BEFORE THE WASHINGTON UTILITIES AND TRANSPORTATION COMMISSION In the Matter of the Investigation on the Commission's Own Motion into the Propriety and Adequacy of Certain Depreciation Rates of US WEST COMMUNICATIONS, INC. And the Changes, if and, That Should be Ordered to Such Depreciation Rates. Docket No. UT-951425 PREFILED SUPPLEMENTAL AND REBUTTAL TESTIMONY OF CHARLES W. KING ON BEHALF OF PUBLIC COUNSEL SECTION OFFICE OF THE ATTORNEY GENERAL AND TELECOMMUNICATIONS RATEPAYERS FOR COST-EFFECTIVE AND EQUITABLE RATES (TRACER) July 23, 1997 TABLE OF CONTENTS I. Introduction 1 II. Supplemental Testimony 2 III. Direct Testimony of Lawrence K. Vanston 5 IV. Direct Testimony of William R. Easton 26 V. Direct Testimony of Robert G. Harris 37 EXHIBITS Revised Exhibit_____(CWK-4) US WEST Depreciation Rates Revised Exhibit_____(CWK-11) Reserve Deficiencies Exhibit_____(CWK-12) Response to WUTC Request No. 02-0007 Exhibit_____(CWK-13) Metallic Cable Retirement Ratios SUPPLEMENTAL AND REBUTTAL TESTIMONY OF CHARLES W. KING I. Introduction Q. Please state your name, address and position. A. My name is Charles W. King. My office address is Suite 410, 1220 L Street, N.W., Washington, DC 20005. I am President of the economic and management consulting firm of Snavely King Majoros O'Connor & Lee, Inc. Q. For whom are you appearing in these proceedings? A. I am appearing on behalf of the Public Counsel Section of the Office of the Attorney General of the State of Washington and on behalf of Telecommunications Ratepayers for Cost-effective and Equitable Rates ("TRACER"). Q. Are you the same Charles W. King who submitted direct testimo-ny in this docket on behalf of Public Counsel and TRACER on June 23, 1997? A. Yes. I am. Q. Did that testimony contain a description of your qualifica-tions and experience? A. Yes. It did. Q. What is the purpose of your supplemental testimony? A. The purpose of my supplemental testimony is to update two of the exhibits in my initial direct testimony based on responses received from US WEST Communications, Inc. ("US WEST") after WUTC Docket No. UT-951425 Supplemental and Rebuttal Testimony of Charles W. King that testimony was filed. Q. What is the purpose of your rebuttal testimony? A. The purpose of my rebuttal testimony is to respond to the testimony of the US WEST witnesses in this case, Lawrence Vanston, William Easton and Robert Harris, that was filed simultaneously with mine, on June 23, 1997 II. Supplemental Testimony Q. What was the data request to which US WEST responded after your initial testimony? A. Data request no. 02-0007 was as follows: Please provide a calculation of US WEST Washington plant and equipment remaining lives as of 1/1/97, calculated in accordance with the methodology approved by the Commis- sion in Docket UT-940641, assuming the projection lives shown in Ex.___(TLS-2) filed in this proceeding, Docket UT-951425. Q. What was US WEST's response? A. US WEST's response, transmitted on June 27, 1997, is attached hereto as Exhibit_____(CWK-12). It provides the remaining life, depreciation reserve percentage, future net salvage and depreciation rate applicable to the plant on January 1, 1997 for each account under currently prescribed rates, under currently prescribed live and salvage parameters (Attachment A), under the FCC approved parameters (Attachment B), and under the Staff recommended parameters (Attachment C). Q. Why does this response require a revision to your earlier exhibits? A. In preparing my Exhibits_____(CWK-4) and _____(CWK-11), I was forced to estimate the reserve percentages and remaining lives and as of January 1, 1997 by trending forward the January 1, 1996 reserves and FCC remaining lives (which I had from earlier data responses) based on assumptions as to retirements and additions during 1996. The January 1, 1997 reserve percentages and remaining lives are critical to the develop- ment of the remaining life depreciation rates shown in Exhibit_____(CWK-4) and the theoretical reserves and reserve deficiency estimates shown in Exhibit_____(CWK-11). For this reason, I am now required to revise my earlier exhibits to reflect the new, actual data provided by US WEST. Additionally, my earlier Exhibit_____(CWK-4) neglected to recognize that in addition to depreciation, the Company is collecting $22.3 million annually in the reserve deficiency amortization approved by the Commission in Docket No. UT-940641. The revised exhibit shows that amortization and the effective composite depreciation and amortization rates under three sets of projection lives and future net salvage parame- ters. Q. What do these revised exhibits show? A. Revised Exhibit_____(CWK-4) shows that the composite deprecia- tion rate under the life and salvage parameters proposed by the Staff in Exhibit_____(TLS-2) is 7.0 percent. This is a downward revision from the 7.1 percent shown in my initial testimony.Even though it is derived directly from US WEST data, this revised composite rate still does not reconcile with the 6.9 percent rate alleged by US WEST witness William Easton on page 2 of his testimony. Additionally, revised Exhibit_____(CWK-4) shows that the composite depreciation and amortization rate under the Staff's projection life parameters is 7.5 percent, equal to the national average rate cited on page 45 of Mr. Easton's direct testimony (WRE-T). The composite rate under the FCC parame- ters is 8.3 percent, well above the national average and higher than 9 of the 13 other US WEST states as reported on page 44 of Mr. Easton's testimony. As noted in my initial testimony, depreciation reserve deficiencies are extremely volatile because they are affected by minor changes in remaining lives (increases in remaining lives reduce the deficiency and vice versa). This character- istic is demonstrated by the revisions in Exhibit _____(CWK-11). Page 1 of the revised exhibit shows that there is a reserve excess under present life and salvage parameters of $120.5 million, down from the $160.1 million shown in my initial exhibit. Page 2 of revised Exhibit_____(CWK-11) shows a dramatic reversal in the reserve condition under the Staff's life and salvage parameters from that initially estimated. My June 23 exhibit had indicated that if Staff's lives and salvage parameters were adopted, there would be a reserve deficiency of $100 million. Now, it appears there would be a reserve excess of $24.5 million. Finally, revised page 3 of Exhibit_____(CWK-11) shows that the reserve deficiency under the FCC life and salvage parameters would be only $33.8 million, which is a consider- able reduction from the $220 million initially estimated in my June 23 testimony. III. Direct Testimony of Lawrence K. Vanston (LKV-T) Q. Please describe the testimony of Lawrence K. Vanston. A. Dr. Vanston sponsors the studies that purport to underlie the projection life parameters proposed by Mr. Easton. The latest of those studies is attached to Mr. Easton's testimony as Exhibit WRE-2. Dr. Vanston's basic thesis is that technology advance, new services, regulatory policy and competition are rendering the existing local telephone network obsolete. Technological obsolescence, he argues, tends to create an "avalanche" of retire-ments that cannot be predicted by traditional mortality analysis. In Dr. Vanston's opinion, such analysis is there- fore irrele-vant and results in service life estimates that are too long. In lieu of mortality analysis, or any reference to historical data, Dr. Vanston proposes that Fisher-Pry substi- tution analysis be employed as the basis for estimating service lives. Fisher-Pry forecasts the remaining life of existing plant based on estimates of the rate of substitution and a forecast of the time when half the old technology has been replaced. Dr. Vanston surveys the factors of technology change, competition and new services demand that he believes will cause the early retirement of virtually all copper cables, digital and analog circuit equipment, and digital switching plant. Dr. Vanston then discusses each of the major categories of plant. He presents three scenarios for the retirement of copper cable and expresses the view that US WEST will likely experience the early scenario, which predicts that 50 percent of all copper will be retired by 2002, 75 percent by 2004 and 100 percent by 2011. On this basis, he believes that US WEST's proposal for a 15 year projection life for aerial and underground copper cable and 20 years for buried copper is conservative, that is, possibly too long. Dr. Vanston believes that the introduction of Synchronous Optical Network ("SONET") will reduce the remaining life of digital circuit plant to 3.7 years, which implies that US WEST's projection life of 10 years for this account is also conservative. With respect to switching plant, Dr. Vanston believes that the introduction of Asychronous Transfer Mode ("ATM") switching will render existing digital switches obsolete within the next decade. He predicts that on January 1, 1995, the average remaining life of this plant was 6.3 years, which he believes is consistent with US WEST's proposed projection life of 10 years. Dr. Vanston concludes with a rebuttal of Mr. Spinks' life estimates. He asserts that Mr. Spinks' analysis is backward looking and fails to recognize the new influences of techno- logical change, competition, and new services. Q. What is your general assessment of Dr. Vanston's substitution analysis? A. Dr. Vanston's substitution analysis is simplistic and biased toward underestimating the service life of plant. Its basic fallacy is the assumption that the effect of each new technol- ogy is to replace an old technology and cause its retirement. While this relationship holds in some instances, it is by no means universal. The reality is much more complex. Contrary to Dr. Vanston, technological advances often contribute to lengthen- ing service lives by augmenting the capacity of existing plant so that it can continue to meet new service requirements. Moreover, even when one technology does replace another, the effect may not be to retire the old plant but rather to reassign it to uses for which it is better suited. Q. Can you provide examples of technologies that augment the capacity of old plant? A. Yes. Consider, for example, loop carrier systems. The first of these systems, the analog SLC 96, permitted 96 individual loops to be handled over five copper wire pairs, thus expand- ing enormously the capacity of the existing cables. The effect of this technology was to increase the life of copper cables that otherwise would have been replaced by larger cables. SLC 96 systems have since been replaced by digital carrier systems that provide even more capacity on existing copper cables, permitting them to remain in use for longer periods of time. Moreover, a yet newer technology, Asynchronous Digital Subscriber Line, ("ADSL"), now permits existing copper cables to provide broadband service up to 6 Mbps and may permit signals up to 55 mbps.See footnote 6, page 15, direct testimony of C.W.King. This technology not only increases the carrying capacity of existing copper cables but allows them to compete with AT&T's 128 Kbps wireless internet access service and may permit them to provide video services. Q. Can you provide examples of technology replacements that have resulted in reassignments of plant, rather than retirements? A. A good example is that cited by Dr. Vanston at page 33 of his testimony, the replacement of copper interoffice cables with fiber cables using SONET transmission technology. Dr. Vanston's 1994 study indicated that fiber had virtually replaced copper for interoffice applications.See Transforming the Local Exchange Network, by Lawrence K. Vanston, 1994 Edition, Exhibits C-4 through C-9. pages 167-173. Yet, these copper interoffice cables were not retired. Instead, they were reassigned to short-haul feeder applications for which copper continues to be most efficient. Long-haul copper feeders have also been replaced by fiber, but again the effect has not been to retire the copper plant. Rather, the copper feeders have been reassigned to distribution service where they continue to be the most cost-effective technology. The evidence of this reassignment is provided in Exhib- it_____(CWK-13), which shows both graphically and with data the ratios of copper cable plant retirements to plant in service in US WEST's Washington service territory. These ratios have been less than .01 (1 percent) for the composite of all copper plant throughout this period. The aerial and underground copper cable ratios have declined, and the buried cable ratios have remained stable for the last three years. These negligible and declining ratios are observed during a period when fiber has been replacing copper for both interof- fice and feeder applications. Dr. Vanston's substitution analyses erroneously assumes that substitution of fiber for copper-based technologies would result in the retirement of copper plant. The result is the extraordinary forecasting errors shown in Exhibits_____(CWK-9) and_____(CWK-10) and discussed on pages 24 and 25 of my direct testimony. Q. Is there any validity to Fisher-Pry substitution analysis as a forecasting tool? A. Yes. Fisher-Pry is probably a valid forecasting tool under two conditions. First, the hypothesis of substitution must be reasonable, that is, there must be evidence that the effect of the new technology is to replace fully the old technology. Second, the substitution must have progressed to the point where its rate can be measured. When these two conditions are met, then Fisher-Pry is probably a good template for extrapo- lating the future course of the substitution. These two condition have held in each of rather limited examples where Dr. Vanston purports to have successfully mapped the replacement of one technology with another. For example, at page 14 of his testimony, Dr. Vanston boasts that in 1989 he accurately forecasted the rate at which digital switching would replace analog switching during the next six years. However, in 1989, it had become fairly evident that the manufacturers of analog switches would no longer support them and, more to the point, about 45 percent of access lines had already been converted from analog to digital switches. The substitution was thus well under way, and Dr. Vanston successfully extrapolated it through six additional years. The same is true of the other instances where Dr. Vanston purports to have developed quantitative forecasts of the substitution of one technology for another. The following table shows the proportion of substitution that had already taken place when Dr. Vanston prepared his 1994 extrapolations for which he has provided statistical reliability data in the GTE case, Docket No. UT-961632: Substitution RefPage number in Transforming the Local Exchange Network, 1994 Edition. 1995 % Substituted Interoffice (IO) Fiber/Baseband 168 92.6% IO Fiber/Digital Carrier 171 82.0% IO Fiber/Analog Carrier 173 99.8% Feeder Fiber/Baseband 183 15.0% Interoffice SONET 85 32.4% Feeder SONET 87 11.9% Digital/EM Switching 194 98.1% Digital/Analog SPC Switching 192 68.9% Q. On what specific substitutions does Dr. Vanston base the life forecasts he advocates in this docket? A. Dr. Vanston plant life recommendations are based primarily on three substitutions: • Fiber cable for copper cable, • ATM switching for digital switches, and • SONET technology for circuit equipment. Q. Do the two conditions you describe, reasonable expectation of substitution and a historical record of substitution, apply to the first of these alleged substitutions, fiber for copper cable? A. No. As I have noted, there is no evidence to date that the deployment of fiber optic cable has resulted in the retirement of any significant amount of copper cable plant in US WEST's network, or that it will anytime soon. To date, the effect of fiber deployment has been to reassign copper cable to other applications, principally distribution cables and short haul feeder routes. There is no current technological or cost justification for fiber to replace copper in these applica- tions. The only reason that fiber might replace copper for distribution purposes is to provide broadband services such as high-speed data and video. As discussed above, the ADSL technology allows copper to provide higher-speed data service without its removal. As regards video, I pointed out in my initial testimony that the plans of all of the Bell Companies and GTE for widespread video services have been canceled. US WEST withdrew its applications for five video dialtone systems in mid-1995.US WEST has indeed entered the cable television market through the merger with Continental Cablevision, Inc., but by means of conventional cable TV systems. Because fiber has been deployed as a complement to, rather than as a substitute for copper cable, there is no record of substitution that provides the basis for estimating the rate at which one technology will replace the other. Exhibit_____(CWK-13) demonstrates that the substitution of fiber for copper, if it occurs at all, has not yet begun. With no track record of substitution, there is no trend or pattern from which a future rate of substitution can be extrapolated. Q. If no fiber has yet been substituted for copper, how does Dr. Vanston develop his forecasts of the displacement of copper cable by fiber? A. Dr. Vanston's predictions concerning the replacement of copper cable by fiber are based on his speculations as to effects of technology, competition, new services development and changing regulation on cable plant deployment. There is no quantita- tive basis whatever for his forecasts of copper replacement and retirement. Q. Are the two conditions you have described, an expectation of substitution and a record of early substitutions, evident with respect to digital switching? A. No. There is no expectation of substitution in switching because there is no evidence of a substitute technology. Dr. Vanston refers to ATM technology as the "next major switching generation," yet he lists it as a "key driver" for elements that make up only 13 percent of switching investment.Vanston testimony at 48. Retirement of the rest of the elements are driven by "life cycle," the deployment of SONET, digital line carriers, and fiber to the loop. As demonstrated on page 1 of Exhibit_____(CWK-5), the most recent life cycle indications for digital switching suggest a projection life of 28.3 years. Dr. Vanston does not explain how SONET and digital line carriers, both of which have been deployed for a number of years, would hasten the retirement of the central processors or the line interfaces. As for fiber in the loop, it has effectively disappeared as a deployment strategy with the cancellation of US WEST's plans to offer integrated voice and video services. More to the point, Dr. Vanston testifies that the average remaining life of the digital switching plant in service on January 1, 1995 was 6.3 years. It is now 2.5 years later, so that the 1995 plant base should have only 3.8 years of remaining life. If so, one would expect the much-touted "avalanche" of digital switch retirements to be on the point of descending. Yet, in response to a data request, US WEST could list only three switches that are scheduled for retire- ment, all of them analog. Since almost no digital switches have been retired,US WEST has retired three DMS-10 digital switches. These are very small, early generation digital switches that cannot easily accommodate equal access. and none are scheduled to retire, there is little record on which to project a rate of retirement or a mid-point in the replace- ment cycle, which are the two quantitative inputs required for a Fisher-Pry substitution analysis. Q. If there are no data on digital switch retirements, how did Dr. Vanston develop his remaining life estimates for this account? A. The entire quantification is presented on page 48 of his testimony. Dr. Vanston has separated digital switching investment into six components. He has then speculated as to the reasons for the replacement of each segment and assumed a remaining life for each segment. The 6.3 year remaining life for the account is a composite of this process. While a great amount of discussion underlies these component estimates, none of it reflects verifiable, objective data. Q. Are the two conditions you have described, an expectation of substitution and a record of early substitutions, evident with respect to digital circuit equipment? A. The replacement of copper-based digital circuit equipment with fiber-based SONET technology follows the replacement of copper cables with fiber cables. That is because SONET is the transmission standard for the carrier systems that operate on fiber cables. Thus, if a copper cable is replaced by a fiber cable, the digital circuit equipment used to multiplex a number of channels into a single signal is likewise replaced. As noted earlier, fiber is replacing copper for certain applications, but it is not causing its retirement. Rather, it is resulting in the reassignment of copper cables to other applications. In some cases, this reassignment may result in the removal of the copper-based circuit equipment.This would be the case, for example, when a feeder cable equipped with digital carrier equipment is reassigned to distribu- tion application, in which each wire pair serves a single end-user circuit. Thus, copper-based circuit plant is being replaced by fiber-based equipment, and there is a reasonable expectation that this replacement will continue. This expectation is effectively recognized by all of the parties in this docket. Each calls for a much shorter life for circuit equipment than for cables: 12 years by the Staff, 11 years by the FCC, and 10 years by the Company. Unlike copper cables and digital switches, there is some track record for the replacement of copper-based digital circuit plant by fiber-based equipment. That is because fiber circuit equipment, whether SONET or not, is mutually exclusive from the circuit equipment used on copper cables. Fiber-based circuit equipment has been replacing copper- based equipment ever since the initiation of fiber as a transmission medium. The record of past retirements does not, however, support any of the projection life proposals in this docket. Page 2 of Exhibit_____(CWK-6) shows that the average expired life of the digital circuit equipment retired during the three years 1992-1994 was 29.2 years. Clearly circuit equipment retire- ments have yet to catch up with the expectations of replace- ment of any of the parties. The Company has apparently relied upon Dr. Vanston's estimate that the projection life for circuit plant should be in the range of 6 to 9 years.Vanston testimony at 46. Its 10 year proposed life is thus "conservative." However, Dr. Vanston's circuit equipment analysis is seriously flawed by his apparent failure to recognize one critical characteristic of the accounting for this type of plant. Unlike other types of plant, cables and switches for example, when circuit equipment is removed from one location it is not all retired. The "hardwire" casing and connections are retired, but the modular "plug-in" units are not. They can be reinstalled in other locations, so they continue in service. These plug-in units, which are depreci- ated on a "cradle-to-grave" rather than location life basis, account for the majority of the cost of a typical digital carrier system. Dr. Vanston's recommended average remaining life of 3.7 yearsId. cannot have recognized this characteristic. This flaw in Dr. Vanston's analysis is similar to his failure to recognize the reassignment of copper cable: replacement does not mean retirement. In each case, the replaced plant lives on, continues in use, generates revenue, and incurs deprecia- tion expense. Q. You have criticized Dr. Vanston for not basing his life estimates on objective, verifiable data. What data are available? A. The most objective and verifiable data available are the records of recent retirements. Beyond that, there are the construction and retirement plans of the Company. Q. What is Dr. Vanston's reaction to the use of such data in estimating lives? A. Dr. Vanston refers to any recognition of past or current retirement activity as "the traditional method of estimating depreciation lives." He asserts that past retirement data provides no useful evidence of the likely pattern of future retirements. To support this thesis, he cites the example of the replacement of crossbar by electronic switching, which occurred in an avalanche of retirements in the late 1970s and early 1980s. He believes that any reference to past retire- ments would likely repeat the error that would have resulted had crossbar lives been prescribed on the basis of prior retirements in the 1970s: a disastrous understatement of depreciation. Q. What is your reaction to this position? A. I agree with Dr. Vanston that life indications based on past retirements should not be slavishly employed to predict future life expectancy. No one is proposing to do so. With the sole exception of fiber submarine cable, all of the projection life proposals by Staff are significantly shorter than the most recent life indications shown in US WEST's depreciation studies.See initial direct testimony of C.W.King at 9. Dr. Vanston is flatly incorrect in his assertion at page 52 of his testimony that Mr. Spinks has looked at only one variable, that of past retirements. If that were the case, Mr. Spinks would have proposed much longer lives. The value of past retirements is that they provide at least some check on the initial trends that can be input into the Fisher-Pry substitution model. In my Exhibits_____(CWK-9) and_____(CWK-10), I attempted to do that with respect to Dr. Vanston's 1994 forecasts of the retirement of digital switch- ing and copper cable plant. As noted in that testimony, I found forecasting errors up to 90 percent. These errors suggest that Dr. Vanston has exaggerated the rate at which these technologies will be subject to obsolescence. While Company construction plans are less objective and verifiable than past retirements, they have the virtue of looking into the future. Dr. Vanston insists that only a future-oriented examination is relevant to establishing account service lives. Yet, even here, there is no support whatever for Dr. Vanston's claimed "avalanche" of retirements. To the contrary, retirements are predicted to decline, not increase, through 2001.Id. at 19. Instead of relying on any objective and verifiable data whatever, Dr. Vanston would have this Commission base its service life prescriptions on rhetoric. The rhetoric is Dr. Vanston's extensive speculations as to technological, competi- tive and new service developments occurring in the first and second decades of the next century. There is no way to test this rhetoric. It cannot be challenged because it relates to developments beyond the current planning horizon of any telephone company. But the key point is this: it cannot be verified either. It is simply not subject to objective scrutiny. Q. What is your recommendation with respect to Dr. Vanston's testimony and studies? This Commission is being asked to prescribe depreciation lives that have a revenue requirement impact measured in the hundreds of millions of dollars. It cannot prescribe depreci- ation parameters resulting in double-digit rate increases that are based solely on rhetorical speculations that have no foundation whatever in objective, verifiable data. For this reason, I recommend that Dr. Vanston's testimony and support- ing studies be given no weight in the Commission's consider- ation of projection lives. Q. At page 17 and again at page 51 of his direct testimony, Dr Vanston asserts that the FCC and several state commissions have accepted the Fisher-Pry substitu-tion methodol-ogy. Does this testimony accurately reflect the extent of acceptance of the TFI studies by regulators across the country? A. No. The FCC might have cited Fisher-Pry as an analysis technique, but it certainly has not adopted TFI's service life projections. The FCC has prescribed ranges of service lives for each plant account within which it will regard individual company parameters as reasonable. A comparison of the FCC projection life ranges with those of TFI is as follows: Projection Lives Account FCC RangesFCC Docket No. 92-296, Orders released 6/28/94 and 5/4/95. TFI Ranges"Depreciation Lives for Telecommunications Equipment: Review and Update", by Lawrence K Vanston and Ray L. Hodges, Technologies Futures, Inc, (Exhibit WRE-2), page 33. Digital Switching 16 - 18 9 - 11 Digital Circuit 11 - 13 8 - 9 Metallic Aerial & Buried Cable 20 - 26 14 - 16 Metallic Underground Cable 25 - 30 14 - 16 Fiber Cable 25 - 30 15 - 20 On May 7, 1997, the FCC rejected the TFI projection lives that had been proposed by the United States Telephone Associa- tion for use in the calculation of the Total Factor Productiv- ity variable in its price cap formula.FCC CC Docket Nos. 94-1 and 96-262, para. 62. Nor can TFI claim widespread acceptance of its projection lives at the state level. To begin with, TFI was forcefully rejected by this Commission in Docket No. UT-940641: The Technology Futures, Inc, studies on which the Company relies for revising the service lives of seven categories of plant are not an adequate basis for revising those lives. The studies do not relate to U S WEST. They are generic to the entire industry. Moreover, the model used by Technology Futures, Inc., called the Fisher-Pry model, was developed to estimate the plant life spans based upon forecasts of the rate of change in the adop- tion of substitute technologies. They are based largely on conjecture, subjective assumptions, and assertions as to rates of plant obsolescence, technological innovation, and new service require- ments that are incapable of test or verification. The model uses only a single variable to explain rates of change; it is not a sophisticated econome- tric model. No probability statistics are provided with the forecasts. No regulatory enti-ties use the forecasts to determine service lives for existing plant.Docket No. UT-940641, Fifth Supplemental Order, On Remand, Granting Petition in Part, April 11, 1996, page 22. In