20211013Initial Comments.pdf

Benjamin J. Otto (ISB No. 8292)
710 N 6m Street
Boise,ID 83701
Ph: [phone]xr12
botto@idahoconservation. org
Attorney for the Idaho Conservation League
BEFORE THE IDAHO PUBLIC UTILITIES COMMISSION
j;t,ir* il l\/t+-tt-,n-i.i1*l Y LU
Iil;t fiilT t3 PH 12: 5 i,
1,, l i 5:,.'-:, .'.;:..,-i.,i.,,'i i: :, .l*i,"i;.'i.3SiOli
IN THE MATTER OF IDAHO
POWER COMPAI\Y'S
APPLICATION TO INITIATE A
MULTI.PHASE COLLABORATIYE
PROCESS FOR THE STUDY OF
COSTS, BENEFITS, AI\D
COMPENSATION OF NET EXCESS
ENERGY ASSOCIATED WITH
CUSTOMER ON.SITE
GENERATION
LPC-E-2I-2t
ICL TNITIAL COMMENT
CASE NO. IPC-E-2I.2I
IDAHO CONSERVATION LEAGUE
INITIAL COMMENTS ON STUDY
TRAMEWORK
)
)
)
)
)
)
)
)
)
)
The Idaho Conservation League (ICL) submits the following Initial Comments on the
proposed customer on-site generation study framework, hereinafter the Value of Solar or VOS
study. ICL's comments begin with trvo overarching recommendations that will create a credible
and fair study - use avoided cost principles and retain a neutral third party to conduct the study.
We then provide recommendations to reorganize and clariff the contents of the study in order to
create a more useful basis to consider future customer-owned generation program options.
I. Scope and Objectives
The "Primary Objective" in the proposed framework is a good start towards determining
the appropriate method of valuing these exports is the appropriate scope for this solar study and a
critical step for developing a credible and fair study. The scope of the study should focus on the
exports from on-site generation onto the electric system because this is the key feature that
distinguishes customer-owned generation from how other customers use the electric system. The
Commission recognized this customer distinction when it created the new customer-generator
L
October 13,2021
classes of Schedules 6 & 8l and when it approved Idaho Power's request to allow non-exporting
customer generators to remain on the same rate schedules for consuming utility service as other
customers of the same class.2
Building upon this distinction recognized by the Commission, the scope of this solar
study should acknowledge and respect the unique relationship between the utility and its
generator customers. Customers, including [CL, invest in self-generation systems primarily to
control their own energy bills, as the Commission heard in the l3 hours of public testimony
regarding the prior solar study settlement.3 Customer-owned generation is unique in that some
portion of the generation serves the customer's own needs and some portion flows beyond the
customer's meter onto the grid. At other times, customers consume utility-provided energy. This
two-way relationship between the customer and utility implicates distinct areas of the
Commission's jurisdiction and regulatory practices.
The scope and sub-objectives for the study, proposed by Idaho Power, confusingly mixes
the distinct concepts of consumption and generation and their relevant regulatory tools. The
Company's Application asserts that "recommendations to modiff the existing offering should
focus on cost-of-service principles, while identifuing the value of excess net energy to ensure
equitable compensation for on-site generation."4 The sub-objectives in the proposed framework
continue this theme by focusing on "the Company's ability to recover costs from self-generating
customers."s This proposal applies regulatory concepts applicable to consumption - cost of
service and revenue requirement - to the question of valuing generation. The more appropriate
1 OrderNo. 34046, Case No. IPC-E-17-13, pp. l5 -19 (May 9, 2018).
2 Order No. 34955, Case No. IPC-E-20-30 (Mar. 9, 2021).
3 OrderNo. 345}9,CaseNo.IPC-E-18-15, pp.3 - 4 (Dec.20,2019).
a Application, Case No. IPC-E-21-21, p. 6 (June 25,2021).
s Study Framework for Party Comments, CaseNo. IPC-E-21-21, p. 5 (Sep. 30,2O2l).
2
rPC-E-21-21
TCL INITIAL COMMENT
October 13,2021
concept for valuing the service that customer-generators provide to the system is Idaho Power's
avoided costs that result from the locally produced energy that flows directly to neighboring
properties.
Because there are unique characteristics of customer-generators that distinguish them
from all other utility customers, the scope of this solar study should not include either the
portion of customer generation that serves the customer's own needs or the customer-generator's
consumption of utility service.6 Consumption of the customer's own solar generation is not a
public use and therefore not within the Commission's jurisdiction. Therefore, how on-site
customer-generation impacts consumption of utility service is outside the scope of this study.
In addition, although the solar customer's consumption of utility-provided energy is
within the Commission's jurisdiction to establish 'Just, reasonable, or sufficient rates"T that issue
is appropriately addressed in a general rate case where the Commission holistically considers
how all customers consume utility service. Because the customer-generator's consumption of
utility supplied service is not the key distinguishing feature of this class of customers, the issues
of cost of service and rate designs for consumption are outside the scope of this study.
The unique issues related to customer generation are only relevant when an extra bit of
energy flows across the meter onto the electric system. At this point, the Commission's
jurisdiction takes on a new flavor because the customer-generator is providing a product for
public use. Like other generation resources, the value of the service customer-generators provide
to the system is appropriately captured by the concept of avoided costs.
t I.c. $ 6l-5ol
'I.c. $ 6r-502
rPC-E-zt-21
ICL INITIAL COMMENT
3
October 13,2021
Accordingly, ICL's comments on the solar study framework will focus on the appropriate
methods to value generation using avoided cost principles and we will leave issues of customer
consumption, measured by cost of service and revenue requirement, to the next general rate case.
By clearly distinguishing between consumption and generation, and applying the correct tool to
each, the study is more likely to be "understandable to an average customer, but its analysis must
be able to withstand expert scrutiny."8
II. Use a Neutral Third-Party to Conduct the Study
ICL recommends that the Commission direct Idaho Power to use a neutral third-party to
conduct the study. In our experience of several years of net-metering dockets, we have
consistently heard from the public that it does not trust Idaho Power to conduct a fair study for
two primary reasons. First, the nature of the rate of return on invested capital regulatory model
creates a strong incentive for utilities to discourage any non-utility generation. To be clear, ICL
is not imputing any nefarious actions onto Idaho Power, rather we are just observing the obvious
incentive structure. Using a neutral third-party to conduct the study would show the public and
stakeholders that Idaho Power also understands this inherent incentive structure and is taking
credible steps to address this issue.
Second, Idaho Power's Application continues the long trend of making unsubstantiated
assertions about subsidization, cost shifts, and customer-generators not paying their fair share
and ICL worries that a solar study conducted by Idaho Power will be unfairly biased against
customer-generators. Idaho Power has yet to provide a comprehensive cost of service study that
documents how customer-generators' consumption of utility power results in cost shifts that are
any more impactful than the normal variation of consumption among members of a rate class.
8 Order 34509, supra at 9
LPC-E-ZI-21
ICL INITIAL COMMENT
4
October 13,2021
The Company's attempt to address potential cost shifts through a fixed cost study in IPC-E-18-
l6 did not result in any greater understanding among customers, the public, or the Commission.e
Further, the Commission has consistently recognized that the "benefits that on-site generation
provide to the Company's infrastructure and resource allocation, once quantified, may well prove
to oupace any alleged costs, increases in fixed-cost responsibility or decreases in net excess
energy compensation credit".l0 In 2019, the Commission also clearly stated that the prior solar
sefflement that attempted to value customer-generator exports was not based on substantial and
competent evidence.ll Nevertheless, Idaho Power continues to assert the need to "limit subsidies
by implementing a more equitable pricing and compensation structure."l2 This statement
evidences a clear bias against customer-generators and ignores the Commission's prior orders to
study this issue in a credible and fair manner so as to build a substantial and competent record
upon which to assess whether any subsidy exists. As the Commission heard in the public
testimony in IPC-E-I8-15, Idaho Power's repeated unfounded assertions about customer-
generators shirking their responsibilities undercuts the Company's credibility to produce a fair
study. Using a neutral third party would break this cycle and build a higher level of credibility
and fairness into the process.
Utilities and public utility commissions around the country have opted to commission
third-party studies of the value of solar rather than conduct the study in-house. Two utilities,
Austin Energy and Arizona Public Service Company, have solicited multiple VOS studies from
Clean Power Research and SAIC, respectively, to determine the value of solar for utility
s Order 34608, Case No. IPC-E-I8-16, pp. 6 - 7 (Mar. 31, 2020).
10 Order 34046, supra at 19.
" Order 34546, Case No. IPC-E-I8-15, p. 4 (Feb. 6,2020).
12 Application, supra at 6; Study Framework supra at 6.
5
[PC-E-ZI-21
ICL INITIAL COMMENT
October 13,2021
customers.l3 In Minnesota, the Minnesota Public Utilities Commission relied on a third-party
study completed by Clean Power Research when it set the value of solar for Minnesota utilities.la
Finally, in its most recent proceeding to update the net metering tariff in California, the
California PUC not only commissioned a third party to conduct a study and issue a tariff
proposal, but also asked intervenors in the case to comment on the third-party proposal and
provide their own proposals.l5 Idaho Power should follow this precedent by soliciting a neutral
third party to conduct its VOS study.
III. Study components
After describing the goals and objectives, the proposed framework turns to the actual
contents of the solar study. This section uses terms that are unnecessarily confusing and proposes
a sequence of issues that places excessive weight on Idaho Power's interests, such as revenue
requirement, rather than the broader public interest, such as the credible and fair calculation of
the value of customer-generator exports. ICL's comments below recommend reorganizing and
clariSing the study framework so it is more understandable to the public and more likely to
result in a useful basis to consider future changes to the customer-generator program.
The framework should begin with defining the range of measurement intervals applied to
each category in the VOS as well as the criteria to assess the accuracy of, and ability to
implement, each measurement interval option. The second step is to identifu all relevant
13 Clean Power Research,2014 Value of Solar Executive Summary, p. 3 (Dec. 12,2Ol3),
http://www.austintexas.gov/edims/document.cfin?id:202758 (last accessed Oct. I1,2021); SAIC,20l3 Updated
Solar PV Value Report, p. 7 (May 10, 2013), https://www.azsolarcenter.org/images/docs/reports/SolarValueStudy-
SAIC-20 I 3-05.pdf (last accessed Oct. I l, 2021).
1a Minn. Pub. Utilities Comm'n, Order Approving Distributed Solar Value Methodology, Docket No. E-999/\,I-14-
65, p.9 (Apr. 1,2014).
1s Cal. Pub. Utilities Comm'n, Order Instituting Rulemaking to Revisit Net Energy Metering Tariffs Pursuant to
Decision l6-01-044, and to Address Other Issues Related to Net Metering, Rulemaking No. 20-08-020, p. 4 (Nov.
t9,2020).
5
IPC-E-21-21 October 13,2021
ICL INITIAL COMMENT
categories for the value of solar and assess them in regards to each measurement interval and at
various levels of cumulative customer-generation on the system. The third step of the framework
should address program implementation issues including the method to return value to customers
and options for future updates to the customer-generator program. The fourth step should
consider program participation issues including the size of eligible systems and how to enable
multi-family housing residents to access the on-site generation program.
l. Meosurement Interval
The concept of net exports, and respect for the appropriate scope of the study, requires
establishing a time period to compare a customer's consumption with any generation that crosses
onto Idaho Power's system. The length of the measurement interval is likely the most important
concept to establishing a credible and fair study design that is underskndable to the public
because it influences both the value of the exports as well as the ability for a customer to
understand their investment decision. We support the proposal to study a range of measurement
intervals in order to build a complete record and foster understanding.
The study framework, however, identifies inapplicable criteria to assess the appropriate
time interval - the "class revenue requirement", the "export credit payments", and the "bill
impacts."l6 These criteria have little meaning to the public and do not assist in understanding
how the value of the components of the value of solar change in regards to the measurement
interval. For example, revenue requirement commonly refers to the revenue Idaho Power should
collect to compensate for providing service to customers. The concept of revenue requirement is
not relevant until the study determines the amount of exports, when they occur, the value they
provide, and the method customers will be credited for providing this value.
16 study Framework, supra at 7-8
rPC-E-2t-21
ICL INITIAL COMMENT
7
October 13,2021
The proposal includes the confusing term "credit payments" which mixes two distinct
methods of compensation - a credit, as is the current method, and a payment, which is an old
method Idaho Power specifically asked to end in2013.t7 Considering the bill impacts to
customer-generators is important, but is appropriately done after assessing the value of exports
and various program implementation issues. Finally, the focus on "existing customers" ignores
the fact that existing customer generators are protected from future program changes due the
Commission's determination that they can remain in the legacy net metering program through
2045.
To produce a more credible and fair solar study that properly focuses on exports, is
understandable to the public, and can withstand expert scrutiny, we recommend the solar study
apply the following criteria to assess which measurement interval is most appropriate to inform
the future design of the customer-generation programs:
l. ability to accurately match generation exports with the value to the system
2. ability for customers to understand how the program rules and credits influence
their investment decision
3. ability of system-providers to give clear and reliable forecast of performance to
potential system buyers
4. ability of each measurement intervalto support the implementation of a range of
crediting options
The solar study should apply these criteria to examine the impact on export value in
different time intervals - monthly, hourly, as well as a sub-hourly option - as discussed in more
detail below.
17 Order 32846, Case No. IPC-E-12-17, p. 15 (July 3,2013).
8
LPC-E-ZI-21
ICL INITIAL COMMENT
October 13,2021
Monthly: This is the current program design and can serve as a benchmark against which
a different interval is assessed according to the criteria above.
Hourly: This is an appropriate time measurement as utility system costs vary by the hour
as does the output of the customer-generation system. Based on our conversations with
customers and system-providers, the study should address methods and tools available to
implement this interval in a manner that allows for customer understanding of the long-term
performance of their system and impact on their energy bills. As the Commission recognized in
IPC-E-18-15, moving to hourly netting is a significant change to the economic value of
customer-owned generation and something that must be supported by a robust record.ls Studying
the change to hourly neffing in relation to the criteria above is a key step in developing this
record.
"Separate Chqnnet': This term is a measurement method, not an interval. The study
framework filed by Staff does not accurately represent ICL's position. ICL did not suggest
"lnstantaneous Net Energy Measurement" because that collection of words mixes a measurement
interval with a measurement method. We note Idaho Power's Application appears to really
suggest they meant an instantaneous time interval: "separate channel, which is sometimes
referred to as instantaneous. "le
Referring to Idaho Power's wording, we recommended in our September 22 informal
comments to use the term "instantaneous," but noted this concept could undercut a fundamental
tenet of customer-generation, the ability to self-supply energy. In that meeting we were relieved
to learn Idaho Power is not proposing to study a program design whereby the Company would
18 Order 34546, supra at 6 - 7
1e Application, supra at 7.
LPC-E-2I-21
ICL INITIAL COMMENT
9
October 13,2021
separately measure all customer-owned generation as an export distinct from customer
consumption. That possible program design would frustrate a fundamental concept of customer-
owned generation, the ability to meet your own energy needs through independent investments.
To better clarifu our recommendation: The study should examine a measurement interval
that is shorter than hourly, but must maintain the fundamental concept that customer-owned
generation first serves the customers' energy needs. Because Idaho Power knows the technical
abilities of its metering system, we suggest the Commission direct Idaho Power to document the
appropriate time-interval that allows customers to plainly see what portion of their system serves
their needs and what portion is exported to the grid. Then the study can assess this sub-hourly
time interval in regards to the three criteria we suggest above.
2. Cotegoriesfor the Value of Solar
To fully capture all of the benefits that customer-owned generation provides to the utility
and its customers, the study design should include as many categories as possible within the
value of solar (VOS) analysis. For each category, the study should address monthly, hourly, and
sub-hourly intervals as described above. Also for each category and each time interval, the study
should assess the value at various levels of growth in the customer-generation program ranging
from current levels to l0 times current and 25 times current.
Researchers Koami Soulemane Hayibo and Joshua M. Pearce at Michigan Technological
University recently compiled a review of several VOS studies across the country to identiff a
generalized model for calculating the true value of distributed solar generation. As indicated in
their 2021 paper, the best practice to evaluate the dollar value of solar per kilowatt-hour of
electricity produced is to sum up the calculated avoided costs, or solar benefits, and then divide
LPC-E-ZI-2r
ICL TNITIAL COMMENT
10
October 13,2021
that sum by the total amount of energy produced during the analysis period discounted by the
associated discount factor. 20
The researchers found that the most common components included in a value of solar
study as avoided costs are: energy production costs (operation and maintenance), electricity
generation capacity costs, transmission capacity costs, distribution capacity costs, fuel costs,
environmental costs, ancillary costs including voltage control benefits, solar integration costs,
market price reduction benefits, economic development value or job creation, health liability
costs, and value of increased security.2l Out of these component categories, Hayibo and Pearce
argued that avoided O&M fixed and variable costs, avoided fuel cost, avoided generation
capacity cost, avoided reserve capacity cost, avoided distribution cost, avoided environmental
cost, and avoided health liability cost are the minimum categories necessary to calculate the true
value of solar. Please refer to ICL Exhibit 501 , the Hayibo and Pearce study, for further details
on the methods to calculate each component of the VOS.
ICL also recommends that, in order to represent fully the entire value of distributed solar
generation, the solar study must include values associated with the resilience and reliability of
the grid, health benefits including those from reduced carbon dioxide emissions and air pollution,
and local economic benefits such as job creation from thriving solar markets.
In addition, ICL makes the following recommendations: In each category in the VOS, the
study should include a time horizon for the avoided costs. Solar systems are the vast majority of
customer-generation resources and typically last for 20 years or more. Regarding the vague
concept of "firmness" and the system level concept of integration, the study should assess
20 Koami S. Hayibo & Joshua M. Pearce, A Review of the Yalue of Solar Methodologtwith a Case Study of the I).5.
VOS, 137 Renewable and Sustainable Energy Reviews (2021).
2r ld,. at2.
LI
IPC-E-21-21 October 13,2021
ICL INITIAL COMMENT
whether distribution circuits with some meaningful amount of customer-generator exports show
an overall variability that adds to or avoids costs for Idaho Power. Because customer-generation
systems sit on the distribution circuits and are unlikely to reach the bulk transmission system, we
recommend the study focus on avoiding distribution level costs and line losses distinct from
transmission costs and line losses. This necessarily requires focusing on the specific locations on
the distribution system the exports occur and recognizing this location value in any export credit
rate. Lastly, the category of environmental benefits that arise from customer-generator exports
and accrue to the utility system should include the utility's reduced costs associated with
compliance with Clean Air Act requirements at its fossil fuel generation facilities as well as the
utility's reduced costs associated with future greenhouse gas regulation. Because these
environmental benefits impact rates for all customers, the environmental and health benefits
category should be moved in the framework from a side issue to a core component of the VOS.
3. Program Implementation Issues
Implementing the customer-owned generation program will affect Idaho Power
differently than the progtam participants. The Company's proposed framework already covers
issues like collecting revenue associated with customer credits and accounting for these credits in
the billing system. The public interest primarily focuses on basic rules governing how customers
can participate in the customer-generation program which is not yet well represented in the
framework. Therefore, ICL's comments in this section are structured as common questions we
have consistently heard that the study should answer.
q. How will customer-generotors be fairly compensatedfor their exports?
The proposed framework includes a section "Billing Structure" and proposes to "explain
how potential customer-generators and on-site system installers will have accurate and adequate
data and information". Explaining an outcome is not the same as studying options. ICL
L2
[PC-E-ZI-21
ICL TNITIAL COMMENT
October 13,2021
recommends the solar study examine the options for providing credits for exports to customers.
Those options should address issues such as: which bill components the credits can offset,
whether credits can be used to offset other accounts held by the same customer, and the ability of
customers to donate credits to other customers.
b. Do the credits expire?
The framework includes a section covering the potential for customer credits to expire,
which is a distinct change in policy from the current program. ICL believes that credits are the
property of the customer and not Idaho Power. Therefore, any regulatory mechanism that would
take away the value of a customer's credits could raise thorny legal questions. The study should
assess options that preserve the customer's value in any accumulated credits including the ability
to transfer them to other accounts, whether held by the customer or not.
c. How will the program changefrom the current net metering program to o successor?
ICL appreciates the careful attention the Commission gave this issue when adopting the
criteria for existing customers to remain in the legacy net metering program. Customers who
invest in generation systems make a significant financial investment to meet their own energy
needs. And due to the exclusive service territory allocated to Idaho Power by state law,
customers are limited to only one program that could enable them to enjoy the benefits of their
personal investment. The study should examine options to allow for a predictable and fair
transition between the current program and any successor program. The Utah Public Service
Commission recognized this dynamic and created three categories of customer-generators22:
"Net-Metered Customers" who remain in the legacy net metering program over the long-term,
"Transition Customers" who interconnected during the years-long process to adopt a successor to
22 For a simple description of the Utah transition structure please see this website: https://rooftopsolar.utah.gov/
13
IPC-E-21-21 October 13,2021
ICL INITIAL COMMENT
net metering and are subject to transitional rates and program rules, and "Post-Transition
Customers" who interconnect after the Commission adopted a successor program and are
immediately subject to the new rates and rules. While Idaho will need different rates and rules
than those adopted in Utah, ICL recommends the solar study include options to address
transitional and post-transition customer-generators.
d. How will the successor progrom change in the future?
The proposed framework presupposes a biennial or annual update to the export credit
rate.23lCL submits that this is an overly restrictive framing of this critical issue. Instead of
assuming a one- or two-year update period, the study should first assess whether establishing a
timeline for in this one area is fair, just, and reasonable when there is no similar timeline for
updating the Company's overall electric rates. The study should consider how the timing of
updates impacts the ability of customers to make informed decisions about a product that lasts
for 20 years or more. The study should consider how the timing of updates impacts the providers
of customer-owned system ability to give rigorous information and informed forecasts to their
potential customers. Once the study considers the propriety of frequent updates unique to this
one customer-facing program, the study should examine the variety of processes available to
make these updates. For example, it should examine whether the updates should align with the
resource planning cycle even though that process is merely an acknowledgment of a process and
not a decision on any of the contents. Another option is to connect any updates to objective
criteria such as changes to the costs Idaho Power avoids by receiving exports from customer-
generators. Because any changes to the customer-generation program have historically caused
23 Study Framework, supra at 20.
[PC-E-ZI-21
ICL INITIAL COMMENT
L4
October 13,2021
enorrnous public interest, ICL recommends the Commission pay close attention to these issues of
fundamental program credibi I ity and fairness.
4. Participotion Rules
o. System Cap
The eligibility cap for participants in the customer-generator program is an important
issue, but not one that needs to be studied as part of the value of solar study. Rather, this issue
can and should be addressed immediately in a separate docket. The question is straightforward:
should the program limit customers to a somewhat arbitrary system size, or should customers be
able to invest in systems that match their own energy needs? To help reduce the size and
complexity of the solar study, ICL encourages the Commission to direct Idaho Power to initiate a
separate docket to consider this issue. We note that the Commission's previous decision to
separate out other system eligibility issues from the previous value of solar proceeding, like
interconnection standards in IPC-E-20-26 and smart inverter settings in IPC-E-20-30, enabled a
quick and non-controversial resolution of those issues that did not directly impact the value of
exports from customer-generation.
b. Multifamily Housing
The solar study should examine potential program designs that will facilitate multifamily
solar installation. Residents of multifamily buildings face unique participation challenges for
customer generation. Residents who own their units may not own the rooftop space and thus
must rely on the building owner or similar entities to manage the solar installation and distribute
the value produced by the on-site generation system. Residents who rent their units have little to
no control over landlord decisions to install solar and landlords have little incentive to pay for
solar installations because renters generally pay their own utility costs. Even if landlords choose
to install solar, renters are unlikely to see any financial benefits unless there exists an avenue for
15
rPC-E-21-21
ICL INITIAL COMMENT
October 13,2021
renters to subscribe to the solar production. The solar study should explore options to address
this market failure so that residents of multi-family housing have an equal opportunity to
participate in a customer-generation program,
One program structure that will facilitate multifamily customer-generation for both
owner- and renter-occupied buildings is a subscribership program where residents can purchase a
subscription to a particular number of rooftop solar panels.2a A subscribership program will
allow building developers/owners to recoup the installation costs of the panels while still
permitting residents to benefit financially from the solar production. A subscribership program
for renters will also benefit low-income customers who are more likely to live in rented units and
are less able to afford individual home solar installations. The study should examine program
designs that ensure that multifamily solar participants receive a value for exports that is
equivalent to the value received by individual system owners.
IV. Other issues
The proposed study framework includes several other issues raised by parties. ICL takes
no position at this point on those issues and perspectives. Rather, we look forward to reviewing
parties' formal initial comments before responding.
Similarly, the Commission made it crystal clear in Orders 34509 and 34546 that
stakeholders must listen to and incorporate the public input and perspectives in this process. ICL
looks forward to learning from the public about what issues are important to them. In particular,
we look forward to learning if this proposed framework is understandable to the generalpublic.
2a The Virginia State Corporation Commission recently adopted a similar program where
multifamily residents can subscribe to shared solar. ,See Commonwealth of Virginia, ex rel. State
Corporation Commission, Ex Parte: In the matter of establishing regulations for a multi-family
shared solar program pursuont to $ 56-585.1: 12 of the Code of Virginia, Case No. PUR-2020-
00124, Order Adopting Rules, p. 12 (Dec. 23, 2020).
15IPC-E-ZI-ZI October 13,2021
ICL INITIAL COMMENT
It could be the case that the parties will need to reassess the entire approach to this issue and
develop a framework that is less reliant on technical jargon and complex regulatory concepts. At
the end of the day, the framework could be reduced to answering some simple questions:
l. How will potential solar owners know the program is fair and predictable?
2. What are the rules potential solar owners need to follow to participate in the
program?
3. How can program participants continue to engage in any future updates to the
program?
4. How does the customer-generation program align with and support ldaho Power's
commitment to 100% clean energy?
V. Conclusion
ICL appreciates the Commission's phased approach to addressing this program that
enables customers to make personal investments to meet their own energy needs. We look
forward to working collaboratively with all stakeholders to develop a credible and fair solar
study.
Respectfully submitted this l3th day of October,202l.
/s/ Beniamin J Otto
Idaho Conservation League
With technical assistance from
Dainee Gibson-Webb, Conservation Analyst
Emma Sperry, Climate Fellow
lPc-E-zt-21
ICL INTTIAL COMMENT
L7
October 13,2021
Idaho Conservation League
Exhibit 501
Koami S. Hayibo & Joshua M. Pearce, A Review of the Value of Solar Methodologtwith a Case StuSt of the U.S.
VOS,137 Renewable and Sustainable Energy Reviews (2021).
rPC-E-21-21
ICL INITIAL COMMENT
18
October 13,2021
Renewable and Sustainable Energy Reviews 137 (2021) 110599
Contmts lists available at ScimceDirect
ETSEVIER
Renewable and Sustainable Energy Reviews
journal homepage: http://www.elsevier.com/locate/rser
ll
i
Ja
iollmrt Iffi!A review of the value of solar methodology with a case study of the U.
S. VOS
Koami Soulemane Hayibo', Joshua M. Pearce"'b'''.
5 Dryawt ol Eleca|ical & co,npta Engin4riJ,g" Mitldgort T1/'hrlologirol Univarity, Ha!$tot\ ML, UsA
b oquutt ol uatoiols Sciance & r]l{d,tgiig, Mbhig@rtTffjhnologicol Univdlry, Hot@\ Mt, UsA
c Phorwotaies @itN@oavlaeahg, Schod of HrrticolEttgirrrrirg, Aolr Uniyan$, Fttrlotd
ARTICLE INFO ABSTRACT
Distributed generadon with solar photovoltaic (PV) technology is economically competitive ifnet metered in the
U.s. Yet ther€ is evidence that net metering is misrepresenting the true value of distributed solar generation so
that the value of solar (VOS) is becomlng the preferred method for evaluating economicr of grid-tied Pv. vOS
calculations are challenging and there is widespread disagreemert ln the literatue on the methods and data
needed To overcome these limitatioru, this study reviews past vos studies to develop a gmeralized model that
considers rralistic future avoided costs and liabilides. The approach used here is bottom-up modeling where the
final VOS for a utility system is cdculated The avolded costs considered arer plant O&M 6xed and variable; fuel;
generation capacity, rcserve capacity, transrnisslon capacity, disuibution capacity, and environmental and hedth
liability, The vOS repr€s€ns the sum of these avoided costs. Each zub.component of the VOS has a sensitivity
analysis run on the core variables and these sensitivities are applied for the total VOS. The resuls show that grid-
tied utility customers arc being grossly under<ompensated in most of the U.S. as the value of solar eclipses the
net metering rate as well as two-tieted rates. It can be concluded that substantial future work is needed for
regulatory reform to ensure that Srid-dd solar PV ownets are not unjustly subsidizing U,S. electric utilities.
Kcrwrlla
Udltty policy
Photovoltsic
Dtstrlbuted tseration
Value of solar
Net metering
Economics
l. Introducdon
Solar photovoltaic (PV) technologies have had a rapid industrial
leaming curve [l-4], which has resulted in continuous cost reducdons
and improved economics [5,6]. This constant cost reduction pressure
has resulted in a spot price of polysilicon Chinese-manufactured PV
modules of only US$0.18,/W as of April 2020 171. Ttrere art several
technical improvements, which are both already available and slated to
drive the costs further down such as black silicon [8 10]. The Intema-
tional Renewable Energy Agency (IRENA) can thus confidently predict
that PV prices will fall by another 60% in the next decade [11]. How-
ever, even at current prices, any scale ofPV provides a levelized cost of
electricity (LCOE) t12l lower than the net metered cost of grid elec-
tricity t13l and this will only improve with stora e costs declining
[14-18]. Specifically, Pv already provides a lower Ievelized cost of
electricity 112,79,201 than coal-fired electricity 113,21,22), In addition,
PV technology can be inherently distributed (e.8. each electricity con-
sumer produces some or all of their electricity on site thus becoming
'prosumers'). Distributed generation with PV has several technical
advantages, including improved reliability, reduced transmission losses
a23,241, enhanced volage profile, reduced transmission and distribu-
tion losses [25], transmission and distribution infrastructures defer-
ment, and enhanced power quality 1261. As PV prices decline, prices of
conventional fossil fuel-based dectricity production are increasing due
to agrng infrastructure 127-291, increased regulations (in some juris-
dictions) [30-33], fossil fuel scarcity [34 36], and pollution costs
137411, Thus, PV represents a threat to convendonal udllty buslness
models [42] and there is evidence that some utilides are manipulating
rates to discourage distributed generation with solar [43], while others
are embracing it such as Austin Texas or the state of Minnesota [44].
Rates structures vary widely throughout the U.S [45-48]. and there has
been significant effort to determine the actual value of solar (VOS)
electricity.
This shift towards VOS is fueled by cridcisms of its predecessor [49],
net metering, that ls misrepresendng the true value of distributed solar
generation t50-52]. VOS is more representative of the electricity cost
because under a Value ofSolar Tariff (VOST) scheme, the utility pur-
chases part of, or the whole net solar photovolaic electricity generation
from its customers, therefore dissociating the VOST from the electricity
* Correspondlng author. Department of Materids Science & Engineering, Michigan Technological University, Houghton, MI, USA.
E-moall ailihtsses: khayibo@mnr,edu (KS. Hayibo), pearce@mru,edtt (J.M. Pearce).
hrl.ps:/ /doi.org/10. I 0l 6/j.rser.2020. I 10599
Received 26 April 2020; Received in revised form 4 September 2020; Accepted 22 November 202O
Available online 4 December 2020
f364-O321,/O 2020 Elsevier Ltd. All rights reserved.
Nomcnclature:Iep
Ip
K
M
n
o
PLl
PLlO
a
s
S6
Us
Up
Up
U7
Uy
vos
vx
B
Cp
C6
C11
Cpv
C7
D
DB
Dg
Dpv
E
F
FB
h
H6
H67
Hn
Hp
Hs
i
Ig
Ip
Burner tip fuel price t$/MMStul
Distribution capacity MWI
Utility generation capacity [p,u.]
Hedth cost of natural tas I$AWhl
PV capadty for year'n' [k!V]
Transmission capacity [p.u.]
Utility Discount rate
Environmenal discount rate
Heat rate degradation rate
Degradadon rate of PV
Environmental cost t$,/MMBtuI
Utility discount factor
Environmenal discount factor
Number of hours in the analysis period
Heat rate of combined cycle tas turbine [BtuAWh]
Heat rate of peaker combustion turbine [BtuAWh]
Heat rate for year n tBtuAWhI
Heat rate of the plant tBhlAwhl
Solar heat rate [BtuAWh]
Number of yea$ in analysis period
Installation cost of combined cycle gas turbine [$,zkW]
Investment on distribution capacity per year without PV
r$1
Investrent on distribution capacity per year with PV t$l
lnstallation cost ofpeaker combustion turbine [$2114I]
Growth rate
Reserve capacity margin
nth year of andysis period
Ouput of the PV tkWhl
lst year load capacity [kWl
10th year load capacity [kW]
Distribution cost [$/k1 I]
PV fleet shape [kW]
Solar capacity cost [$AW]
Utillty cost t$I
Utility fixed operadon and maintenance cost [$AW
Utility price t$AlVhI
Utility transmission capacity cost t$AWl
Utility variables operation and maintenance cost [$,4rWh]
Value of solar I$AWhl
V1: Avoided operadon and maintenance fixed cost [$]Vz:
Avoided operation and maintenance variable cost [$]Vg:
Avoided fuel cost [$]V+: Avoided teneration capacity cost
[$]Vs: Avoided reserve Capacity cost [$]Ve: Avoided
transmission capacity cost [$]Vz: Avoided distribution cost
[$]Vg: Avoided environmental cost [$]Vq: Avoided health
liabilig t$l
K.S. Hayibo utdJ.M. Pwe
retail price [51,53]. Performing a complete VOS calculadon, however, is
challenging. One of the main challentes is data availability and accuracy
[54,55]. Three data challenges have been identified by Ref. [55] that
are: 1) the time granularity of the solar irradiation daa, 2) the origin of
the data, modeled versus measured, and 3) the daA measurement ac-
curacy. Other challenges faced by utilities while assessing the VOS are
which components to include in the cdculations, and what calculations
method to assess the value of each components [56]. The possible
componmts across the literature that are suggested to be included in a
VOS as avoided costs and solar benefits are: energy producdon coste
(operadon and maintenance) [4547,57-$3], dectridty tmeration
capaclty costf [45-47,50,57-63], tranrmirslon capacity cofts
l4s-47,50,57 --61,631, dietribution capaclty costs [4H7,5 o,57 43),
fuel coots [4547,50,57,60-63], environmental costs 145,47,57,58,
60-631, ancillary including voltate control beneEe 147,57-59,631,
colar integradon coctc [47], market price reduction benefiB [47,
601, economic development value or job creation 146,47,57,60,617,
health liabillty costs [57,60,64], and value offurcr€ased secudty [47,
571. A guidebook has been developed by the United States' lnterstate
Renewable Energy Council (IREC) for the calculation of several of the
VOS components [57]. These methods have been further developed by
the U.S. National Renewable Energy Laboratory (NREL) [58]. NREL has
provided more deailed calculation methods than the guidebook from
the IREC with a different level of accuracy. The methods with a higher
level of accuracy are more complicated to implement and require a
higher level ofdata granularity. A qualiative study on VOS performed in
2014 suggested the inclusion ofall relevant components in a VOS studies
[64]. The calculation of the VOS can be done annually, as in the case of
Ausdn Energy [50,53], or can be ffxed for a selected period, as per the
case ofMinnesota state's VOS (25 years) [45,53]. lhere are recently an
increasing number of studies looking into extemality-based componens
of VOS especially environmental costs and hedth liability coss 165-471.ltis is because a country with high solar PV penetration rate provides a
healthy population according to a Genn:rn study [68J. An estimated
average of 1424 lives could be saved each summer in the Eastern United
States, and $13.1 billion in terms ofhealth savings ifthe total electricity
teneration capacity in the Eastem United States included lTVo of solar
Ratnablc ail Su*aindle Enogr Ruizws 137 (2021) 11059
PV 1691. For the entire U.S. if coal-flred electridty were repliaced with
solar generatio& roughly 52,000 prematur€ American deaths would be
prevented from reduced air poUution alone [70]. Not surprisingly, the
latest report from North Carolina Clean Energy Technology C.enter
found out that there are policy changes on VOS across the United States
with 46 states, in addidon ofDC considering making significant chanSes
in their solar policies and might be transidoninS to a VOS model in
cominS years [63].
this indicates VOS is the way of the future for grid integrated PV, but
how exactly should solar be valued on the modem grid? In this study the
VOS literature is reviewed, and a teneralized model ls developed aking
realistic future avoided costs and liabilities into account from the liter-
ature. lte approach used here ls a bonom-up modellng where the f,nal
vdue of solar to a utility system is calculated. this model factors in the
existing parameters, that have been identified in VOS studies in different
U.S, juridictions. The approach smrts from the existing formul,a to
calculate the levelized cost of electricity from solar PV technology [1 2]
and updates the formula by adding the avolded and opportunity costs
and the effect of different ertemdities. The costs considered in the study
are: avoided plant operation and maintenance (O&M) fixed cost; avoi-
ded O&M variable cost; avoided fuel cosq avoided generation @pacity
cost, avolded reserve capacity cos! avoided transmission capacity cost,
avoided distribution capacity cosg avoided environmental cost, and the
avoided health liability cost. The value of solar riepresents the sum of
these costs. Each subcomponent of the VOS has a sensitivity analysis
nrn on the core variables and these sensitivities are applied for the total
VOS. These sensitivides are limited by the best available data on the
variables in the literature and future work is needed to quantify the
secondary costs that would lead to an even hlgher VOS. The conservative
results developed here are presented and discussed in the context of
digning policy and regul,adons with appropriate compensation for
Pv-asset owners and electric utility clstomers.
2
K,S, Hryho ndJ.M. Pme
2. Methods,/theory
2.1. Awidet plott O&M - fixed cost N)
Ihe use of solar energy results in a displacement of eneryy produc-
tion from conventional energy sources. The avoided cost of plant oper-
ation and maintenance (Vr) I$] deperds on the energy saved by using
solar PV for electricity tenerirtion instead of conventional energy gan-
eration processes. Equadon (1 ) describes the calculation of the capacity
of solar PV (Cpv) [klM throughout the lifetime of the solar Pv system.
Dtrring the first year of operation, the installed solar PV system is
considered to not have suffered anydegradation. Therefore, the capacity
has a value of one. The degradation of the insalled solar PV system is
expressed by the degradation rate of PV (Dpv) and for a marginal year
(n), the marginal capacity of the installed PV rystem for that year would
be:
cPv: (l - DPv)' (1)
The fixed O&M cost is directly linked to the need for new conven-
tional electridty generadon plants. If the construction ofnew conven-
donal generators in the location of interest can be avoided, there is no
need to include the fixed O&M in the vduation of solar for this location.
To calculate the value of the fixed O&M (Vil, the value of the udlity cost
(Ud t$I needs to be known first. The utility cost depends on four pa-
rameters, the capacity of solar PV (Cpv) mentioned above, the utility
capacity (Cc) tp.u.l, the utility fixed O&M cost (Ur) t$AVVl, and the
utility discount factor (F). To calculate this utility cos! first the rado of
the capacity of solar to the utility capacity is calculated. Ttris ratio is then
multiplied by the utility fixed O&M cosl A discount is applied to the
result by multiplying it by the utility discount factor [71]. The discount
factor (E) depends on the year and can be cdculated by using the dis-
count rate (D). The discount factor for year (n) is [45]:
F:I
ir +D)'
The discount rate used in the formula describes the uncertainty and
the fluctuation of the value of money in time. The value of the discount
rate differs when considered from a utility point of view or a societal
point of view and can highly impact the utility cost, Whil€ considering
the economics ofsolar PV systems [57], has suStested the use ofa dis-
count rate lower than the value used by the utility.
Ur.:gr*C!v*, (3)'Cc
The avoided plant O&M fixed cost (Vr) is then calculated by sum-
ming the utility cost for all the years included in the analysis period.
v,: I[,u. (4)
2.2. Awided plnt O&M - voiable cost U)
The udlity cost for the avoided variable O&M cost (Vd t$l is
calculated by multiplying the utility variable O&M cost (Uv) t$Awhl
by the energy saved by using solar PV systems or the ouput of the soliar
Pv system (o) [kWh], and the result is discounted by the discount factor
(E).
Uc:lJv*O*F (5)
fire avoided variable O&M (Vd cost is the sum of the utility cost over
the analysis period:
vr:l,ou" (6)
2.3. Awidet@ cost (Vs)
Rwable Mil &r*aingD/b FMgt Rsiils 137 (2021) 110599
require the knowledge of the equivalent heat rate of a marginal solar.
According to Ref. [72], the heat rate [BtuAWh] describes how much
fuel-energy, on average, a generator uses in order to produce I kWh of
electridty. It is typically used in the energy calculation of thermal-based
plants and is therefore misleading for the calculation of solar energy
producdon. Since the method evaluates the avoided cost from
thermal-based pLants, however, it is applied to solar PV generation, Ihe
heat rate (Hs) tBtu/kWh] of solar PV or displaced fuel heat rate during
the first marginal year is calculated as:
Hs (7)
In the equation above, the heat rate (Hp) tBtuAWhl represent the
real value of the udlity plant's heat rate during the operation hours of
the solar PV systems over the analysis period and the parameter (S) [kW]
describes the PV fleet shape that is the hourly PV fleet shape producdon
over the houn (h) in the analysis period.
After the heat rate for the flrst year has been calculiated, the heat rate
for the succeeding years in the analysis period can be calculated by the
followlng equadon [45] :
H,:HIQ - Dt)' (8)
The primary use of heat rates is the assessment of the thermal con-
version efficiency of fuel into electricity by conventional power plants.
As a rezult, it is natural to deduce that the rate at which the heat rate
(Ds) decreases corresponds to the effrciency lost rate of the power plant
t731.
The udlity price (Up) depends on the heat rates and can be calculated
once the heat rate is known as:
-. B"H"u,=-G (e)
Another parameter to account for is the bumer dp price (B)
t$/MMBtul. the bumer dp price describes the cost of buming fuel to
create heat in any fuel-buming equipment [74].
Ihe avoided fuel cost (Vr) t0] is calculated in a similar way as the
value of the fixed O&M. First, the utility cost is calculated by multiplying
the value of the per unit PV output (O) by the utility price (Up). The
result is then discounted by the discount factor. ltre discount factor used
in the case ofthe avoided fuel cost depends on the treasury yieH tasl.
The avoided fuel cost is obtained by summing up the utility cost over the
analysis period.
Uc: Up*O*F
:Dl(r,.s)
5is
(2)
v':\'0u,,
(10)
(11)
2.4. Avoided gawation cqacity cost Ue)
The installation of solar systems reduces the generation of electricity
from new plans. I'his is represented by the avoided capacity cost. To
cdculate the avoided generadon capacity cost, the solar capacity cost
(Sd t$AW1 needs to be known. T\ryo variables are essential to evaluate
the solar capacity cost, the cost of peaker combustion turbine (Ip)
t$Awl and the installed capital cost (Ic) I$AWI. The cost of peaker
combustion turbine (Ip) is the cost associated with the operation of a
turbine that function only when the electricity demand is at its highest.
The insalled capial cost 0d describes the cost of combined cycle gas
turbine updated by the cost based on the heat rate. The solar capacity
can be calculated as follows [75]:
sc:/c* (ur-nryL.r_n, (r2)
Hcr [BtuAWh] and Hc tBtu/kwhl are respectively the heat rate of
the peaker combustion turbine, and the combined cyde gas turbine,
After the calculation of the solar capacity cost (Sd, the udlity cost can beAdditionalln the calculation of the utility price (Up) t$AWhI
3
K.S. Hayfio @tilJ,M. Peocc
obained by first, multiplying the ratio of solar PV capacity (Cpv) and
utility generation @pacity (Cd by the value of solar capadty cost (Sc).
Theq the result is discounted by the discount factor (F) to obain the
final value of the utility cort. And as in the previous cases the value of
avoided generation capacity is the sum of the utility cost overs the
analysis period.
Ur=5r*C!v*, (f3)'Cc
vn:l'ou" (r4)
2.5, Avoided ramte cqacity cost (V;)
Ihe calculation of the avoided reserve capacity cost (Vr) [$] follows
the same pattem as the avoided cost of generation capacity. But in this
case, the effective solar capacity, that is the ratio of the solar PV capacity
(Cpv) and utility gmeration capacity (C6) is muldply by the solar ca-
padty cost, then the result is multiplied by the reserve capacity margln
(M) to obtain the utility costs. After that, the utility cost is discounted as
previously described by the discount factor (F). Then, the avoided
reserve capacity is calculated by adding up the utility cost over the
analysis period [58].
Ur:5r*fu*Y*P (15)" "Ce
vt=\'ou" (16)
2.6. Avoidcd ransnission cqacity cos- N6)
The avoided transmission capadty cost (V6,) [$] cdculadon is also
performed similarly to the avolded generadon capacity cost. This cost
describes the losses that are avoided when electricity does not have to be
transported on long distance because of installed solar systems. It is
calculated by ffrst muldplying the utilty tran$nission capacity cost (U1)
l$/kwl Uy the solar PV capaciry (Cpv). The result is then divided by the
transmission capacity (Cr) [p.u.] and the discount factor (F) is applied to
obtain the udlity cost for a marginal year. The avoided transmission cost
is calculated by the sum, over the years in the analysis period, of the
corresponding utility costs [76],
IJ. -- gr*C!' * P (17),Cr
v6: tiur (r8)
2.7. Awfuleil disoib,ttion cqacity cost (Vz)
The two major variables that influence the avoided distribution ca-
padty cost (Vz) t$] are the peak growth rate (K) and the system wide
costs, The system wide costs account for several financial aspects of a
distribution plant, among whidr, overhead lines and devices, under-
ground cables, line transformers, leased property, streetlights, poles,
towers etc. [77].
All the deferrable system wide costs throughout a year have been
summed up and the rezult divided bythe yearlypeak load increase in kW
over a total period ofa decade to obtain the distribution cost per growth
ofdemand.
The ratio of the 10th year peak load (PLrd tk$4 and the lst year
peak load (PLl) IkWl are used in the calculation of the growth rate (K)
of demand. Ttre exprcssion of the growth rate (K) is as follows [45,78]:
x:13it - 1 (re), PLI'
The distribution capital cost (Q) t$AW1 is utility owned daa and
depends on the utility, and the growth rate (K) that can be obtained by
Rwdle atrl Sul$oi,ublc E tdg Rf,iilt ,37 (m21) 1ro599
using the previous formula, fui escalation factor is necessary to evaluate
the distribution cost for deferrd consecutive years [79].
After obtaining the distribution cost (Q) from the utility and Fowth
rate (IO calculated, the distribudon capaciry (Co) IkWl can be calculated
from the growth rate. the result is then multiplied by the digtribution
cost and discounted by the dlscount factor (F) to 8et the discounted cost
for a particular year. ltre discounted cost for the andysis period can in
tum be used to calctlate the invesment during each year 0o) [$] of the
analysis period [45].
lo:Co*Q*F (20)
When there is no other ganeration rystem than solar PV that
comprircd the installed @pacity, the investment per year 0op) t$] in
terms of deferred distribudon can be calculated from the invesunent
deferred [45].
Iop:Co*Q*DF (in rcrms of deferred distribution) (21)
After obtaining the yearly investment without Pv (Ip) and the yearly
investment in terms ofdderred distribution (Iop), the utility cost can be
obtained by dividing the difrerence between the yearly invesunent
wlthout PV and the yearly investrnent with PV by the distribudon ca-
padg (Cd. This udlity cost can be called the dderred cost per kW of
solar. lhis dderred oost per kW of solar is discounted by the discount
factor (F), multiplied by the solar PV capacity, and summed up over the
analysis period to obtain the avoided distribudon capadty cost.
u"=|fu*p*grn e2)'Co
vr:l,our, (23)
2.8. Awirlcd qutunnstnl cost Ua)
The three major pollutants that are consldered in the calculadon of
the avoided environmental cost (V8) [$] are: Sreenhouse Sases (GHGs),
pollutants zulfur dioxide, nitrogen oxide, and hazardous particulates
t801.
The two parameters that influences the cost linked to CO2 and other
greenhouse gasses' emission are the socid cost of CO2 and the gas
emission factor [81]. With these two variables, the cost of avoided COz
can be calculated in dollars and then the real value linked to this cost is
obtained by converting the previously calculated value in current value
of dollars. This is done by multiplying the externality cost of CO2 by the
consumer price index (CPI) [82]. The obtained result is then muldplied
by the general escalation rate for the following years [80]. The cost of
CO2 for every year is obtained by multiplying the previous value by
pounds of COz per kWh. the same logic is applied to the other polluants
to calculate the related costs and the cost related to all three categories
of pollutant are added up to get the environmental cost (E) t$/MMBtul,
By multiplying the environmental cost by the solar heat rate (Hs), the
utility cost (Uj is obtained, An environmental discount factor (Fs) is
applied to the utility factor. the environmental discount factor (FE) is
defined as follows [83]:
_lh. -- (24)'"-(t+DE)"
Here, Dg is the environmental discount rate taken from the Social
Cost of Carbon report [811.
Uc : E*Hs*Fe*O (25)
v,:l'ou,
2.9. AwidedhealhhobW cost Ue)
(26)
4
The use of solar PV systems prevents part of the emissions of
R.S. Halho ailLJ,M. P@e
pollutants from tetting into the air. This can in tum result in geat health
benefis. The harmfrrl pollutants that greatly impact human health are
NO, and SO2. These nivo chemicals react with other compounds when
they are released in the air to form a heavy and harmful product that is
called pardculate matter PM2.s, [84-86]. Pardculate matter PM2.5, can
cause diseases such as lung cancer and cardiopulmonary diseases [87]. It
is difficult to evaluate the cost related to the avoided health liabilities
and the saved lives. Several works have investigated the calculation of
the cost of human hedth related to electridty production through fossil
fuels [88-91]. Nevertheless, the most relevant approach is the work of
[91] because the methods accounts for changes of the cost at a regional
and plant level. Ttris has been made possible because of data collected by
EPA on the emission level of facilities through the Clean Air Markets
Program. The result obtained by Ref. [91] is conservative as it does not
include environmenal impacts over the long term (e.g. climate change)
[66, 68, 69, 92]. The calculadon of the cost of health liability by Ref. [9 1 ]
depends on the quandty of pollutants emitted [tons,/year] during a year,
the cost ofa unit mass ofemission for each pollutant in [$/tons], and the
annual gross load [kWh,zyear].
The health cost of energy produced by fossil fuel sources (Cn)
t$Awhl obained by Ref. [91] are used to calculate the utility cost. The
utility cost (Uc) is the product of the health cost by the PV systems
output (O), that is discounted by the environmental discount factor (Fd,
Uc=Cn*O*Fr (27)
The avoided health liability cost (Vg) t$l is then calculated by:
vo:l'ou" (2s)
2.10. VahE of nlor UOS)
There are three different ways to represent the value of solar. It can
be expressed either as the annual cost [$] over the analysis period or the
Iifetime of the installed solar photovoltaic system, or as the cost per unit
of solar PV power insalled t$A!Vl, or finally as the cost of generated
electricity by the solar rystem I$AWhl [58]. The most commonly used
metric to express the VOS is the cost ofelectricity generated by the solar
system [$AWh] because it is user friendly and is the same metric used
by utilities on electridty bills [58]. To calctlate the levelized value of
VOS per kilowaft-hour of electricity produced, the sum of the value of all
the avoided cost is cdculated and then divided by the toal amount of
energy produced (O) during the analysis period discounted by the dis-
count factor (E).
vOS-V,*Vz+Vt*Vn:Vs+Vu+Vl+V*+V' eg)D;(O*r)
where:
V1: Avoided O&M fixed cost.
V2: Avoided O&M variable cost.
V2: Avoided fuel cost.
Vl: Avoided gmeration capacity cost.
V5: Avoided reserve capacity cost.
V6: Avoided transmission capacity cost.
V7: Avoided distribution cost,
Vs: Avoided environmental cost.
Ve: Avoided health liability cost.
O: Output of the solar PV rystem.
F: Udlity discount factor.
3. S€ositivity
The calculation of VOS requires several parameters that come from
different sources. Some parameters are locadon dependent, while other
parameters are state dependent, and there are parameters that are udlity
dependent. Many ofthese parameters can also change from one year to
Rawoble @ld su,foi,ublc Engat Riltcws 137 (2021) 1 10599
another. As a result, there are wide differences in the calculation ofVOS
across the literature [56]. The udlity-related parameters that can change
from one VOS calculation to another are the number of years in the
andysis period (i), the udlity discount rate (D), the utility degradation
rate, the udlity o&M fixed, and variable costs, the o&M cost escaladon
rate, the hourly heat rate (Hp), the heat rate degradation rate (Dd, th€
reserve capacity margin (M), the transmission capacity cost (Ur), the
peak load ofyear I CPLI) and year l0 (PLro), the distribution cost (Q),
the distribution cost escaladon factor (Go), and the distribudon capadty
(Co). Parameters such as the cost of peaker combusdon turbine (Ip), the
cost of combine cycle gas turbine (Id, the heat rate of peaker combus-
tion turbine (Hcr), and the heat rate of combine cycle gas turbine (Hd
can be either obained from the utility or from the U.S. Energy Infor-
mation Agency. The solar PV fleet (S) can also be obtained from the
utility or by simulation using the open sounce Solar Advisory Model
(SAITO (https :,/,/gi thub. com,u NREL,/SAM) [45]. Other variables that can
affect the VOS but are not controlled by the utility are the PV degra-
dation rate (Dpy), the environmental discount factor (Fd, the environ-
mental cost of conventional energy, the health cost of conventional
energy, and the cost of natural gas on the energy market. Table 1 sum-
marizes high and low estimates of the values for the variables that are
required to perform a VOS calculadon and the VOS component they are
used to cdculate.
3.1. Numbo of yeors in analysis pclriod
Ihe number ofyears in the analysis period varies and can be as low as
20 years, and as high as 30 years or more 112,571. The typical warranty
provided by solar panels manufacturer is 25 years. As a resrlt, it is
reasonable to set the lowest value of the analysis period to 25 years.
AIso, solar modules have proved to continue to reliably deliver energy
30 years after the installation ofthe system [57], therefore, 30 years has
been set as the higher value ofthe analysis period in this study. IGyes
et al. have pointed out that utility planning is often over shorter time
periods (e.9. 10-20 years) [57]. However, economic decisions should be
made over the entire life of the physical project not an arbitrary cutoff
date [r02] and there are existing methods to estimate the load growth
on the utility side as it is usually done for convmdonal energy genera-
tors [53].
3.2, PV systan ilegrailation mte
the degradation rate of PV panels overtime depends on the location
of operadon as well as climate conditions (temperature, wind speed,
dust, etc.). A statistical study conducted by the National Renewable
Energy Laboratory [93] has found the value ofthe PV system degrada-
tion rate to be comprised between 0.5026 and l%. These two values are
the boundaries that will be used as low and high values for the sensi-
tivity analysis on the PV rystem degradation rate.
3.3. UWdixowrtmu
The discount rate is used to assess the change in money value over-
time. This value can change depending not only on the locadon, but also,
on the utility. A discount rate value as high as 9% can be used or a value
as low as the lnfladon rate might be used. lte discount rate used by
utilities are usually in the high values, but the social discount rate is
closer to the infladon rate [57]. As a result, 9% will be considered as the
high-end value of the discount rate while the current inlLation rate of
2.18% will be considered for the lowest value. It is important to note that
the value of the inllation rate changes with time and if this value is
chosen as the diseount rate it should be updated regularly for new cal-
culations of the VOS. Also, the value of the inlladon rate can be sub-
jected to ongoing events. The value of the infladon rate of 2.1896 was
chosen at a date before the coronavirus outbreak in the United States
that is ongoing. the outbreak has brought the inflation rate to as low as
5
K.S. Hayibo utd J,M. P@tr
Table I
Assumptions used for required variables for a VOS calculation.
Rwsblc Mil Strstoindle Bugr Ruim 137 (202r) 110599
Variable High Sotrce lrw
6timte
Sotrce VOS componmts
stimate
Degradetion rate of PV (Dpv) [%]
Distribution capecity (Co) tkwl
Disaibution cmt (Q) [S/kW]
Enviroment discount rate (Dd [%]
Enviromental Cmt (E) [$/metric tom of CO2]
Health cost of mtual em (CJt$/tflhl
Hmt rate degrodation rate (Dd [%]
Heat rate of combined cycle gas (Hq) tBtuAWhl
Hmt rste of peaks combstion tubine (Ho) tBtu^Whl
Instsllation cepital c6t of combined cyde gas turbine (ld
t$AwI
k$Ellation cost of peaka combution tubine (Ip) [$/k$rl
Lmd Growth Rate (() [%]
Nmbe of yem in analysis period
Reserue capacity marSin (M) [%]
Solar Heat Rate (Hd tBtu^Whl
Trarumission capacity cost (Ur) [SAW]
utiliry Discount rare (D) t%l
Utility fixed O&M cosl (Ur) [S/kW]
Utility viliable O&M c6t (Uv) [$AWhl
ll oll
tsTI
lr()lI
Is7]
1E.E6
0.01153
7.44
o.00215
All componmts
Avoided distribudon coet (Vz)
Avoided distribution cost (Vz)
Avoided enviromental cst (V8)
Avoided oviromental cost (Va)
Avoided health liability cost (Vg)
oAvoided fud cost (Vg)
oAvoided mvirommtal coot (Vs)
oAvoided gmeratlon capacity cost (Vr)
oAvoided rswe upacity c6t (Vs)
oAvoided goeration capacity cost (Vr)
oAvoided rserye capscity cost (Vs)
oAvoided goeration capacity cost (Vr)
.Avoided reserye epacity c6t (Vs)
.Avoided gmeration cspacity cost (V4)
oAvoided rcwe epacity st (Vs)
Avoided distribution capacity cost (Vz)
All components
Avoided reserue cepacity (vs)
oAvoided fuel cmt (Vs)
.Avoided 8€nerstion capacity c6t (Va)
oAvoided merve caDacity cst (Vs)
oAvoided mvirommtal cost (V8)
Avoided tranmision cspacity (V6)
.Avoided plmtr O&M fixed cost (Vr)
.Avoided plants O&M vriable (V,
.Avoided Smeration capacity cost (Va)
eAvolded rsere capacity cost (Vs)
oAvoided transmission capacity cost
(vo)
.Avoided distribution epacity c6t (V7)
Avoided O&M fixed cost (Vr)
Avoided O&M vuiable cost (V2)
I
429,000
1104
2.5
[62-8e]
0.02s
o,2
Ie3]
[es]
Ies]
[81]
[81]
tell
Ie6l
leTl
1e7l
Ielt]
Ie13]
llool
0.5
237,OOO
678
5
ll 2-2:rI
0.025
0.05
-0.94
Is7,93,94]
Ies]
Ies]
[81 ]
[81]
[et]
Ie5l
7627
11,138
896
1496
t.77
30
36
8mo
Iee]
PV indutry
wffimtis
[1oo]
[5:i]
lqel
Is7]
['r5l
Ies]
Ies]
lesl
25
13
17.895
2.18
130.lj5
9
0.25%. This value will not be used to run a sensitivity analysis because of
the special conditions in which it occurred.
3.4. hwiromwral cost
The environmental cost associated with electriciry production
through conventional energy sources depends on the cost associated
with the pollution from carbon dioxide (COz), carbon monoxide (CO),
nitrogen oxide (NOJ, and hazardous particulates (PM). The environ-
mental cost of carbon dioxide dominates the cost of the other compo-
nents. Different estimates of the CO2 cost are given by the EPA [81]. The
cost of CO, NO, and PM depends on state laws. The lowest value and
highest value used for the cost ofCO, NO*, and PM were chosen from the
state ofMinne$ota [l03], It has been hypothesized that ifconventional
energy sources are being used to produce electricity in the future, the
effects on environment are going to worsen (e.g. lower quality fuel,
higher embodied energies, etc.), therefore the environmental cost will
be expected to increase. This will be investigated by raising the envi-
ronmental cost while analyzing the sensitivity of VOS to the environ-
mental cost. This will show the trend of the impact of the environmental
cost on the VOS and in the future, the values will need to be updated
because the environmental cost is likely to exced the maximum used
value in this srudy.
3.5. Heahh liability cost
The health liability cost is a new calculated VOS component intrc.
duced by this study. This component has been mentioned by several
studies but was not incorporated in the calculation due to lack ofdata for
the evaluation 157 ,66,67,lO4). The health and mortality impacts of coal
in pardcular are so severe an ethical case can be made for the industries
elimination [1 05]. For example, Bumey estimated that 26,610 American
lives were saved htween 2005 and 2016 by a conversion of coal-fired
units to natural gas in the U.S U06]. More lives as well as non-Iethal
health impacts would be avoided with a greater transition from coal
to solar [70]. The values used here were obtained from the study of [91]
that found the value of health impact cost of natural gas to be
$0.0254Wh, As previously hypothesized, the use of fossil fuel energy
sources in the future will increase the emissions, and the cost of health
care has been escalating faster than inflation [106] thus increasing the
cost of derived hedth liability. Several increase rates lvill be investi-
gated. Although it should be pointed out the approach taken here was
extremely conservative as the potential for chmate/greenhouse gas
emission liability [107,108] was left for future work as discussed below.
3.6, Odur pownetos
Ttre other par:rmeters are utility related and in case of absence of
utility data, generic values from the U.S. Sovemment agencies is used as
indicated in Table 1 and run through realistic p€rcent increases or de-
creases to determine their effect on the VOS components.
3.7. Seraitivity onarisis
A sensitivity analysis has been run on each of the nine VOS compo-
nents as well as on the VOS. For each component, the sensitivity has
been analyzed for some of its parameters wherever data was available.
The evaluation of the variability of the VOS components has been per-
formed for each parameter. The sensitivity of a component to one of its
parameters is determined by mainaining an average value of the other
parameters and varying the srudied parameter from its lowest value to
its hiShest value. The different values that are obtained for the VOS
component are thelt plotted to show its variation according to the
parameter studied. A correlation study between the different parameters
5
K.S. Htyibo onl J.M. Pwce
has not been conductd because there was no evident relationship be-
tween these parameters. Most of the par:rmeters are set by the utilities
and is often not disclosed openly. An interaction study between the
parameters and how their interaction afiects the VOS components would
be interesting for future studies where udlity data are available,
A similar process has been used for the sensitivity analysis of the
main VOS. The main VOS's variability has been studied according to the
nine VOS components. For each component for whtch the sensidvity of
the VOS is analyzed, average values of the other components are
mainained while the studled component's value is varied from its
lowest value to its highest vdue.
4. Rcsults and diccufslon
The simulation resuls are ploned first for each VoS components. For
each component, sensitivides on the different input variables have been
investigated. Then the sensitivity of the overall VOS to each of the VOS
components has been analyzed.
4.1. Awided OAM fixd cost U)
Fig. I shows the results for the avoided O&M fixed cost (Vr). The
sensitivity has been plotted for five parameters: the utility O&M fixed
cost, the utility O&M cost escalation, the PV degradation rate, the udlity
discount rate, and the utility degradation rate. According to the results,
the avoided O&M cost is highly sensitive to the utility O&M Exed cost
and O&M cost escaladon. When the utility O&M fixed cost increases, the
avoided O&M cost increases accordingly and an increase in the O&M
escalation rate obviously increases the avoided O&M cost because it
increases the udlity fixed O&M cost over the analysis period. V1 is also
sensitive to the utitty discount rate and decreases when the discount
rate incr€ases. This means that using a discount rate close to the social
1.20
1.00
0.80
Rwable otil Su*aindle Endgl RdiM 137 (2o2r) llosw
discount rate while conducting a VoS study will increase the avoided
O&M cost while using a higher discount rate will lower the cost. This is
in accordance with the recommendation of [57] that is the use of a
discount rate lower than that of the utility in a distributed solar gener-
adon economic calculadon. Also, the avoided O&M fixed cost is not very
sensitive to the utiltty degradation rate or the PV degradation rate.
Nevertheless, its value is slightly reduced when the PV degradation rate
increases.
4.2. Awided O&M vsiable cost U,
The parameters for which the avoided O&M variable cost's (Vz)
sensitivlty has been studied are: the udlity O&M variable cost, the utility
O&M cost escalation, the PV degradation rate, and the utility discount
rate. I'he sensitivity of the avoided O&M to its parameters are plotted in
Fig. 2. Fig. 2 shows a similar variation trend of V2 as compared to the
case of the avoided fixed O&M cosL It is highly sensitive to the utility
variable O&M cost, and the O&M cost escdation. Ihe avoided variable
O&M cost increases when the variable O&M, or the o&M cost escalation
rate is incre:lsd but decreases with the increase of the discount rate, and
the PV degradation rate.
4.3. AwidedMcost(Vi)
ln the case of the avoided fuel cost (Vg), the variable considered for
the sensitivity andysis are the heat rate degradation rate, the natural gas
price fluchration rate and the PV degradation rate. While the avoided
fuel cost has shown to be not very dependent on the heat rate degra-
dadon rate or the PV degradation rate, tlris value changes very quickly
with a change in the natural gas price as in Fig. 3. This is an important
factor that should be carefully considered while conducting a VOS study
because the price of natural gas is not fixed and varies according to
0.60
0.40
0.20
=
v!r
0.00
-lO0o/o -80o/o 4@/o 40o/o
.+-Utility fixed O&M co$
-r.-O&M co$ escalation
+-PV degadatiourate
-20o/o 0o/o 2U/o
PERCENT CHANGE f/ol
-r- utility discourt rate
, Utility degradati@ rate
4ff/o 600/o 80o/o l00o/o
Fit. l. Sensitivity of avoided O&M fixed cost (Vr) in terms of LCOE (oAWh) to its parameters in percent change.
7
K.S. Hayho utdJ.M. Pw
1.60
1.40
1.20
1.00
0,80
0.60
0.40
0.20
vu
a{
0.00
-100o/o -8@/o 40o/o
several parameters that are not controlled by the utility such as, the
economy, the weather, market supply and demand U09,ll0l. The
equivalent heat rate degradation rate expresses the degradation of the
utility plant's efficiency over the analysis period and when the efEciency
decreases, there is a slitht decrease in the avoided fuel cost. Another
value for which the avoided fuel's sensitivity could have been studied is
the equivalent heat rate for solar, which was not analyzed in detail here
because of the Iack of utility daa. Ihis is left for future work.
4,4, Avoiilel gatration capacity cost U4)
The sensitivity ofthe avoided generation capacity cost (V+) has been
plotted in Fig. 4 for the discount rate, the utility degradatioq and the PV
degradation rate. The Vr VOS component does not have a high vari-
ability to the PV degradation rate even though it shows a decreasing
trend with the increase of PV degradation. But it reacs sharply to the
utility deSradation rate. This is because the generation capacity ofthe
utility is highly impacted by the utility degradation. Alsq as previously
observed, when the discount rate grows far from the social discount rate,
the avoided generation capacity cost decreases.
4,5. Avoideil rauve capacity cost (Vs)
The avoided reserve capacity cost (Vs) expresses the reserve
component of the generation capacity; therefore, it can have a value of
zero when there is no reserve capacity planned by the utility as shown in
Fig. 5. V5 is highly sensitive to the reserve margin and the result shows
that the more generation capacity is reserved, the more the avoided
generation capacity cost increases. On the other hand, the avoided
reserve capacity cost is lrot very sensitive to the discount rate coDpard
to its sensitivity to the other pararDeters. Vs's value goes up when the
utility degradation rate increases and goes down when the Pv
+Utility variable O&M cost
-*-O&M cost escalatiou
40o/o -2V/o 0o/o 2@/o
PERCENT CHANGE [%]
+ utility discorurt fate
* - PV delradation rate
40o/o 6@/o 80o/o lO{Jo/o
Fig. 2. Sensitivity of avoided O&M variable cost (V, in terms of LCOE (O/kWh) to its parameters in percent change.
degradation rate increases.
4.6. Awidcd ransrnission cqacity cost U6)
ltree parameters have been analyzed in the sensitivity study of V6:
the discount rate, the transrnission capacity cost, and the PV degradation
rate. Ihe parameter it is the most sensitive to is the transmission ca-
pacity cost. Obviously, when the Eansmission is low cost in a locadon,
the avoided cost associated will be low. The results shown in Fig. 6 make
it clear that the avoided transrnission capacity cost does not change with
the PV degradadon rate or the discount rate. Ttris is because the utility
tran^smission capacity has been assumed to be constant over the analysis
period, and the transmission capacity degradation rate has not been
considered because utility data on this parameter was not available.
4.7. Awided distibtttion cqacity cost Uz)
Ihe avoided distribution capacity cost (Vz) is one of the most
complicated VOS componmts to evaluate. As shown in Fig. 7, its
sensitivity has been studied for six variables: the load growth rate, the
distribudon capacity, the distribution capacity cost, the utility discount
rate, the distribution cost escaladon, and the PV degradatio[ rate. But it
depends on more than six parameters. The growth rate, for example is
cdculated from utility daa, mainly, the load for the past ten years of
operation [45,111], Here, the sensitivity has been analyzed on the
growth rate directly to be as widely applicable as possible. Another
parameter is the number of deferred years that is also a utility owned
daa,
Itre avoided distribution capacity cost naturally increases with the
distribudon capital cost Fig. 7 shows that the avoided distribution ca-
pacity cost does not fluctuate with the distribution capacity at all, but it
is highly sensitive to the discount rate, the distribution cost, and the
8
Raewable onil Susoinable EnaE/ RuiM 137 (2021) 110599
K.S, Hayho otrl t.M, Pwce
4.00
3.50
3.00
2.50
=>
Yu 2.00
1.50
1.00
0.50
0.00
-l0oo/o
-t-Heat rate degradation rate
-BV/o -600/o AV/o
*Gas Price fluchration --*-PV degadationrate
40o/o 6V/o 80o/o l00o/o-2V/o 0o/o 20o/o
PERCENT CHANGE [o/o]
Pig. 3. Sensitivity of avoided fuel cost (Vs) in terms of LGOE (c,/kYtlh) to its parameters in percent change.
distribution cost escalation rate. It cian even shift to a negative vdue
when the discount rate is too low. This shows that choosing the discount
during a VOS study must be a trade-offbetween the social discount rate
and the utility discount rate. It is interesting to note that the avoided
distribudon capacity cost goes down when the distribudon cost escala-
don in increasing. A possible oelanadon for this observation is that
when a utility has enough distribudon capacity, it will purchase less
power from solar PV rystems owners, therdore the price toes down. The
same reasoning can be used to explain the decreases ofthe cost when the
load growth toes up. Finalln V7 shows a slight decrease with the in-
crease of the PV degradation rate.
4,8. Avoided envimtwrrrltul cost Ns)
The second most complicated component of the VOS calculation is
the avoided environmental cost (Vs). Ttre sensitivity has been analyzed
for the three environmental discount rate scenarios provided by the EPA
[81 ]. For each scenario, a sensitivity analysis has been conducted on the
environmental cost increase rate. Vs will increase when the chosen
environmental discount rate is low but overall, each of the three EPA
scenarios show an increase when the environmental cost increase rate
Soes up as seen in Fig. 8. This is useful to see how the avoided envi-
ronmental costs might change in the future. Environmental extemalitjes
are volatile and changing quickly [66]. Ifit is assumed that ln the future,
the environmental impact of conventional energy producdon technol-
ogies will increase, then the costs of the environmental extemalities will
increase as well 0041. On the other hand, an increase in distributed
renewable energy generation could lead to a decrease or stabilization of
the avoided environmental cosL
4.9. Awilled health UabW cost (vs)
Ihe avoided health liability cost, Ve, depends on three values, the
health cost increase rate, the environmental discount rate, and the PV
degradation (see Fig. 9). This cost does not fluctuate with the PV
degradation rate but is very sensitive to the other two parameters. The
environmental discount rate used here is the same as the environmenal
discount rate used in the evaluation ofthe avoided environmental cost's
smsitivity study. As a result, the avoided hedth liability cost decreases
when the environmental discount rate Soes up as is the case for the
avoided environmental cost.
4.10. VOS
After the sensitivity analysis of each VOS component, the main VOS
value has been studied to find out how the impact of different compo-
nents compare to one another and which components have more vari-
ability. Fig. l0 shows that the VOS is, in decreasing order, sensitive to
the avoided environmental cost (Vs), avoided health liability cost (Vs),
avoided transmission capacity cost (Vo), avoided fuel cost (Vr), avoided
distribution capacity cost (Vz), avoided O&M variable cost (Vz), avoided
reserve capacity cost (Vs), avoided O&M fixed cost (v1), and avoided
generation capacity cost (Vt)
the contribution of each VOS component to the overall VOS depends
on the case. Ihe lowest VOS value cdculated with the assumpdons used
in this study in term of LCOE is 9.3704Wh while the highest value
calculated is 50.650/kWh. This variation observed in the VOS value
comes from the fact that the parameters values considered from this
study are chosen to have the lowest and the highest value ofa VOS. The
values ofcalculated VOS using utility data are highly likely to be located
within this interval. It is also clear based on the values shown in Fig. 10,
that the VOS exceeds the net meterinS rates (when they are even
9
Rmuabb otil Su*atudle Enag Rcvicn s 1 37 (m21 ) 1 10599
K&Hqi&,o@,ilJ.n PNte
3.05
3.04
3.03
3.4
3.01
3.m
2.99
2.98
2.97
B{u.
t
-l()o/o -BVh 6eh
+Utility dsccrot ratc
+P\y' dcgraddionratc
AV/o 4@/s 0o/o zWh
PERCENTCHANGE[7d
Rawd& od gffi BsA Rcy,,fln 137 (ZXll) 110599
lWo 6@/o 8O/o l0ii9o
lW. 6eh EVh l00Vo
nt 4. S@idvity of evoldcd gcoeradm caFdty ct (V, h tems of IOOE (CAlfr) to its Frmctcrr lo perreot change
t.60
1.40
tr0
t.@
0.80
0.6{)
0.{0
0.20
0.00
-r- Utility dcgndatim r*e
+RcscrvcDfigiD
E{u
-100o/o 3@h ff/o 4Wo AVh Vh 2e/o
PERCENT CHANGE Pol
ftg. 5. Scosldvity of avoldcd rcserE epactty c6t (Vd h t€ru of ICOE (lAWh) to tts paEmcten h pcrceot charye.
+utility disccril r8lc +-utility dcgradatio rate +-PV dcgradrtim rrie
l0
K&flq&ooldl.N"Puu
-100% {096 4W 4W.-2W A9$ 2W{
PERCEIII CIIANCTE[%l
lWc AW. 0Vo 2V.
PERCENTCIIANCE [06]
Ragl&b cld3rffi BgI,, irldan r37 (gEr) rlfft99
10?t 6grt t09t tw%
4Wt 60 EWo lfi)7o
EuB
ro
7.00
6.00
5.00
4.00
3.00
2.fi
1.00
0.00
1
i
I
I
Ii
1
I
I
1
I
I
I
I
I
I
I
T
I
I
II
i
i
I
I
Ff& 6. sdtlvlty of added trurobdo crpadty cc (Vc) ln tcm of IrOE (aAWt) to tg porl'D.fiert ln per@t dreD8e.
3.00
2.fi
2.00
r.50
Eug
=
1.00
0.50
0.00
-1fl)9{6 .8l0lh 6Vh
ng. 7. Sddvtt!, of avoilded dlfilbudm c.podty cct (Vz) ln tcrm of IIOE (tAWh) to its penoctcrr l! peiod drfiga
.+Iw
-a-DiscqlDtraE
+Utility tmirdo cedty cct
+Loadgmu,thralc
..r-Distrihrtiocryadty
+Disfrihrtioceitdcod
.*-Utilitydtcodrrte
-r-Dirfiihrtio oost Gsdstim
.-PV&guhlimraE
11
tr,.S. Hafho @tilJ.M, P@ce RJ,f..,vo,/ our Stttut4D/b Fi.aAt Reaiil. 1 37 eQl ) 1 10599
4V/o 6V/o $P/o l0oo/o
20.00
18.00
16.00
14.00
12.00
10.00
8.00
6.00
4.00
2.00
=>{g
o
+2.5% Discouot rote
-a-3%DiscomtraE
+5ToDiscountraE
0.00
-1000h 8V/o 4U/o 4V/o -2P/o 0oA 2V/o
PERCENT CI{ANGE [7ol
Fig. 8. Sensidvtty of avoided envlronmertal cost (Vs) ln terms of LCOB (oAWh) to its paEmetets ln pcrcent change.
E.00
7.00
6.00
5.00
4.00
3.00
2.00
1.00
F
{Ir
o\
0.00
-100o/o -$tr/o -600/o -40o/o
available as shown in Table 2) ln the U,S. Thus, it can be concluded that
even when Srid-tied solar owners arc provided with a full net metered
rate for electricity fed bad< onto the grid they are effectively subaidizing
the electric udlity,/other customers.
For the low VOS value case shown in Fig. I 1, the avoided distribudon
cost (Vz), and the avoided reserve capacitycost (Vs) has no contribudon
+Health liatility cos increase rate ,
+Envir@meotd discount rate
;
+PV deggadationrate '
-ZU/o Oo/o 20o/o 40o/o 6ff/o E0o/o l00o/o
PERCENT CHANGE [7d
Ftg. 9. Sensitivity of avoided health liability cost (Vq) in temrs of LCOE (CAWh) to its parameters in perceot ctunge.
in the VOS value. The avoided g€neradon @pacity cost (Vr) and the
avoided health liability cost (Vq) represent most of the VOS value fol-
lowed by the avoided environmental cost (Va) and avoided fuel cost
(vg).
Ihe contribudon of the avoided environmental (V8) cost increases
with the VOS value as it becomes the largest contributor to the overall
t2
K.S. Hcyibo o,lil J.M. Pwce
15.00
-100o/o -80o/o
Table 2
Comparison of VOS rates and net metering rates for some U.S. States.
State vos Net MetsinS
Ratilobb @il Sustsinfrle Fnogr Reviavs 137 (m21) 1 10599
+-Avcid O&M fixed Cost (Vl)
* Avoided O&M varioble cct (V2)
--.-- Avoided fuel cosr (V3)
Avoi&d geueration capacity cost 1v-4).+Avoi&d rcserve cspocity cct (VS)
+Avoided traDsmissio capecity cost (V6)
-Avoided
disrributiou capacity cosr O-.7)..-l-Avoi&d etrvirouefral cGt (V8)
---Avoi&d bealrh liability cosr (!9)
-20o/o 0o/o 20o/o 40o/o 600/o 80o/o l00o/o
PERCENT CHANGE [o/o]
45.00
40.00
35.00
30.00
25.00
20.00
u:l
cr,o
-600/o -40o/o
Fig. 10. Sensitivity of VoS LCOE (oAWh) to aU the components in this study, in percent change.
Mimesota
Awtin (Tems)
Maine
New Jereey
Pennsylvanis
13.s0awh
lo.7a/kwh
$.7aftWh
2s.6-28oAWh
28.2-3t.8c/
kwh
r9.4o,/kwh
Minimm value of (40AWh) [l l sl
Washintton D.
c.
value followed by the health liability (Vs) cost as shown in Fig. 12
representing a middle VOS value. ltre avoided Seneration capacity
cost's (V4) is reduced as well as the contribution of the avoided fuel cost
(Vg).
Fig. l3 represents the contribution of each of the VOS components to
the overall value in the case ofthe highest obtained value in the scope of
this study. The avoided environmental cost (Vs), avoided health liability
cost (Vq), and avoided transmission capacity cost (V6) represant 69% of
the total cost.
Ihe evolution of the cost percentage contribution of each VOS
throughout Figs. 1 1, Figure 12, and fig. 13 shows the level of uncer-
tainty of the VOS in respect to the corresponding component.
The lowest and highest LCOE VOS values obtained from the as-
sumptions made in this study are respectively 9.37i/kV,lh and 50,650,/
kWh. The existing VOS studies results fall into this interval. The sample
calculation made by Ref. [45] for Minnesota is 13.50,2kWh while [46]
calculated a VOS of lo.7CAWh for Austin Energy. These values are in
the lower spectrum of the result of this study because of the consider-
ations made. They incorporate less VOS components than the present
study, and this study focuses on sensitivity, therdore higher values of
parameters have been considered. Other results summarized by
Rd. I47l have found the VOS to be 33.7O,zkWh in Maine, between 25.6
and 3l.80,zkWh in New Jersey and Pennsylvania [48], and l9.40AWh
in Washington DC. ln general, the VOS is much higher than the net
meterin8 costs as even the hi8hest costs observed at the residential level
pay [50,62,1 I 2]. The residential net metering rates are also the highest
as compared to commercial and industrial rates so the laner two are
even more unjustly compensated for installing solar. Overall, this in-
dicates that utilities are under-compensating customers with
grid<onnected PV rystems if they are only payinS net metering rates, as
displayed in Table 2, Table 2 shows a comparison between VOS rates
and net meterinS rates in the U.S. states mentioned above, wherever
data is available. As only a tiny fraction of utilities (3%) are paying full
net meterint rates anyway [43], there is a need for regulators to ensure
that solar customers are being adequately compensated for the value of
solar electricity they are sharing with the grid [42]. Substantial future
work is needed to ensure that solar PV owners are not subsidizing
non-solar electricity customers.
5. Future work
This study has covered a vast number of existing VOS components,
but some components were not included in this study due to the lack of a
reliable evaluadon methodology. Itese components include the eco-
nomic development cost, the avoided fuel hedge cost, and the avoided
voltage regulation cosL These represent opportunities for future work
once the evaluation methodologies have been developed. Also, there are
some parameters sensitivities that would provide insights with multiple
utility data sets. These parameters include the analysis period, the
hourly solar heat rate and solar Pv fleet, and the lO-years load profile.
Future studies can focus on incorporating the sensitivities of these
Approximstely rt-soawh (1.2-r.60awh)
[113]
12.r6-l4.66cawh Ir l4]
13
f,"'S. Hatfu @tilt,M, P@Tc
AvoidcdO&M 6red
Cost,0.951/twh, 306
parameten into the modcl or can use the foundadon of this model to
build on new VOS studier according to a spedfic locadon and available
daa from utilides. Another limltation to this stndy is that it does not
include the effect ofthe load match factor, and loss saving factor.
As the results show theenvironmental and health coss can dwarf the
technical costs and thereby determtne the VOS. lhere are also second
Rawwdrbord$ffie Pna13'Rcn es137 QA2,) 110599
AvoidedO&M
vrhble co*,
O.22t14rWh52Vc
AvoidedO&IU
vuiable cost,
l.07lkt@ 4o/o
Avoided resenre
capacity cost,
0.79|kWhr 3o/o
order effects that can be used to obtaln a more accurate VOS values. For
examplg the negadve impact of polludon from convendond fossil fuel
electricity generation on crcp yielde t106I as well as PV producdon
could also be considered in future work to glve a more accurate Vs. ln
addidon, as greater percentages ofPV are applied to the grid the avoided
costs wil ciante and there is a need for a dynamicVOS akin to dynamic
Avoidcd dis0ihrtion-.
cqacity cost
o.mt/twh, (P/6 Avoidcd rcscrrrc
capacity cost,
0.00r/twh,0./6
f[. f 1. Contribution of eactr VOS cdnpdrcnt to the orrcrall VOS IrOE - Low Cost Sccnario,
Avoidedo&Ivl fixed
Cost,0.35llkwh, 4olo
AvoitLrd dislribuio
capacity cost,
l.7,tlkwh" 60/o
Fig. 12. c.imtribudor of eadr VOS componmt to the o\rerall VOS LCOE - Middle C6t Scenado
l4
K.S. Haybo @tit J.M, P@cc
AvoidcdO&It{ fued
Cost, l.54tAWh. 3016
carbon life<ycle andyses needed for real energy economlcs [1 16]. This
compledty will be further enhanced by the introducdon of PV and
storaSe systems [1 r 7] as it will deperd on size [1 I 8] and power flow
management and scheduling [1 I 9,1 20].
Perhaps the most urgort need for future work is acorrate esdmations
ofthe value ofavoided GHG liability costs because the magnitude ofthe
pot€ndal liability [107,108] could overwhelm other subcomponents of
the VOS. ltris is because as the realities of climate change have become
mor€ established, a method gaining traction to account for the negative
externalides is climate litigation [107,108,121-131]. For udlity VOS
analysis this is pardcularly complex as it is difficult to know where to
draw the box around environmental costs. As some studles have
concluded there is liability for past emissions as well as for harm done ln
other nations [122]. Liability for disastrous events is also challenging to
predlct [f26]. Combining both other nadons and disaster creates lia-
bility pot€ndal that could become enonnous with prioridzadon given to
victims that are losing their land, culture, and lives due to dimate
change [1271. Tort-based lawsuits are already posible from a legal
poilrt ofview [126], butthere are otherlegd methods that could be used
to reduce climate change such as public nuisance laws [28]. Some
authors have argrred a'polluters pay prindple' for carbon emissions
[129]. Other studles have concluded that emltters such as convendonal
fossil fuel power plant operators should be forced to buy long term in-
surance in order to cover their share ofclimate change costs for mlni-
midng risks in case of insolvmdes [r30]. Determining what such
insurance premiums should be is another area of substandd future
wod(. Determining what the greenhouse gas liabllity costs are for con-
vendonal electricity generators (as well as potendal avoided insurance
costs) that can be avoided with PV is extremely challenging. Ihese es-
dmates will become easier lvith time as climate change lmpact studies
become more granular thereby assigrfng specific costs to speclflc
amounts of emissions. In addidon, realidng these climate llabiltty costs
ln courtrooms will become more likely. As lGane points out it is dear
that as the negative impacts of climate change grow mone pronounced,
the fossil-fuel based electricity industry faces a future that wlU be less
accepdng of current practices and that will increase economic (and
Rav*dc o,,,il St/a,d/ttfrb Enag Rcyi,.ws 1 37 (mA ) 1 rc599
AvoidedO&M
variable cost,
l.gzl,fwh" 1r/.
maybe even industry exlstentid) risks t13fl. Avoiding these risks has
rcal value, which should be included ln the VOS in the future.
6. Conclucionf
This study deuronstrated a deailed method for valuing the incor-
poration ofsolar PV-generated electddty into the 8dd and analyzed the
sensidvity of each VOS component to its input parameters, and the
overall sensidvity of the VOS to the each of its components. Several
components have been found to be sensidve to the utility discount rate,
namely the avoided O&M fixed cost; avoided O&M variable cost; avoi-
ded generadon capadty cost, and the avoided distribudon capacity cost.
Except for the avoided distribudon 6pacity, the other components'
value decreases with the increase of the udlity discount rate. the dis-
tribudon capadty is more sensldve to the discount rate than the other
components. It increases with the discount rate and can be negadve lf
the discount rate is very low. Itis has shown the necessity of carefully
choosing the discount rate for VOS studies. Most of the VOS values do
not have a high variability to the solar PV degradation rate even though
its increase slightly rcduces the value of each component, and the
overall VOS. The environmental cost and the health Uability clst are
sensidve to the cost incr€ase rate that can be ded to the emissions impa.ct
ofthe conventional €nerg:f sources, Thes€ two costs are likely to increase
in the future with the worseoing of the emission of fossil fuel sources and
more information about lts €frects, which increases potential emissions
Itability for utilities. Finally, spedffc c,se firdies could provide addl-
donal sensidvldes on the few areas ofthe VOS that were not evduated in
this paper to create better VOS models. Overall the results ofthis study
indicate that grid-tied udlity customers are being grossly under-
compensatd in most of the U.S. as the value of solar eclipses the net
metering rate. The implicadons of this sensitivity analysis demand a
reevduadon of the comporsadon for U.S, PV pnosumers as the VOS is
much hlgher than net meterinS or any lesser compensadon schenres.
Substantial future work is needed for regulatory reform to ensure that
solar owners are not unJustly subsidizing U.S. electric utilides. In addi-
tion, future wort can obtain an even more accurate (and higher) value of
Avoided frrc1 cost,
3.851/k$/b, 8elo
Avoided rcserve
- -capacity cost
1.58r/kwt,3%
Avoidcd disnibutiou
caprcity cost,
3.50r/kwh, 7olo
Fit 13. Contributton of each VOS component to the overall VOS LCOE - High Cost ftenario.
15
K.S. Hayibo @ilJ.M, Pcoce
VOS by evaluating economic development cosB, the avoided fuel hedge
costs, the avoided volage regulation costs, secondary health and envi-
ronmental effects such as increased crop yields from PV-reduced
pollution, and accurate esdmations of the value of avoided GHG liabil-
Ity costs or avoided GHG emissions liability insuriance.
Declaration of competing interest
The authors dedare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper,
Acknowledtmmts
This research was supportd by the Richard Witte Endowment.
Referencee
[fl Yu CF, ven Serk WGJHM, Alsm EA Utrevdint the photovoltaic t€chnolos/
leaming curue by incorporation of input price ch8ry€s md scale effects, Renew
Sustain Enety Rev 2017i75t324-37. https://doi.or8,/10.1016,/j.
rser.201 0.09.00 1.
[2] Hong $ Chu"S Y, Woo C. Scmio andysis for etimeting dre leming rate of
photovoltaic powel generation based on learnlng curue theory in South Xorea.
Energ 2015;798G9. https:,/,/doi.org.,/10.1016,/j.enerty.20'14.1 0.050.
[3] Trappey ArC, Ttappey CV, Tan H, Liu PtlY, Li S.I, Lin L.C. The detmimts of
photovoltalc rystm co$G: an evaluation ucing a hierarchical leming cwe
model. J Clean Prod 2016;172:77@-76, hltps://doi.orgll 0.1 016,/j.
jclepro.20l 5.08.095.
[4] Moulein I. Photovoltaic lmint nte Btimation: issues and implicationr. Roew
Srstain Energ5r Rev ?-Ol6;65:il?-24. hftpy / / doi.otg/ 1O.1O'l 6/ j.
rser.20l 6.06.O70.
[5] Feldman D, Barbose G, Margolis R, Wiser R, Darghouth N, Goodrich A.
Photovoltaic (PV) pricint trends. Historical, recent, and neAr-term projectioN.
Golden, CO, USA: National Renewable EnerSy Laboratory; 201 2.
[6] Barbose Cl,, Darthouth NR, Millstein D, Lacommare KH, Disanti N. Widiss R.
Trackin8 the sun X: the installed price of residential and non-residential
photovoltaic systems in the United States. [,awrence Berkley National [,aboraiory;
2077.
[7] PvitriShts. Pvinsi8hts. 2020. http:/,/pvinsithrs.com,/. [Accessed 6 April 2020].
[8] I&o[ M, Otto M, Kisebler I Fiidrel I! Wehrspohn n, Ernst-Bemhard xley, et al.
Black silicon for rolar cell applicatioE Photonic for mlu energpr rystems IV, vol.
8438. Brussels, BelSium: htemadonal Soctety for Opti6 8nd Photonics 2012.
httpst / / doi.org/ 10.l 1 17 / 12.922380.
[9] Bmon A. Co6t reduction ln the solu industry. Mets Today 2015;lE:2-3. hnps:,//
doi.oryl 0.1 O1 6/j.manod.201 4.1 0.O22.
[10] Modancse C, Iain€ H, Pffinm T, Savin tI, Peme J. Economic advmtagc of dry.
etched black silicon in pesiveted mifts reu cell (PERC) photovoltaic
matrufactuint, Ens$s 2018;1 1 :2337, htrpst / / doi - or E/ 1 o.33 / erl 1 092337.
[11] Reut€ts. Solil c6ts to fall further, powering global demd - Irma. Reuters;
2017. h.tlps: / / ww w. reuters.com/article/singapore-energy-solar-idUS
L4N1lvrY2F8. [Accessed 7 AWil 2O2O7.
[12] Branker K, Pethak l[.rM, Pffic JM. A r€vi€w of sler photovoltalc levelized cmt
of elstricity. R€nff Sust ln EnerSy R€v 2011;15:4470-82. https,//doi.org/
10.1O16/j.rser.201 1.07.104.
[f3] Richard C. New wird and rolar cheaper thm existing coal ond 8as.2018. htr
p:,//w.windpowemonthly.com,/ilticle/I49'l 146?utm_source.. website& utm
_medium =social. [Acressed 7 AWil20201.
ll4l Tflo q Atbim K, DeAngelis F, Hemandez J, tom lllt Oduftomaiya A An
conomic analyeis of residentlal photovoltaic systm widr lithium ion battery
stcage in th€ United Stat€s, Roew S$taitr Enugy Rev 2018;9:1057-66.
https:,/ /doi. org,/1 0.1 0 I 6,zj.rser.201 8.06.055.
[15] Uu H, Azutalam D, Chapman AC, Vsbit c. Technmonomic feasibility
assessment of grid{dectiorl lnt J Electr Power Energy Syst 2019;109:40}12.
httpst/ /doi.ort/10.1 01 6/j.ijepes.201 9.01 .045.
lf6l Schill W-P, Zerahn A Kw F. Solar prosumage: an eonomic disusion of
challengec and opportunides. L"* lowitrsch J, editor. Energy transition: financintcoMer CGomership in rmewablm. Chm: Springer Intemational Publishing;
2019. p. 70!31, https:,/,/doi.or8,/10.1OO7 /978.3-379-93518-8_29.
[17] von App€n J, &au M. Strategic decisim making of distibudon network
operatoE md invstoE in residentisl photovoltaic battery stonge systems. Appl
Eneryy 20lE;230:540-50. https:,/,/doi.or8,/10.1016,/j.apenerty.20l8.08.o43.
[18] Marczinkowckl HM, Oster$trd PA. Residential versus comunal combimtion of
photovoltaic md batteqr in smart msty systetr. Enerty 2018;152:466-75.
httpsr,//doi.ortll 0. 1016/j.enerty.2018.03. I 53.
[19] fd CS, Mccullch MD. Leveliz€d st of el*tricity for solar photovoltaic md
elecrical energpr storage. Appl Energ5r 2017;190:19l-203. hltps://doi.or8..
I 0.1016/j.apenerty.201 6. I 2. I 53.
[20] Ksng MH, Rohatti A. Qumtitative melysis of the levelized c6t of elstricity ofcmercial scale photovoltalcs system in the US. Sol Ensty Mats Sol Cell
20l6il94t7l-7. hnps:,//doi.or8,'10. 101 6,/j.solmal 2016.04.046.
Rswob/l dt l SutuinfrL Ensgr R€viws 1 37 (mA) 1 rc599
[21] International Renewable EnerSy ASency. Renewable power teneration costs in
20'17. 2018. Abu Dhabi, UAF-
[22] Dudley D. Renewable €lrergy wlU B€ consistently cheaper than foestl fuels by
2020. 2018. Report Claim. Forbec, https:/,/ww.forbes.com,/sites,/dominicdudle
y/2018/Ol/13/rcnewable-energy-cost€ffective-fossil-fuels-2020/. [Accessed 7
April 20201.
[23] B8nsje B, Islam sM. Rdiability basd optimum location of disrributed
tenmtioL Inr J El8tr Pows Ener8y Syst 2011;33:1470+, https://doi.ort/
1 0. 1 01 6/j.ijepes.20l 1.06.029.
[24] Liu I, Bao H, Liu H. Siting md sizing of disributed gmuation based on the
minimum t?nrmi0sion losr€s cosl IEEE Power Ent Automat Conf 20ll;3:22-5.
https:/,/doi.or8,/10. 1 109P8AM.201 1.6135006. Wuhrn, ehlne: 2O11.
[25] Sead NMd, Sujod MZ, Hul MlnS l" Abas MR Jadln MS, Ishak I&, ct d. Impacts of
photovoltaic dlstibuted goeration lcatim and slze m distribution powe
rystem network UPEDS 2018;9:905. https:/,/doi.or8l10.11591/ijpeds.v9.i2.
pp905-913.
[26] Barter PP, D€ Mello Rw. D€t€rmininS the impact of dirtributed teneration on
pows systms. I. Redid distrtbudon systems. Power englnuing soci€ty ffis
meetint (cat No.0OCH37l34), vol.3. Sesttle, lllA USA rEnn; 20OO. p. 1645-56.
https:,/,/doi.ort,/l 0. I 1 09lP8SS.20O0.868775. 2OOO.
[27] Brown RE, Wllk IlL. Th€ €cono[dcs ofaging infrastructure. IEEE Powq Enerty
Mag 2006;{:36-43. https:,//doi.or8l10.1 I og/MPAE.2006.1632452.
[2el U l Guo J. Widom ebout age [aging deticity infrastructue]. IEEE Pows
Enerry Mag 2006;4:,14-51. hnps:,//doi.or8,/10.1 109,4trPAE.2006.1632453.
[29] Willis HL, Schrieber RR. Ating power delivery infrastructures. second ed. Bma
Raton: CRC Press/Taylor & Francis; 2O17,
[30] Pudasaine D, Ktm J-H, Seo Y{. Mscuy emission trend influenced by stringmt
air pollutants regulation for coal-fued power plsnts in Korea. Ah6 Envirm
2009l43262il4. https:,//doi.or&/1 0. I 0 1 6,/j.atmosenv.20O9.06.O07.
[3U Celebi M, creves F, Rusell C. Potential coal plant retirements: 2012 update.
Battle Group 2012:13,
[32] Rallo M, fopez-Anton MA, Contr€res ML, Meroto-Valer MM. M€rcury policy end
regulations for coal-fired power plmts. Envlron Scl Pollut R6 2012;19:108/+-96.
https://doi.ortl10. 1OO7ls1 1 356-01 1-0658-2.
[33] G€rrard MB, Welton S. US fed€rd climte dung€ l8w in obama's second tem.Tlmd Envirm I8w 2014;3:111-25. https:,/,/doi.or&/10.1017/
s20471 02514000016.
[34] De Cian E, Sferra 4 Tavoni M, The lnlluence ofeconomic growth, population, aad
fossll fu€l scrrdty on energy lnvestmcnt. Oimadc GhaDge 2016;136:39-55.
https:/,/doi.or8,/l 0. 1007/s1 0584-01 3-O902-5.
[35] Iciegler E, Mouratiadou I, Luderer G, Bauer N, Calvin K, Decian E, et al. RoSE:
roadmaps towards sustainable energy futures and climate protection: a synthesis
of results from the rose project. The RoSE Projecl of the Potsdam lnstitute for
Climate Impact Research; 2013.
[36] Murphy D.r. The implicatione of the declining eneryp, retum on investment of oil
production. Phil TraN Math Phy6 Eng fti 2014;372. https:/,/doi.orgl10.1098,/
sta.20l 3.0 I 26. 20130126.
[37] Wilson E, Friedmam SI, Pollat MF. Research for deployment incorpomtint
rislq replation, and liability for carbon capture and sequestrauon. Envirotr Sci
Tecbnol 20O7;41:5945-52. https:,//doi.orI' t I O.1O2l / eso6227 2t,
[38] Burtraw D, Palmer K PauI A Beasley B, Woeman M. Rdiebility in the U.S.
elBtricity industry undu new mvircmotal rcguladm. Eneryy Pol 2013;62:
107H1. https://doi.orgl1O. I 016/j.eDpol.20l3.06.07O.
[39] Itatson LF, Haers D, PatiioEchevri D. Fuel prics, emision standuds, md
tssation cts for coal rc natural tas power plants. Envimn Sci Technol 2013;
47 14926-33. https:,/,/doi.org,/1 0. 1 021,/es4001 642.
[40] Lim J, Mastraryelo E, Burtraw D. Re8ulating 8le€nhouse gases fron coal powe
plants under the dean sir acL J Arsoc Environ R€sour Ecm 2014i7t97-134.
https:,/,/doi.ortl1 0. I 086/6760Ii8.
[41] Burtraw D, Linn J, Palmer K, Paul A. The costs and colsequences of clean air acl
regulation of CO, from power plants. Am Econ Rev 2014;704:55742.
[42] kehoda E, Peetce JM, S&ely C. Policis to ovscom€ barrlem for renewable
magy distibuted gmmtion: a ese study of udlity stuctm md reSulatory
rcdm6 in Michitm. EnerSies 2019;12i674. hnps: / / doi.o.B/1O.339O/
enl2040674.
t43l Schely C, Iouie EP, Pearce JM. F.xamining interconnection ud net metcing
policy for disributed generation in the United Stetes. Renewable Ener8y Focus
2o77i?.2-23tlH. htlps:,/,/doi.or8,/1 0. 1076/ j.ref .2017.O9.OO2.
t44l NREI- Vdue+f-Solar tariffs I state, lml md ribal tovmenh. NREL;2019.
https:/,/ww.uel.to!,/state-local-tribal/basics-valueof-solar-tariffs.hrml.
tAccessed 9 April 20201.
[45] Noris BL, Putnam MC, Hoff TE. Minnesota value of solar: methodology Clean
Power Research; 201 4.
[46] Clean Power Research Value of solar at Austin enerSy. Austin, 'tX, USA: Clean
Power Research; 2Ol 4. 201 3.
[47] Hdm 4 Cook JJ, Aal8r Al Coughlin JW, Mow B. Disributed soler photoyoltaic
ctst-bercfit framework study: consideratioro md reoms for Oklshoma. 2019.
https:,/,/doi.or8,/ I 0. 277 2 / 1 561 51 2,
[48] Perez R, Noris BL, Hoff TE. Thc value of distributed solar electric generation to
New Jersey and Pennsylvania. Clean Power Research;2012.
[49] Brom fr Buym J. Valmtion of distributed mlar a qualitative view. Elcts J
2014i27t2748. https:,//doi.org,/l 0. l0l6/j.rej.2014. I 1.0O5.
[5O] Ribago KR, Libby L, Harvey T, EnerSy A, Norris BL, Hoff TL,, et al. Desitnint
austin ENERGY'S solar tariff using a distributed PV value calcrlator. 2012.
Austin. fi, USA.
16
K.S. Hayho @il.l.M. P@c.
[5U Farrell J. Mimesota's value of solar, Mimesota, USA: lnstilule for Local Self-
Reliance; 2014.
[52] PoullikkasrLAoompara$veaf,sessnentofnetmeteritrtmdfeedlnterlfir]remes
for residmtial PV systems. Surtsin Ensgy Tedmol Ass€ss 2013;3:1-8. hnps://
doi.ortl10. 1016/j.sera.2013.04.001.
[53] Taylor M, Mclsen J, Cory K, Devldovich T, Sterling J, Makhyoun M, Value of
solar. ProSram d6ign and implemmtation con lderationt. Goldeq CO (Unlted
Stotes): USA: National Renewable Energy Lab. (MEL); 20f5. https://doi.or&/
I 0.217 2 / l2l5OO5, GolderU CO.
[54] Munoz FD, Mills AD. Endogeoous assessnot of the elncity value of rclar P\/ in
teneration invEtmot planning sndies. IEEE Tlans Sustain En€rty 2015i6:
157{-85. hnps:/,/doi.ortl10.1 1og./tste.20l 5.2456019.
l55l Gani D, Siorhrtlsi & Dglholtrl P. Drtr drllenges in e3timting the capecity value
of solar pilrotovoltaiG. IEEE J Photovolteics 2017i7:106L73. hftps:,/,/doi.or8l
1 0. 1 1 o9/JPHOTOV.2017.2695328.
[56] Stanton T. Review of state net enerty meterint and successor rate desitN.
National Retulatory Research Imtitute; 201 9.
[57] Keyes JB, R6bago KR. A REGULATOR'S guidebook: calculating the benefits and
costs of distributed solar generation. lnterstate Renewable Energy Council, Inc.;
2013,
[58] Denholrn P, Margolis R, Pdmintis B, Berows C, Ibanez q Bird L, et 61. Methods
for analyzing the benefit and costs of dlstributed photovoltaic g€nereton to tie
U.s. Electric udlity system- Goldm, CO, USA: Nationd Renewable Energy
laborator,i 2014, https://doi.org/ 10.2772/ 1 159357.
[59] Blackbum c, Mage C, Rd V. Soler veluation ard the modem udllty's erp6nslon
into disEibuted toffatiolL Electr J 20l4i27tlH,2. https://doi.or8l10.1016/j.
tej.2013.12.002.
[60] Pitt D, MidDud G. Ass€.ring the vdue of dirributed mlar enerjpr generetiotr
Cun Sustein Raewabl€ Eneryy R€p 2015;2:105-13. https:/ /doi.orE/1O.1OO7 /
s4O5l8-01 5-OO3O-0.
[61] Ilarari S, Kauftnan N. Assessint the value of disrributed solar. Yale Center for
Business md the Enviroment; 2017. p. 21.
[62] Orrell AC, Homer JS, Tary Y. Di6tributed Sosation valuadon and compensadon.
2018. https:,//doi.or Y 10.2172/ 156127 3.
[63] Proudlove A, Lips B, Sarkisian D. The 50 states of solar: 2019 policy review Q4
2019 quarterly report. NC CLEAN ENERGY TECIINOLOGY CENTER; 2020.
[ff] Bfown P& O'sullivm FM. Spadal and tmporal varladon in the vdue of solar
power acrocs United Stat€r el€ctrlcity marteB. Renew Sutain Enerty Rev 202q
121:10959{. https://doi.org,/10.1016./j.rser.2019,109594.
[65] Siler-Evans K Azevedo tr" Morgm MG. Apt J. Regiond variationc in the healtb,
enviromantal, md dlmrte benefts of wind tnd solar geoeratim. Proc Nad Acad
Sci USA 2O13;110:11768-.73. hrQs:,/./doi.or8,/10.1O7 3 /pnas.122197 81 10.
[66] Milbtein D, Wiser & Bolints M, Barbo6e G. The dimat€ md elr{uality bene6B
of wind and solar power in lhc United Stat$. Nat Ens8y 2OU;2t17134. htps:/ /
doi.org,u10. 1038/nener8y.20l 7. 134.
[67] wiser & Millstein D, M8i T, Macknick J, csrpents Jl cohen S, et al. The
envircmental and public health baneftf of adrieving hith p€neErdoB of slrr
energy in the United Strt€s. Enerlt 2016;113:472{6. https:/ /doi.o19/10.1016/|
ener8y.2016.07.068.
[68] Mtillendorff C von, Welsch tL Mearuring renewable oegy extmalldes:
evidmce frm sbj€cdve well-belnt date. Irnd Econ 2017i93i7@-26. htlps:/./
doi.orgl10.3368/1e.93. 1. I 09.
[69] Abel D, Holloway ! llartey M, Rnrshsj A Brintmm G, Duan P, et el. Potmtial
air quality benefits from incrased rolar photoyoltaic elcEicity goeration in the
Esstem Unlted Stetes. Atnoo Environ 2018;175:.6*74. https:/,/doi.org,/10.1016,/
j.atm6env.2017.l 1.049.
[70] Prehoda EW, Pearce JM. Potcntial lives mved by replaciry coal with solar
photovolteic electricity production in the US. Renew Sustain En€r8X/ Rev 201280:
710-5. https://doi.or&/1 0.1 0l 6/j.rser.2017.M.O94.
[71] Borenstein S. The market value and cost of solar photovoltaic electricity
production. Escholarhip 20O8;39.
f72l U.S. ElA. Annual energy outlook 2015. U.S. Ener8y lnformation Administration;
201 5.
[73] CalifomiaEnergyComDlssiotl" Heatrates. Heatrates.2020. htlps:,//ww2.ener8y.
ca.gov,/almanac/electricity-data,/web-qfer,/Heat-Rates-cms.php. [Acceeeed 6
Msrch 20201.
[74] Deaver P, Rhyne I, Bender S, OSlesby RP. EstimatinS burner tip prices, uses, and
poteDtial issues. California EnerSy Commission; 2013.
t75l Baker E Fowlle M, Irmoine D, Reynolds SS. Ite economics ofeolar elecricity.
Ann Rev Rsour Econ 2013151387-,1,26, htrps:,/,/doi.or&/10.1 I 46lamuev-
resource-091 91 2-1 5 I 843.
[76] tlacerola I, Libeman I. ComparinS a value of solar (VoS) Tariff to nel meterin8.
Master. Du-ke University; 201 5,
[77] reportFERc FoRM No. 1: ,nnual report ofm.Jor decEic utilities, licorsees and
otheE md supplemstal 2020.
[78] Gotham DJ, Lu L, Wu F, Nderitu DG, Phillips 1A, Preckel PV, et al. MISO enerty
and peak demand forecastint for system planninS. 2018,
[79] Butts G. Escalation: how much is enouSh?. 2OO7. p. 41. Cocoa, lL, United States.
[80] SivarsEm D, Moore MR. Economic performance of grid{omected photovottaics
ln Cellfmia and Texas (United Stal6): the inllrence of renewable erreqy and
dimate policie. Enagy Pol 2012;4*27447. https: /,/doi.org,/1 0. 1 01 6,/j.
enpol.201 2.06.01 9.
[8L] Technical Support Docment. I'echnical update of the social cost of carbon for
regulatory impact analysis under execulive order 12866. 2016.
Ra,ar/obb @til Stttuinfrl' E og R.l,ia,,s 137 (2021) 110599
U.S. Bl.s. Consumer price iDdq U.S. City average all urban consumers (CPI-U): all
itms, 19E2-84 2020, https:,/,/w.bls.8ov/reSions/midwest/data/consume
rpriceindexhistorical-us-table.pdf. [Accessed 4 April 20201.
Sorrels JL, Walton TG. Chapt€r 2 - cost estimation: concepts and methodoloty. U.
S. Environmental Protection ASency: 2017.
Yim SHL, Barett SRH, Public herltt impacb of cortrbusdon emissione in the
United Xingdom. Bnviron Sci Technol 2012t46t42914, hrtpy/ /doi.orillo.l02t /
es2040416.
Caiazzo F, Ashok A" WaiE IA, Yim SHL, Bsrrett SRH. Air pollution and wly
deeths in lhe Unlted Stat6. Part t qmtifying the impact of -ajor stoff in
2005. Atmoe Environ 2013;79:l9E-208. https:,/,/doi.or8l10. 1016,/j
atmosenv.20l 3.05.08 l.
Housrl lC, Bonett SRE Air pollution and early deaths in th€ United Stet6. Part
tr: attribution of PM2.5 crposure to anisrions spcdes, drne, location rnd sector.
Atnos Envlron 2014;9*61r7. hqy/ /d,oi.otE/1O.7O76/j.
atmoseDv.201 4. I 0.033.
Jffiett M, Bumett RT, Pope GA Ito Iq Thurston G, Krewski D, et 8L lrry-term
ozone expoei[e and mort llty. N Ertl J Med 20O9;360:1085.45. https:/,/doi.orgl
I 0. 1 O56/NE"rMoa0803894.
Muler NA Mendelsohn & Nordhaus W. Environmmtal actounting for pollution
in the United States economy. Am Eaon Rev 2011;101:164$75.https://doi.otg/
1 0.1 257,/aer.101.5. I 649.
tlidden costs of energy : rnpriced coroequences of energy production and ue.
Washington, D.C: National Academies Press; 2010.
Rabl l', Spadaro W, Public health lmpact of alr lnlludor and lnplications for the
en€ryy ryrteE Artru Rev Eoergy Envirci2AOq25t6Ol-27. https://doi.ar&/
10. 1 1 46,/annuev.enerty.25. 1.601.
Mechol 4 Rir.} S. Economic valrc of U.S. fofsil fuel electricity health impact .
Environ Int 2013;52:7H0, https:,/,/doi.or8l10.1016,/j.envint.2012.03.0O3.
Carlame CPUS, Cllmte change law: a decade of flu md m uncstain futue.
SSRN J 2019. https:,/,/doi.ort/1 0.213g,/ssm.34938 1 2.
Jordan DC, Kurtz SR. Reliability and geographic trends of 50,000 photovoltaic
systems in the USA: preprint. Amsterdam, Netherlands: National Renewable
Energy Laboratory; 2014. p. 10,
Phtnikarides .1" Kindyni N, Makrids G, Geo4hiou GE. Revlew of photovolteic
degradauon rate m€tfiodolo8i€s. Renew Sustain Energy Rev 2014;40:14&52.
https:,//doi.or8,/l O.1 0l 6/j.rser.201 4.07. I 55,
U.S. EIA. Capital cost estimates for utility scale electriciry teneratint plants. U.S.
Enerty Information Administration: 2016.
The het mte of power Senerators. SciencinS; 2017. https:,/,/sciencint.com/heat-
rale-power-S,eoerators-7958684.htmI. [Acccssed 30 Mrrdl 2020].
U.S. EIA" SAS outpuL T.ble 82 avmte t€sted heet rat6 by prlme mover aad
enetgy source, 2008 - 201t (Dtu per kllowatthour), 2020. https:/,/m.eia.tov,/
elecriciry/annual/html,/epa-08_O1.html. [Acctssed .l March 2020].
U.S. EIA" Electricity Senerator cost data from surrey form EIA{60. Construction
Co6t Data for Electric Gosatffi Installed in 2Ol7 2O2O, httpst//www.eia.Eov/
electricity/generatorcosts/. [Accsed 30 Mech 2020].
Nw Council. Seventh northwest conseryation and electric power plm. NW
Comcil; 2020.
U.S. EIA. Roerve electric gmaating capacity helps keep the lights on - today in
Ensgy - U.S. Energy Information Administradon (ElA), 2012. https:,//ww.eia.
tov,/todayinenerty/detail.php?id:651O. [Accessed 30 Msrdt 2020].
Tmnsmission cost manrgement I comBcial. AEP Enerly;2018. https://m.
aepenerSy.com/201 8 / 03 / 08 / f ebruary-201 8-edition,/. [Accwd 3O March 2020].
Mian MA. Project economics and decision analysis. second ed., vol. l Tulsa:
Pennwell Corporation; 201 1.
STATE OF MINNFSOTA PUBLIC UTILITIES COMMISSION. Notice Ol updated
environmental exlemality values. 2017.
Akored€ MF, Hi?aD H, Pouesmaeil E Disttbutcd ensty rcsouc€s end benents
to the enviroment. Renew Surtain Eneryy Rev 2010;14:72+'3+. https:/ /doi.org/
1O.l 0l 6,/j.rser.2009. I 0.025.
Petrce JM. Towards quanttfiable metrics wanandnS industry-wide corporate
death penaldes. Soc Sci 2019;8(2):62. https:,/,/doi.org,/10.3390,/socsci8020062.
Bmey J.{. The downrtresm air pollufion impacts of the traruition from coal to
natual tas in the United States, Nat Surtain 202q3:152.60. hrrps:,/,/doi.ort,/
I 0.1 038/s4 I 89ll-01 9-0453-5.
Allen M. Liability for climate chan8e. Na(ure 2003;421(6926).891-2.
Heidri N, Pearce JM. A review of grmhoue gas emisslon liabilides as the value
of renewable sreryy fo mldgarint lawsults for dimate change rdeted damages.
Renew SEtain Eneryy Rev 2016;55:899-908. https:/,/doi.ort,/10.1 0l 6/j.
rser.20l S.1 1.025-
Sebalj D, Mesaric J, Dujak D. Predictint natual gas consumption - a literature
review. Central European conference on infomation and intellitent systems;
varazdin, varazdin, Croatia. Varazdin: Faculty of OrSanization and Infomatics
Varazdin; 2017. p. 293-3OO.
U.S. ElA" Factorc afr*tin8 mtual ts prics - U.S. Ensgy Informatiotr
Admlnlstradon (ElA). Naturd Gs Explained; 2020. https:/,/www.eia.gov
/energyexplained/natural-gas/factors-affecting-natual-gas-prices.php. [Acccsed
4 Aprtl 20201.
Cohen MA, Kauzmann PA, Callaway DS. Economic eltects of distributed PV
generalion on Califomia s distribution system26. Enerty lnstitute al Haas: 201 5.
Brcm DP, Sappington DEItl. DeriSniry comp€nfation for distributed mlar
8en€radonls net metednt evs oprinrl? ei 2Ol7;38. hltps: / /doi.ot8/10.554.7
/01 956574.38.3.dbro.
NREL. T€xss I solu reeearch I NREL NREL solar Rcurch; 2020. hrrps:,//M.
Nel.Eov /sol^t /tps/n.html. [Accrss€d 23 July 2020].
t821
t83l
t84l
tssI
t86l
l87l
t88l
TE9]
teol
lerl
l92l
[e31
te4l
I951
l96I
leTl
tesl
teel
u00l
t10u
t1021
trml
[104]
tlosl
TT06I
tl07I
t1o81
t1091
[110]
luu
tl121
t113I
77
K.S, Hry.il|o @titJ.M. PNcc
trr4I
tlrsl
[r16]
utn
tlrsI
trr9I
[120]
Ir2lI
Maim Public Utilidec CommirstoL MPUC: n€t en€r$/ blllltrg.2020. https:
,/,/m.maine.8ovlmpuc,/eltrtrici tylrenewabls/neb/index.shtnl. [Amed 23
July 20201,
NREL, Pennrylvanla I midmartet colar poticies ltr thc Unttcd Stat€8 | soler
research I NREL NRIL Solar Researc! 2020. eccesse4 hnp6://m.trel.tovlso
larlrps,/pa.hhl. [Acccs6€d 23 July 2020].
Keony & law C, P€arce JI{. Towarde real emrgy econmtcs srcrty policy driven
by life<yde carbon emiesior" EnerSy Pol 201q38(4):19698. htlps:/ /doi.ott/
r0.1016/j.enpol.2oo9.1 1.078. AF 1.
Rlfruneau Y, Bedta S, Barru€l R Ploh S. Optirmtl power f,ow memtmcot for
trtd coDnected PtI ryrteDs wtti blttcrier. rnPn Trr$ Sustlin Ene{y 20U;2(3):
309-20, httpe:,2/doi.orgl10. I 109/TSTE.201 1.21 14901.
Ru Y, Klcird J, Mardn€z S. SfGete size detcrDtnadon fc grld-coonected
photovoltrlc cystcme. rFFn Tranr Susain Encr8t 2014{1}6H1. hnp6://doi.
or8,/10.1109/TST8.2012.2199339. Jut 12.
Lu B, Shahldehpour M. Short-tcrm .chedulry of bettery ln I gid-conn€cted I{r/
batt€ry syrtdr. IEEE Tlans Pow6 SyEt 2q)5;2q2):1053-61. http$//doi.orgl
I 0. 1 I o9,/TPWRS.2005.845060.
Mulder G, De Ridder E Six D. Eectrlcity storage for grld<omcted horsdrold
dwelliqs widr PV panels. SoI Enerly 2010 Jul 1;84(7):1284-93. https:,/,/doi.org,/
I 0. 101 6,/j.solener.20l 0.04.0O5.
Prston BJ. The influence of climate chante lititation on govements and the
private strtor. Clim law 20ll Jan l;2(4):485-513.
Rauo&/c od Snffirfrlc Fdagr Rctuns 1 37 (2021 ) 1 10599
Farbs D. Bedc cmpeoradon for the vlc&trf of dtnate ch.trte, Univcrslty of
Calfcnte, B.stdcy Public Lw Releardl Papcr No. 991357. 2006.
Farber D. The case for climate compensation justice for climte chmge victim in
a complu world. Utah tsw Review; 2008.
Hmcck E. Red dawn, blue thmder, puple rain: corporate risk of liability for
global dimate chmge md the SEC disdosue dilemma. Georgetown Environ Law
Rev Winter 2005;17:233-51. 2005,
Healy K, Tapick J. Clmate chante: it's not just a policy issue for corporate
comel - it's a legd problem. Colmbia J Environ taw Erviron L 2004;89:1-23.
Grosmao D. weming UP to e not-slradical idq: tort-besed climate chente
litigation. Colmbia J Environ Law J Environ L. 2003;1.
l(ilinsky J. lntemtional climate chmge liability: a myth or a reelity. J. Transnat'l
L. & Pol'y 2008;18:377.
Farber DA. Tort law in the era of dimate chante, katrina, and 9/11: elelorint
liability for exraordinary risks43. Val. UL Rev.; 2008. p. 1075.
Reimund S. Liability for climate chmge: the benefits, the costs, md the
Easaction costs. Respon*s Glob Wm: Law Econ Sci Clim Chmte 2007;155(6):
1947-52.
Farber DA. ApportioninS climate chante c6ts. UCl"{ J tnviron Pol'y 2008;26:21.
firane J. CliDate clunge and fosdl fud: an €xarrlnadort of rlske for the arrgy
indutry md produccr ststls. Mns Eoctty Sustsh 20f7;/+. htlpst//doi.otg/
10.1557/me.2O17.3.
17?21
t123I
tt24l
Ir25]
[126]
llxtl
Ir28I
tr29I
lr30l
Ir31]
18
CERTIFICATE OF SERVICE
I hereby certiff that on this I 3th day of October 2021 ,l delivered true and correct copies
of the foregoing INITIAL COMMENTS to the following persons via the method of service
noted:
Electronic mail only (See Order 34781):
Idaho Public Utilities Commission
Jan Noriyuki, Secretary
secretary@puc. idaho. gov
_JslBenjamin J Otto_
Idaho Conservation League
courtney(rDc leanenerqyoDDorun ity.com
Erick Shaner
Deputy Attorney General
Erick. shaner@puc. idaho.gov
Idaho Power
Lisa D. Nordstrom
Connie Aschenbrenner
[e-mail]
[e-mail]
[e-mail]
Industriol Customers of Idaho Power
Peter J. Richardson
Richardson Adams, PLLC
[e-mail]
Dr. Don Reading
[e-mail]
IdaHydro
C. Tom Arkoosh
Arkoosh Law Office
[e-mail]
Erin.ceci I @arkoosh.com
Idaho Clean Energt Association
Kevin King
sta ff@ i dahoc I eane ner gy. or g
Clean Energt Opportunityfor ldaho
Michael Heckler
Courtney White
[e-mail]
tPC-E-21-21
ICL INITIAL COMMENT
Kelsey Jae
Law for Conscious Leadership
kelsy@kels),jae.com
Idaho Sloar Owners Network
Joshua Hill
Terri Ottens
[e-mail]
[e-mail]
Micron Technologt Inc
Jim Swier
Micron
[e-mail]
Austin Rueschoff
Thorvald A. Nelson
Austin W. Jensen
Holland & Hart, LLP
darue sch ho ff@ho I I an dh a rt. c om
[e-mail]
awj ensen @ho I I andlrart. com
ac lee@hol landhart.com
elsareanoamari @hol landhart.conr
City of Boise
Ed Jewell
Deputy City Attorney
ej ewell @c ityofbo i se.ore
boisecityattorney@c ityofboise.org
Kiki Leslie A. Tidwell, pro se
[e-mail]
Idaho lrrigation Pumpers Association
Eric L. Olsen
19
October 13,2021
Echo Hawk & Olsen, PLLC
[e-mail]
Richard E. Kuckhorn, pro se
Wesley A. Kucklnrn, pro se
[e-mail]
[e-mail]
ABC Power Inc
Ryan Bushland
[e-mail]
Comet Energt, LLC
George Stanton
Georqe.stanton@cometenerg.v. com
IdaHome Solar, LLC
Tyler Grange
tyler@idahomeso lar.com
lPc-E-21-21
ICL INITIAL COMMENT
20
October 13,2021