76 South Main Street
Akron, Ohio 44308

330-761-4482
Fax: 330-384-5433

 

Raymond L. Evans, P.E.
Vice President,
Environmental and Technologies

December 1, 2014

EPA Docket Center, US EPA

Docket ID No. 
US. Environmental Protection Agency
Mail code: 28221T

1200 Avenue, NW
Washington, DC 20460 7

Dear Sir or Madam:

Re: Docket No. EPA-HQ-OAR-2013-0602
FirstEnergy Corp. Comments on Proposed Carbon Pollution Emission Guidelines for

Existing Stationary Sources: Electric Utility Generating Units

Please ?nd enclosed FirstEnergy?s comments on proposed ?Carbon Pollution Emission
Guidelines for Existing Stationary Sources: Electric Utility Generating Units.?

If you should have any questions, please contact Ms. Michele Somerday at (330) 761-4128 or
email at or Mr. Michael Jirousek at (330) 384-5744, or email
at 

 

Sincerely,

rle/mls
Enclosure
By e?mail: gov

FirstEnergy Corp. Comments on EPA’s proposed
Carbon Pollution Emission Guidelines for Existing Stationary Sources:
Electric Utility Generating Units

INTRODUCTION
FirstEnergy Corp. (FE) is a diversified energy company dedicated to safety, reliability and
operational excellence. Its 10 electric distribution companies form one of the nation's largest
investor-owned electric systems, serving customers in Ohio, Pennsylvania, New Jersey, West
Virginia, Maryland and New York. Its generation subsidiaries currently control nearly 18,000
megawatts of capacity from a diversified mix of scrubbed coal, non-emitting nuclear, natural gas,
hydro and other renewables. The majority of our generation is merchant.

FE has already achieved significant reductions of CO2 emissions. As a company, we expect to
achieve a 25% reduction below 2005 levels by 2015. Even so, it is unclear if the company will
get credit under the Clean Power Plan (CPP) for any of these reductions even though the EPA
press statements emphasized that the rule’s goal is to reduce power sector emissions 30% below
2005 levels by 2030.

In 2009, the President announced his goal of a 17% reduction below 2005 levels by 2020. The
power industry is on track to meet its share of that target. Over the long term, we see dramatic
long term emission reductions in our sector very much in line with the President’s 2050 goals
given the average retirement age of a coal plant, the transformative shift to gas due to the
enormous domestic resource now economically recoverable, MATS and other proposed EPA
rules. Given these facts, we question the need for the current structure of the proposed rule.

The nation’s electric system has been developed and maintained on the core principles of
reliability and affordability. The current emissions trajectory of the electric generation sector
would suggest that reliability and affordability can come with significant reductions in
emissions. However, the proposed Clean Power Plan puts no emphasis on either affordability or
reliability and, in fact, deemphasizes and, depending upon implementation, punishes both. For
example, existing nuclear power is the most reliable, affordable and emission free source of

1

 electricity today, yet the proposal does not appropriately recognize any of these attributes and
actually devalues it in comparison to other generation that is less reliable, less affordable and
relies on quick response backup power that comes with CO2 emissions. It is crucial to maintain
diversity within our generation fleet going forward in order to hedge against potential price
increases and supply disruptions for any particular fuel. From a reliability perspective, it is
essential that base load generation (coal and nuclear) remain a feasible and cost-effective source
of generation to meet existing and future energy needs.

In structuring this rule, EPA doesn’t appear to have fully vetted issues with regard to the broader
system that will be impacted by this rule, such as; transmission capability of both electricity and
natural gas; general infrastructure upgrades; energy storage; and even the contribution of fuel
diversity to reliability and affordability of the system, to name a short few. Virtually all of these
broader system issues fall under the jurisdiction of FERC and/or state utility regulatory
commissions. Given the complexities of the broader system which EPA does not have
jurisdiction over, we believe that prior to EPA approval of any state implementation plan, FERC
(and/or the relevant regional transmission organization (RTO)) should first certify that the plan
will not adversely impact the broader energy system or degrade reliability. Because the system
interconnects multiple states, the reliability impact in one state cascades to an entire region, so it
is vital that FERC/RTO play a role in fully understanding the impact of each individual plan and
be responsible for certifying its impact before EPA approves and a state implements its plan.

FE is also concerned that the Best System of Emission Reduction (BSER) has been developed
without consideration to how electricity market structures are not monolithic throughout the
states. Economic decisions with regard to investment in a unit or other infrastructure will vary
dramatically depending upon whether those occur in a regulated market or in a competitive
merchant market. For example (and further discussed in Block #1 comments), economic
tolerance for investment in heat rate improvements differs significantly depending upon whether
that investment is subject to a regulated rate of return from a state public utility commission
(PUC) or whether the return is totally dependent upon market pricing. While EPA has developed
state specific BSER emission rates, it does not take into account the differing market dynamics in
each state. As noted above, a heat rate that may be economically achievable in a “regulated”
state may not be achievable in a “competitive” state. To ensure fairness and accuracy, EPA

2

 should reconstruct BSER calculations based on the differing market dynamics in each state.
Setting an accurate state specific rate requires such attention to detail.

And lastly as a general comment, FE believes that the rule is faulty in that while EPA relied on
the fact the electric system is an inter-connected system and thus requires an interdependent
approach laid out in the proposal, EPA then developed specific BSER building blocks as
independent silos without calculating their interconnectivity. FE will address this more
specifically in the comments below.

SPECIFIC
FE appreciates the opportunity to further comment on EPA’s Proposed Carbon Pollution
Emission Guidelines for Existing Stationary Sources: Electric Utility Generating Units (EGU).
These comments focus on four key areas of concern:
1) Nuclear
2) Best System of Emission Reduction
3) Maximum State Flexibility
4) Clean Air Act Authority

Nuclear
In its development of the state goals through application of BSER, EPA assumed the license
renewal of all existing nuclear units up to a final life span of 60 years. The license renewal
process is an extremely thorough, multi-year endeavor and, as with any permitting process, the
outcome is certainly not predetermined. NRC cannot commit that it will approve any application
prior to the end of the exhaustive public process. To do so would be inconsistent with the law
and overall good government. EPA cannot and should not presume a licensing outcome that is
currently unknown. EPA’s final 111(d) rule should exclude nuclear units whose license expires
prior to 2030 from its calculation of BSER. Consistent with EPA’s treatment of new nuclear
plants, any unit whose license expires prior to 2030, and receives a license renewal approval
after 2012 should be considered a “new” nuclear unit for the purpose of compliance.

Also with regard to the treatment of nuclear, EPA determined that 5.8% of all existing nuclear
units, regardless of location, are at risk of economic shutdown. At risk nuclear plants vary state

3

 to state, largely dependent upon whether they operate as a merchant unit or a unit regulated by a
PUC. As a result, EPA improperly represented at risk nuclear capacity in setting the standards
for states that have existing nuclear capacity, by applying a uniform 5.8% in each state regardless
of whether a specific unit in a state is at risk for an early closure.

Best System of Emission Reduction
EPA’s interpretation of BSER includes actions that are outside of its jurisdiction and legal
enforceability (i.e., “outside the fence measures”). In fact, no federal agency has statutory
authority to require or enforce the emission reductions contemplated in Building Blocks #3 or #4
(with the possible exception of the NRC authority to regulate nuclear generation). Building
Block #1 is the only block that EPA has authority to regulate. The Clean Air Act does not grant
EPA authority over Building Blocks #2, #3, and #4, which were used to develop and establish
state compliance goals. These Building Blocks are primarily the domain of the state and/or
Regional Transmission Organizations as granted by FERC or the Nuclear Regulatory
Commission (NRC). The Clean Air Act (CAA) does not provide EPA authority to implement or
enforce such energy system programs, nor can a state provide authority to EPA (for programs
which do not currently exist)1 by including certain compliance methods (such as renewable
energy standards or energy efficiency requirements) in a State plan ultimately approved by EPA.
EPA, similarly, lacks authority to fully utilize the building blocks that are the basis of its own
proposal to develop and enforce a federal implementation plan, if it becomes necessary.

EPA has a statutory obligation to ensure that BSER is adequately demonstrated and to show that
the state emission rate goals are achievable, particularly in light of the interconnected nature of
the power system. EPA should ensure that the state emission rate goals in any final rule reflect:
(1) an evaluation of the four BSER Building Blocks to properly reflect the interrelationships of
the various options and potential for impact on power grid reliability and affordability and (2)
appropriate assumptions and conclusions about the level of reductions achievable by each
Building Block. Specifically, Building Blocks #1 and #2 are diametrically opposed; therefore,
both should not be included in the BSER calculation.

1

Gina McCarthy, Administrator U.S. Environmental Protection Agency responses to questions in a U.S. Senate Committee on
Environment and Public Works, July 23, 2014, hearing entitled, "Oversight Hearing: EPA’s Proposed Carbon Pollution
Standards for Existing Power Plants."

4

 In regard to Building Block #2, increasing the utilization of natural gas combined cycle (NGCC)
units will displace coal-fired EGU output and coal-fired EGU heat rates will actually increase as
a result, increasing their CO2 emission rate.2 Coal-fired EGUs are designed to be most efficient
when operated in a steady state at their full load capacity. Increases to the NGCC fleet capacity
factor will necessarily relegate coal-fired EGUs to load following service resulting in more time
operating at unstable and generally lower loads where they are less efficient. Load following
service will also lead to an increase in the number of startups and shutdowns experienced by the
coal-fired fleet. Coal-fired unit startups are lengthy and inefficient operating regimes that will
further increase coal-fired EGU heat rates and CO2 emissions.
In addition, efficiency is poor for low generation levels (a connected plant that is operating at
zero MW output still has to supply station loads) and increases with the level of generation, but
at some optimum level it begins to diminish3. Most power plants are designed so that the
optimum level is close to the rated output.

When capital investments (i.e. heat rate improvements) are made in merchant markets, investors
carefully consider whether the forward looking revenues will cover the costs of the investment.
Given the uncertainties that Building Block #2 introduces into how coal units will be dispatched,
it is unlikely that investors relying on the market for recovery of investments will choose to
invest in these improvements. This dynamic does not exist for regulated generators and points
out how this proposed rule exacerbates the inequity between regulated and restructured states.

Similarly, EPA has also overlooked the negative impact on gas-fired EGU efficiency with
respect to load following units. Under Building Block #2 of the Proposed Guidelines, EPA
would require 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 as well as higher heat
rate simple cycle combustion turbines. This raises a concern that there may not be sufficient
load following generation capable of meeting the load following needs of transmission grids,

2

Most power plants are designed such that when the unit operates at its designed capacity, efficiency is also optimized.
Therefore, reduction of coal-fired units’ output, because of increased utilization of NGCC, will result in degradation of the
effective heat rate of coal-fired EGUs. The result will be an increase in the rate of carbon dioxide emissions from coal-fired
EGUs that see their output dispatched to lower than optimal levels, which is counter to the goal of the proposed rule.
3 Source: http://home.eng.iastate.edu/~jdm/ee553/CostCurves.pdf

5

 such as PJM, as NGCC units typically follow load at an order of magnitude faster than coal-fired
generating plants. A greater utilization of natural gas, oil peaking, or addition of new natural gas
peaking units would have higher heat rates than NGCC plants and some baseload coal
plants. Reliance on peaking units to fill the traditional load following mission could result in an
increase in CO2 emissions that is again contrary to the intent of the proposed rule.
EPA rejected natural gas co-firing or conversion at coal-fired steam EGUs in calculating BSER
stating that “…other approaches could reduce CO2 emissions from existing EGUs at lower cost”
and “…EPA has not proposed at this time to include this option in the BSER and has not
incorporated implementation of the option into proposed state goals.” EPA solicits comment on
whether this option should be considered part of BSER (Fed. Reg. 34876).

FirstEnergy agrees that natural gas co-firing or conversion of coal-fired steam EGUs should not
be considered in determining the BSER due to the impacts on EGU performance and the
availability and delivery of natural gas supplies.

Natural gas combustion will result in higher tube metal temperatures in the furnace and
convection pass than is seen with coal combustion. A unit derate, typically 85%4 of maximum
continuous rating, may be needed to keep heat transfer surfaces within the range of temperatures
for which they are designed and avoid modification or upgrade of materials. Redesign and
replacement of furnace tubing and components such as superheaters, reheaters and other
components would be needed to continue to achieve maximum continuous rating.

The availability of natural gas supply and delivery to EGUs is critical to reliable operation.
Unlike coal, natural gas cannot be stored on site, so any interruption in supply results in
immediate shutdown. The extreme colder than normal conditions, termed the polar vortex,
experienced by many regions of North America in January, 2014, as well as the Southwest Cold
Weather event of February, 20115, exposed the various challenges with fuel supply and delivery

4 Technical Assessment Guide (TAG®)-Power Generation and Storage Technology Options: 2012 Topics. EPRI, Palo Alto,
CA: 2013. 1024063
5 Report on Outages and Curtailments During the Southwest Cold Weather Event of February 1-5, 2011: Causes and
Recommendations, Staffs of FERC & NERC, 2011

6

 related to increased reliance of the power industry on natural gas, according to NERC. High
demand for natural gas exceeded the delivery capacity of the gas transportation system and
resulted in curtailment of fuel delivery to some power plants. The lack of natural gas fuel supply
resulted in extremely high market pricing for electricity and the threat of rolling blackouts for
certain regions.

In the proposed rule, EPA states that it does not propose to find that carbon capture and storage
(CCS) is a component of BSER for CO2 emissions from existing fossil fuel-fired EGUs. EPA
solicits “comment on all aspects of applying CCS to existing fossil fuel-fired EGUs (in either full
or partial configurations)”. (Fed. Reg. at 34876). FirstEnergy agrees CCS should not be a
component of BSER for CO2 emissions from existing fossil fuel-fired EGUs. Partial CCS has
not been adequately demonstrated at full scale for existing units, is not technically feasible and
cannot be implemented at costs that are reasonable.

Building Block #1
Building Block #1 assumes that all affected units can achieve a 6% heat rate improvement (HRI).
FE believes that the methods used to establish the 6% HRI are flawed and set an unrealistic
target.

Heat rate degradation is a normal occurrence in steam plants that results from normal aging of
plant equipment and systems and can be exacerbated by changes in operational duty cycles such
as increased start-ups, shut-downs, operation at other than steady state load and run time at
lower-than-rated capacity. Degradation also occurs when new systems that draw large amounts
of auxiliary power, such as environmental controls, are added to a plant. Changes in coal supply
to meet environmental requirements can also degrade heat rate due to changes in fuel quality.
Most of these types of degradation are not economically recoverable.

The proposed rule includes 4% HRI related to maintenance and operating practices that is based
on an unsubstantiated statistical analysis. Serious flaws in that statistical analysis include several
coal-fired units listed with gross efficiencies over 42% which is impossible; while others are
listed with heat rates under 20% – very unlikely. Another serious flaw is EPA's assumption that
30% of the heat rate variance from the top decile is associated with controllable operating and

7

 maintenance practices that would be cost effective to perform. EPA fails to consider that what
may be cost effective in a state with regulated markets with a guaranteed rate of return on
investments may not be “cost-effective” in states with competitive markets where market prices
determine what is “cost-effective.” EPA only adjusted the data for ambient temperature and
capacity factor, however, there are many factors other than maintenance and operating practices
that likely contributed to heat rate variability on the units in the dataset. These include, among
others, heat rate improvement projects (that results in a double impact because they are included
as potential operating practice heat rate improvements when, in fact, they are heat rate
improvement opportunities that have already been completed); capacity factor changes within
the EPA bands; fuel switching; and additions of environmental controls. FirstEnergy is not
aware of any work EPA performed to 1) validate their statistical approach and variability of the
data set, 2) validate the assumption that 30% of the heat rate variability is due to operation and
maintenance practices, or 3) evaluate that such practices, are cost effective – both in a
competitive and regulated market structure. The proposed rule’s use of a 4% HRI for
maintenance and operating practices is significantly overstated which results in a BSER that is
unachievable.

EPA relies on the Sargent and Lundy (S&L) report prepared for EPA in 2009 to justify an
additional 2% HRI from future plant modifications. A cursory review of the S&L report shows
that it outlines a group of upgrades that have been known and practiced by the industry for years.
S&L specifically stated in their report that “[t]he primary intent of the study was to focus on
methods that have been successfully implemented by the utility industry.” Since utilities have
already completed the actions that S&L includes in the study at many EGUs or have already
determined them to be inapplicable to their EGUs, the additional 2% HRI is unachievable and
should be removed from Building Block #1.6

FE’s analysis concludes a total heat rate improvement up to 1.5% from current operating
parameters is the maximum attainable at an economically justifiable cost for a merchant unit. An
6 Additionally, from the report: “S&L cautions that the costs presented herein are not indicative of those that may be expected
for a specific facility due to variables such as equipment, material, and labor market conditions and site specifications.” And
further that, “The costs should not be used as a basis for project budgeting or financing purposes.” Regardless of the specific
statements by S&L to the contrary, EPA still used the S&L costs as a basis for the cost of compliance. The report characterizes
the costs as “order of magnitude”. There is a substantial difference between $20/ton and $200/ton of CO2 which is beyond the
cost of a new NGCC plant.

8

 S&L case study found a 4% heat rate improvement, including both maintenance and uprate
projects, was possible. However, over half of the heat rate improvements were due to an entire
turbine steam path replacement. Only a 1.7% heat rate improvement could be achieved without
that turbine steam path replacement which many plants have already performed. The second
S&L case study found a 1.2% HRI and included a number of improvements the utility had
already performed – reinforcing the point that plants are already performing many of the heat
rate improvements S&L described in the normal course of business. In addition, the proposed
rule’s assumptions ignore that any heat rate improvement recovered through maintenance or heat
rate improvement projects will deteriorate between maintenance outages (that is how it became
recoverable in the first place). In addition, heat rate improvements attained against the 2012 base
year will be significantly offset by reduced coal plant capacity factors associated with EPA’s
Building Block #2 that shifts dispatch from coal-fired units to NGCC units and future additions
of pollution control devices that utilize station power.

FE agrees with EPA’s findings that total potential CO2 reductions achievable through heat rate
improvements at non coal-fired units are small compared to the potential at coal-fired units and
should not be used in the setting of BSER. (Fed. Reg. 34877)

In general, current market power prices in competitive markets do not support making many of
these capital investments, such as a number of heat rate improvements, and could lead to further
shut downs of coal plants beyond EPA’s assumptions. In the regulated markets, additional costs,
if approved by the state PUC, will be passed on to the customer through higher prices.

Any final rule that relies on heat rate improvements must expressly provide that those changes
do not trigger NSR or NSPS requirements. The looming threat and cost of NSR will further
reduce the economic tolerance in the decision making process. In addition, NSR would also
introduce further time delay that could impede the ability of a state to meet compliance
deadlines.

Building Block #2
Building Block #2 assumes the ability to shift generation from coal-fired plants to NGCC plants,
thereby raising the average NGCC plant capacity factor to 70%. This shifting of generation will

9

 reduce the average efficiency of the coal-fired units. EPA has shown a strong relationship
between lower capacity factors and lower coal plant efficiency. Therefore, one of the impacts of
Building Block #2 will be to offset some of the efficiency improvement efforts taken by coalfired plants to meet Building Block #1, thereby making Building Block #1 even harder to
achieve. This reinforces the necessity to analyze BSER Building Blocks in an integrated manner
rather than individually.

FE operates in the PJM Interconnection RTO which recorded the following NGCC capacity
factors: 2010 – 28.8%, 2011 – 46.8%7, 2012 – 60.4%, 2013 – 51.6%8, and January – June 2014
– 49.4%9. In PJM, capacity factors for NGCC have never approached 70%. In fact, when
natural gas prices were volatile from 2000 through 2008, NGCC capacity factors were typically
well below 20%. While NGCC capacity factors were their highest in 2012, that year was
characterized by milder than normal weather, reduced economic activity, and natural gas prices
that reached low levels of $2/mmbtu not experienced in over a decade. This combination of
factors resulted in NGCC capacity factors that were an anomaly that year.

EEI analysis indicates that the average utilization rate of NGCC capacity in 2012 was 46 percent.
Only 10 percent of these units operated at annual utilization rates of 70% or higher and 19% of
these units operated at utilization rates of at least 70% over the summer season. So, while a 70%
utilization rate may be “technically” feasible, it is unrealistic based on operational experience.
EPA appears to have based its proposed increase in utilization rate on analysis of only 10% of
the NGCC fleet.

Since 2012, natural gas prices have rebounded and have remained around $4/mmbtu. For NGCC
capacity factors to reach the 70% range, natural gas prices would need to return to record low
levels for NGCC units to be economically dispatched in RTO markets such as PJM. Therefore,
from an economic dispatch perspective, EPA’s assumed 70% NGCC utilization rate for BSER is
unrealistic and based on faulty assumptions and expectations.

7

Monitoring Analytics 2011 State of the Market Report for PJM, page 111
Monitoring Analytics 2013 State of the Market Report for PJM, page 188
9 Monitoring Analytics 2014 Quarterly State of the Market Report for PJM, page 181
8

10

 EPA’s assumed 70% utilization rate for all NGCC units ignores permit conditions that may
create regulatory restrictions or artificial barriers to full operation, often referred to as “synthetic
minor” permits. Permits can impose limits on emissions, fuel consumption, or hours of
operation for regulatory reasons or for ease of permitting. NGCC units may not be able legally
to maintain a 70% capacity factor or be able to maintain a high enough capacity factor to help
bring the state’s average capacity factor to a 70% level. This legal impediment combined with
the physical inability of some plants to operate at a 70% capacity factor is a fatal flaw in
Building Block #2.

The assumption that existing pipeline infrastructure can support increased NGCC capacity
nationwide also seems to ignore regional disparities as well as the realities of pipeline markets.
The electric power sector competes for pipeline services with two other major natural gas
consumers: local gas distribution companies serving residential and commercial sectors and
industrial consumers. Gas pipelines are very highly subscribed. FE’s recent survey of pipelines
within PJM determined that Texas Eastern is fully subscribed and Tennessee Gas is unable to
offer “No-Notice” service due to lack of available storage.

Congestion in the electric transmission system does not necessarily coincide with congestion in
the natural gas transmission system. Ongoing changes in gas-electric coordination and direct gas
consumption should be factored into determining “reasonable” levels of NGCC utilization. In
fact, recent experience has shown that gas supplies may not be available for NGCC generation in
shoulder months due to the need to replenish storage or in winter months when pre-empted by
local distribution companies. While the existing natural gas pipeline infrastructure was able to
support 2012 peak utilization, this may not be sufficient evidence that a 70% utilization rate will
be achievable for every state’s existing NGCC fleet in the future.

Operating NGCCs at 70% capacity will be the equivalent of adding 5,200 MW of generation into
the system on an average basis. An assumed heat rate of 8,000-9,000 BTU/kWh equates to an
increase in gas usage of approximately 1 BCF/day, or 365 BCF/year not including any new
combined cycle generation or a 6% increase in Marcellus shale gas production. The
infrastructure to move this additional gas is currently not constructed and most construction
projects are centered on getting gas out of the system not delivery to generation facilities that run

11

 at reduced capacity factors. In fact, PJM reported that over 9,000 MW of gas generation
capacity was offline due to “Confirmed Gas Curtailments” this past January. This equates to
roughly 17% of all gas generating capacity (natural-gas fired generators accounted for 47% of
the unavailable MW).10 FirstEnergy was forced to switch units that historically ran on natural
gas to oil due to the inability of Columbia Gas to supply the units with natural gas on a firm or
intermittent basis.

PJM recently discovered that NGCC units have been “chronically curtailed” over the past six
winters stating that they are currently working on “gas-related contingencies” which would
include switching to oil during significant weather related curtailment events.11
Another study prepared by the staffs at FERC and NERC12 investigated cold weather events in
the southwest in 2011 concluded that at least 12% of the electrical outages attributed to weather
events were actually “occasioned by natural gas curtailments to gas-fired generators and
difficulties in fuel switching.” The authors point out that in some states the priority of
curtailments places the needs of residential and other human needs above those of gas-fired
EGUs. In other words, natural gas-fired generators, including NGCC, will be curtailed before
residential customers and other human services.

Natural gas curtailments cast further doubt on the viability of a 70% capacity factor for NGCC
units assumed in Building Block #2.

NGCC unit capacity factor can only be increased by a market mechanism that forces NGCC
units to offer generation into RTO markets so that they will be dispatched at the proposed rule’s
desired 70% rate. Such a mechanism does not currently exist and there has been no indication
from federal and state regulators that any such mechanism is even being considered. If
implemented, NGCC units would likely be offered as “must-run”, which effectively is a $0 offer.
Must-run offers are uneconomic and well below the NGCC marginal costs. The unintended
consequence of mandating NGCC units as must-run units to meet policy objectives would be a
10

PJM’s May 8 “Analysis of Operational Events and Market Impacts During the January 2014 Cold Weather Events.”
Winter Generation Outage Analysis, PJM Planning Committee, June 5, 2014
12
Report on Outages and Curtailments During the Southwest Cold Weather Event of February 1-5, 2011. Causes and
Recommendations, Staffs at FERC and NERC, August 2011
11

12

 shifting of the supply curve and thus artificially depressing market prices (LMPs) for all
resources.

The Independent Market Monitor (IMM) for PJM has stated that depressing market prices leads
to premature and uneconomic retirements. Coal and nuclear units are already under stress. The
IMM estimates that 14,597 MW of capacity are at risk of retirement in addition to the 24,933
MW that are currently planning to retire.13 The risk of further retirements will only be increased
by additional price suppression due to policy decisions made in any final carbon rules.

PJM analysis concludes that it would likely not be able to meet the winter peak requirement if
comparable generator outages that occurred in January 2014 were to occur in the winter of
2015/2016 coupled with extremely cold temperatures and expected coal plant retirements.14

If market conditions continue to be depressed, generation resources on the margin that are forced
to decide between investments to maintain viability or retirement, will choose retirement. This
will further exacerbate reliability concerns and the volatility of consumer cost.

As far as transmission constraints are concerned, FE offers the following comments from EPRI:

“The changes in the utilization of the various generating plants driven by this proposal
could have a significant impact on transmission reliability due to potential large changes
in power flows across the system and retirement of generation that contributes to
transmission system voltage and frequency performance. The change in generation will
almost certainly require development of new transmission to ensure operational
reliability, but scheduling outages of existing facilities will be difficult if simultaneous
upgrades across many systems are needed such that time lines for commissioning of new
transmission facilities may be delayed. To understand the full reliability, economic, and
financial impacts of the proposed rule, detailed transmission reliability evaluations
should be conducted.”15

13

Monitoring Analytics, 2013 State of the Market Report for PJM, page 1
PJM Capacity Performance, PJM Staff Proposal, August 20, 2014, page 4
15 EPRI Docket ID No. EPA-HQ-OAR-2013-0602
14

13

 Shifting from market-based dispatch of generation to regulatory driven mandates will result in
market distortions and have unintended consequences. The market operates on the principle of
ensuring reliability and affordability, and any changes to system operations that ignore those
principles will by definition degrade reliability and/or affordability. According to an IHS Energy
report “The Value of US Power Supply Diversity,” economic and reliability affects will be felt if
the power supply is arbitrarily changed.

“If the US power sector moved from its current diverse generation mix to the less diverse
generating mix, power price impacts would reduce US GDP by nearly $200 billion, lead
to roughly one million fewer jobs, and reduce the typical household’s annual disposable
income by around $2,100. These negative economic impacts are similar to an economic
downturn. Additional potential negative impacts arise from reducing power supply
diversity by accelerating the retirement of existing power plants before it is economic to
do so. For example, a transition to the reduced diversity case within one decade would
divert around $730 billion of capital from more productive applications in the economy.
The size of the economic impact from accelerating power plant turnover and reducing
supply diversity depends on the deviation from the pace of change dictated by the
underlying economics”.16

In addition, operating NGCC units to a certain capacity factor is currently highly dependent on
the price of natural gas. Economics are what dictate the capacity factors of electric generators.
However, it appears that EPA disregarded the role of economic dispatch and its impact on
capacity factors by utilizing a 70% NGCC unit capacity factor in BSER. Economic dispatch is
what ensures consumers will pay the most affordable rates for electricity. To prioritize capacity
factor over economic dispatch would require changing the current dispatch regime and thus
eliminate affordability as a top priority. Thus, consumers will end up paying more for electricity
than they currently do. We do not believe that is in the interest of our customers.

Finally, mandating baseload operation of NGCC units (at 70% CF) eliminates the availability of
those NGCC units to load follow, serve peak power needs during ramps and on extreme demand

16

IHS Energy, The Value of US Power Supply Diversity, July 2014, page 6

14

 days, and back up intermittent generation. This is especially important since this proposed rule’s
Building Block #3 relies on significant increases in intermittent renewable energy, which is
typically backed up by natural gas generation. Continued displacement of coal and nuclear
baseload resources by historically peaking units, such as NGCC, will result in a shortage of
capacity during peak periods, threatening future reliability.

Building Block #3
EPA must correct numerous faulty assumptions in Building Block #3 and re-calculate state target
rates prior to finalization of this rule.

For example, EPA state emission rates in the East Central Region must be re-calculated due to
the passage of SB310 in Ohio just prior to the publication of EPA’s proposed 111(d) rule. Ohio
SB310 amended energy efficiency and alternative energy resource mandates (including
renewable mandates) by imposing a two-year pause on energy efficiency, peak demand
reduction and renewable energy resource requirements. It also eliminated both the renewable
energy resource in-state requirements and the advanced energy resource requirements, thereby
reducing the former alternative energy resource mandate by half. Due to the timing of Ohio SB
310, EPA did not reflect the impact of this new state law in the BSER Building Block #3
calculations for the East Central Region. EPA must recalculate state goals within the region
based on Ohio’s SB310.

Ohio SB310 states that “because the energy mandates in current law may be unrealistic and
unattainable, it is the intent of the General Assembly to review all energy resources as part of its
efforts to address energy pricing issues. Therefore, it is the intent of the General Assembly to
enact legislation in the future, after taking into account the recommendations of the Energy
Mandates Study Committee that will reduce the mandates in sections 4928.64 and 4928.66 of the
Revised Code and provide greater transparency to electric customers on the costs of future
energy mandates, if there are to be any.” Therefore, EPA’s final 111(d) rule should provide a
mechanism for re-calculating state target rates following future Ohio legislation based on the
recommendations of the Energy Mandates Study Committee.

15

 In developing the target renewable energy generation levels, the EPA calculated a hypothetical
RES requirement for each region by averaging the RES requirement of each state that currently
has an RES requirement within the region. EPA recognizes state expertise in developing
renewable energy goals, thus justifying the use of these goals in calculating BSER goals. The
GHG Abatement Measures Technical Support Document states:

“These state goals and requirements have been developed and implemented with
technical assistance from state-level regulatory agencies and utility commissions such
that they reflect expert assessments of RE technical and economic potential that can be
cost-effectively developed for that state’s electricity consumer”

However, EPA chose to exclude that same expertise by the state of West Virginia simply
because after an exhaustive vetting and legislative process, the state determined that it could not
support a mandatory renewable energy goal. It was the state’s informed decision, developed and
implemented with technical assistance from state-level regulatory agencies, the utility
commission, and interested public and private parties, no different from any other state that EPA
used in the calculation of a region’s renewable potential. By excluding states like West Virginia
that have determined their renewable mandate to be zero, whether through affirmative action or
through a decision not to act on a legislative mandate, EPA disregards its own technical
justification document. EPA cannot and should not pick only those states that have concluded a
specific outcome after study. EPA should accept all states’ “expert assessment of RE technical
and economic potential that can be cost-effectively developed for the state’s electricity
consumer” regardless of what conclusion that expertise leads to. To cherry pick a few states in
order to produce a certain outcome undermines the credibility of EPA’s technical support for this
proposal. Either all states are experts or all states are not. EPA should recalculate the target
renewable energy generation levels under Building Block #3 by including every state in the
calculation, incorporating states such as West Virginia as a zero since it imposes no renewable
energy mandate.

EPA’s approach is also flawed as it does not distinguish between renewable energy that is
generated within the state versus renewable energy imported from a neighboring state. For
example, Washington, DC has a renewable target of 20%, yet it is difficult to imagine 20% of

16

 Washington, DC electricity being generated by allowed renewable energy within the District’s
borders. Washington, DC recognized as much when it allowed for RECs to be procured outside
the District’s borders, but within PJM. And yet, for the purposes of EPA’s proposed rule, it is
assumed that Washington, DC has the potential to achieve a 20% renewable energy requirement.
Renewable energy that is generated within the state is what is most representative of the
capabilities within the state. Thus, to accurately reflect each state’s renewable potential, EPA’s
approach should only be based on in-state renewable sources of generation. EPA should
recalculate state emission rate targets based solely on verifiable in-state renewable sources of
generation.

EPA’s approach is further flawed as it gives equal weight to each state in the region, as opposed
to a weighted average to factor in the different sizes and populations of the various states in the
regions that impact electric consumption and generation. For example, Washington, DC is given
the same weight as Ohio or Pennsylvania whose electric consumption each is 12 to 15 times as
large as Washington, DC and hundreds of times larger in terms of electric generation.

EPA’s approach also ignores its own GHG Abatement Measures Technical Support Document
that provides: “[s]tates within each region exhibit similar profiles of RE potential or have similar
levels of renewable resources.” Clearly landlocked states do NOT “exhibit similar profiles of RE
potential or have similar levels of renewable resources” as states with off-shore capability.
Including New Jersey and Maryland in the same region as West Virginia ignores the obvious
regional differences. For example, Maryland has created a “mechanism to incentivize the
development of up to 500 megawatts (MW) of offshore wind capacity, at least ten nautical miles
off of Maryland's coast” (state of MD website). However, landlocked states like West Virginia
have zero capacity to develop or offer incentives for large scale off-shore renewable projects.
EPA’s Technical Support Document states that the “Northeast region has strong resources offshore” but has placed states with strong off-shore renewable energy capability (i.e. New Jersey,
Maryland and Virginia) in the same region as landlocked states who have zero off-shore
resources. EPA must reconfigure the regions in its proposed rule and recalculate BSER.

EPA’s approach also assumes that renewable programs are emission reduction programs, but the
vast majority include alternative compliance methods or “safety valves” that do not result in

17

 emission reductions. In fact, environmental groups have consistently opposed the use of safetyvalves under the logic that it reduces investment in renewable energy and allows for emissions
that would otherwise not occur without such a mechanism. However, EPA assumes, for the
purpose of setting BSER that a state will achieve its entire renewable requirement through the
procurement of allowed renewable energy and actually achieve the emission reduction used to
calculate BSER goals for individual states. This approach is simply not accurate and results in
artificial inflation of renewable energy assumptions and thus emission reduction assumptions.
EPA must recalculate BSER in a manner that reflects the true emissions impact of all state
renewable energy requirements including calculating the impact of each safety-valve mechanism
(or alternative compliance mechanism). Additionally, within the notice of data availability
(NODA) the EPA requests feedback on ways that state-level RE targets could be set based on
regional potential for renewable energy. EPA relied on historic RE development from the top 16
states. This approach overstates the RE development rate by relying only on data from those
states that have been most successful in developing their renewable generation. Historic RE
development should be based on the experience of all states.

With regard to the nuclear portion of BSER in Building Block #3, please refer to our previous
comments.

Building Block #4
The proposed rule assumes a 1.5% annual Energy Efficiency (EE) gain that is not reasonable.
EPA acknowledges that the projected cumulative EE savings rate are well above the average
savings that most states have actually achieved of 0.58% in 2012. EPA concluded that three
states (AZ, ME, VT) have already achieved the highest level of performance, more than 1.5%
annual incremental retail sales savings. However, EPA failed to explain why AZ, ME, and VT
were successful and how that success can be uniformly duplicated in every other state. EPA
further assumes each state currently below the 1.5% annual savings rate can increase its
incremental savings levels by 0.2% per year. Therefore, EPA assumed that states would start
ramping up EE programs in 2017 in order to reach the target annual EE savings rate no later than
2025.

18

 The proposed rule’s assumptions of 1.5% rate and the 0.20% per year pace of improvement are
too aggressive and unrealistic. In the Greenhouse Gas Technical Support Document (GHG
TSD)17, the 1.5% value is the highest value of the studies referenced. Based on EPRI’s most
recent study, a value of 0.5-0.6% per year is achievable, only about one third of the 1.5% value
used.18 Of the studies referenced, the EPRI study is the more realistic because it is based on a
“bottom up” engineering approach as opposed to the “top down” policy approach performed by
American Council for an Energy-Efficient Economy (ACEEE). The 1.5% annual incremental
savings has only been achieved by three states. Market potential is highly dependent on maturity
of EERS in each state, saturation levels of various programs and technologies, existing level of
state building code standards and what is qualified in each state. The pace of incremental EE
savings slows over time as codes and standards increase and typically, the largest gains are made
earliest in the life of EE programs. As an illustration of this point, in the baseline year 2012
savings values largely were influenced by lighting programs. Widespread adoption of increasing
EISA standards (EISA 2008) have effectively and significantly reduced what can be counted
towards lighting savings. As efficient lighting programs and other technologies saturate
consumer opportunities, there is a diminishing level of potential. The EPRI Study19 reports
potential using a baseline that includes current codes and standards in place at the time the study
was done. The 0.5 - 0.6% incremental annual potential reported from this study will be reduced
by future stricter federal and state standards and local building codes requirements for efficiency.

Projection of the top three states achievements to remaining states is unrealistic and will result in
unachievable and uneconomic goals. It is not appropriate for the EPA to use the experience of 3
out of 50 states to determine a one-size-fits-all nationwide annual incremental savings rate for all
EE programs in all states. Furthermore, the sustainability of past achievements is not guaranteed
going forward, particularly over a long time horizon through 2030. Recently, one of those top
three state’s utility commission, the Arizona Corporation Commission, issued a request for
informal comment on modifying its current rules on energy efficiency to eliminate Arizona’s
aggressive goals of 22% by 2020 and instead incorporates energy efficiency requirements as part

17

Greenhouse Gas Technical Support Document at 5-24
EPRI Report 1025477, “U.S. Energy Efficiency Potential Through 2035”
19 Ibid
18

19

 of a biannual integrated resource planning process.20 The proposed changes also include a focus
of cost effectiveness on ratepayer impacts.21 Arizona Commissioner Gary Pierce was quoted as
saying that

“The rules were set up, and it was pretty easy at first to capture all the low-hanging fruit,
but as we started reaching, these companies, because they are under an order to reach
certain levels of energy efficiency, they were looking for stuff and trying to plug it in no
matter what the costs.”22

This not only highlights that the performance of a limited group of states is not appropriate for
all states, but that the level of savings achieved or projected to be achieved is not necessarily
achievable and shouldn’t be assumed going forward as the high cost to achieve such level of
savings is increasing and causing reconsideration of such requirements due to ratepayer impacts.

Also, Ohio was listed as one of the eleven states that are projected to achieve 2.0% or more by
2020. But recent legislation modified the Ohio targets with the purpose of further study by the
state to ensure that energy efficiency and renewable energy levels are realistic and beneficial to
ratepayers.

“It is the intent of the General Assembly to ensure that customers in Ohio have access to
affordable energy. It is the intent of the General Assembly to incorporate as many forms
of inexpensive, reliable energy sources in the state of Ohio as possible. It is also the
intent of the General Assembly to get a better understanding of how energy mandates
impact jobs and the economy in Ohio and to minimize government mandates. Because the
energy mandates in current law may be unrealistic and unattainable, it is the intent of the
General Assembly to review all energy resources as part of its efforts to address energy
pricing issues. Therefore, it is the intent of the General Assembly to enact legislation in
the future, after taking into account the recommendations of the Energy Mandates Study
Committee, that will reduce the mandates in sections 4928.64 and 4928.66 of the Revised
20 Arizona Corporation Commission, Utilities Division Docket No. E-00000XX-13-0214, November 4, 2014, at R14-2-2404,
Energy Efficiency Goal
21 Ibid at R14-2-2411, Cost-effectiveness.
22 Arizona Daily Star, November 8, 2014, “State Regulators Mull Scrapping Energy Savings Goals”

20

 Code and provide greater transparency to electric customers on the costs of future
energy mandates, if there are to be any.”23

Additionally, Ohio’s legislation creates an opportunity for large customers to opt out of
participating in utility energy efficiency programs.24 FirstEnergy estimates that the potential
volume of customers electing to opt out of efficiency programs will be over a third of its total
sales volume.

A Market Potential Study that was performed for the FirstEnergy Ohio Companies’ Energy
Efficiency and Peak Demand Reduction Filing25 reports that the achievable potential for the
Companies’ territories are in fact less than the current state targets for both the base case and
high case. For the years 2017 through 2026, the average annual incremental savings for the base
case for Ohio Edison Company, The Cleveland Electric Illuminating Company and The Toledo
Edison Company respectively are approximately 0.5%, 0.6% and 0.4%. For the high case, those
values are 0.7%, 0.7% and 0.5%.
Furthermore, in referencing the Pennsylvania 2012 Potential Study26 the GHG TSD27 references
a value of 2.9% as annual incremental achievement. This is incorrect. Table 1-3 and 1-4 from
this Study show program potential of cumulative values for the periods of 2013 – 2016 and
2013-2018 of 2.3% and 3.7%, respectively. On an incremental basis this would be
approximately 0.75% per year (before considering effects of degradation). This is significantly
lower than EPA’s assumption of 2.9%/year. The Pennsylvania value was the highest in the
range of potential studies quoted (see GHG TSD, at Appendix 5-1) used to support EPA’s
assumed 1.5% incremental EE savings.

In addition, rebound effects of EE measures should be considered, particularly in conjunction
with a mass based BSER scenario. More efficient use of energy results in (as stated in GHG
TSD at 5-29) “[a]n improvement in energy efficiency would effectively reduce the cost of a
23

Ohio Senate Bill 310, effective September 12, 2014, Section 3 page 34.
Ibid, Section 4928.6611, page 31 and Section 8, page 37.
25 Market Potential Study Energy Savings and Demand Reduction for Ohio Edison, Toledo Edison and the Illuminating
Company, June 22, 2012, Public Utilities Commission of Ohio Case 12-2190-EL-POR et al. (See Tables 1-1 through 1-9.)
26
Electric Energy Efficiency Potential for Pennsylvania, Final Report, May 10, 2012, Prepared for Pennsylvania PUC
27 Ibid, Appendix 5-1, Table 1
24

21

 service or production input, potentially boosting its demand or production output thus increasing
energy use.” This creates additional growth in demand and energy. Although good for the
economy, this additional demand and energy usage makes a state goal more difficult to achieve.

EPA requested comment on alternative approaches and/or data sources for evaluating costs
associated with the implementation of state demand-side energy efficiency policies (Fed. Reg.
34875). In determining the role of demand side energy efficiency programs as a Building Block
for carbon reduction, cost assumptions should take into consideration that there are correlations
to cost regarding both the level of savings as a percent of sales as well as varying levels of
maturity of state EERS. The EPA assumed first-year net costs of $275/MWh (2011$)28 based on
the 2009 ACEEE national review of data on EE programs costs.29 This study relied on outdated
data from a period of 2001 to 2009 across fourteen states. During this pre-EISA era, low cost
efficient lighting represented a third or more of total savings. For the purposes of Building
Block #4, the time period in consideration for achievement of EE savings is 2017 through 2030.
Four major factors will contribute to higher costs in that future time period than have been
observed in the last two decades: 1) increasing saturation levels from programs in place for
multiple decades, 2) increasing federal and state standards and local building codes which will
effectively make each marginal kWh savings more and more costly to achieve; 3) diminishing
opportunities for marginal kWh savings from new technologies and 4) increasing costs
associated with new technologies. These factors should be taken into consideration when
forecasting costs, particularly, so far out into the future. To further illustrate this point, the
references given in the GHG TSD30 point to additional growing “greenfield” states that have
started programs. These lower costs represent low hanging fruit that will be harvested during the
first implementation cycle and are not representative of savings beyond 2020. For example, after
2020, the baseline for most general service lamps is effectively the CFL. LEDs will never
realize the savings as CFLs did in the residential sector because EISA has effectively saturated
lighting end uses with CFLs and because of the code change in 2020. The incremental savings
of LEDs compared to CFLs is a fraction of that for CFLs compared to incandescent lighting and

28

GHG TSD, pages 5-50
Though not referenced in the GHG TSD, presumably the document referenced is the ACEEE report “Saving Energy CostEffectively: A National Review of the Cost of Energy Saved through Utility-Sector Energy Efficiency Programs,” Fredrich,
Eldridge, York, Witte and Kushier, September 2009
30 GHG TSD, pages 5-51 first paragraph
29

22

 at a significantly higher cost per kilowatt hour of savings. While the EPA has included cost
escalators of 20% to 40%,31 cost escalators alone do not explicitly account for each of these four
factors.

The EPA’s analysis also makes assumptions for participant incremental measure costs to arrive
at total costs of energy efficiency programs. As stated in the GHG TSD32:

“….while program costs are relatively known and consistently reported by the program
administrator, participant costs require significant effort to estimate, and are less
consistently estimated and reported. The ratio between program and participant costs
will vary significantly from one program to the next within a utility’s portfolio.”

There are correlations between participant cost versus program cost and overall levelized cost,
and it is not apparent that these are accounted for by the EPA. For example, a program with high
costs will have a lower ratio of participant costs to programs costs, and conversely a program
with low costs will have a higher ratio. These factors introduce variation from state-to-state and
to assume a 1:1 ratio for all states could cause significant over or under forecasts.
EPA invited comments on all aspects of its goal computation procedure (Fed. Reg. 3489534897). In EPA’s calculation of the contribution of demand side energy efficiency, the
degradation assumption used in developing targets is based on 20 years, while measure life is
based on 10 years –which overstates the potential (see 5-36 GHG TSD). Contrary to EPA
claims, this is not conservative (see GHG TSD 5-38) and results in higher values achievable than
would be calculated if 10 year degradation was assumed in lieu of 20 years. The final rule’s
BSER should be recalculated using 10-year degradation assumptions.

EPA requests comment on Efficiency Measurement and Verification (EM&V) protocols for
energy efficiency (Fed. Reg. 34921). With regard to EM&V protocols, FE supports the
following approach:

31
32

GHG TSD, pages 5-53
GHG TSD, pages 5-51

23

 •

States should be granted the flexibility to determine the protocols for the measurement
and verification of demand side energy efficiency programs

•

No new EM&V protocols which would create undue administrative burden and increases
the cost of energy efficiency

•

Assumptions used by EPA (regarding 2012 achievements for EE) to develop state BSER
goals were based on existing state evaluation methods, therefore compliance EM&V
should be consistent with this methodology

•

Consideration of what can count:
o States should decide what can count towards energy efficiency. For example, in
Ohio, the legislation has specific language that allows CHPs and efficiency
improvements resulting from transmission and distribution investments to count
towards its energy efficiency mandates.
o Energy efficiency should be counted on a gross versus net basis that takes “free
ridership” into consideration.33
o Any savings from changing federal, state standards, and local building codes
should be explicitly supported with protocols defined by EPA or “given” to States
as credits against their obligations.
o Credits for energy efficiency should not be limited to established programs as
long as savings can be measured and verified within accepted protocols.

EPA requests comment on the treatment of export/import power (Fed. Reg. 34922). States
should have ultimate flexibility during implementation, including determining how to treat
export/import power. For energy efficiency savings, states should be able to take credit for
100% of savings regardless of whether they are a net importer or exporter of power. Typically
these programs are funded by state ratepayers or taxpayers, and therefore, the state should
receive credit for the reductions regardless of where the generation is offset.

EPA requests comments on different approaches for providing crediting or administrative
adjustment of CO2 emission rates (Fed. Reg. 34919). In regards to the value of a credit or
33 A free-rider is someone who would have installed an energy-efficiency measure without any program incentives based on the
energy savings, but receives a financial incentive or rebate anyway. Free ridership is very dynamic and changes overtime.
Regardless of free ridership, resulting energy savings are real savings whether or not they can count towards an individual state’s
statutory compliance purposes and regardless of why they occur.

24

 adjustment resulting from energy efficiency, whatever methodology is selected for planning and
compliance purposes should be consistent with how the BSER was calculated to ensure that
BSER is realistic and achievable.

As noted above, achieving the aggressive targets for non-emitting sources is unrealistic.
Combining the unreasonableness of meeting those targets with the limited opportunity for heat
rate improvements and redispatch in Building Blocks #1 and #2, results in serious consequences
both in the short-term and long-term. In their comments, EPRI has determined that any
shortfalls within Building Blocks #3 or #4 would require decreases in fossil generation. EPRI’s
“fossil leverage factor” highlights that the algebra of the compliance equation results in a
multiplier effect. Essentially, they were able to identify that 1 MWhr shortfall of nonemitting
resources must be made up by more than a factor of 2 MWhrs of fossil generation in 20 states
across the United States. For example in West Virginia, for every 1 MWhr of Building Block #3
or #4 that is not achieved, 5.17 MWhrs of fossil generation must compensate for the lack of the
zero emission delivery in 2030. Building Blocks #3 or #4 unrealistic assumption will force
additional decreases in fossil generation, raising serious concerns regarding reliability, planning
and compliance. This leverage factor is significantly increased if nonemitting resources are not
met by the 2020 interim goal. Again, taking West Virginia as an example, for every 1 MWhr of
Building Block #3 or #4 that is not achieved, 23.35 MWhrs of fossil generation must compensate
for the lack of the zero emission delivery in 2020. This “fossil leverage factor” underscores the
reason why the interim goals must be set aside.

EPA also contends that some stakeholders’ believe that the state goals fail to reflect the full
potential, under the BSER, for incremental RE and EE to replace fossil steam generation. By
adding incremental RE and EE generation, this action actually avoids emissions and does not
decrease emissions. Therefore, to subtract equivalent fossil generation from the BSER would be
erroneous.

Maximum State Flexibility
CAA section 111(d) require states, not EPA, to set standards of performance for sources. Not
only do states have the authority to set the standards, but they also have the authority to
determine how the sources within each state will meet those standards. Therefore, states should

25

 have ultimate flexibility in building their state programs and determining what activities can be
included for compliance. These activities include, but are not limited to, the following areas:
•

A state should have authorization to deem that a current state program qualifies as BSER
for that state

•

The states should have sole discretion on how to reach the 2030 compliance target
including timing, glide path, interim targets, etc…, without any requirements from EPA
during that interim period

•

Enforcement of a state plan should be the sole responsibility of the state

•

States should have the discretion to include new Section 111(b) affected NGCC units in
their compliance plans

•

States should have the ability to control what activities count towards compliance and
when these activities can count towards compliance including, but is not limited to, plant
retirements, treatment of nuclear and hydro, energy efficiency measures, renewable
energy, and new 111(b) NGCC units

•

States should have the ability to count any emission reductions that occur after the
baseline year including retirement of fossil-fired generation

•

States should have sole authority over whether or not the state uses a mass-based or ratebased approach towards compliance and how that translation is calculated

•

States should have definitive oversight and control over any multi-state approach or plan

The CAA gives states authority over implementation under section 111(d). This rule should not
impede on state’s power to carry out that authority.

Clean Air Act Authority
EPA’s authority under section 111(d) is limited to issuing “emission guidelines” addressing
factors relevant to the states’ implementation of the BSER that has been adequately
demonstrated to reduce CO2 emissions at a source (inside the fence).
Section 111(d) specifically directs EPA to establish a procedure for states to submit plans
establishing performance standards for existing sources. States possess considerable discretion
and flexibility under the Act in developing standards of performance based on EPA’s emission
guidelines. To the extent EPA’s guidelines are based on replacing equipment to improve the

26

 efficiency of the generating unit, EPA should clearly exempt such activities from being
considered a “modification” for purposes of NSR permitting, particularly in light of EPA’s
Office of Enforcement and Compliance Assurance’s previous focus on these type of projects in
their enforcement cases. These projects have also been the target of third party citizen suits
contending they represent violations of the NSR rules.

Duplicative regulation under Sections 111(b) and 111(d) is not permitted by the CAA. EPA
cannot regulate the same source under both CAA Section 111(b) and CAA Section 111(d). An
EGU regulated under CAA Section 111(b) because it is a “new source”—which modified and
reconstructed sources are defined to be— cannot simultaneously be subject to regulation under
CAA Section 111(d) as an existing source and vice versa. Modified or reconstructed sources that
were previously subject to a state plan under CAA Section 111(d) cannot be required to continue
to be covered by CAA Section 111(d), although states do have discretion to keep those sources
in their CAA Section 111(d) plans if the states choose to do so.

Similarly, EPA is prohibited from regulating pollutants under Section 111(d) from a source
category already regulated under Section 112 of the CAA.

Other
EPA has yet to make available all the documentation the public needs to assess whether the
proposed rule is reasonable, including 21 of the 25 IPM modeling runs EPA relies on to argue
that the Building Blocks are achievable.

Review of the limited modeling results leaves many questions, for example, regarding the
appropriateness of how energy efficiency is represented in the model (both characteristics and
cost), the appropriateness of modeled transmission investment, and the treatment of any
“remaining plant balance” associated with modeled retiring plants. Further, it is not clear from
the limited modeling results presented, how the Building Blocks are integrated into the cases.

IPM’s assumption that a 100% load factor (full reductions in all hours of each year) for EE
resources is unreasonable. For example, an energy efficiency program involving residential light
bulbs only generated reductions when the lights are normally on, likely something far less than

27

 100% of the time. Or, consider a residential refrigerator EE program that results in savings only
when the old refrigerator was normally running (i.e. more efficient operation, fewer hours of
each day) again, far less than 100% of the time. Examples similarly continue in the industrial
sector with HVAC programs with less than 100% load factor and even industrial lighting and/or
motors which would only approach 100% load factor results in the most efficient 24/7
manufacturing facilities.

It is not clear from the limited results and documentation provided by EPA whether the assumed
$44B cost for EE is consistent with “one for one” programs with less than a 100% load factor (in
which case the modeling assumptions are not appropriate) or alternately, the $44B cost
assumption may be understating the spend necessary to accomplish this magnitude of energy
reductions.

Rural Cooperatives and Municipal facilities should be subject to the same requirements as
electric generating units under this rule. Especially in deregulated markets where they would
operation at a competitive advantage, as compared to units subject to the proposed rule. Any
longer-term capacity planning strategies advantage could be misconstrued as market distortion.

Conclusion
Having an electric system that is reliable and affordable is paramount to individual families,
manufacturing, the service industries, economic security and prosperity, and the overall wellbeing of this nation. The current emissions trajectory of the utility industry would suggest that
reliability and affordability can come with significant CO2 emission reductions from the existing
fleet, similar to what EPA’s projects will result from this rule, without the negative consequences
of this proposal. As such we question the need for such a radical retransformation of the electric
system based on building blocks that do not place the utmost value on reliability or affordability.
We recognize the difficulty of EPA’s task of using the Clean Air Act to effectively and
economically regulate GHGs. We agree with statements from many political leaders and leaders
of EPA that the Clean Air Act is not adequately designed to effectively regulate GHGs, so we
caution EPA to be careful in using an inappropriate tool where the results of doing so often come
with negative consequences. Even so, we submit these comments as a means of providing

28

 additional technical support and identifying areas where EPA’s calculations can be made more
accurate. We thank you for the opportunity to comment and engage in this process.

FE is an active member of the Utility Air Regulatory Group (UARG), Edison Electric Institute
(EEI), Utility Solid Waste Activities Group (USWAG), Midwest Ozone Group (MOG), and the
Electric Power Research Institute (EPRI) and incorporates their comments herein by reference.

29