ML19343B838
| ML19343B838 | |
| Person / Time | |
|---|---|
| Site: | Allens Creek File:Houston Lighting and Power Company icon.png |
| Issue date: | 12/18/1980 |
| From: | Perl L NATIONAL ECONOMIC RESEARCH ASSOCIATES, INC. |
| To: | |
| Shared Package | |
| ML19343B832 | List: |
| References | |
| NUDOCS 8012300682 | |
| Download: ML19343B838 (56) | |
Text
.
I DIRECT TESTIMONY OF DR. LEWIS J. PERL ON BEHALF OF HOUSTON LIGHTING & POWER COMPANY RE ENERGY ALTERNATIVES l
i 803g39,,Q Q
l l
1 TESTIMONY 2 OF LEWIS J. PERL 3 ON BEHALF OF 4 HOUSTON LIGHTING & POWER CO.
5 6 Q. Would you please state your name ud address?
7 A. My name is Lewis J. Perl. My business address is Five World Trade 3 Center, New York, New York.
9 Q. What is your present employment?
10 A. I am Senior Vice President of National Economic Research Associates, 11 Inc., an economic consulting firm specializing in the economics of energy,.the 12 environment, antitrust, and labor.
13 Q. Would you briefly describe your educational and employment back-14 ground prior to your association with National Economic Research Associates, 15 Inc.?
16 A. I received my B.S. degree in industrial and labor relations from Cornell 17 University in 1963. I received my M.A. degree in 1968 and my Ph.D. degree in 18 1970, both in economics from the University of California at Berkeley. From 19 1968 and to 1972, I taught economics at the New York State School of 20 Indust.e i 'nd Labor Relations at Cornell University. Both in teaching and in 21 my graduau program, my fields of specialization were labor economics, 22 econometrics and economic history.
23 Q. Would you describe the nature of your consulting work with National 24 Economic Research Associates,Inc.?
25 26 I
nera' !
1 A. National Economic Research Associates, Inc. (NERA) was established 2 in 1961 to offer economic consulting services in a number of fields, with 3 particular emphasis on regulated industries and their problems.
4 Since joining NERA in 1972, I have been responsible for a variety of 5 studies relating to the economics of energy and the environment. These 6 include numerous analyses of the comparative costs of coal and nuclear 7 generation for presentation in licensing and other regulatory proceedings. In 8 addition, I have done analyses of the costs of air and water pollution control 9 for the electrie utility industry, as well as various comparisons of the costs 10 and benefits of specific environmental programs. I have directed a number of 11 econometric studies of the elasticity of demand for electricity, the elasticity 12 of demand for telephone service, and the price of uranium. I have been 13 involved in the development of a model for projecting coal prices and a variety 14 of studies evaluating the economics of alternative energy sources.
15 Q. Have any of your writings been published?
16 A. I am the author of numerous economic papers which have appeared in 17 such professional journals as the Qua-te, Journal of Economics, the Journal 18 of Human Resources, and such publications as Nuclear News, Nations' 19 Business, The Wall Street Journal, The New York Times, and The Los Angeles l 20 Times. In addition, I have given speeches on energy .md environmental related 21 topies before the Atomic Industrial Forum and the World Energy Conference.
22 Q. Of what professional, honorary societies and industry groups are you a 23 member?
24 A. I am a member of the American Economic Association and the Air 25 Pollution Control Association.
26 Q. Dr. Perl, what is the purpose of your testimony?
n e:r $
l l
1
. l l
I 1 A. The purpose of my testimony is to examine the need for and-2 desirability of the Allens Creek nuclear plant as a component of Houston 3 Lighting & Power's (HL&P) generation expansion plan in the next decade. Sly 4 examination of need consists of two basic components. First, I have examined 5 the expected cost of electricity from the Allens Creek facility and from a 6 number of alternative coal-fired electric generating plants. The objective of 7 this part of my testimony is to estimate the savings, if any, to consumers from S constructing Allens Creek and to determine the sensitivity of these savings to 9 uncertainties as to the costs and operating characteristics of both coal and 10 nuclear facilities. In the second part of my testimony, I have attempted to 11 examine the impact of uncertainties as to demand growth and the effect of 12 conservation policies on the need for and desirability of the Allens Creek 13 facility.
14 Q. Would you summarize the results of your analysis?
15 A. Yes. Sly comparison of Allens Creek with coal-fired alternatives 16 suggests that, using the most likely assumptions as to each of the components 17 of cost, the cost of electricity from Allens Creek would be significantly 18 cheaper than that from either subbituminous or lignite coal plants. While 19 there were substantial uncertainties in the costs of both Allens Creek and a 20 coal-fired alternative, the exoected cost of electricity from the Allens Creek 21 facility was still less than those of any of the coal-fired alternatives 22 considered. With respect to the subbituminous alternative, the differences in 23 expected costs were substantial, but the lignite alternative had only slightly 21 higher expected costs than Allens Creek. Thus, assuming that HL&P needs to 25 add some baseload capacity over the next 10 years, there woald be savings to 26 corsumers from building Allens Creek; if significant lignite reserves are ne r.n L
l l
l 1 feasible these savings are modest, but if subbituminous fuel must be used the 2 savings would be substantial.
3 Since there are substantial uncertainties in both coal and nuclear 4 generating costs, a wide range of costs is possible whichever option is 5 selected, and cost uncerteinties are somewhat greater for nuclear than for 6 coal alternatives. As a consequence, no single choice of a generating option 7 provides absolute assurance of minimum cost. Given these uncertainties, a 8 utility trying to maintain low cost and also reduce uncertainty would develop a 9 generation expansion plan containing some mix of coal and nuclear capacity.
10 The second component of this analysis indicated that the need for and 11 the economic desirability of the Allens Creek facility were insensitive to the 1
12 uncertainties in load forecasting or the success of conservation and load man-13 agement policies. Because of the very substantial dependence of HL&P's 14 current system on oil- or gas-fired capacity and the very high expected costs 15 of those fuels, there would be substantial economic advantages to the 16 construction of the Allens Creek facility, even if conservation policies were 17 sufficient to eliminate all of the growth in demand expected over the next i
18 decade. This would be true even if conservation programs significantly 19 reduced system load factors. Moreover, load management policies, to the 20 extent they were successful in shifting peak demand to off-peak periods, 21 would increase the economic savings from the construction of Allens Creek.
22 The cost-minimizing generation expansion plan for HL&P includes an i
23 enormous expansion of new baseload capacity, either coal- or nuclear-fired.
24 These capacity additions would be required to meet load growth and to reduce l
25 the system's uneconomic dependence upon oil and natural gas. It seems 26 unlikely that all of these capacity requirements can be met, even if Allens nie r a'.
1 Creek and all feasible coal additions are made. This reflects both environ-2 mental and fuel supply limits on coal expansion. If, as seems likely, the 3 expansion of coal capacity is limited, the development of Allens Creek is best 4 viewed as a means for reducing dependence on oil and gas. In this context, the 5 economic advantages of constructing this plant are substantial and implausibly 6 large increases in the capital costs of constructing this facility would be 7 required to reverse this economic advantage.
8 Q. Dr. Perl, can you summarize your analysis of the comparative costs of 9 Allens Creek and electricity from coal-fired plants?
10 A. Yes. This analysis involved comparing the levelized cost of electricity 11 from the Allens Creek plant, assuming completion in 1989 with the costs of 12 electricity from either a subbituminous or a lignite coal plant completed in the 13 same time frame. The results of this analysis are described in Figure 1.
14 Capital costs, nonfuel operating and maintenance (O&M) costs and capacity 15 factors for each of these alternative facilities were derived based upon a 16 statistical analysis of data for plants completed over the last decade. Fuel 17 costs for both Allens Creek and each of the coal alternatives were derivec ist from coal and uranium demand and supply models developed by NERA.
19 Using the most likely values for the cost components of each of these 20 alternatives, the levelized costs of electricity from the Allens Creek facility 21 would be 9.4 cents per kilowatt-hour while that from a subbituminous coal 22 plant would be 11.6 cents per kilowatt-hour and that from a lignite-fired coal 23 plant,10.4 cents per kilowatt-hour. Assuming an equivalent lifetime supply of 24 energy from both the coal and nuclear alternatives, the discounted present 25 value of the savings from Allens Creek would be $418 million if the alternative 26 is a subbituminous plant and $192 million if the alternative is a lignite plant.
n!e:r11
)
1 The present value of total electricity costs for each of the three alternatives 2 considered is summarized in Figure 2.
3 As is indicated in Figure 1, the costs of coal and nuclear plants 4 represent a substantially different mixture of capital and operating costs. For 5 the nuclear facility, capital costs constitute nearly two-thirds of total 6 electricity costs. However, for the coal facilities, capital costs constitute less 7 than one-third of total operating costs. The analysis suggests that the lower 8 operating costs of nuclear plants offset their higher capital costs over the 9 lives of these facilities.
10 It is quite important to note that each of the components of cost for 11 both Allens Creek and its alternatives is subject to a substantial margin of 12 uncertainty. For capital costs, capacity factors and nonfuel operating and 13 maintenance costs, this reflects a substantial variation from unit to tmit in 14 historic costs and performance. For fuel costs, this reflects uncertainties as 15 to the future demand and supply of these fuels.
16 In choosing among the alter.it.tives several approaches can be used to 17 account for these uncertainties. One approach is to assign a probability to 18 each of the possible values for each cost component and use .these probability 19 distributions to estimate the exoected costs of electricity for each of the 20 alternatives. These expected values may be different from deterministic l
21 values described above because of the skewed distribution of specific cost 22 components and because the formulae for levelized costs tend to produce 23 skewed cost distributions, even when each of the cost components is sym-24 metrically distributed. A comparison of expected costs for each of these 25 alternatives is described in Figure 3. When the comparison is based on 26 expected values, the levelized cost of electricity from Allens Creek, at 10.2 ILera
I cents per kilowatt-hour, is still less than that from either the subbituminous 2 plant (11.8 cents) or the lignite plant (10.7, cents). In each case, the expected 3 values of levelized cost are somewhdt above the deterministic values de-4 scribed in Figure 1. This reflects the skewness of the probability distributions 5 underlying these values. In addition, as a result of somewhat greater skewness 6 in the probability distribution for nuclear than for coal costs, the expected 7 cost for Allens Cr.ek is closer to the expected costs of the coal-fired 8 alternatives than was the case for the deterministic values. Despite this 9 narrowing of the range, there are still substantial savings in exoected costs 10 from constructing Allens Creek. The present value of these savings is $315 11 million when the alternative is a subbituminous plant and $100 million when 12 the alternative is a lignite plant. These results are summarized in Figure 4.
13 While a comparison of expected costs still favors the nuclear option, 14 there is substantial uncertainty surrounding these expected values. This un-15 certainty has several impc tant implications. First, whichever of these 16 options is selected, there is a significant probability that costs will be either 17 greater or less than the expected values described above. Thus, for the 13 nuclear plant, an 80 percent confidence interval of cost produces a cost band 19 ranging from 7.5 to 13.3 cents per kilowatt-hour. For coal alternatives, the 20 band, dile narrower, is still quite wide--from 9.8 to 14.1 cents fer the 21 subbituminous alternative and from 8.5 to 13.1 cents for the lignite alterna-22 tive. These confidence bands are described in Figure 3.
23 Second, since these probability distributions exhibit substantial over- j 24 lap, tnere is no way to guarantee that the alternatives selected will be, in.
25 fact, lowest cost. Thus, for a subbituminous alternative, there is a 77 percent 26 chance that coal costs exceed nuclear costs, but a 23 percent that coal costs 1
Die 2:a' l
1 are less than or equal to nuclear costs. For the lignite alternative, there is a 2 62 percent chance that nuclear costs are less than coal costs but a 38 percent 3 chance that coal costs are less than or equal to nuclear costs.
4 FinaHy, given the substantial uncertainties in cost for both coal and 5 nuclear plants and the fact that these uncertainties are largely uncorrelated 6 with one another, it may be desirable for HL&P to consider a generation 7 expansion plan which contains both coal and nuclear units. While such a plan -
8 would have expected costs which were higher than those of an all-nuclear
-9 system, the associated variability in cost would be substantially reduced.
10 HL&P's current expansion plan represents such a mix; 30 percent of planned 11 adc" ns over the next 10 years are nuclear and 70 percent are coat 12 Q. Can you now describe in somewhat more detail your analysis of the 13 need for the Allens Creek facility and its sensitivity to conservation and load 14 management policies?
15 A. In addition to the cost factors described above, in general the 16 economic desirability of the Allens Creek facility depends upon the expected 17 future demand for electricity in the HL&P service territory, the availability of 18 coal-fired alternatives and the cost and availability of existing generating 19 facilities. If expected demand growth is rapid,. this increases the need for 20 baseload generation and, since Allens Creek is the most economic source of 21 baseload generation, for this facility in particular. In addition to demand 1 22 growth, demand for Allens Creek (or other baseload plants) depends on the l 23 cost of electricity from existing generating facilities. Where these costs are 24 high, as is the case with gas and oil units, it may be economic to construct 25 Allens Creek or other baseload facilities even in the absence of any growth in )
l 26 de mand. F!nally, the availability of coal alternatives to Allens Creek may be n! era
I l
l
_g.
1 limited by fuel availability and environmental requirements. If overall 2 requirements for additional generating capacity are sufficiently large, these 3 requirements may exceed maximum feasible coal additions. In this case, the 4 economic advantages of Allens Creek would be based not on a comparison with 5 ' coal costs, but with the costs of existing oil-fired or gas-fired capacity.
l 6 In order to determine the sensitivity of the need for Allens Creek to 7 load growth, the cost of existing facilities, and coal availability, it was S necessary to solve a relatively complex optimization problem. This optimi-9 zation problem involved determining the mix of capacity necessary to meet 10 forecasted loads on the HL&P system at lowest cost and the sensitivity of this 11 mix of capacity additions to alternative assumptions regarding load growth, 12 cost and capacity availability.
13 In order to solve this optimization problem, I developed a linear 14 programming model of the HL&P system. Given a specification of the 15 expected demands for electricity by time period and a description of the 16 availability and operating costs of existing generating plants, this linear 17 programming model was used to determine the pattern of capacity additions 18 and generation needed to meet these demands at lowest cost. The results of 19 this analysis are summarized in Figure 5.
20 In the first case, this linear programming model was used to estimate 21 capacity and generating requirements assuming that the company's forecast of 22 growth in demand was accurate. Using this forecast, peak demand in 1990 23 would be 15,050 megawatts and total energy requirements would be 86.9 24 thousand gigawatt-hours. Existing capacity in 1980 is 11,"63 megawatts and 25 plants currently under construction and those planned for constructioti during 26 the 1980s would add 6,340 megawatts to this total, bringing capacity in 1990 nie r a'
1 1 to 18,103 megawatts. Planned capacity additions, consist of Allens Creek 2 (1,130 megawatts), other nuclear capacity (770 megawatts), and various coal 3 units (4,440 megawatts). Under these plans, the system reserve margin on 4 peak would be 20.3 percent with and 12.8 percent without Allens Creek. In 5 evaluating the need for capacity additions,1990 oil prices (in 1979 dollars) 6 were assumed to be $25.56 per barrel and gas prices were $4.60 per MCF.
7 While these prices are probably too low, assumptions of higher prices would 8 simply reinforce the conclusions reached.
9 When the linear programming model was used to test this plan, all of 10 the additions proposed by HL&P were made, but the model results indicated 11 that an additional 6,742 megawatts of coal capacity would be required to meet 12 load at minimum cost. Thus, the cost-minimizing expansion plan required 13 13,082 megawatts of capacity additions, and, if feasible nuclear additions 14 were limited to Allens Creek and the South Texas Project,11,182 megawatts 15 of this capacity would be coal-fired. Moreover, if Allens Creek were not 16 available,12,297 megawatts of coal capacity would be required to meet 17 demand at lowest cost. This expansion plan is compared with the current plan, la with and without Allens Creek, in Figure 5. This result clearly indicates that 19 the decision to construct Allens Creek is justified given HL&P's load forecast 1 20 and the cost of existing operating capacity.
21 Mereover, it seems unlikely that the amount of coal fired capacity i 22 required by the cost minimizing plan, either with or without Allens Creek, 23 could be constructed over the next ten years. First, it exceeds current plans 24 by nearly 7,000 megawatts and planning and constructing this amount of coal 25 fired capacity would prove difficult with HL&P's current resources. Under the l l
26 cost minimizing expansion plan 52.0 million tons of lignite coal would be IleT a'
{
1 required annually by 1990. A 20-year supply of coal for these plants would 2 require a reserve of 1.0 billion tons, which is more than 55 percent of all 3 proven reserves of Texas lignite coal. Since there are many competitors for 4 this coal,it is unlikely that HL&P can secure a coal reserve of this magnitude 5 except at exorbitant prices. While subbituminous coal reserves are not 6 similarly limited, energy from subbituminous coal plants would be significantly 7 more expendive.
8 In addition to constraints on coal supply, the operation of the large 9 volume of coal capacity specified above would pose significant environmental 10 problems. Total sulfur oxide emissions from these plants, even with maximum 11 feasible controls, would be 87,000 tons annually. These emissions are likely to 12 create problems in complying with PSD or non-attainment provisions of the .
13 Clean Air Act, 14 These factors suggest that the amount of coal capacity available to 15 meet load growth and achieve an economic reduction in oil dependence, with 16 or without Allens Creek, is likely to be constrained well below the amount 17 required to minimize cost. The existence of these const:aints increases the 18 need for the Allens Creek facility.
19 To illustrate the effect of coal constraints on need, I have used the 20 linear programming model to calculate the expected savings to consumers in 21 1990 from constructing Allens Creek with and without such constraints.
22 Assuming unlimited ability to - conrtruct lignite capacity, the savings to 23 consumers from constructing Allens Creek would be $47.9 million annually by 24 1990. On the other hand, if lignite and other coal additions are constrained to 25 6,540 megawatts over this period (which is still 2000 megawatts in excess of 26 current phns), the annual savings from constructing Allens Creek would be i
n.e r a' !
l l
1 $245 million. In estimating these savings, I have used levelized annual capital 3 costs for Allens Creek and its alternatives which assume no further inflation 3 after 1990. These estimates are summarized in Figure 6.
4 The very large consumer savings from Allens Creek which -occur if 5 coal additions are constrained imply that substantial increases in the capital 6 cost of the Allens Creek facility could be experienced without eliminating its 7 advantage. Thus, the expected capital cost of Allens Creek is $1,855 per 8 kilowatt. If unlimited coal expansion were feasible, increasing this cost by 22 9 percent to $2,261 would cause the linear programming model to reject Allens 10 Creek as part of the least cost plan. On the other hand, assuming coal 11 additions are constrained to 6,540 megawatts, the capital cost of Allens Creek 12 would have to increase to $4048 per kilowatt before it would be removed from 13 the cost-minimizing plm Taese numbers are summarized in Figure 7.
14 Q. What conclusion did you draw from this analysis?
15 A. Given expected load growth, the Allens Creek facility is an econ- .
16 omically desirable component of HL&P's generation expansion plan. Moreover, 17 since the economically optimal generation expansion plan requires more coal 18 additions than are likely to be feasible, the expected savings to consumers 19 from constructing Allens Creek substantially exceeds the difference in cost 20 between coal and nuclear plants, and reflects instead the difference in cost 21 between operating existing oil- or gas-fired plants and the levelized cost of 22 electricity from Allens Creek. .
23 Q. What would be the effect on this result of reductions in demand 24 projections which would result if forecasted growth were too high or if explicit 25 conservation policies could reduce growth below forecasted levels?
26 nera' l
1 A. Reduction in demand growth, within a plausible range, does not appear 2 to alter these conclusions. To test this, the demand forecasts described above 3 were first reduced to the low end of the range employed by HL&P. This 4 resulted in a peak demand of 13,550 megawatts in 1990. This is well below the 5 low end of Dr. Anderson's forecast range which suggests a minimum of 14,044 6 megawatts in 1990. Even at this demand level, however,11,539 megawatts of 7 capacity additions were required to meet 1990 demands at lowest cost.' This is 8 still 5,199 megawatts in excess of current plants. Of this total,1,130 9 megawatts represents Allens Creek, 770 represents other nuclear and the 10 balance ,of 9,639 represents coal capacity. Even this represents far more cor1 11 capacity than is likely to prove feasible over this period. The results of this 12 analysis are summarized in Figure 8.
13 While still lower growth does not appear to be plausible, a similar 14 analysis was also done assuming no growth in electricity between 1980 and 15 1990. In this case peak demand in 1990 would be 10,200 megawatts. Even in 16 the absence of any load growth, the optimal generation expansion plan called 17 for the addition of 8,268 megawatts of additional capacity and this plan 13 included Allens Creek, South Texas Project and 6,368 megawatts of coal 19 capacity. Even at this lower. load forecast, the minimum cost expansion plan 20 included nearly 2,000 megawatts of capacity in excess of current HL&P plans.
21 In short, the need for the Allens Creek generating facility appears to 22 be independent of forecasted growth in electricity demands in the Houston ,
f 23 area. Since no conservation program or combination of nonconventional 24 alternatives to electric generation is likely to produce zero growth in electric 25 generation for the IIL&P Service territory, it would appear that the need for 26 l l
i Il&I'!EL
1 the Allens Creek facility is essentially independent of the effectiveness of 2 such conservation policies.
3 It should be noted that in addition to reducing overall load growth, 4 most conservation policies are likely to erode the system load factor. This 5 occurs because such policies tend to be more effective in reducing off-peak 6 than in reducing peak demands. In order to test the sensitivity of my 7 conclusions to conservation policies which reduce the system load factor, I 8 have examined the optimal generation expansion plan for the HL&P system 9 --assuming both no growth in load between now and 1990 and an erosion of the 10 current load factor from 66 to 60 percent. By comparison with the no growth 11 case described above, this reduces overall energy demand by over 10 percent 12 between 1980 and 1990, while leaving peak demand unchanged. As shown in 13 Figure 9, even with this erosion in total energy requirements, the optimal 14 generation expansion plan requires 7,083 megawatts of additional capacity 15 including Allens Creek, the South Texas Project and 5,183 megawatts of coal 16 capacity. This is stillin excess of current planned additions and suggests that 17 even this erosion in energy demands would not eliminate the economie 18 desirability of Allens Creek or other comoonents of HL&P's current generation 19 expansion plan.
20 Q. Have you evaluated the effect on these conclusions of load man-21 agement policies which increase rather than decrease load factor?
22 A. While the most likely direct effect of conservation policies is to 23 reduce the system load factor, there are load management policies which 24 might improve system load factor by shifting some demands from the peak to 25 the off-peak periods. In particular, time-of-day, seasonal or tempera-26 ture-sensitive pricing might accomplish this result. In order to determine the 11e:Pla'
1 sensitivity of the demands for Allens Creek to load management policies, I 2 assumed no growth in energy requirements for the period 1980 to 1990 and, in 3 addition, assumed an. improvement in the system load factor so that the 4 overall load factor was 100 percent. This change reduced peak demand to 5 6,721 megawatts, which is well below the current level. While such an 6 improvement is obviously not attainable in practice, I considered this case in 7 order to test the limits of the sensitivity of demand for Allens Creek to the 8 load factor. In this case, the requirements for capacity additions were 7,677 9 megawatts of additional capacity, which included Allens Creek, South Texas 10 Project and 5,777 megawatts of coal capacity. This is still in excess of 11 current plans and indicates that even the most extreme improvement in 12 system level factor combined with a "no growth" conservation program would 13 not eliminate the need for Allens Creek.
14 The overall results of this analysis can be summarized briefly. It 15 would appear that the economic desirability of the Allens Creek facility is 16 essentially insensitive to conservation policies, even those that would reduce 17 to zero the expected load growth of the system over the period 1980 to 1990.
18 Even if these conservation policies also tended to erode the system load 19 factor, tnis would still not eliminate the need for or the economic desirability 20 of the Allens Creek facility. Load management policies which improve the 21 system load factor would tend to increase the system's requirements for 22 capacity additions and, consequently, the need for the Allens Creek facility.
23 Given the very substantial dependance of the current HL&P system on 24 capacity fired by oil and gas, the need for capacity additions, either coal or 25 nuclear capacity, are very large. Assuming that no nuclear capacity beyond 26 Allens Creek and South Texas Project can be constructed over the next nePR
1 decade, the requirement.s for coal capacity probably exceed the amount which 2 can be built, given the civironmental requirements and the ability to secure 3 reliable coal reserves.
4 , Q. Can you describe in somewhat more detail the methodology used to 5 derive the capital costs, operating and maintenance costs and capacity factors 6 used in this analysis for coal and nuclear plants?
7 A. Yes. For each of these factors, estimates used in this study were 8 based on a statistical analysis of data on a sample of generating units 9 constructed over the last 15 years. Multiple regression analysis was used to 10 relate capital costs, operating and maintenance costs and capacity factors for 11 each of these plants to a number of plant character 3 tics. Equations derived 12 from this analysis were used to predict costs and operating characteristics for 13 Allens Creek and for subbituminous and lignite coal-fired units.
14 The equation for projecting nuclear capital costs is described in Table 15 1. As indicated, the determinants of costs include: the date on which the unit 16 received its construction permit; the time required to license the unit; the 17 wage rate of construction labor in the area in which the plant was constructed; 18 the size of the unit in megawatts; the number of nuclear units previously 19 licensed by the architect / engineer (A/E) responsible for designing or con-20 structing the facility; a variable which distinguishes between single unit.
21 plants, the first unit of a multiple unit plant, and the second or subsequent unit 22 of a multiple unit plant; a variable which distinguishes between plants built in 23 the Northeast section of the country and elsewhere;. and a variable which 24 distinguishes between units built with and without cooling towers. Taken 25 together, these variables account for nearly 90 percent of the historie 26 variation in construction costs for nuclear facilities.
rLe r a'
l l
1 For coal plants, the determinants of capital cost, as described in Table 2 1,, are the date the unit was completed, the size of the unit in megawatts, a 3 ' variable distinguishing between the first unit at a site and second or -
4 subsequent units, the wage rate prevailing in the area in which the plant is 5 constructed, a variables distinguishing between plants built with and without l 6 scrubbers, and variables which distinguish among plants built in various regions 7 of the country.
8 For both coal and nuclear units, these equations express the log of 9 construction costs per kilowatt of capacity as a linear function of these deter-10 minants. As a result of this choice of functional form, each unit change in the 11 determinants of cost is associated with a constant percentage change in cost 12 per kilowatt. Where the determinants of cost are also measured in logarith-13 mic terms, as is the case for wages and unit size, each percentage change in 14 the independent variable produces a constant percentage change in cost per 15 kilowatt. The cost projected by these equations are expressed in constant 16 1979 dollars, exclusive of allowance for funds used during construction.
17 In using these equations to projedt the capital costs for Allens Creek 18 and the alternative coal unit, several assumptions were made. First, it was 19 assumed that, except for escalation in input costs, there would be no 20 additional escalaticn in construction costs for coal and nuclear plants beyond 21 that which had occurred through 1979. Thus, costs for units completed in 1979 22 were used to forecast costs of units completed in 1989, adjusting only for I 23 escalation in labor and materials costs in the intervening period. Although 24 costs for both coal and nuclear plents have escalated over the past few years j
25 at a cate which substantially exceeded the rate of escalation in labor and 26 Ibe. rd
1 materials costs', we did not feel that there was any empirical basis for 2 extrapolating the historic rate of cost escalation into the future.
3 Second, for coal facilities, it was assumed that the construction would 4 take seven years from start to completion, whereas for the Allens Creek 5 facility it was assumed that seven and a half years would be required. During 6 the construction period, costs were assumed to escalate for both facilities at a
~
7 rate of 3.5 percent per year and interest on funds used during construction was 8 assumed to equal 7.5 percent per year. In computing the book value of these 9 facilities at the time of completion, I assumed that interest expenses on 6i 10 percent of construction costs would be included in rates during the construc-11 tion period and the balance would be capitalized. For the nuclear facility, 12 licensing time was set equal to 27 months and A/E experience was derived by 13 calculating the number of units licensed by Ebasco prior to Allens Creek.
14 Third, in deriving the capital costs for the coal facility, a separate 15 equation was used to estimate scrubber costs. This reflects the fact that 16 scrubber costs are very sensitive to the sulfur content of the fuels used and 17 the percent of sulfur reduction required and, consequently,.the simple dichot-18 omous scrubber variable in the econometric equations would not suffice to 19 estimats these costs. For particulates a similar approach was used. The 20 estimates of scrubber and particulate cost as derived from engineering models 21 are described in Table 2.
22 Fourth, the coal facilities which were examined in the regression 23 analysis included bituminous, subbituminous, and lignite plants but most of the 24 plants analyzed were bituminous facilities. Because of the lower heat content 25 of subbituminous and lignite fuels, the capital costs of these facilities tend to i
26 be somewhat higher than those of bituminous plants. Using data derived from nie r a'
. , . _ - - _ . - _ - _ . - - . - _ . . - . . - . . , . . , - . - - . _ - . - , - - - , ~ , , . -
1 a study by United Engineers and Constructors, I estimated that, each 10 ,
2 percent decline in the heat content of the fuel used in a coal-fired plant 3 increased the capital cost by 0.11 percent and this was used to adjust base 4 capital costs to reflect the heat contents of the fuels used in the subbitumin-5 ous and lignite plants studied.
6 Fifth, the size variable used in predicting the capital costs of coal 7 plants measures gross capacity. For nuclear plants, the size variable measure 8 the net electric capacity. In order to express costs in comparable terms, costs 9 for both coal and nuclear facilities were expressed per kilowatt of net 10 capacity and, for coal plants, net capacity was assumed to be equal to 90 11 percent of gross capacity.
12 Finally, it should be noted that a lignite coal facility constructed by 13 HL&P would, in all likelihood, be situated in or near the lignite coal fields'.
14 This would make this facility somewhat more distant from HL&P's load center 15 than either a subbituminous plant or the Allens Creek facility. To account for 16 this difference, the capital costs of any additional transmission facilities 17 needed to bring power from each of these facilities to the HL&P system are 13 included in the plant capital cost estimate.
19 The resultant capital cost estimates for the Allens Creek facility, a 20 subbituminous coal plant, and a lignite coal plant are described in Table 2. In 21 each case, these costs reflect the gross book value of these plants at the time 22 of completion. It should be noted that these costs include direct construction 23 costs and accumulated allowance for funds used during construction on that 24 portion of construction work in progress not included in the rate base during 25 the construction period. In addition to the mid-point values, which are $1,855 26 per kilowatt for Allens Creek, $1,020 per kilowatt for the subbituminous plant l
n;e PR ;
20- f i
1 and $1,107 per kilowatt for the lignite plant, this table also describes the 80 2 percent confidence band around these estimates. This 80 percent confidence 3 band has been derived using the standard error of estimate from the regression 4 results described above. For the Allens Creek plant, the range of projected 5 capital costs is from $1,440 to $2,389 per ki~owatt, for the subbituminous plant 6 it is from $762 to $1,365 per kilowatt, and for the lignite plant it is from $827 7 to $1,482 per kilowatt.
8 The procedure used to derive nonfuel operating and maintenance costs 9 for coal and nuclear plant:5 is quite similar to that used to derive capital costs.
10 The O&M cost equations are described in Table 3. For coal plants, the 11 determinants of O&M costs are unit size, the number of units at a site, 12 prevailing wage levels for utility employees, variables which distinguish among 13 various regions of the country, and a variable reflecting the date on which the 14 plant came on-line. The same variables are used to predict the O&M costs for 15 nuclear plants but, in addition, there appears to be a relationship between the 16 age of a nuclear plant and its operating and maintenance costs. For sampled 17 coal plants, the regression equation explained 90 percent of the variation in 18 annual operating costs while for nuclear plants the equation explained 87 19 percent of this variation. For both coal and nuclear plants, the O&M costs of 20 the most recent plants appear to be somewhat lower than plants built earlier.
21 In projecting costs for future facilities, however, I did not assume that this 22 declining cost trend would contmue. In addition, the data for nuclear plants 23 suggest that O&M costs rise with age, but that the rise largely levels-out by 24 the fif th or sixth year of operation and declines thereafter. Since data on 25 nuclear plants with more than 5 to 6 years of operation are limited, I assumed 26 that O&M costs remained constant, except for the effects of inflation, after
+ '
-- " TN'
1 the fifth year. Plant O&M costs for both coal and nuclear facilities, are 1 1
2 described in Table 4. The costs in this table are expressed both in constant 3 .1979 dollars and in levelized dollars over the life of these facilities. In 4 estimating levelized costs, I assumed that labor and materials inputs to these 5 operating and maintenance costs would inflate at 5.5 percent per year over the 6 life of these facilities. As with the capital costs, I have indicated both the 7 mid-point of costs and the 80 percent confidence interval. This interval was 8 derived based upon tne standard error of these regression equations.
9 In addition to plant operating and maintenance costs, scrubber and 10 particulate removal systems impose additional O&M costs on coal-fired 11 facilities. The O&M costs used to develop the estimate described above do not 12 reflect these scrubber or particulate costs, since most of the facilities on 13 which these estimates were based do not have scrubbers or the type of 14 particulate systems required by current environmental regulation. The O&M 15 costs for coal facilities were augmented by scrubber and particulate O&M .
16 costs derived from an engineering model of these costs. Table 4 also contains 17 these estimates.
18 In order to express capital and nonfuel operating and maintenance 19 costs per kilowatt-hour of generation, it is necessary to determine the annual 20 generation which will be achieved by either the nuclear facility or the coal 21 alternative. For coal facilities, I derived the achievable capacity factor from 22 historie data on the equivalent availability of coal-fired units. Equivalent 23 availability measures the percentage of time a unit is partially or fully 24 available to meet electricity load. For periods in which a unit is partially 25 available to meet load, equivalent availability reflects the percentage of total 26 n.e r a'
I capacity which is available. If a unit could be used fully whenever it was 2 available, annual generation would equal:
3 4 MWH = EA
- HOURS 5
6 where: MWH = annual generation in megawatt-hours; 7 EA = equivalent availability; 3
CAP = net capacity in megawatts; g HOURS = hours in the year (8,760 normally and 10 8,784 for leap years).
11 12 In these cases, unit capacity factor (the ratio of generation to the product of capacity and hours in the year) would equal equivalent availability. In 13 14 practice, even the most economic baseIcad units achieve capacity factors 15 which average about 90 percent of equivalent availability. This reflects 16 system loading constraints which prevent even the most economic units from 17 being fully utilized. . Consequently, for coal units, the maximum achievable 18 capacity factor was set equal to 90 percent of equivalent availability. The 19 equation used to forecast the equivalent availability factor of individual units, 20 which is summarized in Table 5, relates equivalent availability to size, age, i
l 21 first-year equivalent availability and sulfur content. Equivalent availability 22 appears to decline as size increases, to increase with age and to decrease with 23 sulfur content.
24 For nuclear units, I estimated the relationship between capacity factor 25 and plant characteristics directly, and this relationship is also described in 26 Table 5. The capacity factor for nuclear units appears to be related to size, n,e r a
1 agt , and vintage of the unit. Capacity factor appears to be a decreasing 2 function of size, to increase with age, and to be better for more recent units.
3 The capacity factor equation described here reflects only the oerformance of 4 BWR units which appear to exhibit a somewhat different pattern of per-5 formance than PWR units. Allens Creek is a BWR unit.
6 The predicted capacity factors for coal units and for the Allens Creek 7 facility are described in Table 6 along with their confidence intervals. For the 8 coal facility, the predicted capacity factor was 68 percent. For nuclear 9 plants, two capacity factors were estimated. When the vintage trend was not 10 extrapolated, the projected capacity factor was 61 percent. When vintage was 11 extrapolated into the late 80s the projected capacity factor was 73 percent.
12 Giving relatively modest weight to this vintage effect, we assumed a 65 13 percent capacity facter for the nuclear unit.
14 Q. Dr. Perl, can you indicate how the nuclear fuel estimates used in your 15 testimony were derived?
16 A. Yes. The cost of nuclear fuel consists of five separate components:
17 the costs of uranium oxide or, enrichment, conversion, fuel fabrication, and 18 waste disposal.
19 With respect to the first of these factors, NERA has developed a 20 model which estimates uranium oxide prices as a function of both supply and 21 demand factors. In this model, the demand for uranium oxide is derived from 22 projections of nuclear generating capacity coming on-line through the end of 23 the century. This demand is adjusted judgmentally to exclude imports and to 24 include exports and, therefore, represents demand for U.S. uranium oxide 25 production. The initial supply of uranium oxide used in this model is derived 26 from the existing data on U.S. reserves, classified by their cost of extraction.
n/e.r a l 1
1
l 1 Future uranium reserves are projected using an econometric equation which 2 relates discoveries of uranium to exploratory and developmental drilling 3 activity. As modelled, supply in any one year is a function only of reserves 4 previously discovered on which mines and mills have been established. This 5 estimate of supply, in conjunction with an estimate of demand provides an 6 estimate of equilibrium price for the year in question. The equilibrium price 7 of uranium in the initial year, in turn, determines the mining industry's 9 investments in new mines and mills and the level of exploratory effort.
9 Development of new mines and mills and the discovery of new reserves then 10 determine reserves available in future years. Sequential analysis of supply and 11 demand over time provides a trajectory of uranium oxide prices. The prices 12 used in this analysis on a year-by-year basis.through 2020 are described in 13 Table 7. A range of prices has been developed reflecting alternative forecasts 14 of demand.
15 Estimates of the costs of each of the other components of the nuclear 16 fuel cycle were derived based upon testimony of prepared by NUS for 17 proceedings scheduled by the Nuclear Regulatory Commission (NRC) on the 18 recycling of uranium and plutonium. The range of estimates for each of the 19 components are described in Table 8. Since these estimates reflect forecasts 20 of costs prepared by the NRC, the electric utility industry manufacturers of
(
21 nuclear fuel, and by intervenors in this proceeding, it seems reasonable to 22 suppose that these values reflect the full range of anticipated costs.
23 Q. Can you indicate how you derived costs of coal fuel used in this 24 analysis?
25 A. Yes. With respect to costs of coal, NERA also has developed a model .
26 which projects equilibrium coal prices as a function of both supply and demand ner/a
1 factors. In this model data from the U.S. Bureau of Mines for each of 22 2 regions of the country are used to derive estimates of coal supply by sulfur 3 content and heat content. Reserves in each region are then further subdivided 4 by extraction cost to create coal supply curves for each coal type and region.
5 In addition to these data on supply, the model derives demands for coal by 6 sulfur and heat content for each of 21 regions of the United States. Using 7 data on transportation cost from each supply to each demand region, the 3 model then derives the equilibrium level of production for each coal reserve.
9 A pattern of production is derived which minimizes the national aggregate 10 cost of meeting these coal demands. Based upon the supply curves discussed 11 above and these estimates of coal demand, equilibrium prices for each fuel are
~
12 derived.
13 In estimating coal prices for subbituminous and lignite plants in Texas, 14 I examined the forecast of equilibrium prices to determine the lowest cost 15 source of fuel for coal plants constructed in the HL&P service territory. For 16 lignite plants, this involved using Texas lignite coal. For subbituminous plants, 17 . coals from either New Mexico or from Wyoming could be lowest cost, 18 depending upon future demand, and consequently, the prices used reflect an 19 average of prices for coal from each of these regions. With respect to lignite, 20 the plant is assumed to be a mine mouth facility and so no transport costs are 21 involved. With respect to subbituminous coal, the transport costs from either 22 New Mexico or Wyoming to a site in Texas are calculated from equations 23 which forecast rail freight rates as a function of region and distance. Prices 24 of these fuels, at the mine and delivered, are described in Table 9. In deriving 25 the most likely value used in this analysis, I used estimates of the most likely 26 nega' 4
1 demand for coal while upper and lower bounds reflect higi:er and lower coal 2 demands.
3 Q. Can you describe how the levelized cost of electricity was derived 4 from these cost components?
5 A. Yes. The levelized cost of electricity is a constant annual charge for 6 electricity which would have the same discounted present value as actual 7 electricity charges over the life of the plant. This levelized cost was derived 8 from the cost components described above using the following formula:
9 10 h CC
N 1 gfy (1+dN 14 15 .
16 where: CC = the book value of the capital cost of the unit per kilowatt of.
17 capacity; IS AFg= the annualization factor for capital cost in year I; 19 CF;= the capacity factor of the plant in year I; 20 OM g = nonfuel operating and maintenance costs per kilowatt of 21 capacity; 22 FC = fuel costs per kilowatt-b.our of generation; 23 d= the discount rate; 24 N= assumed life for the facility.
25 26 6
nera'
1 With the exception of the annualization factor and the discount rate, 2 all of the terms in this equation are described above. The annualization factor 3 is a number which when multiplied by the initial book value of the facility, 4 provides the cost of interest, amortization, taxes insurance, decommissioning, 5 and periodic decontamination. In calculating this factor, interest, amortiza-6 tion and taxes are based upon financial data supplied by HL&P which are 7 described in Table 10. The interest costs during the operating life of these 8 facilities have been adjusted upward to include the costs of interest incurred 9 during the construction period.
10 In estimating cecommissioning costs, I assumed that these would equal 11 10 percent of the initial cost of the facility plus the effect of nominal 12 escalation over the life of the facility. A sinking fund was used to derive an 13 annual decommissioning charge. A similar approach was used to fund 14 decontamination charges occurring in the 15th and 25th years. The decom-15 missioning costs are derived from a study done by the Atomic Industrial Forum 3 (AIF) and the decontamination costs are from a study by Sargent & Lundy.
17 The adjustments to the annualization factor to reflect decommissioning and 13 decontamination are shown in Table 11.
19 Q. Dr. Perl, would you summarize your conclusions?
20 A. Yes. HL&P requires very substantial additions of baseload generating 21 capacity over the next decade. These capacity additions, which exceed 22 current plans, are required both to meet anticipated load growth and to reduce 23 oil dependence. Since the expected cost of electricity from Allens Creek is 24 lower than any of the alternative capacity additions examined, and lower than 25 the operating costs of existing gas or oil plants, this facility should be a part of 26 HL&P's expansion plan. Given the system's current substantial dependence on n e r'a'
l l
I i ,
1 oil and gas and probable constraints on expanding coal capacity significantly 2 beyond current plans, the desirability of Allens Creek is largely independent of I
3 conservation end toad management policies.
4 5
6 7
8 9
10 11 12
, 13 4
14 15 16 17 18 19 20 21 22 23 24 25 .
26 nera
l R@m1 Levelized Bectricity Cost Frotn Allern Creek And Coal Alternatives 1
(Deterministic Values) l l
l
, ._ l r
C AM Cost l Fued Coot L casocitycost Levedland CantsikWh 12.00 -.
11.55 (W '
10,36 I
9.35 '(I24 (W
8.00 .
l (2.85) (7.54) (5.85)
- .a -i.F ; l
, ,~av 1, 4
,',7
. h.rs.t:w *;.,;. : I
- d 2.w,.
W- l 4.00 2. .,;M2 %
~3 ~.
,. m.e > . - . &
\ 3,~ . l-j
- qn w . .
1 ? ;a. ,
.,; :::q;m e: +;; x . +
. - QQh :: -
-i( R . . . .](K27)l l
- f. :pgi:l ~
-< :-i, 2:
. . , . ~s . ,
- I
- f; ,- .
Y, n *;;a_
e ,;. -
r ' e;7 a '- .
s
';; "- ax < ,-
n '. _
i " l 0 l AIIens Subbituminous Lignite i Creek Pfant Pfant l l
l l
l 1
neI'R
Figure 2 Discounted Present Value Of Electricity Costs From AIIens Creek And Coal Alternatives (Deterministic Values)
., CostSavines From AasensCreek Millions of Dollars 2,196 g,
n-W' (4LM J970 1.778 .
1,700 -
0 AIIens Subbituminous Lignite Creek Plant Plant nern l l
M gure 3 Probabilistic Distribution Of Levelized Electricity Cost From Allens Creek And Coal Alternatives A 90th Pweentlie w Levedized Cants /kWFr . Mean voeue a 10th Percentile Vefue 14.07A 13.27A 13.14A 13.00 =
12.ss' 11.'00 -
10.72*
10.19' 9.82*
9.00 - 1 S.53*
\
7.48*
7.00 =
?
O Allens Sunbituminous Lignite Creek Plant Plant nera*
I Figure 4 Discounted Pmsent Value Of Electricity Costs From Allens Creek And Coal Altematives (Expected Values)
- .r Cost Savings w Prom AasensCreen j
I Millions i of Dollars
(
l 2,300 -
. ; ~,
m .
r .,.,
c, 2.038 1,938
~ 5fd'$ ,
(3gg) 1,800 -
l l
0 Aliens Subbituminous UJnite Creek Pfant Plant l
1 ne r a'
Figure 5 Alternative Capacity Expansion Plans For HL&P Through 1990 Medium Growth Capacity Expansion M Existing Capacity'
^"*"*C'**"
24.000 -
20,000 =
18,103 AHene Creek jg,gyg __
16,000 =
- 15.050 Coal Coal Coal I44401 IA4401 t11,1821 12,000 - otner reucieer Other Nucmar Other Nucteer
. Caestb. - cm- ~ .. cm
'113306* ~I m 'If,astt
.- , ; w 8,000 = ,
. . 3-
,4-4,000 - N #- 0"'
ans ant- . ame Gas, . Gasr- . .Gae .
- ~
~[g p !g3733* ,
- !E873f' 0
HL&P Capacity Expansion Plano Cost-Minimi:ing With Without Plan AIIens Allens Creek Creek
Fignue 6 Savings in Electricity Costs in 1990 From Allens Creek Millions of Dollars 245.2 240 = '
16 0 -
I 80 -
47.S M
O No . Coal Additions Constraint Limited to on Coat 6,5to gw The PH"
Figure 7 CapitalCost At Which AIIens Creek Becomes Uneconomic fno Dollars Q. ca,t,ee,,en in Caeitet
,agg,,,c,,,,
Per To 8ecome Uneconomic Kiloseert 4,048 4,000 . ,
15p:+ .1; rw Y. *s. :.
.;:q
~ ~ ~
w ge w ,
- ywg y.
. ~ 1;Tg+~ -
?l(
,, o .n Y ca-TYIk;;.~g .. yy, 0 52 ep^&&{4
- w .R.
?~.G 3.000 - H 6. 4 . N-. .r.
- m. . wir;..m n ;
- mp
{me*;fEh?.' -
T.=i.y;:p.f.
. ~ . .
d-
' di%yk.3 2,261 r~
" ~..~ c Ca. '*r y-
&; -r .w
~
e... r .,
2
.. .2 2,000 =
LS55 -
i 1,000 -
0 Capital Cost Capital Cast At Which Allens of AIIensCreek Creek Becomes Uneconomic:
l l No Coal Additions Constraint Limited to l on Coal 6,540 MW l
i nera :
I
I Figure 8 Capacity Expansion Plans For HL&P Under Alternative Growth Assumptions Capacity Expanseon Existing Capacity
- I'*""C'**"
24,000 = .
23.302 Allene Creek 20,031 20,000 ~ Allons Creek Coal Coat 18.103 Allens Creek: (11,182) (9,6391 Coad 16,000 " ca.
. . 16,3681 l
14,4401 , ,
12,000 - Otw Nucteer Other Nuckem Otw Nuclear OtherNucaser ca,s ;,: . :g _: ca.g, . ,
coesc 11;3901- 11,390F _ lt,890F firMr
- . . c. . -
8,000 =
- W ~- -
f
~ '
, ow? - . . ow - ' oss , .ow ane ' ad ame . ane -
Gear , Gap - . Gems -
Gams -
~ ^
4,000 "
.Is,s738 Is,s73r ' tas73E Is,arst 1
0 HL& P Cost-Minimizing Expension Plan: 1 Capacity Medium Low No Plan Growth Growth Growth (Peak'O'Demands# (Peak Demands (Peak Demands (Peak Demands 15,050 M W ) 13,550 MW) 10,200 MW)
1 Figure 9 '
Capacity Expansion Plans For HL&P Under No Growth And Allemative Load Factor Assumptions ;
I Capacity Expansion
! Extsting Caoacsty 0 """'* Peak Demand f aliens Creek 1,130 Mw j Other Nucfear 770 MW
! Megswatts ,
24,000 -
20,031 19,44o , i 20,000 - Alien creen 18A46 .
Allens Cream 18.103 Allens Creen Allens Cream ,
i 16 /l00 - Coed Coad ,
. Coal 16,M f5,183) (5,7771
[4,4401 Cther Nucseer Other Nucseer Other Nucteer 12,000 - other Nucteer
~ CW ^ Cast- - 8'
. Ciner ) jw ,
17,33e8, - 1t2908 gg , gg -
^ - - -=_ , ,
';;; n -
l
- i
- . x .~y s. - l 8,000 = . . - . ;,
~
~
r i
^ - .
' - . . One ~ est . . mr _
~ OIK: - u ame ame Qg,,, .
j
. ,ame ; an. - Gas- l 4,000 -
145r32 IN lasrat l l .,
0 .. . . . _ .
HL & P Cost-Minimizing Expansion Plan:
Capacity Expansion 00 * 60% 1007.
PIen (Peak Demands 15,050 MW)
FYor $r Yn, (Peak Demands (Peak Demands (Peak Comands 10,200 m) 10,200 MW) 6.721 MW) l
I TABLE 1 !
Page 1 of 2 )
REGRESS 10NS RELATING CAPfrAL COS'IS FOR NUCLEAR AND COAL PLANTS TO SELECTED DETERMINAN'IS 1 Nuclear Capital Costs Coal Capital Costs Regression Regression Variable Co- Variable Co-Variable Mean efficient t-Statistic 2 Mean efficient t-Statistic 2 (1) (2) (3) (4) (5) (6)
Constant 1.0000 6.2601 -
1.0000 5.1914 -
Ln of Size in MW 3 6.7100 -0.1957 -1.6817 6.0967 -0.1142 -4.1529 Ln of Wage Rate
- 2.3522 0.5087 2.5970 2.3591 0.5908 4.8079 First Unit Indicator s 6.4000 0.0776 1.2523 0.2618 0.2417 8.2938 Subsequent Unit Indicator 8 0.3400 -0.1702 -2.5547 Cooling Tower Indicator 7 0.4000 0.0646 1.3568 Ln of Licensing Time s 2.7118 0.2744 2.2362 Ln of AE Experience 8 1.7966 -0.0907 -4.4220 Sequence Group 11a 0.1000 -0.8342 -5.4887 Sequ1nce Group 2t o 0.1000 -0.6304 -3.9904 Sequ:nce Group 318 0.2000 -0.4543 -3.5348 Sequ;nce Group 410 0.3000 -0.3823 -3.3443 Sequsnee Group 513 0.2000 -0.1781 -1.8994 Scrubber Indicator 11 0.0901 0.1157 2.2927 Optrating Date Indicator it'2 0.2704 -0.2274 -6.1384 Optrating Date Indicator 213 0.2446 -0.1296 -3.6767 Op: rating Date Indicator 314 0.2060 0.2020 4.7783 Northeast Indicator 15 0.3200 0.2184 4.3113 Mountain / Pacific Indicatort s 0.0987 0.1752 3.9376 Southwest Indicator 15 0.0429 -0.3124 -4.3497 Number of Observations 50 233 R-Squared O.8981 0.6097 Standard Error 0.1363 0.1904 11 e 1a
TABLE 1 Pago 2 of 2 REGRESSIONS RELATING CAPITAL COSTS FOR NUCLEAR AND COAL PLANTS TO SELECTED DETERMINAN'IS SOURCES AND NOTES
- The dependent variable in the regression is the naturallog of capital cost per kilowatt. Actual costs have been adjusted to reflect the costs of constructing the plants at labor and materials costs prevailing in 1979 with no allowance for funds used during construction. Results are based on units completed through 1978.
2 A t-statistic is the ratio of the mean of the coefficient to its standard error.
It measures the reliability with which the coefficient is measured. A t-statistic of 1.96 or higher indicates that the coefficient is significantly differ-ent from zero at the 5 percent level. A t-statistic of 1.64 or higher indicates significance at the 10 percent level
' For nuclear units, size reflects the net design el'ectric rating as reported to the NRC at the time the unit received its operating permit. For coal units, data reflecting the Gross Generatne Nameplate Rating as reported to the FPC was multiplied by 0.9 to produce a net rating.
- The regional wage rate is average wages plus employer contributions to funds, for all building trades in 1976, expressed in dollars per hour for the Bureau of Labor Statistics region in which the plant is located. The source is U.S.
Bureau of Labor Statistics, Union Wages and Hours: Building Trades. July 1.
1976, Table 12, Bulletin 1972.
'Tiie indicator equals 1 if the unit is the first unit of a series. built at a site and 0 otherwise.
'The indicator equals 1 if the unit is a subsequent unit built at a site and 0 otherwise.
- The indicator equals 1 if the unit has a cooling tower and 0 otherwise.
8 Licensing time is the number of months from date of application for con-struction permit to date of receipt of construction permit.
' AE experience is the number of reactors built or under construction by the architect-engineer at the time the unit received its construction permit.
- The units in the sample were arranged chronologically by date of construction permit. Units were then grouped chronologically to reflect major shifts in cost occuring over time. Sequence Group 1 includes the first five units of the sample: Sequence Group 2 includes the next five units of the sample; Sequence Group 3 includes the next 10 units; Sequence Group 4 includes the next 15 units; and Sequence Group 5 includes the next 10 units. The last five units in the sample represent a reference group.
- The indicator equals 1 if the unit has a scrubber and 0 otherwise.
- The indicator equals 1 if the unit began commercial operation between 1965 and 1968 and 0 otherwise.
l'The indicator equals 1 if the unit began commercial operation between 1969 and 1971 and 0 otherwise. Units completed from 1972 to 1975 constitute a
- *. reference group.
The indicator equals 1 if the unit began commercial operation between 1976 and 1978 and 0 otherwise.
Is These indicator variables equal 1 if the plant is in the given region of the country and 0 if it is not in that region. See Table 12, for the definitions of the regions used. All other regions constitute a reference group.
neTa
___.__ __ _ __ _ ~ __
1 TABLE 2 Page1of2 Elfr! MATED CAPfrAL COS'!15 FOR ALLENS CREEK AND COAL ALTERNATIVES Allens Subbituminous Lignite -
Creek Plant Plant (1) (2) (3)
Initial Capital Cat (Net of AFDC) '1979 $/k W Plant 1 791.93 311.38 320.96 Scrubber ' -
31.55 94.00 ESP 3 -
45.57 27.55 Total Generating Plant 791.93 439.20 442.61 1990 Completed Cat Adjusted fer Transtr.ission Cacital Costs ' Current 5/kW)
Generating Plant
- 1794.99 959.58 967.03 Transmission 3 59,98 59.98 134.85 Total 1854.97 1019.56 1101.38 1990 Completed Cost Adjusted Fce Transmission Capital Costs and Transmission Lesses:
Medium 8 1854.97 1019.56 1107.05 Low 7 1440.17 7S1.52 326.37 High 7 2389.22 1365.02 1482.15 tLe T a
TAHLt2 Paga 2 of 2 ESTIMATED CAPITAL COSTS FOR ALLENS CREEK AND COAL ALTERNATIVES SOURCES AND NOTES 1 Based on capital cost regressions in Table 1. For Allens Creek, the following assumptions were made: Size = 1,130 MW; Wage = $9.06; Licensing time = 27 months; AE experience = 13. For the subbituminous and lignite plants, costs were assumed to be 1.03 and 1.06 times the costs based on the regression to reflect differences in heat content between bituminous and subbituminous or lignite fuels. For the coal alternatives, we assumed that two 600-MW units would be constructed in the Southwest region.
2 Based on the following equation:
(113.90 + 5.75PR - 92.45FR] RW.5908 RS .1142 where:
RW = ratio of regional wage to national wage; RS = unit size relative to 600 MW; PR = pounds of sulfur removed per MMBtu; FR = ratio of outlet sulfur to inlet sulfur.
8 Particulate costs were derived from Stearns-Rogers Incorporated," Economic Evaluation of Fabric Filtration Versus Electrostatic Precipitation for Ultra-high Particulate Collection Efficiency," June 1978.
" Completed capital costs were calculated by adding escalation and interest to initial capital costs. Estimates assume 8.5 percent annual escalation from mid-1979 through 1986 and 7.5 percent escalation thereafter. Interest during construction is assumed to be 7.5 percent per year. Construction time is 7.0 years for coal and 7.5 years for nuclear.
sEstimates provided by J. Greenwaite of Hl.&P, May 1979.
8 Lignite costs were divided by .9953 to acjust for additional transmission losses.
7 Derived by increasing and decreasing the logarithm of medium costs by 1.28 standard errors. This reflects an 80 percent confidence band around the medium value. To reflect the uncertainties in projecting future costs, the standard errors of regression equations excluding the date variable were used. Nuclear estimates assume a standard error of 0.1977 and coal estimates assume a standard error of 0.2280, i
l l
l nera
TABLE 3
. Paga 1 of 2 REGRESSIONS RELATING OPERATING AND MAINTENANCE COSTS FOR NUCLEAR AND C,OAL PLANTS TO SELECTED DETERMINANT 5 Nuclear O&M Costs Coal O&M Costs Regression Regression Variable Co- Variable Co-Variable Mean efficient t-Statistic
- Mean efficient t-Statistic *
(1) (2) (3) (4) (5) (6)
Constant 1.000 -1.4339 -2.08 1.000 -6.1335 13.00 Ln (Unit Size) 3 1.526 0.3591 6.97 1.476 1.8169 14.34 1/(Unit Size) 3 0.269 2.0000 5.21 Ln (Number of Units) 0.164 0.6106 5.62 0.502 .
0.9210 17.14 Regional Wage Index
- 1.052 1.3957 4.51 1.020 1.8678 4.76 Wage Escalation Index' 1.807 1.7487 13.93 1.636 1.1842 18.71 Northeastern Region Indicator' O.470 0.3969 6.31 0.230 0.1223 1.93 Southern Region Indicator' O.113 0.2662 2.56 0.376 0.0912 1.15 Western Region Indicator' O.122 0.1798 2.18 Ln (Vintage) ' 2.078 -0.1381 -2.16 1.339 -0.2538 -6.16 Ln (Age -1)' 1.554 -0.8211 -3.34 .
1/(Age +1)' O.248 -2.9540 -3.51 l
Number of Observations 168 335 R-Squared 0.8743 0.9026 Standard Error 0.3144 0.3303 n eT'a
TABLE 3 Paga 2 of 2 REGRESSIONS RELATING OPERATING AND MAINTENANCE COS'I5 FOR NUCLEAR AND COAL PLANTS TO SELECTED DETERMINANTS SOURCES AND NOTES I The dependent variable is the natural log of cost in millions of current
. dollars. With the exception of the wage indices, the 1965-1974 data are from Federal Power Commission (now Federal Energy Regulatory Commission),
Steam-Electric Plant Construction and Annual Production Expenses and the 1975-1977 data are from Schedule 432 of Federal Power Commission Form One, " Annual Report of Privately Owned Electric Utilities, Classes A and B."
2A t-statistic is the ratio of the mean of the coefficient to its standard error. It measures the reliability with which the coefficient is measured.
A t-statistic of 1.96 or higher indicates that the coefficient is significantly different from zero at the 5 percent level A t-statistic of 1.64 or higher indicates significance at the 10 percent level.
3 Unit size is the gross size of the unit in hundreds of MW.
" Ratio of 1969 median earnings for electric and gas utilities employees for the SMSA nearest the plant to the median value of this variable for the U.S.. Source is U.S. Bureau of the Census, Census of the Pooulation: 1970, Vol.1, Table 188.
swage index for each year with the 1965 as the base to measure inflation.
Source is U.S. Bureau of Labor Statistics, Employment and Earnings. U.S.,
1909-78, Bulletin 1312-11, pp. 730-731.
8these indicator variables equal 1 if the plant is in the given region of the country and 0 if it is not in that region. See Table 12 for the definitions of the regions used. All other regions constitute a reference group.
7 Vintage equals the year the plant came on line minus 1963 for coal and minus 1959 for nuclear (so that the earliest plant year has a value of 1).
8 Age is the average age of the plant, calculated as the average of the ages of the individual units weighted by unit size.
l l
l 1
I l
1 net 3
TABLE 4 Pags 1of2 OPERATION AND MAINTENANCE COSTS FOR ALLENS CREEK AND COAL ALTERNATIVES l
! Low Medium gHI h (1) (2) (3)
-($/KW)
FIXED O&M COSTS Plant O&M Costs (1979 $) 1 Allens Creek 3.44 12 .63 18.87 Suobituminous Plant 4.25 6.48 9.90 Lignite Plant 4.25 6.48 9.90 Plant O&M Costs (1990 $) 2 Allens Creex 16.73 25.03 37.42 Subbituminous Plant 8.42 12.85 19.63 Lignite Plant 8.42 12.85 19.63 Transmission O&M Costs (1990 $)3 Allens Creek 1.13 1.69 2.53 Subbituminous Plant 1.11 1.69 2.58 Lignite Plant 2.49 3.79 5.79 O&M Costs Adjusted for Transmission O&M (1990 $)
Allens Creek 17.86 26.72 39.95 Subbituminous Plant 9.53 14.54 22.21 i Lignite Plant 10.91 16.64 25.42 Levelized O&M Cost Adjusted '
I for Transmission Losses (Current S)*
Auens Creex 30.66 45.87 68.59 Subbituminous Plant 16.36 24.96 38.13 Lignite Plant 18.82 28.70 43.34
, SCRUBBER O&M COSTS ' mills /kWh)
I 1979 Mills /kWh
- l Succituminous Plant 1.82 i Lignite Plant 2.37 Levelized Current Mills /kWh Adjusted for Transmission Losses
- Succituminous Plant 5.79 l Lignite Flant 7.59 l
l l
t n e la i
TABLE 4 Paga 2 of 2 OPERATION AND MAINTENANCE COSTS FOR ALLENS CREEK AND COAL ALTERNATIVES SOURCES AND NOTES
- Medium costs were calculated from the regression equations documented in Table 3. For Allens Creek, costs reflect a single unit 1130 MW plant; the coal plants consist of two 600-MW units. The regional wage index equals .9597, the wage escalation index equals 2.21 and the plants are in the South. In addition to the regession forecast, nuclear costs include a
$.40/kW premium for nuclear insurance. Vintage eguals 18 for Allens Creek and 14 for coal. O&M costs reflect the levelized impact of in-creases in real costs through year 10 and constant costs thereafter. Low and high costs are calculated as 2128 standard errors from the mean.
,This reflects an 80 percent confidence band around the medium value.
~1979 dollar costs are escalated at 6.28 percent annually from 1979 3
through 1986 and 5.5 percent annually thereafter.
Medium costs are based on information provided by J. Greenwaite, HL&P. Low and high costs assume the ratio of transmission to plant
, costs is the same as for the medium case.
1990 dollar costs are levelized over the 33-year operating life of the plant assuming an inflation rate of 5.5 percent and a discount rate of 11.05 percent. Lovelized plant costs for lignite are divided by .9953 to
, adjust for additional transmiss,!on losses.
Based on the equation:
Scrubber O&M = (1 + Energy Penalty)( Lime Cost + Ash Disposal + Other O&M) where:
Energy Penalty = [.0317 .0310
- f I Lime Cost = (1.01 * .0005
- 10-3* ESC 79; Other O&M = (.1290 + .0011* IS ' .0112*05 .1172* HR*10~3* ESC 79; Ash Disposal = (.0005* AC*AP*.2] *HR*10~3" ESC 79; OS = Jutlet sulfur = .445 and .4624 lbs. SOz/MMBtu for cabbituminous and lignite respectively;
- IS = Inlet sulfur = 1.4952 and 2.23 ?8 lbs. SO 2/MMBtu for subbituminous and lignite respectively; LP = Lime price = $40/ Ton; HR= Heat rate = 10,000 Btu /kWh; AC= Ash content = 8.14 and 12.78 lbs./MMBtu for subbituminous and lignite respectively; AP = Ash disposal price = $4/ Ton; ESC 79 = factor to convert mid-1977 to mid-1979 dollaa = 1.1517.
n el'a
TABLE 5 Pago 1 of 8 REGRESSIONS RELATING NUCLEAR CAPACfrY FACTOR AND COAL EQUIVALENT AVAILABILITY FACTOR TO SELECTED DETERMINAN'IS Nuclear Capacity Factor 1 Coal equivalent Availability Factor z Regression Regression Variable Co- Variable Co-Variable Mean efficient t-Statistic 3 Mean efficient t-Statistic 8 (1) (2) (3) (4) (5) (6)
Constant 1.0000 -161.7059 -
1.0000 30.92202 -
1/Ln (1+ Vintage) " 0.3815 617.2019 2.3363 Size Indicator 1 8 0.3786 211.1855 1.9767 Size Indicator 2 s 0.3107 245.5147 2.0380 7
Size Indicator 3 0.1553 341.3599 1.3842 Size 1/Ln (1+ Vintage) 0.1505 -574.0762 -2.0496 Size 2/Ln (1+ Vintage) 0.1166 -669.7551 -2.1010 Size 3/Ln (1+ Vintage) 0.0556 -952.5255 -1.3949 1/ Log (1* Age) s 1.9081 -3.7548 -1.9474 1.4943 5.7776 2.3692 First Year Equivalent Availability Factor' 63.6315 0.3730 5.1403 Sulfur Content 18 22.1889' -0.3865 -4.0580 1/ Size 11 0.1327 75.7579 2.0509 Number of Observations 103 270 R-Squared 0.1064 0.1814 Standard Error 14.0213 14.2439 i
l't e '1" a l l
l
TABLE 5 Paga 2 cf 2 REGRESSIONS RELATING NUCLEAR CAPACTIT FACTOR AND COAL EQUIVALENT AVAILABILITY FACTOR TO SELECTED DETERMINANTS S
_OURCES AND NOTEC 1
The regression is based on the observed capacity factor for each year from 1961-1978 for commercial boiling water reactors completed through 1978.
Initial partial years were excluded. The dependent variable is capacity factor as a percent.
2The regression is based on the observed equivalent availabi'ity factor on coal units over 600 MW for each year through 1977. Initial partial years and the first full year of operation were excluded. The dependent variable is equivalent availability factor as a percent.
3A t-statistic is the ratio of the mean of the coefficient to its standard error.
It measures the reliability with which the coefficient is measured. A t-statistic of 1.96 or higher indicates that the coefficient is significantly different from zero at the 5 percent level. A t-statistic of 1.64 or higher indicates significance at the 10 percent level
' Vintage equals the year the plant came on line less 1959, so the earliest plant has a value of 1.
5The indicator equals 1 if net design rating is between 600 MW and 799 MW and 0 otherwise.
'The indicator equals 1 if net design rating is between 800 MW and 999 MW and 0 otherwise.
The indicator equals 1 if net design rating is 1,000 MW or greater and 0 otherwise.
8 Age is the number of full operating years between the commercial operation date and the year in which the equivalent availability factor is measured. A unit's first calendar year of operation (beginning the first New Year's Day on which the unit was in commercial operation) is counted as year one.
SObserved equivalent availability factor for the first full year of operation.
15 Sulfur content is expressed as percent times 10.
11 Size is measured as hundreds of megawatts. Megawatts is the FPC name-plate rating.
ITe1 a
LEVELIZED CAPACITY FACTORS FOR ALLENS CREEK AND COAL ALTERNATIVES Caoacity Factor Low ' .\ledium a g1
' Percent)
(1) (2) (3)
Allens Creek 2
No Vintage Trend 43 61 79 3
Vintage Trend Extrapolated to 1986 55 73 91 47 65 83 NERA Estimate 50 68 37 Coal Alternatives *
~
SOURCES AND NOTES 1 Low and high values are calculated as 1.28 standard errces from the medium. This reflects an 80 percent confidence band around the medium value.
2 Based on regression equation in Table 5, assuming vintage equals 78 and size is greater than 1000 31W.
3 Based on regression equation in Table 5, assuming vintage equals 86 and size is greater than 1000 31W.
- The gress equivalent availability factor is based on regression equa-tion in Table 5, assuming size is 600 51W and using the mean value of the regression data for the first year equivalent availacility factor.
The net capacity factee is estimated assuming that net capacity is 90 percent of gross capacity and that the maximum capacity*
factor is 90 percent of the equivalent availability factor.
i neea j
TABLE 7 t
FORECASTS OF URANIUM OX1DE PRICES 0.2 Percent Tails Assay 0.3 Percent Tails Assay Year Low Medium High Low Medium ph (1979 Dollars 7 Eld)
(1) (2) (3) (4) (5) (6) 1988 21.88 23.21 24.61 24.55 26.03 27.61 1989 23.08 24.83 26.74 25.88 27.86 29.98 1990 24.33 26.58 29.03 27.28 29.81 32.56 1991 27.00 29.93 33.19 30.03 33.29 36.91 1992 28.24 31.77 35.74 31.54 35.49 39.94 1993 29.90 34.15 38.98 33.43 38.18 43.58 1994 31.64 36.67 42.48 35.39 41.00 47.52 1995 33.43 39.31 46.23 37.42 44.00 51.74 1996 35.28 42.11 50.27 39.52 47.17 56.29 1997 37.21 45.08 54.59 41.71 50.50 61.17 1998 39.22 48.21 59.26 43.96 54.03 66.42 1999 41.29 51.51 S4.25 4G.31 57.77 72.05 2000 43.44 55.00 69.62 48.74 61.71 78.11 2001 41.19 52.91 67.98 47.08 60.47 77.70 2002 43.52 56.73 - 73.97 49.40 64.41 83.96 2003 44.82 59.30 78.46 50.88 67.31 89.07 2004 46.16 61.98 83.22 52.39 70.36 94.48 2005 46.31 63.12 86.00 52.57 71.64 97.63 ;
90.54 2006 47.34 65.47 53.73 74.32 102.78 2007 48.22 67.68 94.97 54.74 76.82 107.82 2008 49.10 69.93 99.61 55.74 79.39 113.07 2009 49.99 72.26 104.45 56.74 82.02 118.56 2010 50.88 74.65 109.49 57.76 84.73 124.29 2011 51.51 76.68 114.14 58.46 87.04 129.57 2012 51.65 78.03 117.89 58.63 88.58 133.83 2013 52.78 80.94 124.09 59.92 91.88 140.88 2014 52.95 82.39 128.20 60.11 93.53 145.53 2015 52.55 83.13 131.27 59.77 94.37 149.02 2016 52.17 83.60 133.97 59.22 94.89 152.07
~
2017 51.10 83.12 135.17 58.02 94.35 153.45 2018 49.37 81.47 134.48 56.04 92.50 152.65 2019 47.41 79.40 133.01 53.81 90.14 150.99 2020 46.15 78.45 133.36 52.39 89.06 151.39 Levelized Price, 1988-2020 36.30 47.39 62.97 40.94 53.50 71.17 SOURCES AND NOTES 1 Derived from NERA model of uranium oxide supply and demand.
Demand for uranium oxide is based on forecasts in U. S. Department of Energy, Grand Junction Office, Statistical Data of the Uranium Industry. January 1.1978.
nera l J
I i
ESTIMATED Furl PRICES FOR ALLENS CREEK Fuel Component Units Low Medium High (1) (2) (3)
Constant Dollar Case (1979 $/ Unit)
Uranium Oxide - 0.2% Tails 2 Lbs. 36.30 47.39 62.97 Uranium Oxide - 0.3% Tails 2 Lbs. 40.94 53.50 71.17 Conversion 3 Kg. 5.21 6.51 7.82 Enrichment 3 SWU 97.72 143.33 260.59 Fabrication 3 Kg. 117.27 156.36 195.44 Transportation and Disposal 3 Kg. 143.33 277.56 364.83 Component Purchase Current Dollar Case
- Date 5 (Current 5/ Unit' Uranium Oxide - 0.2% Tails Lbs. 7/88 62.12 81.10 107.76 Uranium Oxide - 0.3% Tails Lbs 7/88 70.06 91.56 121.80 Conversion Kg. 7/88 8.92 11.14 13.38 Enrichment SWU 7/89 176.43 258.78 470.49 Fabrication Kg. 1/90 217.47 289.96 362.44 Transportation and Disposal Kg. 7/98 418.99 723.67 1,066.48 SOURCES AND NOTES 1
Reflects the cost of each component adjusted for the : ffect of any real escalation over the life of the plant. Only Uranium Oxide prices are assumed to escalate in real terms.
2 See Table 7.
3 Prices are from draft testimony by NUS Corporation in the Matter of Generic and Environmental Statement on Mixed Oxide Fuel. NUS prices assume no recycling will take place. NUS estimates in 1975 dollars were escalated at the general rate of inflation to convert them into 1979 dollar estimates.
'Mid-1979 dollar prices were escalated to date at which spent at 6.28 percent annually through 1986 and 5.5 percent annually thereafter. In calculating levelized nuclear fuel prices, these component costs were adjusted for the effect of inflation over the life of the plant, sComponent purchase dates are based on a January 1990 start date.
n eerza
TABLE 9 Pago 1 of 2 ESTIMATED FUEL PRICES FOR THE COAL ALTERNATIVES TO ALLENS CREEK Subbituminous Lignite Low Medium M Low 31edium g (1) (2) (3) (4) (5) (6)
F.O.B. Coal Prices in 1979 $/ Ton 1990: Texas
- 11.32 11.92 13.26 New Mexico
- 11.77 11.77 11.77 Wester:t Wyoming
- 8.13 8.13 8.13 Average 9.95 9.95 9.95 2000: Texas
- 15.99 *20.19 23.96 New Mexico
- 13.00 20. 57 32.58 Western Wyoming
- 8.98 12.35 17.23 Average. 10.99 16.46 24. 90 Levelized F.O.B. Cost
- 12.54 17.44 25.26 17.37 22.01 Leve11 zed Transport Costs in 1979 $/ Ton # ,
New Mexico 13.06 18.06 18.06 Western Wyoming 22.49 22.49 22.49 Average 20.27 20.27 20.27 Levelized Delivered -
Fuel Cost 1979 $/ Ton' 32.31 37.71 45.53 17.37 22.01 28.65 1979 Mills /kWh ' 20.61 23.69 23.60 14.52 13.38 23.92 Current Mills /kWh 8
65.62 75.43 91.07 46.22 58.50 76.14 i
n e-r a l
TABLE 9 Paga 2 of 2 ESTIMATED FUEL PRICE 5 FOR THE COAL ALTERNATIVES TO ALLENS CREEK SOURCES AND NOTES 1 Reflects price forecasts in 1990 and 2000 from the NERA Electricity Supply Optimization Model 2Subbituminous coal prices reflect an average of New Mexico and Western Wyoming prices. For the years between 1990 and 2000 prices are estimated based on the growth rates for the decade. After 2000, annual prices for subbituminous coal assume a real growth rate of 1.5 percent per year; annual prices for lignite after 2000 assume a real growth rate of 1.5 percent in the low case, 2.0 percent in the medium case and 2.5 percent in the high case.
Levelized prices in 1979 dollars are estimated assuming a discount rate of 5.05 percent and a 33-year plant life.
2 Derived from the equation:
$/ Ton = 2.536 + 0.01348
- miles, assuming that New Mexico coal must be transported 1,152 miles, Western Wyoming coal must be transported 1,479 miles, and there is no real escalation in transport costs.
- The sum of transport cost and costs F.O.B. the mine.
51979 mills /kWh = (1979 $/ Ton
- Heat Rate
- 0.5)/ Heat Content. Heat content is 8,139 Btu /Lb. for subbituminous and 6,170 Stu/Lb. for lignite coal.
Plant heat rates are 10,220 for subbituminous and 10,303 for lignite and reflect scrubber energy penalties of 2.0 percent for subbituminous and 2.55 percent for lignite. For the lignite plant, costs are further adjusted for transmission losses by dividing costs by 0.9953. Levelized costs in current dollars reflect adjustments for inflation of 6.28 percent annually from 1979 to 1986 and 5.5 percent thereafter. Adjustment includes levelized effect of inflation over the life of the plant at a discount rate of 11.05 percent.
I 1
l i
1 11 e T a
TABLE 10 FINANCIAL ASSUMPTIONS USED IN PERFORMING COAL / NUCLEAR COST COMPARISONS Share Yield *
' Percent)
(1) (2)
Caoital Structure Long Term Debt .50.0 8.75 f 10.0 8.75 l Preferred Stock Common Stock 40.0 14.50 Discount Rate 11.05 AFDC Post 1978 7.50 Tax Rates Federal Income Tax . 46.00 Property Tax Through 1982 1.35 Post 1982 1.38 Tax Treatment Book Life 33 Years Tax Life:
Coal 23 Years Nuclear 16 Years Method of Depreciation double declining balance for 2 years, sum of years' digits for
~
remaining life, normalized Investment Tax Credit:
Rate 10.00 Treatment flow through Proportion of Plants Depreciable for Tax Purposes Coal 0.83 Nuclear 0.82 Through Post Escalation Rates 1986 1986
' Percent)
(1) (2)
General 6.5 5.5 ,
Capital 8.5 7.5 l O&M 7.5 5.5
)
I 1
n eT a
TABLE 11 DERIVATION OF NUCLEAR FIXED CHARGE RATE ADJUSTED FOR DECONTAMINATION AND DECOMMESSIONING COSTS Completed Capital Cost Excluding Transmission ($/KW) 1,794.99 Decontamination Cost ($/KW)*
Year 15 111.83 Year 25 191.02 Decommissioning Cost ($/KW)2 Year 33 922.86 Levelized Annual Charge ($/KW)'
Decontamination 4.24 Decommissioning 3.31 Levelized Annual Fixed Charge Rate (Percent)
Unadjusted 17.06 Adjusted for decommissioning 17.24 Adjusted for decommissioning and decontamination 17.48 SOURCES AND NOTES
' Based on decontamination costs of $24/KW in 1977 dollars oc-curring once in 15th year and once in 25th year of operation
- from a study by Sargent & Lundy Engineers, Nuclear Versus Coal Economic Study, Draft Report prepared for Arizona Puolic Ser-vice Company, Report SL - 3725, Ao'ril 1979. Exhibit III - 21.
Costs were escalated to the year of expenditure at the following general rates of escalation: 1978, 7.4 percent; 1979 through 1986, 6.5 percent annually; and post-1986,5.5 percent annually.
2 Decomissioning costs were estimated at 10 percent of costs at start of construction and escalated to the year of expenditure at the general rates of escalation in footnote 1. This estimate re-flects a conservative interpretation of results developed by Atomic Industrial Forum, National Environmental Studies Project, An Engineering Evaluation of Nuclear Power Reactor Decommissioning Alternatives, (AIF/NESP-009-0095R), Washington, D.C.,1976.
3 Based on discount rate of 11.05 percent.
n e'I' a
r ,:
DEFINITIONS OF REGIONS USED IN REGRESSION ANALYSES TABTl 1 REGIONS Northeast:
Connecticut, Maine, Maryland, Massachusetts, New Hampshire, New Jersey, New York, Pennsylvania, Rhode Island, Vermont, Washington, D.C.
Mountain / Pacific:
Alaska, Arizona, California, Colorado, Idaho, Montana, Nevada, New Mexico, Oregon, Utah, Washington, Wyoming Southwest:
Arkansas, Louisiana, Oklahoma, Texas TABLE 3 REGIONS ,
Northeast:
Connecticut, Maine, Massachusetts, New Hampshire, New Jersey, New York, Pennsylvania, Rhode Island, Vermont South:
Alabama, Arkansas, Florida, Georgia, Kentucky, Louisiana, Maryland, Mississippi, North Carolina, Oklahoma, South Carolina, Tennessee, Texas, ;
Virginia, West Virginia 1 West:
Arizona, California, Colorado, Idaho, Montana, Nevada, New Mexico, Oregon, Utah, Washington, Wyoming 1
n e'1' a l
, - . ._ _. ... .. ._.____ __ ._ _ --.