20040712Yankel Rebuttal.pdf

KEL
SSOCIA TES. IN
",,"
lX"tCE1VED
~ P
. ".
". i--... L
2084 JUL , 2 APi to: 29814 Lake Road
li.C- .. I
T i t~;Ji LJ ;:3 LI C
Bay Village, Ohio 44140
"'
IlL i tIt ~ co "'1t'.l1 55 ION
Telephone [phone]
Fax [phone]
BEFORE THE IDAHO PUBLIC UTILITIES COMMISSION
IN THE MATTER OF THE APPLICATION
OF A VISTA CORPORATION FOR THE
AUTHORITY TO INCREASE ITS RATES
AND CHARGES FOR ELECTRIC AND
NATURAL GAS SERVICE TO ELECTRIC
AND NATURAL GAS CUSTOMER IN
THE STATE OF IDAHO
CASE NO. VU-04-0 1
COEUR SILVER VALLEY
REBUTTAL TESTIMONY OF
ANTHONY 1. Y ANKEL
July 12, 2004
PLEASE STATE YOUR NAME, ADDRESS, AND EMPLOYMENT.
I am Anthony J. Yanke!. I am President ofYankel and Associates, Inc. My
address is 29814 Lake Road, Bay Village, Ohio, 44140.
ARE YOU THE SAME ANTHONY 1. Y ANKEL THAT HAS PROVIDED
DIRECT TESTIMONY IN TIllS CASE?
Yes.
WHAT IS THE PURPOSE OF YOUR REBUTTAL TESTIMONY?
I address certain issues brought up by the Staff with respect to Schedule 25.
Q. PLEASE SUMMARIZE YOUR CONCLUSIONS WITH RESPECT TO THE
ASSIGNMENT OF COSTS TO SCHEDULE 25 CUSTOMERS AS WELL AS THE RATE
DESIGN FOR THAT CUSTOMER CLASS.
A. The Staff's cost-or-service study (like the Companys) fails to properly address the
assignment/allocation of certain primary distribution related costs to Schedule 25. If data is
utilized that is more reflective of cost causation, the rate of return for Schedule 25 comes out to
be above the jurisdictional average.
Yankel, DI
Coeur
Although the Staff's rate design is somewhat of an improvement over that proposed by
the Company with respect to recognizing the benefits of load factor for Schedule 25 customers
there is still a good deal of room for improvement. I develop a rate design for Schedule 25
customers that is similar to that recently approved on the Idaho Power system, which far better
reflects a rate differential between high and low load factor customers.
Yankel, D I
Coeur
COST -OF -SERVICE STUDY
Q. DO YOU AGREE WITH THE STAFF'S COST-OF-SERVICE ANALYSIS?
A. No. The cost-of service study used by the Staffis simply the Companys cost study
with the inclusion of the Staff's (as opposed to the Companys) revenue requirement numbers.
Basically, the Staff did not challenge any of the Companys methodology. Admittedly, I did not
challenge a great deal of the study either, but my review was limited to only one group of 14
customers with very specific characteristics.
In 1994, the summer peak was only 88% of the winter peakl while in the 2002 test year
data used in this case, the summer peak is approximately 4% higher than the winter peak2
Primarily, this change has been brought about by an increase in air-conditioning load, which has
prompted the Company to begin including cooling degree values in its load normalization
calculations. It is my understanding that the load research data used in this case was gathered
over 10 years ago, during a time when this system was winter peaking. I mention this because
the Companys load research data impacts the residential classThe Companys cost-of-service
study lists the Residential class (like Schedule 25) as being significantly below cost-of-service.
A lot of the disparity that the Companys cost-of-setvice study is showing for the Residential
class could simply be an artifact of the outdated data being used by the Company that reflects a
very different load profile. Like Schedule 25, a lot more review should go into the data used to
develop cost-of-service studies before they are used to disproportionately raise rates for anyone
class of customers.
1 July 1994 peak was 1270 MW while the December peak was 1436 MW.2 Page "ILK -78 of the workpapers provided by Tara L. Knox.3 It does not impact the cost-of-selVice for Schedule 25 as they are all measured hourly.
Yankel, D I
Coeur
Q. HAS THE STAFF DISAGREED WITH YOUR POSITION WITH RESPECT TO
DEVELOPING MORE OF A DIRECT ASSIGNMENT FOR CERTAIN DISTRIBUTION
COSTS TO THE SCHEDULE 25 CUSTOMERS?
A. No, I assume that the lack of inclusion of direct assignment data for these distribution
costs associated with Schedule 25 was more of an oversight or lack of data, than a deliberate
disagreement with the treatment. Most rate analysts would agree that it is far more
appropriate/accurate to directly assign costs than it is to allocate costs.
Q. CAN DISTORTIONS IN COST -OF-SERVICE RESULT IF DIRECT
ASSIGNMENTS ARE NOT MADE?
A. Yes, significant distortions can occur if direct assignments are not made. Potlatch-
Lewiston is a good example. This is by far the largest customer on the system and is three times
the size of all Schedule 25 customers combined. The Company either allocates distribution plant
on the basis of Non-Coincident Peak (NCP) or it is directly assigned. Potlatch-Lewistons share
of the Idaho NCP is 20%4, Potlatch-Lewiston is directly assigned only $70921 of Account 361
costs (Structures and Improvements), but if these costs were to be simply allocated on the basis
ofNCP, Potlatch-Lewiston would be allocated $5198835 or over 7-times the actual cost
incurred. Potlatch-Lewiston does not even use any Account 364 (Poles, Towers & Fixtures), but
4 Exhibit 16 Schedule 2 page 31 line21.
5 $2627000 times 19.790/0 equals $519883.
Yankel, D I
Coeur
an allocation based upon NCP would place a burden upon this facility of $11283269. Simply
put, it is inappropriate to allocate costs to large customers on the basis ofNCP when it is possible
to directly assign or more accurately define cost causation.
Q. IS THE DATA YOU USED TO ASSIGN/ALLOCATE COSTS TO SCHEUDLE 25
A TRUE DIRECT ASSIGNMENT?
A. No. A true direct assignment would assign only costs. As a surrogate for cost
causation, I chose to use the actual miles of primary distribution line used to serve these
customers and the assumption that costs per circuit mile average out to be the same. If the
Company can produce actual cost figures for the 21 miles of the primary distribution lines that is
used to serve all Schedule 25 customers7 (compared to 3857 total primary miles in Idaho), then
this data should be substituted.
There should be no question that using actual miles of primary distribution line is far
more accurate for this customer group than the simple choice of using NCP data to allocate these
costs. The NCP data would suggest that an average of 60 miles of primary distribution line was
associated with eachof the Schedule 25 customers (includingPotlatch-Lewiston) when in fact
there is only 21 miles of primary distribution (overhead plus underground) that is used to serve
all Schedule 25 customers (including Potlatch-Lewiston).
Q. SHOULD SCHEDULE 25 BE SINGLED OUT TO GET MORE THAN THE
AVERAGE RATE INCREASE?
6 $57015000 times 19.79% equals $11283269.7 See Exhibit 306.
Yankel, D I
Coeur
A. No. The difference in the choice of reflecting the relative number of miles of primary
circuits compared to the simplistic application ofNCP is the sole difference that pegs the rate
return for Schedule 25 at significantly below average cost-or-service, versus sljghtly above
average cost -of-service. It is this difference that should have been recognized in the Company
cost study, before recommendations were made to disproportionately increase rates for Schedule
25. A fluke in the cost -of-service study or the lack of quality data should not be the cause of a
disproportionate increase to any class-especially, when better data is available.
Yankel, DI
Coeur
RATE DESIGN
Q. DO YOU AGREE WITH THE STAFF'S OVERALL POSITION WITH RESPECT
TO RATE DESIGN FOR SCHEDULE 25?
A. I have some concerns with some of the comments made by the Staff regarding the
rate design for Schedule 25 customers. Specifically, I disagree with Mr. Schunkes proposal for
the next case to gather additional information so that the Company can provide "a proposal to
eliminate the declining block rates in Schedules 21 and 25,,Although I welcome the
development of additional data, I do not believe that its intended purpose should be the
elimination of the declining block ratesThe data should be allowed to speak for itself and if
the data suggests that there should be more declining blocks or steeper declining blocks, then so
be it.
Q. DO YOU AGREE WITH THE RATE DESIGN DEVELOPED BY THE STAFF
FOR SCHEDULE 25?
A. No. At the outset, I should say that I agree with Dr. Peseaus assessment that
Potlatcb-Lewiston should not be included in the Schedule 25 rates. This facility should be
treated separately as there are no other customers that have load characteristics that are remotely
similar. My comments will address rate design for only 14 customers-all but the Potlatch-
Lewiston load.
8 Schunke' s direct testimony at page 4 lines 13 and 14.
Yankel, DI
Coeur
My primary disagreement with the Staff's proposed rate design for Schedule 25 is that in
spite of the inclusion of a declining block energy rate, it still places very little reward (via lower
rates) for higher load factor usage. In my direct testimony, I attempted to address this concern in
a general way. Now that a more probable revenue requirement is being addressed, it is possible
to put a numerical value to the rate design concepts I proposed in order to provide some reward
to higher load factor customers.
Q. HOW WOULD YOU DEVELOP A RATE DESIGN FOR SCHEDULE 25 THAT
BETTER REWARDS lllGH LOAD FACTOR CUSTOMERS?
A. In my direct testimony, I proposed a ratio between demand costs and tail block energy
costs of at least 120: 1 in order to be somewhat consistent with the rate design for similar
customers in the Idaho Power service area as recently adopted by this Commission. As you may
recall, I calculated a ratio of 78: 1 for the existing Schedule 25 rates and a ratio of 80: 1 for the
Company proposed Schedule 25 rates. The Staff proposal of a second demand block rate of
$2.75 per kW and 3.268 cents per kWh for the tail block energy rate produces a ratio of84:
some improvement, but still a far cry ITom the rate design on the Idaho Power system.
Instead of the Staffs proposal of $9000 for the first 3000 kW and $2.75 per kW for each
additional kW, I propose that the initial 3000 kW be priced at $10500 and that each additional
kW be priced at $3.25 per kW. This demand charge is still less than half of the demand cost
calculated by the Company of $7.02 per kW per month for Schedule 25 customers and it serves
several purposes: 1) It is a rate that is similar9 to the rate being charged to Idaho Power Schedule
9 Idaho Powers Schedule 19 rate has a $3.21 demand charge in the summer and a $2.64 demand charge in
the winter, but additionally has a Basic Load Capacity charge of an additional $0.37 per kW of annual peak
Yankel, DI
Coeur
19 customers of$3.21 per kW; 2) It places more charges on the demand component so that
higher load factor customers will receive more of a benefit; and 3) It allows a ratio of the demand
charge to the tail block energy rate to be sufficiently larger without forcing the tail block energy
rate itself to be significantly reduced. The net impact of this rate design would be to place
approximately half of the increase upon the demand component.
The Staff's tail block energy rate is 3.268 cents per kWh. Using the ratio I previously
recommended between demand and tail block energy rates of 120, my proposed tail block energy
rate becomes 2.710 cents per kWh. Although there is not a huge difference between these two
tail block rates, there is sufficient difference to cause the ratio of demand to energy charges (120)
to be similar to the emphasis that is placed upon load factor in the Idaho Power system. This
proposed tail block rate is about 6% 10 below the current energy rate for Schedule 25-meaning
that the tail block rate would have a slight decrease. If desired, a higher tail block could be
developed, but in order to maintain the ratio of demand to tail block energy rate of 120, this
would entail raising the demand charge further. It cannot be forgotten that all customers will pay
the demand rate as well as the initial and tail block energy rate, so just because one proportion of
the rate is going down, it does not mean that the overall bill is being reduced-just the price
signals will be arranged differently.
The last rate component to be addressed is the initial energy block. The last rate
component must do two things: 1) it must make sense; and 2) it must result in a rate such that
when taken in total, all of the rate components produce the revenue requirement for the schedule.
I have targeted the average rate increase calculated by the Staff of 15.78% because I believe that
demand that that effectively increases both the demand an winter demand charges by more than $0.37 per
montWybilling demand-depending upon the difference between the monthly billing demand and annual
demand.
710 cents divided by 2.874 cents equals 94.3%.
Yankel, DI
Coeur
Schedule 25 should get no more than the average rate increase. In order to get this percentage
increase from Schedule 25 with the above proposed rate components, an initial energy block rate
of 4.33 cents per kWh is required. This rate is well within the realm of reason and is 12% greater
than the initial block rate proposed by the Staff.
Q. PLEASE SUMMARIZE YOUR RATE DESIGN FOR SCHEDULE 25 AND WHY
YOU BELIEVE THAT IT IS BETTER THAN THAT PROPOSED BY EITHER THE
COMPANY OR THE STAFF.
A. There is hardly any reward under the present Schedule 25 rate design for high load
factor customers. The days of the energy constrained utility in Idaho are numbered. More
emphasis should be placed upon demand charges in the A vista service territory compared to the
past. I believe that Idaho Powers new rates can serve as a model for rate design in the A vista
service area. The demand charge that I have proposed is essentially the same as that for Idaho
Powers Schedule 19 and the tail block energy rate is designed to hit a target ratio that is
representative of rate design on the Idaho Power system. In some respects, the proposal I am
making may seem radical, but the perceived change is more a result of where we have been as
opposed to where we should be going-the historical rate design was greatly lacking in its ability
to reward high load factor customers.
Q. DOES THIS CONCLUDE YOUR REBUTTAL TESTIMONY?
A. Yes.
Yankel, DI
Coeur
EXHIBIT 306
CONFIDENTIAL
CO
M
P
A
R
I
S
O
N

O
F

P
R
O
P
O
S
E
D

R
A
T
E
S

F
O
R

S
C
H
E
D
U
L
E

2
5
St
a
f
f
St
a
f
f
Co
e
u
r
Co
e
u
r
Bi
l
l
i
n
g
Pr
e
s
e
n
t
Pr
e
s
e
n
t
Pr
o
p
o
s
e
d
Pr
o
p
o
s
e
d
Pr
o
p
o
s
e
d
Pr
o
p
o
s
e
d
De
t
e
r
m

i
n
a
n
t
s
Ra
t
e
s
Re
v
e
n
u
e
Ra
t
e
s
Re
v
e
n
u
e
Ra
t
e
s
Re
v
e
n
u
e
Bi
l
l
s

3
,
00
0

o
r

l
e
s
s

k
W
16
8
50
0
26
0
00
0
00
0
51
2
00
0
$1
0
50
0
76
4
00
0
Gr
e
a
t
e
r

t
h
a
n

3
00
0

k
W
28
5
49
3
$2
.
$6
4
2
35
9
$2
.
$7
8
5
,
10
6
$3
.
$9
2
7
85
2
Bl
o
c
k

1

p
e
r

k
W
h
00
0
00
0
$0
.
02
8
7
4
41
4
16
0
$0
.
03
8
7
3
25
3
,
32
0
$0
.
04
3
3
0
63
7
20
0
Bl
o
c
k

2

p
e
r

k
W
h
21
9
70
7
,
4
8
1
$0
.
02
8
7
4
31
4
39
3
$0
.
03
2
6
8
18
0
,
04
0
$0
.
02
7
1
0
95
4
07
3
Pr
i
m
a
r
y

V
o
l
t
a
g
e

D
i
s
c
o
u
n
t
:l
l
i
.
U
1
.
Q
$1
0
,
4
7
5
10
2
$1
2
57
4
65
6
$1
2
12
7
31
5
Pe
r
c
e
n
t
a
g
e

I
n
c
r
e
a
s
e
20
.
04
%
15
.
77
%
;:
:
;
.
:
Certificate of Service
I HEREBY CERTIFY that on this 9th day of July 2004, I caused to be served a true and
correct copy of the foregoing document to the individual addressed below:
Jean Jewell
Idaho Public Utilities Commission
472 W. Washington St.
Boise, ill 83720-0074
Dennis E. Peseau, Ph.
Utility Resources, Inc.
1500 Liberty Street SE, Ste 250
Salem, OR 97302
Scott Woodbury
John Hammond
Idaho Public Utilities Commission
472 W. Washington St.
Boise, ill 83720-0074
Conley E. Ward
Givens Pursley LLP
601 W. Bannock St.
Boise, ill 83701-2720
David J. Meyer
Senior Vice President and General Counsel
A vista Corporation
1411 E. Mission Ave., MSC-
Spokane, W 1\ 99220-3727
Brad M. Purdy
Attorney at Law
2019 N. 17th St.
Boise, ill 83702
Kelly Norwood
Vice President, State and Federal Regulation
1\vista Utilities
1411 E. Mission Ave., MSC-
Spokane, W A 99220-3727
Michael Karp
147 Appaloosa Lane
Bellingham, W A 98229
Charles L.A. Cox
Evans, Keane
111 Main St.
Kellogg, ill 83837
~jF