REDACTED Direct Testimony of J. Richard Hornby Synapse Energy ...

Pi]:

)I:

u.

**

I

i1-l ;[I 22 A;/

BEFORE THE ARKANSAS PUBLIC SERVICE COMMISSION

-!a"

I

,

.

I . -

IN THE MATTER OF ENTERGY ARKANSAS, INC.'S REQUEST FOR APPROVAL OF THE ACQUISITION OF NEW CAPACITY TO SERVE ITS RETAIL CUSTOMERS

(1

"s.

d

) ) ) DOCKET NO. 06-152-U ) )

REDACTED

Direct Testimony of

J. Richard Hornby Synapse Energy Economics, Inc. On behalf of the General Staff of the Arkansas Public Service Commission

January 24,2007

'37

Entergy Arkansas, Inc. Docket No. 06-152-U Redacted Direct Testimony of J. Richard Hornby 1

Q.

Please state your name, position and business address.

2

A.

My name is J. Richard Hornby. I am a Senior Consultant at Synapse Energy Economics, Inc, 22 Pearl Street, Cambridge, MA 02139.

3 4

Q-

On whose behalf are you testifying in this case?

5

A.

I am testifying on behalf of the General Staff of the Arkansas Public Service Commission (General Staff).

6

7

Q.

Please describe Synapse Energy Economics.

8

A.

Synapse Energy Economics (Synapse) is a research and consulting firm specializing in energy and environmental issues, including electric generation,

9 10

transmission and distribution system reliability, market power, electricity market

11

prices, stranded costs, efficiency, renewable energy, environmental quality, and

12

nuclear power.

13

Q.

Please summarize your work experience and educational background.

14

A.

I am a consultant specializing in planning, market structure, ratemaking and gas

15

supply/fuel procurement in the electric and gas industries. Over the past twenty

16

years I have presented expert testimony and provided litigation support on these

17

issues in approximately 100 proceedings in over thirty jurisdictions in the United

18

States and Canada. Over this period my clients have included staff of public

19

utility commissions, state energy offices, consumer advocate offices and

20

marketers.

21

Prior to joining Synapse in 2006, I was a Principal with CRA International,

22

formerly Tabors Caramanis & Associates. From 1986 to 1998 I worked with the

23

Tellus Institute (formerly Energy Systems Research Group); initially as Manager

24

of the Natural Gas Program and subsequently as Director of their Energy Group.

25

Prior to 1986 I was Assistant Deputy Minister of Energy for the Province of Nova

26

Scotia.

1

Entergy Arkansas, Inc. Docket No. 06-152-U Redacted Direct Testimony of J. Richard Hornhy 1

I have a Master of Science in Energy Technology and Policy from the

2

Massachusetts Institute of Technology and a Bachelor of Industrial Engineering

3

from the Technical University of Nova Scotia, now merged with Dalhousie

4

University.

5

I have attached my resume to this testimony as Exhibit JRH-1.

6

Q.

What is the purpose of your testimony?

7

A.

Entergy Arkansas Inc. (EA1 or the Company) is requesting a finding that it has a

8

need for new capacity which can best be met through the acquisition of load-

9

following capacity using a combined cycle gas turbine (CCGT) technology.

10

Synapse was retained to assist the General Staff in the review of the request. In

11

particular, Synapse was asked to assist in determining whether the proposed

12

acquisition of CCGT capacity is in the public interest.

13

Q.

request?

14 15

What data sources did you rely upon to prepare your review of EAI’s

A.

I relied primarily on the Direct Testimony, exhibits and workpapers ofthe EA1

16

witnesses Robert R. Cooper and Kurtis Castleberry as well as their responses to

17

information requests, in both this proceeding and the base rate proceeding Docket

18

No. 06-101-U. In addition I reviewed the Commission’s Orders in Dolcket No.

19

03-028-U and various relevant materials filed in that proceeding.

20

Q-

need to acquire new load-following capacity.

21

22

Please summarize the rationale underlying EAI’s request for a finding of

A.

EAI’s request is based upon the application of its resource planning principles to

23

the results of its resource planning analyses. EA1 draws four key conclusions

24

from those analyses and principles:

25

e

The quantity of capacity that EA1 has under its long-term control is

26

forecast to be substantially less than the quantity of capacity that it should

27

hold to ensure reliable service to its firm customers, i.e. forecast peak 2

Entergy Arkansas, Inc. Docket No. 06-1524 Redacted Direct Testimony of J. Richard Hornby 1

requirements plus 15.25% reserve margin. The forecast shortfall in 2007

2

is 1,462 MW. The shortage is in capacity appropriate for meeting “load-following” requirements and peaking requirements. EA1 forecasts a shorta.ge of 670 MW of load-following capacity in 2007. EA1 is currently meeting its need for this capacity through purchases from other Operating Companies under the Entergy System Agreement and from purchases under short-term power purchase agreements (PPAs). The Company wishes to reduce its reliance on those sources.

10

@

Gas-fired CCGT capacity would be the most economic type of load-

11

following capability to acquire. This capacity would also increase the

12

diversity of EAI’s portfolio of resources.

13

Q.

request.

14 15 16

Please summarize the major conclusions from your review of the Company’s

A.

The major conclusions from my review of the Company’s request are summarized below.

17

First, the Company has a shortage of capacity under its long-term control.

18

Second, the shortage occurs in two categories of capacity, load-following

19

and peaking. CCGT capacity appears to have the most appropriate

20

operational characteristics for load-following.

21

Third, the Company’s proposal to increase the quantity of load-following

22

capacity under its long-term control by acquiring a CCGT resource is

23

consistent with good resource planning principles. However, EA1 will

24

need to demonstrate that the specific quantity and type of capacity it

25

actually acquires will enable it to provide reliable service at reasonable

26

rates over the long-term, considering the operational characteristics and

27

economics of its entire portfolio of existing resources. 3


Fourth, based upon the gas prices expected over the 2007 to 201 3 timeframe, and the capacity factor at which the new resource is expected to operate during that period, retail customers would not receive a net savings relative to a CCGT resource if EA1 were to use wholesale baseload capacity (WBL) that is currently assigned to wholesale and being sold to the Operating Companies under short-term PPAsto serve retail load. However, it is possible that a portion of that WBL capacity may be the most economic resource for the Company’s next long-term acquisition of capacity to meet firm retail requirements.

10

Q.

Please summarize your recommendations based upon those conclusions.

11

A.

I recommend that the Commission issue an order declaring that EAI:

12

0

has demonstrated a need to acquire capacity for load-following;

13

e

bears the burden of demonstrating that the specific quantity and type of

14

capacity that it acquires for load-following will enable it to provide

15

reliable service at reasonable rates; and

16

e

seek Commission approval before entering any new long-term PPAs for WBL capacity.

17 18

e

19

evaluate the cost effectiveness of the WBL capacity as part of its next acquisition.

20 21

Aggregate Shortfall in Capacity under EA1 Long-term Control

22

Q.

term control.

23 24

Please summarize EAI’s estimate of the shortfall in capacity under its long-

A.

Mr. Cooper presents his estimate of the forecast shortfall in his Direct Testimony.

25

He calculates the shortfall as the difference between the quantity of capacity EA1

26

must have available in order to ensure reliable service to firm retail customers, 4

Entergy Arkansas, Inc. Docket No. 06-1524 Redacted Direct Testimony of J. Richard Hornby

1

“firm requirements”, and the capacity that it actually has available for that

2

purpose, “capability” or firm resources. He presents his estimates for each year

3

over the 2007 to 2016 planning horizon in Exhibit RRC-1.

4

EAI’s firm requirements equal forecast peak requirements of firm retail customers

5

plus a planning reserve margin of 15.25%. Its estimate of 2007 firm requirements

6

is 5,630 MW. This is projected to increase to 6,609 MW by 2016. EAI’s

7

forecasts of peak demand, and its planning reserve of 15.25%, are reasonable.

8

For the resource, or capability, side of the equation in this calculation, EAT

9

considers only capacity that it either owns or controls under long-term PPAs,

10

which it defines as contracts with terms longer than three years. Mr. Castleberry

11

states that EA1 has 4,168 MW of firm resources available to serve retail load,

12

Exhibit KWC-1. Mr. Cooper indicates that the quantity of capacity available to

13

serve retail load will increase to 4,329 in 2009 when the Lake Catherine units

14

would be available to return from inactive reserve status.

15

Based upon those projected levels of firm requirements and capability, EA1 is

16

forecasting at shortfall of 1,462 MW in 2007 that will increase to 2,280 MW by

17

2016.

18

Q.

How has EA1 been ensuring reliable service given these shortfalls?

19

A.

EA1 has been, and is currently, meeting those shortfalls through purchases of

20

capacity from other Operating Companies under the System Agreement and under

21

under short-term PPAs with terms ranging between one and three years. EA1 is

22

proposing to reduce its dependence on those two sources by either purchasing, or

23

entering a long-term contract for, capacity.

24

Q.

that encourage retail customers to reduce or shift their peak demand?

25

26

Could the magnitude of the forecast shortfall be reduced through programs

A.

Yes, but not in the short-term. EA1 should certainly be working with retail

27

customers to identify programs that would lead to cost-effective reductions in

28

peak load, and hence reductions in the shortfall. The Commission recognized the 5


need to implement conservation, energy efficiency and demand response

2

programs in its recent ruling adopting Rules jbr Conservation and Energy

3

EfJiciency Programs. However, it takes time for such programs to be designed

4

and implemented. Thus it is not realistic to expect them to produce a material

5

reduction in EAI’s load in the short-term.

6

Q.

capacity under its long-term control?

7 8

What is the conclusion from your review of EAI’s estimate of the shortfall in

A.

My conclusion is that the Company has a shortage of capacity under its long-term control.

9 10

Q.

Please summarize EAI’s proposed response to this shortfall.

11

A.

EA1 is seeking a finding that its need for new capacity will best be met through

12

the acquisition of load-following capacity using CCGT technology. The

13

Company does not indicate that it is seeking approval in this phase for a specific

14

quantity of CCGT capacity, but Mr. Castleherry indicates that it has issued an

15

RFP for approximately 500 MW of capacity.

16

Q.

Have you reviewed EAI’s support for its proposal?

17

A.

Yes. In the balance of my testimony I describe my review of the materials that

18

EA1 has provided to support its proposal. My analyses assume 500 MW of new

19

capacity for illustrative purposes. I realize that EA1 may ultimately propose a

20

different quantity, higher or lower, in their acquisition filing.

21

Composition of Shortfall in Capacity under EA1 Long-term Control

22

Q.

capacity under its long-term control.

23

24

25

Please summarize EAI’s analysis of the composition of the shortfall in

A.

In his Direct Testimony Mr. Cooper presents an analysis of the composition of the

shortfall, expressed in terms of the categories of retail load that EA1 serves and

’ Docket No. 06-004-R. 6

Entergy Arkansas, Inc. Docket No. 06-152-U Redacted Direct Testimony of J. Richard Hornby

1

the types of capacity it has available to serve each category. He presents an

2

illustration of that analysis in Exhibit RRC-,2 using projections for 2007. Mr. Cooper begins by presenting the projected distribution or shape of the

3

4

firm requirements (MW) in each hour of the year, grouped from the lowest load in

5

an hour to the highest load in an hour. This load distribution curve is a graph that

6

shows the percent of annual hours during which retail load will be at a given

7

level. It is important to note that this not a graph of hourly load in chronological

8

order and as such it does not give one the full flavor of hour-to-hour variation in

9

load to which EA1 must respond by increasing and decreasing the generation from

10

load-following capacity. Mr. Cooper divides the load duration curve into four segments, according

11 12

to the category of load and the type of capacity that would be used to serve that

13

category in a “model generating portfolio.” His analysis considers three basic

14

categories of load and corresponding types of capacity - baseload, load-following

15

(further separated into core and seasonal) and peak. The key characteristics of

16

each category from a resource planning perspective are summarized below.

17

Baseload. EA1 defines this segment as the level of load that occurs in

18

85% of the hours of the year. In a model generating portfolio this segment

19

would be served by units with relatively high fixed costs and relatively

20

low variable costs, operated at a relatively steady level and high capacity

21

factor2, e.g., 80% or more.

22

Load-following. This segment of load varies substantially from hour to

23

hour during most hours of the year. The capacity used to serve this

24

segment must have the flexibility to operate at a wide range of output

25

levels and to vary its level of operation quickly. In a model generation

2

Capacity factor is the ratio of annual generation from a unit divided by the maximum annual generation that unit could generate. Thus, it is a useful indicator of a unit’s annual average

utilization. The higher a unit’s capacity factor, the more electricity it generates and the lower its unit fixed cost of production, since its absolute fixed cost is being spread over more generation. 7


portfolio this capacity would have lower fixed costs than base load

2

capacity but higher unit variable costs, since it will operate at a lower

3

capacity factor than baseload.

4

e

Peak load. This segment of load consists ofthe extreme hourly peaks that

5

occur in a very few hours of the year. The capacity used to serve this

6

segment must have the flexibility to operate at very high output levels with

7

short notice for short periods. This segment would ideally be served by

8

capacity with very low fixed costs and high variable costs, because it will

9

operate at a very low capacity factor, e.g. less than 5%.

10

Q.

the shortfall in capacity under EAI’s long-term control in 2007.

11 12

Please summarize the results of Mr. Cooper’s analysis of the composition of

A.

Mr. Cooper’s analysis indicates that, as compared to a “model generating

13

portfolio,” EA1 would have a shortage of load-following capacity (670 MW) and

14

a shortage of peaking capacity (1,564 MW) under its control in 2007. The results

15

of Mr. Cooper’s comparison are summarized in Table 1 below. -~

Table 1 Excess (Shortage) in MW Under Long-Term Control Relative to Model Portfolio

r l-----Category

Requirements

Resources

Baseload

Core Load-Following

Seasonal LoadFollowing Peak + Reserves

!

732

Excess (Shortfall)

771

4‘71

I 409

2,368

804

Total

8

(1,564)


1

Q.

Is Exhibit RRC-2 sufficient proof that it is reasonable for EA1 to acquire a specific quantity of load-following capacity?

2

A.

No. That Exhibit simply indicates that, as compared to an ideal or model portfolio, it has an excess of baseload and a. shortage of load-following and peaking capacity. One cannot conclude from that comparison that EA1 should acquire 670 MW of load-following capacity, or some alternative quantity, any more than one can conclude that 771 MW of the Company’s existing baseload capacity committed to firm retail requirements is “excess” to those requirements. In fact, EA1 has had a mismatch between its load categories and

9 10

corresponding capacity levels for several years. This was recognized in the

11

Docket No. 03-028-U proceeding for example. Although Mr. Castleberry has

12

categorized EAI’s existing capacity as either baseload, intermediate, or peaking,

13

the reality is that EA1 has some latitude to use certain units categorized as

14

baseload to provide load-following. The determination of whether a specific unit

15

categorized as baseload could be used in an intermediate role, for load-following,

16

will depend upon the operational characteristics of that specific unit as well as on

17

the economics of using it in that role.

18

Q.

Does Mr. Cooper’s Direct Testimony provide adequate support for the

19

acquisition of a specific quantity of load-following capacity in response to its

20

shortfall?

21

A.

No. In his Direct Testimony Mr. Cooper provides a few anecdotal examples of

22

the magnitude of the load-following capability that EA1 requires. However, he

23

does not demonstrate the extent to which a combination of its existing baseload

24

capacity, intermediate capacity, and purchased capacity could, or could not,

25

adequately provide that capability in the future. In addition, his economic

26

analyses, which I discuss in detail below, do not provide adequate support for

27

acquisition of a specific quantity of load-following capacity.

9


Q.

merits of the proposed acquisition?

2

3

Will the quantity of load-following capacity be a key factor in evaluating the

A.

Yes. Ultimately the quantity of load-following capacity to be acquired will

4

determine the absolute economic benefit or cost of the acquisition, which in turn

5

will determine the impact of the acquisition on retail rates over the planning

6

horizon. However, the quantity is a key input to the economic evaluation, as I will

7

discuss later. It does make a difference whether EA1 acquires 500 MW now, and

8

nothing further for several years, as opposed to, for example, 300 MW now and

9

200 MW in a few years time.

10

The implications of the quantity that is acquired in 2007 are illustrated in

11

my Exhibit JRH-2. This bar chart shows the owned and purchase resources as of

12

2007, baseload, load-following, and peaking, assuming 500 MW of new capacity.

13

From 2008 onward the top section of the bar in each year shows the additional

14

capacity that EA1 would need to meet its reliability requirements. That chart

15

indicates that if EA1 were to acquire 500 MW of load-following capacity in 2007,

16

it likely would not be in a position to acquire another significant quantity of any

17

other type of capacity on a long-term basis until 201 0. In addition, that quantity

18

of capacity will affect the quantity of its purchases under PPAs during that time

19

and could affect the capacity factor at which its baseload capacity operates.

20

Q.

Do you agree with EA1 that CCGT capacity has better operational

characteristics for providing load-following than other types of capacity?

21 22

A.

Yes.

23

Q.

What do you conclude from reviewing EAI’s analysis of the composition of the shortfall in capacity under its long-term control?

24 25

A.

I agree with the Company that its shortage occurs in two categories of capacity,

26

load-following, and peaking. However, EA1 has not provided sufficient evidence

27

to support the specific quantity of load-following capacity it will acquire. I agree

10


that CCGT capacity has better operational characteristics for load-following than

2

other types of capacity.

3

4

Acquisition of Load-Following Capacity Versus Purchase Under PPAS

5

Q.

following capacity under its long-term control.

6 7

Please summarize EAI’s rationale for increasing the quantity of load-

A.

EA1 is currently relying upon purchases under short-term PPAs and purchases

8

under the Entergy System Agreement for a substantial portion of its load-

9

following capacity. The Company wishes to reduce its reliance on those sources

10

by increasing the quantity of load-following capacity under its long-term control.

11

Q.

under its long-term control?

12 13

Do you agree with EAI’s general goal of increasing the quantity of capacity

A.

Yes. In his Direct Testimony Mr. Cooper discusses the principles that guide

14

EAI’s approach to resource planning. EA1 also provided data request responses

15

which outline these principles in more detail. One of EAI’s goals under these

16

principles is to acquire a portion of its total capacity requirements via short-term

17

PPAs, but to limit the size of that portion. E:AI maintains that it has, or is about to,

18

reach that limit. In addition, EA1 wishes to reduce its dependence on the other

19

Operating Companies, which makes sense given its plan to exit the System

20

Agreement as of 20 13.

21

Q.

Did either Mr. Castleberry or Mr. Cooper provide an economic analysis of

22

EAI’s proposal to replace load-following capacity from PPAs with load-

23

following capacity under its long-term control?

24

A.

No. In his Direct Testimony Mr. Castleberry does discuss the general

25

implications of a continued high reliance on PPAs for load-following capacity,

26

including the risk of insufficient supply and/or higher prices in the future.

27

However neither he nor Mr. Cooper provide an economic analysis comparing the 11


1

cost of load-following capacity under PPA!j to the cost of load-following capacity

2

under EAI’s long-term control .

3

Q.

PPAs with load-following capacity under its long-term control?

4 5

Did you review the economics of EA1 replacing load-following capacity from

A.

Yes. I prepared an estimate using information that EA1 provided regarding the

6

charges it is paying under its PPAs and my understanding of the range of likely

7

costs for acquiring CCGT capacity. Based upon those assumptions, EA1 may pay

8

somewhat more per MWh for load-following generation from CCGT under its

9

control as compared to purchasing under PPAs, in the order of

10

However, the impact on retail rates of that slightly higher unit

11

minimal. For example, if EA1 were to acquire 500 MW of load-following

12

capacity and use it at a

13

component of retail rates would be less than 1%.

capacity factor, the impact on the supply

14

That potential differential and impact should be assessed in terms of the

15

benefit of having capacity under EAI’s control in the long-term as discussed by

16

Mr. Castleberry. In addition, the capacity charges under PPAs may increase over

17

the planning horizon as supply becomes tighter over the region. Based upon these

18

factors the benefits of EA1 acquiring some quantity of CCGT appear to justify this

19

slight increase. However, when EA1 files for approval of an actual acquisition it

20

should include an analysis of the impact on rates of this displacement as part of its

21

economic justification.

22

Q.

load-following capacity under its long-term control?

23 24

What is your conclusion regarding EAI’s goal of increasing the quantity of

A.

The Company’s proposal to increase the quantity of load-following capacity

25

under its long-term control by acquiring a CCGT resource is consistent with good

26

resource planning principles. However, EA.1 will need to demonstrate that the

27

specific quantity and type of capacity it actually acquires will enable it to provide

28

reliable service at reasonable rates.

12


2

Economic Evaluation of Acquisition of Alternative Types of Capacity

3

Q.

and its economic evaluation of those resources.

4 5

Please describe the candidate resources that EA1 considered for acquisition

A.

In his Direct Testimony Mr. Cooper considers three types of capacity resources, a

6

new coal unit, a new gas-fired CCGT unit, and the existing Lake Catherine Unit

7

4. In his economic evaluation he calculates the unit cost of generation from each

8

unit, expressed as a 30 year levelized cost per MWh. He estimates those unit

9

costs for two different forecast of gas prices, a constant $5/MMBtu in 2006$ and

10

a constant $7/MMBtu in 2006$. Mr. Cooper calculates the unit cost of power

11

from each unit over a range of capacity factors, or levels of annual output. Mr.

12

Cooper presents the results of his calculations as graphs in Exhibit RRC-4.

13

Q.

What are the key assumptions in Mr. Cooper’s economic evaluation?

14

A.

The key assumptions in Mr. Coopers economic evaluation are the future price of

15

natural gas and the capacity factor at which the new resource is likely to operate.

16

The other assumptions used in the calculation are also important, but there is

17

much less uncertainty associated with them.

18

Q.

Please comment on the forecast gas price assumptions that Mr. Cooper uses.

19

A.

The $5 forecast gas price that Mr. Cooper uses is comparable to the base case

20

long-term forecast of the Energy Information Administration (EIA) in Annual

21

Energy Outlook 2006 (AE02006), when expressed on a long-term levelized basis,

22

e.g., 2007 - 2030. The $7 forecast gas price that he uses is somewhat higher than

23

the AE02006 high case long-term forecast, again when expressed on a levelized

24

basis. Overall, the two gas price forecasts that he uses are reasonable estimates

25

for a long-term analysis. As I will discuss in more detail below, a price of

26

approximately $7/MMBtu is most appropriate for more detailed annual analyses

27

over the 2007 to 2016 period. That price is consistent with current NYMEX

28

prices through 2012. 13


Q.

Please comment on the capacity factor assumptions that Mr. Cooper uses.

2

A.

Mr. Cooper presents the unit cost of generation from each candidate unit over a

3

range of capacity factors, from 10% to 80%. This broad range is somewhat

4

puzzling, since EA1 maintains that it needs load-following capacity. It implies

5

that new coal capacity would have the same load-following operational

6

characteristics as a CCGT unit. He does not provide information on, and

7

evaluation of, the capacity factor at which a new load-following resource is most

8

likely to actually operate in each year over the planning horizon and beyond.

9

Instead, he states that a new (baseload) resource added to EAI’s portfolio is not

10

expected to have an annual capacity factor in excess of 70% and that a new load-

11

following resource is expected to operate at or below 45%.

12

Q.

would operate at or below 45% over the planning horizon?

13 14

Did EA1 provide any analysis to confirm that a new load-following resource

A.

No. It is interesting to note that the illustrative core load-following capacity in the

15

“model generating portfolio” that Mr. Cooper presents in Exhibit RRC-2 would

16

have a capacity factor greater than

17

Q.

would operate at or below 45% over the planning horizon?

18 19

Did you prepare an analysis to confirm that a new load-following resource

A.

Yes. I did this by estimating the load in each hour of 2006, assuming that EA1

20

added 500 MW of load following capacity in 2007 and only added peaking

21

capacity from 2008 onward. My estimate of the load in 2016 is equal to the 2007

22

load in each hour increased by 17%, the cumulative increase in peak demand that

23

EA1 is projecting in Exhibit RRC-1. My results are presented as a figure in my

24

Exhibit JRH-3.

25

That analysis indicates that as load increases over time the

tor of the 500 MW increases from approximately

to

26 27

28

Q.

Why do you place so much emphasis on determining the capacity factor at

which the new load-following capacity is likely to operate? 14


A.

Capacity factor is a key input assumption that affects the relative economics of

2

the candidate resources, as indicated in Exhibit INC-4. If EA1 can indicate, with

3

a high degree of confidence, the range of capacity factors at which the new

4

resource will operate, it will reduce the uncertainty associated with the selection

5

of the preferred resource.

Q.

6

Based on the analyses prepared by EAI, and your review of the likely

7

capacity factor, is CCGT capacity preferable to a new coal plant or further

8

major investments in the Lake Catherine unit?

9

A.

Yes.

10

Q.

Are there other candidate resources available to EA1 that Mr. Cooper did not discuss in his Direct Testimony?

11

12

A.

Yes. EA1 did not consider existing EA1 capacity currently allocated to wholesale

13

sales that will become available as existing contracts with those customers expire.

14

Mr. Castleberry identifies this WBL capacity in Exhibit KWC-1. Approximately

15

200 MW of WBL capacity that EA1 was selling under short-term PPAs became

16

available by the end of 2006.3

17

Q.

following needs?

18

19

Has EA1 evaluated the economics of using WBL capacity to meet EAI’s load-

A.

Yes. In response to Staff Data Request PSG 2-3, EA1 provided an analysis of

20

using 200 MW of WBL capacity to provide load-following, at a

21

capacity factor, as opposed to selling it to the other Operating Companies. EA1

22

evaluated the economics of those two alternatives over a range of gas prices. EAI’s analysis indicates that using ‘WBL capacity would not be in the best

23 24

interests of Arkansas retail customers, largely because any savings it realized

25

from doing so would be offset by an increase in its annual Production Cost

26

Equalization Payment. In fact, EA1 estimates that the total offset would be such

3

Docket No. 05-1 16-U, Direct Testimony of Mr. Hugh McDonald, page 13. 15

Entergy Arkansas, Inc. Docket No. 06-152-U Redacted Direct Testimony of J. Richard Hornby that there would never be a net savings to retail customers from using the WBL capacity to serve retail load as long as the Production Cost Equalization Formula is in effect.

4

Q.

Did you prepare any further analyses in order to test EAI’s conclusions?

5

A.

Yes. I prepared a year-by-year comparison, for 2007 through 2016, of the unit

6

cost of two alternatives. The first alternative assumes EA1 acquires 500 MW of

7

CCGT in 2007 and sells the 200 MW of WBL to the Other Operating Companies.

8

The second alternative assumes EA1 meets its load-following needs through a

9

combination of 200 MW of WBL capacity and 300 MW of CCGT capacity. In

10

my analysis I calculate the net unit cost ($/MWh) of generation from each

11

alternative in each year, based on the capacity factors I identified earlier, and a

12

gas price of $7/MMBtu. (This gas price reilects the NYMEX forward prices for

13

gas at the Henry Hub through 2012 as of mid-January.). In addition, I estimated

14

the impact of the Production Cost Equalization Formula on the net unit cost of

15

each alternative in each year. The results of my analysis are consistent with those of EAI. The 500 MW

16 17

CCGT alternative has a lower unit cost in each year the Production Cost

18

Equalization Formula is in effect through 201 3. These results are presented

19

graphically in my Exhibit JRH-4.

20

Q.

EAI’s need for load-following capacity over the next several years?

21 22

Given these factors, is the WBL capacity a reasonable candidate for meeting

A.

No. It appears that that there would not be a net savings to retail customers if EA1

23

used the WBL capacity available to it for load-following. However, if EA1 were

24

to need baseload capacity as part of its next acquisition, in the 201 0 timeframe for

25

example, a portion of this WBL capacity may be cost-effective in that role. EA1

26

should evaluate the cost effectiveness of the WBL capacity as part of its next

27

acquisition. In addition it should seek Commission approval before entering into

28

any new long-term PPAs for WBL capacity.

16


Q.

evaluation of the candidate resources available to EA1 for load-following.

2 3

Please summarize the major conclusions from your review of the economic

A.

Based on my review of the Company’s analyses and my own analyses, I have two

4

main conclusions. First, CCGT capacity is preferable to the other candidate

5

resources, Le., a new coal plant, future investment in the Lake Catherine unit, and

6

use of WBL capacity. However, EA1 will need to demonstrate that the specific

7

quantity and type of capacity it actually acquires will enable it to provide reliable

8

service at reasonable rates over the long-term, considering the operational

9

characteristics and economics of its entire portfolio of existing resources. Second,

10

it is possible that a portion of that WBL capacity may be the most economic

11

resource for the Company’s next long-term acquisition of capacity to meet firm

12

retail requirements.

13

Q.

What do you recommend based upon the conclusions from your review?

14

A.

I recommend that the Commission issue an order declaring that EAI:

15

e

has demonstrated a need to acquire capacity for load-following; and

16

e

bears the burden of demonstrating that the specific quantity and type of

17

capacity that it acquires for load-following will enable it to provide

18

reliable service at reasonable rates.

19

e

seek Commission approval before entering any new long-term PPAs for

WBL capacity.

20

21

e

22

evaluate the cost effectiveness of the WBL capacity as part of its next acquisition.

23

Q.

Does this complete your Direct Testimony?

24

A.

Yes.

17

CERTIFICATE OF SERVICE I hereby certify that a copy of the foregoing has been served on all parties of record by forwarding the same by first class mail, postage prepaid, this 24'h day of January, 2007. -

Valerie F. Boyce

'