20131106Comment.pdf

Comments to the Idaho PUC regarding the Idaho Power 2013 IRP
By Courtney White
A fundamental rote of the IRP process is to consider what the future may look Like and what
alternative paths forward would best serve needs.With regard to energy,we are facing new
tevels of change and uncertainty.The 2013 IRP process has not yet adapted to evaluate the
impact of technology changes and the vatue of option-preserving alternatives,as exptained
by the comments in this document.This increases the need for Idaho Power,the pubLic,and
the PUC to give diligent and open-minded consideration to alternatives at any decision point
before further vesting in specific resource aLternatives.
Contents
A.Risks,Uncertainties,and the Value of Flexibility
1 The value of staging investments to meet demand
2 The vaLue of fLexibility when forecasts are wrong
3.The demand forecast is more uncertain than presented
4.The cost of planning for worst case scenarios
5.The value of leveraging improvements in technoLogy
B.Alternative Resources /Expected changes
1.The IRP process did not incorporate the value of projected improvements in
storage technology
2.The opportunities for demand management are not adequateLy considered
3.Energy efficiency was not adequately considered as a resource
4.The Resource Alternatives Analysis of solar was invalid due to poor assumptions
C.Given increased uncertainty,new approaches are needed
1.The process shouLd value fLexibility.
2.New approaches to valuing storage should be adopted.
3.The process should seek to understand the conditions in which alternative paths
forward wouLd better serve customer needs.
4.The process shoutd be a venue for seeking out better options.
5.The process and financial incentives should adapt to enable more objective
consideration of atternatives
A.Risks,Uncertainties,and the Value of Flexibility
The IRP addresses risk in its section Risk AnaLysis and Results.This analysis identified only
four variabtes:natural gas prices,customer toad,hydroelectric conditions,and a carbon
adder.There are numerous variables that effect the value of alternative resources that are
not adequately vatued in the comparison of alternative resources.Here are some examples:
An analysis of alternative resources should reflect savings gained when
capacity can be installed incrementally to meet demand growth,thereby
delaying capital costs,overhead,and variable costs.
/Lumpy 200 MW Capacity Investment
2.The value of flexibility when forecasts are wrong.When the future is uncertain,
options which preserve fLexibility have higher vatue.The IRP does not adequately address
this fundamental difference in aLternative paths.If,for exampLe,actuaL demand growth
were less than projected,atternatives which alLow the utitity to postpone additionaL
investments woutd be advantageous.The cost of adding unnecessary capacity couLd be
avoided.There is a
financiaL vaLue to Issue:The lumpier the investment,the higher the cost of mis-forecasting.
flexibility which the
IRP process is
faiting to capture in
its comparison of
aLternatives.
1.The value of staging investments to meet demand.Resources vary in the degree to
which investment can be Issue:The cost of excessstagedtobestfit
demand growth.In the
Resource Alternative
Analysis,the 2013 IRP
does not address the
advantages of deLaying
the fixed and variabLe
costs associated with
resources which can be
staged incrementaLly to
fit the demand growth.
200 MW
c.1.
—
This chart isfor exemplary purposes.Theflexibility to stage capacity additions to fit demand
growth is specific to each alternative.
If actual requirements are less than projected,some alternatives preserve the option to
adapt,which could save customers significant $by preventing investment in unnecessary
infrastructure.
Lumpy 200 MW Capacity Investment
This chart is for exemplary purposes.The flexibility to stage capacity additions to fit demand
growth is specific to each alternative.
Over the most
recent 5 years (2012
/2007),the
annuatized growth in
number of
residentiat
customers has been
0.8%.The 2013 IRP
projects residentiat
customers to grow
by an annuaLized
rate of 1.6%over
the coming five
years.This
projection seems
aggressive.
3500
1500 Year
13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
The above chart is hypotheticaL for ittustration only.Betow is a chart showing the proposed
capacity additions relative to the forecasted peak hour need as described on pages 107-108
of the 2013 1RP.
Note that,if in Proposed capacity plan relative
year 17 the to peak hour forecast
demand were
tower than MW
projected,the 4500
proposed excess Proposed
capacity in year 4000
18 is even greater
Note also that the
proposed capacity
plan assumes 150 3000MWofDemand For illustration:330MW planning margin
Response in year
16 and 17,which 2500
goes to zero in
year 18.2000
Peak-Hour +330MW
--Existing
Resources
3.The demand forecast is more uncertain than presented.There are numerous factors
and assumptions which go into the demand forecasting,and we ratepayers have limited
representation as the Company form utates its projections.As a test of whether the process
is objective and the range of uncertainty is being adequately presented,let’s consider the
projected growth in
number of
residentiaL The 2013 IRP forecasts an accelerated growth in #
customers.of residential customers relative to recent years.
550,000
500,000
450,000
Actual #residential customers _e-’
::::.
300,000 If future growth were equal to recent
5-year annualized growth rate
250,000
200,000 -
<2 0)“s,‘)<)0)
9,9,9,9,9,9,9,
3
To verify whether the methodotogy has worked in the past,tets took back at a prior IRP.In
the 2006 IRP,the number of residentiat customers was projected to grow by 68,761 from
2005 to 2012.This projected increased was 72%higher than actuat.
For example,the increase in the number of residential
customers grew much less than projected in the 2006 IRP
Increase in It of residential customers.2005-2012
The 2013 IRP states that the utility needs
$5800 in capital costs for each new residential
customer.The over estimate in number of
customers correlates to $167 million over
estimate in capital requirements.
40,008
1
2006 IRP projected increase Actual increase
Simitarty,consider the hypotheticat that the growth rate in number of residentiat customers
is constant rather than acceterated in the coming years.The 2013 IRP projects the number
of residentiat customers to be 494,980 in 2018.lIthe growth rate in residentiat customers
continues at the same rate as the prior 5 years,the projection woutd be in excess of actuat
by over 60,000.At $5800 per customer,this error margin would corretate with
approximately $350 mittion in capital costs.This anatysis is not a prediction but a
demonstration of the vatue associated with atternative paths that mitigate the risk of
forecast errors.
The 2013 IRP identifies Low and High growth toad projects as a means of addressing the
range of uncertainty.For exampte,the “Low”scenario (page 14,Appendix A)is 4.5%tess
than the Expected”scenario in 2017.Is that an adequate range?With regard to the
exampte variable of residentiat customers,the actuat number in 2012 was 6.5%tess than
projected in the 2006 IRP.
The future is not becoming easier to predict but more uncertain.An alternative which
allows investments to be staged more closely with actual demand has value because it
reduces the risk of investing in excess capacity.The 2013 IRP is inconclusive because it
does not adequately account for this value.
68,761
4
4.The cost of contingency capacity requirements were not adequately evaluated.
The 2013 IRP on page 106 describes the Level of excess capacity needed:
Idaho Power maintains 330 MW of transmission import capacity above the forecast
peak toad to cover the worst single planning contingency.The worst single planning
contingency is defined as an unexpected toss equat to Idaho Power’s share of two
units at the Jim Bridger coal facility.
The levet of excess capacity needed for contingency planning varies by alternative resource.
The Resource Alternatives Analysis incLudes reliability risks for some alternatives;for
example,the risk that the sun doesnt shine translates into a cost in the IRP’s comparison of
solar to alternative resources.However,the cost of risks associated with disrupted
operation of the Bridger coal facilities is not integrated into the cost of continuing to rely on
these alternatives.The level of contingent capacity varies across different resource
strategies,and a diversified array of resources will tend to have lower risk than reliance on
large facilities,particutarly ones that are minority-owned.The 2013 IRP is inconclusive
because risks are selectively translated into costs for some alternatives yet not for others.
B.The IRP process did not adequately address expected changes
The IRP process is misleading by presenting a 20-year forecast that omits trends and
technologies which have a material impact on the forecast.
1.The IRP process did not incorporate the value of projected improvements in storage
technology.The IRP process did not adequately address the full benefits of storage,the
wide range of options,or the changing nature of this technology.
The PUC in California,for example,recently announced its requirement that the state’s
large investor-owned utilities must buy a total of 1.3 gigawatts of storage.The process
arriving at this decision did not isolate the cost effectiveness of storage as a new resource
alternative based on todays technology and costs,instead it considered the impact of
improved cost effectiveness for renewable resources,the need for emergency capacity,and
the impact on grid optimization.The San Francisco Chronicle quotes one commissioner in its
October 17 coverage http://www.sfate.com/business/articte/CPUC-s-new-energy-storage-rutes-aim-to-aid-grid
4904875.php:
“Energy storage has the potential to be a game changer for our electric grid,and I
fully support the goals of grid optimization,integration of renewable energy and
reduction of greenhouse gas emissions,”said Commissioner Mark Ferron.
Storage technology will improve,and Idaho Power’s rate payers deserve to benefit from
these improvements.As reported on UPI.com,posted October 18,2013:
“Integrating renewables is complex,and the complexity is increasing as more solar
and wind comes online,”Carla Peterman,also a CPUC commissioner,who previousty
5
served on the CaLifornia Energy Commission,was quoted as saying by the San
Jose Mercury News.
“Storage has a tot of potentiaL,but theres no one soLution.We need to deveLop this
market to see what the potentiaL is,’Peterman said.
CaLifornia’s mandate is expected to spur innovation in emerging storage technologies,
from batteries to flywheeL energy storage,a technique that acceLerates a rotor to
high speeds thus creating a kinetic battery.
“This is transformative,”Chet Lyons,an energy storage consuLtant in Boston toLd the
Mercury News.“It’s going to have a huge impact on the development of the storage
industry,and other state reguLators are looking at this as a precedent.”
During the IRP pLanning horizon,nations around the worLd witt be investing to provide
etectricity to growing populations,driving improvements in distributed generation and
storage technology,which wiLt improve the cost effectiveness of sotar and wind here in
Idaho.Avista,for exampte,notes in its 2013 the potentiat vatue of storage (section 6-10):
Increasing amounts of sotar and wind generation on the eLectric grid makes energy
storage technoLogies attractive from an operationat perspective.The technoLogies
coutd be an ideal way to smooth out renewable generation variability and assist in
toad fottowing and regutation needs.The technology atso coutd meet peak demand,
provide voLtage support,relieve transmission congestion,take power during over
supply events,and supply other non-energy needs for the system.Over time,storage
may become an important part of the nation’s grid.
The flexibility to take advantage of technology improvements has financial value,which
should be considered in comparing alternative resource strategies.
2.The opportunities for demand management are not adequately considered.Time-of-
use pricing,for exampLe,has potentiat to substantiatty change the peak toad forecast.As
utitities around the world strive to tink pricing to costs,behaviors change and new markets
are created for technologies that enable customers to optimize the timing of their
electricity demand.Rather than project strong and unchecked growth in peak demand,as
shown in page 29 of Appendix A (below),the IRP process should give greater consideration to
strategies and trends that could better manage the shape of the demand curve over the 20-
year planning horizon.
6
Table 10.System summer peak load growth (MW)
Annual Growth Rate
Growth 2013 2017 2022 2032 2013—2032
95ePercenlile 3382 3596 3,881 4,418 1.4%
gO Percentile 3,344 3,555 3,835 4,365 1.4%
In a free market,a flim would not forecast
189 3,387 3,651 4,147 1.4%
unmitigated growth in peak demand for ..
the next 20 years.Competitive pressure I
would prompt firms to pursue demand-
side alternatives to reduce peak loads.
2.200
1,800
1,400 4.
1.000 4
1977 1982 1997 1902 1997 2002 2007 2012 2017 2022 2027 2032
Aclual less AsIans Actual —50th Percentile —90th Percentile —95th PercentIle
3.Energy efficiency was not adequately considered as a resource.The 2012 Idaho
Energy PLan specificaLLy instructs Idaho and Idaho utiLities to consider energy efficiency as a
resource,as stated on page 9:
When acquiring resources,Idaho and Idaho utiLities shoutd give priority to cost-
effective and prudent:(1)conservation,energy efficiency,and demand response;and
(2)renewabLe resources,recognizing that these atone wilt not futfitt Idaho’s growing
energy requirements and that these resources ptay a rote in addition to conventionat
resources in providing for Idaho’s energy needs.
However,the Resource Alternative AnaLysis,for which the results are presented on page 84
of the 2013 IRP,does not identify conservation,energy efficiency,or demand response as
resources which coutd be compared to the aLternatives presented.In the portfotio of
resources considered in Section 8,energy efficiency was not treated as a scatabte resource.
Instead,it is presented as a reduction in toad.The IRP process shoutd fottow the state
guidetine to give priority to energy efficiency as an atternative when acquiring resources.
4.The Resource Alternatives Analysis of solar PV included poor assumptions.
>The anatysis assumes solar panels exclusively face south,which is tess efficient for
meeting peak demand than facing southwest.Idaho Power onty considered atternatives
in which the utitity,not the customer,owned the sotar panels.This begs the question -
why woutd Idaho Power choose to orient the panets in a manner that makes them tess
cost effective under the methodoLogy Idaho Power uses to evatuate this resource?
>The cost estimates for solar presented in the 2013 IRP are high.Residents and tocat
Idaho sotar instatters are scratching their heads at how high the estimates in the IRP are
retative to current and expected market prices.Even if one tooks at the source data used
by the IRP,The NREL Cost and Performance Data for Power Generation Technot ogles,the
costs appear high.For exampLe,betow is a comparison of the capitat costs which the
7
2013 IRP used for assessing Distributed SoLar PV (page 85)reLative to the capitat costs
presented in the NREL source data,page 38.
The 2013 IRP assumes Plant Capital for Distributed Solar PV is
much higher than the Capital Costs for Residential Solar PV
reported in the NREL source data.
Again,projected decreases are not reflected in the IRP.
Capital Costs $4,340
$1 kW $3,750 $3,460
2025
2013 IRP
Note:The NREL study presents data in constant 2009 dollars while the IRP presents data in
2013 dollars,though this does not explain substantially higher cost estimates used by the IRP
Given the average price of a sotar PV panet has declined 60%since the beginning of 201 1
(source:Sotar Energy Facts:Q2 2013,by SEIA),the substantiatty higher estimates used in
the IRP might be expLained if the 2013 IRP used otd data.In order to adequateLy consider
soLar PV aLternatives,the IRP process shoutd be forward-Looking through use of expected
prices,and shouLd tap the expertise of LocaL professionats and LocaL research.
Solar PV costs are projected to continue declining.The NREL source data used by the
IRP projects decLines in soLar PV.McKinsey &Company pubLished the resuLts of its in-
depth study of solar PV in the 2012 McKinsey QuarterLy,which states:
PV prices are expected to continue to faLL -even though subsidies are expected to dry
up -as manufacturing capacity doubLes over the next three to five years and
underlying costs drop by as much as 10 percent annualLy untiL 2020.Indeed,our
analysis suggests that by the end of the decade,costs couLd decLine to $1 per watt
peak (Wp)for a fuLly instaLLed residentiaL system.
Betow are McKinsey &Companys projections of instatLed system costs of soLar PV:
NREL Study
$
Industrialization will yield significant cost reductions.,
c-SI mulDcrystallIne soIar-photoIta1c system
PoIioon B Modile I C1I •Vfer •Balence of sym (BO
4.0 -
2.6
Po)y[ccn
ci.clh.
Produc1Mty
Prccur.m.nt lncr.m.nt&
Sc&
Cpflrntred
rØam
lrren,.rtu
tchnccqf
4%Proosrnn Imp ernents
I _-.J-’__i_
0pImed
S8tffi 5%
tveized cost ofenergy;assumptions:y%weighted average cost of capita!,annual operations and maintenance equivalent to r%of
system costs 0.9%degradation per year,constant 2011 dollars,;%margin at module level (engineering,procurement,and construction
margin included in BOS costs).
Source:Indusy experts;Photon;GIM Research National Renewable EnergyLaboratory;US Enegynformationkrrniii{tration;
Enerdata;press search;companyWeb sites;McKinsey analysis
The IRP process is unable to vatidate what alternatives most cost effectiveLy meet customer
needs given it did not objectiveLy consider the current and dectining costs of alternatives
and the vatue of flexibility to adapt to actuat demand,stage investments,and take
advantage of technotogy improvements.
C.Given increased uncertainty,new approaches are needed
The energy sector is facing new tevets of uncertainty,thus the planning process needs to
adapt.Betow are a few notes on how to do so.
1.The process should value flexibility.When analyzed purely on costs,bulk buys and
lumpy”investments typically took attractive due to the economies of scale.However,the
ability to stage investments to meet demand,the ability to avoid unnecessary investments
when forecasts go awry,the abitity to leverage improvements in technology,and the abitity
to reduce the tevet &cost of capacity ptanning margin alt have vatue.
2.New approaches to valuing storage should be adopted.Avista summarized this
challenge and opportunity in its 2103 IRP,section 6-10:
One of the biggest obstacles to energy storage is quantifying and property vaLuing its
benefits.At a minimum,the vatue of storage is the spread or difference between the
Best-in-class installed system cost (no margins)
$per watt peak,2011 dollars
Levelized cost of electricity1
$per kilowatt hour 2011 dollars
2.8
2.4
2,0
1,6
1.2
0-s
0.35
0.34
0.28
0.26
0,24
0.20
0,18
0.16
0.14
012
0.10
0.08
n
2011—15 2016-20
9
vatue of energy in on versus off-peak hours (toad factoring),minus the tosses.Since
the technotogy can meet reguLation,toad foLlowing,and operating reserves,there is
vatue beyond toad factoring.Vatuing these benefits requires new system modeling
toots.
The Idaho Power IRP process would atso benefit from new system modeting toots for vatuing
storage.
3.The process should seek to understand the conditions in which alternative paths
forward would better serve customer needs.As the comptexity of variables and range of
uncertainties grow,firms often use their ptanning process to anatyze what conditions woutd
have to occur to make one path forward better than another.If,for example,the carbon
adder were the onty significant variabte,the 2013 IRP’s effort to identify the trigger point at
which solar PV were favorabte woutd be usefut.As described eartier,there are numerous
other variables which affect the uttimate cost effectiveness of atternatives -dectining PV
costs,storage technotogies,demand management,the abitity to detay investment in
unnecessary capacity,the retiabitity of demand forecasts,etc.The IRP process shoutd think
more broadty and consider what-woutd-it-take for the resource atternatives prioritized by
state policy to be viabte.
4.The process should be a venue for seeking out better options.The above what-woutd
it-take approach is not onty a toot for vatuing atternatives in the face of uncertainty but also
for buitding understanding and for innovating better solutions.The 2012 Idaho Energy plan
encourages us to consider these opportunities and maintain Itexibitity to Leverage them:
The PLan aLso encourages monitoring new energy technoLogies and cooperation
between stakeholders to altow us to more quickLy respond to possibLe opportunities
and risks.
Rather than omit discussion of transformative changes in the energy sector,our IRP process
couLd become a venue for professionals within and outside of Idaho Power to buiLd
understanding of those changes and to explore opportunities for att stakehotders to benefit
from them.
5.The process and financial incentives should adapt to enable more objective
consideration of alternatives.As an investor owned utiLity,Idaho Power has an
accountabitity to optimize its return on capitaL.If the financial incentives avaitabte to the
company favor investments in infrastructure,it is chaLLenging for the company to be
objective in its consideration of aLternatives.This probLem is magnified by the increasing
compLexity and subjectivity of planning for an uncertain future.I betieve that Idaho Power
and other Idaho stakehotders have the capacity to be Leaders in exptoring incentive
structures that enabte both the company and its ratepayers to benefit from more objective
pursuit of atternative resources.
Because the IRP process did not adequatety identify or vatue atternative paths forward,the
need for ditigent and open-minded consideration of alternatives is paramount before making
further commitments to specific resource strategies,such as the proposed investments
which presume a continued retiance on operation of the Bridger faciLities.
10