20080627Said direct.pdf

E¡
Ii: 3ö
BEFORE THE IDAHO PUBLIC UTILITIES COMMISSION
IN THE MATTER OF THE APPLICATION
OF IDAHO POWER COMPANY FOR
AUTHORITY TO INCREASE ITS RATES
AN CHAGES FOR ELECTRIC SERVICE.
CASE NO. IPC-E-08-10
IDAHO POWER COMPANY
DIRECT TESTIMONY
OF
GREGORY W. SAID
1 Q.Please state your name and business address.
2 A.My name is Gregory W. Said and my business
3 address is 1221 West Idaho Street, Boise, Idaho.
4 Q.By whom are you employed and in what
5 capacity?
6 A.I am employed by Idaho Power Company as the
7 Manager of Revenue Requirement in the Pricing and
8 Regulatory Services Department.
9 Q.Please describe your educational background.
10 A.In May of 1975, I received a Bachelor of
11 Science degree in Mathematics with honors from Boise State
12 University. In 1999, I attended the Public Utility
13 Executives Course at the University of Idaho.
14 Q.Please describe your work experience with
15 Idaho Power Company.
16 A.I became employed by Idaho Power Company in
17 1980 as an analyst in the Resource Planning Department. In
18 1985, the Company applied for a general revenue requirement
19 increase. I was the Company witness addressing power
20 supply expenses.
21 In August of 1989, after nine years in the Resource
22 Planning Department, I was offered and I accepted a
23 position in the Company's Rate Department. With the
24 Company's application for a temporary rate increase in
SAID, DI 1
Idaho Power Company
1 1992, my responsibilities as a witness were expanded.
2 While I continued to be the Company witness concerning
3 power supply expenses, I also sponsored the Company's rate
4 computations and proposed tariff schedules in that case.
5 Because of my combined Resource Planning and Rate
6 Department experience, I was asked to design a Power Cost
7 Adjustment ("PCA") which would impact customers' rates
8 based upon cnanges in the Company's net power supply
9 expenses. I presented my recommendations to the Idaho
10 Public Utilities Commission in 1992, at which time the
11 Commission established the PCA as an annual adjustment to
12 the Company's rates. I sponsored the Company's annual PCA
13 adjustment in each of the years 1996 through 2003. I
14 continue to supervise PCA-related regulatory filings.
15 In 1996, I was promoted to Director of Revenue
16 Requirement and in 2002 I was promoted to Manager of
17 Revenue Requirement. I have managed the preparation of
18 revenue requirement information for regulatory proceedings
19 since 1996.
20 Q.What topics will you discuss in your
21 testimony in this proceeding?
22 A.My testimony can be divided into four
23 sections addressing (1) power supply expense modeling, (2)
24 PCA changes resulting from base rate changes, (3) revenue
SAID, DI 2
Idaho Power Company
1 requirement adjustments that I provided to Ms. Schwendiman,
2 and (4) revenue requirement observations and conclusions.
3 POWER SUPPLY EXPENSE MODELING
4 Q.What role does power supply expense modeling
5 play in a general rate case?
6 A.Power supply expense modeling in a general
7 rate case provides the Commission with a view of "normal"
8 expectations for fuel expense (FERC accounts 501 and 547),
9 purchased power expense (FERC account 555), and surplus
10 sales revenue (FERC account 447). Power supply investment,
11 depreciation expense,and operating and maintenance
12 expenses are reflected in other FERC accounts that are not
13 addressed by power supply expense modeling.
14 Q.Please def ine the term "variable power
15 supply expenses" as the Company and the Commission have
16 used the term historically.
17 A.The Company and the Commission have
18 traditionally used the term "variable power supply
19 expenses" to refer to the sum of fuel expenses (FERC
20 accounts 501 and 547) and purchased power expenses (FERC
21 account 555) excluding expenses due to purchases from PURPA
22 qualifying facilities ("PURPA") minus surplus sales
23 revenues (FERC account 447). Because surplus sales
24 revenues are subtracted from fuel and purchased power
SAID, DI 3
Idaho Power Company
1 supply expenses, variable power supply expenses are also
2 referred to as net power supply expenses. For ratemaking
3 purposes, PURPA expenses have been qUantified separately
4 from variable power supply expenses and are treated as
5 fixed inputs to power supply modeling rather than variable
6 outputs.
7 Q.How are variable power supply expenses
8 "normalized" for ratemaking purposes?
9 A.Variable power supply expenses are
10 determined for each water condition dating back to 1928.
11 In this case, 80 water conditions have been evaluated. The
12 average of the variable power supply expenses over the
13 range of hydro conditions is considered "normal" or
14 representative of the possible circumstances the Company
15 might encounter for ratemaking purposes. The Idaho Public
16 Utilities Commission first adopted this method of averaging
17 a representative range of power supply expenses associated
18 with multiple water conditions to determine normalized
19 power supply expenses in 1981.
20 Q.Have you supervised the preparation of
21 normalized variable power supply expense modeling to
22 reflect the current test year 2008 characteristics?
23 A.Yes. Under my supervision and at my
24 request, a power supply simulation that is representative
SAID, DI 4
Idaho Power Company
1 of the test year 2008 variable power supply expenses
2 associated with 80 separate water conditions was prepared.
3 This year the analysis includes water conditions
4 corresponding to years 1928 through 2007. The average of
5 the variable power supply expenses related to each of the
6 80 water conditions represents the normalization of
7 variable power supply expenses.
8 Q.Please describe the simulation of test year
9 2008 variable power supply expenses.
10 A.The simulation of test year 2008 variable
11 power supply expenses reflects 2008 normalized loads and
12 resources that include 127 average megawatts of PURPA
13 generation. The 2008 PURPA generation amount includes a
14 reduction of 62 average megawatts from the PURPA generation
15 amount used in 2007. This reduction is due to a number of
16 PURPA proj ects changing their contracts to delay their on-
17 line dates beyond December 2008.
18 Q.Have you supervised the preparation of an
19 exhibit to demonstrate the normalization of variable power
20 supply expenses for the test year 2008?
21 A.Yes. Exhibit No. 47 shows the results of
22 the variable power supply expense normalization modeling
23 for the test year 2008.
SAID, DI 5
Idaho Power Company
1 Page 1 of Exhibit No. 47 shows the summary results
2 containing the 80-year average variable power supply
3 generation sources and expenses. Pages 2 through 81
4 contain results for each of the 8 0 individual water
5 conditions 1928 through 2007.
6 Q.How has the annual PURPA expense changed
7 since the last general rate case that used a 2007 test
8 year?
9 A.The annual PURPA expense for test year 2008
10 has decreased from $93.1 million to $63.3 million
11 reflecting the delay of 62 average megawatts of anticipated
12 PURPA proj ects that I mentioned earlier in my testimony.
13 Q.Have you supervised the preparation of an
14 exhibit detailing the test year 2008 PURPA project
15 generation and expenses?
16 A.Yes. I supervised the preparation of
17 Exhibit No. 48 which consists of one page. Column 1 of
18 Exhibit No. 48 shows the generation and expenses associated
19 with contracted PURPA projects that will be on-line during
20 the test year 2008.
21 Q.What are the corresponding variable power
22 supply expenses for the 2008 test year based upon this
23 level of PURPA generation and expense?
SAID, DI 6
Idaho Power Company
1 A.The normalized variable power supply expense
2 for the 2008 test year as shown on Page 1 of Exhibit No. 47
3 is $88.4 million. This amount is $47.5 million greater
4 than the Company's filed 2007 test year normalized variable
5 power supply expenses and $53.5 million greater than the
6 2007 test year normalized variable power supply expenses as
7 determined in Order No. 30508 based upon a stipulation of
8 the parties in Case No. IPC-E-07-08.
9 Q.What do the $29.8 million decrease in PURPA
10 expense and the $53.5 million increase in normalized
11 variable power supply expense indicate with respect to the
12 change in net total power supply expense from test year
13 2007 to test year 2008?
14 A.It indicates a $23.7 million net total power
15 supply expense increase ($53.5 million additional expense
16 $29.8 million reduction in expense = $23.7 million net
17 increase) to serve increased test year 2008 loads with
18 reduced PURPA generation sources. Base on those amounts, I
19 instructed Ms. Schwendiman to use the $88.4 million of net
20 variable power supply expense as shown on page 1 of Exhibit
21 No. 47 and the corresponding PURPA expense of $63.3 million
22 as shown on page 1 of Exhibit No. 48 in her quantification
23 of the Company's 2008 revenue requirement. This represents
24 a total PURPA and variable power supply expense of $151.7
SAID, DI 7
Idaho Power Company
1 million ($88.4 million + $63.3 million = $151.7 million),
2 which is an increase of $23.7 million from the 2007 test
3 year determination of $128.0 million.
4 Q.Has there been any change in the Company's
5 system load since the last general rate case, IPC-E-07-08?
6 A.Yes. The Company's 2007 annual normalized
7 system load used in the IPC-E-07-08 general rate case was
8 15.6 million megawatt-hours ("MWh"). The Company's 2008
9 annual normalized system load used in this case is 15.9
10 million MWh, an approximate 1. 9 percent (15.9 million MWh /
11 15.6 million MWh = 1.92 percent) increase in system load.
12 Q.Please recap the change in total PURPA and
13 variable power supply expenses that corresponds to the 1.9
14 percent higher loads of 2008 and the reduction in
15 contracted PURPA resources.
16 A.The Company's determination of normalized
17 variable power supply expenses for the test year 2008 in
18 this case is $88.4 million (page 1 of Exhibit No. 47). The
19 corresponding 2008 PURPA expense is $63.3 million (page 1
20 of Exhibit No. 48) for a total 2008 PURPA and variable
21 power supply expense of $151.7 million ($88.4 million +
22 $63.3 million = $151.7 million). The Commission adopted a
23 2007 normalized variable power supply expense for the test
24 year 2007 of $34.9 million. The corresponding test year
SAID, DI 8
Idaho Power Company
1 2007 PURPA expense was $93.1 million for a total 2007 PURPA
2 and variable power supply expense of $128.0 million ($34.9
3 million + $93.1 million = $128.0 million). Total
4 normalized PURPA and variable power supply expenses have
5 grown by $23.7 million ($151.7 million - $128.0 million =
6 $23.7 million) .
7 Q.Have the modeled market prices of energy
8 changed in the last year?
9 A.Yes. Modeled market prices for energy sold
10 as surplus are slightly lower than market prices last year.
11 In the IPC-E-07-08 case, monthly-modeled surplus sales
12 prices fluctuated from $21 per MWh to $118 per MWh
13 depending on market conditions . The annual fluctuation of
14 modeled surplus sales prices in that case was from $34 per
15 MWh to $73 per MWh. In this case, monthly-modeled surplus
16 sales prices fluctuate from $16 per MWh to $104 per MWh.
17 The annual fluctuation of modeled surplus sales prices in
18 this case is from $30 per MWh to $79 per MWh. Because of
19 the additional load and reduction of PURPA generation,
20 surpluses have been reduced during the highest market price
21 periods of time bringing the averaged weighted price for
22 surplus sales down. With the load growth that the Company
23 has experienced and the reduction of PURPA generation, the
24 normalized volume of surplus sales has decreased from 3.0
SAID, DI 9
Idaho Power Company
1 million MWh to 2.4 million MWh.
2 Modeled market prices for energy purchased are also
3 slightly lower than market prices last year. In the IPC-E-
4 07-08 case, monthly-modeled purchased power prices
5 fluctuated from $15 per MWh to $165 per MWh depending on
6 market conditions. Anual fluctuation of modeled purchased
7 power prices in that case was from $42 per MW to $116 per
8 MWh. In this case, monthly-modeled purchased power prices
9 fluctuate from $13 per MWh to $93 per MWh. The annual
10 fluctuation of modeled purchased power prices in this case
11 is from $22 per MWh to $81 per MWh. While there has been a
12 slight decrease in the modeled purchased power prices, the
13 normalized volume of purchased power has increased from 401
14 thousand MWh to 472 thousand MWh due to seasonal load
15 growth.
16 Q.Have fuel prices for Company-owned coal-
17 fired generating plants changed over the last two years?
18 A.Yes. The cost of coal at the Bridger plant
19 has increased from $14.51 per megawatt-hour to $16.12 per
20 megawatt-hour. The cost of coal at the Boardman plant has
21 increased from $13.91 per megawatt-hour to $14.36 per
22 megawatt-hour. The cost of coal at the Valmy plant has
23 increased from $22.06 per megawatt-hour to $24.12 per
24 megawatt-hour. Coal price increases are the result of a
SAID, DI 10
Idaho Power Company
1 number of factors, principally, the costs of mining and
2 transportation. Higher costs for steel , explosives, tires,
3 and diesel fuel as well as higher costs to remove
4 overburden associated with deeper coal seams have combined
5 to drive coal mining costs higher. Once mined, coal is
6 transported via railroad cars, again at higher costs than
7 in 2007. Higher mining costs and higher transportation
8 costs result in higher ultimate fuel costs. The fuel cost
9 for the Boardman coal-fired plant has not increased at the
10 same pace as the fuel costs at the Bridger and Valmy plants
11 based upon a below-market price contract that will expire
12 at the end of 2008.
13 Q.Have modeled variable gas prices for
14 Company-owned plants changed over the last two years?
15 A.Yes. For test year 2007, the Company
16 modeled gas prices at $98.32 per megawatt-hour for the two
17 smaller Danskin units and $86.45 per megawatt-hour for the
18 Bennett Mountain unit. Modeled variable gas prices for the
19 2008 test year are $79.90 per megawatt-hour for the three
20 Danskin units and $81.96 per megawatt-hour at Bennett
21 Mountain. The reduction in variable gas prices for the
22 three Danskin units reflects the addition of Danskin unit 1
23 that has a lower heat rate than the older Danskin units.
24 The reduction in the Bennett Mountain variable gas rate is
SAID, DI 11
Idaho Power Company
1 reflective of a slight reduction after the post-hurricane
2 spikes in gas prices.
3 Q.In light of load growth, PURPA resource
4 decline, market price changes, and fuel cost changes, do
5 you believe the Company's modeled power supply expenses
6 represent a reasonable estimate of normalized power supply
7 expenses for the test year 2008?
8 A.Yes, I do.
9 Q.Please summarize the Company's sources of
10 energy as shown on page 1 of Exhibit No. 47.
11 A.From the summary information contained on
12 page 1 of Exhibit No. 47, it can be seen that for the test
13 year 2008, Company-owned hydro generation supplies 8.7
14 million MWh while Company-owned thermal generation supplies
15 7.4 million MWh (Bridger 5.1, Boardman 0.4, and Valmy 1.9) .
16 This is essentially the same generation output from
17 Company-owned resources that was envisioned in the 2007
18 test year. Danskin and Bennett Mountain, as peaking
19 plants, operate intermittently, but offer significant
20 contribution at important times when resources and
21 purchases are inadequate to serve peak loads.
22 Purchases of power come from three sources: market
23 purchases, contract purchases other than PURPA, and PURPA
24 purchases. PURPA purchases are assumed at fixed normalized
SAID, DI 12
Idaho Power Company
1 levels amounting to nearly 1.1 million MWh. Because the
2 PURPA purchases are fixed inputs to power supply modeling,
3 they are not shown on the variable output summary, however,
4 when combined wi th the market and other contract purchases
5 of 1.0 million MWh, total purchases amount to 2.1 million
6 MWh (1.1 million MWh + 1.0 million MWh) .
7 Total hydro and coal-fired generation amounts and
8 purchases add up to 18.2 million MWH (8.7 + 7.4 + 2.1 =
9 18.2). Hydro generation contributes approximately 48
10 percent (8.7 million MWh / 18.2 million MWh = 48 percent)
11 of the generation mix, thermal generation contributes
12 approximately 41 percent (7.4 million MWh / 18.2 million
13 MWh = 41 percent), and purchases contribute approximately
14 11 percent (2.1 million MWh / 18.2 million MWh = 11
15 percent).
16 Q.How is the energy from the resources you
17 just described used?
18 A.Of the over 18.2 million MWh consumed, 15.9
19 million MWh are utilized for system loads while over 2.3
20 million MWh are sold as surplus. With load growth and the
21 reduction in PURPA generation, surplus sales have been
22 reduced from the 2.9 million MWh anticipated in the 2007
23 test year.
SAID, DI 13
Idaho Power Company
1 Q.Please summarize the expense and revenue
2 information associated with the normalized power supply
3 operations that you have just described.
4 A.Exhibi t No. 47 contains variable expense and
5 revenue information limited to FERC accounts 501, Fuel
6 (coal); 547, Fuel (gas); 555, Purchased Power; and 447,
7 Sales for Resale. Hydro generation has no assumed fuel
8 expense. Coal expenses of $133.4 million are comprised of
9 Bridger at $82.1 million, Valmy at $45.3 million and
10 Boardman at $6.0 million. Gas expenses amount to $7.1
11 million. Purchased power expenses, not including PURPA,
12 amount to $58.1 million while surplus sales amount to
13 $110.2 million. Altogether, net variable power supply
14 expenses amount to $88.4 million ($133.4 million + $7.1
15 million + $58.1 million - $110.2 million = $88.4 million) .
16 PCA CHAGES
17 Q.How do base level PCA expenses differ from
18 test year variable power supply expenses?
19 A.Base level PCA expenses differ from test
20 year variable power supply expenses in two ways. First,
21 base level PCA expenses include PURPA expenses. Second,
22 base level PCA expenses are determined for an April through
23 March time frame rather than a calendar year. April
24 represents the beginning of the runoff period that provides
SAID, DI 14
Idaho Power Company
1 the basis for the PCA proj ection.
2 Q.What is the base level of PCA expenses for
3 test year 2008?
4 A.In this case, normalized power supply
5 expenses amount to $88.4 million and normalized PURPA
6 expenses amount to $63.3 million. The sum, $151.7 million,
7 represents the new base PCA expense level.
8 Q.Are you sponsoring an exhibit that shows the
9 derivation of the appropriate new PCA regression formula to
10 be used for projecting the next year's PCA expenses?
11 A.Yes. Exhibit No. 49 was prepared under my
12 supervision to show the derivation of the new PCA
13 regression formula.
14 Q.Please describe Exhibit No. 49.
15 A.Exhibit No. 49 consists of six columns.
16 Column 1 shows the number of the observation from 1 to 79.
17 Column 2 contains the PCA year corresponding to each
18 observation; observation 1 is 1928, observation 2 is 1929,
19 and so on through observation 79 which is 2006. Because
20 the PCA year is for months April through March of the
21 following year, there are only 79 observations instead of
22 the 80 conditions represented in Exhibit No. 47. Column 3
23 contains the April through July runoff measured at Brownlee
24 Dam for each of the observation years 1928 through 2006.
SAID, DI 15
Idaho Power Company
1 Column 4 contains the natural logarithm of the runoff value
2 contained in Column 3. Column 5 contains the April through
3 March annual power supply expense based upon data from
4 Exhibit No. 47, but reflecting PCA-year totals rather than
5 calendar year totals. Finally, Column 6 contains the
6 regression predicted value of April through March annual
7 power supply expenses .
8 To the right of the columns is summary output of
9 certain regression statistics (such as r-square) and
10 formula coefficients.
11 Q.Please describe the new PCA regression
12 formula based upon Exhibit No. 49.
13 A.The basic PCA formula takes the following
14 form: Annual PCA expense = Cl - C2 * ln (Brownlee runoff)
15 + C3. The values of CL, C2 and C3 are constant with the
16 only variable being April through July runoff measured at
17 Brownlee Dam. The equation without C3 is used to predict
18 net power supply expenses and is the direct result of the
19 regression analysis contained in Exhibit No. 49. The
20 constant CL represents the prediction of annual net power
21 supply expense that would occur if there was zero April
22 through July runoff at Brownlee. The value of Cl is
23 $2,595,771,216. In reality, the lowest April through July
24 runoff measured at Brownlee contained in the observations
SAID, DI 16
Idaho Power Company
1 is 1.93 million acre-feet which occurred in the 1992
2 observation.
3 Because the regression provides a linear fit of a
4 non-linear transformation, the value of C2 is somewhat
5 difficult to explain. Observed Brownlee runoff data in
6 acre-feet is first transformed by the natural logarithm
7 function. For each unit increase in the natural logarithm
8 of the Brownlee runoff data the proj ection of annual power
9 supply expenses will be reduced by C2, which is
10 $162,707,198. The average natural logarithm of Brownlee
11 runoff values, based upon the observations contained in
12 Exhibit No. 49, is 15.41. This value corresponds to a
13 runoff of approximately 4.9 million acre-feet (e A 15.41 =
14 4,925,814 million acre-feet). With a runoff of 4.9 million
15 acre-feet and a natural logarithm of 15.41, the projected
16 net power supply expenses would be $88,453,295
17 ($2,595,771,216 - ($162,707,198 * 15.41)). An increase of
18 1 to the natural logarithm would result if the runoff was
19 approximately 13.4 million acre-feet (In(13,389,749) equals
20 16.41 which equals 15.41 + 1.0). With a runoff of
21 13,389,749 acre-feet, the net power supply expenses would
22 be $162,707,198 less than $88,453,295 making the projection
23 of power supply expenses a negative $74,253,903
24 ($2,595,771,216 - ($162,707,198 * 16.41) = -$74,253,903).
SAID, DI 17
Idaho Power Company
1 The natural logarithms of observed Brownlee runoff
2 ranged from 14.47 (1992 runoff) to 16.25 (1984 runoff).
3 The difference, 1.78 (16.25 - 14.47), multiplied by
4 $162,707,198, equals approximately $290 million, which
5 represents the change in proj ected power supply expenses
6 from the highest water case (1984) to the lowest water case
7 (1992) .
8 The value of C3 is $63,269,889, which is the
9 normalized PURPA expense. Because the normalized PURPA
10 expense is quantified separately from net variable power
11 supply expenses, it is added to net variable power supply
12 expenses to determine the PCA expenses.
13 Q.What is the new PCA regression equation with
14 values inserted for the constants?
15 A.The new PCA regression equation is:
16 Annual PCA expense = $2,595,771,216
17 - $162,707,198 * ln (Brownlee runoff)
18 + $63,269,889.
19 Q.How does the range in proj ected power supply
20 expenses from high condition to low condition resulting
21 from this regression equation compare to the corresponding
22 range of proj ected power supply expenses based upon the
23 previous regression equation?
SAID, DI 18
Idaho Power Company
1 A.The predictions of power supply expenses
2 based upon the regression observations contained in the
3 previous regression analysis ranged by $333 million from
4 the highest estimate to lowest estimate of power supply
5 expenses. The current range varies by only $290 million as
6 a result of slightly lower market price assumptions which
7 have reduced the volatility in power supply expenses.
8 Q.Please describe what is meant by the term
9 "embedded" cost.
10 A.The term "embedded" cost refers to an
11 average cost that is "embedded" in the rates and charges
12 paid by the Company's customers. Included within all
13 customer class rates is an embedded component related to
14 the total of PURPA and variable power supply expenses.
15 There would also be embedded components related to other
16 generation related expenses, transmission related expenses,
17 distribution related expenses, general and administrative
18 expenses, and returns. All customer classes have the same
19 embedded PURPA and variable power supply cost because no
20 customer class has preferential rights to energy. As a
21 result, the embedded rate for PURPA and power supply
22 expenses as reflected as a component of the overall rate is
23 determined by dividing the test year total PURPA and
24 variable power supply expenses by the total system load.
SAID, DI 19
Idaho Power Company
1 Q.What is the embedded total PURPA and
2 variable power supply expense rate at the generation level
3 as derived from data contained in Exhibit No. 47?
4 A.The embedded total PURPA and variable power
5 supply expense rate at generation level is $9.56 per
6 megawatt-hour ($151,691,135 / 15,863,628 megawatt-hours =
7 $9.56 per megawatt-hour) .
8 Q.How does the embedded total PURPA and
9 variable power supply expense rate compare to the
10 Commission approved Load Growth Adjustment Rate ("LGAR")?
11 A.The Commission approved LGAR is $62.79 per
12 MWh, but is only applied to one-half of load growth in the
13 2008 PCA year making the rate effectively $31.40 per MWh.
14 Q.Do you have a recommendation for the
15 appropriate level for the LGAR beginning in April 2009?
16 A.No . Per Order No. 30508, the Commi ss ion has
17 directed the Commission Staff, the Company and interested
18 parties to convene workshops to seek agreement as to the
19 appropriate LGAR methodology to be used after March 2009.
20 Q.Did Commission Order No. 30215 direct the
21 Company to update marginal cost studies and line loss data
22 in general rate proceedings?
23 A.Yes.
SAID, DI 20
Idaho Power Company
1 Q.Please define "marginal" costs with relation
2 to PURPA and variable power supply costs.
3 A."Marginal" costs refer to a very specific
4 computational method of determining incremental costs for a
5 hypothetical situation where no model inputs change other
6 than load. Rather than measuring the change in total PURPA
7 and variable power supply expenses from one year to the
8 next and dividing by the change in load from the first year
9 to the next, marginal costs are determined based upon a
10 hypothetical instantaneous load change and the resulting
11 modeled expense change to serve that load change. In
12 recent analyses, marginal costs are also based upon a five-
13 year average.
14 Q.At your direction, did the Company prepare
15 marginal cost analyses in conjunction with this case?
16 A.Yes. Exhibit No. 50 contains a
17 quantification of the five-year average marginal energy
18 cost at generation level (i. e. including line losses) as
19 $65.98 per megawatt-hour using standard marginal cost
20 methodology and 2008 through 2012 data. The annual
21 marginal cost for the single year 2008 is $56.48 per
22 megawatt-hour.
23 Q.Do you recommend any additional PCA
24 computational changes with the establishment of the new PCA
SAID, DI 21
Idaho Power Company
1 regression formula?
2 A.Yes. There are two PCA computational
3 factors that need to be updated as a result of the current
4 review of power supply expenses. First, for PCA proj ection
5 calculations, a new normalized Base Power Cost must be
6 determined for inclusion in rate Schedule 55. Second, a
7 new Idaho jurisdictional percentage must be determined.
8 Q.Have you supervised the development of an
9 exhibit to determine the PCA computational factors you have
10 just mentioned?
11 A.Yes. Exhibit No. 51 is a one-page exhibit
12 detailing the appropriate computation of the PCA factors I
13 have outlined.
14 Q.What is the first computation shown on
15 Exhibit No. 51?
16 A.The first computation details the normalized
17 Base Power Cost computation. The new normalized PCA
18 expense for the 2008 test year is $151.7 million compared
19 to the previous $128.0 million settlement value from the
20 2007 test year.
21 The normalized Base Power Cost is equal to the
22 $151.7 million normalized PCA expense divided by the
23 normalized system sales value of 14,465,151 MWh. The
24 resulting Base Power Cost is 1.04867 cents per kWh or
SAID, DI 22
Idaho Power Company
1 $10.49 per megawatt-hour.
2 Q.Please discuss the Idaho jurisdictional
3 percentage computation contained in Exhibit No. 51.
4 A.The Idaho jurisdictional firm load
5 (15,036,726 MWh) divided by the system firm load number
6 (15,863,628 MWh) results in an Idaho jurisdictional
7 percentage of 94.8 percent. This is up from 94. 7 percent
8 in 2007 due to a slightly higher growth rate in Idaho than
9 in Oregon.
10 RE REQUIRET ADJUSTMNTS
11 Q.Please describe your role in the preparation
12 of the Company's proposed 2008 revenue requirement.
13 A.As the Manager of Revenue Requirement, I
14 evaluated the concerns the parties in the IPC-E-07-08 rate
15 case expressed with regard to the Company's presentation of
16 test year data in that case. Based upon the parties'
17 strongly expressed desire to have an auditable starting
18 point and explicit methods of adjusting starting values to
19 the test year, I directed the Company's efforts to respond
20 to those requests. The results of those efforts are
21 reflected in the exhibits of Ms. Schwendiman. The
22 auditable starting point is 2007 actual data. That data
23 has been adjusted to reflect normalized power supply
24 expenses as approved in the 2007 case and to remove
SAID, DI 23
Idaho Power Company
1 expenses that are typically not considered for ratemaking
2 purposes such as certain memberships or advertizing
3 expenses.
4 Given adjusted 2007 data, several methods are then
5 utilized to adjust historical 2007 data to test year 2008
6 levels. These methods have been primarily described by Ms.
7 Smi th in her testimony in this case. The primary methods
8 used to adjust historical 2007 data to the 2008 test year
9 include trending of plant investments less than $2 million
10 using a compound growth rate, using known and measurable
11 adjustments for plant investments of greater that $2
12 million, and basing the growth of expenses and revenues
13 upon compound growth rates. I was part of the senior
14 management team that assisted Ms. Smith in developing these
15 methods to adjust historical 2007 data to 2008 test year
16 levels.
17 Q.In addition to the methods of adjusting 2007
18 data to the 2008 levels described by Ms. Smith, are there
19 some specific methods for adjusting 2007 data to the 2008
20 test year that you provided to Ms. Schwendiman?
21 A.Yes. I instructed Ms. Schwendiman to make
22 additional adjustments to reflect 2008 power supply
23 expenses, fuel inventories, imputed revenues for
24 annualizing adjustments associated with plant additions
SAID, DI 24
Idaho Power Company
1 greater than $2 million, and contributions in aid of
2 construction ("CIAC"). I have previously described the
3 power supply expense levels that I instructed Ms.
4 Schwendiman to use.
5 Q.Please describe the fuel inventory
6 adjustment that you instructed Ms. Schwendiman to use.
7 A.I instructed Ms. Schwendiman to adj ust fuel
8 inventory dollars to reflect a 26 -day inventory at the
9 Bridger Plant and 60-day inventories at both Valmy and
10 Boardman. Because Bridger is a mine-mouth plant, fewer
11 days of fuel inventory is required.
12 Q.Did you instruct Ms. Schwendiman to include
13 imputed revenue associated with annualized plant additions
14 of greater than $2 million?
15 A.Yes. The Commission in Order No. 29505
16 issued in Case No. IPC-E-03-l3 stated that "it is critical
17 to match revenues and expenses to these plant additions" in
18 reference to known and measurable additions. In Order No.
19 29505, the Commission used a proxy for additional revenues
20 stating that the Company had "not adequately quantified"
21 such additional revenues. In its next rate case, Case No.
22 IPC-E-05-28, the Company introduced a methodology for
23 imputing revenues. The Company used this same methodology
24 in the preparation of its revenue requirement in the next
SAID, DI 25
Idaho Power Company
1 rate case, Case No. IPC-E-07-08. Both cases were settled.
2 In its filing in this Case the Company has included a
3 quantification of revenues associated with annualizing
4 adjustments to transmission and distribution plant
5 determined in the same manner submitted in the 2005 and
6 2007 rate cases.
7 Q.Please describe the Company's method of
8 quantifying revenues associated with the annualizing
9 adj ustments to plant.
10 A.In order to estimate the additional revenues
11 that the Company would receive as a result of adding the
12 plant reflected in the annualizing adjustments, I requested
13 the preparation of Exhibit No. 52. Page 1 of Exhibit No.
14 52 shows the quantification of the revenue credit
15 associated with the annualizing plant adjustment. Page 2
16 of Exhibit No. 52 shows the planned use of those additional
17 facilities annualized in the Company' s2008 test year.
18 Based upon the system anticipated loads to be served via
19 those facilities by year end 2008 (128,479 MWh) and the
20 system average revenue per MWh ($15.56 per MWh), the
21 imputed revenue associated with the annualized transmission
22 and distribution additions is $1,489,324 for the Idaho
23 jurisdiction. This is an approximate 11.6 percent
24 reduction to the Idaho jurisdictional revenue requirement
SAID, DI 26
Idaho Power Company
1 resulting from these additional investments. Most of the
2 annualized investments in this case are for the purposes of
3 system reliability, compliance, or environmental
4 improvement rather than being related to load growth.
5 Q.What instruction did you give Ms.
6 Schwendiman with regard to CIAC?
7 A.I instructed Ms. Schwendiman to adjust
8 actual 2007 CIAC to 2008 levels based upon the method used
9 to adj ust the corresponding plant investments for those
10 specific accounts from 2007 to 2008 levels. Ms. Smith
11 discusses the methods used to adjust plant financial data.
12 RENU REQUIRENT OBSERVATIONS AN CONCLUSIONS
13 Q.Please summarize why Idaho Power Company is
14 utilizing a 2008 test year.
15 A.The fundamental reason that Idaho Power is
16 utilizing a 2008 test year is to address current concerns
17 regarding regulatory lag. In prior rate cases, rates
18 resulting from a test year were implemented five months
19 after completion of the test year (2003 test year rates
20 became effective June 1, 2004, and 2005 test year rates
21 became effective June 1, 2006). Rates implemented in March
22 2008 were based upon a settlement stipulation that did not
23 specify a precise test year. A 2008 test year in this case
24 will allow for rates based upon a 2008 test year to become
SAID, DI 27
Idaho Power Company
1 effective early in 2009, shortly following the test year.
2 Q.In your opinion, given normal conditions in
3 2009, will implementing rates based upon a 2008 test year
4 allow the Company to earn its authorized rate of return in
5 2009?
6 A.No. Based upon recent experience where the
7 Company is making large investments in all aspects of its
8 business at the same time that costs are rising, I do not
9 envision that revenues that the Company will receive based
10 upon a 2008 test year will keep pace with the revenue
11 requirements driven by investment levels and expenses in
12 2009.
13 Q.Does that conclude your testimony?
14 A.Yes, it does.
SAID, DI 28
Idaho Power Company