NAVY AIRCRAFT CARRIERS  - GAO/NSIAD-98-1 - August 1998 NAVY AIRCRAFT CARRIERS
Cost-Effectiveness of Conventionally and Nuclear-Powered Aircraft Carriers
GAO/NSIAD-98-1 -- August 1998

LIFE-CYCLE COSTS FOR NUCLEAR-POWERED AIRCRAFT CARRIERS ARE GREATER THAN FOR CONVENTIONALLY POWERED CARRIERS ============================================================ Chapter 3 A nuclear-powered carrier costs about $8.1 billion, or about 58 percent, more than a conventionally powered carrier to acquire, operate and support for 50 years, and then to inactivate. The investment cost for a nuclear-powered carrier is more than $6.4 billion, which we estimate is more than double that for a conventionally powered carrier. Annually, the costs to operate and support a nuclear carrier are almost 34 percent higher than those to operate and support a conventional carrier. In addition, it will cost the Navy considerably more to inactivate and dispose of a nuclear carrier (CVN) than a conventional carrier (CV) primarily because the extensive work necessary to remove spent nuclear fuel from the reactor plant and remove and dispose of the radiologically contaminated reactor plant and other system components. (See table 3.1.) Table 3.1 Life-Cycle Costs for Conventional and Nuclear Aircraft Carriers (based on a 50-year service life) (Fiscal year 1997 dollars in millions) Cost category CV CVN ------------------------------------------------------ ------ ------ Investment cost\a Ship acquisition cost $2,050 $4,059 Midlife modernization cost 866 2,382 Total investment cost $2,916 $6,441 Average annual investment cost $58 $129 Operating and support cost Direct operating and support cost $10,43 $11,67 6 7 Indirect operating and support cost 688 3,205 Total operating and support cost $11,12 $14,88 5 2 Average annual operating and support cost $222 $298 Inactivation/disposal cost Inactivation/disposal cost $53 $887 Spent nuclear fuel storage cost n/a 13 Total inactivation/disposal cost $53 $899 Average annual inactivation/disposal cost $1 $18 ====================================================================== Total life-cycle cost $14,09 $22,22 4 2 Average annual life-cycle cost $282 $444 ---------------------------------------------------------------------- Note: Numbers may not add due to rounding. \a CVN investment cost includes all nuclear fuel cost; CV fuel is included in operations and support activities. Source: Our analysis of Navy data. WHAT IS A LIFE-CYCLE COST ANALYSIS? ---------------------------------------------------------- Chapter 3:1 One of the most persistent challenges facing DOD is the ability to provide adequate resources for the acquisition, operations, and support of its systems and equipment. A life-cycle cost analysis provides managers with important information for a number of purposes, including improved long-range planning and budgeting, decisions about replacing aging equipment, and comparisons of competing programs and systems. A keystone to sound acquisition decisions is having an estimate of the total cost of a program or equipment over its full life (i.e., life-cycle cost). For this reason, a life-cycle cost analysis has been part of the DOD acquisition system for many years. The life-cycle cost is the sum total of direct, indirect, recurring, and nonrecurring costs of a system over its entire life through its disposal. The important point is that the total cost is not just the initial near term cost. Typically, a life-cycle cost analyst will focus attention on the annual operating and support cost because it usually accounts for the largest share of the total cost. The most common cost components that are included in a life-cycle cost analysis are shown in figure 3.1. Figure 3.1: Life-Cycle Cost Components Source: GAO. We developed a life-cycle cost model to estimate the life-cycle cost for a nuclear- and a conventionally powered aircraft carrier. We used the Nimitz-class carrier as our baseline for the nuclear carrier. We selected the Kennedy-class\1 as the baseline for a conventional carrier because it is comparable in size to the Nimitz-class carrier, it can employ an air wing that is similar to that on a Nimitz-class carrier, and there was adequate historical cost data available to build our cost estimates. Detailed information about the methodologies we used to develop our cost estimates is discussed in this chapter and appendix I. -------------------- \1 For our analysis, the Kennedy-class includes the CV-63, CV-64, CV-66, and CV-67 as these carriers are similar in size, displacement, crew size, and maintenance plans, and are often grouped together. INVESTMENT COSTS ARE HIGHER FOR NUCLEAR-POWERED CARRIERS THAN FOR CONVENTIONALLY POWERED CARRIERS ---------------------------------------------------------- Chapter 3:2 Ship acquisition and midlife modernization costs for a nuclear-powered carrier are double that estimated for a conventionally powered carrier. (See table 3.2.) Table 3.2 Investment Costs for Conventional and Nuclear Aircraft Carriers (Fiscal year 1997 dollars in millions) Investment category CV CVN ------------------------------------------------------ ------ ------ Ship acquisition $2,050 $4,059 Midlife modernization 866 2,382 ====================================================================== Total $2,916 $6,441 ---------------------------------------------------------------------- Note: Numbers may not add due to rounding. Source: Our analysis of Navy data. ACQUISITION COST -------------------------------------------------------- Chapter 3:2.1 Historically, nuclear carriers have cost more to acquire than conventional carriers, for several reasons. First, they are larger and heavier than conventional carriers. Second, the acquisition cost for a nuclear carrier includes nuclear fuel cost. Finally, the unique industrial base, specialized nuclear suppliers, and the Naval Nuclear Propulsion Program's exacting and stringent environmental, health, and safety standards add to this cost. For example, an unavoidably high cost overhead structure (engineering, quality assurance, and production control) and costly production work are required in the naval nuclear propulsion industry. Shipbuilders must follow šnon-deviationē plans (i.e., no deviation from the approved plans without government approval). We obtained acquisition cost data for 27 carriers that were authorized or built since 1942. Figure 3.2 shows, in constant dollars, the acquisition cost per ton\2 of conventionally and nuclear-powered carriers. Acquisition costs for the Nimitz-class nuclear-powered carriers have averaged about $3.9 billion\3 while the last conventionally powered carrier, the U.S.S. John F. Kennedy (CV-67), cost $2.1 billion. Regardless of the year the ship was commissioned, nuclear carriers cost more than conventional carriers. Figure 3.2: Acquisition Cost per Ton for Conventional and Nuclear Carriers (fiscal year 1997 dollars) Source: Navy. The acquisition costs of about $2.1 billion for the conventional carrier and $4.1 billion for the nuclear carrier were based on an estimating methodology similar to one used by us in an earlier study\4 and by the Center for Naval Analyses. Historical acquisition cost per ton and the ship's displacement tonnage were used to provide rough order of magnitude estimates for the acquisition costs of a Kennedy-class and a Nimitz-class ship. (See table 3.3 and app. I for a detailed discussion of the methodology.) Table 3.3 Acquisition Cost Estimates for the Conventionally and Nuclear-Powered Carrier (Fiscal year 1997 dollars) Ship Cost per ton\a displacement\b Estimated cost ---------------------- -------------- -------------- -------------- CV $35,191 58,268 $2,050,500,000 CVN $51,549 78,741 $4,059,000,000 ---------------------------------------------------------------------- Note: Numbers may not add due to rounding. \a We determined the ratio between the cost per ton of CV-67 and CVN-68 (less any nuclear fuel related cost) and applied this ratio to the average cost per ton for the CVN-68 class to estimate the cost per ton for a newly acquired conventional carrier. We used the average cost per ton for the CVN-68 class as the estimated cost per ton for the nuclear carrier. \b Displacement as measured in light tons. Source: Our analysis of Navy data. -------------------- \2 To allow for increases in both ship weight and cost, the acquisition cost per ton is produced by dividing the ship's acquisition cost by its tonnage. We used the ship's light displacement tonnage that did not include weight due to the ship's fuel, water, or ammunition. \3 Includes nuclear fuel cost. \4 Navy's Aircraft Carrier Program: Investment Strategy Options (GAO/NSIAD-95-17, Jan. 1995). MIDLIFE MODERNIZATION -------------------------------------------------------- Chapter 3:2.2 Our analysis shows that it will cost the Navy nearly three times as much to refuel, overhaul, and modernize a nuclear carrier than it will a conventional carrier. Both carrier types require some level of midlife modernization to allow the Navy to use the ships for nearly 50 years. When a nuclear carrier has been in service for nearly 24 years,\5 it will undergo a 2-1/2 year refueling complex overhaul (RCOH), which includes refueling the reactor plant, repairing the propulsion plant, restoring the ship's material condition, and performing modernization efforts. The nuclear carrier is expected to operate another 24 years after the RCOH. Similarly, at about age 30, a conventional carrier can undergo a 2-1/2 year Service Life Extension Program (SLEP) which includes restoring the ship's condition, installing system upgrades and performing modernization efforts. After a SLEP, the conventional carrier should operate an additional 15 years or more. In both cases, the actual work required for midlife modernization will vary for each individual ship, depending on the ship's condition. Figure 3.3 shows the estimated RCOH cost for the first two Nimitz-class carriers and the actual cost for SLEPs performed on the Kennedy-class\6 conventional carriers. We used the Navy's best estimate for the RCOH planned for the CVN-68 as the estimated midlife modernization cost for the nuclear carrier. For our cost model, we used the average historical cost for SLEPs performed on the Kennedy-class conventional carriers as the estimated midlife modernization cost for the conventional carrier. Figure 3.3: Midlife Modernization Cost for CVN-68, CVN-69 and Kennedy-class Conventional Carriers Note: We used estimated cost for nuclear carriers and actual cost for the conventional carriers. Source: Navy. -------------------- \5 Given current operating tempo, its nuclear fuel is expected to last about 24 years before it needs to be replaced. Midlife modernization will take place at the time of refueling. \6 The CV-66 did not undergo a SLEP. NUCLEAR CARRIERS ARE MORE EXPENSIVE TO OPERATE AND SUPPORT THAN CONVENTIONAL CARRIERS ---------------------------------------------------------- Chapter 3:3 We estimate that it will cost about $14.9 billion to operate and support a nuclear carrier over its lifetime, which is nearly 34 percent more than the $11.1 billion we estimate it will cost to operate and support a conventional carrier. As shown in table 3.4, the cost for a conventionally powered carrier's fossil fuel is more than offset by the added cost for a nuclear-powered carrier's personnel and depot maintenance. A major cost difference between the two carrier types is in the indirect cost category for support activities provided by DOE to the nuclear carriers. Table 3.4 Life-Cycle Direct and Indirect Operating and Support Costs for a Conventionally and Nuclear-Powered Carrier (Fiscal year 1997 dollars in millions) Cost category CV CVN ------------------------------------------------------ ------ ------ Direct operating and support costs Personnel $4,636 $5,206 Fossil fuel 738 n/a Depot maintenance\a 4,130 5,746 Other\b 933 724 Total direct operating and support costs $10,43 $11,67 6 7 Indirect operating and support costs Training $161 $1,107 Fossil fuel delivery 469 n/a Nuclear support activities n/a 2,045 Other\c 58 53 Total indirect operating and support costs $688 $3,205 ====================================================================== Total operating and support costs $11,12 $14,88 5 2 ---------------------------------------------------------------------- Note: Numbers may not add due to rounding. \a Includes routine maintenance, repairs, and ship modernization work but not the cost of midlife modernization. \b Includes a number of direct unit cost categories such as spare parts, supplies, and intermediate maintenance. \c Includes a number of indirect support cost categories such as publications, ammunition handling, and technical services. Source: Our analysis of Navy data. DIRECT OPERATION AND SUPPORT COSTS -------------------------------------------------------- Chapter 3:3.1 Nuclear carriers have higher personnel costs than conventional carriers primarily because they require additional personnel and their nuclear personnel receive special pay and bonuses. Depot maintenance costs are greater for a nuclear carrier because more labor is needed to maintain it than is needed for a conventional carrier. In a separate analysis, we found that the total cost related to nuclear fuel over a nuclear carrier's lifetime exceeded the lifetime cost for a conventional carrier's fossil fuel. We did not review the reasons for the cost differences for other elements in the direct cost category because it included costs from a number of subcategories such as travel, spare parts, and intermediate maintenance. In most cases, the cost difference for each individual subcategory was not significant enough to warrant detailed review. PERSONNEL COSTS ------------------------------------------------------ Chapter 3:3.1.1 The personnel cost for a nuclear carrier is estimated at about $5.2 billion over its lifetime compared to about $4.6 billion for a conventional carrier (see table 3.5). Our estimated personnel cost is based on historical personnel cost reported by the Navy's Visibility and Management of Operating and Support Cost (VAMOSC) Management Information System database for the CV-67 class and CVN-68 class carriers as well as an additional 30.6 percent to account for accrued retirement cost, which is not captured by that database. Table 3.5 Personnel Costs for Conventionally and Nuclear-Powered Carriers (Fiscal year 1997 dollars in millions) CV CVN ------------------------------------------------------ ------ ------ Annual personnel cost $71 $80 Annual accrued retirement 22 24 Total annual $93 $104 Life-cycle cost $4,636 $5,206 ---------------------------------------------------------------------- Note: Numbers may not add due to rounding. Source: Our analysis of Navy data. To determine why the nuclear-powered carrier's personnel cost is higher, we analyzed ship manpower documents. On the basis of our work, we are reasonably certain that most of the cost difference can be attributed to the different propulsion plants and not some other ship characteristic. Our work further indicates that the higher cost are the result of three factors: increased personnel needed to operate the nuclear propulsion plant, higher grade structure for propulsion plant personnel and bonuses, and special duty pay for nuclear personnel. A nuclear carrier has a requirement for nearly 3,400 personnel compared to about 3,200 for a conventional carrier. A majority of the added personnel can be traced to the departments that operate the propulsion plant.\7 Table 3.6 provides a comparative summary of the required propulsion plant personnel for nuclear and conventional carriers. Table 3.6 Propulsion Plant Personnel for Conventional and Nuclear Carriers Personnel CV CVN Difference ------------------------------------------ ------ ------ ---------- Officers\a 22 33 11 Enlisted 597 716 119 ====================================================================== Total 619 749 130 ---------------------------------------------------------------------- \a Includes warrant officers. Source: Navy. Some of the higher personnel cost for the nuclear carrier can be attributed to the grade structure of propulsion plant personnel. Nuclear propulsion plant personnel are a higher grade level than propulsion plant personnel for a conventional carrier. As shown in figure 3.4, nearly 50 percent of a nuclear carrier's enlisted propulsion plant personnel are E-5 and above whereas 75 percent of a conventional carrier's propulsion personnel are E-4 and below. Figure 3.4: Grade Structure for the Enlisted Propulsion Plant Personnel for Conventionally and Nuclear-Powered Carriers Source: Our analysis of Navy data. The higher personnel cost for a nuclear carrier can also be attributed to special pay and bonuses. The Navy uses a variety of incentive pay and bonuses to attract and retain nuclear personnel (see table 3.7). Table 3.7 Special Pay and Bonus Incentives Available for Nuclear Personnel in Fiscal Year 1997 Enlisted ---------------------------------- ---------------------------------- Enlistment bonus Up to $4,000 Selective reenlistment bonus Up to $30,000\a Special duty assignment pay Up to $275 per month\b Officers Accession bonus Up to $8,000 Annual incentive Up to $12,000\c ---------------------------------------------------------------------- \a Variable bonus based on award level. Nominal bonus of $22,000 for a 6-year reenlistment of an E-5 with 24 months in service. \b An E-5 would receive $150 per month. \c Variable bonus based on active duty commitment length. Source: Navy. -------------------- \7 This includes the Engineering Department for the conventionally powered carrier and the combination of the Engineering and Reactor Departments for the nuclear-powered carrier. FUEL COSTS ------------------------------------------------------ Chapter 3:3.1.2 We estimate it costs $738 million to provide fossil fuel over a conventional carrier's 50-year life. Historical data indicate that a conventional carrier uses about 500,000 barrels of fossil fuel each year or about 25 million barrels over its lifetime. Our estimate was developed by multiplying the estimated barrels of fuel consumed by $29.52, which was the average per barrel price the Navy paid for fossil fuel from fiscal year 1991-95. Because of historical interest and speculation as to escalation in fossil fuel prices, we examined trends in fossil fuel prices and performed a sensitivity analysis of a conventional carrier's life-cycle cost if fuel prices were to double (see app. I). The fuel cost for a conventional carrier is clearly visible as an operating and support cost. Conversely, the fuel cost for a nuclear carrier is included within the investment cost (e.g., acquisition, midlife modernization) and, therefore, is not clearly identified. We compared the costs of fossil and nuclear fuel and found that the life-cycle cost for nuclear fuel is greater than for fossil fuel. Given current operating tempo, the nuclear fuel cores for the Nimitz-class carrier are expected to provide enough power for about 24 years. When the initial fuel cores are depleted, the cores are removed and replacement fuel cores are installed. Replacement cores will be removed when the ship is inactivated at the end of its service life. To provide a comparison of fuel cost, the Navy identified portions of the investment costs that directly relate to the initial and replacement cores (see table 3.8). Table 3.8 Nuclear Fuel Cost for a Nimitz-class Carrier (Fiscal year 1997 dollars in millions) Total cost ------------------------------------------------------ -------------- Initial core ---------------------------------------------------------------------- Uranium $24 Fuel core procurement 308 Fuel installation 12 Fuel removal at midlife 150 Replacement core ---------------------------------------------------------------------- Uranium 24 Fuel core procurement 308 Fuel installation 64 Fuel removal at inactivation 85 ====================================================================== Life-cycle nuclear fuel cost $975 ---------------------------------------------------------------------- Note: Numbers may not add due to rounding. Source: Navy. The uranium costs shown were based on what it cost to manufacture during the late 1980s, when it was last manufactured for the Navy. The uranium used to fuel nuclear reactors is supplied by DOE.\8 DOE ceased production of defense grade uranium in 1991. There is sufficient uranium to operate the Navy's nuclear force for decades. Excess uranium can be reconfigured and sold to private utility companies for use in commercial reactor plants. An alternative cost methodology based on the opportunity cost for the uranium appears in appendix I. Table 3.9 provides a comparative summary of fuel costs for conventionally and nuclear-powered carriers. Table 3.9 Comparison of Life-Cycle Fuel Cost for Conventionally and Nuclear-Powered Carriers (Fiscal year 1997 dollars in millions) Total Annual life ized Fuel type cost cost ------------------------------------------------------ ------ ------ CVN nuclear fuel $975 $19.5 CV fossil fuel $738 $14.8 ---------------------------------------------------------------------- Source: Our analysis of Navy data. -------------------- \8 All domestic enrichment services were handled by DOE until 1993, when these operations were transferred to the United States Enrichment Corporation. DEPOT MAINTENANCE COST ------------------------------------------------------ Chapter 3:3.1.3 We estimate that the life-cycle cost for depot maintenance for a conventional carrier is about $4.1 billion compared to about $5.7 billion for a nuclear carrier. Depot maintenance activities include routine maintenance, repairs, alterations, and fleet modernization expected to be performed over the life of the carrier. Midlife modernization activities, although performed at the depot-level, are not included in our estimated depot maintenance cost. Rather, they are included as part of investment costs. Our cost estimates were not based on the historical depot maintenance costs captured by the Navy's VAMOSC database for several reasons. First, the cost data collected for the nuclear carriers reflected depot maintenance performed under the Engineered Operating Cycle (EOC) maintenance plan. Since the Navy is changing its depot maintenance strategy for nuclear carriers, we did not believe that historical costs would provide the best estimate for depot maintenance costs. Second, the VAMOSC data, which captured costs for each fiscal year between 1985 and 1994 and the 10-year average cost, could lead to over- or underestimating this cost because of the carriers' age. For instance, the average depot maintenance cost for nuclear carriers reflected maintenance and fleet modernization work performed on nuclear ships that had been in commission, on average, 10 years or less. Conversely, the average depot maintenance cost for conventional carriers reflected maintenance and fleet modernization work performed on carriers that had been in commission for an average 25 years. Because of these reasons, we developed depot maintenance cost estimates based on the Navy's notional plans for routine maintenance and fleet modernization for conventionally and nuclear-powered carriers. Estimates of how often depot maintenance will be needed (interval), how many months the ship will be in maintenance (duration), and how much shipyard labor (labor workdays) will be needed for the carrier are guided by the Chief of Naval Operations. Figure 3.5 shows the notional depot maintenance cycle for the two carrier types. Figure 3.5: Notional Depot Maintenance Cycle for Conventionally and Nuclear-Powered Carriers (in months) Source: Navy. To estimate depot maintenance cost, we determined the number and types of depot maintenance periods that would occur over each of the carrier's 50-year service life. Based on the Navy's notional plans, we determined that the nuclear carrier would need about 38 percent more workdays of labor for anticipated depot-level maintenance and fleet modernization over its 50-year life than the conventional carrier. Next, we developed an estimated cost for each type of depot maintenance by multiplying the number of labor workdays\9 by the composite labor workday rates\10 for public and private shipyards. We estimated additional depot costs for materials, centrally procured equipment, spare parts, and other miscellaneous items based on our analysis of historical costs for these items. Table 3.10 summarizes the estimated depot maintenance and fleet modernization costs for nuclear and conventional carriers. Table 3.10 Life-Cycle Cost for Depot Maintenance of a Conventionally and a Nuclear-Powered Carrier (Fiscal year 1997 dollars in millions) Number in Cost Life- lifeti per cycle Depot maintenance type me type cost ---------------------------------------------- ------ ------ ------ CV SRA 17 $51.3 $872 COH 6 543.0 3,258 ====================================================================== Total $4,130 CVN PIA1 2 $188.3 $377 PIA2 4 214.6 858 PIA3 9 240.8 2,167 DPIA1 1 327.7 328 DPIA2 2 376.8 754 DPIA3 3 421.2 1,263 ====================================================================== Total $5,746 ---------------------------------------------------------------------- Note: Numbers may not add due to rounding. Source: Our analysis of Navy data. The actual depot maintenance cost for the nuclear carrier could be more than our estimate and lower than our estimate for the conventional carrier because we used composite or average shipyard labor rates. However, our earlier work\11 showed that the actual labor cost for nuclear work was higher than the labor cost for nonnuclear work. Specifically, we found that the average cost of $213 per workday for nuclear labor (then-year dollars) was 25 percent more than the average cost for nonnuclear labor, which was about $170 per workday. We also found that the applied overhead for nuclear work was an average of $303 per workday, nearly 60 percent more than the average overhead of $189 per workday for nonnuclear work. Maintenance of nuclear ships costs more due to the complex nature of nuclear work. Shipyards must provide a greater level of service, pay higher costs for specially trained and skilled workers, and maintain specialized shipyard departments that support nuclear work, such as radiological controls, nuclear engineering, nuclear planning, and nuclear quality assurance. In addition, shipyards must make provisions to package and dispose of nuclear waste that is generated during maintenance. The cost to process low-level waste generated during maintenance was included in the depot maintenance cost estimate. Parts and other materials used in nuclear systems are often of a unique design and require specialized material. -------------------- \9 The Navy provided labor workdays estimated for both maintenance and fleet modernization for each type of depot maintenance. \10 The Navy provided composite public and private shipyard rates that are the average cost for labor and overhead. \11 Nuclear-powered Ships: Accounting for Shipyard Costs and Nuclear Waste Disposal Plans (GAO/NSIAD-92-256, July 1, 1992). INDIRECT OPERATION AND SUPPORT COSTS -------------------------------------------------------- Chapter 3:3.2 A significant sustaining base is needed to support both the nuclear and the conventional carriers. Some examples include logistics services, training, engineering, and software support. The VAMOSC database captures many of the costs for these activities and reports a portion of the costs as an indirect operating and support cost for each ship. However, we identified several supporting activities and functions that were not captured or were partially captured by that database, such as training, fossil fuel delivery and nuclear support activities. Indirect support costs for a nuclear carrier are significantly more than those for a conventional carrier. We estimate a nuclear carrier's indirect support cost--$3.2 billion--is nearly five times more than the estimated $0.69 billion for a conventional carrier. This difference is primarily due to several expensive activities that support the nuclear carrier but are not needed for the conventional carrier. TRAINING COST ------------------------------------------------------ Chapter 3:3.2.1 We estimate that the Navy spends nearly $1 billion more to provide a steady flow of trained personnel to operate and maintain a nuclear carrier's propulsion plant than it does to train personnel to operate and maintain a conventional carrier's propulsion plant. The primary reason is the nuclear carrier has a greater requirement for personnel with specialized skills than a conventional carrier does. Our estimate was not based on the historical VAMOSC database because it does not capture certain training costs that are central to the differences in the propulsion system, nor does its allocation method provide visibility to reasons for the cost difference. Therefore, we developed estimates for initial and specialized training pipeline costs for both carrier types. To estimate indirect training cost, we used a methodology similar to one developed by the Navy's Center for Cost Analysis. The methodology is based on determining the annual training requirement needed to provide a steady flow of skilled personnel that are required in the propulsion plant departments for each carrier type. The annual training requirement is determined by the number of billets, crew turnover, and attrition.\12 For our analysis, we determined the annual training requirement initial skills training for four enlisted ratings (machinist's mates, electrician's mates, electronics technicians and boiler technicians)\13 and the annual training requirement for additional specialized\14 training beyond the initial level. We selected these skill types because they accounted for most of the difference between the required skills for propulsion plant departments of the two types of carriers. We also determined the annual training requirement for specialized officers skills.\15 Using information provided by the Navy, we identified the training courses required for these skills, and the Chief of Naval Education and Training provided the cost per graduate for each course. The Chief of Naval Education and Training did not have cost information for the required 26 weeks for practical training required for specialized officer and enlisted nuclear skills. Therefore, we estimated the cost per officer and enlisted student based on Navy budget data. Table 3.11 compares the indirect training costs. Table 3.11 Propulsion Plant Pipeline Training Costs for a Conventionally and Nuclear- Powered Carrier (Fiscal year 1997 dollars in millions) Initia Specia l l Life- traini traini Annual cycle ng ng cost cost -------------------------------------- ------ ------ ------ ------ Conventional carrier Enlisted $2.50 $0.71 $3.21 $160.6 2 Officers \a ====================================================================== Total $2.50 $0.71 $3.21 $160.6 2 Nuclear carrier Enlisted $4.28 $16.47 $20.75 $1,037 .57 Officers \a 1.38 1.38 69.10 ====================================================================== Total $4.28 $17.85 $22.13 $1,106 .67 ---------------------------------------------------------------------- Note: Numbers may not add due to rounding. \a Our model assumed the training cost for officer's initial training would be the same for both carriers. Source: Our analysis of Navy data. -------------------- \12 Annual training requirement = ((billet /turnover) *attrition). \13 As of October 1, 1996, the boiler technicians rating was merged with the machinist's mates rating. \14 We focused on Navy Enlisted Classifications (NECs) that are earned after required training has been satisfied. \15 We focused on additional qualification designation (AQD) for officer billets, which indicated a requirement for skills and knowledge beyond those implicit by the officer's billet classification or submarine specialty. FOSSIL FUEL DELIVERY ------------------------------------------------------ Chapter 3:3.2.2 We estimate that it costs about $469 million to deliver fossil fuel to a conventional carrier over its lifetime. Our estimate is based on a Navy methodology\16 that allocates a portion of the annual operating and support cost for these facilities and ships to each barrel of fossil fuel issued. For example, since Navy depots store fuels other than fossil fuel, the total cost to operate and maintain these facilities is allocated based on the proportion of fossil fuel to total fuel issued from each depot. This cost is then divided by the number of barrels of fossil fuel delivered to produce the delivery cost per barrel. The same method was used to determine the cost per barrel for the Navy ships and oilers operated by the Military Sealift Command (MSC) that deliver fossil fuel. Table 3.12 shows how the fossil fuel delivery cost of $18.77 per barrel cost was computed. Assuming a conventional carrier uses about 500,000 barrels of fossil fuel per year, or 25 million barrels over its lifetime, we estimate that it will cost about $469 million to deliver fuel to the conventionally powered carrier. Table 3.12 Cost to Deliver Fossil Fuel to a Conventionally Powered Carrier (Fiscal year 1997 dollars) Portion Cost Annual allocated allocated Activity/ operating and to fuel for fuel Barrels of Cost per ship type support delivery delivery fuel issued barrel ------------ -------------- ---------- ------------ ------------ ---------- Fleet supply $54,295,049 41.77% $22,679,042 10,526,000 $2.19 centers\a Ships Navy ships\b $294,480,540 45-65% $144,815,575 MSC oilers $273,057,960 100% $273,057,960 ================================================================================ Total $567,538,500 $417,873,535 25,198,595 $16.58 ================================================================================ Total per $18.77 barrel delivery cost Number of 25,000,000 barrels delivered over a conventiona l carrier's lifetime ================================================================================ Total cost $469,250,0 to deliver 00 -------------------------------------------------------------------------------- Note: Numbers may not add due to rounding. \a The cost per barrel for the fleet supply centers also includes an allocation of $0.04 for the annual operation of service craft. \b Navy ships included are the AO, AOE, and AOR class ships. The percentage of the ship cost allocated to the delivery of fossil fuel varied from 45 to 65 percent. Source: Our analysis of Navy data. -------------------- \16 The Navy's cost methodology, which has been used for many years, includes an annualized acquisition cost as well as the annual operating and support cost. However, our analysis does not include an allocation of acquisition cost because we did not have comparable acquisition cost data for facilities (i.e., DOE laboratories) that support the delivery of nuclear power to the Navy's fleet. NUCLEAR SUPPORT ACTIVITIES ------------------------------------------------------ Chapter 3:3.2.3 We estimate that it will cost about $2.04 billion to provide supporting services to the nuclear carriers. Most of this cost is due to the work performed by the activities of the Bettis and Knolls Atomic Power Laboratories, large research and engineering facilities that are solely dedicated to support the Navy's nuclear propulsion program. More than half of the laboratories' activities are funded through the DOE appropriation for the Naval Nuclear Propulsion Program. DOE's budget for the Program averaged $731 million annually during the 1990s. Program activities are primarily focused on operational research, development, and testing\17 for the purpose of gaining a better understanding of reactor behavior fundamentals, evaluating reactor performance, and verifying and improving the accuracy of models. The laboratories evaluate cladding, structural, and component materials suitable for use in the operating nuclear plant and develop and test nuclear fuel. The laboratories also evaluate equipment and systems that transfer, convert, store, control, and measure power that the naval reactor has created. The DOE-funded activities are an integral and necessary component to providing effective military nuclear propulsion plants to the Navy and to ensure their safe and reliable operation. In addition to the support provided by DOE, the Navy also budgets about $200 million in operation and maintenance funds to provide essential technical and logistical support for its operational reactors. The types of activities include routine maintenance and engineering support, inspection and refurbishment of reactor plant components, safety surveillance at shipyards, and power plant safety assessments. Our cost estimate was based on allocating a portion of the annual cost for these activities to a nuclear carrier. For the DOE-funded activities, we allocated 5 percent of DOE's average funding between fiscal year 1991 and 1997 and 2.08 percent of the Navy's average funding between fiscal year 1994 and 1996 for Navy support activities (see table 3.13). Table 3.13 Cost to Provide Nuclear Support Activities to a Nuclear Carrier (Fiscal year 1997 dollars in millions) Percen t Allocated Life- Annual alloca annual cycle cost ted cost cost ---------------------------------- ------ ------ ---------- ------ Energy-funded $731.0 5.0\a $36.6 $1,828 Navy-funded 208.4 2.08\b 4.3 217 ====================================================================== Total $2,045 ---------------------------------------------------------------------- Note: Numbers may not add due to rounding. \a DOE activities were allocated on the basis of a nuclear carrier's demand for power in relationship to other nuclear ships in the Navy's fleet. (See app. I for additional information regarding the relative demand for power nuclear ships.) \b The Navy-funded activities were allocated based on our analysis of the number of reactor plants supported by these funds. Source: Our analysis of DOE and Navy data. -------------------- \17 The Navy's research development test and evaluation budget is the source for obtaining funds needed for specific development, test, and evaluation of new reactors. For example, about $413 million (then-year dollars) of the budget for fiscal years 1994 to 1998 will fund the laboratories' development of the reactor for the New Attack Submarine. These funds are not included in this discussion. NUCLEAR CARRIERS ARE MORE COSTLY TO INACTIVATE AND DISPOSE OF THAN CONVENTIONAL CARRIERS ---------------------------------------------------------- Chapter 3:4 Nuclear carriers are significantly more expensive to inactivate and dispose of than conventional carriers. The cost to inactivate and dispose of a nuclear carrier is estimated at $887 million. In addition, it will cost the Navy an estimated additional $13 million to provide long-term storage\18 of the spent nuclear fuel (SNF) after it is removed from the carrier's reactor plant. On the basis of Navy data, we estimate that the cost to inactivate and dispose of a conventional carrier is $52.6 million. -------------------- \18 Radioactive materials will need safe storage for thousands of years. Our estimate is based on the radioactive materials storage requirement during the first 100 years after a carrier is commissioned. CARRIER DISPOSAL COST -------------------------------------------------------- Chapter 3:4.1 A conventional carrier can be placed in the reserve fleet or retained as a mobilization\19 asset at the end of its service life. When the Navy no longer needs a conventional carrier, it can transfer the carrier to the Maritime Administration, sell the carrier to a private firm or foreign government, or sell the carrier for its scrap value. Our estimate of $52.6 million to inactivate and dispose of a conventional carrier is based on Navy data and includes the cost to place the carrier in reduced mobilization status, 3 years maintenance in a reduced mobilization status, and final disposal cost less scrap value. These are not realistic options for a nuclear aircraft carrier because of its nuclear propulsion systems. A nuclear ship is constructed with a nuclear power plant inside a section of the ship called the reactor compartment. The components of the nuclear power plant include a high-strength reactor vessel, heat exchangers (steam generator), and associated piping, pumps, and valves. Each reactor plant contains over 100 tons of lead shielding, part of which is made radioactive by contact with the radioactive material. At the end of its useful service life, the nuclear carrier and its radioactive materials must be disposed of. Although a nuclear carrier has never been disposed of, the basic steps necessary to dispose of a carrier would be similar to those performed on nuclear submarines and surface ships. The first step is defueling the reactor plant. The highly radioactive SNF is removed from the reactor and sent to the Naval Reactor Facility, located at DOE's Idaho National Engineering and Environmental Laboratory for examination and temporary storage. (Disposition of spent nuclear fuel is discussed later in this chapter.) Next, piping systems, tanks, and vessels are drained; the radioactive systems are sealed; and the reactor compartment is sealed and enclosed in a high integrity steel package. Reactor compartments removed from submarines have been transported by barge from Puget Sound Naval Shipyard to DOE's Hanford, Washington, site for final burial. The Navy provided us with a cost range of between $818.6 million and $955.5 million to dispose of the first Nimitz-class nuclear-powered carrier. We used the mid-point cost in our analysis--$887 million. Most of the cost can be attributed to defueling and removing contaminated nuclear equipment and material. This estimate did not include the cost associated with storing the SNF or any cost associated with maintaining oversight of the reactor plant's burial site in Hanford. -------------------- \19 The Navy keeps three carriers in Mobilization B status. When a carrier is taken out of active service, it is placed in a Mobilization B status; the oldest carrier in Mobilization B status is then disposed of. SPENT NUCLEAR FUEL STORAGE -------------------------------------------------------- Chapter 3:4.2 Spent nuclear fuel (SNF) will be removed from the carrier's reactor plant twice during the Nimitz-class carrier's service life--at its midlife and at inactivation. Because it is highly radioactive, SNF will need to be safely stored for thousands of years. Based on estimates recently provided in DOD's official comments on our draft report, the Naval Nuclear Propulsion Program now estimates it will cost about $13 million to safely store the SNF during the first 100 years after a nuclear-powered carrier is decommissioned using a new dry storage method (see table 3.14). We were unable to verify the accuracy and completeness of this estimate but we do know that the new method promises to be significantly less expensive than the method formerly used, called the wet storage method. Table 3.14 Navy Cost Estimate for the Dry Storage of a Nuclear-Powered Carrier's Spent Nuclear Fuel (Fiscal year 1997 dollars in millions) Cost -------------------------------------------------------------- ------ Initial cores Hardware per ship set of cores $ 4.8 Operational costs for 75 years of dry storage 1.8 Replacement cores Hardware per ship set of cores 4.8 Operational costs for 50 years of dry storage 1.7 ====================================================================== Total cost $13.0 ---------------------------------------------------------------------- Note: Numbers may not add due to rounding. Source: Naval Nuclear Propulsion Program. The Navy has been temporarily storing its SNF using a wet storage method. Under this method the nuclear propulsion program stores the fuel in special pools located at DOE's Idaho National Engineering and Environmental Laboratory. The water in the pools serves the dual purpose of acting as the barrier for the radiation and dispersing the heat in the SNF. Using this method, DOE estimated that it would cost about $306 thousand to receive and place the Nimitz-class cores into the storage pools and an additional $1.144 million for each year the cores are stored. The storage cost begins to accumulate after the first cores are removed during the carrier's refueling complex overhaul near its 25th year of service. The storage cost will double when the replacement cores are removed upon carrier inactivation. Temporary storage costs for naval SNF are likely to change as the Navy transitions\20 to the dry storage method. Table 3.14 reflects the anticipated savings from adopting the new method of spent nuclear fuel storage. Ultimately, SNF will have to be permanently disposed of, which will present an extremely difficult challenge because it will remain dangerous for thousands of years. The national strategy focuses on disposal of SNF generated by civilian nuclear power plants and high-level waste in a geologic repository. DOE is responsible for developing an underground repository. However, DOE does not expect that the repository will be operational until about 2010,\21 more than 10 years behind that envisioned. Thus, estimating the cost is complicated because the current repository plans were not based on disposing defense-grade SNF, such as that from naval reactor plants, but on the high-level waste that is generated from reprocessing defense-grade SNF. Thus, we did not estimate final disposal costs for the carrier's SNF but instead focused on the current storage practices. -------------------- \20 In October 1995, the state of Idaho, the Navy, and DOE reached an agreement regarding the shipment and storage of SNF in Idaho. As a result, all SNF located at DOE's Idaho site will be placed in dry storage by 2023 and all SNF will be removed from Idaho by 2035. \21 Nuclear Waste: Foreign Countries' Approaches to High-Level Waste Storage and Disposal (GAO/RCED-94-172, Aug. 4, 1994) and Nuclear Waste: Comprehensive Review of the Disposal Program Is Needed (GAO/RCED-94-299, Sept. 27, 1994). AGENCY COMMENTS AND OUR EVALUATION ---------------------------------------------------------- Chapter 3:5 DOD partially agreed with our life-cycle analysis. DOD agreed that a thorough understanding of total life-cycle costs is key to allocating scarce resources. However, DOD disagreed that comparing the life-cycle costs of conventionally powered carriers such as the U.S.S. John F. Kennedy with Nimitz-class nuclear-powered carriers was appropriate because of differences in the age, size, and capabilities of the carriers. DOD agreed that the life-cycle cost of nuclear-powered carriers is greater than conventionally powered carriers but that the premium is not as large as estimated by us. DOD did not agree with our approach of making cost-per-ton comparisons between nuclear-powered Nimitz-class carriers and conventionally powered carriers such as the Kennedy. DOD believed that it would be more appropriate to compare conventionally and nuclear-powered carriers with equivalent capabilities. While the nuclear-powered Nimitz-class carrier and the conventionally powered Kennedy-class carrier are not identical, we chose to compare them because the Kennedy was the last and largest conventionally powered carrier built, it employs an airwing of comparable size to that of the Nimitz-class carriers, and there was adequate historical data available. Further, both classes of carriers have performed the same missions for more than two decades. Our estimate of the difference in costs between the two types of carriers is greater than DOD's estimate primarily because of differing methodologies. Our acquisition cost estimate was based on a cost-per-ton methodology, which is an accepted method for estimating these costs and has been used by the Navy and others. The actual acquisition cost for the Kennedy, adjusted for inflation, is virtually the same as the acquisition costs used in our estimate. According to the Navy, it estimated the cost for a "new" conventionally powered carrier with the capabilities of the newest Nimitz-class carriers and assumed the conventionally powered carrier would have a larger displacement than a Nimitz-class carrier. It stated that the conventionally powered carrier's cost was based on actual manhours adjusted to reflect manhour and material growth over 30 years. Navy officials told us they assumed the conventionally powered carrier would be constructed at Newport News Shipbuilding and therefore used that company's cost factors (for example, labor rates, overhead rates, and material rates). In addition, because the Navy did not include the cost of the Kennedy's SLEP when it calculated the average cost for a CV SLEP, its estimate was greater than ours. Our operating and support costs were based on historical data for the conventionally powered Kennedy-class and the nuclear-powered Nimitz-class. The DOD estimate is based on estimates of a much larger conventionally powered carrier as discussed above. We also used different methodologies for estimating indirect costs (see app. VII, comments 33, 34, and 35). The Navy chose to estimate personnel cost using its Manpower Billet Cost Factor Model. This model is intended to estimate the full manpower cost, including indirect cost, such as training. We did not use this model because much of the nuclear training costs are not captured by that model. Instead, we used validated historical costs to estimate direct personnel costs. We separately estimated training pipeline cost using a methodology developed by the Navy's Center for Cost Analysis. The Navy also used a different method for allocating the annual cost of DOE's laboratories that support its nuclear fleet. We allocated this cost based on the demand/consumption of nuclear power. Finally, the Navy's new estimate to inactivate and dispose of the CVN is nearly 40 percent less than the estimate used in our analysis. We did not use the Navy's newer estimate because no evidence was provided or found to support the significant reductions in cost to the original estimate provided to us. DOE concurred with DOD's comments addressing estimates of costs associated with nuclear reactor plant support activities and the storage of SNF. These comments and our evaluation of them are discussed in appendix VII.