This paper compares the work by Mark Jacobson et al. of 100% renewables to the rigors of long-run utility system planning. This comparison to integrated resource planning (IRP) allows comparison between assumptions used by Jacobson to results from real-world planning studies. Seven criteria are proposed for designing such a study.

1. Contents lists available at ScienceDirect
The Electricity Journal
journal homepage: www.elsevier.com/locate/tej
100% renewables study has limited relevance for carbon policy
Robert J. Procter
Independent Energy Economist, Portland, OR 97206, United States
A R T I C L E I N F O
Keywords:
100% renewables
Carbon
Greenhouse gasses
WWS-only
Deep de-carbonization
100 by ’50
A B S T R A C T
This paper compares the work by Mark Jacobson et al. of 100% renewables to the rigors of long-run utility
system planning. This comparison to integrated resource planning (IRP) allows comparison between assump-
tions used by Jacobson to results from real-world planning studies. Seven criteria are proposed for designing
such a study.
1. Introduction
According to the Sierra Club, 36 cities are working towards 100%
renewable energy; 25 more have committed themselves to that goal,
and five have reached it.1
In addition, one county and one state have
made that same commitment.2
Sen. Jeff Merkley of Oregon is one co-
sponsoring a bill titled “100 by ’50 Act.3
; This movement also goes by
the slogan “100 by 50.” Part of the momentum arises from research,
including The Solutions Project, that argues for the wholesale trans-
formation of the entire energy system (not just electric utilities) of the
United States and 138 other countries to one that relies only on a
limited set of 100% renewables, referred to as wind, water, and sunlight
(WWS-only), by 2050.
Heard et al. (Heard) paints a challenging picture for deep reductions
in carbon emissions. Calling on multiple sources, he notes that more
than 1.2 billion people lack access to electricity. That, in addition to
population growth of around 2 billion more people by century’s end,
primarily outside Organization for Economic Co-operation and
Development (OECD) countries, where projected gains in electrification
is more than triple that of OECD countries, will make accomplishing
deep cuts in carbon emissions challenging.
We’ve been allowed to ignore how electricity is produced and de-
livered. Our indifference to the complexity of the electric power system
is a barrier to informed public policy on greenhouse gas (GHG) related
emissions, especially CO2. The importance of reliable power has risen
with the increase in digital communication. As a result, studies that
argue that it’s prudent to rapidly shift wholly out of fossil fuels and into
WWS-only for all our electricity needs must be held to a very high
standard.
This paper compares a set of implicit and explicit assumptions in
research by Jacobson et al. (Ref4)4
and in [20] to system planning work
by Portland General Electric (PGE) and the Northwest Power Planning
Council (NWPPC). These are but two examples of how utilities and the
NWPPC do utility expansion planning using integrated resource plan-
ning (IRP).
Data on existing generation capacity and projected additions will
also help broaden the real-world context for the policy recommenda-
tions in Ref4. Sections 2 and 3 together will examine Ref4 to utility
planning in Oregon and the Pacific Northwest (PNW). In particular,
important implicit and explicit assumptions required by Ref4’s con-
clusions are compared to results from PGE’s 2016 IRP and to the most
recent power plan from the NWPPC. These planning analyses are both
long-term and representative of current practice for both utility-level
and regional electric power planning efforts. Augmenting those studies
are selected research papers. The selected research papers are a subset
of the rather large body of work that currently exists that address as-
sumptions and results in Ref4. Finally, Section 4 presents a set of con-
clusions and recommendations.
2. An overview of Ref4
Ref4 argues that it is both technically and economically feasible to
https://doi.org/10.1016/j.tej.2017.11.010
E-mail address: proctereconomics@gmail.com.
1
Is Your City #ReadyFor100? See: http://www.sierraclub.org/ready-for-100/cities-ready-for-100.
2
Ibid.
3
Transitioning America to Clean 100% Renewable Energy for All by 2050. See: https://www.merkaey.senate/100by50.
4
Mark Z. Jacobson et al. 100% Clean and Renewable Wind, Water, and Sunlight (WWS) All- Sector Energy Roadmaps for 139 Countries of the World, June 2017, See: http://web.
stanford.edu/group/efmh/jacobson/Articles/I/USStatesWWS.pdf.
The Electricity Journal 31 (2018) 67–77
Available online 12 April 2018
1040-6190/ © 2018 Elsevier Inc. All rights reserved.
T

2. switch to a WWS-only portfolio in every application where fossil fuels
are currently used. To be clear, the argument is that every current use of
fossil fuels would be replaced by either electricity or hydrogen in every
sector of the economy of 139 countries simultaneously. Industrial
processes would no longer use fossil fuels. Products would no longer
contains fossil fuels. All transportation, including trains, planes, and
ships, would no longer use fossil fuels. Natural gas would no longer be
used for heating or cooking. Same with propane and all other fuel oils.
In addition, we would stop burning wood and other biomass to generate
electricity and/or heat.
Conceptually Ref4’s analysis presumes there is one utility spanning
the lower 48 states with new high-voltage direct-current transmission
that forms a super-grid across which electricity flows from generators
anywhere to loads everywhere. To accomplish coast-to-coast co-
ordination of the power system, Ref4 implicitly assumes the existing
institutional framework (organizations, balancing authorities, state-
level utility requirements, state statutes, and administrative proce-
dures) have been costlessly transformed to allow one entity to manage
operations that spans the continental U.S. As a result, he essentially
presumes that the contiguous 48 states comprise one coast-to-coast
utility balancing authority (BA).5
His estimates of the levelized cost of energy (LCOE) of WWS-only
and business-as-usual (BAU) for the United States appear in Table 1.
However, Ref4 notes, “The electric power cost of WWS [-only] in 2050
is not directly comparable with the BAU electric power cost, because
the latter does not integrate transportation, heating/cooling, or in-
dustry energy costs.6
In addition to his observation that comparing the LCOE’s of WWS-
only to BAU is an apples-to-oranges comparison (which he then does),
we will see that numerous costs have either not been counted or have
been assumed away. Further, a difference of $0.42/MWh over the long
time horizon used in that study is insignificant, and well within the
range of uncertainty that exists in the real-world, even though Ref4
implicitly assumed perfect information. Assuming that all states can
rapidly shift to WWS-only is hypothetical. While bench research can be
used to assess how various technologies interface or when alternative
policies may be effective/ineffective, analysis such as that in IRP is
essential to examining how to modify an existing utility power system.
Those trained in neoclassical price theory will note the positive
correlation between key assumption made in the partial-equilibrium
comparative statics (PECS) model of price theory and the structure of
Ref4’s methodology. PECS provides a powerful tool to help structure
policy analysis. However, blindly applying it without examining how
the results are altered when its assumptions of perfect information, zero
transactions costs, instantaneous transformation, and infinite divisi-
bility do not hold forms a shaky foundation upon which to base policy.
To be clear, LCOE under BAU and WWS-only portfolios are not es-
timates of a utility’s revenue requirements (revenues it needs to cover
costs), nor are they prices (rates) customers would pay for electricity.
Rather, they are estimates of the cost of two competing stand-alone
generation portfolios at the customer’s meter (Ref4 claims transmission
and distribution costs and lines losses have been included). Why stand-
alone? Typically, LCOE for different generation technologies are an
input to an analysis that examines ways a utility can go about meeting
future loads. These stand-alone costs are estimates of the stand-alone
production costs of the new technology. Below we will see that least
cost utility planning required of investor-owned utilities in Oregon and
by the NWPPC (and in other states) requires that the analysis account
for all the costs arising from changing the existing power system, which
is different than the LCOE of various technologies.
Ref4 at times assumes an extant utility exists (he notes that new
wind turbines would be located near existing ones) that needs mod-
ifying, while at other times he argues no such entity exists (as when he
asserts that integration costs are zero under the WWS-only). To be
credible, adding renewable generation to an existing power system
needs to account for risk and the costs of altering the existing power
system. More will be said about this and related issues in Section 3.
Turning to an overview of Ref4’s modeling, Loftus et al. (Loftus)
described it as “Top–down, scenario-based back-casting.7
A goal for
reducing carbon emissions is pre-selected and the acceptable technol-
ogies are pre-defined. The analysis results in an energy system that is
consistent with the pre-selected target using the pre-determined types
of generators. What Loftus points out is that Ref4 begins with the result
that is sought.
Ref4 identified reduced expenses for health care due to switching to
WWS-only of $1425/per person per year, and climate cost savings of
$7434/per person per year (those cost savings are taken at face value
herein). Avoiding these externalities is the basis of his argument that
the WWS-only portfolio is economic. However, these health care and
climate costs do not constitute a thorough examination of social costs
and benefits of the two portfolios. First, there is no substantive treat-
ment of externalities from WWS-only, from resource extraction to fab-
rication, shipping, construction and operation, and decommissioning.
Second, Ref4 implicitly assumes the accounting costs used to calculate
LCOE are reasonable estimates of the shadow prices of those factors of
production. Third, as we will see, numerous costs have been assumed
away when evidence suggests they in fact are positive.
As I was completing the draft of this paper, the paper by Clack et al.
at [14] became available. While there are several references to that
work in this paper, my focus is on a comparison between the structure
of Ref4’s analysis and how utilitiy planning is practiced in Oregon and
by the NWPPC.
It’s puzzling that Ref4 argued so forcefully for a WWS-only energy
system considering arguments made in a paper he co-authored with two
students (hereafter, HFJ). HFJ noted, “Because the approaches that
have been employed in moderate penetration [of renewables] regimes
may not be extendable to systems with very high penetrations, care
must be taken to place these methodologies into the proper context and
to formulate methodologies that can be applied to systems with very
high penetrations of intermittent renewables8
3. Further examination of Ref4’s analysis
This section examines Ref4 across eight aspects of utility planning.
These are: risk and adequacy of utility resource planning; commercial
availability; electricity demand; generation supply; costs; transmission
& distribution; system operation and reliability, and carbon policy.
Table 1
LCOE for BAU and WWS-only portfolios − US. ($/MWh).
(a) 2013 LCOE of BAU
(Electricity only)
(b) 2050 LCOE of BAU
(Electricity only)
(c) 2050 LCOE of
WWS (All energy)
10.19 10.04 9.62
Jacobson (June 2017), pp. 123–125.
5
For an explanation of Balancing Area or Authority, refer to: Glossary of Terms Used in
NERC Reliability Standards, Revised Aug. 1, 2017. See: http://www.nerc.com/files/
glossary_of_terms.pdf.
6
Jacobson (June 2017), p. 122.
7
Peter J. Loftus, Armond M. Cohen, Jane C. S. Long, and Jesse D. Jenkins, A critical
review of global decarbonization scenarios: what do they tell us about feasibility? Climate
Change, Nov. 6, 2014 p. See: http://onlinelibrary.wiley.com/doi/10.1002/wcc.324/full.
8
Elaine K. Hart, Eric D. Stoutenburg, and Mark Z. Jacobson, The Potential of
Intermittent Renewables to Meet Electric Power Demand: Current Methods and Emerging
Analytical Techniques, Proceedings of the IEEE, Vol. 100, No. 2, February 2012, p.323.
See: https://web.stanford.edu/group/efmh/jacobson/Articles/I/CombiningRenew/
HartIEEE2012. pdf.
R.J. Procter The Electricity Journal 31 (2018) 67–77
68

3. 3.1. Risk and adequacy of utility resource planning
Prior to evaluating the economics of potential portfolios, the utility
must first make the case that it needs to invest in new generation re-
sources. If no need exists, an economic analysis of options is irrelevant.
Ref4 answers the determination of need in the first sentence to his June
2017 study, “The seriousness of global air-pollution, climate, and en-
ergy-security problems requires a massive, virtually immediate trans-
formation of the world’s energy infrastructure to 100% clean, renew-
able energy producing zero emissions.9
The issue of need is non-trivial. For example, PGE’s 2016 IRP,
Docket LC66, before the Oregon Public Utility Commission
(Commission), proposed acquiring 175 aMW renewables ahead of need.
As the Citizen’s Utility Board (CUB) testimony attests, the determina-
tion of need comes before any assessment of economics. To be sure, had
Ref4 positioned that analysis as “what if” designed to evaluate tech-
nologies and various carbon reductions strategies, this review and cri-
tique would be structured differently.
While Ref4 presumes that some unidentified entity is able to
manage a coast-to-coast power system, PGE’s 2016 IRP10
is grounded in
its existing utility system. Lund11
et al. (Lund) also starts his analysis
with the electricity and energy system in place in Denmark, as does the
7th power plan developed by the NWPPC.
A robust risk analysis is crucial to evaluating power costs and as-
sessing how well the changed power system will function. Even if risk
isn’t endogenous, scenario analysis could provide useful information for
any entity contemplating enacting laws requiring high penetration of
renewables, including but not limited to, 100%. For example, PGE
models 23 different portfolios under 23 future states-of-nature. LCOE
for each technology is an input to the analysis. Complying with
guidelines established by the Commission, PGE uses net present value
revenue requirements (NPVRR) and three risk metrics. The preferred
portfolio is the one that has the highest weighted score after summing
the NPVRR, and the three measures of risk across the different future
states-of-nature. Furthermore, Joskow correctly pointed to the incon-
sistencies in using LCOE to compare dispatchable to non-dispatchable
generating technologies.12
On the topic of economic modeling, the necessary level of ag-
gregation and abstraction should be determined by the question(s)
being asked. When it comes to evaluating when and how to augment a
utility’s existing system, using models that are capable of representing
the long-run needs are essential. Keep in mind that the use of such
models abstracts from the day-to-day, let alone the second-by-second
operation of that system. These are planning models.
As such, they abstract from actual system operations. Turning to
the NWPPC regional power planning, the models used are at the
level of the PNW region. Even so, the level of rigor to adequately
capture power system operations is significant, in no small part due
to the fact that an extant power system exists, and any realistic
evaluation of that system, be it adding intermittent renewables,
increments of energy efficiency, DR, and the like must reasonably
take into consideration that extant system. Risks endogenous to the
NWPPC analysis includes unregulated river flows, temperatures (as
they affect electricity loads), forced outages on thermal generating
units and variability in wind generation, as well as different en-
vironmental policies. System operation is modeled picking ran-
domly from these stochastic variables thousands of times.
3.2. Commercial availability
For the WWS-only portfolio to pass an economic test, one ne-
cessary condition is that the technologies are commercially avail-
able. Edward Dodge13
(Dodge) argued that Ref4 offers no proof that
machines now using fossil fuels can be powered by hydrogen in a
reliable and cost-effective way, that he ignores land use issues, and
has “cherry-picked” technologies for both wind and solar that are
not the norm in the U.S. Clack also identified real-world deploy-
ment as one of the shortcoming of Ref4’s analysis, correctly framing
the problem “… it is critical that the scope of the challenge to
achieve this in the real world is accurately defined and clearly
communicated.14
Table 2 presents the technology classification used by the NWPPC’s
regional portfolio model. Only technologies classified “primary” (deemed
proven, commercially available, and deployable on a large scale in the PNW
at the start of the planning period) are allowed to meet load. It’s worth
noting that energy storage technologies are not available to the regional
power modeling, nor are upgrades to existing hydro, or new hydro., off-
shore wind, solar+battery storage, storage technologies broadly, tidal, or
wave generated electricity, all of which Ref4 assumes must be pursued now.
3.3. Electricity demand
In the residential sector, Ref4 implicitly forecasts electricity loads to
drop 26% by 2050 with WWS-only (249,200 MW compared to
336,800 MW under BAU). It’s sobering to note that total btu con-
sumption of natural gas in the residential sector of the U.S.15
is slightly
higher (4.694*1015
btu’s) than the amount of electricity, (4.388*1015
btu’s).
How realistic is it to expect that natural gas in homes will be re-
placed by electricity? Evidence from the PNW suggests that builders
and consumers will be reluctant to behave in that way. Turning to the
PNW, which comprises all of Idaho, Oregon, Washington and the
western-most portion of Montana, this region has some of the lowest
electric rates in the U.S. (see Table 3).16
Even so, the market share of
residences using electricity for space and water heating has declined,
while the fraction using natural gas has risen to approximately 58%.
Further, the NWPPC found that if consumers were able to choose the
lowest-cost system, regional electricity usage would drop by about
1000 GWh per year or 114 MWa by 2035.17
This result suggests that
9
Jacobson (June 2017), p. 2.
10
Follow this link to information on their IRP, https://www.portlandgeneral.com/our-
company/energy-strategy/resource-planning/integrated-resource-planning.
11
H. Lund, B.V. Mathiesen, Energy system analysis of 100% renewable energy sys-
tems: The case of Denmark in years 2030 and 2050, Energy 34 (2009), p. 526. See: http://
www.ewp.rpi.edu/hartford/∼ernesto/F2010/EP2/Materials4Students/Farrell/
Lund2009.pdf.
12
Paul L. Joskow, “Comparing the Costs of Intermittent and Dispatchable Electricity
Generating Technologies,” Center for Energy and Environmental Policy Research,
September 2010. See: https://www.researchgate.net/profile/Paul_Joskow/publication/
227357598_Comparing_the_Costs_of_Intermittent_and_Dispatchable_Electricity_
Generating_Technologies/links/0f31753985daf553a3000000/Comparing-the-Costs-of-
Intermittent-and-Dispatchable-Electricity-Generating-Technologies.pdf?origin=
publication_detail.
13
Edward Dodge, Critique of the 100 Percent Renewable Energy for New York Plan,
the energycollective, Nov. 17, 2013. See: http://www.theenergycollective.com/ed-
dodge/301031/critique-100-renewable-energy-new-york-plan.
14
Christopher T. M. Clacka, Staffan A. Qvist, Jay Apt, Morgan Bazilian, Adam R.
Brandt, Ken Caldeira, Steven J. Davis, Victor Diakov, Mark A. Handschy, Paul D. H. Hines,
Paulina Jaramillo, Daniel M. Kammen, Jane C. S. Long, M. Granger Morgan, Adam Reed,
Varun Sivaram, James Sweeney, George R. Tynan, David G. Victor, John P. Weyant and
Jay F. Whitacre, Evaluation of a proposal for reliable low-cost grid power with 100%
wind, water, and solar, Journal of the National Academy of Sciences, p. 6723. See: http://
www.pnas.org/content/early/2017/06/16/1610381114.full.pdf?with-ds = yes.
15
CE1.1 Summary Totals and intensities, U.S. homes, 2009 Residential Energy
Consumption Survey, Independent Statistics & Analysis, U.S. Energy Information Agency,
Release Date December 14, 2012. See https://www.eia.gov/consumption/residential/
data/2009/index.php?view=consumption.
16
The EIA reports an average price of $10.41/MWh. Table 1.2. Summary Statistics for
the United States, 2005–2015, EIA. See: https://www.eia.gov/electricity/annual/html/
epa_01_02. html.
17
Appendix N: Direct Use of Natural Gas, Seventh Northwest Conservation and
Electric Power Plan, May, 2016., p. N-10. See: https://www.nwcouncil.org/media/
7149904/7thplanfinal_appdixn_duofnatgas.pdf.
R.J. Procter The Electricity Journal 31 (2018) 67–77
69

4. unless statutes are changed to mandate electricity be substituted for the
direct use of natural gas, gas will likely continue to be the fuel of choice
where available.
Even absent the electrification of all current fossil fuel uses by
homes and businesses within its service territory, PGE’s IRP uses a
forecast of average annual load growth of 1.2% over the 2017–2050
timeframe,18
and its reference case forecast has been adjusted for en-
ergy efficiency investments. For the PNW, NWPPC forecasts 0.4% to
0.8% average annual growth rate.19
Loftus notes that energy intensity also affects forecasted energy use.
It declined 0.9% per year over the period 1990–2005. Ref4 assumes
annual reductions exceeding 10%.
3.4. Generation supply
Table 4 presents Ref4’s calculations of the number of new gen-
erators by fuel types in 2050.20
New capacity equivalent to three new
hydro plants, each with an installed capacity of 1300 MW (noted in the
study documentation) is required. For reference, when Grand Coulee
Dam was constructed, its initial installed capacity was approximately
2000 MW. To be fair, Ref4 notes that a variety of WWS-only portfolios
probably exist. While this caveat is well meaning, at the very minimum
it calls out for sensitivity analysis on how variations in the WWS-only
portfolio impact the results. This contrasts with the NWPPC analysis
that excludes new hydro plants and upgrades to existing hydro.
Over the next 33 years, Ref4 envisions approximately 5.8 million
MW of new generation is needed (plus an additional 600 GW for
peaking and system stability). He reports that as of 2013, 2.71% of that
amount is currently installed, or approximately 1600 MW.21
Since the U.S. installed capacity totals approximate 1.2 million
MW.22
and if only about 1600 MW of that amount (distributed PV is
10,000 MW) are usable for WWS-only in 2050, this also suggests a
significant stranded assets problem.23
While any sunk costs are ex-
cluded from a social benefit-cost analysis, they may find their way into
the costs used to set rates. As a result, in the real-world of utility ex-
pansion planning, cost recovery, and rate setting, stranded assets re-
main a significant issue. Any utility facing such a problem will file
testimony arguing that yet-to-be-depreciated plant costs should remain
in rate base while customer groups will argue for their removal, arguing
that those plants are no longer be used and useful.
We can also compare Ref4’s large increase in installed generation to
current expansion plans. Table 5 lists generating plants in various
stages of development as of 2015. Even if all these plants are con-
structed, it represents a total addition of about 200,000 MW, which is
roughly a 17% increase, about the same fraction of new plants on line
during 2006–2014.24
Adding 4.8 million MW of new generation over
the next 33 years implies an annual average increase in installed ca-
pacity of roughly 139,000 MW.
While Ref4 does not explicitly address plant ownership, plant
ownership impacts financing and tax exposure, and therefore costs. It
also has implications for policy at various levels of government. For
example, in the PNW there are about 156 publicly owned utilities. In
Oregon, there are 36 public utilities (including co-ops), all of which are
exempt from economic regulation by the Commission. Table 6, suggests
that administrative procedures and statutes must be broad-based,
especially in light of the fact that the overwhelming fraction of new
generation is being developed by non-utilities.
While Ref4 argues that new high voltage transmission and storage
resolve all reliability issues, PGE found that with increases in the pro-
portion of their portfolio comprised of intermittent renewables, they
will also need new dispatchable resources to adequately maintain re-
liability. As a result, PGE’s IRP includes a plan to acquire 400 MW of
additional dispatchable resources. The traits of such a resource
Table 2
Technology classification.
Source: NWPPC, Table H − 1: Classification of Generating Resources, p. H-4
Primary Secondary Long-Term
Natural Gas Combined Cycle Combustion Turbine Biogas Technologies (landfill, wastewater treatment, animal
waste, etc.)
Enhanced Geothermal
Natural Gas Reciprocating Engine Biomass Woody Residue Off-Shore Wind
Natural Gas Simple Cycle (Aeroderivative Gas Turbine, Frame Gas
Turbine)
Conventional Geothermal Small Modular Nuclear Reactors (SMRs)
On-Shore Wind New Hydro Solar + Bettery Storage
Utility-Scale PV Upgrade to Existing Hydro Energy Storage Technologies
Energy Storage Technologies Tidal
Waste Heat Recovery and Combined Heat and Power (CHP) Wave
Table 3
Overall electricity prices ($/MWh).
Source: “Table 5.6.A. Average Price of Electricity to
Ultimate Customers by End-Use Sector,” Electric Power
Monthly, U.S. Energy Information Agency. See: https://
www.eia.gov/electricity/monthly/epm_table_-
grapher.php?t = epmt_5_6_a.
Region/State All Sectors
Idaho 7.96
Montana 8.72
Oregon 8.81
Washington 7.81
New England 16.38
Mid-Atlantic 12.17
Mountain 9.11
Alaska 19.71
Hawaii 25.93
18
Figure ES-1: Reference case forecast by class: 2017 to 2050, Portland General
Electric 2016 Integrated Resource Plan, Executive Summary, p. 6. See: https://www.
portlandgeneral.com/our-company/energy-strategy/resource-planning/integrated-
resource-planning.
19
CHAPTER 7: ELECTRICITY DEMAND FORECAST, Seventh Northwest Conservation
and Electric Power Plan, May 2016. P. 7-3. See: https://www.nwcouncil.org/media/
7149931/7thplanfinal_chap07_demandforecast.pdf.
20
Mark Z. Jacobson, Mark A. Delucchi, Guillaume Bazouin, Zack A. F. Bauer, Christa
C. Heavey, Emma Fisher, Sean B. Morris, Diniana J. Y. Piekutowski, Taylor A. Vencilla
and Tim W. Yeskoo, 100% clean and renewable wind, water, and sunlight (WWS) all-
sector energy roadmaps for the 50 United States, Energy Environ. Sci., 2015, 8, 2093, p.
2098.
21
Ibid, Table 2.
22
Table 4.3. Existing Capacity by Energy Source, 2015 (Megawatts), U.S. Energy
Information Administration. See: https://www.eia.gov/electricity/annual/html/epa_04_
03. html.
23
Ref4 argues that any stranded costs from shifting to WWS-only from BAU are
compensated for by the avoidance of externalities of BAU. However, this is a specious
argument. Stranded assets will be addressed in a rate case and as such should be included
as a cost exposure to its ratepayers.
24
Table 1.3 Generating Capacity Additions, 2008-2014, America’s Electricity
Generating Capacity, 2015 update, American Public Power Association, p. 8.
R.J. Procter The Electricity Journal 31 (2018) 67–77
70

5. resemble a combustion turbine (CT). If some other option or group
thereof with the characteristics of a CT (rapid ramps, low operating
costs), such as hydro or demand response (DR) or perhaps storage, PGE
will consider those approaches.25
At this point in time, reaching that
much new flexible resource means at least two new CTs. More will be
said on this issue in 3–7.
3.5. Costs
Table 7 presents an estimate of the total capital program for new
generation only. This infrastructure construction project has an esti-
mated capital cost of just over $4.5 Trillion (T). T&D investments need
to be added, plus other equipment, such as storage devices, in order to
estimate the total capital cost of WWS-only. Data on functional
spending from the Edison Electric Institute (EEI) indicate that for 2013,
distribution-related investments total approximately 21%, and trans-
mission spending accounts for about 17% of total capital investment by
IOUs, for a total T&D fraction of 38%.26
These may be conservative estimates going forward due to higher
costs for new transmission to off-shore wind sites, reconfiguration of the
distribution system to enable significant amounts of DG, plus con-
structing new long-distance high-voltage DC transmission, and the cost
of storage devices. Adding 38% to the $4.5T results in a total capital
program of roughly $6.2T, a conservative estimate over the next 33
years. That implies an average annual capital investment of approxi-
mately $188 billion/yr. EEI indicates that annual capital programs in
the range of $100 billion are not unheard of. However, $188 billion/
year represents a doubling of annual investments every year for the
next 33 years. We may now be able to better see why Clack27
argues
that Ref4’s portfolio is cost-prohibitive in the real world.
While Ref4 argues that sufficient economic utility-scale storage ex-
ists to help maintain system reliability. PGE found that utility-scale
storage is not yet affordable and dependable enough to “…store the vast
amounts of [electricity] needed over weeks to reliably satisfy de-
mand…”28
Black & Veatch developed costs for PGE for 2-h and 4-hour
lithium-ion battery packs. When PGE compared those costs to an
equivalent amount of capacity from a 50 MW frame CT, they concluded
that the battery packs added about $2 M/yr. to costs, keeping capacity
contribution constant.29,30
Capacity factor can have a major impact on power system costs.
Ref4 does not address this issue in any kind of explicit detail. In con-
trast, PGE examined capacity factors of wind plants as loads changed.
Table 8 models a day in August. The results indicate that capacity
factors are negatively correlated with loads. Under high loads, the
frequency of zero output more than doubled and the frequency of a 20%
capacity factor grew by 50% and the frequency of a 90% capacity factor
dropped by half. This isn’t the direction of change that supports meeting
loads with intermittent renewables when storage remains too ex-
pensive.
PGE also found that as the proportion of output from renewable
generation increases, ramp rates for dispatchable generation increase,
as does wind plant output forecast errors. As a result, the output from
dispatchable generation changes more frequently resulting in less effi-
cient operation, more emissions, and greater costs on a dollar per MWh
basis.
Table 4
Number of generators in 2050 in the US. (x 1000).
On shore wind Off-Shore wind Wave Geo Hydro Tidal Res. PV Comm/Gov. PV Utility PV CSP
328 156 35 10 0.003 9 75,100 2747 46.5 2.3
Source: Ref20, et al., Energy Environ.
Table 5
Generation in various stages of development, US.
Total (MW) Fossil Fuel (%) Renewable (%)
Under Construction a
43,551 41 46
Permitted b
48,551 60 40
Pending Application c
79,263 43 41
Proposed d
200,273 26 65
Note: Rounded to nearest whole number. Nuclear excluded.
a
“TABLE 2.1 Plants Under Construction, Fuel Type, America’s Electricity Generation
Capacity 2015 Update, p. 11.
b
Ibid, “TABLE 2.2 Permitted Plants, Fuel Type,” p. 12.
c
Ibid, “TABLE 2.3 Pending Application Plants, Fuel Type,” p. 12.
d
Ibid, “TABLE 2.4 Proposed Plants, Fuel Type,” p. 13.
Table 6
New generation by ownership type, US. (%).
Non-Utility IOU Public Co-Op Federal
Under Construction a
58.9 25.4 10.6 2.2 3.0
Permitted b
89.1 4.5 2.4 3.9 –
Pending Application c
83.5 14.1 2.4 3.2 –
Proposed d
84.5 5.4 6.2 1.4 2.5
Note: Rounded to nearest whole number. Nuclear excluded.
a
“TABLE 5.1 Plants Under Construction by Ownership,” p. 24.
b
Ibid, “TABLE 5.2 Permitted Plants, by Ownership,” p. 24.
c
Ibid, “TABLE 5.3 Pending Application Plants by Ownership,” p. 25.
d
bid, “TABLE 5.4 Proposed Plants, by Ownership, p. 25.
Table 7
Capital cost of new generation, U.S.
Sources: Unless otherwise noted, $/kW cost are from “Levelized Cost of Energy—Key
Assumptions,”LAZARD'S LEVELIZED COST OF ENERGY ANALYSIS—VERSION 10.0,
pp.18-20.
Cost
($/kW)
Quantity (MW)
c
Total Expenditure
(×10^6)
Residential PV 2000 22,000 $44,000
Comm./Govt. PV 2100 99,000 $207,900
Utility PV, Crystalline 1400 2,100,000 $312
Geothermal 4250 21,000 $9.5
On-shore wind 1250 1,600,000 $2,000,000
Off-shore wind 2750 780,000 $2,145,000
Hydro a
2400 4000 $9600
Wave b
3000 27,000 $81,000
Tidal b
3400 8800 $29,920
TOTAL $4,517,741
a
Jacobson (2017), Table S24, p. 103.
b
Jacobson (2017), Table S26, p. 107.
c
Jacobson (2015), Table 2, p. 2098.
25
Further discussion of this issue appears in Chapter 5: Resource Adequacy of PGE’s
IRP.
26
EEI Industry Capital Expenditures, Slide #3. See: http://www.eei.org/
resourcesandmedia/industrydataanalysis/industryfinancialanalysis/
QtrlyFinancialUpdates/Documents/EEI_Industry_Capex_Functional_2014.07.00. pptx.
27
Clack et al.
28
Ibid, p.6723.
29
PGE 2016 IRP, p. 246.
30
Ibid, p. 238.
R.J. Procter The Electricity Journal 31 (2018) 67–77
71

6. PGE also found that spikes in wind plant output likely requires
ramping down two CTs and hydro, and the hour-ahead wind forecast
error led to a need for greater upward flexibility at the 5-min level.31
This is why PGE found that new dispatchable generation is needed to
integrate more intermittent generation at high penetrations. Even at
25% RPS, about 400 MW of new dispatchable generation is needed to
avoid significant real-time imbalances between output and loads. Keep
in mind that these results are obtained from long-run planning analyses
performed using system operations modeling. How the power system
actually operates in real-time will be determined by facts during real-
time operations.
Ref4 implicitly assumes no costs need be included in system cost asso-
ciated with needed changes to existing laws and rules that guide power
system operation. Several citations identify such cost categories. One place
these costs will show up is in the costs submitted to the utility commission in
a subsequent rate case. They will also show up in cost reimbursements from
any fund that exists to support intervenor costs, such as those by the Citizens
Utility Board in Oregon. For example, one citation identified twelve trans-
action costs for the Kyoto Protocol.32
The World Bank identified transaction
cost of low-carbon policies, and they appear in Appendix A.33
3.6. Transmission and distribution
Ref4 understands that the proper way to determine T&D system
expansion is to perform an analysis of the optimal modifications to the
existing system. Given the “bench research” nature of their analysis,
those calculations are outside the scope of his analysis. Despite this
constraint, Ref4 argues that the LCOEs are for a system “…that we think
plausibly, reliably will match supply and demand.34
Clack calls Ref4
out on this point, noting especially the absence of both a power flow
study and any discussion of transmission constraints.
Ref4 also argues that a WWS-only generation portfolio will not have
integration (interconnection) costs. In contrast, PGE’s IRP notes that
transmission services will be needed for all new renewables either from
their transmission business or from the Bonneville Power
Administration (BPA), which owns 75% of the high-voltage transmis-
sion in the PNW. In either case, transmission charges are costs to PGE’s
power business and as such will be reflected in the LCOE for the various
generation options.
In addition to transmission rates, BPA has a set of procedures that
govern the technical aspects of interconnecting new generation, es-
tablishing costs of system upgrades and interconnection facilities, and
creating a queue priority based on the order in which requests are re-
ceived.35
Further, the NWPPC noted that each BA takes a different
approach to estimating integration costs, and a partial list appears in
Appendix B. Ref4 assumes these costs do not exist.
3.7. System integration, operations and reliability
A study conducted for the Western Electric Coordinating Council
(WECC) suggest that evaluating the ability of the existing power system
to use the output from renewables must examine the sub-region. For
example, a 2015 report indicated that unless mitigation is provided,
within WECC the need for curtailment of generation and/or load rises
at an increasing rate with the proportion of the generation portfolio.
However, the different sub-regions faced different challenges. While
California wouldn’t be able to use 50–60% of the potential output of the
next solar PV installation (over and above the levels assumed in the
analysis) under the high-renewables scenario, they note that the
amount of displacement observed in California and the Southwest were
much greater than that observed in both the Basin and Rocky Mountain
sub-regions. They also noted that it appears that the type of plants
(solar PV, wind, geothermal, etc.) and geographic distribution within
the sub-region drive the differences in the frequency of displacement
between what they call their common case and the high renewables
case.
Table 9 illustrates the impact of different strategies to reduce the
need to curtail renewable generation in different WECC sub-regions.
One key assumption was that each utility BA within each sub-region
allowed the sub-region to control its resources to optimize system op-
erations across the sub-region. The authors noted that such an as-
sumption is heroic considering that bilateral transactions predominant
within WECC.
Information from an MIT Symposium reinforces the understanding
that instability occurs in fractions of a second.36
They concluded that
“Intermittent renewables present integration challenges at all time-
scales for the power system. As renewable penetration increases, system
stability on the timescales of fractions of a second will increasingly
matter as much as backup capacity at the minutes to hours scales.”
Modeiling in PGE’s, 2016 IRP provide insights into the challenge
they face maintaining reliability as the proportion of intermittent re-
newables increases. For example, they found that reducing generation
may be more difficult than increasing it, if capacity needs are primarily
met with non-dispatchable generation (e.g., WWS-only). They sug-
gested that one way around this operational problem is scheduling
more generation day-ahead to retain the ability to decrease generation.
They also found that at the higher RPS of 50% in 2040, the curtailment
potential (recall, it was 3.3% under 25% RPS), rose to 18%, roughly a
factor of five with a doubling of the installed intermittent generation.37
PGE found that the cause of their system’s reduced capability to
Table 8
Capacity factor frequency under two load scenarios.
Capacity Factor (%) Low Load (%) High Load (%)
0 15 32
20 8 12
40 2 2
60 2 2
80 12 2
90 18 8
Table 9
Renewable curtailment, high renewables case (%).
Source: “Table 3. Impacts of new investments on regional renewable curtailment, High
Renewables Case,” Western Interconnection Flexibility Assessment, Final Report, NREL,
December 2015, p. xxvi. See: https://www.wecc.biz/Reliability/WECC_Flexibility_
Assessment_Report_2016-01-11.pdf.
Basin California Northwest Rockies Southwest
Reference 0.4 8.7 5.6 0.6 7.3
+600 MW 2 h. Storage 0.4 7.2 5.7 0.6 6.2
+600 MW 6 h. Storage 0.4 5.8 5.8 0.6 5.1
+600 MW 12 h. Storage 0.4 5.8 5.7 0.6 5.1
+600 MW CCCT 0.4 8.7 5.6 0.6 7.3
31
Ibid, p. 138.
32
Axel Michaelowa, Frank Jotzo, Transaction costs, institutional rigidities and the size
of the clean development mechanism, Energy Policy 33 (2005) p. 513. See:.http://www.
sciencedirect.com/science/article/pii/S030142150300257X.
33
Luis Mundaca, Mathilde Mansoz, Lena Neij, and Govinda R Timilsina, Transaction
costs of low-carbon technologies and policies: The diverging literature, The World Bank,
Development Research Group Environment and Energy Team, August 2013. See: http://
documents.worldbank.org/curated/en/903121468162285896/pdf/WPS6565. pdf.
34
Ibid, p. 119.
35
Presentation by Nick Peck, Generation Resource Interconnection & Integration. BPA
Transmission Services, June 30, 2016, slide no. 8. See: https://www.bpa.gov/
PublicInvolvement/Cal/doc/Generation-Resource-Interconnection-and-Integration.pdf.
36
MIT Energy Initiative, April 20, 2011, p. 13. See: https://energy.mit.edu/wp-
content/uploads/2012/03/MITEI-RP-2011-001.pdf.
37
PGE 2016 IRP, p. 145.
R.J. Procter The Electricity Journal 31 (2018) 67–77
72

7. meet system peaks in 2021 (it declined by as much as 2000 MW) was
due to an increased demand for balancing reserves, both INC and
DEC.38
If no new dispatchable generation is added, system imbalances
increase upwards of 3000 MWh/yr., but they can be reduced by up to
two-thirds with the acquisition of new dispatchable generation. How-
ever, this implies that new generation alone doesn’t mitigate all the
real-time imbalances, or the oversupply issues at 50% RPS.
This potential for inadequate reserves can be dire for generators, the
majority being non-utilities. If DEC balancing reserves are insufficient,
and they are being provided by BPA, it limits generators to their
schedules when intermittent generators are over-generating.39
That
means those renewables lose revenue due to reduced output and that is
especially problematic when they are counting on production tax
credits tied to the level of output. If BPA is providing INC balancing
reserves, the consequences can be even greater, since BPA auto-
matically curtails transmission schedules for each non-federal generator
that is under generating. Appendix C contains 10 graphs that illustrate
how renewables curtailment changes within each WECC sub-region as
the fraction of renewables increases in 2024 from state-level RPS as of
2014 for the common case to renewables penetration roughly at least
double the common case fractions.40
About a year ago, California increased its required amount of fre-
quency regulation. It has been reported that the California Independent
System Operator (CAISO) roughly doubled the required amount both
day-ahead and real-time. As a result, the price of this service went from
about $5/MWh to $15/MWh. The CAISO indicated that the reason for
the increase was driven by the potential for greater renewables output
volatility.41
Replacing fossil fuel generators (which are synchronous) with WWS-
only (which are non-synchronous) will likely exacerbate frequency
control issues along the distribution system. In 2016, Australia had a
transmission line go out that caused 445 MW of wind generators to go
off-line. There wasn’t sufficient synchronous generation available to
keep frequency control within prescribed limits. Nearly 2 millions
customers lost power.42
In the future, other devices than CTs may be
integrated into the grid that aid managing distribution frequency.
However, it doesn’t appear that day has yet arrived, at least not in
WECC.
All this is in keeping with Heard’s observation that “Projected 100%
renewable electricity systems are incomplete in the absence of evidence
that essential, regulated ancillary services, will be maintained.43
Clack
dryly noted that its fairly easy to match instantaneous energy demands
with variable generation if one assumes a nationally integrated grid,
loads that can be shifted in time, and the existence of large amounts of
inexpensive storage.44
Trainer noted that in 2006, for 300 hours there
was almost no wind energy over the whole of Ireland, the UK, and
Germany, in midwinter, the best time of the year for wind energy. For
about 120 continuous hours, UK capacity averaged about 3% while
electricity demand reached its peak high for the year. Jacobson at [20]
sweeps away these operational concerns arguing that all grid
integration problems are resolved through a combination of storage,
DR, and ramping down other WWS-only generators.45
3.8. Carbon policy46
One tool of carbon policy that has seen extensive use throughout
WECC are RPS. PGE’s IRP discusses how it plans to meet its RPS ob-
ligation, not just today, but throughout the next 23 years. However, in
keeping with the necessary condition to show that a need exists before
deciding to allow cost recovery for capital investments, interveners
submitted testimony opposing PGE’s plan to acquire more renewables
in excess of need.47,48
While a thorough examination of different policies to reduce carbon
is beyond the scope of this paper, it is worth noting that studies have
generally indicated that carbon must be priced for high-penetration
renewables to be cost-effective on their own. Elliston concluded that
carbon needed to be priced for the 100% renewable portfolios to be
economic (he allowed for a broader range of renewables than WWS-
only).49
If the discount rate was 5%, carbon prices needed to be in the
range of $50-65/mega-ton of CO2 equivalent (MTCO2e). If the discount
rate rose to 10%, carbon prices needed to increase to between
$70–$100/MTCO2e. Otherwise, replacing existing fossil-fueled gen-
erators with newer fossil-fueled generation was the least-cost approach
when carbon prices are below these levels.
PGE found that parity (between its preferred portfolio in its 2016
IRP and one with more new wind generation) was reached at a price of
about $500/ton (real levelized 2016$).50
They suspect this price level is
needed due primarily to two factors: (a) there already would be a high
penetration of renewables to meet the 50% RPS by 2040; and, (b) all
the portfolios evaluated already show significant drops in CO2 intensity,
most likely from shuttering their Boardman coal plant in 2020, ceasing
power purchases from Coalstrip units 3 and 4 in Montana by 2035, and
the significant amount of energy efficiency that will have been added
on top of what already exists. By 2050, both the Preferred portfolio and
Wind 2018 Long have approximately the same quantity of CO2 emis-
sions, about 6 million short tons.
A study published earlier this year included results of analysis
performed by the NWPPC on six different carbon reduction policies.
Among its conclusions was that the RPS mandate was the costliest ap-
proach to carbon reduction, on a dollar-per-mega ton of CO2 reduction
basis.51
38
PGE 2016 IRP, p. 130.
39
Operational Controls for Balancing Reserves, Limiting Plant Output to Scheduled
Values and Curtailing Schedules to Actual Plant Generation, Bonneville Power
Administration, Oct. 1, 2015. See: https://www.bpa.gov/Projects/Initiatives/Wind/
Pages/operational-controls.aspx.
40
The specific fractions of renewables used for both scenarios for each WECC sub-
region can be found in Fig. 4 . Renewable portfolios analyzed in the flexibility assessment,
of the Western Interconnection Flexibility Assessment, Final Report, p. xiii.
41
Jeff St. John, July 5, 2016. See: https://www.greentechmedia.com/articles/read/in-
california-solar-and-wind-boosts-the-price-for-frequency-regulation.
42
B.P. Heard, B.W., Brook, T.M.L. Wigley, and C.J.A. Bradshaw, Burden of Proof: A
Comprehensive Review of the Feasibility of 100% Renewable Electricity Systems,
Renewable and Sustainable Energy Review, 76(2007), p. 1125. See: https://www.
sciencedirect.com/science/article/pii/S1364032117304495.
43
Ibid.
44
Clack, p. 6723.
45
Jacobson (2015), p. 2104.
46
While this citation is somewhat old, it still provides an overview of environmental
regulation of electric utilities. Jim Lazar and David Farnsworth, Incorporating
Environmental Costs in Electric Rates, Working to Ensure Affordable Compliance with
Public Health and Environmental Regulations, October 2011. See: http://www.raponli-
ne.org/wp-content/uploads/2016/05/rap-lazarfarnsworth-in-
corporatingenvironmentalcostsinelectricrates-2011–10. pdf.
47
Driving that point home is Staff’s recommendation that the commissioners not ac-
knowledge PGE’s plan to acquire 175 aMW renewables before the time they are out of
physical compliance with the RPS. Staff argued that doing so would maintain a rate-
making principle that the customers who give rise to a cost should also bear that cost. See:
J. P. Batmale, PORTLAND GENERAL ELECTRIC: (Docket No. LC 66) Acknowledgement of
2016 Integrated Resource Plan, PUBLIC UTILITY COMMISSION OF OREGON STAFF
REPORT PUBLIC MEETING DATE: August 8, 2017, July 27, 2017, P. 26. See: http://
edocs.puc.state.or.us/efdocs/HAU/lc66hau165349. pdf.
48
Filings by all parties to docket LC66, PGE’s 2016 IRP filing, may be found at: http://
apps.puc.state.or.us/edockets/docket.asp?DocketID=20423.
49
Based on the data in Table S34 of p. 123 of the Supplemental Information to
Jacobson (June 2017) found the WWS-only portfolio for Australia had a 2050 all-energy
LCOE of $9.26/MWh compared to 2050 BAU-electricity of $10.06/MWh.
50
PGE 2016 IRP, p. 326.
51
Robert J. Procter, Cutting Carbon Emissions form Electricity Generation, The
Electricity Journal, Volume 30, Issue 2, March 2017, Pages 41–46. See: http://www.
sciencedirect.com/science/article/pii/S1040619017300118.
R.J. Procter The Electricity Journal 31 (2018) 67–77
73

8. 4. Conclusions and policy implications
Bench research such as that by Ref4 and [20] can help scope targets,
plans, critical paths, suggest where policies may be aligned or in con-
flict, plus suggest insights of a purely engineering nature. What they
don’t do is substitute for the very difficult and messy process of real-
world utility policy development and implementation. For the reasons
outlined in this paper, activists, county commissions, city councils, and
legislators need to move with caution when dealing with proposals
mandating or promoting WWS-only as a carbon reduction policy.
PECS can be a valuable tool for helping the analyst frame policy
analysis. However, once we step out of the textbook and into the real-
world of actual utility operations, it becomes essential to step back and
question if that particular model is adequate to the task at hand. This
paper has primarily, though not solely, focused on utility policy ana-
lysis practiced by the Commission and the NWPPC. Comparing as-
sumptions made in Ref4 to these long-run real-world electric utility
planning efforts has identified numerous examples where the simpli-
fying assumptions made in Ref4 do not support the conclusion that
WWS-only is economically and technically viable, at this time.
A robust utility expansion planning analysis must examine what
impacts various policies may have on renewables adoption and CO2
emissions once Ref4’s assumptions of perfect information, zero trans-
action costs, and instantaneous transformation are replaced with more
realistic assumptions.
Ref4 projects the U.S. will need to transform the current physical
power system with about 1.2 million MW of generation, to one with an
installed capacity nearly six times greater. The institutional require-
ment to support that transformation, plus operate a coast-to-coast BA,
are also complicated. Just for starters, a new organization would need
to be formed that has the authority to manage loads and generation
across the continental U.S.
We know how critical it is to reduce CO2 (and other GHG)
emissions. Yet, our resources are not unlimited while our problems are
many. The severity of the problem of GHG emissions in light of our
limited resources means we must be rigorous in our analysis of how to
go about reducing those emissions. This is especially daunting con-
sidering that the electrical system is the most critical of critical infra-
structures.
Seven criteria should guide any study purporting to demonstrate
that a high-level penetration of renewables is economic. These are:
A Detail the steps required to shift from the utility’s existing system to
one with a high-penetration of intermittent renewables;
B Models must account for existing statutes and administrative rules;
C Evaluate risks, including but not necessarily limited to (1) loads; (2)
generation output, especially of intermittent renewables; (3) fuel
prices; and (4) CO2 costs;
D Use sub-hourly modeling of power system;
E Use commercially available technologies and plausible loads;
F Include the equipment costs required by two-way communication;
and
G Use NPVRR or PVRR to evaluate costs, including specific attention
to stranded costs, integration costs, ancillary services, T&D costs,
transaction costs, and costs associated with prospective, but not yet
implemented, carbon policies.
It is incumbent on us to move forward developing solutions to
carbon emissions. When it comes to how energy is produced and used,
doing so requires complex analysis of highly technical systems.
Understanding the trade-offs and needs must be informed by both solid
science and an appreciation for the immense complexity of the gov-
erning infrastructure of rules, procedures, statutes, and institutions. The
alternative is to pursue approaches that may both cost more and pro-
duce poorer results.
Appendix A. Illustrative transaction costs related to renewable energy technology a
Source Scale Citation
Search and pre-feasibility, negotiation and development, approval and administrative
procedures (planning phase)
9% of total investment
costs
Skytte et al. (2003, pp.
66–67) b
Monitoring (accounting and verification), enforcement and adjustment costs
(production phase)
7% of total investment
costs
Costs undertaken by obligated parties beyond costs of meeting the obligation itself;
including costs to handle quota obligation on behalf of end-users (for the specific
TGC scheme in Sweden)
2.5% of total activity
costs
Oikonomou and Mundaca
(2008, pp. 224–225) c
18% of the total costs Kåberger et al. (2004, p.
687) d
Renewable Portfolio Standard, Texas (US) 2.9 % of the value of the
Renewable Elec.
Langniss (2003, p. 230) e
German Renewable Act 1.3% of the value of the
Renewable Elec.
Langniss (2003, p. 231) f
Notes:
a
Excerpted from Table 3 on p. 21 of the World Bank Report (see footnote 33).
b
Skytte, K., Meibom P., Uyterlinde M.A, Lescot, D., Hoffmann, T. and del Rio P. (2003). “Challenges for Investment in Renewable Electricity in
the European Union”. Background report in the ADMIRE REBUS project supported by the European Commission. November 2003.
c
Oikonomou, V. and Mundaca, L. (2008). “Tradable White Certificates Schemes: What Can We Learn from Tradable Green Certificate Schemes?”
Energy Efficiency. DOI 10.1007/s12053-008-9017-7.
d
Kåberger, T., Sterner, T., Zamanian, M., and Jurgensen, A. (2004). “Economic Efficiency of Compulsory Green Electricity Quotas in Sweden”.
Energy & Environment, 15(4), 675–697.
e
Langniss, O., Wiser, R., (2003). “The Renewables Portfolio Standard in Texas: an Early Assessment.” Energy Policy. Volume 31 (2002). Pages
527-535.
f
Langniss, O. (2003). “Governance Structures for Promoting Renewable Energy Sources”. Lund University. Department of Technology and
Society.
R.J. Procter The Electricity Journal 31 (2018) 67–77
74

9. Appendix B. Illustrative integration costs
Source: “A Review of Variable Generation Integration Charges,” Table 4, pp. 12–18,//and Table 5, pp. 19–25. While these costs are at least
several years old now, they may have changed. However, they remain illustrative of integration costs of intermittent renewables for several PNW
entities.
Utility Integration Charge
Avista 7% of the published avoided-cost rate for wind qualifying facilities under PURPA (capped at $6.50/MWh).
BC Hydro $10/MWh (for the 2012 IRP)
BPA Wind: $1.23 per kW-month Solar: $0.21 per kW-month
Idaho Power 8% of the published avoided-cost rate for wind qualifying facilities (QFs) under PURPA (capped at $6.50/
MWh).
Northwestern Energy
(Montana)
$11.28/MWh
Portland General Electric $9.15/MWh (2014$, included in 2011 IRP update)
PacifiCorp $9.70/MWh (2010$)
Puget Sound Energy $1.55/kW-month of transmission reservation capacity for generators with hourly scheduling intervals
• →30% discount available for 30-min scheduling intervals
• →50% discount available for 15-min scheduling intervals
R.J. Procter The Electricity Journal 31 (2018) 67–77
75

10. Appendix C. Sub-hourly curtailment from inadequate system flexibility.
See Fig. A1.
Fig. A1. curtailment experienced as a result of flexibility reserve shortages in the High Renewables Case, Western Interconnection Flexibility Assessment, Final Report,” NREL, December
2015, p. 185. See: https://www.wecc.biz/Reliability/WECC_Flexibility_Assessment_Report_2016-01-11.pdf.
Source: “Figure 85. Sub hourly curtailment experienced as a result of flexibility reserve shortages in the High Renewables Case, Western Interconnection Flexibility Assessment, Final
Report,” NREL, December 2015, p. 185. See: https://www.wecc.biz/Reliability/WECC_Flexibility_Assessment_Report_2016-01-11.pdf
R.J. Procter The Electricity Journal 31 (2018) 67–77
76

11. Robert J. Procter is an independent energy economist in Portland, Oregon. He
previously worked for the Bonneville Power Administration and the Oregon
PUC. Among the efforts he had led were examining whether bias exists in the
utility bidding process, smart grid planning, demand response policy, electricity
pricing, and determining the cost-effectiveness of conservation in new residential
construction. He has taught economics at colleges in Portland and has been a guest
lecturer in graduate seminars for the Hatfield School of Government at Portland
State. He holds a Ph.D. from Michigan State, and M.S. from Purdue, and a B.A. from
UC Berkeley.
R.J. Procter The Electricity Journal 31 (2018) 67–77
77