Pacific Enters. & Subsidiaries v. Commissioner

101 T.C. No. 1, 101 T.C. 1, 1993 U.S. Tax Ct. LEXIS 42

• Court: United States Tax Court
• Decided: July 12, 1993
• Docket: Docket No. 5295-91
• Status: Published

Opinion

Cohen, Judge:

Respondent determined the following deficiencies in, and additions to, petitioner’s Federal income taxes:

Year Deficiency Addition to tax sec. 6661

1977 -0-

1978 $6,695,614

1979 3,246

1980 6,949,217

1981 1,594,037

1982 -0-

1985 82,639,888 $5,354,116

Unless otherwise indicated, all section references are to the Internal Revenue Code in effect for the years in issue, and all Rule references are to the Tax Court Rules of Practice and Procedure.

After concessions, including respondent’s concession that the section 6661 addition to tax does not apply, the issues for decision are:

(1) Whether the cushion gas contained in underground gas storage reservoirs and the line pack gas contained in gas pipelines are items of inventory or capital assets;

(2) whether the 1985 and 1986 adjustments to the amounts of cushion gas in petitioner’s reservoirs were changes in estimates or changes in accounting methods requiring the Secretary’s approval under section 446(e); and

(3) whether the standard for depreciation of cushion and line pack gases is economic or physical recoverability.

We must also determine the total and recoverable amounts of cushion gas and line pack gas for the years in issue.

FINDINGS OF FACT

At the time of filing the petition, Pacific Enterprises & Subsidiaries (petitioner) was a publicly held California corporation with its principal office in Los Angeles, California. Petitioner, a holding company, owned two gas utility subsidiaries, Southern California Gas Co. (SoCalGas) and Pacific Lighting Gas Supply Co. (PLGS). In 1985, these subsidiaries merged, and petitioner remained as the holding company for the surviving entity, SoCalGas. The objective of both utilities was to provide safe, reliable, and efficient natural gas service to their respective customers.

During the years in issue, SoCalGas was a public utility that operated underground gas storage reservoirs and an extensive gas pipeline transmission and distribution system in order to provide natural gas service to residential, commercial, and industrial customers in southern and central California. SoCalGas was the largest natural gas distribution utility in the United States with over 3 million residential customers, 160,000 commercial customers, and 26,000 industrial customers.

PLGS was a public utility engaged in acquiring, transporting, sorting, and exchanging natural gas for the purpose of selling the gas to SoCalGas for resale; SoCalGas purchased all of the available gas of PLGS.

On November 26, 1985, PLGS merged with SoCalGas, with SoCalGas’ being the surviving corporation. Before the merger, petitioner included both subsidiaries in petitioner’s consolidated Federal income tax returns, which were prepared under the accrual method of accounting; after the merger, petitioner included SoCalGas, the surviving subsidiary, in the consolidated returns. Before the merger, for Federal income tax, regulatory, financial, and other accounting purposes, SoCalGas used the first in, first out (FIFO) method of accounting to identify its natural gas inventories and used the cost method to value such inventories. PLGS used the last in, first out (lifo) method to identify its natural gas inventories and used the cost method to value such inventories. As a result of the merger, SoCalGas adopted the LIFO method to identify its natural gas inventories in 1985. Both utilities used historical cost to value gas held as capital assets.

Supply and Demand for Natural Gas

Thq demand for natural gas fluctuated hour to hour and seasop to season. Peak demand occurred during the morning and évening hours and in the winter season (beginning in November and ending in March). On cold days, the use of natural gas by residential and commercial customers of SoCalGas was as much as seven times greater than on warm days, primarily because of increased use of natural gas for space and water heating. The supply of natural gas, on the other hand, was relatively constant.

SoCalGas and PLGS received most of their natural gas supplies from El Paso Natural Gas Co. (El Paso) and Transwestern Pipeline Co. (Transwestern). Other suppliers included California producers, Federal offshore properties, other out-of-State suppliers, and other subsidiaries of petitioner. Because the demand for gas fluctuated and the supply was relatively constant, underground storage of natural gas was the principal method of balancing the supply and demand for gas.

Underground Storage Reservoirs

Like many reservoirs, petitioner’s underground storage reservoirs for natural gas were rock formations and not open caverns; the reservoirs had varying densities, consistencies, sizes, and depths that nevertheless were permeable and porous. “Permeability” is a measure of the ease with which gas or any other fluid can move through porous rock. “Porosity” of a storage reservoir is a measure of the pore or void spaces between grains of sand and rock where gas is stored.

The detailed plans for petitioner’s underground storage reservoirs were made prior to the installation of wells and equipment for withdrawing natural gas. Among the variables that petitioner considered in choosing the reservoirs for storage were: (1) The number of wells to be drilled and operated, (2) the amount of compression horsepower to be installed, and (3) the maximum pressure to which the reservoir could be raised. For safety and economic reasons, maximum reservoir pressure was set at the original pressure encountered when the discovery well was drilled in the reservoir. The volume of petitioner’s reservoirs changed gradually over time due to production of oil and water by other users of the reservoirs and due to pressures of the aquifers surrounding the reservoirs.

Before the 1985 merger, SoCalGas operated two underground storage reservoirs (Honor Rancho and Playa Del Rey) and PLGS operated four (La Goleta, East Whittier, West Montebello, and Aliso Canyon). SoCalGas operated all six after the merger. The reservoirs were all located in southern California.

The reservoirs were designed to meet the demand on an extreme peak day, which is a day on which demand for gas is high due to extreme weather conditions. SoCalGas used the coldest day in California during the last 56 years as the extreme peak day for designing its reservoirs; that day occurred in 1937.

Categories of Natural Gas in Reservoirs

Natural gas stored in storage reservoirs is divided into working gas and cushion gas. These categories of gas are defined as follows:

1. Working gas: The volume of natural gas in the reservoirs above the designed level of cushion gas. Working gas may or may not be completely withdrawn during any particular output season.

2. Cushion gas: The minimum volume of gas required in an underground storage reservoir to provide the pressure necessary to deliver working gas volume to meet customer demands. Cushion gas is subdivided as follows:

a. Recoverable cushion gas: The volume of cushion gas that will be recovered when the reservoir is abandoned.

b. Nonrecover able cushion gas: The volume of cushion gas that will remain in place in the reservoir after the reservoir is abandoned.

The natural gas contained in a reservoir at discovery of the reservoir is termed “native gas”. Native gas can be either recoverable or nonrecoverable.

Honor Rancho

The Honor Rancho reservoir was obtained in the early 1970s, when SoCalGas found it necessary to increase its underground gas storage capacity. The reservoir was to be used primarily for seasonal storage. SoCalGas leased Honor Rancho from the County of Los Angeles for $144,886 per year prior to September 8, 1980, and for $252,248 per year after September 7, 1980. This reservoir was chosen because it met the following capabilities: It had a working gas storage capacity of 20 to 30 billion cubic feet (bcf), a maximum deliv-erability of 1 bcf of gas per day through the end of January, and a minimum deliverability of 60 million cubic feet (mmcf) of gas per day at the end of a winter withdrawal season. Deliverability is the measure of the volume of gas flowing out of a reservoir or pipeline at a given pressure. Deliverability decreases as gas is withdrawn.

In connection with the acquisition of Honor Rancho, James W. Fairchild of Fairchild, Ancell & Wells, Inc. (Fairchild), conducted a computerized simulation to estimate Honor Ran-cho’s working gas capacity and potential deliverability in order to determine if they were within the range of interest of SoCalGas. Fairchild relied on industry-wide gas injection and withdrawal patterns because Honor Rancho had previously been operated as an oil field and had no history as a gas storage reservoir. Based on the simulation, SoCalGas initially estimated Honor Rancho’s gas storage capacity as follows:

In bcf

22.5 Working gas .

Cushion gas: Recoverable .... io co

N onrecoverable iq oó

Total capacity .... io co

When Honor Rancho was purchased, it contained 7.5 bcf, or $397,178, of native gas that was included in the purchase price of the reservoir.

Injection of natural gas into the Honor Rancho reservoir began in September 1975 while the field was still producing oil and gas. SoCalGas used compressors for injection but not to withdraw gas from its reservoirs, because the additional investment in equipment for withdrawing gas was not economically beneficial.

After SoCalGas began gas storage operations at Honor Rancho, it determined, in 1978, that it could inject approximately 2.5 bcf of additional gas, which represented an increase in working gas capacity.

In 1983, the senior management of SoCalGas decided to increase the minimum deliverability rate of gas from 60 mmcf per day to 100 mmcf per day. The established minimum wellhead flowing pressure at Honor Rancho was 850 pounds per square inch gauge (psig), the internal pressure per square inch that would be measured with a pressure gauge. The other pressure measurement used in the natural gas industry is pounds per square inch absolute (psia), which is 14.7 pounds per square inch (psi) greater than a pressure measured as psig. Psia takes account of the background pressure caused by the atmosphere, and, in southern California, the atmospheric pressure is 14.7 psi.

During the winter of 1979-80, SoCalGas withdrew more working gas than ever before and discovered that, when approximately 17.5 bcf of Working gas was withdrawn, the field did not operate according to projections.

During 1983 through 1985, SoCalGas began to review the operations of all of its underground gas storage reservoirs. As part of this effort, during the winter of 1984-85, SoCalGas intentionally withdrew working gas from the Honor Rancho reservoir to test its deliverability at low storage volumes. SoCalGas concluded that the cushion gas in the reservoir was insufficient to maintain working gas capacities above 17.5 bcf, so the company reclassified 7.5 bcf of gas, or $31,057,500, from working gas (inventory) to recoverable cushion gas (capital asset). The Honor Rancho volume changes are summarized as follows:

Honor Rancho Gas Volumes (In bcf)

Initial estimate 1978 revision 1985 revision Change 1978-85

Working gas capacity 22.5 25.0 17.5 (7.5)

Recoverable cushion gas 3.5 3.5 11.0 7.5

Nonrecoverable cushion gas 8.5 8.5 8.5 0.0

Total capacity 34.5 37.0 37.0 0.0

The reclassification was reviewed and approved by a vice president and a senior vice president of SoCalGas. The reclassification had no effect on income in 1985. The reclassification was prospectively reflected in the financial and regulatory accounting records of SoCalGas but was not retroactively reflected because SoCalGas treated the reclassification as a change in an accounting estimate.

Playa Del Rey (PDR)

SoCalGas acquired the PDR reservoir from the Federal Government in 1953. The PDR storage reservoir is located approximately 6,500 feet below the surface in a partially depleted oil reservoir near the Marina Del Rey yacht harbor. SoCalGas used PDR to balance the daily variation in demand. SoCalGas established minimum and maximum deliverability rates for PDR at 200 mmcf and 600 mmcf per day, respectively. SoCalGas established the minimum wellhead flowing pressure for PDR at 650 psig. PDR is not a single reservoir but interconnected reservoir areas known as the townsite area, the Del Rey gas cap, and the main storage area. Only the main storage area is used by PDR for gas storage operations. Initially, SoCalGas estimated the total capacity of PDR to be 7 bcf, comprised of 3.9 bcf of working gas, 1.9 bcf of recoverable cushion gas, and 1.2 bcf of nonrecoverable cushion gas.

In December 1986, based on a computer simulation of PDR, SoCalGas concluded that 1.3 bcf of working gas, or $564,980, was to be reclassified as recoverable cushion gas. The PDR volumes are summarized as follows:

Playa Del Rey Gas Volumes (In bcf)

Initial 1986 estimate revision Change

Working gas capacity CO H bi 0} CO

Recoverable cushion gas CO H k> 0} H

Nonrecoverable cushion gas © 6 to H

Total capacity 7.0 7.0 0.0

The recommendation to reclassify the gas was reviewed and approved by a vice president and a senior vice president of SoCalGas. As with Honor Rancho, SoCalGas did not retroactively restate its financial or regulatory accounting records, because SoCalGas treated the reclassification as a change in an accounting estimate.

Regulated, Accounting for Gas in Reservoirs

SoCalGas and PLGS were regulated by the California Public Utilities Commission (cpuc). cpuc requires utility subsidiaries to operate their storage reservoirs in a “reasonable and prudent” manner, requires gas utilities to justify all expenditures as just and reasonable, establishes the rates charged for gas, and requires the utilities to provide safe and reliable service to customers, cpuc also requires utility subsidiaries to use the cpuc Uniform System of Accounts. Neither SoCalGas nor plgs was regulated by the Federal Energy Regulatory Commission (FERC), but the CPUC system of accounts was based on accounts set up under the Federal Power Commission, the predecessor of FERC.

Accounting for the various categories of gas in petitioner’s reservoirs under the CPUC Uniform System of Accounts was as follows:

Type of gas CPUC account

Working gas Inventory

Recoverable cushion gas Nondepreciable capital asset

Nonrecoverable cushion gas Depreciable capital asset

Recoverable line pack gas Nondepreciable capital asset

Nonrecoverable line pack gas Depreciable capital asset

Other gas in the pipelines Inventory

CPUC determined the rates that SoCalGas charged based on estimated future revenues and costs. Certain assets and other fixed costs, such as the original cost of the completed utility plant, less depreciation and deferred taxes, were aggregated to form a “rate base”. Variable costs, including labor, gas, and maintenance, were aggregated to form “expenses”. SoCalGas revenue was calculated as expenses plus a rate of return on the “rate base”.

Controlling the Supply of Gas

SoCalGas followed a priority system established by CPUC for emergency restrictions in gas supplies. The order of priority in the system, from highest to lowest, was as follows:

Priority 1: Residential and small commercial customers with no alternative fuel source;

priority 2: commercial and industrial customers with no alternative fuel source;

priority 3: commercial gas boiler users;

priority 4: industrial customers with the ability to switch to fuel oil to meet their boiler fuel needs; and

priority 5: utility electric generators with the ability to switch to alternate fuels, such as Southern California Edison Co.

In gas shortages and emergencies, the gas supplies were to be curtailed starting with the lowest priorities. Thus, priority 4 and 5 customers were referred to by petitioner as “interruptible customers”.

Due to seasonal demand fluctuations, SoCalGas injected gas into reservoirs in the summer for withdrawal in the winter. Compressors are used to inject working gas into transmission pipelines, whereas working gas flows out freely without the aid of equipment during withdrawal.

The withdrawal of cushion gas never occurred at the Honor Rancho reservoir and occurred only once at the pdr reservoir, in 1969, during testing of that reservoir. Some structural damage to PDR occurred as a result of the withdrawal of the cushion gas. Rather than continue to withdraw cushion gas and further damage the reservoirs, SoCalGas and PLGS would curtail deliveries to “interruptible customers”.

Abandonment of Storage Reservoirs

The “Ten Section Field” reservoir was acquired in the late 1970s by PLGS in a joint effort with Pacific Gas & Electric Co. Ten bcf of gas was injected into the reservoir after acquisition. Because Ten Section Field was never used, in 1983, PLGS abandoned it. Upon abandonment, 6 bcf of gas was withdrawn and put into the PLGS/SoCalGas distribution system; the remaining 4 bcf of gas was sold in place with the reservoir. CPUC approved the sale of Ten Section Field with the 4 bcf of cushion gas still in place, because it would not have been cost effective to remove that gas.

At the time of trial of this case in August 1992, petitioner was in the process of abandoning the East Whittier reservoir.

Pipeline System

In addition to the reservoirs, SoCalGas and PLGS owned the pipelines that serviced natural gas customers. Before the 1985 merger, the pipeline systems of SoCalGas and the pipeline system of PLGS were operated as an integrated system.

The SoCalGas and PLGS pipeline system operated as follows: natural gas was received from out-of-State suppliers or from other sources and injected into a large-diameter-pipe system called the transmission system. The transmission system also received gas from underground storage. Regardless of its origin, the gas injected into the pipeline traveled through the pipeline to various compressor stations that received the gas, compressed it, and released it forward into the pipeline at pressures in excess of the original pressure. There were 12 or 13 compressor stations in the transmission system during the years in issue. The pressure differential, or gradient, between these compressors caused the gas to move along each section of the transmission system from one compressor to the next. Natural gas was delivered from the transmission system into a series of smaller diameter pipes called the distribution system. The distribution system transported the gas closer to the locations of customers. Regulators were used to lower the gas pressure between the transmission system and the distribution system. From the distribution system, gas moved through additional regulator stations into the distribution network. The distribution network branched out into successively smaller diameter pipe to customers’ meters.

In the transmission and distribution systems, the molecules of gas already in the pipeline were pushed forward toward the customers; these molecules were constantly replaced so that a constant volume of gas remained in the pipelines. This volume of gas was “line pack” gas. Line pack gas was, thus, the minimum volume of gas in the pipeline that was needed to deliver gas to customers. The compressors would not have functioned and the system would not have delivered gas to customers unless line pack gas was constantly present.

Depending on the size and location of the pipe in the system, the transmission system pipelines operated at maximum pressures ranging from 125 to 400 psig. Distribution system pipelines operated at maximum pressures ranging from 150 to 175 psig, and distribution network pipelines operated at pressures ranging from 40 to 60 psig. Gas regulators at each residence reduced the gas pressure from a range of 5 to 25 psig to just a few ounces psig.

For each of the years 1977 through 1985, the volume of line pack gas in the transmission system was approximately 1.4 bcf, and the volume of line pack gas in the distribution lines was 56 mmcf. Petitioner treated 75 percent of the total volume of gas in the pipelines as a capital asset and the remaining 25 percent as inventory.

The term “line pack gas” was also used by oil and gas companies to refer to the “packing and drafting” of gas in the transmission system. Because natural gas is compressible, in addition to using the underground reservoirs, SoCalGas would compress some of the excess gas deliveries into the pipeline, or “pack the line”, and then later withdraw that gas, or “draft the line”. The two line packing terms are not synonymous. Petitioner accounted for the “packed” and “drafted” gas as inventory, whereas line pack gas was accounted for as a depreciable capital asset.

OPINION

We must first decide whether petitioner was required to treat cushion gas and line pack gas as inventory as opposed to capital assets. We must next decide 4f petitioner’s 1985 reclassification of a portion of its working gas to cushion gas was a change in accounting method. We must then determine the correct volumes of cushion gas and line pack gas in petitioner’s reservoirs and pipelines. Finally, we must decide whether the depreciation of cushion and line pack gases is based on economic or physical recovery, and, under that standard, the volumes of nonrecoverable cushion gas and line pack gas.

Cushion Gas and Line Pack Gas as Inventory or as Capital Assets

Petitioner excluded from inventory the cushion gas kept in its underground reservoirs and the line pack gas kept in its pipelines; petitioner capitalized these items. Respondent determined that all of the gas held in petitioner’s reservoirs arid pipelines was inventory, except for the nonrecoverable portions of those gases.

Respondent possesses broad discretion to determine whether a particular method of accounting clearly reflects income and, if it does not, to require a change to a method that, in respondent’s opinion, does clearly reflect income. Capitol Fed. Sav. & Loan v. Commissioner, 96 T.C. 204, 209 (1991). Thus, the issue becomes whether petitioner’s accounting for the gas as a capital asset clearly reflects income.

Section 446(a) provides that “Taxable income shall be computed under the method of accounting on the basis of which the taxpayer regularly computes his income in keeping his books.” Section 446(b), an exception to this general rule, provides that “If no method of accounting has been regularly used by the taxpayer, or if the method used does not clearly reflect income, the computation of taxable income shall be made under such method as, in the opinion of the Secretary, does clearly reflect income.” Section 471(a) provides:

Whenever in the opinion of the Secretary the use of inventories is necessary in order clearly to determine the income of any taxpayer, inventories shall be taken by such taxpayer on such basis as the Secretary may prescribe as conforming as nearly as may be to the best accounting practice in the trade or business and as most clearly reflecting the income.

Whether the particular accounting method clearly reflects income is a question of fact. Coors v. Commissioner, 60 T.C. 368, 394 (1973), affd. sub nom. Adolph Coors Co. v. Commissioner, 519 F.2d 1280 (10th Cir. 1975). If a method of accounting reflects a consistent application of Generally Accepted Accounting Principles (gaap), it will ordinarily be regarded as clearly reflecting income. Hallmark Cards, Inc. v. Commissioner, 90 T.C. 26, 31 (1988). But see Thor Power Tool Co. v. Commissioner, 439 U.S. 522 (1979). If a taxpayer’s method of accounting is set forth in the Internal Revenue Code or in the regulations, respondent may not reject that method as not providing a clear reflection of income if the taxpayer has applied that method on a consistent basis. Hallmark Cards, Inc. v. Commissioner, supra.

A large portion of the gas in question is classified as recoverable and is nondepreciable. The effect of respondent’s classification of that gas as inventory in this case occurs because petitioner used a FIFO inventory system and was subject to large price fluctuations in purchasing its natural gas.

Under FIFO accounting, the cost of gas in ending inventory was based on the volume in thousands of cubic feet (mcf) of gas in ending inventory and running back through the volume and cost per mcf of each month’s deliveries to storage until the appropriate volume in mcf was achieved. For example, if the ending inventory was 20 million mcf and the volume and cost per mcf of monthly deliveries to storage during the year were as shown below, the value of ending inventory would be $49 million, calculated as follows:

Month Deliveries to storage Cost per mcf Inventory volume Inventory cost

December 2,000,000 mcf $2.60 2,000,000 mcf $5,200,000

November 1,500,000 2.60 1,500,000 3,900,000

October 4,000,000 2.50 4,000,000 10,000,000

September 4,500,000 2.50 4,500,000 11,250,000

August 2,500,000 2.30 2,500,000 5,750,000

July 3,000,000 2.30 3,000,000 6,900,000

June 4,000,000 2.40 2,500,000 6,000,000

20,000,000 49,000,000

If petitioner’s cushion gas and line pack gas were also included in ending inventory, those volumes would be valued at the latest prices as shown above, rather than at lower historical costs. The result of the changed valuation would be a decrease in the cost of gas sold and an increase in taxable income, which would be particularly notable in periods of rising prices.

Petitioner presents a dual argument to support its claim that the cushion gas and line pack gas should not be treated as inventory. First, petitioner argues that the volumes of cushion and line pack gas do not fit the definition of inventory or merchandise; these volumes of gas were not held primarily for sale to customers in the ordinary course of petitioner’s business. See Wood v. Commissioner, 16 T.C. 213, 225-226 (1951). Second, petitioner argues that accounting for cushion gas and line pack gas as capital assets clearly reflected income and was consistently applied in accordance with GAAP, in accordance with prevalent industry practice, as allowed by CPUC, and in accordance with the Internal Revenue Code.

Transwestern Pipeline Co. v. United States, 225 Ct. Cl. 399, 639 F.2d 679 (1980) (per curiam), considered the issue of whether line pack gas is inventory and held that the volume of gas was not inventory but rather a capital asset. The issue in Transwestern was whether the cost of line pack gas in the taxpayer’s pipeline system should be treated as a depreciable capital expenditure or as inventory. The court found that (1) line pack gas was an essential component without which the taxpayer’s pipeline could not operate; (2) a vast majority of the line pack gas would be lost on abandonment of the transmission system; (3) line pack gas met the definition of a fixed asset under GAAP, under industry standards, and under the system of accounts of the Federal Power Commission (now FERC); and (4) capitalization and depreciation of line pack gas clearly reflected income under the matching principle. Transwestern Pipeline Co. v. United States, supra at 681, 685. The trial judge in Transwestern found:

The delivery of natural gas from the system is dependent on a minimum pressure level prevailing in the system at all times; and this, in turn, is dependent on the presence of line pack in the system. The system will not deliver gas at the contractual requirements of Transwestern’s customers unless the line pack is constantly present. * * *

When the line pack is viewed in the manner outlined in the preceding paragraph, line pack seems to be as much of an integral part of Transwestern’s pipeline system — and, therefore, as much a part of the basic capital asset that the pipeline system represents — as the pipe in the pipeline, or the compressors, or any other essential component without which the pipeline system could not operate.

[Id. at 684.]

In Arkla, Inc. v. United States, 765 F.2d 487 (5th Cir. 1985), revg. 592 F. Supp. 502 (W.D. La. 1984), the court addressed the capital-asset-versus-inventory aspects of cushion gas. In Arkla, the Commissioner conceded that nonrecoverable cushion gas was a capital asset that was eligible for an investment tax credit (itc). Therefore, the court addressed only the question of whether recoverable cushion gas was eligible for that credit. The District Court concluded that both recoverable and nonrecoverable cushion gas were capital assets, depreciable and eligible for ITC. 592 F. Supp. at 510-511. The Court of Appeals for the Fifth Circuit agreed that the cushion gas was a capital asset but found that (1) recoverable cushion gas was not depreciable because the taxpayer could not suffer an economic loss by holding the gas as a capital asset and (2) recoverable cushion gas was not eligible for the itc because it lacked a deter-mínate useful life. 765 F.2d at 490. Petitioner’s position in this case is consistent with that court’s conclusion that the cushion gas is a capital asset.

Respondent asserts that the physical similarity or fungibility between recoverable cushion gas and working gas molecules means that recoverable cushion gas is sold to customers in the ordinary course of business. The volumes of cushion and working gas are material here, however, not the individual molecules of gas. As the Court of Appeals for the Fifth Circuit noted in Arkla: “That recoverable and nonrecoverable cushion gas are physically identical does not require them to be treated identically for depreciation purposes.” Id. The same is true in distinguishing between recoverable cushion gas and working gas. That the two categories of gas are physically identical does not require them to be treated identically; cushion gas is used for a different purpose than working gas. See Latimer-Looney Chevrolet, Inc. v. Commissioner, 19 T.C. 120, 125 (1952) (certain cars removed from inventory were considered capital assets because they were used as company cars rather than being held for sale to customers). The Court stated that the cars were not inventory because “it is not the nature of the property itself which is determinative * * * but rather the purpose for which the property is held.” Id.; see also Honeywell, Inc. v. Commissioner, T.C. Memo. 1992-453.

Petitioner argues and the evidence establishes that cushion gas and line pack gas were not held by petitioner with the purpose of selling the gas to customers in the ordinary course of business. The volume of recoverable cushion gas of SoCalGas would not be sold until abandonment of the reservoirs or in emergencies. Gas in the reservoirs never fell below the cushion gas volume at Honor Rancho and fell below cushion gas volume at PDR only once, during testing of that reservoir. That cushion gas would be available for sale at the time the reservoir was abandoned does not transform the gas into working gas (inventory) sold in the ordinary course of business. Most capital assets may be sold upon termination of the business in which they are used. Recoverable cushion gas is, in effect, the salvageable portion of total cushion gas.

Respondent argues that the accounting method used for cushion gas was not consistently applied because petitioner reclassified certain portions of working gas (inventory) in 1985. This argument will be addressed below. Prior to 1985, petitioner consistently applied its accounting method in determining the volumes of cushion gas, working gas, and line pack gas.

We conclude that cushion gas and line pack gas are capital assets and not inventory.

Change in Method of Accounting

In 1985, SoCalGas reclassified 7.5 bcf of gas at Honor Ran-cho and, in 1986, reclassified 1.3 bcf of gas at PDR from working gas (inventory) to cushion gas (capital asset), based on engineering reports that the volumes of gas that were previously carried on the books as cushion gas were too low. Based on the engineering claims, petitioner has asked for a recomputation of prior year taxes to establish the following overpayments of taxes:

Year Overpayment

1977 . $2,562,019

1978 . 2,283,817

1979 . 836,302

1980 . 8,502,740

1981 . 355,096

1982 . 11,790,935

Total . 26,330,909

Although 1986 is not before the Court, whether the reclassification of 1.3 bcf of gas at PDR was proper affects the overpayment claims for 1977 through 1982; therefore, the reclassification of gas at PDR will be considered for this limited purpose. Petitioner’s claims in this case are that it was proper to reclassify 7.5 bcf of gas at Honor Rancho and 1.3 bcf of gas at PDR from working gas (inventory) to cushion gas (capital asset) and that these reclassifications were not changes in its method of accounting. Petitioner’s requested adjustments stem from the FIFO method of accounting prior to the 1985 merger. Petitioner’s suggested reclassification of working gas as cushion gas removes lower value inventory gas from both beginning and ending inventory categories and places that gas in a capital asset category. This, in effect, would cause the cost of gas sold value to be higher in each of the years from 1977 to 1982.

Respondent has disallowed the $26 million refund, contending that petitioner changed its method of accounting when it reclassified working gas to recoverable cushion gas without the consent of the Secretary. (Respondent has conceded that the 1985 deficiency should be reduced by $31 million because the reclassification had no effect on income for that year.)

Our initial inquiry is whether the 1985 and 1986 reclassi-fications at Honor Rancho and PDR are changes in method of accounting, because these would require consent of the Secretary before the changes could be implemented by the taxpayer. The consent was not obtained by SoCalGas.

Section 446(e) requires that “a taxpayer who changes the method of accounting on the basis of which he regularly computes his income in keeping his books shall, before computing his taxable income under the new method, secure consent of the Secretary.”

Section 1.446-1(e)(2), Income Tax Regs., provides in part:

(2)(i) Except as otherwise expressly provided in chapter 1 of the Code and the regulations thereunder, a taxpayer who changes the method of accounting employed in keeping his books shall, before computing his income upon such new method for purposes of taxation, secure the consent of the Commissioner. Consent must be secured whether or not such method is proper or is permitted under the Internal Revenue Code or the regulations thereunder.

(ii)(a) A change in the method of accounting includes a change in the overall plan of accounting for gross income or deductions or a change in the treatment of any material item used in such overall plan. Although a method of accounting may exist under this definition without the necessity of a pattern of consistent treatment of an item, in most instances a method of accounting is not established for an item without such consistent treatment. A material item is any item which involves the proper time for the inclusion of the item in income or the taking of a deduction. * * *

(b) A change in method of accounting does not include correction of mathematical or posting errors, or errors in the computation of tax liability (such as errors in computation of the foreign tax credit, net operating loss, percentage depletion or investment credit). Also, a change in method of accounting does not include adjustment of any item of income or deduction which does not involve the proper time for the inclusion of the item of income or the taking of a deduction. * * * A change in the method of accounting also does not include a change in treatment resulting from a change in underlying facts. On the other hand, for example, a correction to require depreciation in lieu of a deduction for the cost of a class of depreciable assets which had been consistently treated as an expense in the year of purchase involves the question of the proper timing of an item, and is to he treated as a change in method of accounting.

(c) A change in an overall plan or system of identifying or valuing items in inventory is a change in method of accounting. Also a change in the treatment of any material item used in the overall plan for identifying or valuing items in inventory is a change in method of accounting.

(iii) A change in the method of accounting may be illustrated by the following examples:

‡ ❖ % ífc ^

Example (4). From 1968 through 1970, a taxpayer has fairly allocated indirect overhead costs to the value of inventories on a fixed percentage of direct costs. If the ratio of indirect overhead costs to direct costs increased in 1971, a change in the underlying facts has occurred. Accordingly, an increase in the percentage in 1971 to fairly reflect the increase in the relative level of indirect overhead costs is not a change in method of accounting but is a change in treatment resulting from a change in the underlying facts.

‡ # * # * * 5}«

Example (6). A taxpayer in the manufacturing business has for many taxable years valued its inventories at cost. However, cost has been improperly computed since no overhead costs have been included in valuing the inventories at cost. The failure to allocate an appropriate portion of overhead to the value of inventories is contrary to the requirement of the Internal Revenue Code and the regulations thereunder. A change requiring appropriate allocation of overhead is a change in method of accounting because it involves a change in the treatment of a material item used in the overall practice of identifying or valuing items in inventory.

Example (7). A taxpayer has for many taxable years valued certain inventories by a method which provides for deducting 20 percent of the cost of the inventory items in determining the final inventory valuation. The 20 percent adjustment is taken as a “reserve for price changes.” Although this method is not a proper method of valuing inventories under the Internal Revenue Code or the regulations thereunder, it involves the treatment of a material item used in the overall practice of valuing inventory. A change in such practice or procedure is a change of method of accounting for inventories.

In considering whether petitioner’s reclassifications of cushion gas are barred by section 446(e), it is not sufficient that petitioner merely show the correctness of a new method; that fact alone cannot justify a change without the Secretary’s consent. Southern Pac. Transp. Co. v. Commissioner, 75 T.C. 497, 681 (1980), supplemented by 82 T.C. 122 (1984). We also explained in Southern Pacific:

In addition, consent is required when a taxpayer, in a court proceeding, retroactively attempts to alter the manner in which he accounted for an item on his tax return. If the alteration constitutes a change in the taxpayer’s method of accounting, the taxpayer cannot prevail if consent for the change has not been secured. Casey v. Commissioner, [38 T.C. 357] supra at 385-386 [(1962)]; Cubic Corp. v. United States, an unreported case (S.D. Cal. 1974, 34 AFTR 2d 74-5895, 74-2 USTC par. 9667), affd. per curiam 541 F.2d 829 (9th Cir. 1976). [Id. at 682.]

Petitioner cites section 1.446-l(e)(2)(ii)(&), Income Tax Regs., for the proposition that “A change in the method of accounting * * * does not include a change in treatment resulting from a change in underlying facts.” Petitioner asserts that SoCalGas did not change its method of accounting when it reestimated the “underlying fact” of the actual cushion gas volume in its reservoirs. Petitioner relies on ESCO Corp. v. United States, 750 F.2d 1466 (9th Cir. 1985).

In ESCO Corp., the taxpayer adopted a more sophisticated forecasting methodology that allowed the taxpayer to estimate its workers’ compensation claim expenses accurately and to deduct workers’ compensation claim expenses in the current year. The Court of Appeals for the Ninth Circuit held that the taxpayer had not changed its method of accounting when it—

deducted only a portion of its accrued expenses * * * because of insufficient statistical data and forecasting methodologies, the use of more sophisticated techniques * * * cannot be considered a change in accounting method. The new techniques more accurately predicted * * * [the taxpayer’s] expenses and allowed it to avoid the underaccruals it had been experiencing. * * * [Id. at 1470.]

Petitioner also relies on Baltimore & O.R.R. v. United States, 221 Ct. Cl. 16, 603 F.2d 165 (1979). In that case, the taxpayer adopted a more accurate formula for estimating the salvage value of rails; the court held that the taxpayer had not changed its method of accounting by adopting the more accurate estimates of salvage value. Id. at 171.

Petitioner likewise asserts that the reclassifications can be considered corrections of mathematical or posting errors under section 1.446-l(e)(2)(ii)(6), Income Tax Regs. Petitioner relies on the cases of Standard Oil Co. (Indiana) v. Commissioner, 77 T.C. 349 (1981); Underhill v. Commissioner, 45 T.C. 489 (1966); and Gimbel Bros., Inc. v. United States, 210 Ct. Cl. 17, 535 F.2d 14 (1976).

In Standard Oil Co., 77 T.C. at 383, the Court held that the taxpayer did not change its method of accounting when it began deducting intangible drilling costs (IDC’s) that it had erroneously capitalized. The taxpayer was complying with a previously chosen method of accounting that required deduction rather than capitalization of the IDC’s. Id. at 382-383.

In Underhill, 45 T.C. at 491, the taxpayer changed from prorating investment income to the cost-recovery method of recording income on certain speculative investments. The Court concluded that the reclassification was not a change in accounting method under section 446(e), because the reclassification did not involve the proper method or time of reporting an item of income. Id. at 496. The taxpayer “had no choice in determining whether an obligation was speculative or nonspeculative, nor was there any doubt or choice about the method or time of reporting income once that determination was made.” Id. at 497.

In Gimbel Brothers, 535 F.2d at 23, the Court of Claims concluded that the taxpayer had not changed its method of accounting when the taxpayer reclassified certain rotating charge accounts from the accrual method to the installment method, because the charge accounts should have been recorded under that method initially.

These cases, cited by petitioner, do not involve inventory identification or valuation. Section 446(e) and the related regulations, however, specifically deal with inventory changes. In particular, section 1.446-l(e)(2)(ii)(c), Income Tax Regs., provides that a change in the treatment of a material item that is used in identifying or valuing items of inventory is a change in the method of accounting.

Respondent points to example (7) of that regulation, supra p. 19, which states that a change from an incorrect inventory method to a correct one is a change in method of accounting. Respondent also cites several Tax Court cases that deal with the question of whether a change in inventory value is a change in accounting method. Three cases deal with section 481 but are pertinent because section 481 defers to section 446(e) for the definition of a change in method of accounting. The cases are Hamilton Indus. Inc. v. Commissioner, 97 T.C. 120 (1991); Wayne Bolt & Nut Co. v. Commissioner, 93 T.C. 500 (1989); and Primo Pants Co. v. Commissioner, 78 T.C. 705 (1982).

In Hamilton Indus. Inc. v. Commissioner, supra, the taxpayer attempted to shield the recognition of gain on inventory acquired in a bargain purchase by treating that inventory and subsequently acquired raw materials and manufactured goods as a single item of inventory under the LIFO method. The Court concluded that this practice was unacceptable for tax purposes and that it constituted a change in accounting method from that previously used by the taxpayer. Id. at 127.

In Wayne Bolt & Nut Co. v. Commissioner, supra, the taxpayer had utilized for a number of years a sampling method for determining the value of its ending inventory. When the taxpayer finally took a complete physical count of its inventory, it discovered that approximately $2 million worth of inventory that had been previously written off was actually still in inventory. To correct the problem, the taxpayer increased its opening and ending inventories. The Commissioner did not challenge the correction by the taxpayer but did maintain that the correction was a change in the method of accounting and that an adjustment should be made under section 481, recapturing in income the cost of items mistakenly written off in prior years. The Court agreed. Id. at 513.

In Primo Pants Co. v. Commissioner, supra, the taxpayer consistently valued its inventories as a percentage of cost when its inventories should have been valued at full cost. The taxpayer attempted to avoid a section 481 adjustment by alleging that the practice was not a change in accounting method. The Court disagreed and held that deferral of income until final closing inventory was corrected was a timing question that constituted a change in accounting method. Id. at 725.

Diebold, Inc. v. United States, 891 F.2d 1579 (Fed. Cir. 1989), involving reclassification of inventory as a capital asset, also supports respondent’s position. The Court of Appeals found that, although the reclassification was between two preexisting accounts, the inventory was not similar to or in the same category as other items in the capital asset account; thus, the reclassification was a change in accounting method under section 446(e). Id. at 1582-1583. The court stated: “Section 446(e) prohibits taxpayers from unilaterally amending their returns simply because they have discovered that a different method of accounting yields a lower tax liability than the method they originally chose.” Id. at 1583. The court disallowed the taxpayer’s refund claim because the taxpayer did not attempt to obtain the Secretary’s consent for reclassifying the inventory. Id. at 1584.

We hold that petitioner’s reclassification of a portion of working gas (inventory) to capital asset accounts was a change in accounting method because, under section 1.446(e)(2)(ii)(a) and (c), Income Tax Regs., the reclassification changed a material item that was used in the identification of inventory. The reclassification is material, not only because of the large dollar amount involved but also because it was a change that affected the timing of income. By reclassifying inventory as a capital asset, $26 million of income would be deferred until abandonment of the reservoir. Such a change in accounting required approval of the Secretary. Because approval was not requested, the Commissioner has authority under section 446(e) to disallow the reclassification. The reclassification can be disallowed even though the original method of accounting is incorrect. Diebold, Inc. v. United States, 891 F.2d 1579, 1583 (Fed. Cir. 1989). We agree with respondent that petitioner’s claims of overpayment for 1977 through 1982 should be disallowed because they are based on an impermissible change in method of accounting.

Volume of Cushion Gas

We must next determine whether the capitalized volumes of cushion gas were too high and a portion should be reclassified to working gas as respondent has determined.

Both petitioner and respondent employed experts to provide the Court with evidence on the appropriate cushion gas volumes in the Honor Rancho and pdr reservoirs. Petitioner obtained reports from E.A. Breitenbach of Scientific Software-Intercomp, Inc. (SSI); Fairchild; and M. Rasin Tek (Tek) of Texon, Inc. (Texon). Respondent obtained reports from Allan Spivak of Intera West Consultants (Intera) and Barry L. Evans of B.L. Evans & Associates, Inc. (Evans). Respondent also employed experts George L. Donkin of J.W. Wilson & Associates, Inc., and David A.T. Donohue of International Human Resources Development Corp., who did not calculate their own results but commented on the results of Intera and Evans. We have considered the qualifications and experience of the parties’ experts and their particular knowledge and experience in determining volumes of natural gas in reservoirs and pipelines, as well as the substance and reasoning of their reports and testimony.

The charts below summarize petitioner’s tax return position, petitioner’s affirmative claims, and both parties’ experts’ findings on the volumes of cushion gas in petitioner’s reservoirs:

Honor Rancho

Cushion Gas Volumes by Year

(All Volumes Are Stated in bcf)

P’s experts R’s experts

Year P’s tax return¹ P’s affirmative claims SSI Fairchild Texon Intera Evans

1977 12.0 19.5 16.4 12.0 17.20

1978 12.0 19.5 16.8 12.0 19.50

1979 12.0 19.5 17.6 12.0 19.50

1980 12.0 19.5 17.8 19.5 19.20

1981 12.0 19.5 17.9 19.5 18.50

1982 12.0 19.5 17.9 19.5 19.50

1983 12.0 19.5 20.0 19.5 19.50

1984 12.0 19.5 19.5 19.5 19.50

1985 19.5 19.5 19.1 19.5 19.50 12.8-18.1 14.8-17.7

Playa Del Rey

Cushion Gas Volumes by Year

(All Volumes Are Stated in bcf)

P’s experts R’s experts

Year P’s tax return

1a P’s affirmative claims SSI Fairchild Texon Intera Evans

5.544 5.5 6.91-4.65 1977 rH

5.208 5.5 6.04-4.55 1978 tH

5.360 5.5 6.48-4.57 1979 tH

5.549 5.5 6.65-4.57 1980 t-I

5.464 5.5 6.25-4.65 1981 i-I

5.518 5.5 6.65-4.45 1982 i — I

5.724 5.5 5.61-4.65 1983 t*H

5.561 5.5 5.92-4.65 1984 tH

5.472 5.5 5.61-4.65 3.8 3.8 1985 t-I

As detailed above, both petitioner’s and respondent’s experts have opined that cushion gas volumes capitalized by petitioner were not too high. Thus, for 1977 through 1985, we conclude that cushion gases (capital assets) were 12 bcf of gas at Honor Rancho and 3.1 bcf of gas at PDR. We reject respondent’s determination that some of the volume of cushion gas should be reclassified as working gas. For 1985, however, 7.5 bcf of gas at Honor Rancho will be adjusted from cushion gas back to working gas, based on our holding that a change in method of accounting occurred in 1985.

Volume of Line Pack Gas

Petitioner reported that, of the total volume of gas in the pipelines, at least 75 percent of that gas was line pack gas and that the other 25 percent was inventory gas. Petitioner now claims that the volume of line pack gas was greater than 75 percent. Only the volumes of line pack gas in the transmission system are in dispute, because the parties agree that the distribution system pipelines contained 56 mmcf of line pack gas.

Petitioner employed experts Tek of Texon; Henry H. Rachford, Jr., of Stoner Associates, Inc. (Stoner); and Harry R. Littrell of SSI to determine the appropriate volumes of line pack gas in its transmission pipeline system. Respondent employed from ZEI, Inc. (ZEI), the following experts: Donald E. Anderson; James M. Iocca; and Walter J. Woods, Jr. Petitioner’s estimates and those of the experts are summarized below:

Line Pack Volumes

(All Volumes Are Stated in mcf)

Transmission System

Petitioner’s experts

Petitioner’s Year tax return¹ Respondent’s expert — ZEI Stoner SSI² Texon

1977 1,361,923 1,340,000 1,745,000 1,870,000 to 1,911,000 1,464,808

1978 1,361,680 1,348,000 1,793,000 1,961,000 to 1,979,000 1,512,244

1979 1,361,522 1,975,000 1,516,000 1,847,000 1,853,000 1,708,511

Petitioner’s experts

Petitioner’s Respondent’s Year tax return¹ expert — ZEI Stoner SSI² Texon

1980 1,374,665 1,551,000 1,796,000 1,852,000 to 1,990,000 1,652,779

1981 1,375,895 1,403,000 1,761,000 1,857,000 to 2,042,000 1,567,480

1982 1,375,991 1,438,000 1,796,000 1,818,000 to 2,159,000 1,930,346

1983 1,376,092 1,496,000 1,793,000 1,847,000 to 2,015,000 1,671,263

1984 1,375,027 1,479,000 1,779,000 1,839,000 to 1,918,000 1,651,002

1985 1,378,390 1,399,000 1,817,000 1,933,000 to 2,065,000 1,558,150

Respondent’s expert, ZEI, calculated line pack gas by taking the volume of gas in the system at midnight on December 31 of each year and adjusting that volume by 19 percent to account for inventory gas that was also in the line. (SoCalGas and PLGS used 25 percent as the adjustment for inventory gas in the line.) ZEI then adjusted the December 31 volumes by 12 percent to account for the additional line pack gas needed in the month of December to support anticipated increases in the consumption of gas. ZEI arrived at the 12-percent figure by comparing the consumption in the month of December to the consumption for the rest of the year.

Petitioner’s experts disagreed with ZEl’s 12-percent adjustment for the following reasons: Stoner presented evidence that the December 31 supply was not higher than the demand in 1977, 1979, 1980, and 1983. SSI pointed out that the December 31 volume of line pack gas would not be higher in anticipation of large consumption on January 1 because industrial and commercial gas use (a large component of the SoCalGas customer basé) decreased on holidays. In addition, the calculations and data summaries by SSI indicate that (1) line pack gas is actually greatest in late summer months rather than at December 31 as suggested by ZEI and (2) the 12-percent increase would have been only a 3-percent increase if ZEI had compared the December 31 consumption with yearly averages instead of comparing the consumption from the whole month of December with yearly averages.

Under GAAP and for income tax accounting, assets are valued at acquisition or, as with stocks, on the last day of the fiscal year. Here, the fiscal year ended December 31. Thus, if the minimum amount of line pack gas on December 31, 1977, that was needed to provide enough pressure to deliver gas on January 1, 1978, was 1,361,923 cubic feet, then that is the value that should be reported on petitioner’s tax return, not an average yearly value. Taking the above into account, zei’s calculations (plus 12 percent) support petitioner’s tax return position but do not support petitioner’s affirmative claims to increase line pack gas. Petitioner’s experts also support petitioner’s original tax return position.

As to petitioner’s affirmative request that the volume of line pack gas be increased from its tax return position, such adjustment would be a change in accounting method requiring the Secretary’s approval because, like the reclassification of cushion gas, changes to the method of identifying line pack gas would defer recognition of income. These changes would constitute a change in method of accounting under section 446(e). Therefore, the volumes of line pack gas carmot be increased here from the amounts originally reported by petitioner.

Recoverability Defined

The parties agree that nonrecoverable cushion gas and nonrecoverable pipeline gas are properly classifiable as capital assets that are subject to depreciation. However, the parties disagree on the volumes of petitioner’s gas that were nonrecoverable and on the standard for determining those volumes of nonrecoverable gas.

Petitioner maintains that a standard of economic recoverability is appropriate, and respondent asserts that the standard should be physical recoverability. Economic recoverability means the use of technically feasible methods of recovery as long as market prices for the sale of gas recovered are greater than the direct cost of recovery. Physical recoverability means the use of current technology and associated equipment to recover the gas regardless of cost. Respondent’s alternative argument is that the amounts of nonrecoverable cushion and pipeline gas have been overstated by petitioner, even under the economic recoverability standards.

Transwestern Pipeline Co. v. United States, 225 Ct. Cl. 399, 639 F.2d 679 (1980) (per curiam), and Arkla, Inc. v. United States, 765 F.2d 487 (5th Cir. 1985), revg. 592 F. Supp. 502 (W.D. La. 1984), both indicate that economic recoverability is the proper standard. Transwestern, 639 F.2d at 685, states that recoverable line pack gas is the gas that “could reasonably have been foreseen as economically recoverable at the end of the system’s useful life.” In Arkla, 765 F.2d at 490, the court denied depreciation of recoverable cushion gas because that gas could be “commercially removed from the storage facility at the end of the facility’s useful life” so that the gas “does not represent a potential economic loss” to the utility. In addition, the Court of Appeals for the Fourth Circuit, in an oil and gas case involving depreciation of improvements to oil and gas wells, relied on the economic life of the property rather than on the physical life in calculating annual depreciation. Burnet v. Petroleum Exploration, 61 F.2d 273, 275 (4th Cir. 1932), affd. 288 U.S. 467 (1933).

Respondent suggests three reasons for adopting a physical recoverability standard. First, the physical recoverability standard eliminates uncertainty by providing a reasonably definite line between recoverable and nonrecoverable gas, whereas economic recoverability would set a moving standard requiring yearly adjustments to the nonrecoverable volume. Second, physical recoverability avoids the necessity of resolving whether current or future economic conditions should govern recoverability of gas, whereas economic recoverability necessitates forecasting economic conditions. Third, physical recoverability promotes a reasonably uniform result by relying on objective criteria rather than on experts’ estimates of changing economic conditions. Respondent has cited no authority for these positions, and we have found none.

Economic standards may change with time but so does the technology for physical recoverability. Physical recovery, as shown by the expert testimony in this case, is as much a matter of opinion as economic forecasting. We conclude that the dividing line between recoverable and nonrecoverable gas should be determined realistically; i.e., by a standard of economic feasibility. Recoverability based on physical capability rather than on economic feasibility is a fiction that cannot be justified. Thus, petitioner may depreciate the volume of gas that would cost more to extract than its value in the market.

Cushion Gas Volumes

Petitioner had no direct methods to measure the volumes of gas that would be nonrecoverable if operations were abandoned. SoCalGas estimated that nonrecoverable cushion gas volumes were 8.5 bcf at Honor Rancho and 1.2 bcf at pdr.

Respondent presented expert evidence of an overstatement of nonrecoverable cushion gas at PDR; however, respondent did not determine this overstatement in the deficiency notice and conceded on brief that the PDR nonrecoverable cushion gas was not challenged.

In the notice of deficiency, respondent determined that the nonrecoverable cushion gas volume at Honor Rancho was 4.7 bcf under a physical recoverability standard, and respondent now asserts, in the alternative, that the volume is 5.2 bcf under an economic recoverability standard.

The expert testimony on nonrecoverable cushion gas at Honor Rancho is summarized as follows:

Honor Rancho

Breakdown of Cushion Gas Into the

Nonrecoverable Portions

(All Volumes Are Stated in bcf)

Respondent’s experts Petitioner’s experts

Tax Year return Intera Evans ■ SSI Fairchild Texon

6.94 1977

7.40 1978

7.40 1979

7.40 1980

7.40 1981

7.40 1982

7.40 ,1983

7.40 1984 | ¡

7.40 1985 cq CD

In calculating the above volumes for the Honor Rancho reservoir, the experts from SSI and Intera used computerized simulations of the reservoir, Fairchild used results from a reservoir model he developed when Honor Rancho was first purchased and used pressure - and volume analyses (P/Z curves), Texon used only P/Z curves, and Evans used the computer simulation generated by Intera. Differences among the experts resulted from differing minimum deliverability rates and differing economic factors. Petitioner and its experts used a minimum deliverability of 100 mmcf per day and did not consider the rental of extraction equipment in their economic analyses, whereas respondent’s experts performed analyses using three different deliverabilities of 100 mmcf, 25 mmcf, and 0 mmcf per day and included the rental of compressors and other equipment for extracting gas.

The experts agreed that computer simulations of reservoirs tend to give a more accurate analysis of the actual volume of cushion gas in a reservoir than the analysis derived from P/Z curves; thus, we give more weight to the conclusions derived from the SSI and Intera computer simulation. To test SSI’s computerized reservoir model and Intera’s revised reservoir model, the experts used the models to simulate past performance of the reservoirs and compared those simulations with the reservoirs’ actual performance history. This process is commonly called “history matching”. We conclude that the SSI simulation had the most accurate history match.

SSI applied an economic recoverability formula to the results of its computer simulation. The formula deflated to 1975 dollars the 1991 costs of depleting petitioner’s East Wittier reservoir (which was in the process of abandonment in 1992). SSI considered the lease costs and property tax costs in its calculations. Next, SSI calculated the daily cost of operating the reservoirs using those 1975 values. Daily cost divided by the actual 1975 intrastate gas price yielded the volume of production (the economic limit) for those daily costs. The computer simulation indicated that 8.4 bcf of gas would be left in the Honor Rancho reservoir when the economic limit was reached.

Respondent argues that the 100 psi wellhead pressure used by SSI in its computer simulation was too high. The SSI rebuttal report included a graph that indicates that using wellhead pressure at Honor Rancho of 50 psi rather than 100 psi would increase the economic recoverability by only .35 bcf, and SSI expressed doubt that the 50 psi flowing pressure was feasible. Other experts indicated that the 50 psi wellhead pressure was not feasible without compression equipment.

Evans used the Intera computer simulation but applied an economic recoverability formula to the data. In performing his economic analysis, Evans used sales prices and costs for 1985 rather than the prices and costs available to SoCalGas at the acquisition dates of the reservoirs. The use of 1985 data is incorrect because the volume of nonrecoverable cushion gas is determined at the date of acquisition, not later.

Determining the recoverable portion of cushion gas is essentially the same as determining the salvageable portion of a capital asset. The regulations discussing salvage value state:

(c) Salvage. (1) Salvage value is the amount (determined at the time of acquisition) which is estimated will be realizable upon sale or other disposition of an asset when it is no longer useful in the taxpayer’s trade or business or in the production of his income and is to be retired from service by the taxpayer. Salvage value shall not he changed at any time after the determination made at the time of acquisition merely because of changes in price levels. * * * [Sec. 1.167(a)-l(c)(l), Income Tax Regs.]

Evans also erred in using a value for property taxes of $5,400 a month (IVz percent of gross revenue) because he did not provide support for that percentage. Evidence established that property taxes paid by SoCalGas were at least $600,000 a year, or $50,000 a month.

Finally, Evans did not address whether CPUC would accept the costs of recovering gas at the lower pressures indicated in his expert report. Because petitioner’s gas sales are regulated by CPUC, this is a relevant consideration. CPUC had accepted, for purposes of establishing rates, petitioner’s depreciation deductions.

Fairchild did not provide data to support his conclusion that the economic limit for recovery at Honor Rancho should be 8.5 bcf. Texon and Intera used a physical recoverability standard.

Of the experts evaluating cushion gas, SSI provided the most reliable analysis of the volume at Honor Rancho, ssi evaluated nonrecoverable cushion gas at the date of acquisition and considered actual lease costs and property taxes. SSI did not evaluate the cost of renting compressors because such costs were not incurred in the abandonment of petitioner’s other reservoirs. We conclude that the most accurate determination of nonrecoverable cushion gas at Honor Rancho is as calculated by SSI, 8.4 bcf.

Line Pack Gas Volumes

Petitioner depreciated line pack gas as if all gas in that category was nonrecoverable at abandonment of pipeline operations. Respondent disallowed depreciation deductions on line pack gas, claiming that none of the line pack gas was nonrecoverable. Respondent based this conclusion on a physical recoverability standard but also presented expert reports addressing economic recovery of line pack gas.

ZEI determined the nonrecoverable portions of gas separately for the transmission system and the distribution system, whereas petitioner’s experts opined on the transmission system only. ZEI determined that 50 percent of the 56 mmcf of line pack gas in the distribution system was recoverable- and nondepreciable. Petitioner has not challenged this conclusion. Based on the only evidence in the record, we conclude that the correct amount is 50 percent of the total.

The positions of the parties and of their experts on the amount of line pack gas in petitioner’s transmission system pipelines that was economically nonrecoverable are summarized as follows:

SoCalGas PLGS

Petitioner’s tax return 100% 100%

Notice of deficiency -0--0-

Petitioner’s experts:

Stoner 39% 33%

SSI 13% to 39% 13% to 39%

Texon 90% 90%

Respondent’s expert:

ZEI 7% to 19% 7% to 19%

Petitioner’s experts assumed that the gas would flow freely from the reservoir down to a minimum operating pressure of 200 psi in determining nonrecoverable portions of line pack gas. Any remaining gas would be vented into the air or otherwise disposed of for safety reasons. Because price and cost information was not available for the year in which the transmission system was acquired, all of the experts used current costs in their economic analyses.

Stoner proposed that 39 percent of line pack gas in SoCalGas’ system and 33 percent of line pack gas in PLGS’ system were nonrecoverable. Stoner determined the volumes of line pack gas using a commercially available software package, GAS Steady State, which the industry, CPUC, and FERC also use. Stoner then imposed economic constraints that the gas would flow freely from the pipelines until pressure dropped below 200 psi. Stoner stated that expelling additional gas with other methods of recovery, such as injecting nitrogen and liquids into the pipelines, would not be economically feasible because nitrogen was too costly and liquids would have trapped gas in the pipeline system requiring additional equipment to separate the gas from the liquid. The report did not provide a calculation for arriving at this conclusion.

SSI proposed that 39 percent of line pack gas was nonrecoverable. SSI used a gas network simulation computer program, TGNET, to simulate the functioning of SoCalGas and PLGS transmission systems and to determine the volumes of line pack gas in those systems. The program simulated pipeline conditions in trimesters:

Winter. November through February

Early summer . March through June

Late summer. July through October

The simulations used a minimum pressure range of 200 to 250 psig, but SSI warned that these calculated volumes may be too low to ensure safe and reliable operation of the pipeline.

For determining the nonrecoverable portion of line pack gas, SSI rejected the use of portable compressors and displacement of gas by air, water, or nitrogen because these methods were either technically infeasible or too costly; however, the SSI report did not provide us with the calculations for arriving at this conclusion. Instead, SSI relied on flowing gas to the existing customer base of SoCalGas until the pressure in the transmission pipelines reached 200 psig. This analysis resulted in approximately 755 mmcf of gas remaining in the transmission network, or 39 percent of total line pack gas as calculated under TGNET.

Although SSI questioned whether it was economically feasible to modify pressure regulators in the transmission system to operate the transmission system at 60 psig, SSI performed an alternate calculation at that pressure. Reducing the pressure to 60 psig resulted in approximately 254 mmcf of nonrecoverable line pack gas, which is 13 percent of SSI’s line pack gas estimate or 19 percent of petitioner’s reported line pack gas.

ZEI proposed that between 7 and 19 percent of line pack gas was nonrecoverable. ZEI analyzed line pack gas recovery under four options; under each option, the gas was sold to different potential purchasers of gas. The ZEI report provided details of the cost assumptions and calculations used in arriving at economic recoverability. ZEl’s estimates, using portable compressors to withdraw gas down to a system pressure of 50 psia or 35.3 psig, yielded recoveries of 81 to 93 percent of the line pack gas over a maximum 49-day period.

Of the experts evaluating line pack gas, ZEI provided the most reliable evaluation of line pack gas recoverability. The ZEI report provided a logical breakdown of the costs that were considered in determining an economic limit for recovering gas. Thus, we conclude that the economically nonrecoverable volume of line pack gas in the transmission system was 19 percent. The 19-percent value is also supported by SSI, because, under reduced pressures (60 psig), SSI determined that 19 percent of petitioner’s reported volumes would be nonrecoverable.

The parties agree that the nonrecoverable portion of petitioner’s line pack gas is subject to depreciation and may be taken into account for purposes of the ITC. Although petitioner’s arguments in its opening brief on the eligibility of recoverable gas for the ITC are unclear, petitioner states in its reply brief that “the Gas Company has neither depreciated nor sought to depreciate recoverable cushion gas or line pack.” Nondepreciable capital assets fail to meet the definition of “section 38 property” for which the ITC is allowed. Secs. 38, 46; see also Arkla, Inc. v. United States, 27 Fed. Cl. 226 (1992).

Pursuant to the parties’ stipulations, the net operating loss adjustment will be recomputed based on the findings in this opinion.

Decision will be entered under Rule 155.

1 Although these volumes of cushion gas were not expressly listed on petitioner’s Federal income tax returns, the parties have agreed that these are the volumes of gas used by petitioner in determining taxable income.

Although these volumes of cushion gas were not expressly listed on petitioner s Federal income tax returns, the parties have agreed that these are the volumes of gas used by petitioner in determining taxable income.

1 Although these volumes of line pack gas were not expressly listed on petitioner’s Federal income tax returns, the parties have agreed that these are the volumes of gas used by petitioner in determining taxable income.

2 SSI’s overall conclusion was that the transmission system contained approximately 1,900,000 mcf of total line pack gas for each of the years 1977 through 1985.