PUBLIC DISCLOSURE DOCUMENT

Transcription

1 DOCUMENT STATE OF GEORGIA BEFORE THE GEORGIA PUBLIC SERVICE COMMISSION In Re: Georgia Power Company s Application for Decertification of Plant Branch Units and and Mitchell Unit C, Application for Certification of the Power Purchase Agreements with BE Alabama LLC from the Tenaska Lindsay Hill Generation Station and with Southern Power Company from the Harris, West Georgia and Dahlberg Electric Generation Plants and Updated Integrated Resource Plan ) ) ) ) ) ) ) ) ) DOCKET NO. DIRECT TESTIMONY OF DAVID A. SCHLISSEL ON BEHALF OF SOUTHERN ALLIANCE FOR CLEAN ENERGY NOVEMBER, 0

2 Docket No. 0 0 Q. What are your name, position and business address? A. My name is David A. Schlissel. I am the President of Schlissel Technical Consulting, Inc. My business address is Horace Road, Belmont, Massachusetts 0. Q. On whose behalf are you testifying in this case? A. I am testifying on behalf of the Southern Alliance for Clean Energy ( SACE ). Q. Please summarize your educational background and work experience. A. I graduated from the Massachusetts Institute of Technology in with a Bachelor of Science degree in engineering. In, I received a Master of Science degree in engineering from Stanford University. In, I received a law degree from Stanford University. In addition, I studied nuclear engineering at the Massachusetts Institute of Technology during the years -. Since, I have been retained by governmental bodies, publicly-owned utilities, and private organizations in states to prepare analyses and expert testimony on engineering and economic issues related to electric utilities. My recent clients have included the U.S. Department of Justice, the Attorney General and the Governor of the State of New York, state consumer advocates, and national and local environmental organizations. I have filed expert testimony before the Georgia Public Service Commission ( Commission ) and state regulatory commissions in Arizona, Texas, New Jersey, Connecticut, Kansas, New Mexico, California, New York, North Carolina, South Carolina, Maine, Illinois, Vermont, Indiana, Ohio, Massachusetts, Missouri, Rhode Island, Wisconsin, Iowa, South Dakota, Minnesota, Michigan, Florida, North Dakota, Mississippi, Maryland, Virginia, Arkansas, Louisiana, Colorado, New Mexico, Oregon, and West Virginia, and before an Atomic Safety & Licensing Board of the U.S. Nuclear Regulatory Commission. Page

3 Docket No. 0 0 A copy of my current resume is included as Exhibit. Additional information on my work experience is available at Q. Have you testified previously before this Commission? A. Yes. I presented testimony in Commission Dockets Nos. -U and 00-U. Q. What is the purpose of your testimony? A. Schlissel Technical Consulting was retained to review and evaluate Georgia Power Company s ( Georgia Power or the Company ) August, 0 Application for the decertification of Plant Branch Units and and Mitchell Unit C and approval of related accounting requests; certification of the four power purchase agreements ( PPAs ) and approval of cost recovery requests; and approval of the updated Integrated Resource Plan ( IRP ), including the 0 Unit Retirement Study. My analysis focuses on the Company s decertification request and Unit Retirement Study. Q. Please summarize your conclusions. A. My conclusions are as follows:. Georgia Power s Unit Retirement Study is insufficient because it (a) relies on an unreasonably high range of natural gas prices; (b) fails to consider the potential for significantly higher coal prices; (c) fails to reflect the performance- and cost-related risks associated with continued operation of aging coal units; and (d) compares retrofitting existing coal units to building new gas-fired capacity instead of the actual alternatives of the four PPAs, for which the Company is seeking certification in this proceeding.. Despite the above flaws, which favor the continued operation of coal, in most of the scenarios in the Unit Retirement Study, the retirement of each coal unit is the lower cost alternative or is xxxxxxxxxxxxxxxxxxxxxxx than the option of retrofitting and continuing to operate the unit. Page

4 Docket No The Company should file a revised Unit Retirement Study, preferably by its December, 0 amended application deadline with the Commission, and the parties to this docket should be afforded an opportunity to review that revised Study.. Georgia Power s decision to retire and decertify Plant Branch Units and is reasonable and its decertification request should be approved. Q. What research have you undertaken in preparing this testimony? A. I have reviewed Georgia Power s Application and supporting documents, the Company s testimony, and its response to the data requests submitted by the Staff of the Public Service Commission. In addition, as part of my ongoing work, I regularly review materials related to the operations, costs, and financial and economic risks associated with existing coal-fired power plants. Natural Gas Prices Included in the Retirement Study Q. What natural gas prices did the Company use in its 0 Unit Retirement Study? A. Georgia Power uses natural gas prices that were based on Henry Hub gas price forecasts prepared by Charles River Associates ( CRA ) in October 00 and revised in April 0. Q. How do the Henry Hub natural gas price forecasts, on which the Company s gas prices are based, compare to current NYMEX futures prices and the natural gas price forecasts in AEO0 from the Energy Information Administration of the U.S. Department of Energy? A. As shown in Figure, which compares the Company s natural gas price forecasts for a $0/ton CO price to the NYMEX futures prices and AEO0 forecasts, the Henry Hub natural gas price forecasts used by Georgia Power in its 0 Unit The Company has designated these forecasts trade secret. Page

5 Docket No. Retirement Study are RE DACTED than both current NYMEX futures prices and the gas price forecast in the EIA s AEO0 report. Figure : Georgia Power Henry Hub Natural Gas Price Forecasts vs. Current NYMEX Gas Price Futures and AEO0 FIGURE REDACTED 0 As Figure illustrates, even Georgia Power s Low Gas Price Forecast is approximately xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx current NYMEX futures. The Company s Moderate Gas Price Forecast is REDACTEDxxxxxxxxx x than current NYMEX gas price futures and its High Gas Price Forecast is REDAC TEDx x xx the current NYMEX futures prices (that is, xxxxxxx REDxxxxxxxxxxxA CTED). Page

6 Docket No. 0 0 Q. Are you aware of any other utility that has also relied on CRA in developing the natural gas prices used in recent resource planning assessments like Georgia Power s 0 Unit Retirement Study? A. Yes. Ameren Missouri used CRA for help in developing the natural gas prices that it used in its 0 IRP filing with the Missouri Public Service Commission earlier this year. Q. Did Ameren Missouri make public the natural gas prices that it used in its IRP analyses? A. Yes. The fuel prices that Ameren Missouri used in its IRP analyses (including Henry Hub natural gas prices) were included in the public portion of the company s IRP filing. Q. How do the Henry Hub natural gas prices that Ameren Missouri used in its 0 IRP compare to the natural gas prices that Georgia Power used in its 0 Unit Retirement Study? A. As shown below in Figure, Ameren Missouri did not use any natural gas prices in its IRP analyses that xx x xxxxx Georgia Power s High Gas Price Forecast for a $0/ton CO price. The low Henry Hub natural gas prices used by Ameren Missouri xxxxxxxxxxx Georgia Power s Low Price Forecast and Ameren s high Henry Hub natural gas prices xxxxxxxxxxxxxxx Georgia Power s Moderate Forecast. Page

7 Docket No. Figure : Georgia Power s Henry Hub Natural Gas Price Forecasts vs. Ameren Missouri s Gas Price Forecasts, Current NYMEX Gas Price Futures, and AEO0 FIGURE REDACTED 0 Q. What information has Ameren provided the Missouri Commission about the natural gas prices it used in its 0 IRP analyses? A. Ameren has indicated that the natural gas prices it used in its 0 IRP analysis may be too high. It provided information to the Missouri Commission that showed that the Henry Hub natural gas prices that it used in its 0 IRP were consistent with the AEO00 forecasts but were above the AEO0 forecasts. Consequently, Ameren intends to include an update of its natural gas price assumptions in its 0 IRP Annual Update. Ameren Missouri 0 Response to Comments of Parties, Exhibit A at, Missouri Public Service Commission Case No. EO-0-0, available at Page

8 Docket No. 0 Q. Figures and compare the Henry Hub natural gas prices that Georgia Power used in its scenarios with a $0/ton CO price with recent NYMEX futures and the natural gas price forecasts in the EIA s AEO 0 report. Would the results be very different if you compared the Henry Hub natural gas prices that Georgia Power used in the scenarios with the $0/ton, $0/ton or $0/ton CO prices? A. No. Q. What action do you recommend Georgia Power take concerning its natural gas prices assumptions in its Unit Retirement Study? A. I recommend that, like Ameren Missouri, Georgia Power update its natural gas prices assumptions in a revised Unit Retirement Study. The range of natural gas prices that Georgia Power used in its 0 Unit Retirement Study are too high and need to be reduced to a more reasonable range. Q. Should the Commission be concerned that Georgia Power will be unreasonably exposed to natural gas price volatility if coal units other than Branch Units and are retired in the coming years? A. No. I do believe that utilities and regulatory commissions should be concerned about natural gas price volatility. However, as shown in Table, below, Georgia Power is currently heavily dependent on coal to power its generating facilities. 0 Table : Georgia Power s Generation Fuel Mix Given this heavy dependence on coal and the lower percentage of gas in its fuel mix, Georgia Power would not become overly dependent on natural gas even if it were to retire all,00 MW of coal capacity considered in the 0 Unit Page

9 Docket No. 0 0 Retirement Study and entered into the four PPAs for replacement power. For example, the Company s fuel mix in 00 would still have been more than 0 percent coal even if the generation from the units considered in the 0 Unit Retirement Study had been replaced by generation from gas-fired facilities. Thus, replacing these coal-fired units with generation from natural gas-fired facilities would give the Company a more balanced fuel mix. Q. What is your understanding of the size of the domestic U.S. natural gas reserves and how much low cost shale gas can be produced? A. My understanding is that the U.S. has significant economically recoverable U.S. natural gas reserves. The only question about the size that I have seen among reputable sources is the level of increase in shale gas reserves in the U.S. In other words, whether shale gas will mean merely a significant increase in economically recoverable U.S. reserves or whether that increase will be even larger. For example, the new supplies of natural gas that have been identified since 00 have been described (by Entergy Corporation, for example) as a structural change and a seismic shift in the natural gas market. This structural change has two important impacts on utility resource planning. First, as a result of the existing and expected supply glut, current and projected prices of natural gas have been reduced. Second, the dramatically larger domestic supplies of natural gas should be able to accommodate increased demands for any fuel switching due to federal or state regulation environmental regulations without causing significant increases in natural gas prices. Similarly, in early June 00, the American Gas Association and an independent organization of natural gas experts known as the Potential Gas Committee, the authority on gas supplies, released a report concluding that the natural gas Report and Recommendation Concerning the Little Gypsy Unit Repowering Project, submitted by Entergy Louisiana to the Louisiana Public Service Commission, April, 00. Page

10 Docket No reserves in the United States are percent higher than previously believed. The new estimates show an exceptionally strong and optimistic gas supply picture for the nation, according to a summary of the report. A subsequent report in 0 reported a further percent increase in domestic gas reserves and again noted that: Consequently, our present assessment, strengthened by robust domestic production levels and a growing base of proved resources, demonstrates an exceptionally strong and optimistic gas supply picture for the nation. Additionally, a June 0 MIT Study on the Future of Natural Gas similarly concluded that the U.S. has huge reserves of low cost natural gas: There are abundant supplies of natural gas in the world, and many of these supplies can be developed and produced at relatively low cost. In the U.S., despite their relative maturity, natural gas resources continue to grow, and the development of low-cost and abundant unconventional natural gas resources, particularly shale gas, has a material impact on future availability and price. * * * * Globally there are abundant supplies of natural gas, much of which can be developed at relatively low cost. The mean production of remaining recoverable resource in this report is,00 Tcf, 0 times current annual global natural gas consumption, with low and high projections of,00 Tcf and 0,00 Tcf, respectively. Of the mean projection, approximately,000 Tcf could be developed economically with a natural gas price at or below $/Million British thermal units (MMBtu) at the export point. Unconventional natural gas, and particularly shale gas, will make an important contribution to future U.S. energy supply and CO emission-reduction efforts. Assessments of the recoverable volumes of shale gas in the U.S. have increased dramatically over the last five years, and continue to grow. The mean projection of the recoverable shale gas resource in this report is approximately Estimate Places Natural Gas Reserves percent Higher, N.Y. Times, June, 00, available at Potential Gas Committee, April, 0 Press Release, available at Page

11 Docket No Tcf, with low and high projections of 0 Tcf and 0 Tcf, respectively. Of the mean projection, approximately 00 Tcf could be economically developed with a natural gas price at or below $/MMBtu at the wellhead. While the pace of shale technology development has been very rapid over the past few years, there are still many scientific and technological challenges to overcome before we can be confident that this very large resource base is being developed in an optimum manner. Q. Have you seen any filings by utilities and reports that suggest that the increased supplies of natural gas will lead to less volatility in natural gas prices? A. Yes. For example, Xcel Energy explained in its 00 Resource Plan filing to the Minnesota Public Utilities Commission that the dramatically increased supply of natural gas should dampen price volatility: Economically recoverable shale gas has been a major contributor to increasing reserves and declining natural gas prices.. * * * * A long-term lower price for natural gas will produce significant benefits to our customers. It will reduce the production cost at both current and new resources. In addition to lowering the cost of energy from our natural gas-fired facilities, the lower cost of energy is expected to put downward pressure on wind prices, which are a close competitor. Lower natural gas production costs also reduce the integration costs of wind on our system since our ability to follow the wind with flexible gas generation becomes less expensive. Today s natural gas forecasts also predict reduced price volatility. The Commission has expressed concern in the past that more extensive use of natural gas for electric generation would hamper the supply and increase the cost of natural gas for residential heating customers. The substantial increase in supply due to the ability to economically recover shale gas may result in the ability to expand natural gas-fired generation while reducing the cost to all users of natural gas. Still, natural gas is a commodity that comes with some price volatility and the impacts of federal regulations on MIT Future of Natural Gas, June 0, at pages and. Available at Page 0

12 Docket No. 0 0 shale extraction will be a key factor in whether the same level of volatility that we have seen in the past decade returns. A recent report from the Bipartisan Policy Center and American Clean Skies Foundation s Task Force on Ensuring Stable Natural Gas Markets has similarly noted that: Recent developments allowing for the economic extraction of natural gas from shale formations reduce the susceptibility of gas markets to price instability and provide an opportunity to expand the efficient use of natural gas in the United States. And: The currently understood and projected shale gas resource has allowed the United States to project a significant increase in economically recoverable gas resources for the first time in the last years. And for the first time since the 0s, it now appears that deliverability (i.e., available production) could be adequate to meet increasing gas demand, meaning that the United States will no longer be in the tight supply/demand regime that has historically made natural gas markets vulnerable to price instability. Q. Are there other actions that a utility can take to mitigate the risk of natural gas price uncertainty and volatility besides retrofitting aging coal-fired power plants? A. Yes. Many utilities regularly limit their exposure to natural gas price uncertainty and volatility through financial or physical hedging. In addition, energy efficiency (both for electricity and for natural gas) and renewable technologies are reasonable alternatives for limiting dependence on natural gas. Repowering older natural gas-fired units (in particular, older combustion turbines) with newer, more efficient combined cycle technology is another option. Xcel Energy Minnesota 00 Resource Plan, at pages - to -. Available at Id, at page of, available at Id, at page of. Page

13 Docket No. 0 0 Coal Prices Included in the Retirement Study Q. Are there any other biases in the fuel prices that Georgia Power has used in its 0 Unit Retirement Study? A. Yes. The Company has used an REDACTEDxxx range of future coal prices in the 0 Unit Retirement Study with REDAC differences between the low and high coal price forecasts and has assumed that coal prices will REDACTED xxxxxxxx xxxxxx. 0 I believe that a wider range of potential coal prices should have been considered. In particular, there is an increasing emphasis on exporting domestic U.S. coal at the very same time that traditional sources are being depleted. This is expected to lead to upward pressure on coal prices as Central Appalachian reserves are depleted and mining in the Powder River Basin ( PRB ) intensifies due to rising domestic and international demands and reduced supplies at other sources. For example, a presentation by John Drexler, Senior V.P. and C.F.O. of Arch Coal, Inc., at the BMO Capital Markets 0 Global Metals/Mining Conference in February 0, noted the following: Even modest increases in export activity can have significant market implications: Arguably the most significant driver in the 00 market run-up was a million ton increase in exports from 00 to 00. U.S. exports appear to be in the midst of an even greater expansion at present. The market implications of such an increase could prove dramatic. 0 See the Trade Secret CRA Scenario Fuel Forecast Documentation 00 (Revised April 0), at pages to. See, for example, Scott Learn, Mining companies aim to export coal to China through Northwest points, The Oregonian, September, 00. Available at l. In Slide No.. Page

14 Docket No. 0 Indeed, there are indications that intensified mining efforts will lead to rising costs of production even in the PRB. In 00, the U.S. Geological Survey ( USGS ) issued a study of the PRB s Gillette coal beds. This study, which reflected forty years of USGS research on coal reserve methodology throughout the United States, concluded that the methods used by the United States government to calculate coal reserves had significantly overstated the amount of economically recoverable coal. The study explained that as existing mines and new mines in the area are more intensively exploited, production costs would rise substantially, perhaps to a level that could not be covered by the market price. This is an important observation as the Gillette coal bed contains most of the coal produced in the PRB and, overall, accounts for % of the nation s coal production. Coal Plant Operating Performance and Cost Uncertainty Q. How old are the coal units that Georgia Power examined in its 0 Unit Retirement Study? A. As shown in Table, below, only Plant Yates Units and and Plant McIntosh Unit are younger than years old and even Plant Yates Units and are already years old. United States Geological Survey, Assessment of Coal Geology Resources and Reserves in the Gillette Coalfield River Basin, Wyoming, Open-File Report The study offers precise calculations for existing mines in the Gillette coal beds as well as cost curves based on various production levels. These models allow for a dynamic understanding of the relationship between rising costs of production and the need for higher coal prices in the market place. Page

15 Docket No. Table : Ages of Georgia Power Coal-Fired Units Evaluated in 0 Unit Retirement Study 0 Q. How long does Georgia Power assume in its 0 Unit Retirement Study that it will be able to continue to operate these coal units? A. Georgia Power assumes that it will be able to continue to operate each of these coal units until 00, at which time: Plant Branch Units and will be and years old. Plant Yates Units - will be to 0 years old. Plant Yates Units and will be years old. Plant Hammond Units - will be - years old. Plant Kraft Units - will be - years old. Plant McManus Units and will be and years old. Page

16 Docket No. 0 0 Plant McIntosh Unit will be years old. Q. How does Georgia Power s assumption concerning unit life spans (i.e. service lives ) in the 0 Unit Retirement Study compare with the assumptions in its 00 Depreciation Study? A. The Company s Retirement Study assumes a much longer life span than does its Depreciation Study. The Depreciation Study that the Company presented in its 00 rate case explained the following concerning the life spans that Georgia Power assumes for its fossil-fired units: Life span is the time between the initial in-service date of a unit and its date of retirement, or removal from service. The life span used in this analysis was years for the controlled-emission generating facilities of Plant Bowen, Plant Hammond Unit, Plant Scherer, and Plant Wansley, an increase of five years over the life span upon which present depreciation is based. A life span of years was used for Plant Branch and Plant Yates Units -, the same as in the last study. A retirement date of December 0 was used for the not-fully-controlled-emission facilities of Plant Hammond Units -, Plant Yates Units -, Plant McIntosh, and Plant Kraft. The retirement date of Plant Mitchell Unit was extended from the last study by years to 0 due to its anticipated conversion to biomass. The calculated composite average service life, employing direct weighting, of this functional group was 0. years. In sum, the Company does not assume a service life for depreciation purposes for any of its coal units longer than years, even for those units that are controlledemission generating facilities. In the Unit Retirement Study, however, the Company assumes that units at Plant Branch, Yates, Hammond, Kraft and McManus will operate well beyond their th year. Georgia Power Company Depreciation Rate Study As of December, 00, prepared by American Appraisal at, Georgia Public Service Commission Docket No. (filed on July, 00). Page

17 Docket No. 0 0 Q. Does the Company assume that the operating performance of any of these units will degrade as they age? A. No. Without providing any supporting evidence, Georgia Power assumes that there will be no REDACTEDxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx. Instead, in each of its analyses, the Company assumes the xxxxxxxxxxxxxxxxxxxxxxxxxredacted xxxxxxxxxxxxxxxxxxxxxxxxxxxx. Q. Has the Company provided evidence to support its conclusion that these units will continue to operate efficiently as they age? A. No. SACE specifically requested such evidence but did not receive the requested information. Q. Does the Company assume that the cost of operating and maintaining its coal units will increase significantly as they age? A. No. The Company assumes, with some minor exceptions, that the recurring annual operating and maintenance ( O&M ) costs for each coal-fired unit will increase only at the rate of inflation during the period Q. Did the Company provide evidence to support this assumption? A. No. Q. What is the actual operating experience in the U.S. with large (that is, greater than 00 MW) coal units? A. It is my understanding that the oldest operating coal unit of 00 MW or larger in the U.S. is Detroit Edison s Trenton Channel Unit, which is only about years. Consequently, there is no actual experience with units of this size in terms of how their operating performance may degrade or their O&M costs may increase as they age through their 0s, 0s and 0s. Unfortunately, Georgia Power does not Page

18 Docket No. 0 0 reflect this performance and cost uncertainty in its analyses in the 0 Unit Retirement Study. Q. Is Detroit Edison planning to continue to operate Trenton Channel Unit? A. The Company has told the Michigan Public Service Commission that it assuming for planning purposes that the unit will be retired in 0, but that decision is not definite. Alternatives Considered in the Retirement Study Q. Did Georgia Power compare the cost of retrofitting and continuing to operate each of the coal units considered in its 0 Unit Retirement Study to the cost of power under the proposed four PPAs? A. No. Georgia Power compared the costs of retrofitting and continuing to operate each of the coal units to the cost of a self-build option. Q. Do the Company s analyses show that the cost of power under the PPAs would be lower than self-building new units if the existing coal units were retired instead of retrofitted? A. Yes. Georgia Power s response to HR---B clearly shows that the costs of the PPAs are xxxxredactedxxxxxxx the cost of self-building a new combined cycle unit. Q. Should Georgia Power have compared the costs of retrofitting and continuing to operate each of the coal units to the projected cost of purchasing power under the four proposed PPAs? A. Yes. The 0 Unit Retirement Study should have compared the cost of retrofitting and continuing to operate each of the coal units to the replacement option that the Company actually intends to use if each unit is retired rather than Direct Testimony of Angela P. Wojtowicz on behalf of the Detroit Edison Company, Michigan Public Service Commission Case No. U-, at page, lines -. Available at Page

19 Docket No to some hypothetical alternative that will not be built. This could have an important impact on the results of the analyses in the Unit Retirement Study. 0 Unit Retirement Study Results Q. What is your conclusion concerning the Retirement Study and its results? A. I have described four biases in the Company s 0 Unit Retirement Study: () the use of a very high range of natural gas prices, () the failure to consider any significant uncertainty in the future price of coal, () the failure to consider any uncertainty in the future operating performance and operating costs for each of the coal units, and () the failure to compare the cost of retrofitting and continued operation to the projected cost of power under the PPAs that actually would be used in place of any retired units. These biased assumptions favor coal in the analyses. Nevertheless, in many scenarios, the retirement of each coal unit is still the lower cost alternative or is xxxxxxxxxxxxxxxxxx than the option of retrofitting and continuing to operate the unit. For example: Retirement of Plant Branch Units & is the lower cost option with the Company s low natural gas prices (which are not that low) for both the 0 and 0 compliance analyses. Retirement of Units & also is the lower cost option with the $0 and $0 CO prices. Retirement of Plant Yates Units & is the lower cost option with the Company s low natural gas prices (which are not that low) for both the 0 and 0 compliance analyses. Retirement of Units & also is the lower cost option with the $0 and $0 CO prices. Retirement of Plant Yates Unit is the lower cost option in all scenarios in both the 0 and 0 compliance analyses except for the extreme scenario with high fuel prices and a $0 CO price. Retirement of Plant Yates Units - is similarly the lower cost option in all scenarios in the 0 compliance analysis except for the Moderate and High fuel price scenarios with a $0 CO price. Moreover, the retrofitting and continued operation of Units - is redacted less expensive than retirement in the Moderate Fuel. In the 0 Compliance Analysis, retirement of Plant Yates Units - is the lower cost option in all scenarios except for the High Fuel Price/$0 CO price scenario. Page

20 Docket No. 0 0 Q. What about the results for Plant Hammond Units -? A. Continued operation of Plant Hammond Units - is generally the lower cost option in all scenarios in the 0 and 0 Compliance Analyses except for the Low Fuel Price/$0 CO price scenario. Q. What about the results of the fuel switching analyses for Plant Kraft Units -, Plant McIntosh Unit, and Plant McManus Units and? A. The Company s analyses show that fuel switching is the lower cost option in all scenarios examined but it is not clear what the results would be with more reasonable assumptions concerning natural gas prices, future unit operating performance and costs, and the actual alternative to continued operation (that is, including PPAs in the comparison instead of the self-building of a new unit). Q. What impact would correcting for the flaws you have found have on the results of the Company s 0 Unit Retirement Study? A. Although I have not performed the analyses, I believe that, in general, retiring each of the coal units would likely be an even more economic option in a larger number of scenarios if the flaws that I have identified are fixed. The Potential for Excess Generating Capacity Q. What will be the impact on the reserve margins if the Company purchases capacity from the four proposed PPAs but fails to retire a significant amount of coal unit capacity in addition to Plant Branch Units and? A. As shown in Table below, the Company would have xxxxxxredacted through 0 if it enters into the PPAs and does not retire the approximately,0 MW of coal-fired capacity associated with Plant Branch Units and and Plant Yates Units -. Page

21 Docket No. Table : Georgia Power s Need for Capacity with Branch Units and, Yates Units - and the Four Proposed PPAs. TABLE REDACTED 0 Consequently, the Commission should not certify the PPAs unless the Company has made a decision to retire these units because, if the coal units are not retired, ratepayers will be forced to pay for both the costs of the PPAs and the costs of retrofitting the coal units. Conclusions Q. What is your recommendation concerning Georgia Power Company s 0 Unit Retirement Study? A. Georgia Power should present new economic analyses that () reflect a lower range of natural gas prices; () reflect a wider range of potential coal prices; () examine the potential impact of degrading operating performance and significantly higher O&M costs due to unit aging; and () compare the cost of retrofitting and continuing to operate each coal unit with the cost of power under the proposed PPAs instead of site specific alternatives. Page 0

22 Docket No. 0 Q. Should Georgia Power include all of its existing coal-fired units in such a revised Unit Retirement Study? A. Yes. The Company should perform unit retirement studies for all of its existing coal units, especially given the expectation that natural gas prices will remain low for the foreseeable future. Q. Do you agree with the Company s conclusion that Plant Branch Units and should be decertified? A. Yes. The Company s economic analyses show that retiring Plant Branch Units and is the lower cost option (that is, has a lower NPV) with the xxxxxxxxxxxxx xxxxxxxxxxxxxxxxxx REDACTED. Q. Does this conclude your testimony? A. Yes. Page