20081204Goins Rebuttal.pdf

Department of Energy
Washington, DC 20585
R., EC,.EfHI:'"" ,.., ì)
20080EC -4 AM 10: 03
IDAHO PUBLiC
UTILITIES COMMISSiON
December 3, 2008
VIA OVERNIGHT SERVICE
Ms. Jean Jewell
Commission Secretary
Idaho Public Utilities Commission
472 WWashington
Boise, ID 83720-0074
RE: Case No. IPC-E-08-10
Dear Ms. Jewell:
Enclosed please find:
(i) an original and 10 copies of the Rebuttal Testimony of Dr. Dennis W. Goins on behalf
of the United States Deparent of Energy in the above-captioned proceeding;
(2) an additional copy of this testimony, that I request be date-stamped and retued in
the enclosed postage paid envelope;
(3) a disk upon which ths testimony is set out in computer searchable form.
If you have any questions concerning this fiing, please contact me at [phone].
Sincerely yours,~Artur Perry B tier
Attorney for the United States
Departent of Energy
Offce of the General Counsel
United States Deparent of Energy
1000 Independence A venue SW
Washington, D.C. 20585
Artur.Bruder(ihq .doe.gov
[phone]
(1 Pnnt wi so Ink on recycled par
CERTIFICATE OF SERVICE - IDAHO PUC CASE NO. IPC-E-08-10
I hereby certfy that, this 3d day of December, 2008, I have served or caused to be served
a true and correct copy of the attached Testimony of Dr. Dennis W. Goins on behalf of
the United States Departent of Energy upon each of the individuals listed below, by: (I)
placing the same in the United States Mail, addressed to each of the persons listed below
at the street address set out below; (2) electronically trnsmitting the same to each of the
persons named below at the email address set out below; (3) sending an original and ten
(i 0) copies of the same via Federal Express to the Secretary of the Commission.
Ms. Jean Jewell, Secretary
Idaho Public Utilties Commission
472 W. Washington
Boise, ID 83702
jean.jeweii~puc.idaho.gov
Barton L. Kline
Lisa D. Nordstrom
Idaho Power Company
1221 W. Idaho St. (83702)
P.O. Box 70
Boise, ID 83707-0070
bkl ine(idahopower. com
Inordstrom(iidahopower .com
JohnR. Gale
Vice President, Regulatory Affairs
Idaho Power Company
1221 W. Idaho St. (83702)
P.O. Box 70
Boise, ID 83707-0070
rgaie~idahopower.com;
Weldon Stutzman
Neil Price
Deputy Attorneys General
Idaho Public Servce Commission
472 W. Washington (83702)
PO Box 83720
Boise, il 83720-0074
weldon.stutzan~puc.idaho.gov
neiL.price(ipuc.idaho.gov
Peter J. Richardson
Richardson & O'Lear
515 N. 27th St.
P.O. Box 7218
Boise, ID 83702
peter~richardsonandoleary
~~
~
,i:
~S..oc.
;0rn('rn
:2rn(~
Dr. Don Reading
Ben Johnson Associates
6070 Hil Road
Boise, il 83703
dreading~indspring.com
Randall C. Budge
Eric L. Olsen
Racine, Olson, Nye, Budge &
Bailey, Charered
P.O. Box 1390
201 E. Center
Pocatello, il 83204- i 39 I
rcbCfracinelaw. net
elo(iracinelaw.net
Anthony Yankel
29814 Lake Road
Bay Village, OR 44 I 40
yanel~attbi.com
Michael Kur, Esq.
Kurt J. Boehm, Esq.
Boehm, Kurt & Lowry
36 E. Seventh Street, Suite 15 iO
Cincinnati, OH 45202
mkurt~BKLlawfirm.com
kboehm~ BKLlawfirm.com
Conley E. Ward
Michael Creamer
Givens Pursley LLP
601 W. Bannock Street
POBox 2720
Boise, ID 83701-2720
cew(igivenspursley.co
Dennis E. Peseau, Ph.D.
Utility Resources, Inc.
1500 Liberty Street, Suite 250
Salem, OR 97302
dpcseau(icxcitc.com
-2-
Brad M. Purdy
A ttorney at Law
2019 N.17thSt.
Boise, Idao 83702
bmpurdy(§hotmail.com
Ken Miler
Snake River Alliance
Box 1731
Boise, ID 8370 I
kmiler~snakeriverai lance.org
Kevin Higgins
Energy Strategies LLC
Parkside Towers
215 South State Street
Suite 200
Salt Lake City, UT 84111
khiggins(§energystrat.com
~-)
Arhur Perry Bru ~sq.
Office of the General Counsel
United States Deparent of Energy
Washington, DC 20585
[phone]
-3-
RECEIVED
Arthur Perry Bruder (admitted pro hac vice)
1000 Independence Ave. SW
Washington, D.C. 20585
phone: [phone]
FAX: [phone]
arthur .brudertWhg. doe .gov
Attorney/Representative for the
United States Department of Energy
2009 DEC -4 AM 10: 03
IDAHO PUfe'UC
UTILITIES C01IHIiISSION
STATE OF IDAHO
BEFORE THE
IDAHO PUBLIC UTILITIES COMMISSION
CASE NO. IPC-E-08-10
IN THE MATTER OF THE APPLICATION OF
IDAHO POWER COMPANY
FOR AUTHORITY TO INCREASE ITS RATES AND
CHARGES FOR ELECTRIC SERVICE TO ELECTRIC
CUSTOMERS IN THE STATE OF IDAHO
REBUTTAL TESTIMONY OF
DR. DENNIS W. GOINS
ON BEHALF OF THE
U.S. DEPARTMENT OF ENERGY
December 3, 2008
TABLE OF CONTENTS
Page
INTRODUCTION ................................................................................................................... 1
LIP A WITNESS Y ANKEL ..................................................................................................... 2
STAFF WITNESS HESSING .................................................................................................. 9
Case No. IPC-E-08-10
Dennis W. Goins - DOE. Reb
Pagei
2 Q.
3
4 A.
5
6
7 Q.
8
9 A.
10 Q.
11 A.
12
13
14
15
16
17 Q.
18 A.
19
20
21
22
23
24
INTRODUCTION
PLEASE STATE YOUR NAME, OCCUPATION; AND BUSINESS
ADDRESS.
My name is Denns W. Goins. I operate Potomac Management Group, an
economic and management consulting firm. My business address is 5801
Westchester Street, Alexandra, Virgina 223 I O.
DID YOU FILE DIRECT TESTIMONY IN THIS CASE ON OCTOBER 28,
2008?
Yes.
ON WHOSE BEHALF AR YOU TESTIFYING IN THIS PROCEEDING?
I am testifying on behalf of the U.S. Deparent of Energy (DOE) representing
the Federal Executive Agencies (FEA), which is comprised of all Federal facilities
served by Idaho Power Company (IPC). Two of the larger FEA facilties are the
Deparent of Energy's Idaho National Laboratory (DOE) and Mountain Home
Air Force Base. IPC serves DOE under a special contract, and serves the bulk of
Mountain Home AFB' s load under Schedule i 9 Large Power Service.
WHAT IS THE PURPOSE OF YOUR REBUTTAL TESTIMONY?
The purose of my rebuttl testimony is to respond to certain conclusions and
recommendations made in direct testimony fied by Anthony Yanel on behalf of
the Idaho Irrgation Pupers Association, Inc. (IIP A) and Keith Hessing on behalf
of the Staff of the Idaho Public Utilities Commission. In particular, I wil respond
to:
1. IIP A witness Yankel s attempt to recognize the impact ofload growth in
IPC's cost-of-service analysis. As I discuss later, witness Yankels effort
Case No. IPC-E-08-10
Dennis W. Goins - DOE -Reb
Page 1
2
3
4
5
6
7
is essentially a roundabout way to argue for vintage pricing of IPC's
assets-that is, assignig entitlement to older, lower-cost assets to existing
loads and then charging new loads for the higher marginal cost of capacity
additions. Vintage pricing is both arbitra and economically ineffcient,
and should be rejected. Witness Yankel's vintage pricing adjustments to
IPC's cost-of-servce methodology suffer from the same deficiencies and
should be rejected.
8
9
10
11
12
13
14
15
16
17
2. Staff witness Hessing's uncritical adoption of IPC's 3CP/12CP cost-of-
service methodology. He relies on results from IPC's cost study to
develop his recommended revenue spread based on the Staff s proposed
revenue requirement. However, the cost study on which he relies ignores
numerous deficiencies in IPC's costing methodology that I identified in
my direct testimony. Because witness Hessing relies on an improper and
unreasonable allocation of IPC's costs, his recommended revenue spread
does not properly track IPC' s actual cost of serving retail customers in
Idaho. His recommended higher-than-average rate increases for higher
load factor' classes and special contract customers should be rejected.
18 LIP A WITNESS YANKEL
19 Q.
20
PLEASE DESCRIBE WITNESS Y ANKEL'S PRIMARY CRITICISM OF
IPC'S COST-OF-SERVICE METHODOLOGY.
21
22
A.Witness Yankel presents data showing that loads for the irgation class have
grown veiy little in the past 25 years relative to load growth for other classes. He
1 Load factor refers to the ratio of a customer's average demand to peak demand during a specified period.
For example, if a customer uses 2,190 kWh per month, the customer's average demand is 3 kW in a typical
month (2,190 kWh divided by 730 hours per month). If the customer's maximum monthly peak demand is
10 kW, the customer's monthly load factor is 30 percent (3 kW divided by 10 kW). We generally classify
low load factor customers as those with load factors below the system average load factor, and high load
factor customers as those with load factors above the system average load factor. IPC's annual system load
factor typically falls between 55 and 60 percent.
Case No. IPC-E-08-10
Dennis W. Goins - DOE - Reb
Page 2
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15 Q.
16
17 A.
18
19
also shows that during this 25-year period, IPC's plant-in-service has more than
doubled to meet load growth, leading to signficant rate increases to pay for the
additional costs of IPC' s expanding asset base. He then concludes that customers
whose load growth caused the need for additional capacity should pay for the
resulting higher cost of service. According to witness Yankel, because irrgation
loads have grown little in 25 years, irrgation customers should bear little if any of
the higher cost of new capacity (generation, transmission, and distribution) to
meet load growth. He criticizes IPC's costing methodology because, in his
opi11on:
.. .the Company's cost of service study inappropriately allocates a
significant portion of this growth to the Irrgation class. Given the
obvious fact that growth and the cost of growth are not being fueled by
the Irrgators, the allocation of significant portions of the cost of ths
growth to the Irrgators is on its face counter-intuitive.2
HOW DOES WITNESS YANKEL PROPOSE TO FIX THIS ALLEGED
PROBLEM?
He proposes modifying IPC's cost-of-service methodology to address what he
calls backward-looking costs and forward-looking costs. His direct testimony
addresses his proposed solution using these two cost concepts:
2 See the direct testimony ofIIA witness Anthony J. Yanel at 9:9-12.
Case No. IPC-E-08-10
Dennis W. Goins - DOE - Reb
Page 3
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19 Q.
20
21
22 A.
23
24
The simplest way to correct the Company's Base Case study would be
to continue to define "backward-lookig" costs base on test year usage
levels and "forward-lookig costs" at the anticipated increase in usage
levels in the Company IRP. The "backward-lookig costs" would
simply be costs as they exist today and allocated on the basis of to day's
energy or 12-CP as is presently done in the Company's "base" cost of
service study. The "forward lookig costs" would be developed using
the same weighting factors used by the Company associated with the
cost of the anticipated growth, but would be allocated on the basis of
only the growth that is anticipated from each rate schedule over a futue
ten year period. The relative share of historic costs and anticipated
costs related to growth would then be averaged using the Company's
existing procedures in order to develop a composite allocation factor
for use in spreading test year costs for allocation puroses. In ths
manner, the methodology would be exactly the same as the Company's
Base Case, but the marginal costs would be tied to the marginal/new
usage and not to the present level (status quo) ofusage.3 (Emphasis in
originaL.)
DOES IPC'S COSTING METHODOLOGY FAIL TO ASSIGN COSTS
PROPERLY TO REFLECT FACTORS DRIVING ITS NEED FOR NEW
CAPACITY?
Yes. However, witness Yankel does not correctly identify deficiencies in IPC's
costing methodology that cause the problem with respect to the allocation of
production costs, nor does he propose a reasonable and proper solution. I agree
3 Ibid. at 19:13 - 20:2.
Case No. IPC-E-08-10
Dennis W. Goins - DOE - Reb
Page 4
1
2
3 Q.
4
5 A.
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23 Q.
24 A.
25
with him that IPC's recommended costing methodology is seriously deficient.
But I disagree regarding how to fix IPC's methodology.
is WITNESS YANKEL'S CRITICISM OF IPC'S COST STUDY BASED
ON A VALID PREMISE?
No. Witness Yanel implicitly assumes that if a class has little or no load growth,
then it also should have little if any growth in costs assigned to it. This
assumption ignores a fundamental tenet of cost of service-namely, costs should
be assigned using allocation factors that reasonably link costs and cost-drvers.
Consider IPC's production costs. As I noted in my direct testimony, the key
driver underlying IPC's need for new production resources (both new capacity and
expensive purchased power) is peak demand in summer months. As a result, a
reasonable cost-of-service methodology should ensure that the allocation of these
higher sumer-related costs should be linked to sumer peak demands. That is,
customer classes that use electricity primarily in sumer peak months-
regardless whether their loads have grown or remained stable in the past 25
years-should expect to see significantly higher rates as IPC adds production
resources. Instead, as my direct testimony shows, IPC's classification and
allocation of production costs focuses on year-round average demands and energy
usage, not summer peak demands. Not surrisingly-albeit incorrectly, IPC's
costing methodology implies huge rate increases for high load factor classes, yet
little if any increase for most lower load factor classes that contrbute heavily to
summer peak demands relative to their demands in off-peak months.
DOES ALL LOAD GROWTH CAUSE IPC TO ADD CAPACITY?
No. In tring to lin load growth to cost assignent, witness Yankel ignores the
link between the timing of electrc loads and IPC's need for new production
Case No. IPC-E-08-10
Dennis W. Goins - DOE - Reb
PageS
1
2
3
4
5
6
7
8
9
10
11
12 Q.
13
14 A.
15
16
17
18
19
20
21
22
23
resources. For example, demands in peak periods drive capacity requirements,
while load growth in off-peak hours may have little if any effect on IPC's need for
new production capacity. In criticizing IPC's costing methodology, witness
Yankel never mentions that the bulk of the irrgation class' annual kWh usage and
highest class peaks occur in peak hours in peak sumer months. In fact, to the
extent that IPC's costing methodology understates production costs attrbutable to
summer peak demands, costs assigned to the irrgation class are understated-not
overstated at witness Yankel contends. Ths result was confirmed in the cost-of-
service analyses presented in my direct testimony, which indicated that the rate
increase for the irgation class should be significantly greater than the increase
proposed by IPC.
is WITNESS YANKEL'S PROPOSED FIX WORSE THAN THE
ALLEGED PROBLEM IT is DESIGNED TO CURE?
Yes. His proposed fix uses an average of growth-adjusted forward-looking costs
and historical backward-looking costs to allocate test-year costs. This scheme is
nothing more than a variant of vintage pricing, which attempts to insulate classes
with little or no load growth from any cost responsibility for new capacity-even
if their peak demands coincide with other demands that are driving the need for
new capacity. Instead of resorting to vintage pricing, the cure for the alleged
problem that witness Yankel describes is twofold:
. Properly classify and allocate production costs to emphasize sumer
peak demands as the drving force behind IPC's recent and planed
production resource additions.
Case No. IPC-E-08-10
Dennis W. Goins - DOE - Reb
Page 6
1
2
3
4
5 Q.
6 A.
7
8
9
10
11
12
13
14
15
16
17
18
19 Q.
20 A.
21
22
23
. Shift load to off-peak periods to avoid contributing to system peak
demands. Simply, stated, cost responsibility assigned to irrgators in
any properly designed cost-of-service study wil decline as irgators
shift more load to off-peak hours.
is VINTAGE PRICING ECONOMICALLY EFFICIENT?
No. Vintage pricing assumes specific customers have entitlement to lower cost
assets simply by virte of when they became market participants. Under witness
Yankel's scheme, the higher marginal cost of new capacity additions is assigned
primarly to classes with significant load growth, while no-growth classes are
primarily assigned much lower historical embedded costs. Charging some
customers prices based on marginal capacity costs while charging other customers
using the same capacity prices based on historical embedded costs is both
economically ineffcient and discriinatory.Moreover, vintage pncing
encourages no-growth classes to use more energy in high-cost peak periods.
Under witness Yankel's scheme, irrgation customers would be encouraged to use
more energy in sumer peak periods, even while ¡PC is promoting its Peak
Rewards program to encourage irrgation customers to shift loads to off-peak
hours.
HAS VINTAGE PRICING BEEN LARGELY DISCREDITED?
Yes. Vintage pricing has generally been recognized as bad regulatory policy. 4 For
example, in a 1993 report in which it discussed whether new or existing
customers should be primarily responsible for the cost of transmission capacity
additions, the National Regulatory Research Institute (NRR) stated:
4 I do not consider direct cost assignments (for example, directly charging customers for the incremental
cost of interconnection upgrades that a utility does not normally provide) as a form of vintage pricing.
Case No. IPC-E-08-10
Dennis W. Goins - DOE - Reb
Page 7
2
3
4
5
6
7
8
9
10
11
12
13
14 Q.
15
16 A.
17
18
19
20 Q.
21 A.
22
23
24
. . .In most cases, transmission line upgrades are likely to be constrcted
not only to serve the curent applicant, but also to serve projected futue
applicants, and to provide more reliable service to existing wholesale,
retail, and transmission customers. FERC might choose to protect
existing wholesale, retail, and transmission customers at all costs. This
would result in a system of vintage pricing, with existing customers
paying a depreciated embedded cost of old plant and new customers
paying the full incremental cost of new plant. One outcome of such
pricing is that the old customers would benefit, enjoying increased
reliability and the opportty to increase their own wholesale or retail
purchases or transmission service without paying any part of the cost
of service of the new plant. Such vintage pricing makes for bad
economics, whether or not practicable or feasible.5 (Emphasis added.)
DID WITNESS YANKEL CONDUCT A COST STUDY THAT
INCORPORATED HIS PROPOSED FIX?
Yes. He prepared a cost study that included growth-adjusted allocation factors for
generation and transmission costs but not distribution costs.6 As expected, his
study generally implies rate increases similar to those proposed by IPC except for
the irrgation class.
DO YOU AGREE WITH THE RESULTS FROM HIS COST STUDY?
No. Even if one agreed with the premise of witness Yanel's growth-adjusted
cost study, one could not agree with his study's results because of a serious error
he introduced in his analysis. Specifically, in developing his growth-adjusted
allocation factors, witness Yankel used annual class energy growt instead of
5 Kenneth W. Costello et ai., A Synopsis of the Energy Policy Act of 1992: New Tasksfor State Public
Utility Commissions, The National Regulatory Research Institute, Columbus, Ohio, June 1993 at 31.6 See Yankel direct testimony at Exhibit No. 302.
Case No. IPC-E-08-10
Dennis W. Goins - DOE - Reb
Page 8
1
2
3
4
5
6
7 Q.
8
9 A.
10
11
12
13
14
15
16
17
18
19
20 Q.
21
22 A.
23
sumer peak demand growth-the priar drver underlying IPC's need for
additional production resources. Moreover, he did not to correct the classification
and allocation errors embodied in IPC's costing methodology, thereby ensurng
that his results did not reflect proper links between customer demands and cost
allocation. The Commission should reject his growth-adjusted cost-of-service
study and associated revenue spread.
SHOULD A COST STUDY ATTEMPT TO ADDRESS THE IMPACT OF
LOAD GROWTH ON RISING COSTS?
Yes. But such attempts should be direct, transparent, and economically rationaL.
Assigning asset entitlements to specific customers or classes through vintage
pricing is not a proper mechanism to address load growth in a cost study. (Neither
is the convoluted blending of average and marginal cost concepts as occurs in
IPC's 3CP/12CP cost study.) Instead, in an embedded cost analysis, load growth
can be properly addressed by linkg as closely as possible allocation factors to
cost drivers. For example, if sumer peak demands are principal drivers behind
rising production costs, then fixed production cost should be classified as
demand-related costs and allocated using schemes that emphasize sumer peak
demands-not energy usage (thnk IPC's 3CPIl2CP methodology).
STAFF WITNESS HESSING
DOES STAFF RECOMMEND IPC'S PROPOSED 3CP/12CP COST-OF-
SERVICE METHODOLOGY?
Yes. Staff witness Hessing recommends that the Commission adopt IPC's
3CPIl2CP methodology.?
? See the direct testimony of Staff witness Keith Hessing at 8:25 - 9: 1.
Case No. IPC-E-08-10
Dennis W. Goins - DOE - Reb
Page 9
1 Q.
2
3 A.
4
5
6
7
8
9
10
11
12
13
14
15 Q.
16 A.
17
18
19
20
21
22
23
24
WHAT is THE COMMON THEME BETWEEN COST-OF-SERVICE
POSITIONS TAKEN BY WITNESSES YANKEL AND HESSING?
Both witnesses-consistent with IPC's witness Tatum-recommend
disproportionate allocations of production-related costs to loads occurng in non-
sumer months and off-peak hours. Their recommendation is premised on
support (either explicit or implicit) for effectively allocating steam and hydro
production-related costs away from the sumer on-peak loads though a
combination of classification and allocation technques that improperly focus on
energy usage. While witness Yankel' s position is understandable-the irrgation
class benefits from disproportionate cost allocations to off-peak loads-I do not
understand why Staff supports a costing methodology that fails to emphasize
sumer peak demands as the principal driver behind IPC' s rising production
costs. Moreover, the costing methodologies advocated by IIP A and Staff put a
completely uneasonable emphasis on energy usage as a principal cost drver.
HOW DOES THIS OVER-EMPHASIS ON ENERGY OCCUR?
The problem starts with IPC's (and Staffs) use of load factor to classify steam
and hydro production plant. Under ths load factor classification scheme, IPC
(and Staff) classify almost 60 percent of these fixed costs as energy-related costs.
No witness in this case has provided a valid reason why such a high percentage of
IPC's fixed production costs should be classified as variable energy costs. That
should surrise nobody since the classification cannot be justified on either
economic or engineering grounds. As I noted earlier, load factor is simply the
relationship between average and peak demands. Whle system load factor may
infuence the tyes and operation of system production resources, it does not
Case No. IPC-E-08-10
Dennis W. Goins - DOE - Reb
Page 10
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
reflect the driving force underlying the amount of required production resources-
namely peak demands.8
IPC's and Staffs over-emphasis on energy as a cost driver is then exacerbated
by the use of 12CP factors in the 3CPI12CP methodology to allocate the 40
percent of steam and hydro fixed costs that are classified as demand costs. A
12CP allocation method is a hybrid between strct demand and energy allocation
approaches since it diminishes the importance of annual peaks through averaging,
thereby indirectly recognizing annual energy usage in assignng cost responsibility
for fixed production costs. As a result, the 12CP method shifts cost responsibility
to energy-intensive, high load factor customers. More importantly, except for
utilities with similar monthy peaks, the 12CP method largely ignores the role of
annual system peaks in driving the need for and operation of production capacity.9
For example, in IPC's case, the 12CP approach almost totally ignores the role that
hydro plant in paricular plays in serving IPC's sumer peak loads. Combining a
load factor scheme that improperly classifies 60 percent of IPC's steam and hydro
plant costs as energy costs with an allocation method that mutes the importance of
summer peak demands in allocating the remainng demand-related 40 percent of
these costs results in further unjustified shifts of IPC's production costs to high
load factor customers.
8 While I support classifying all steam and hydro plant costs as demand-related costs, in my direct testimony
I presented a rational alternative to IPC's (and Staffs) load factor classification scheme. My alternative
classified IPC's steam and hydro plant costs as 57.10 percent demand and 42.90 percent energy. I offered
this alternative simply as an option if the Commission decides that some part ofIPC's fixed steam and
hydro plant costs should be classified as energy-related costs.
9 A 12CP allocation method is sometimes
justified as a means of reflecting the value of production capacity
in all months-including months with maximum demands far below the utility's anual system peak.
However, a utility does not plan for and add production capacity to meet the average of its 12 monthly
system peaks. Its capacity additions are driven by its need to meet annual system peaks. Capacity used to
meet annual system peaks then becomes available to meet peak demands in all other months.
Case No. IPC-E-08-10
Dennis W. Goins - DOE - Reb
Page 11
2
3 A.
4
5 Q.
6
7 A.
8
9
10
11
12 Q.
13
14 A.
15
16
17
18
19 Q.
20
21
22 A.
23
Q.DID STAFF PERFORM A COST-OF-SERVICE STUDY USING THE
3CP/12CP METHODOLOGY?
Yes. The cost-of-service analysis that Staff conducted (see Staff Exhibit No. 130)
was based on a lower jurisdictional revenue requirement than IPC proposed.
DID STAFF CORRCT ANY OF THE DEFICIENCIES IN IPC'S COST
STUDY THAT YOU IDENTIFIED IN YOUR DIRECT TESTIMONY?
No. Staff witness Hessing made no changes in IPC's cost study with respect to
the classification and allocation of IPC's retail costs. Staffs cost-of-service study
is simply IPC' s cost study with Staff s lower jursdictional revenue requirement.
As a result, Staffs cost study suffers from the same fatal flaws as IPC's cost
study-flaws that I discussed in detail in my direct testimony.
DID STAFF PROPOSE A REVENUE SPREAD THAT REFLECTS
RESULTS FROM ITS COST-OF-SERVICE ANALYSIS?
Yes. Staffs proposed revenue spread is quite similar to IPC's proposed spread
although the percentage increases are smaller because of Staffs lower revenue
requirement. In particular, Staffs proposed revenue spread calls for increases
well above the system average increase for higher load factor customers, including
special contract customers.
HOW DOES STAFF JUSTIFY ITS RECOMMENDED HIGHER-THAN-
AVERAGE RATE INCREASES FOR HIGHER LOAD FACTOR
CUSTOMERS?
Staff implies that because average rates (prices) for higher load factor customers
(around $30 dollars per MW) are less than IPC's marginal power supply costs
Case No. IPC-E-08-10
Dennis W. Goins - DOE - Reb
Page 12
2
3 Q.
4
5
6 A.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
(around $60 per MW according to witness Hessing), above average increases for
higher load factor customers should be expected. 10
DOES THIS RELATIONSHIP BETWEEN MARGINAL POWER SUPPLY
COSTS AND AVERAGE RATES SUPPORT STAFF'S RECOMMENDED
REVENUE SPREAD?
No. Witness Hessing's cost and rate comparison ignores two key factors:
. ¡PC's test-year average fuel and purchased power expense is less than
$18 per MW-well below the average price witness Hessing claims
that higher load factor customer pay and far below his estimated $60
per MW marginal power supply cost. Under ¡PC's rates, customers
are charged average power supply costs-not marginal costs.
Moreover, marginal power supply costs are not constant-they may
vary signficantly by hour, day, or month. In many hours of the year,
the average price IPC charges for energy may be signficantly above
its marginal cost of energy.
. In a properly designed cost-of-service study, high load factor
customers should receive most of the off-system sales revenue credits
resulting from off-peak sales attbutable to excess block energy
purchases and available system hydro capacity. This does not happen
in ¡PC's cost study-which Staff recommends-because off-system
sales revenue credits are allocated without reference to when such
sales occur (primarly off-peak months). As I pointed out in my direct
testimony, because IPC's costing methodology does not properly lin
sales revenue credits to off-peak energy usage, a disproportionately
small share of the off-system sales revenue credits is assigned to
10 Ibid. at 11:2-7.
Case No. IPC-E-08-10
Dennis W. Goins - DOE - Reb
Page 13
1
2
3
4
5
6 Q.
7
8
9
10 A.
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
higher load factor customers and a disproportionately high share is
assigned to lower load factor customers whose electrcity usage occurs
primarily in summer peak months. As a result, IPC's cost of serving
higher load factor customers is overstated and understated for low
load factor classes.
HAS THE COMMISSION TRAITIONALLY RELIED ON COST
STUDIES THAT ALLOCATE A DISPROPORTIONATE AMOUNT OF
IPC'S STEAM AND HYDRO PLANT COSTS TO NON-SUMMER, OFF-
PEAK LOADS?
No. For 25 years the Commssion has consistently adopted what is commonly
referred to as a weighted 12CP methodology to allocate demand-related steam and
hydro production plant costs to rate classes. The weights used to develop class
allocation factors are IPC's monthly marginal generation capacity costs. These
weighted 12CP allocation factors emphasize peak demands in sumer months
when IPC's marginal generation capacity costs are highest. As a result, weighted
12CP factors allocate substantially more costs to rate classes that use the IPC
system durig sumer peak hours. Because IPC's growing sumer peak
demands are driving its need for production resources, the Commission should not
support any cost-of-service methodology that does not emphasize the importce
of sumer on-peak demands in drving the need for capacity resources-for
example, the costing methodologies proposed by witnesses Yankel and Hessing.
Instead, the Commission should adopt my recommended classification of
production costs and my recommended weighted 12CP methodology, which
emphasizes summer peak demands, to allocate IPC's production costs to rate
classes.
Case No. IPC-E-08-10
Dennis W. Goins - DOE - Reb
Page 14
1 Q.
2
3 A.
4
5
6 Q.
7
8
9 A.
10
11
12
13
14
15
16
17
18
19
20
21
22
SHOULD THE COMMISSION REJECT STAFF'S RECOMMENDED
COST STUDY AND ASSOCIATED REVENUE SPREAD?
Yes. Staff s cost study is fatally flawed, and its recommended revenue spread-
which is derived from the cost study results-does not reasonably track IPC's cost
of serving each customer class.
HAVE YOU CHANGED ANY RECOMMENDATIONS IN YOUR DIRECT
TESTIMONY AFTER REVIEWING THE TESTIMONY FILED BY
WITNESSES YANKEL AND HESSING?
No. I stil recommend the following:
1. Reject IPC's 3CP/12CP seriously and probably fatally flawed cost-of-
service study. i I
2. Reject IPC's classification of steam and hydro production plant costs as
demand- and energy-related costs. Instead, all steam and hydro production
plant costs should be classified as demand-related costs.
3. If the Commission allows IPC to classify steam and hydro plant costs into
demand and energy cost components, then system load factor should not
be used to determine the energy cost component. Instead, as an
alternative, I recommend classifying 57.10 percent of these plant costs as
demand and 42.90 percent as energy. (I described how these percentages
are derived in my direct testimony.)
4. Reject IPC's classification of Account 555 purchased power costs.
Instead, they should be classified using the same alternative classification
11 Throughout my direct testimony I focused on IPC's 3CPI12CP cost study since IPC recommends this
study. However, IPC's Base Case and Modified Base Case cost studies suffer from deficiencies
comparable to those in the 3CPI12CP cost study. As a result, neither the Base Case nor the Modified Base
Case studies should be used for setting IPC's rates in this case.
Case No. IPC-E-08-10
Dennis W. Goins - DOE - Reb
Page 15
2
scheme I propose for classifying steam and hydro plant costs (that is, 57.10
percent demand and 42.90 percent energy.)
3
4
5
6
7
5. Reject IPC's proposed assignment of all demand-related hydro plant costs
to the baseload capacity category. Instead, I recommend assigng 50
percent of demand-related hydro costs to the baseload plant category
(which is allocated on the basis of 12CP demands) and 50 percent to the
peaking category (which is allocated on the basis of3CP demands).
8
9
10
11
12
6. Reject IPC's proposed assignment of demand-related purchased power
costs to baseload and peaking capacity categories on the basis of how it
assigns production plant to these categories. Instead, I recommend using
the same 50/50 demand and energy split for demand-related Account 555
costs that I recommend for assigng demand-related hydro plant costs.
13
14
15
16
7 Reject IPC's marginal-cost-weighted allocation of energy costs in its
3CP/12CP study. Instead, an unweighted energy cost allocation should be
used to ensure that higher load factor classes are assigned a higher
percentage of the lower fuel costs associated with baseload capacity.
17
18
19
8. Require IPC to allocate demand-related production costs using a weighted
12CP method. I presented results from two Wl2CP cost studies that I
performed in Exhbit Nos. 6 i 0 and 611.
20
21
22
23
24
9. Reject any revenue spread that is based on 3CPIl2CP cost study results.
Instead, results from my Exhibit No. 6l 1 should be used as a staring point
in developing a revenue spread for any rate change the Commssion
approves in this case.12 At a minimum, these results support an across-the-
board revenue spread instead of the higher-than-average increases for
12 This includes Staffs proposed revenue requirement.
Case No. IPC-E-08-10
Dennis W. Goins - DOE - Reb
Page 16
1
2
3
4
5 Q.
6 Á.
higher load factor and special contract customers proposed by IPC and
Staff.
10. Require ¡PC to retain the services of a reputable outside firm to examine,
evaluate, and recommend necessar changes to its cost-of-service modeL.
DOES THIS COMPLETE YOUR REBUTTAL TESTIMONY?
Yes.
Case No. IPC-E-08-10
Dennis W. Goins - DOE - Reb
Page 17