2 reduction
strategy, this section reviews and comments on three proposals made thus far pertaining to
environmental dispatch.
A.
Summary of RTO Market-Based CO2 Reduction Proposals
In a paper entitled, A Market-Based Regional Approach to Valuing and Reducing GHG
Emissions from Power Sector, released in April 2014 just before the Proposed Rule was issued,
Judy Chang, Jurgen Weiss, PhD., and Yingxia Yang, PhD., of The Brattle Group, made a fairly
detailed proposal for CO2 reduction using RTO markets.361
359
79 Fed. Reg. at 34,862.
Id. at 34,865.
361
Judy Chang, Jurgen Weiss & Yingxia Yang, A Market-Based Regional Approach to Valuing and Reducing GHG
Emissions from the Power Sector (April 2014), available at
http://www.brattle.com/system/publications/pdfs/000/005/003/original/A_Marketbased_Regional_Approach_to_Valuing_and_Reducing_GHG_Emissions_from_Power_Sector_Chang_Weiss_Yang
_Apr_2014.pdf?1397577641. The paper was prepared for Great River Energy, a generation and transmission
cooperative.
360
191
The Brattle Group proposal has RTOs assess a charge on “each participating generator for CO2
emissions at a rate equal to the carbon price (in $/ton) multiplied by each generating source’s
emission rate (in tons per MWh). By doing so, the CO2 emissions become an additional variable
cost for CO2-emitting generators.”362 (Brattle Group, page 3.) The paper goes on to say that
“generators would therefore include these costs in their offers” as a means to recover such
charges. The determination of the initial carbon fee would be accomplished by the states within
the RTO in conjunction with an independent facilitator, and would include a mechanism for
adjusting carbon prices if expected emissions reductions are not achieved. The Brattle Group
recommends that the carbon fees collected by the RTO be refunded to the load-serving entities
(“LSEs”) in a neutral manner not tied to any incentives, such as based on LSE share of total load.
This refund would partially, but not fully, offset the costs.
A hypothetical example illustrates how this proposal might work. In this simplified scenario,
generators always offer to sell at a price equal to their variable cost of producing that unit of
electricity. There are three generating plants. Plant A emits one ton of CO2 per MWh and has a
variable cost of $50/MWh. Plant B emits two tons of CO2 per MWh and has a variable cost of
$35/MWh. Plant C emits four tons of CO2 per MWh and has a variable cost of $30/MWh. Each
plant can generate up to 100 MWh per hour. During this hypothetical hour, the total demand for
electric energy is 225 MWh.
Hypothetical Example Without CO2 Pricing:
Because each plant offers to sell electricity at a price equal to its variable cost, Plants C and B
would be dispatched first at their full 100 MWh capability. Plant A would be dispatched at just
25 MWh and would set the clearing price at $50.
A total of 625 tons of CO2 would be emitted (400 from Plant C, 200 from Plant B, and 25 from
Plant A).
Plant C would earn a “profit” (i.e., inframarginal revenues) of $2,000 (($50 - $30) x 100 MWh),
Plant B would earn $1,500 (($50-$35) x 100 MWh) in profits, and Plant A would earn zero
because its cost is equal to the clearing price. The cost to consumers would be $11,250 ($50
clearing price x 225 MWh).
Hypothetical Example With CO2 Pricing of $10 per ton:
The variable cost per MWh would increase by $10 multiplied by the number of tons per MWh.
Plant A would now offer to sell for $60 ($50 plus $10 x 1 ton), Plant B would offer at $55 ($35
362
Id. at 3.
192
plus $10 x 2 tons), and Plant C would offer at $70 ($30 plus $10 x 4 tons). Plants A and B
would be dispatched at 100 MWh, and Plant C would be dispatched at 25 MWh. Plant C would
set the clearing price at $70.
A total of 400 tons of CO2 would be emitted (100 from Plant A, 200 from Plant B, and 100
from Plant C), for a total reduction of 225 tons of CO2 in that hour.
Plant C would earn zero profits, a decrease of $2,000, Plant B would earn $1,500 ($70 - $55 x
100 MWh), a decrease of $500, and Plant A would earn $1,000 ($70 - $60 x 100 MWh),
compared to zero without a carbon price.
Under the Brattle Group proposal, the LSEs would receive a refund of $1,000 from Plant A,
$2,000 from Plant B, and $1,000 from Plant C for a total of $4,000.
Consumers incur a cost of $11,750 ($70 x 225 MWh - $4,000 refund.) The costs to consumers
would increase by $500 under carbon pricing.
These hypothetical examples are summarized in the following table:
Table 22: Hypothetical CO2 Emissions Reductions and
Costs to Consumers from Environmental Dispatch
Without Carbon Pricing
Plant
A
Emissions Rate
(tons/MWh)
1
Cost ($/MWh)
$50
LMP ($/MWh) $50
Dispatch
(MWh)
25
Profits
0
Emissions
(tons)
25
Cost to Consumers
Refund
B
C
2
$35
4
$30
With Carbon Pricing ($10/ton)
A
B
C
Total
100
100
225
$1,500 $2,000 $3,500
200
Net Cost
400
1
$60
$70
2
$55
100
$1,000
100
25
$1,500 $0
625
100
$11,250
0
$1,000
$11,250
4
$70
Total
200
225
$2,500
100
400
$15,750
$2,000 $1,000 $4,000
Net Cost
$11,750
The Brattle Group states that imposing this carbon fee would have two types of effects. First, in
the near term, it would increase the offer price for higher-emitting resources, and (all other things
being equal) such resources would be displaced by lower-emitting resources that have a lower
193
(or no) carbon-cost adder. As a result, there would be a shift in the dispatch from higheremitting to lower-emitting plants, as is illustrated in the hypothetical example above. Second,
lower-emitting or non-emitting (i.e., CO2-free) resources would earn greater profits because the
market-clearing price would be set by higher-emitting plants. Over the longer-term, these profit
differentials would “alter the generation mix through generator entry and retirement to efficiently
accomplish the intent of EPA.”363
The Brattle Group states that its proposal would be more efficient than direct emissions
limitations on individual generators and that it uses an “existing market system that is well
equipped to determine least-cost solutions under constraints.”
Two other, less-detailed proposals have also been released. The ISO/RTO Council, whose
membership includes nine ISOs and RTOs in the U.S. and Canada, released a paper entitled,
EPA CO2 Rule—ISO/RTO Council Reliability Safety Valve and Regional Compliance
Measurement and Protocols, which contains two proposals.364 First is the RSV proposal
discussed in Section XXII, which allows for compliance and enforcement flexibility while
longer-term reliability solutions are pursued. The second is not a proposal for carbon pricing per
se, but a recommendation that states be given the option to measure compliance with Section
111(d) on a regional basis, as EPA has proposed. The ISO/RTO Council contends RTOs and
their use of SCED are the optimal means for such regional compliance measures, asserting that
“supply bids submitted by generators effectively internalize compliance costs while still ensuring
least cost compliance with environmental requirements.”365 In other words, if a generator can
reduce CO2 emissions at a lower cost than another generator, all things being equal, that
generator will be dispatched earlier in the stack.
A third recently released proposal is from the non-profit environmental group Clean Air Task
Force entitled, Power Switch: An Effective, Affordable Approach to Reducing Carbon Pollution
363
Brattle Group at 3. This approach differs from the concept of a “shadow price.” A shadow price is not an actual
fee assessed against a market participant, but is added only within the dispatch algorithm as a means of altering the
dispatch order. The result of a shadow price would be to move a plant with a greater offer price but lower carbon
dioxide emissions higher in the dispatch stack than a plant with a lower offer price and higher emissions. The
Brattle Group proposal has a direct fee rather than a shadow price because under a shadow price, higher CO2
emitting generators could adjust their offers downward to maintain their dispatch position, but an actual fee that
directly increases a generator’s costs would need to be incorporated into its offer price.
364
RTO/ISO Council, EPA CO2 Rule—ISO/RTO Council Reliability Safety Valve and Regional Compliance
Measurement and Proposals (Jan. 28, 2014), available at
http://www.isorto.org/Documents/Report/20140128_IRCProposal-ReliabilitySafetyValveRegionalComplianceMeasurement_EPA-C02Rule.pdf.
365
Id. at 5-6.
194
from Existing Fossil-Fueled Power Plants.366 The CATF paper is centered on a recommendation
that EPA offer a model interstate emission credit trading rule for adoption by the states.
However, it also discusses one option for meeting the emissions guidelines “through the
redispatch of existing electric resources by an Independent System Operator (ISO),” with the
Brattle Group proposal referenced as a means to achieve such redispatch.367 The CATF paper
notes that because of the different regulatory and market structures in which the states operate,
compliance flexibility would be needed. For the RTO-operated markets, the CATF paper
concludes that “[s]tates in RTO markets already benefit from transparent market -based energy
prices” and that “[a] market-based emission credit pricing mechanism tied to electric generating
output would complement the market prices for energy and provide covered generators with a
market price signal regarding the value of changes in generation dispatch, unit commitment, unit
retirement and alternative compliance options.” 368
The CATF paper provides the results of an analysis conducted by the NorthBridge Group of the
potential CO2 emissions reductions from this proposal by 2020, which concludes that the vast
majority of such reductions would come from shifts in dispatch (accounting for 212 out of the
308 metric tons that are projected by NorthBridge to be reduced below 2020 forecast levels.) 369
B.
Critiques of RTO Market CO2 Reduction Proposals
The Brattle Group proposal would have many potential complications, a number of which the
proposal itself identifies. A major question is whether EPA’s legal authority allows the
imposition of a direct fee or tax on merchant generators to be included in the CO2 emission
reduction guidelines.370
Other complications include the difficulty of identifying an optimal carbon price that achieves
emissions reductions while not causing a level of retirements that threatens reliability; the
appropriate price to pay generators or charge load from non-participating states located within
the RTO; whether to subject new generators complying with NSPS to the same pricing rules as
existing generators; and how to handle generators with legacy long-term contracts that are priced
outside of the RTO markets.
366
Clean Air Task Force, Power Switch: An Effective, Affordable Approach to Reducing Carbon Pollution from
Existing Fossil-Fueled Power Plants (Feb. 2014), available at http://www.catf.us/resources/publications/view/194.
367
Id. at 17.
368
Id. at 16.
369
Id. at 22 (Figure 12).
370
This is also likely to be a concern for the use of environmental dispatch within non-RTO regions.
195
Besides the issues acknowledged in the Brattle Group proposal itself, there are flaws in the RTOoperated markets that may interfere with the ability of a carbon fee to achieve the goals of
reducing CO2emissions while minimizing costs. Such flaws are identified and discussed below.
1.
Prices and Costs Are Not Always Aligned.
The carbon fee would be layered upon an energy market where offers to sell electricity can vary
significantly from the underlying production costs. Generation owners may have different
reasons for offering at prices above or even below the actual costs. For example, an owner of a
fleet of plants may choose to offer a marginal plant at a higher price to drive up the clearing price
and thus the earnings for its other plants, even if it means that one plant is less likely to be
dispatched (known as economic withholding). This absence of a direct cost and price connection
and the use of different offer strategies by generators are reflected in the volatility of wholesale
electric power prices. According to EIA, “Power prices formed in RTOs tend to be spikier than
those formed in markets featuring bilateral trading between market participants (Pacific
Northwest and Southeast).” 371
Below are two further hypothetical scenarios showing the more complex implementation issues
in markets operated by RTOs. In these scenarios, a generation owner employs a strategy of
bidding a single marginal plant significantly above its cost to improve the earnings of the other
plants in its fleet, and no CO2 reductions are achieved. While many different scenarios may
occur, and in some hours, the strategy may be successful in reducing CO2 emissions, this
example shows the outcome within the RTO-operated markets can be highly unpredictable.
Hypothetical Example of Generator Strategic Price Offer Without Carbon Pricing
This scenario uses the same three plants as in the prior example: Plant A emits one ton of CO2
per MWh and has a variable cost of $50/MWh; Plant B emits two tons of CO2 per MWh and has
a variable cost of $35/MWh; and Plant C emits four tons and has a variable cost of $30/MWh.
In this case, Plants A and B are owned by the same entity. To maximize their earnings, the
owner offers Plant A at $65 or $15 above its actual cost and Plant B at its cost of $35; and Plant
C is offered at $30 by another owner. The dispatch order and CO2 emissions would be the same
as in the prior example: Plants B and C are each dispatched first with 100 MWh from each and
Plant A is dispatched at 25 MWh. Total emissions are 225 MWh. This is illustrated in Table 23.
371
http://www.eia.gov/todayinenergy/prices.cfm
196
As shown, plant C earns a profit of $3,500 (($65-$30) x 100), Plant B earns $3,000 (($65-$35) x
100) and Plant A now earns $375. Total profits are $6,875 and costs to consumers are $14,625
(225 MWh x $65 clearing price), significantly above a scenario without the use of a strategic
offer. All of the generators have a financial interest in this strategy and have no reason to change
their or others’ behavior.
Hypothetical Example of Generator Strategic Price Offer With Carbon Pricing
As in the prior scenario, there is a carbon price of $10 per ton. The owner of Plant A uses the
same strategy, but adds more to their offer to account for the carbon pricing and higher costs.
The owner therefore adds $20 to the cost as part of their strategic bidding strategy and offers to
sell electricity from Plant A at $80 ($60 cost with carbon pricing plus $20 adder). Plants B and
C offer at their costs plus the carbon fees ($55 and $70 respectively including the carbon fees).
In this case, Plant A is dispatched at only 25 MWh and earns a profit of only $500 ($80$60)*25)), which is $500 below the carbon pricing scenario without strategic offer prices as
shown in Table 23. But Plant B, under the same ownership, now earns $1,000 more than
without strategic bidding, with $2,500 in profits ($80-$55) x 100)). There is a net gain of $500
for the owner of A and B. Plant C also increases its profits by $1,000.
These scenarios are summarized in the table below. As a result of Plant A, the lowest emission
plant, being offered at the highest cost with the use of strategic bidding, the emissions are higher
in the strategic bidding scenario—625 tons vs 400 tons. In fact, in this scenario, because of the
adjustments to the strategic offer in the face of the carbon pricing, the CO2 emissions remained
the same as without carbon pricing. Costs to consumers are actually reduced under the strategic
bidding scenario, because the higher CO2 emissions under the strategic bidding scenario produce
a greater refund. But the strategic bidding has prevented the policy from achieving its goal of
lowering CO2 emissions.
197
Table 23: Pricing of Strategic Bidding
Strategic Bidding
Without Carbon Pricing
Plant
A
Emissions
(tons)
1
Cost
$50
Offer
$65
LMP
$65
Dispatch
(MWh)
25
Profits
$375
Emissions
25
Cost to Consumers
Refund
B
2
$35
$35
$65
Strategic Bidding
With Carbon Pricing
A
B
C
C
4
$30
$30
$65
Total
100
100
225
$3,000 $3,500 $6,875
200
400
625
$14,625
0
Net Cost
$14,625
Without Strategic Bidding (from Table 17)
Cost
$50
$35
$30
Offer
$65
$35
$30
LMP
$65
$65
$65
Total
Dispatch
25
100
100
225
Profits
$375 $3,000 $3,500 $6,875
Emissions
25
200
400
625
Net Cost
$11,250
Changes Due to Strategic Bidding
Profits
$375 $1,500 $1,500 $3,375
Emissions
0
0
0
0
Net Cost
$3,375
1
$60
$80
$80
2
$55
$55
$80
4
$70
$70
$80
25
$500
25
100
$2,500
200
100
$1,000
400
$250
$2,000
$4,000
$60
$80
$80
25
$500
100
$55
100
$2,500
200
($500) $1,000
(75)
0
Total
225
$4,000
625
$18,000
$6,250
$11,750
$70
100
$1,000
100
$1,000
300
Total
225
$4,000
400
$11,750
$1,500
225
0
There are additional factors besides economic withholding or other forms of strategic bidding
that may cause the a carbon fee, working in conjunction with the RTO market prices, to not
necessarily be the least-cost means of compliance mechanism for the Proposed Rule’s
environmental redispatch building block. First, financial entities can play a significant role in the
energy markets and price formation. There are tools available in the RTO markets, such as
virtual bids, that allow financial entities that do not own resources to “buy” and “sell” electricity,
arbitraging prices between the real-time and day-ahead market or between two different sets of
pricing points (such as through Up-to-Congestion transactions). (Congestion here refers to
transmission congestion that can cause price differentials in the RTO energy markets.) Such
transactions are aimed only at taking advantage of price differentials between the day-ahead and
198
real-time markets or between congestion revenues in two different power flow directions. These
transactions can be marginal and set the clearing price. But because these sales and purchases do
not involve actual generation, they can influence energy prices separately from any carbon
pricing.
There are several scenarios where a generator may run even if its offer price is above the clearing
price (also known as “out-of-merit” dispatch). Such generators are then compensated for the
difference between their offer and the clearing price through what are known as “uplift”
payments.372 For example, system operations may not be accurately modeled, requiring the use
of generating units that are manually dispatched in real time to ensure reliability, even if their
offers exceeded the clearing price. Units that did clear could be manually curtailed. Another
scenario occurs where a unit committed in the day-ahead market does not clear the real-time
market, but must continue to run because of minimum run times or an inability to ramp down
quickly. This unit would also receive uplift payments. Such scenarios could entail the operation
of higher CO2 emitting units than anticipated by the environmental dispatch program.
The CATF paper contends that in the near term, the primary intended benefit of environmental
dispatch would be to shift dispatch, especially between existing coal and natural gas plants,
which in EPA’s Proposal would entail natural gas dispatch at least at 70 percent of its capacity.
Similarly, in its analysis of the potential for increasing natural gas dispatch, provided in the
CATF proposal, the Northbridge Group projects an increase in the average capacity factor for
combined cycle natural gas units from 48 to 65 percent and reductions in the average coal unit
capacity factor from 67 to 58 percent by 2020. Several factors, however, could make such a shift
difficult to achieve in practice. Not all natural gas capacity can act as baseload capacity. A
portion of natural gas plants are best used as peaker plants because of their ramping capabilities.
With an increase in variable, renewable energy penetration, a good portion of the gas-fired
capacity is going to need to be used as flexible ramping capacity.
Lack of access to natural gas, due to limits on pipeline and storage capacity, can also impede the
ability to increase the dispatch, and this does not appear to be incorporated into the NorthBridge
Group analysis.
Following the recent shift in coal and gas prices between 2008 and 2012, with coal prices
increasing by 31 cents per MMBtu (a 15 percent increase between those years) and natural gas
falling by 5.6 cents (a 62 percent decrease between 2008 and 2013) NGCC plant capacity factors
372
For a more detailed discussion of uplift, see Staff Analysis of Uplift in RTO and ISO Markets, Federal Energy
Regulatory Commission, August 2014, http://www.ferc.gov/legal/staff-reports/2014/08-13-14-uplift.pdf
199
increased from 40.1 to 51 percent between 2008 and 2012, 373 an 11 percentage point change.
EPA is assuming that another 20 percentage point increase is feasible. The fee assessed on CO2
emissions to achieve this dramatic an increase could be significant.
The ISO/RTO Council proposal for RTO-wide compliance measures is centered on the argument
that SCED would “optimize the efficiency and effectiveness of a compliance program across a
broad fleet of generators and demand response resources.” 374 While SCED does dispatch those
units with the lower price offers first, it does not reduce the total costs paid for electricity.
Because of the use of a single-clearing price, if the compliance measures undertaken increase the
costs of a marginal unit and that cost is reflected in the offer, those costs would apply to all
kilowatt-hours consumed within a given hour. A single-clearing price approach is therefore
inherently more expensive than a regime where were the cost is recovered only for the electricity
generated by an individual plant. Compounding this single-clearing price effect is the fact that
generators do not necessarily bid their actual costs and may decide to bid substantially above
their costs, subject only to the relevant market rules and market power mitigation measures.
C.
The Regional Greenhouse Gas Initiative Is Not an RTO-Operated
Program, but a Voluntary Program.
A brief discussion of RGGI is provided here as it is often cited as an example of a CO2 emissions
reduction program that is also market-based. But it is important to note that RGGI is not an
RTO-operated program. Rather it is a cooperative effort of a group of states also located within
three RTOs (ISO-NE, NYISO and PJM). Moreover, the extent to which the RGGI emission
allowance prices have been reflected in generators’ offers and therefore affect RTO dispatch
does not appear to have been studied in the RGGI Market Monitor Reports. The reports of
benefits from RGGI cover state energy efficiency and renewable energy programs that states
have funded with the proceeds from the sale of allowances. 375 The Analysis Group conducted a
study of RGGI in 2011, but this study only modeled the dispatch in the RTO markets and did not
observe actual behaviors. 376
373
U.S. Energy Information Administration, Table 6.7.A. Capacity Factors for Utility Scale Generators Primarily
Using Fossil Fuels, January 2008-May 2014
374
ISO/RTO Council, EPA CO2 Rule at 6.
375
See “RGGI Benefits,” http://www.rggi.org/rggi_benefits
376
The Economic Impacts of the Regional Greenhouse Gas Initiative on Ten Northeast and Mid-Atlantic States,
Nov. 15, 2011, The Analysis Group,
http://www.analysisgroup.com/uploadedFiles/Publishing/Articles/Economic_Impact_RGGI_Report.pdf
200
Despite this absence of a real world analysis of how the RGGI’s CO 2 allowance costs have
actually affected generator offers and dispatch within the RTO energy markets, acting FERC
Chairman Cheryl LaFleur stated in July 2014 that the RTOs “have been able to successfully
accommodate the requirements of the Regional Greenhouse Gas Initiative (RGGI) into their
market designs. Generators that must purchase emissions allowances under RGGI are able to
include the cost of the allowances in their market bids, and those costs are reflected in the
economic dispatch.”377 Similarly, the Proposed Rule asserts that “operators of EGUs subject to
CO2 emissions limits in RGGI now include the cost of RGGI CO 2 allowances in those EGUs’
variable costs,” and that: “[T]he PJM market monitor publishes breakdowns of wholesale energy
prices, including a CO2 emission allowance cost component.”378 But the market monitor is
simply estimating the cost components of the marginal unit by multiplying the carbon emission
allowance price by the emissions of that unit’s technology type. This is not an analysis of how
actual offers or the dispatch itself was affected by the carbon allowance prices. Moreover, in
2013, the CO2 emission allowance accounted for 0.3 percent of the LMP while the mark-up (or
the difference between costs and price offers) was a negative two percent of the LMP. 379 For the
first half of 2014, the CO 2 emission allowance was the same percentage while the mark-up was
2.5 percent.380 In other words, the markup added or subtracted from the generators’ actual costs
was much more of a driver of the offers than the emission allowance cost, illustrating the
complexity of how offers are formed and how the emission allowance will affect these offers.
D.
Summary of APPA Position on Environmental Dispatch
APPA is not taking a position on state decisions to develop cap-and-trade or CO2 pricing
programs as a means to reduce CO2 emissions. But APPA recommends against any management
of such program by an RTO and emphasize the impediments posed by the RTO markets to
implementing such programs. If such a program is implemented in an RTO region, it will face
challenges in achieving its goals and will likely increase the costs to consumers more than if it
were not left in the RTO’s hands. Therefore, APPA urges that the states proposing such an
approach develop a proposal for a non-RTO regional entity to manage its implementation.
377
Responses of Acting Chairman Cheryl A. LaFleur to Committee on Energy & Commerce Subcommittee on
Energy & Power Preliminary Questions for the Federal Energy Regulatory Commission,
http://docs.house.gov/meetings/IF/IF03/20140729/102558/HHRG-113-IF03-Wstate-LaFleurC-20140729SD001.pdf
378
79 Fed. Reg. at 34,862 & n.129.
379
Monitoring Analytics, 2013 State of the Market Report for PJM, Table 3-63,
http://www.monitoringanalytics.com/reports/pjm_state_of_the_market/2013.shtml
380
Monitoring Analytics, 2014 Quarterly State of the Market Report for PJM: January – June, Table 3-65,
http://www.monitoringanalytics.com/reports/PJM_State_of_the_Market/2014/2014q2-som-pjm-sec3.pdf
201
XXV.
RTO-Operated Mandatory Capacity Markets Pose Significant
Barriers to New Generation Resource Development and Thus to
Implementation of the Proposed Rule.
Although the building blocks in the Proposed Rule incorporate an increase in dispatch from
natural-gas fired units, but do not explicitly include the construction of new natural-gas units,
under the portfolio approach, compliance with the proposed guidelines will likely require a
partial turnover of the existing fleet and the construction of new natural-gas fired generation
resources. Such resources will be needed both to replace retiring coal plants and meet the
flexible ramping needs required to support an increase in intermittent renewable resources. This
section discusses how the RTO-operated mandatory capacity markets will greatly impede such
new natural gas generation and are also likely to pose barriers to new renewable resources, as
described in the next subsection.
The difficulties imposed by the mandatory capacity markets for new resource development were
identified in an APPA-commissioned paper released in May 2014 entitled, Markets Matter:
Expect a Bumpy Ride on the Road to Reduced CO 2 Emissions, by Cliff Hamal of Navigant
Economics (“Navigant Paper”). 381 (See Attachment 3.) The Navigant paper prepared for APPA
found that the mandatory capacity markets “are already floundering over existing challenges and
will be severely stressed by the added complexity of maintaining reliability while shifting to a
lower CO2 emission portfolio.”382
A.
Background on RTO-Operated Capacity Markets
The RTO-operated capacity markets provide payments to owners of power plants who agree to
stand ready to supply power when needed or to customers who agree to curtail power use when
called upon (known as demand response). The RTO-operated capacity markets in the midAtlantic (operated by the PJM Interconnection or “PJM”), New England (operated by ISO New
England or “ISO NE”), and New York City and the lower Hudson Valley (operated by the New
York ISO or “NYISO”) are “mandatory markets,” because all resources must be bought and sold
through these markets. See Attachment 5 (Capacity Markets Fact Sheet) for a more detailed
description of these markets.
The RTOs hold periodic auctions where capacity is offered and purchased, typically once a year .
These auctions produce a single price per MW that will be paid to all capacity resources,
381
382
Available at: http://appanet.files.cms-plus.com/PDFs/Markets_Matter_--_Hamal_Report.pdf
Id. at 1.
202
regardless of the type and cost. All customers within the RTO region pay the costs of these
capacity payments, though there is no requirement that the generation owners actually use the
revenue to build new power plants.
B.
RTO-Operated Mandatory Capacity Markets Have Not Been Effective in
Leading to the Construction of New, More Efficient Resources at a
Reasonable Cost to Consumers.
There are numerous flaws in the mandatory capacity markets that have made them ineffective
and costly constructs that will likely impede the development or retention of more efficient and
lower CO2 emitting resources, as follows:





Different resources have different costs, which are not reflected in the capacity market
prices. A 50-year old coal plant is paid the same amount per MW and for the same
duration as a brand new highly efficient combined-cycle natural gas plant.
The vast majority of the revenue collected through capacity markets has been paid to
older, existing units, although many older plants have paid off much of their fixed costs
and are therefore earning windfall profits. For example, only 9 percent of the $72 billion
in revenue committed through the PJM capacity markets is for new resources, demand
response, or energy efficiency. 383
Financing of newer units at a reasonable capital cost requires a long-term steady stream
of revenue, such as that provided by a long-term contract and not the capacity market.
The capacity markets do not distinguish between technology types or specific locations
on the grid. As a result, critical needs are not addressed, including adequate flexible
ramping capability to match the variability of renewable resources, reliability gaps
created by retiring coal plants, and the coordination of natural gas infrastructure and
delivery with the significant expansion of natural gas generation. The RTOs often create
systems of side payments to ensure reliability, such as direct payments through what are
known as reliability-must-run agreements to coal plants to remain in place to ensure
reliability.
Capacity markets use zonal price differentials on the theory that higher prices will act as
a “signal” for the development of new generation or transmission. But such higher prices
are not effective signals because owners of existing generation have no financial interest
in building new resources and lowering prices for their existing units. In fact, they have
383
Monitoring Analytics, 2014 Quarterly State of the Market Report for PJM: January – June, Table 5-13,
http://www.monitoringanalytics.com/reports/PJM_State_of_the_Market/2014/2014q2-som-pjm-sec5.pdf
203

an incentive to hinder new entrants, and have an established track record of using RTO
regulatory processes and litigation to do so.
Investors seek steady and predictable revenue flows, not fluctuating prices and many
other factors influence the decision to build, including land and transmission availability,
local acceptance, and environmental rules.
The ineffectiveness of the capacity markets as a tool in resource development was confirmed by
a recent study by Christensen Associates commissioned by the Electric Markets Research
Foundation. The study’s authors concluded that RTO markets “do not and cannot address longterm capacity needs.” The study also found that “[b]ilateral forward contracting remains key
under any market design for locking in revenues and facilitating financing of new resources.
Contrary to this key necessity, however, RTO markets include some design elements that impede
long-term investments and long-term bilateral contracts.”384
C.
Recent Mandatory Capacity Market Developments Create Direct
Impediments to New Resources.
Not only are mandatory capacity markets generally ineffective in new resource development, but
the markets can actually serve as barriers to new capacity development. These markets all have
some type of MOPR or “buyer-side mitigation” (BSM) that imposes a floor price on offers to sell
from new resources, making it more difficult for these new plants to clear the auctions.385
When mandatory capacity markets were created, the states, public power, and cooperative
utilities negotiated specific and express exemptions from these MOPRs in PJM and ISO-NE for
resources built by load serving utilities to self-supply their own loads. While FERC initially
approved these provisions as just and reasonable, in 2011, the Commission effectively stripped
these self-supply provisions out from the relevant PJM and ISO-NE market rules. FERC did so
based on arguments made by the generators and supported by these RTOs that the provisions
allowed load-serving utilities to exercise “buyer-side market power.” APPA believes these
arguments are unsubstantiated and the MOPR in these RTOs is an unnecessary and inefficient
384
Ensuring Adequate Power Supplies for Tomorrow’s Electricity Needs, Christensen Associates Energy Consulting
LLC, June 16, 2014,
http://www.emrf.net/uploads/3/1/7/1/3171840/ensuring_adequate_power_supplies_for_emrf_final.pdf
385
These rules are not only a concern for the current RTO mandatory capacity markets. Although MISO, the
CAISO, and ERCOT do not currently have mandatory capacity markets, merchant generation owners are frequently
advocating for such a construct to be adopted in these RTOs. For example, three large merchant generation owners,
Exelon, Dynegy and NextEra filed a motion in August requesting that MISO implement a mandatory, forward
capacity market with a MOPR as a means to address pending supply shortfalls in the MISO region. See Indicated
Capacity Suppliers Motion for Expedited Action, FERC Docket ER11-4081-001 (Aug. 25, 2014).
204
barrier to entry. These RTO proposals and their acceptance by FERC overturned previously
negotiated agreements and provide a dramatic example of the absence of a commitment by RTOs
to respect all stakeholder interests.
The impetus for these rule changes was actions taken by several states located within RTOs that
had become frustrated with the lack of new, more efficient generation in their states given the
billions of dollars spent by their utility customers on capacity payments. These states had sought
to take control of their energy resource future and protect their residents from high electricity
prices. New Jersey, Maryland, and Connecticut all established competitive bidding processes for
the procurement of capacity in their states using long-term bilateral contracts. Fearful of the
lower prices that would result from the entry of new generation resulting from these state efforts,
owners of existing power plants sought to block this new generation by obtaining approval from
FERC for a significant tightening of the MOPRs in PJM and ISO-NE. As part of the new MOPR
provisions, FERC eliminated the aforementioned carefully negotiated self-supply and state
resource exemptions.386 In addition, FERC also determined in 2012 a new, more efficient
natural-gas plant under long-term contract in the NYISO was subject to mitigation because the
plant had an unfair advantage by signing a contract that reduced its risks and was procured
through a “discriminatory” process that was only open to new generation.387
Imposition of a MOPR and BSM makes it more likely that new resources will fail to clear the
capacity auctions and that the LSEs would pay twice for new capacity (once for the plant and a
second time through the market). This risk makes financing for such new plants more difficult to
obtain, which raises the cost of capital. Not only do these rules adversely affect the ability of
state utility commissions to ensure reliable service in their states, but they also raise barriers to
resource development by public power utilities, because a MOPR greatly increases the financial
risk these entities face in constructing or procuring new resources. These rules are therefore
impediments to the development of new, cleaner resources, potentially including renewable
energy.
In separate cases, federal district courts in Maryland and New Jersey also invalidated these state
contracts because FERC has jurisdiction over wholesale power rates and states cannot take
actions that impact wholesale power markets. These decisions were appealed to the U.S. Courts
386
PJM Interconnection, 135 FERC ¶ 61,022 (2011) (Order Accepting Proposed Tariff Revisions, Subject To
Conditions, and Addressing Related Complaint),; PJM Interconnection, 135 FERC ¶ 61,029 (2011) (Order On Paper
Hearing And Order On Rehearing).
387
New York Indep. System Operator, 140 FERC ¶ 61,189 (2012) (Order on Complaint).
205
of Appeals for the Third and Fourth Circuits, and both Circuits upheld the district court
decisions388
Not only do the mandatory capacity markets impede implementation of the Proposed Rule by
erecting barriers to new resource development, but further jurisdictional complications exist
where the utilities were restructured at the retail level and therefore no longer own the
generation. Such “retail access states” tend to be located in RTOs with mandatory capacity
markets. These difficulties are summed up in the Navigant Paper (at page 2):
Merchant generators are for-profit companies that sell their energy at
market prices and are neither under traditional utility ownership nor
subject to state price regulation. They do not have any obligation to serve
customers (load) and face minimal price regulation at the federal level.
Even without consideration of CO2 emission reduction objectives, the
combination of RTO-run energy markets and domination of merchant
generators has led to significant difficulties in providing longer-term
investment incentives to maintain proper levels of generating capacity.”
Even in markets with fully-regulated, vertically-integrated utilities, the
challenges of integrating CO2 reduction policies will be significant, but
those utilities have the means to balance all of the options and develop a
comprehensive strategy to meet these environmental goals. In markets
that have restructured at the wholesale and retail level, and eliminated
cost-of-service rates for electricity supply, resource decisions will be
driven by competing firms in response to short term price signals (and
expectations of future short-term price signals).[389]
FERC Commissioner Tony Clark aptly described this difficulty during an April 2014 technical
conference: “The region of the country that continues to vex me more than any other is in those
restructured regions, the Northeast part of the country, the politics and the levels of government
and the stages of restructuring or not restructuring and how they match up offers a very unique
388
PPL EnergyPlus, LLC v. Nazarian, 753 F.3d 467 (4th Cir. 2014), aff’g PPL EnergyPlus v. Nazarian, 974 F.
Supp. 2d 790 (D. Md. 2013); PPL EnergyPlus, LLC v. Solomon, No. 13-4330, 2014 U.S. App. LEXIS 17557 (3d
Cir. Sept. 11, 2014), aff’g PPL EnergyPlus v. Hanna, 977 F. Supp. 2d 372 (D. N.J. 2013).
389
Navigant Paper at 7.
206
set of challenges because you just don't have some of those command and control type levers
that you have in some other regions of the country.”390
Ironically, those entities that remain vertically-integrated (either through asset ownership or
through contracts) and maintain an obligation to serve retail customers regardless of whether
their states have implemented retail restructuring—namely public power and cooperative
utilities—have had their ability to procure new resources to supply their customers impeded by
mandatory capacity market rules.
The Proposed Rule does not directly address this concern, however. Instead it notes that “states
committed to Integrated Resource Planning (IRP) would be able to establish their CO 2 reduction
plans within that framework, while states with a more deregulated power sector system could
develop CO2 reduction plans within that specific framework.” 391 The Proposed Rule later states
that “in the U.S. electricity system the demand for electricity services is met, on both a shortterm and longer-term basis and in both regulated and deregulated contexts, through integrated
consideration of a wide variety of possible options, coordinated by some combination of utilities,
regulators, system operators, and market mechanisms.” 392 These vague and highly generalized
statements do not specify how LSEs and a state within an RTO market, that have not been able
to successfully contract for new, lower-emission resources, can implement these guidelines
without significant reforms to the mandatory capacity markets.
XXVI.
Public Power’s “One Unit” Utility Members
EPA specifically asks for comment on “whether there are special considerations affecting small
rural cooperative or municipal utilities that might merit adjustments to this proposal, and if so,
possible adjustments that should be considered.” 393
In June 2011, EPA convened a Small Entity Representative (SER) panel with respect to its
rulemaking on NSPS for new and existing units. This process was held pursuant to the
requirements of the Small Business Regulatory Enforcement and Fairness Act (SBREFA), which
is intended to provide small businesses flexibility in meeting federal regulations. Though this
process was not completed by EPA in 2011, APPA filed comments that are relevant in this
proceeding. More than 90 percent of public power utilities qualify as small businesses under
390
FERC Technical Conference, Winter 2013-2014 Operations and Market Performance in Regional Transmission
Organizations and Independent System Operators, Docket No. AD14-8-000, April 1, 2014, Transcript pages 286-7.
391
79 Fed. Reg. at 34,834.
392
Id. at 34,881.
393
79 Fed. Reg. at 34,887.
207
SBREFA. APPA also offers these comments for review under the Regulatory Flexibility Act
(RFA) and calls for consideration by EPA, as well as by state agencies, of the many special
governance issues, bond, debt management, and local governmental obligations that are to be
considered under the RFA.
In this context, state agencies may want to set up subcategories or special practices and other
measures to address unit concerns—and in particular, those utilities with small single units.
These should include unit-by-unit determinations by state agencies that factor the size of unit,
age, remaining useful life, and feasibility of the reduction measures since they cannot easily
accomplish the reduction goals. In addition to the pollution reduction methods, state agencies
should allow for reasonable methods of monitoring and verification for energy efficiency to
avoid measures that are very burdensome and/or expansive for very small utilities. APPA points
to the analysis provided by Professor Bradford Cornell for FTI Consulting that addresses the
unique impacts of CO2 reduction measures for public power (municipal and electric coop) as
well as some other for profit utilities.394
Specifically, EPA needs to be aware of the fact that there are numerous small municipal systems
that have only one generation resource under 100 MW today. The implications for those
communities under the Proposed Rule are particularly grave as those municipal systems do not
have the flexibility to rely on other units. The small municipal systems that fall into this
category are listed below. The list of public power utilities with only one unit with coal capacity
that warrant special consideration by their state agencies is below.
Table 24: Public Power Utilities That Have Only Coal Capacity
Util
Code
1050
1192
4280
5742
7222
8449
8543
9286
394
UTILITY_NAME
City of Azusa
City of Banning - (CA)
Conway Corporation
Eldridge City Utilities
City of Gillette
Henderson City Utility Commission
Hibbing Public Utilities Commission
Illinois Municipal Elec Agency*
State
CA
CA
AR
IA
WY
KY
MN
IL
Fuel
Type
Coal
Coal
Coal
Coal
Coal
Coal
Coal
Coal
Capacity
(MW)
34.1325
22.755
72
8.94875
26.726
405
35.9
432.912
http://www.fticonsulting.com/global2/media/collateral/united-states/the-impact-of-a-fleet-emission-rate.pdf
208
Util
Code
9667
10704
11235
11833
12807
12840
13470
14194
14268
15989
17177
18715
19883
20382
21704
24431
26253
40576
40603
40604
50000
50002
UTILITY_NAME
City of Jasper - (IN)
Lansing Board of Water and Light
Lafayette Public Power Authority
Municipal Energy Agency of MS
Michigan South Central Power Agency
Town of Montezuma - (IN)
City of New Madrid - (MO)
City of Orrville - (OH)
City of Owensboro - (KY)
City of Richmond - (IN)
City of Sikeston - (MO)
Texas Municipal Power Agency
City of Virginia
City of West Memphis - (AR)
MSR Public Power Agency
Utah Municipal Power Agency
Louisiana Energy & Power Authority
Intermountain Power Agency
Wyoming Municipal Power Agency
Heartland Consumers Power District
Northern Illinois Municipal Power
Agency
Kentucky Municipal Power Agency
TOTAL
State
IN
MI
LA
MS
MI
IN
MO
OH
KY
IN
MO
TX
MN
AR
CA
UT
LA
UT
WY
SD
Fuel
Type
Coal
Coal
Coal
Coal
Coal
Coal
Coal
Coal
Coal
Coal
Coal
Coal
Coal
Coal
Coal
Coal
Coal
Coal
Coal
Coal
Capacity
(MW)
14.5
529.7
279
43.2
55
3.8745
650
84.5
445.3
93.9
261
453.5
30.2
36
159.3405
18.73125
111.6
1640
50.727
140.58
IL
KY
Coal
Coal
141.28
141.28
6,421.59
All public power coal capacity
34,539.00
Source: Energy Information Administration, Form EIA-860, 2012 data
Since Section 111(d) calls for the consideration of compliance costs, EPA must take into account
the unique impacts that any final rule will have on public power utilities owning only coal-fired
power plants. APPA also seeks to draw to EPA’s attention the 2014 paper395 written by
Professor Bradford Cornell, for FTI Consulting, regarding the potential impacts to public power,
395
http://www.fticonsulting.com/global2/media/collateral/united-states/the-impact-of-a-fleet-emission-rate.pdf
209
rural electric cooperative utilities, and other utilities as a result of a regulatory system with no
effective control technologies or other options to reduce CO 2 economically. This detailed paper
may be found at the FTI website.
XXVII. The Final EPA Rule Should Respect the Importance of U.S.Canadian Electricity Generation Resources for Both Countries.
The North American electricity market does not exclusively reflect generation of electricity in
the U.S. Electricity plays an integral role in the strong U.S.-Canada energy relationship. There
are more than 35 electric transmission interconnections between the U.S. and Canada, which
together make a very integrated North American grid. The two nations have engaged in
electricity trade for many decades. This practice has resulted in positive business decisions, a
wider diversity of supply, and a mutual willingness to provide voltage support in transmissiondependent areas on both sides of the border. The map below, provided by Canadian Electricity
Association, reflects the interconnection between U.S. and Canadian electricity providers.
Figure 26: Electricity Exports and Imports Between Canada and the U.S. (2013)
210
The Proposed Rule does not appear to give full opportunity to Canadian electric utilities to
provide electricity across the border to U.S. states for these purposes. EPA should address this
issue in its rulemaking and provide full opportunity for Canada to continue to be eligible to sell
its electricity across the border. There is also a question as to whether Canadian exports of hydro
or nuclear electricity to U.S. states can be used to meet state goals. Between five and ten percent
of Canada’s total electric generation is exported to the U.S. Most of this comes from British
Columbia, Manitoba, Ontario, and Quebec and is either hydro- or nuclear-based. The Canadian
government and Canadian Electricity Association (CEA) assert that just the exports from Quebec
alone have averted 53 million metric tons of CO2 emissions or roughly the equivalent of 13
million vehicles from the road.
The states of California, Minnesota, Vermont and Wisconsin have accepted imported electricity
from Canada to meet their various “renewable” energy definitions.396 APPA recommends that
imports of all non-CO2 emitting sources imported from Canada be eligible for compliance in a
state plan under the Proposal.
Further, the Canadian government has its own version of NSPS for new and existing coal-fired
power plants. A any final rule from EPA should address this issue in a manner that recognizes
the interconnected nature of the North American grid and that is consistent with the U.S.—
Canadian trade policies set by both the General Agreement on Tariffs and Trade (GATT) and the
North American Free Trade Agreement (NAFTA). The Canadian Electricity Association’s
“Reducing GHG Emissions Under EPA’s Section 111(d) Guidelines” (January 2014) paper is
included in Attachment 6 to explain the complexity of this trade issue.
XXVIII. Carbon Capture and Sequestration on Existing Power Plants
APPA agrees with EPA and compliments it for not proposing that carbon capture and
sequestration (CCS) is adequately demonstrated or commercially available at any scale for
existing power plants. APPA filed extensive comments397 on this issue in response to the
proposed NSPS for CO2 emissions from new power plants under Section 111(b) of the CAA.
APPA’s comments in that proceeding provided a number of white papers and other information
illustrating that CCS is not adequately demonstrated. If a power plant is located where CCS
396
http://www.leg.state.vt.us/docs/2010/Acts/ACT159.pdf and
https://www.edockets.state.mn.us/EFiling/edockets/searchDocuments.do?method=showPoup&documentld={51AC
B5CO-3C14-48EA-A8DO-BCAOD7EDFA89}&doumentTitle=20113-60294-01
397
Docket OAR-2013-0495
211
could be established, then those reductions should warrant compliance with the interim and 2030
deadlines in this Proposal. It is highly unlikely that a typical existing power plant located outside
the oil and gas production regions of the U.S. could do this. If it is possible, such a plant should
be deemed compliant with all interim and final obligations even if the power plant only achieved
all its possible reductions at the time of the final deadline.
Between 2014 and 2030, there is a possibility that new technology might emerge that could
achieve a reduction in CO2 emissions from EGUs or that could result in CO 2 destruction. If this
were to happen, states should be allowed to revise their plans to reflect the use of this new
technology as compliance without having to complete the other measures in the plan.
XXIX.
The NSPS Process for Existing Plants Does Not Require Automatic
Revisions Every Eight Years.
The NSPS process allows EPA to consider revisions to the standards every eight years. There
has been some confusion as to whether this reconsideration and possible revision can occur
automatically. APPA’s view is that such action is not automatic and we support the comments
of the Utility Air Regulatory Group on this issue.
XXX.
Miscellaneous Issues.
APPA endorses the detailed comments and observations offered by UARG on the mass and rate based standards from both the Proposed Rule and the November 6, 2014, NODA.398 APPA joins
UARG in responding to the NODA requesting comments on co-firing of natural gas and goal
generating units.399 Natural gas conversion and co-firing are not BSER for coal-fired utilities
and should not be included in the Proposed Rule. In addition, APPA defers to UARG’s response
to EPA’s questions about Part 75 monitoring and reporting requirements and incorporates those
recommendations by reference. 400
398
http://www2.epa.gov/carbon-pollution-standards/clean-power-plan-proposed-rule-translation-state-specific-ratebased-co2
399
Including work submitted by Lowell Smith for UARG demonstrating that natural gas conversion and co-firing
are too expensive to include as BSER.
400
Including Continuous Emissions Monitoring (CEMs) and the ability to use Appendix G in lieu of CEMS.
212
XXXI.
Potential Constitutional Issues Raised by Proposed Rule.
As previously discussed, the Technical Support Document entitled State Plan Considerations
identifies four distinct pathways available for states to develop their plans to meet the goals:




Rate-based CO2 emission limits applied to affected EGUs;
Mass-based CO2 emission limits applied to affected EGUs;
State-driven portfolio approach; or
Utility-driven portfolio approach.
Only the last option is genuinely workable, however.
Direct emission limits, whether rate- or mass-based, are illusory approaches, because there is no
technology available that will allow direct emission limits to be effectively imposed and met by
existing plants. CCS is acknowledged to be off the table, and there is no other technology. The
suggested heat rate improvements of 4-6 percent are largely unobtainable because: (1) utilities
have been making efficiency improvements whenever economical and no realistic heat rate
improvements exist, and (2) some heat rate improvements may trigger NSR. Further, even if the
4-6 percent improvement was attainable, it would be inadequate to satisfy direct emission limits
imposed by the goals set by EPA.
A state-driven portfolio approach is unworkable for the states and electric utility industry
because it is based on a misunderstanding that the fleet of generation serving any given state is
interchangeable, regardless of unit ownership or even location within different RTOs. For
example, a state cannot mandate the redispatch of a NGCC unit owned by one utility in one RTO
to offset the retirement or reduced dispatch of a coal unit owned by another utility in another
RTO, even if both units are in that state. Indeed, redispatch from an electrical standpoint,
because the two plants are in separate RTOs, could not be effectively achieved.
Moreover, a state cannot compel a multi-state utility (such as a municipal joint action agency) to
reallocate to that state’s portfolio the renewable energy credits or energy efficiency credits from
the utility’s out-of-state generation resources that the utility—or another state—is using toward
compliance with the rule in the other state. Indeed, as described below, such action by EPA and
the state would raise the prospect of an unconstitutional taking of the utility’s property or
abrogation of the utility’s contracts.
Instead, the only “option” of the four state approaches that can possibly work —legally and
practically—is the utility-driven portfolio approach. Within each state, each utility’s CO 2
emissions should be evaluated against the building blocks based on the ability of each utility to
implement those building blocks within its own system. Furthermore, in the event that a state
213
has excess credits, those should be the property of the utility and available for it to use toward
compliance in another state.
A.
The Proposed Rule’s Allocation of Renewable Energy Credits Potentially
Raises Constitutional Concerns Under the Fifth Amendment Takings
Clause.
The Fifth Amendment to the Constitution states: “[N]or shall private property be taken for
public use, without just compensation.”401 That Clause is made applicable to the States by the
Fourteenth Amendment.402 The Proposed Rule’s approach of allowing states to use utilityowned renewable energy and renewable energy credits (RECs) to meet a state CO 2 goal creates
the potential for an unconstitutional taking of property without just compensation. Specifically,
it would allow “… for renewable energy measures, consistent with existing state RPS
[Renewable Portfolio Standards] policies, a state…[to] take into account all of the CO2 emission
reductions from renewable energy measures implemented by the state, whether they occur in th e
state or in other states….”403
Furthermore, by suggesting that a state with a renewable measure in place that causes renewable
investment in another state has the right to claim those out-of-state renewables as a CO2 offset in
its plan, the Proposal doubles down by reaching across state borders to confiscate renewable
energy and RECs of utilities and developers. The state-driven portfolio approach is one which
EPA identifies as a structure for states to use to develop their compliance plans. 404 Under such a
plan, individual states must assume the responsibility for meeting their specific CO 2 reduction
goal established by EPA, which is based on the generation located within each state’s borders.
Under the proposal, states are encouraged to use renewable energy as one of four building blocks
to meet the compliance goal. Indeed, on average, EPA relies on renewable energy for more than
30 percent of the CO2 reduction to be achieved. 405
While the Proposal touts flexibility and offers “options” to states in developing compliance
plans, the fact is that it sets up a structure in which states are empowered to confiscate for their
401
U.S. Const. amend. V.
See Chicago, B. & Q. R. Co. v. Chicago, 166 U. S. 226 (1897).
403
79 Fed. Reg. at 34,921-34,922; EPA, State Plan Considerations at 34, 37 (Technical Support Document).
404
79 Fed. Reg. at 34,901; State Plan Considerations at5, section II.
405
Brattle Group policy brief, EPA’s Proposed Clean Power Plan: Implications for States and the Electric Industry,
June 2014, p. 3, Table 1, available at
http://www.brattle.com/system/publications/pdfs/000/005/025/original/EPA%27s_Proposed_Clean_Power_Plan__Implications_for_States_and_the_Electric_Industry.pdf?1403791723
402
214
own regulatory goals renewable energy and RECs—which are the property of utilities and
developers—and use them to satisfy their individual CO2 goal. Under a state-driven portfolio
approach, state plans which include renewable energy for compliance will open the door for
states to simply count all the renewable energy generated in the state and the accompanying
RECs406 to meet the compliance goal. As a result, the utility that owns the renewable
energy/RECs will be prohibited from using those resources elsewhere because the Proposal
expressly prohibits “double counting.”407 This is especially true in states that have a renewable
energy mandate that already requires Renewable Energy Credit (REC) retirement for compliance
with that separate state goal.
For example, in Minnesota, the state has imposed a Renewable Energy Standard (RES). It is
served by a variety of fossil and renewable generation, both in-state and out-of state, and by a
variety of utilities based in and outside of Minnesota. However, Minnesota’s CO2 goal is based
on the generation located only in the state. The proposal allows Minnesota to structure its state
plan to count toward compliance all of the RES RECs used in the state by out-of-state entities.
This is problematic for an APPA member, MRES, a joint action agency that provides power to
Minnesota public power utilities and has no generation contributing to the CO2 emissions located
within the state. MRES does, however, provide RE to its member utilities based on the state’s
mandate. That RE and the associated RECs come from contracts between MRES and wind
developers, and has been bought and paid for by MRES members in not only Minnesota, but also
Iowa, North Dakota, and South Dakota. Furthermore, that RE is located in several states,
including Minnesota, Iowa, and North Dakota. Allowing the state of Minnesota to offset the
emissions of its in-state EGUs with RE from a utility like MRES that does not even emit any
CO2 in the state appears to constitute a taking of MRES property for a public purpose without
any compensation.
While MRES may have an RES compliance obligation in Minnesota, that does not entitle EPA to
authorize the state of Minnesota to take the RE paid for by MRES members and their customers
(which has not been used to meet Minnesota’s RES, i.e. excess RECs) to meet its state goal to
offset the CO2 emitted by others. Furthermore, if neighboring states such as Iowa and North
Dakota where MRES has contracts for RE and RECs take a similar approach, the same
renewable energy and RECs could be claimed by multiple states to meet their state compliance
plan under EPA’s flawed reasoning that a state policy that encourages the construction of
renewables can claim credit for those renewables even if they are located out-of-state. This sets
406
Under this scenario, it is not clear whether a state plan would honor contracts under which in-state renewable
resources and RECs are sold to an out-of-state entity. The preamble is silent on this point, and does not indicate any
indication that it recognizes that the RE/RECs are the property of utilities and not states.
407
79 F.R. at 34,922.
215
up disputes between states regarding which state’s policy effectively induced the construction of
the RE. Under either case, this element of the Proposed Rule is appears to be unconstitutional
taking of private property without compensation, in violation of the Fifth Amendment.
B.
The Proposed Rule’s Allocation of Renewable Energy Credits Potentially
Raises Constitutional Concerns Under the Article 1 Contracts Clause.
The Contract Clause in Article 1, Section 10, clause 1 of the Constitution states: “No State shall
… pass any … Law impairing the Obligation of Contracts[.]” 408 By authorizing state plans that
enable states to take the renewable energy and/or RECs of utilities and others, the Proposed Rule
also potentially violates the Contracts Clause. APPA member MRES has several contracts with
a number of individual entities for the output of wind projects and their associated RECs, totaling
85 MW.409 A state-driven portfolio approach that adopts EPA’s suggestion to use the renewable
energy and RECs located in the state to satisfy a state goal will take that RE and RECs out of the
hands of the purchaser, such as MRES, and into the hands of the state to meet its objectives. The
Proposed Rule authorizes state plans that would have this very result.410
The Proposal authorizes states to take RE and RECs from their owners by disregarding the
contractual rights of both the seller and the purchaser. It vitiates the obligation of the seller to
deliver the RE and RECs to the purchaser, and substitutes the state as the beneficiary of the
renewable contract (again, without compensation). This constitutes an undeniable “substantial
impairment of a contractual relationship.” 411 The construct of the state-driven portfolio
approach, including its provisions allowing states to interfere with existing contracts for RE and
RECs—both in-state and out-of-state—appears to violate the Contracts Clause.
Specifically, EPA, in the Proposed Rule, essentially invites states to create a substantial
impairment in the renewable energy contracts of utilities. First, it is undeniable that many
utilities have such contractual relationships with renewable power producers, and that a
regulation that allows the state to step in and use that RE to meet its CO 2 reduction obligation
would constitute a change in law that impairs that contractual relationship. That action satisfies
the first of the two elements in the test of a Contract Clause violation. 412 The second element,
whether the impairment is substantial, while typically the subject of controversy, is also
408
U.S. Const., art. I, § 10, cl. 1.
MRES also has a contract for non-emitting nuclear power from the Point Beach facility in Wisconsin, along with
a portion of the environmental attributes associated with that power.
410
79 Fed. Reg. at 34921-34922.
411
See General Motors Corp. v. Romein, 503 U.S. 181, 186, 112 S.Ct. 1105, 1110, 117 L.Ed.2d 328, 337 (1992).
412
See id.
409
216
undeniable.413 Where the regulation establishes the mechanism for the state to unilaterally use a
utility’s resources for its own benefit, there can be little doubt that there is a virtual “destruction
of contractual expectations” and thus, a substantial impairment. 414
These identified potential violations of the Fifth Amendment’s Takings Clause and Article 1’s
Contracts Clause raise constitutional concerns about the Proposed Rule that EPA needs to
address before finalizing it. The agency must also reexamine the Proposed Rule to address any
other potential constitutional defects, such as Commerce Clause issues, created by its crafting of
optional compliance paths for states.
XXXII. Conclusion
APPA urges EPA to withdraw and re-propose the rule in a manner that comports with its
statutory authority. If EPA decides instead to move forward with this Proposal, then APPA
strongly recommends that the Proposed Rule be modified to:









Allow states to choose a baseline that accurately reflects their unique circumstances.
Provide full credit for investments already made that reduce or offset CO 2 emissions.
Fix the errors and revise the assumptions in the computations of the four building blocks
in a manner that reflects what can realistically be accomplished and ensures greater
equity among the states.
Provide a streamlined process for new source review determinations and stipulate that an
EGU’s energy efficiency upgrade under a state compliance plan should be considered
greenhouse gas Best Available Control Technology for Prevention of Serious
Deterioration determinations.
Remove under-construction nuclear units from the relevant state baselines.
Allow all non CO2-emitting generation resources to be used for compliance.
Provide states with more time to develop state compliance plans.
Provide more guidance on the development of multi-state plans and interstate
agreements.
Eliminate the interim reduction goal and allow states to determine the emission reduction
trajectory (glide path) to reach their final reduction goal.
413
Id.
Energy Reserves Group, Inc. v. Kansas Power & Light Co., 459 U.S. 400, 411, 74 L. Ed. 2d 569, 103 S. Ct. 697
(1983).
414
217



Allow a state’s final reduction goal, the year to achieve that goal, and/or the glide path to
be adjusted based on the discovery of material changed circumstances, with the burden of
so demonstrating placed on the state.
Include and allow mechanisms to ensure that potentially regulated entities have the
maximum degree of flexibility to comply with state plans at reasonable cost, including
additional reduction or avoidance measures from the energy sector.
Provide for the establishment of a reliability “safety valve” to ensure that compliance
with mandated emission reduction requirements does not inadvertently i mpair system
reliability or conflict with NERC standards.
Such modifications taken together would improve the workability and affordability of the final
rule. APPA appreciates the opportunity to provide these comments and thanks EPA for both the
additional time to prepare comments and for their accessibility to discuss relevant issues and
concerns. APPA looks forward to continuing to work with the Agency on the development of its
final rule.
XXXIII. Attachments
1. APPA’s Meeting the Challenge: Public Power’s Commitment to Reducing Greenhouse
Gases brochure (2014)
2. Aspen Energy’s 2010 Implications of Greater Reliance on Natural Gas for Electricity
Generation
3. Markets Matter: Expect a Bumpy Ride on the Road to Reduced CO 2 Emissions by Cliff
Hamal, Navigant Economics (May 2014)
4. CCS Chart Regarding Adequate Demonstration of Technology for BSER (updated
November 2014)
5. APPA Capacity Markets Fact Sheet (2014)
6. Reducing GHG Emissions under EPA’s Section 111(d) Guidelines from Canadian
Electricity Association (CEA)
7. APPA’s Explanation of Rate Impact Analysis
8. Letter from House Science Committee to EPA Administrator (August 13, 2014)
9. Wyden-Dow $80 Billion list
218
Submitted by:
James J. Nipper
Senior Vice President, Regulatory Affairs and Communications
American Public Power Association
2451 Crystal Drive, Suite 1000
Arlington, VA 22202-4804
202-467-2931
[email protected]
Theresa Pugh
Director of Environmental Services
American Public Power Association
2451 Crystal Drive, Suite 1000
Arlington, VA 22202-4804
202-46-2943
[email protected]
Desmarie M. Waterhouse
Director of Government Relations and Counsel
American Public Power Association
2451 Crystal Drive, Suite 1000
Arlington, VA 22202-4804
202-467-2930
[email protected]
Alex Hoffman
Energy and Environmental Services Manager
American Public Power Association
2451 Crystal Drive, Suite 1000
Arlington, VA 22202-4804
202-467-2956
[email protected]
219