20151207Comment.pdf

Introduction
Avista Utilities has requested the regulatory authority to launch a natural gas deind-se
management (DSM)portfolio and recover the anticipated $1.25 million program expense
through a Schedule 191 surcharge on retail rates.The Company has also proposed revisions to
the Schedule 190 tariff governing the operations of that proposed DSM portfolio.The Company
had previously withdrawn its natural gas DSM portfolio,with Commission approval,due to the
substantial reduction in natural gas avoided cost.At that time the Company represented that
the reduction in the avoided cost rendered it impossible to deliver the cost-effective
performance necessary to benefit consumers.
There are several issues raised by the Company’s proposal that bear closer scrutiny to include:
•The use of an avoided cost forecasts that is substantially higher than the expected case
avoided cost scenario and the medium case avoided cost with carbon legislation
represented by Avista in its most recent (2014)natural gas Integrated Resource Plan’.
•The Company proposes to apply a real (inflation adjusted)discount rate to a nominal
avoided cost forecast.
•The Company’s calculation of both the Utility Cost Test (UCT)and Total Resource Cost
(TRC)test are not compliant with industry standards for such metrics.
•The Company projects the performance of the proposed Idaho natural gas DSM
performance to be far higher than the independently evaluated performance of the
same portfolio within its Washington jurisdiction.
•A brief review of the Company’s supporting work papers has led to the discovery of a
number of spreadsheet errors.
The comments to follow outline some of the major analytical and policy issues raised by
Avista’s filing and conclude with a recommendation for denying the Company’s request
pending a more detailed analytical and policy discussion among a broader cross-section of
interested parties.
Revisions to the Avoided Cost Stream
Avista has chosen to base its cost-effectiveness projections upon an avoided cost stream that is
substantially higher than that contained within its 2014 natural gas IRP,both with and without
the assumption of carbon legislation2.The IRP process provides an opportunity for a fully
transparent and comprehensive analysis of supply and demand-side resource cost with
provisions for public involvement through a Technical Advisory Committee.The IRP allows all
resources to compete on a level playing field so as to lead to a selection of an optimal resource
mix.
In bypassing this process and substituting drastically higher avoided cost values the Company is
unnecessarily creating a biased criterion for resource selection.There have not been any
changes in natural gas markets of such a magnitude so as to justify deviating from the
Commission recognized 2014 IRP results.
In addition to the revision to the core avoided cost forecast itself the Company has added an
estimate of future total (fixed and variable)capacity costs based upon a calculation for
allocating average capacity costs for ratemaking purposes.This approach is inconsistent with
the incremental nature of the calculation of avoided cost stream.This is a substantial change
from accepted approaches for resource planning,DSM business planning and for DSM cost-
effectiveness evaluation.
In addition to the procedural and methodological faults noted above,the Company’s estimate
of avoided cost contains several spreadsheet reference errors3 that:
1.Utilize the Company’s higher winter avoided cost forecast in the calculation of the
annual avoided cost.
2.Layered upon the Company’s revised avoided cost forecast is an estimate of future
monetized carbon costs that are represented as a 2%escalation,but the calculations
actually utilize a 3%annual escalation.
3.Two different discount rates are applied at various points within the calculation of
avoided cost.
4.The avoided cost used in the calculation of cost-effectiveness is not equal to the revised
avoided cost calculation described by the Company and explained within the work
papers accompanying the filing.
The net effect of the revisions outside of the IRP process has been to artificially increase the
cost-effectiveness calculations of the Company’s proposed DSM portfolio.
Applicable Discount Rate
Avista is also proposing to apply a real (inflation adjusted)discount rate to future costs and
benefits.This lower discount rate significantly increases the cost-effectiveness of DSM
programs since measure costs are incurred almost exclusively up-front while the benefits
(primarily the avoided cost of energy)flow over the life of the measure.
However,the Company is proposing to apply the inflation-adjusted real discount rate to a
nominal stream of avoided costs that has not been adjusted for inflation.This incongruity is
without rational justification.The Company’s work papers indicate that the 6.80%discount
rate4 will be reduced to 4.O2%.The application of a real discount rate to a nominal avoided
cost stream is an inconsistency that will again artificially inflate the cost-effectiveness of a DSM
portfolio.
It is also notable that the Company’s present value of avoided cost streams are calculated by
applying the discount rate to beginning-of-year values rather than more conventional end-of
year application.While a deviation from the end-of-year industry standard to a mid-year
discounting convention might be justifiable,a beginning-of-year assumption presumes that the
annual savings from a measure occurs entirely on the first day of each year.This is an
unjustifiable deviation from typical analytical practice creating another upward bias to the cost
effectiveness calculation.
The Utility Cost Test and Total Resource Cost Test Standard
Four basic cost-effectiveness metrics have been standardized for the evaluation of DSM
programs.Specific analytical methodologies have been most notably documented by the
National Action Plan for Energy Efficiency (NAPEE)and the California Energy Commissions
Standard Practice Manual.Each of the four cost-effectiveness tests view DSM programs from a
different perspective.
The UCT test is designed to measure the utility cost to provide energy to the consumer.The
scope of the UCT is restricted to only those costs that would appear on the consumer utility bill.
The incremental cost of the promoted measure borne directly by the customer is not within the
scope of the UCT test.In Avista’s most recent independently reviewed assessment of its
Washington natural gas DSM program,those costs borne directly by the customer are over
twice as high (221%)as the total cost of running the utility program6.
If the objective of a DSM portfolio were to minimize the consumer utility bill without regard to
any other consumer cost the UCT test would be the appropriate metric.But a favorable ratio of
benefits to costs within the UCT does not in any way indicate that the consumers funding the
DSM portfolio will benefit from their investment.
In contrast the TRC test is a metric designed to more comprehensively assess the impact of a
DSM program upon the cost of end-use services borne by the ratepayer population,whether
these costs are remitted through surcharges levied upon the consumer by the utility or in the
form of premiums paid for the equipment or services promoted by the program.Thus a
favorable ratio of TRC benefits to TRC costs indicates that the ratepayer population will pay less
for end-use services with due consideration to all benefits and costs.Similarly a portfolio that is
not TRC cost-effective represents an increase in the customer cost of end-use services.For that
reason the Northwest Power and Conservation Council utilizes essentially the same test as its
sole criteria for DSM cost-effectiveness within its resource planning process.
The Company has referenced that the TRC test includes a valuation of non-energy benefits that
may occur with the adoption of the promoted measure (e.g.increased comfort in comparison
to the baseline measure).The Company failed to note that the TRC test also includes the value
of non-energy costs as well (e.g.discomfort that may be associated with the measure).The net
value of these costs and benefits are a small component of the TRC test for natural gas DSM
programs,much smaller than witnessed in electric DSM programs.Difficulties experienced in
quantifying non-energy impacts may lead to a very modest upwards or downwards bias the
results of the TRC test,but they are important considerations that are not incorporated within
the UCI test and are no reason to abandon the expectation that a utility DSM program should
pass the TRC test.
The Company’s filing references a “...draft revised Memorandum of Understanding ...“with
Staff in referencing its proposal to launch a portfolio which is not TRC cost-effective.There is
no such publically available document,but Avista’s website does contain a Memorandum of
Understanding on the subject of cost-effectiveness tests that states that “...all programs and
individual measures should have the goal of cost-effectiveness from the total resource,utility
and participant perspectives”7.That latter document was signed by Staff,the Idaho Power
Company,Rocky Mountain Power and Avista.This represents a regulatory approach that is
more beneficial to customers.
Calculation of Cost-Effectiveness Metrics
The Company’s methodologies for calculating the UCT and TRC cost-effectiveness metrics are
not compliant with accepted industry methodologies.These industry standard calculations
require the analysis to distinguish between participants who were motivated to adopt a
measure as a result of the program from those who would have adopted the measure even in
the absence of the program.The latter is termed natural adoption and both the customer costs
and benefits of these program participants should be excluded from the cost-effectiveness
calculation.This adjustment is performed by applying a net-to-gross (NTG)factor to the total
program participation.
The National Action Plan for Energy Efficiency (sponsored by the U.S.Environmental Protection
Agency)specifies that:
“A key requirement for cost benefit analysis is estimating the NIG.The NTG adjusts the
cost-effectiveness results so that they only reflect those energy efficiency gains that are
attributed to,and are the direct result of,the energy efficiency program in question.It
gives evaluators an estimate of savings achieved as a direct result of the program
expenditures by removing savings that would have occurred even absent a conservation
program.Establishing the NTG is critical to understanding overall program success and
identifying ways to improve program performance”.8
Similarly the October 2001 California Standard Practice Manual states that for both the UCT and
TRC test:
“The avoided supply costs should be calculated using net program savings,savings net
of changes in energy use that would have happened in the absence of the program”.9
Similarly the same document specifies that the Ratepayer Impact Measure (RIM)test calls for:
“Both the reductions in supply costs and the revenue increases should be calculated
using net energy savings”.’0
Avista’s cost-effectiveness calculations do not include this required net-to-gross adjustment.
Excluding such an adjustment overstates the cost-effectiveness of the UCT,TRC or RIM test.
The degree of the overstatement is particularly acute in the case of the UCT test.Analysis
lacking such an adjustment should not be referred to as either a UCT or TRC calculation.
Comparison to Past Performance
The Company currently offers a natural gas DSM portfolio within its Washington jurisdiction.
Given Avista’s longstanding practice of offering identical programs within its contiguous service
territories it is reasonable to conclude that a prospective Idaho portfolio would perform very
similarly to the existing Washington portfolio.
Avista’s Washington natural gas DSM portfolio has been subject to an external independent
evaluation of acquisition of actual operations during the 2012-2013 program years (the most
recent independently reviewed period).The results of this independent analysis of actual and
verified portfolio performance led to a TRC benefit-to-cost ratio of 0.351;(ratepayer benefits
being only 35%of costs)and a UCT benefit-to-cost ratio of 0.8012 (indicating that the utility
portfolio benefits were 80%of the reduction in the utility cost).
The proposed Idaho portfolio would be expected to perform similarly.The difference between
the cost-effectiveness represented by the Company in its filing and the poor performance of
the Washington portfolio is indicative of the magnitude of the revisions that the Company has
made to previously accepted analytical methodologies.
Schedule 190 Tariff Provisions
The Company has proposed updated language for Schedule 190,governing natural gas DSM
operations.Aspects of this tariff would qualify projects which are definitively cost-ineffective.
Specifically the proposed Schedule 190 tariff grants the authority and obligation to extend an
incentive of $3.00 per first-year therm on any qualifying project with a consumer simple
payback of up to 15 years (and an even longer discounted payback).The Company is proposing
to open eligibility to all projects with measure lives of as short as ten years.Even in the
absence of any utility cost associated with the administration of the DSM program,measures
which require 15 years to obtain a simple payback but with a measure life of only ten years are
clearly cost-ineffective.
Conclusion
Avista’s request to impose a $1.25 million surcharge on Idaho natural gas consumers is based
upon a flawed analysis that includes a non-standard calculation of the UCT,TRC and RIM test
4’
that improperly credits the naturally occurring adoption of efficiency measures to the utility
DSM program,confuses the allocation of total costs for rate recovery purposes with the
calculation of incremental avoided costs for resource planning purposes,incorporates
substantial changes to the avoided cost outside of the IRP process and public opportunity for
involvement,mixes real discount rates and nominal avoided costs and includes several
calculation errors.Avista has projected UCT and TRC performance that is well over twice that
observed in its comparable Washington natural gas DSM portfolio,based upon an independent
evaluation of actual results.
Given the significant policy and analytical questions raised by this filing and in recognition that
there is little (likely no)harm to a reasonable delay we suggest that the Commission deny
Avista’s application to impose a surcharge to fund natural gas DSM programs through Schedule
191 and the related revisions to Schedule 190 pending the opportunity for public review,comment
and discussion by all interested parties.
Footnotes
1.The “expected case avoided cost”scenario is found in Appendix 5.4,page 141 of the 2014
Avista Natural Gas Integrated Resource Plan Appendix.The “carbon legislation —medium
case avoided cost”is found in Appendix 5.4,page 144 of the 2014 Avista Natural Gas
Integrated Resource Plan Appendix.
2.See footnote 1 above.
3..Specific spreadsheet cell references of these errors are available upon request.
4.Page 16 of the Avista Washington /Idaho Demand-Side Management Business Plan.
5.Page 16 of the Avista Washington /Idaho Demand-Side Management Business Plan as
well as the Company’s work papers supporting this filing.
6.Washington UE-140801 Exhibit BWF-3 summarizes the results of Avista’s 2012-2013
Washington DSM portfolio.Per the summary of cost-effectiveness of the natural gas DSM
program for that period the customer incremental cost of promoted measures was
$18,459,614 in comparison to the $8,335,517 total utility program cost (including both
incentive and non-incentive funding).
7.The Company has made this MOU available at
https://www.avistautilities.com/savings/dsm/dsmhistory/Documents/ldaho%2OPUC%20D
SM%20M0U.pdf
8.National Action Plan for Energy Efficiency “Understanding Cost-Effectiveness of Energy
Efficiency Programs;Best Practices,Technical Methods and Emerging Issues for Policy
Makers”,page 4-9
9.The TRC test reference is at page 18 of the October 2001 California Standard Practice
Manual.Reference to the UCT test is contained on page 23 of the same document.
10.Page 13 of the October 2001 California Standard Practice Manual.
11.Washington UE-140801 Exhibit BWF-3,page 3.
12.Washington UE-140801 Exhibit BWF-3,page 3.
13.Per the cost-effectiveness summary contained in Washington UE-140801 Exhibit BWF-3
page 3,the actual natural gas DSM portfolio TRC benefit-to-cost ratio was 0.35 vs.the
Company’s projected 0.93 benefit-to-cost ratio (2.6 times higher).Similarly the same
Washington document references a 0.80 UCT benefit-to-cost ratio in comparison to a 2.02
projection in the Company’s filing (2.5 times higher than the actual Washington
performance).
Jonathan Powell
Spokane,WA 99208
powellacres ©aol.com