SCOTT

Edwin W. King, Jr., for petitioner. Gary F. Walker and Kim A. Palmerino, for respondent.

Petitioner, ADC, manufactured combat boots. Its Puerto Rican subsidiary, ADPR, manufactured the uppers for such boots and transferred them to petitioner at a price determined by petitioner. On its return for its fiscal year 1986, ADPR elected the cost sharing method as prescribed under sec. 936(h)(5)(C), I.R.C. That return was due on June 15, 1987, with an extension duly obtained, and was placed in the mail that day with a metered stamp and proper postage for certified mail with return receipt attached. Although other returns mailed the same day by the same individuals in the same box with metered postage were received within the normal time for the receipt of mail, the two items mailed to the Philadelphia Service Center, including ADPR's return for its fiscal year 1986, were stamped received by that service center well past the normal time for receipt of items mailed June 15, 1987. Respondent was unable to locate the envelope in which ADPR's return was mailed. ADPR again elected the cost sharing method on its 1987 return.

Respondent determined a transfer price from ADPR to ADC on the basis that ADPR had not elected the cost sharing provisions of sec. 936(h)(5)(C). Petitioner determined the transfer prices on the basis that a proper cost sharing election had been made for each year here in issue, and therefore a proper method for determining transfer prices under sec. 936, I.R.C., was to apply an appropriate gross profit markup computed under the methods provided in sec. 482, I.R.C., to all costs of producing the product by ADPR, including the materials cost. ADPR determined that no cost sharing payment to ADC for product area research was required for each of the fiscal years here in issue.

Held: (1) ADPR's return for its fiscal year 1986 was timely filed; therefore it properly elected cost sharing for that year and, accepting respondent's concession, cost sharing is applicable to the prior fiscal year 1985. Cost sharing was properly elected for ADPR's fiscal year 1987; (2) a cost sharing payment for product area research was required to be made by ADPR to ADC for each of the fiscal years here in issue; (3) although a cost sharing payment for ADC's costs for product area research was required from ADPR, the failure to make such payment was not willful and did not revoke its cost sharing election; (4) under the cost sharing method, the sales price of the component product is determined under the appropriate sec. 482 method. Only if none of the three methods preferred in the sec. 482 regulations is appropriate is a fourth method such as “profit splitting” to be used. The evidence here is sufficient to determine ADPR's transfer prices under the cost-plus method as described under the sec. 482 regulations; (5) respondent's determination as to the appropriate allowance for location savings for ADPR is upheld; and (6) the amount ADC paid to ADPR in excess of the equivalent of an amount computed on an arm's-length basis is in effect a loan by ADC to ADPR on which interest should be imputed to ADC.

SCOTT, Judge:

Respondent determined deficiencies in petitioner's Federal income taxes for its fiscal years ending the last week of September 1981, 1985, 1986, and 1987 in the amounts as follows:

+-----------------------+ ¦Fiscal Year¦Deficiency ¦ +-----------+-----------¦ ¦1981 ¦$ 14,688.60¦ +-----------+-----------¦ ¦1985 ¦691,456.14 ¦ +-----------+-----------¦ ¦1986 ¦332,395.00 ¦ +-----------+-----------¦ ¦1987 ¦503,898.00 ¦ +-----------------------+

The issues for decision are: (1) Whether Altama Delta Puerto Rico Corp. (ADPR), a subsidiary of petitioner (ADC), filed a timely Federal income tax return for its fiscal year ending September 27, 1986, claiming an election for the cost sharing method described in section 936(h)(5)(C)(i);

(2) whether ADPR was required to make a payment to ADC for its share of the cost of product area research under section 936(h)(5)(C)(i)(I) for each of the fiscal years here in issue, and if so, the amount of such payments; (3) whether ADPR's failure to make a timely cost sharing payment to ADC for its fiscal years here in issue was due in whole or in part to willful neglect, causing its cost sharing election, if made, to be revoked under section 936(h)(5)(C)(i)(III); (4) what is the proper amount of the transfer price of products transferred from ADPR to ADC and the appropriate section 482 method of determining that price; and (5) the amount of location savings to which ADPR is entitled for each of the fiscal years here in issue; and (6) whether, for petitioner's fiscal years 1985, 1986, and 1987, respondent properly allocated interest income to petitioner from ADPR under the provisions of section 482, and, if so, the proper amounts to be allocated.

FINDINGS OF FACT

Some of the facts have been stipulated and are found accordingly.

Petitioner, a closely held Georgia corporation, had its principal place of business in Atlanta, Georgia, at the time of the filing of its petition in this case. During the fiscal years here at issue, ADC's principal place of business was in Darien, Georgia.

For its fiscal years ending October 3, 1981 (fiscal year 1981), September 28, 1985 (fiscal year 1985), September 27, 1986 (fiscal year 1986), and October 3, 1987 (fiscal year 1987), petitioner filed its Federal income tax returns on an accrual method of accounting with the Internal Revenue Service Center in Atlanta, Georgia.

Petitioner manufactures and sells military boots to the U.S. Department of Defense (DOD) pursuant to Government contract specifications. The DOD dealt with petitioner through its agent at the Defense Personnel Support Center (DPSC) in Philadelphia, Pennsylvania. The DPSC is a division of the Defense Logistics Agency, which is headquartered at the Pentagon in Washington, D.C.

Mr. Aage Jensen (Mr. Jensen) served as petitioner's chief executive officer (CEO) from its inception through the years at issue. Mr. Bill Fleming (Mr. Fleming) served as petitioner's chief financial officer (CFO) from 1970 through 1988. ADC had between 350 and 400 employees during the years at issue.

Altama Delta Puerto Rico Corp., a wholly owned subsidiary of petitioner, was incorporated as a Delaware corporation on March 26, 1982. ADPR was engaged in the business of manufacturing the upper part of the combat boot, which is a component of the boot referred to as “uppers”. ADPR shipped the uppers it manufactured to petitioner for further processing into the final boot product. From its inception through the years here in issue, ADPR's manufacturing plant and principal office were located in Salinas, in the Commonwealth of Puerto Rico.

Mr. Jensen served as ADPR's CEO, and ADC's other officers served as officers for ADPR. The financial statements of ADC were prepared on a consolidated basis with ADPR and a wholly owned foreign sales corporation, Altama Delta Virgin Islands (ADVI). However, each of the corporations filed separate Federal income tax returns for the years here in issue. Both ADC and ADPR are members of the same affiliated group for purposes of sections 936 and 482.

The manager of the ADPR plant in Salinas was Mr. Jose Alvarado Rodriguez (Mr. Alvarado). Mr. Alvarado was responsible for all phases of operation at the Salinas plant.

ADPR leased facilities for its leather and stitching operations from the Puerto Rican Industrial Development Co., an agency of the government of the Commonwealth of Puerto Rico. The yearly rental for the years here in issue was approximately $20,000.

Approximately 80 people were employed by ADPR each year. ADPR experienced approximately a 2-percent employee turnover rate per year during the period 1985 through 1987.

ADPR filed its Federal income tax returns on an accrual method of accounting with the Internal Revenue Service Center in Philadelphia, Pennsylvania, for its fiscal years 1985, 1986, and 1987. ADPR elected possession corporation status under section 936(a) and (e) on its fiscal year 1985 return, which allowed ADPR to claim a tax credit under sections 27(b) and 936(a). ADPR also filed forms 26, Foreign Corporation Reports, with the Department of State for the Commonwealth of Puerto Rico for the calendar years 1982 through 1987.

During the years at issue, the DOD contacted numerous footwear manufacturers to invite them to bid on solicitations for combat boots. The DOD's solicitations for boots contained complete specifications detailing exactly how the boots were to be manufactured and what materials should be utilized by manufacturers. Each solicitation was for a certain delivery destination. The sizes required were to be specified by persons at the delivery destination. All sizes were not required at all destinations. Included in the specifications were patterns for soles and specifications concerning the sole molding process.

Government contracts for boots during the years here in issue contained “set asides” which favored contractors with manufacturing plants in “labor surplus areas” and manufacturers that qualified as small businesses (businesses with fewer than 500 employees). Companies that were not located in labor surplus areas were not allowed to bid for certain contracts for combat boots, because a certain portion of the manufacturing process had to be done in labor surplus areas. During the years here in issue solicitations of bids by the DOD reserved half of the total contracts for companies that incurred most of their costs in high unemployment areas. During the years at issue, both petitioner and ADPR qualified as small business concerns and were located in labor surplus areas which enabled petitioner to qualify for labor surplus preferences.

ADPR manufactured only the upper portion of the boot. Petitioner manufactured uppers at its Darien plant, in addition to performing the lasting, vulcanizing, and finishing on the boots which contained the uppers manufactured by ADPR, as well as the uppers it manufactured in Darien. ADPR sold uppers only to ADC.

Before 1986 petitioner manufactured and sold the standard combat boot (the old boot). After 1985 petitioner manufactured and sold a black leather “DMS” combat boot (the new boot), which was designed and built according to the DOD's new specifications. Petitioner also manufactured the hot weather “jungle” combat boot (the jungle boot) during the years at issue. Uppers for the old boot and for the jungle boot were manufactured at the Darien facility, as well as at ADPR's plant in Salinas, but no uppers for the new boot were manufactured at ADC's Darien facility.

The following schedule shows the number of boots produced in Darien during the years at issue separated into the contracts both stitched and finished in Darien (ADC) and those stitched in Salinas (ADPR) and finished in Darien:

+---------------------------------------------------------------+ ¦Altama Delta Corporation ¦ +---------------------------------------------------------------¦ ¦Costs and Profits by Contract for the 1985 Through 1987 Period ¦ +---------------------------------------------------------------¦ ¦Contracts Stitched in Darien and Finished in Darien ¦ +---------------------------------------------------------------+

+-------+ ¦ ¦ +-------+

Overhead/ Total Operating Boot Sales Pairs Labor cost SGA cost materials Total cost Profit type costs Tropical $ 311,565 $1,873,977 $2,945,296 $4,038,068 $ $233,764 9,091,105 8,857,341 Old boot 7,324,896 268,404 1,146,058 1,801,239 4,132,183 7,079,481 245,415 Civilian 1,052,226 44,563 253,088 397,773 602,328 1,253,189 (200,963) Other 1,276,344 979,047 68,196 107,183 985,465 1,160,844 115,500 sales Total 18,744,571 1,603,579 3,341,319 5,251,491 9,758,045 18,350,855 393,716 [sic]

+---------------------------------------------------------------+ ¦Altama Delta Corporation ¦ +---------------------------------------------------------------¦ ¦Costs and Profits by Contract for the 1985 Through 1987 Period ¦ +---------------------------------------------------------------¦ ¦Contracts Stitched in Salinas and Finished in Darien ¦ +---------------------------------------------------------------¦ ¦ ¦ +---------------------------------------------------------------+

+------------------------------------------------------------------------------------+ ¦ ¦ ¦ ¦ ¦Overhead/ ¦Total ¦ ¦Operating ¦ +--------+-----------+-------+----------+----------+----------+-----------+----------¦ ¦Boot ¦Sales ¦Pairs ¦Labor cost¦SGA cost ¦materials ¦Total cost ¦Profit ¦ ¦type ¦ ¦ ¦ ¦ ¦costs ¦ ¦ ¦ +--------+-----------+-------+----------+----------+----------+-----------+----------¦ ¦ ¦ ¦ ¦ ¦ ¦ ¦ ¦ ¦ +--------+-----------+-------+----------+----------+----------+-----------+----------¦ ¦New boot¦$21,715,413¦452,812¦$1,508,255¦$2,370,497¦$3,969,421¦$19,467,290¦$2,248,123¦ +--------+-----------+-------+----------+----------+----------+-----------+----------¦ ¦Tropical¦6,152,962 ¦214,767¦624,242 ¦981,110 ¦1,124,650 ¦5,838,861 ¦314,101 ¦ +--------+-----------+-------+----------+----------+----------+-----------+----------¦ ¦Old boot¦2,439,061 ¦82,845 ¦240,798 ¦378,458 ¦366,712 ¦2,220,728 ¦218,333 ¦ +--------+-----------+-------+----------+----------+----------+-----------+----------¦ ¦Civilian¦1,275,070 ¦46,378 ¦144,321 ¦226,827 ¦319,317 ¦1,614,281 ¦(339,211) ¦ +--------+-----------+-------+----------+----------+----------+-----------+----------¦ ¦Total ¦31,582,506 ¦796,802¦2,517,616 ¦3,956,892 ¦5,780,099 ¦29,141,160 ¦2,441,346 ¦ ¦ ¦ ¦ ¦ ¦ ¦[sic] ¦ ¦ ¦ +------------------------------------------------------------------------------------+

ADC's bids for Government contracts were prepared primarily by Mr. Jensen and Mr. Fleming. Petitioner's bidding activities required Mr. Jensen and Mr. Fleming to gather the information needed regarding leather prices and other supply costs in order to arrive at ADC's bidding price. Mr. Jensen determined how many uppers would be produced in each location. Mr. Jensen would gather price information from both the Darien and Puerto Rican plants in order to determine the costs that would be involved in each step of the manufacturing process for that particular bid. Mr. Jensen would use information of costs incurred under prior contracts to determine the price necessary to bid on the contract being solicited and to determine whether ADC could bid at a price within the narrow price range acceptable to the Government.

Bids made by petitioner were based on known costing data with a targeted profit margin. Accordingly, the profit on a specific contract could be predicted, except to the extent of variances from standard costing.

During the years at issue, petitioner was awarded at least a portion of seven of the eight contract solicitations offered by the DOD. The DOD's contracts typically ran from approximately 6 to 8 months during the years in issue, but could run for longer than 1 year.

After it was awarded a contract by the Government, petitioner's officers would contact its suppliers to determine the exact cost of the materials needed to meet the contract. The purchase orders for each material used in the boot would reflect the agreed-upon price. Materials were the most costly component of the boot, and the most expensive portion of the boot was the leather upper.

Petitioner's standard cost ranged from $19.29 to $39.63 per pair of boots depending on the contract and type of combat boot. Combat boots that did not meet the Government's specifications were sold at a loss in the civilian market.

An essential material in the manufacture of certain combat boots is the leather used to form the upper. The leather is purchased as a “side”, which is the skin from one-half of a cow divided from head to tail. Although the average size upper pair of boots actually consists of approximately 2-1/2 square feet of leather, the uppers for a pair of boots are cut from approximately 4-1/2 square feet of leather.

The leather was received by ADPR on pallets. Some pallets were shipped to ADPR directly from the tannery, but most of the pallets were shipped to ADPR through the Darien plant. Each pallet contained approximately 3,900 square feet of leather. When leather sides for cutting and stitching at ADPR were shipped first to Darien, they would be scanned by a machine in Darien to verify the accuracy of the square footage stamped on each side. There was no similar machine available in Salinas. When the leather sides were being measured in Darien, if a side was not of satisfactory quantity, it would be rejected and returned by personnel at Darien to the tannery.

After being inspected in Darien, the leather sides were shipped to ADPR in Salinas. Inspectors in Salinas inspected the leather shipped to ADPR through ADC and the leather shipped directly to ADPR in Salinas from the tanneries for imperfections. The imperfections would be marked with white chalk and, if the imperfections were such as to cause a leather side to be unusable, that side would be returned to the tannery for credit. Defects which were marked by inspectors in Salinas could include scars, brands, scratches, slaughter cuts, bites, wrinkles, or folds in the hide. Some defects were allowable on certain parts of the boot but not on others. For example, small scratches or wrinkles would be allowed on the backstay or on a portion of the quarters (the back half of the side, including the leg and one or more ribs). The white chalk marks placed on the hides by the inspectors were for assisting the cutters, but the decision as to how to utilize the imperfect portions of a side was made by the cutter. Leather used for gussets (the tongue of the boot) is a different grade and type of leather and was not inspected prior to the cutting step.

Cutters manually placed metal dies on the hide to stamp out pieces of the boots using a hydraulic press on areas not marked for defects. Each whole and half shoe size in each width had a different set of five dies. The dies for the quarters were reversible to form the left and right sides of the boot. Military specifications required that the vamp (which forms the toe and front sides of the boot) be cut from the “bend” (that area of the side that has the most durable fiber) portion of the hide, and the remaining pieces were cut from the usable leather outside the bend area.

Cutting is considered the key process in manufacturing uppers because of its effect on leather usage. The price for each square foot of leather during the years at issue was approximately $2.50. A hide contains approximately 20 to 26 square feet of leather. Approximately 20 percent of the total leather was lost during the inspection and cutting steps because the unusable leather was scrapped. Approximately 3 to 5 percent of the leather was lost through the remaining production steps. Some of the most experienced and skilled workers in the Salinas plant were cutters, many of whom worked for ADPR from its inception.

After cutting, the leather pieces were sent to the prefitting department. There, the leather pieces were stamped with the contract and lot numbers. Each vamp was manually stretched to smooth it and was inspected for defects. Each piece was marked with lines using templates in a hand-operated machine for the stitchers to follow when the pieces were sewn together. The prefitting department also performed a splitting operation where the edges of the leather pieces were tapered.

The inspected, cut, and marked pieces were then sewn in the stitching department. After stitching, the uppers were sent through a final inspection where inspectors looked for any boots that should have been rejected. Reasons for rejecting a completed upper included visible scratches or scars on the upper, stitching over lot numbers stamped on the outside of the boot, or uneven stitching. After final inspection, the uppers were packed into boxes marked by size of the boots and shipped to ADC in Darien by boat.

The lasting and vulcanizing processes for the boots were performed in Darien. In the lasting step, the stitched uppers were shaped by fitting the upper around a wooden last, a form which is shaped like a human foot and over which a shoe is shaped.

After lasting, the boots were vulcanized, a process in which the rubber sole is attached to the leather upper. In the vulcanizing process, a coat of rubber cement is first painted over the bottom of the upper. This coat is then dried and the process repeated. In the second step, the upper is fitted over a foot-shaped holder and a frame is locked around the upper. A piece of uncured rubber is inserted under the upper portion and the mold is closed. The rubber is then heated under pressure for approximately 12 to 15 minutes to bond the sole to the boot upper. A vulcanizing press is used to affix the rubber sole to the boot. A sole plate or mold sits on top of the vulcanizing press which forms the outsole of the boot.

After the soles were affixed to the uppers, the boot would be sent to the finishing department in Darien, where it would be trimmed of excess thread and buffed. The heel of the boot would be glued, pressed, and nailed to the sole. The edges of the sole were trimmed to remove excess rubber. The boots were then inspected for compliance with military specifications. Employees at the Darien plant completed the final finishing step by gluing the interior heel pad to the insole, coloring the boots to a uniform color, inserting the laces and insoles, and packing the boots into shipping cartons.

The quality of uppers made by ADPR was better than required by specifications. The uppers made by ADPR were superior to the uppers made by ADC.

ADC performed various management functions for ADPR. ADC handled the bookkeeping and accounting functions for both the Darien and the Salinas plants. Bidding was done by ADC. ADC also did purchasing for ADPR and arranged for insurance for products that were shipped from Puerto Rico to Darien. The expenses of the services performed by employees of ADC for ADPR were calculated by petitioner's accounting department, and ADC was paid by ADPR by check.

ADPR paid a management services fee to ADC for accounting, data processing, payroll, scheduling, purchasing services, and other services rendered by ADC to ADPR. For these services, ADPR paid ADC $85,000 in 1985, $109,000 in 1986, and $127,000 in 1987. This amount was calculated by ADC's CFO, Mr. Fleming, based on time sheets submitted by ADC personnel documenting time spent by the various individuals on services for ADPR. ADPR paid vendors by checks drawn on its bank account in Puerto Rico for leather and other raw materials purchased for the manufacture of uppers whether the leather was delivered first to ADC or directly to ADPR.

At the end of each of its fiscal years 1985, 1986, and 1987, the bank balance of ADPR reflected the following cash or cash equivalents:

+------------------------------------+ ¦Fiscal year¦Cash or cash equivalents¦ +-----------+------------------------¦ ¦1985 ¦$ 481,290 ¦ +-----------+------------------------¦ ¦1986 ¦690,751 ¦ +-----------+------------------------¦ ¦1987 ¦2,126,566 ¦ +------------------------------------+

ADPR paid the shipping costs of raw materials to Salinas from Darien, and ADC paid the cost of shipping uppers from Salinas to Darien.

Officers and employees of ADPR set up the production schedule followed by ADPR from leather inspection to shipping of the completed uppers. ADPR's employees determined the machine layout in ADPR's facility, developed a system of quality control for ADPR, and saw that ADPR was in compliance with labor and environmental standards. The hiring, training, timekeeping, and scheduling of employees at the Salinas factory were performed by ADPR employees in Salinas. Information regarding hours worked was transmitted from Salinas to Darien for the use of employees of ADC in Darien in preparing ADPR's payroll and keeping its accounting records.

ADPR maintained an inventory of leather and other raw materials for which it had paid. ADPR maintained a 2-week supply of leather, which amounted to approximately 40,000 square feet of leather. A representative of the DOD came to ADPR's plant in Salinas every Thursday to inspect the uppers that had been manufactured there to determine whether they met the DOD standards. The uppers could not be shipped to ADC in Darien until the DOD inspector signed an approval of the quality of the uppers to be shipped.

Mr. Fleming calculated the price to be paid by ADC to ADPR for the uppers by adding a profit factor to ADPR's cost of manufacturing the uppers. He calculated the profit factor by dividing the total profit on each contract for which ADPR manufactured the uppers between ADC and ADPR based on the portion of manufacturing costs for that contract incurred by each entity. When petitioner and Mr. Alvarado agreed on a purchase price for the uppers, it was understood that the price would more than cover the expected cost of production in Puerto Rico so that ADPR would make a profit.

Starting in 1985, petitioner began to manufacture the new boot. Prior to beginning to manufacture the new boot, petitioner had manufactured the old boot and the jungle boot. Petitioner had owned its own plates or molds for all whole sizes and leased or rented from Ro-Search, a wholly owned subsidiary of Wellco, the half-size molds. In late 1984 or early 1985 when the solicitations for contracts for the new boot were made, there was a substantial change in the specifications from those of the old boot. The design for the new boot was different from the design of the old boot in both the material and the sole design.

Petitioner determined that the necessary new molds for the new boot would require a very substantial outlay. Under date of September 7, 1984, Ro-Search addressed a letter to each ADC, McRae Industries (McRae), Wellco Enterprises (Wellco), and Belleville Shoe Manufacturing (Belleville). The letters proposed that each of the military licensees of Ro-Search enter into an agreement with Ro-Search to pool their old boot molds through Ro-Search. The four boot manufacturers and Ro-Search agreed that Ro-Search would take the existing whole-size molds of all four manufacturers and convert them to the new specifications (the Ro-Search agreement). These molds then became part of the basic inventory of molds that was shared by the four licensees. The half-size molds were to be made by Ro-Search.

Since ADC and the other companies were still making old boots and jungle boots, the agreement stated in part: “We will take your existing whole size molds and convert them, as needed”. Ro-Search designed, converted, or manufactured all of the molds for the new boot, both whole and half sizes. It leased or rented these molds to each of the four boot manufacturers. Petitioner entered into a license agreement with Ro-Search and paid royalties for the use of the molds and research technology of Ro-Search during the years at issue.

Ro-Search owned only one set of the molds but arranged a schedule for delivery of these molds at certain intervals to each manufacturer to meet that manufacturer's delivery requirements.

Ro-Search became responsible for the major maintenance of both the whole and half-size molds. Ro-Search charged petitioner a 40-cent-per-pair technical assistance/mold leasing fee for the molds used in the manufacture of the new boots. Ro-Search agreed to provide a basic inventory of molds sufficient to satisfy the daily combat boot production of petitioner at a minimum production rate of 1,600 pairs of boots each month. In the event total production of military boots increased, Ro-Search agreed to allocate additional molds as necessary. Ro-Search agreed that it would not lease the combat boot molds to any licensees other than the four which had participated in supplying the old molds to Ro-Search and participated in the Ro-Search agreement. The fees paid by the original licensees entitled them to maintenance and replacement of broken molds, new molds when needed, and training for employees using the adapted molds.

Molds could have been manufactured and used by any other company willing to invest the capital necessary to either design and build or purchase the molds. Because the DOD publicly issued detailed specifications for the new boot, the molds could have been developed by any machinery company. Each mold would have cost approximately $8,000. It would have taken from 84 to 109 molds to make up one set of molds for performing a DOD contract. No companies other than the four which had agreements with Ro-Search submitted a bid to the DOD for military boot contracts during the years here at issue.

Solicitations from the DOD for military boots were low in the years immediately prior to those here in issue. During the years at issue, the DOD made solicitations for a much larger number of boots than in prior years. This unusually high demand was due to a number of factors. First, an error in the DOD's newly installed computer inventory system underestimated the number of combat boots in the DOD's inventory, thus generating a larger number of boot contracts. Second, the introduction of the new boot increased quantities and profitability over the old boot. Third, there were a number of barriers to new manufacturers' entering the combat boot market. In considering bids on the contracts a manufacturer which was a small business concern and had a plant located in a labor surplus area was given certain preferences. Manufacturing of the new combat boot would have been difficult for companies that did not lease the sole mold plates from Ro-Search.

The four companies that bid on the contracts for military boots usually bid for a lower number of boots than they could produce. Because of the few bidders for the manufacture of combat boots, a bidder that submitted a price which was not the lowest bid often would be awarded a significant share of the total combat boot award because the full quantity needed by the DOD was not covered by the bid of the lowest bidder. These factors resulted in high profit margins on combat boots for all four combat boot manufacturers during the years here in issue, although there was an overall decline in the profits of the U.S. footwear industry.

ADC's gross profit margins for its fiscal years here at issue were as follows (in millions of dollars; numbers may not add due to rounding):

+------------------------------------------------+ ¦ ¦ ¦Cost of ¦ ¦Gross profit¦ +-----+-----+----------+------------+------------¦ ¦Year ¦Sales¦goods sold¦Gross profit¦margin ¦ +-----+-----+----------+------------+------------¦ ¦1987 ¦$17.8¦$14.8 ¦$3.0 ¦20.7% ¦ +-----+-----+----------+------------+------------¦ ¦1986 ¦15.6 ¦13.3 ¦2.3 ¦17.6 ¦ +-----+-----+----------+------------+------------¦ ¦1985 ¦16.9 ¦13.6 ¦3.4 ¦25.0 ¦ +-----+-----+----------+------------+------------¦ ¦Total¦50.3 ¦41.6 ¦8.7 ¦21.0 ¦ +------------------------------------------------+

ADPR's gross profit margins for its fiscal years 1985 through 1987 using the price at which its products were transferred to ADC were as follows (in millions of dollars; numbers may not add due to rounding):

+------------------------------------------------+ ¦ ¦ ¦Cost of ¦ ¦Gross profit¦ +-----+-----+----------+------------+------------¦ ¦Year ¦Sales¦goods sold¦Gross profit¦margin ¦ +-----+-----+----------+------------+------------¦ ¦1987 ¦$ 7.9¦$ 6.6 ¦$1.3 ¦20.2% ¦ +-----+-----+----------+------------+------------¦ ¦1986 ¦5.7 ¦4.7 ¦1.0 ¦21.2 ¦ +-----+-----+----------+------------+------------¦ ¦1985 ¦3.9 ¦2.9 ¦.9 ¦31.6 ¦ +-----+-----+----------+------------+------------¦ ¦Total¦17.6 ¦14.3 ¦3.2 ¦22.9 ¦ +------------------------------------------------+

The following schedule shows the gross profit margins for McRae and Wellco for each of the years 1985 to 1987 (sales, costs, and gross profit being in millions of dollars; numbers may not add due to rounding):

+--------------------------------------------------+ ¦ ¦ ¦ ¦Cost of ¦Gross ¦Gross profit¦ +-------+-----+-----+----------+------+------------¦ ¦Company¦Year ¦Sales¦goods sold¦profit¦margin ¦ +-------+-----+-----+----------+------+------------¦ ¦McRae ¦1987 ¦$20.9¦$14.6 ¦$ 6.3 ¦43.3% ¦ +-------+-----+-----+----------+------+------------¦ ¦ ¦1986 ¦16.5 ¦12.5 ¦4.1 ¦32.6 ¦ +-------+-----+-----+----------+------+------------¦ ¦ ¦1985 ¦17.3 ¦13.4 ¦3.9 ¦28.7 ¦ +-------+-----+-----+----------+------+------------¦ ¦ ¦Total¦54.8 ¦40.4 ¦14.2 ¦35.2 ¦ +-------+-----+-----+----------+------+------------¦ ¦ ¦ ¦ ¦ ¦ ¦ ¦ +-------+-----+-----+----------+------+------------¦ ¦Wellco ¦1987 ¦19.3 ¦14.1 ¦5.2 ¦36.6 ¦ +-------+-----+-----+----------+------+------------¦ ¦ ¦1986 ¦17.3 ¦13.3 ¦4.1 ¦30.5 ¦ +-------+-----+-----+----------+------+------------¦ ¦ ¦1985 ¦15.4 ¦12.0 ¦3.6 ¦29.6 ¦ +-------+-----+-----+----------+------+------------¦ ¦ ¦Total¦52.2 ¦39.4 ¦12.8 ¦32.4 ¦ +--------------------------------------------------+

The following schedule shows the sales and operating income in millions of dollars and the operating profit margin of Justin Industries for each of the years 1985 to 1987 (numbers may not add due to rounding):

+---------------------------------------+ ¦ ¦ ¦Operating¦Operating profit¦ +-----+------+---------+----------------¦ ¦Year ¦Sales ¦income ¦margin ¦ +-----+------+---------+----------------¦ ¦1987 ¦$109.7¦$ 8.1 ¦7.4% ¦ +-----+------+---------+----------------¦ ¦1986 ¦101.2 ¦7.0 ¦6.9 ¦ +-----+------+---------+----------------¦ ¦1985 ¦103.9 ¦6.8 ¦6.5 ¦ +-----+------+---------+----------------¦ ¦Total¦314.8 ¦21.9 ¦7.0 ¦ +---------------------------------------+

Because of the large expenditures of Justin Industries for sales and marketing, a computation of its gross profit margin averaged about 47 percent during the years 1985 through 1987.

ADPR's first cost sharing election was attached to its return for its fiscal year 1986. ADPR's corporate return for its fiscal year 1986 was originally due to be filed on or before December 15, 1986. However, ADPR, by the timely filing of Form 7004, Application For Automatic Extension of Time To File Corporation Income Tax Return, received an extension of time for the filing of its fiscal year 1986 return to June 15, 1987. The firm of Arthur Young was employed by ADC and its subsidiaries, including ADPR, to audit each entity's financial records and to prepare each entity's tax returns during the fiscal years at issue. The office of Arthur Young located in Jacksonville, Florida, did this work for ADC and ADPR. Mr. Ronald E. Price (Mr. Price) and Mr. Darrell Carlis (Mr. Carlis), both certified public accountants (C.P.A.'s) with Arthur Young, were assigned to prepare the Federal corporate income tax returns for ADC and ADPR.

Around 7 p.m. on Monday, June 15, 1987, Mr. Price and Mr. Carlis personally delivered several tax returns, including the return of ADPR for its fiscal year 1986, to the offices of ADC in Darien, Georgia, for execution by a corporate officer. The Federal income tax return, Form 1120, of ADPR for its fiscal year 1986 contained an election on Form 5712-A under section 936(h)(5) to use the cost sharing method of computing income. ADPR also elected to have the cost sharing method apply to all prior taxable years for which section 936(h) was in effect. The tax returns which Mr. Price and Mr. Carlis presented to Mr. Fleming, a corporate officer of ADC and ADPR, for signature around 8 p.m. were: (1) ADC's Federal income tax return, Form 1120, for its fiscal year 1986; (2) ADC's Georgia State income tax return, form 600, for its fiscal year 1986; (3) ADC's Georgia State estimated income tax return for its fiscal year 1987; (4) ADPR's Federal income tax return, Form 1120, for its fiscal year 1986; (5) ADPR's Federal estimated income tax return for its fiscal year 1987; (6) ADPR's Commonwealth of Puerto Rico income tax return for its fiscal year 1986, and (7) ADVI's Federal income tax return, Form 1120, for its fiscal year 1986.

Mr. Fleming reviewed and signed each of the returns presented to him by Mr. Price and Mr. Carlis and gave the signed returns back to the two C.P.A.'s for mailing. Mr. Price and Mr. Carlis placed each of the returns in an envelope properly addressed to the office where it should have been filed, sealed each envelope, and took the envelopes containing the returns to the post office on Kings Road in Jacksonville, Florida. Mr. Price placed the proper postage to send each return by certified mail on the envelope containing that return by postage meter, prepared and dated a certified mail receipt, and placed a return receipt for each tax return on the envelope containing that return. Mr. Price placed the envelope containing each corporate tax return into the metered mail slot inside the post office on Kings Road in Jacksonville, Florida. The ADC Form 1120, which was mailed on June 15, 1987, to Atlanta, Georgia, was stamped received in Atlanta, Georgia, on June 19, 1987; the ADC Georgia income tax return, form 600, also mailed to Atlanta, Georgia, was stamped received in Atlanta, Georgia, on June 18, 1987; the ADC Georgia estimated income tax return mailed to Atlanta, Georgia, was stamped received in Atlanta, Georgia, on June 19, 1987; the ADPR estimated income tax return mailed to the Department of Treasury, Bureau of Income Tax, San Juan, Puerto Rico, was stamped received in San Juan on June 23, 1987; the ADPR Commonwealth of Puerto Rico income tax return mailed to San Juan, Puerto Rico, was stamped received in San Juan, Puerto Rico, on June 23, 1987; ADVI's Federal income tax return, Form 1120, mailed to the Internal Revenue Service Center in Philadelphia was stamped received in Philadelphia, Pennsylvania, on June 29, 1987; and the ADPR Federal income tax return, Form 1120, for its fiscal year 1986 mailed to the Internal Revenue Service Center in Philadelphia was stamped received by the Philadelphia, Pennsylvania, Internal Revenue Service Center on June 30, 1987.

The envelopes in which all the Federal tax returns, other than the ADPR return for its fiscal year 1986, were mailed were retained by representatives of the IRS. Each of these envelopes showed a private metered postmark of June 15, 1987, and none of these contained an official U.S. Postal Service postmark. The envelope in which ADPR's Federal income tax return for its fiscal year 1986 was mailed to the IRS Philadelphia Service Center could not be found by respondent. At the time of the trial representatives of respondent were still attempting to find that envelope, but had not found it by the time the trial of this case was concluded.

All mail deposited in the inside mail slot before 10 p.m. on June 15, 1987, at the Kings Road post office in Jacksonville, Florida, would, under normal post office procedures, have received a postmark of June 15, 1987.

The Jacksonville, Florida, post office on Kings Road maintained an after-hours window in the lobby of the post office. This window has been open on a 24-hour-a-day basis ever since this mail facility opened in October of 1975. Mr. Price did not know of the availability of this window where he could have had a certified mail receipt stamped when he mailed ADPR's fiscal year 1986 return with the attached cost sharing election.

The U.S. Postal Service has an overnight, a 2-day, and a 3-day standard that it attempts to satisfy in making deliveries of first class mail across the country. This does not mean that all first class mail is delivered within the overnight, 2-, or 3-day standards. It means that the system should have the mail delivered within the established postal service standards. First class mail that is deposited and canceled in the evening at a post office in Jacksonville, Florida, should have been delivered in Philadelphia, Pennsylvania, on the second day, or certainly on the third day, from the mailing date. This applies to both stamped and metered mail. Although it would not be unusual for mail to take more than 3 days to travel to its place of destination in Philadelphia from Jacksonville, it would have been out of the ordinary for an item of mail to take 14 or 15 days to travel that route in June 1987, if it was addressed properly. There is nothing in this record to indicate that there were any abnormal circumstances affecting delivery of mail from Jacksonville, Florida, to Philadelphia, Pennsylvania, on June 15 through 19, 1987.

ADVI's Federal income tax return, Form 1120, for its fiscal year 1986, which Mr. Price mailed to the Philadelphia Service Center on June 15, 1987, from Jacksonville, Florida, was stamped received by the Philadelphia Service Center on June 29, 1987. Respondent retained the envelope that contained ADVI's Form 1120. The Internal Revenue Service subsequently mailed a letter dated August 10, 1987, to petitioner stating that the IRS had not received ADVI's Federal income tax return for its fiscal year 1986. Petitioner's accountants mailed a reply to the IRS' letter transmitting to the IRS a copy of the return receipt for certified mail which they had received with respect to the envelope containing ADVI's Form 1120 for its fiscal year 1986 and a copy of the tax return. The accountants received no response to this letter.

Mr. Price requested Ms. Sheryl Sheffer (Ms. Sheffer), a tax senior at Arthur Young, to determine whether ADPR should make a cost sharing election under section 936(h)(5)(C)(i)(I) for its fiscal year 1986.

Ms. Sheffer determined that the cost sharing method was the best method for ADPR to use. Ms. Sheffer determined that ADPR was required to make a cost sharing payment for the expenditures paid or accrued by ADPR's affiliated group for use of molds under the Ro-Search agreement pursuant to section 936(h)(5)(C)(i)(I). Mr. Price, however, disagreed with Ms. Sheffer's conclusion because Mr. Price determined that royalty payments to Ro-Search related to the vulcanizing process which was performed entirely at the Darien plant, and, hence, no cost sharing payment was required. Mr. Price concluded that the subsidiary was not the beneficiary of the service for which the royalty was paid. Mr. Price further concluded that the cost sharing method election could be made retroactively.

The accounting firm, based on Mr. Price's conclusion and advice, entered “None” on line 7, Schedule P of Forms 5735 relating to cost saving payments for product area research attached to ADPR's 1986 and 1987 Federal tax returns.

During the preparation of ADPR's 1986 return, Mr. Fleming discussed the cost sharing method with representatives of Arthur Young. Mr. Fleming knew that no cost sharing payment was made by ADPR to ADC for product area research. Mr. Fleming relied on the advice of accountants at Arthur Young in determining that no such cost sharing payment should be made. Mr. Fleming was aware that personnel at Arthur Young had researched the topic, and that the accountants had determined that a payment was not necessary.

Mr. Fleming specifically discussed with the accountants for ADPR the ramifications to ADPR of making the cost sharing election. The cost sharing election signed by Mr. Fleming and attached to ADPR's Federal income tax return for its fiscal year 1986 expressly refers to section 936(h)(5)(C)(i) and section 1.936-7(a), Income Tax Regs.

Mr. Fleming signed ADPR's 1986 tax return and the attached cost sharing election. Attached to the 1986 return was Schedule P, Form 5735, signed by Mr. Fleming. The instructions for line 1 on part 1 of Schedule P, Form 5735, attached to ADPR's Federal income tax return for its fiscal year 1986 state: “Enter the amount of product area research expenditures incurred by * * * all * * * members of the affiliated group”.

ADPR did not file a cost sharing election with its Federal income tax return for its fiscal year 1985. ADPR did not make a cost sharing payment to ADC for product area research during the years in issue or in any of the years subsequent. On brief, respondent has conceded in this case that if ADPR made a proper cost sharing election for its fiscal year 1986, that election also applies to ADPR's fiscal year 1985.

The Office of Management and Budget's statistical policy division prepares the Standard Industrial Classification Manual. The three-digit Standard Industrial Classification code (SIC) for “footwear, except rubber” is number 314, which includes the uppers produced in Puerto Rico and the finished boots produced by petitioner.

A statutory notice of deficiency was mailed to petitioner on April 3, 1992. In the statutory notice of deficiency, respondent reduced petitioner's deduction for purchases from ADPR by the amounts of $698,605, $677,594, and $986,959 for its fiscal years 1985, 1986, and 1987, respectively.

The amount of the adjustment to ADPR's markup for each year here in issue was calculated by reducing ADPR's costs by disallowing any allowance for raw materials before applying the profit margin percentage, which respondent further adjusted to 15 percent. The adjustment was computed as follows:

+-------------------------------+ ¦Altama Delta ¦ +-------------------------------¦ ¦Intercompany Pricing Adjustment¦ +-------------------------------¦ ¦Markup on Cost ¦ +-------------------------------¦ ¦ ¦ +-------------------------------+

FY 85 FY 86 FY 87 ADPR's costs per certified financial statements: Labor and overhead $644,711 $823,929 $ 935,779 Selling, general and administration expenses 131,783 173,068 183,885 Total costs allowed in pricing base 776,494 996,997 1,119,664 ADPR's reported profits per financial statements 815,079 827,154 1,154,909 ADPR's correct operating profit (15% markup times 116,474 149,560 167,950 cost) Decrease to petitioner's intercompany purchases 698,605 677,594 986,959

Respondent determined that ADPR did not timely file a Federal income tax return and accompanying election under section 936(h) for its fiscal year 1986 and, therefore, had made no valid election for computing its taxable income under the cost sharing method. For this reason, respondent concluded that the markup should be applied only to manufacturing costs, which did not include materials costs. Respondent further determined that the decrease to the dollar amounts of petitioner's intercompany purchases for its fiscal years 1985 through 1987 were loans to ADPR which required that imputed interest be added to petitioner's income for each year. Petitioner does not contend that the amounts, if any, paid by it to ADPR in excess of an arm's-length purchase price for the uppers should be treated as a capital contribution.

OPINION

The ultimate issue for decision in this case is the amount petitioner should have paid on an arm's-length basis to its subsidiary, ADPR, during the years here in issue for the uppers ADPR transferred to it. The answer to this question controls the amounts, if any, by which petitioner's income for each of the years here in issue will be increased. However, there are numerous subissues involved in the resolution of the ultimate issue.

Before we can begin a discussion of the evidence with respect to the ultimate question, it is necessary to consider on what basis the resolution should be made. Petitioner contends that it made a proper election to determine prices on a cost sharing basis for each of its fiscal years 1985, 1986, and 1987 under the provisions of section 936(h)(5)(F)(i). Respondent recognizes that such an election was made by ADPR for its fiscal year 1987, but contends that no proper election was made for its fiscal years 1985 and 1986. However, respondent concedes that if a proper election was made by ADPR for its fiscal year 1986, that election is properly applicable in this case to its fiscal year 1985. It is therefore necessary to determine whether petitioner made a proper election to compute prices on a cost sharing basis for its fiscal year 1986.

Section 936(h)(5)(F)(i) provides that an election to use the cost sharing method shall be made on or before the due date prescribed by law for filing the income tax return of the electing corporation (including extensions) for its first taxable year after December 31, 1982. ADPR attached to its Federal income tax return for its fiscal year ending September 27, 1986, a cost sharing election. This income tax return was originally due to be filed on or before December 15, 1986, but that date was extended to June 15, 1987. The issue in respect to whether ADPR is entitled to use the cost sharing method, therefore, is controlled by a determination of whether ADPR's Federal income tax return for its fiscal year 1986 was filed or deemed to be filed on or before June 15, 1987.

From the facts, it is completely clear that ADPR's return for its fiscal year 1986 was deposited in a box for the deposit of mail in a U.S. post office in Jacksonville, Florida, by 10 p.m. on June 15, 1987, in a properly addressed envelope with proper metered postage for certified mail and a return receipt request placed on that envelope. It is also clear that there was a metered postmark on the envelope of June 15, 1987. It is unnecessary to further discuss the evidence in this regard, since we not only consider the conclusion from the evidence clear, but respondent on brief has conceded that ADPR's Federal income tax return for its fiscal year 1986 was properly mailed by certified mail by 10 p.m. on June 15, 1987. The mailing of ADPR's return does, of course, carry with it the normal presumption of receipt of the document in due course, which the record here shows would normally be 3 days. This means that ADPR's Federal income tax return would normally have been received by the Philadelphia Service Center on or before June 18, 1987, or certainly not later than June 19, 1987, if the counting of the 3 days started on the 16th, the day following the mailing. The envelope in which the return was mailed has not been located by respondent and, therefore, is not available. From this fact we conclude that this envelope was either lost or destroyed. However, the return receipt received by petitioner from the Philadelphia Service Center bears a stamped date of June 30, 1987, and the return bears a stamped received date by the Philadelphia Service Center of June 30, 1987.

It has long been settled that a received stamp placed on a document by an employee of the Internal Revenue Service carries a presumption that the document was actually received on the date shown by the stamp. See Perkins v. Commissioner, 33 B.T.A. 606 (1935); Wedemeyer v. Commissioner, T.C. Memo. 1990-324, affd. without published opinion 959 F.2d 243 (9th Cir. 1992). However, this presumption is rebuttable, and there have been a number of cases in which, based on the evidence, we have considered the presumption rebutted. Kralstein v. Commissioner, 38 T.C. 810, 819-820 (1962); Oppenheimer v. Commissioner, 16 T.C. 515, 527 (1951); Perkins v. Commissioner, supra at 616, 617. Although these cases were decided prior to the enactment of section 7502, this fact does not affect the holding that a stamped date by the IRS on a document is a presumption of the receipt date that may be rebutted and not conclusive of the receipt date. There is also the presumption here, as above stated, that ADPR's fiscal 1986 Federal income tax return would have been received by the Philadelphia Service Center in due course, which is the presumption we relied upon in Estate of Wood v. Commissioner, 92 T.C. 793 (1989), affd. 909 F.2d 1155 (8th Cir. 1990). In the Estate of Wood case we relied on the Supreme Court case of Hagner v. United States, 285 U.S. 427 (1932). In that case, the Supreme Court pointed out that when a letter properly addressed is mailed at the post office, the presumption is created that it reached its destination in the usual time and was actually received by the person to whom it was addressed. In the Estate of Wood case, we also cited numerous other cases for this same proposition, including Crude Oil Corp. of America v. Commissioner, 161 F.2d 809 (10th Cir. 1947), revg. 6 T.C. 648 (1946), in which it was held that a capital stock tax return containing an election which was properly mailed, but according to the records of the Internal Revenue Service never received, had in fact been received. In Crude Oil Corp. of America v. Commissioner, supra, the Court of Appeals for the Tenth Circuit concluded that the clear proof of the proper mailing of the return was sufficient to overcome the presumption attaching to respondent's determination of lack of receipt of the document.

In the instant case, therefore, it is necessary, based on all the evidence, for us to determine whether the presumption attaching to the date of receipt stamped by an employee of respondent on the return receipt for certified mail and the return has been overcome by the other evidence here present, so that the presumption that the return was received in the normal time after mailing prevails.

Based on all the evidence in this case, we conclude that petitioner has overcome any presumption that attaches to the stamped received date on the return and, therefore, we hold that the tax return was timely filed based on the presumption that it was received by the Philadelphia Service Center in the normal time for receipt. Estate of Wood v. Commissioner, supra.

The evidence here shows that petitioner's accountant shortly before 10 p.m. on June 15, 1987, mailed in the same box at the post office in Jacksonville, Florida, seven different income tax forms in separate envelopes addressed to destinations in Atlanta, Philadelphia, and San Juan, Puerto Rico. All the documents except those mailed to the Philadelphia Service Center were stamped as received in the normal time range for mail to arrive to the destination to which the envelopes were addressed. There was a Form 1120 for ADC mailed to the Atlanta Service Center which was stamped received by that Service Center on June 19, 1987; an ADC form 600 mailed to Atlanta which was stamped received June 18, 1987; an ADC-Georgia Estimated Tax Return mailed to the State of Georgia Income Tax Division in Atlanta which was stamped received on June 19, 1987; an ADPR Commonwealth of Puerto Rico return mailed to San Juan, Puerto Rico, which was stamped received June 23, 1987; an ADPR Estimated Tax Return mailed to San Juan, Puerto Rico, which was stamped received June 23, 1987; an ADVI Form 1120 mailed to the Philadelphia Service Center which was stamped as received June 29, 1987; and the ADPR Form 1120 for that corporation's fiscal year ended September 27, 1986, here in issue, mailed to the Philadelphia Service Center which was stamped received June 30, 1987. The return receipts attached to the returns in each incidence were stamped received on the same date as the received date stamped on the return. The envelope in which each of these documents, except ADPR's Form 1120 for its fiscal year 1986, was mailed was retained. Respondent has been unable to find the envelope in which ADPR's return for its fiscal year 1986 was mailed, and it must be presumed either lost or destroyed.

ADC received a letter dated August 10, 1987, from the Internal Revenue Service Center in Philadelphia, Pennsylvania, stating that no Federal income tax return had been filed by ADVI for its fiscal year 1986. The controller of ADC wrote a letter, which was reviewed by ADC's accountant, stating that the return had been mailed on June 15, 1987. A copy of the metered certified receipt for mailing the return was attached to the letter. No reply was received to this letter. However, the return which was stated not to have been received was in fact received, since it was later located along with the envelope in which it was mailed and is an exhibit in this case. The stamped receipt date of June 29, 1987, appears on the return.

The facts in this case show that there was nothing unusual that happened in the transportation of mail from Jacksonville to Philadelphia during the week of June 15, 1987. The facts in this record are such that they establish that it was highly unlikely that two documents mailed at the same time in Jacksonville, Florida, both would have been delayed beyond the normal receipt time at the Philadelphia Service Center. This leads us to conclude that the problem was not in the mail being delivered to the Philadelphia Service Center untimely, but in the time of stamping the documents at the Philadelphia Service Center. Based on the evidence here present, any other conclusion would be illogical. We, therefore, conclude that no presumption attaches to the stamp placed on the return here involved in the Philadelphia Service Center. We have an admitted timely mailing on June 15, 1987, of the return and the receipt of the return in the Philadelphia Service Center. The facts here present cause the presumption which we have previously discussed; i.e., that a document properly addressed bearing the proper postage which is placed in the U.S. mail is received by the addressee in normal mail delivery time, to control. See Estate of Wood v. Commissioner, supra, and cases there cited at 798-799.

Section 7502(a) provides that when any return which is required to be filed by a prescribed date under the internal revenue laws is, after such date, delivered by U.S. mail to the office with which the return is required to be filed, the date of the U.S. postmark stamped on the cover in which the return is mailed shall be deemed to be the date of delivery if the postmark date falls on or before the prescribed date for the filing of the document. Here, the return was deposited in the U.S. mail on the prescribed date in an envelope bearing a metered postmark, postage prepaid, properly addressed to the office with which the return was to be filed. Section 7502(b) provides that the section shall apply in the case of postmarks not made by the U.S. Postal Service only if and to the extent provided by regulations prescribed by the Secretary.

Since respondent has either lost or destroyed the envelope in which the return here involved was mailed, under most circumstances it could be presumed that the envelope would bear a postmark date of June 15, 1987, since the evidence shows that a document mailed before 10 p.m. at the post office at which this return was mailed would be postmarked on the date mailed. See Hotel Equities Corp. v. Commissioner, 65 T.C. 528, 529 (1975), affd. 449 546 F.2d 725 (7th Cir. 1976), in which we concluded that the postmark date on the envelope in which the document was mailed, which envelope was not kept by respondent would be the date the evidence showed the document was mailed, stating in footnote 3 that

Since the only evidence possibly available to show a postmark date * * * on the envelope was in effect conceded by respondent's attorney to have been destroyed by respondent, we have assumed that a postmark date of * * * [the mailing date] appeared on the envelope. Respondent does not contend that we should do otherwise.

However, in the instant case the envelopes retained by respondent in which the other documents were mailed had only metered postmarks of June 15, 1987. In an affidavit in the record in this case by stipulation of the parties, a post office employee stated that mail deposited before 10:15 p.m. in the slot in which the envelopes containing those documents were deposited would bear the postmark date of the date on which they were deposited. Therefore, if that statement in this affidavit is literally accepted, no distinction was made between stamped and metered mail. Although this record might warrant concluding that had the envelope been retained it would have had a U.S. postmark of June 15, 1987, we will not rely on such a conclusion in view of the many cases we have decided which have discussed postmarks on metered mail as compared to postmarks on stamped mail.

Section 301.7502-1(c)(1)(iii), Proced. & Admin. Regs., provides that if mail bearing a metered postmark is received not later than the time when a document would ordinarily be received if it were postmarked at the same point of origin on the last day prescribed for filing the document, the document will be deemed to be timely filed. Since we have concluded that petitioner has overcome any presumption that the return here involved was received on June 30, 1987, which might attach to the fact that a stamp showing it received on that date appeared on the return, we are left with a return received without the date of receipt being shown. However, as we have pointed out above, the presumption is that since the document was properly mailed, and there were no difficulties with the mails at the time of its mailing, the document was received in the normal time for receipt. We, therefore, conclude that in accordance with the provision of section 301.7502-1(c)(1)(iii), Proced. & Admin. Regs., the return of ADPR for its fiscal year ended September 27, 1986, was timely mailed on June 15, 1987, and was received in the Philadelphia Service Center not later than a document postmarked at Jacksonville, Florida, on June 15, 1987, after 10 p.m. would ordinarily be received. We, therefore, conclude that ADPR's return for its fiscal year 1986 was timely filed.

Since we have concluded that ADPR's return for its fiscal year 1986 was timely filed, ADPR has properly made the cost sharing election provided for in section 936(h)(5)(F)(i) for its fiscal year 1986. Therefore, the transfer prices of uppers transferred from ADPR to ADC are to be computed under the cost sharing election for all the years here involved, unless, as respondent contends, ADPR waived its right to use this method by failure to make payments to its parent company for intangibles.

We have considered the position of respondent with respect to petitioner's failure to make payments for product area research to ADC and for the reasons hereinafter stated conclude that ADPR was required to make such payments, but did not waive its right to its cost sharing election by failing to make those payments.

Section 936(h)(5)(C)(i)(I) provides as follows:

(i) Cost sharing.

(I) Payment of cost sharing.--If * * * [a cost sharing election] is in effect, the electing corporation must make a payment for its share of the cost (if any) of product area research which is paid or accrued by the affiliated group during that taxable year. Such share shall not be less than the same proportion of 110 percent of the cost of such product area research which the amount of “possession sales” bears to the amount of “total sales” of the affiliated group. The cost of product area research paid or accrued solely by the electing corporation in a taxable year (excluding amounts paid directly or indirectly to or on behalf of related persons and excluding amounts paid under any cost sharing agreements with related persons) will reduce (but not below zero) the amount of the electing corporation's cost sharing payment under this method for that year. In the case of intangible property described in subsection (h)(3)(B)(i) which the electing corporation is treated as owning under subclause (II), in no event shall the payment required under this subclause be less than the inclusion or payment which would be required under section 367(d)(2)(A)(ii) or section 482 if the electing corporation were a foreign corporation.

It is necessary that a “product area research” cost be incurred within the “affiliated group” in a particular year in order for ADPR to be required to make a cost sharing payment for that year. The parties have stipulated that ADPR and ADC are members of the same affiliated group for purposes of sections 482 and 936.

Because of the affiliation between ADPR and ADC, if either paid a cost for product area research during any year at issue, ADPR would be required to make its share of the payment under section 936(h)(5)(C)(i)(I). Section 936(h)(5)(C)(i)(I)(a) defines “product area research” as follows:

For purposes of this section, the term “product area research” includes (notwithstanding any provision to the contrary) the research, development and experimental costs, losses, expenses and other related deductions--including amounts paid or accrued for the performance of research or similar activities by another person; qualified research expenses within the meaning of section 41(b); amounts paid or accrued for the use of, or the right to use, research or any of the items specified in subsection (h)(3)(B)(i); and a proper allowance for amounts incurred for the acquisition of any of the items specified in subsection (h)(3)(B)(i)--which are properly apportioned or allocated to the same product area as that in which the electing corporation conducts its activities, and a ratable part of any such costs, losses, expenses and other deductions which cannot definitely be allocated to a particular product area. Section 936(h)(5)(C)(i)(I)(e) defines “product area”:

(e) Product area.--For purposes of this section, the term “product area” shall be defined by reference to the three-digit classification of the Standard Industrial Classification [SIC] code. The Secretary may provide for the aggregation of two or more three-digit classifications where appropriate, and for a classification system other than the Standard Industrial Classification code in appropriate cases. Section 1.936-6(a)(1), Q&A-1, Income Tax Regs., states as follows:

Answer 1: Solely for the purpose of determining the tax consequences of operating in a possession, the Secretary or his delegate has exclusive jurisdiction to decide the proper SIC category under which a product is classified. For this purpose, the product area under which a product is classified will be determined according to the 1972 edition of the SIC code. From time to time and in appropriate cases, the Secretary may prescribe regulations or issue rulings determining the proper SIC category under which a particular product is to be classified, and may prescribe regulations for aggregating two or more three-digit classifications of the SIC code and for classifying product areas according to a system other than under the SIC code.

Under the 1972 edition of the SIC code, the three-digit code for footwear (except rubber) is 314, which includes the uppers produced in Puerto Rico and the finished boots produced by petitioner. Since the Secretary (or his delegate) who has exclusive jurisdiction to decide the proper SIC category under which a product is classified has not issued a ruling to the contrary, we conclude that the uppers produced by ADPR are considered to be in the same product area as the finished boots in determining whether any product area research existed with regard to the affiliated group.

The parties are in disagreement as to whether product area research expenditures existed within the affiliated group with regard to the manufacture of the combat boots. Under section 1.936-6(a)(1), Q&A-3, Income Tax Regs., product area research expenditures not only include expenditures deductible under section 174 or expenses incurred by U.S. affiliates, but also include deductions permitted under section 168 with respect to research property which are not deductible under section 174, payments (such as royalties) for the use of, or the right to use, intangibles listed under section 936(h)(3)(B)(i), an allowance for amounts incurred with respect to the manufacturing intangibles equal to the allowable depreciation or amortization on the intangible property for the year or, in the case of nondepreciable or nonamortizable property, the cost of the acquisition as if it were amortizable over 5 years (provided that the payment is noncontingent), and in the case of any contingent payment made in acquiring nonamortizable property, the amount of the payment when paid or accrued. ADC deducted no product area research costs for the fiscal years 1985 through 1987 under either section 174 or section 168. Therefore, the question here is whether the royalties paid by petitioner to Ro-Search for the use of the molds used in the vulcanizing process are payments for product area research within the meaning of section 936(h)(3)(B)(i).

We find that petitioner had the right to use, through its licensing agreement with Ro-Search, an intangible as defined under section 936(h)(3)(B)(i). This section states that the intangible must be a “patent, invention, formula, process, design, pattern, or knowhow” which had substantial value independent of the services of any individual. It is clear from the record that the design for the molds had substantial value. Mr. Jensen testified that ADC had determined that it would have cost between $4 and $5 million to develop these molds. Dr. Larry Dildine (Dr. Dildine), petitioner's expert witness, testified that it would cost at least $750,000 for a company to develop these molds. Dr. Dildine further testified that the Ro-Search agreement greatly increased the profitability of the combat boot manufacturers. It is clear from the record that the reason petitioner entered into the agreement with Ro-Search was the high cost of manufacturing the molds. While the DOD provided the specifications for the new boot and assisted Ro-Search in the design of the new molds, the record shows that Ro-Search incurred substantial expenses and expended substantial effort in developing the molds. We conclude that the design for these molds constitutes an intangible as described under section 936(h)(3)(B)(i) which the affiliated group had the right to use.

Since the design for these molds constitutes an intangible as described under section 936(h)(3)(B)(i), we conclude that the cost to ADC for the use of these molds was a cost of product area research and that ADPR should have made a cost sharing payment to ADC with respect to the cost of use of the molds. However, under section 936(h)(5)(C)(i)(III), the failure to make timely cost sharing payments for product area research does not revoke ADPR's cost sharing election unless such failure was due in whole or in part to fraud or willful neglect.

Under section 936(h)(5)(C)(i)(III)(a), if failure to make a timely cost sharing payment is due in whole or in part to fraud or willful neglect, the electing corporation shall be deemed to have revoked the cost sharing election made under section 936(h)(5)(A) on the first day of the taxable year for which the cost sharing payment was required. Respondent's position is that petitioner's failure to make a cost sharing payment was due at least in part to willful neglect by petitioner's accounting firm in its research with respect to whether cost sharing payments were required and to the failure of petitioner's officers to further investigate the necessity for such payments.

Respondent contends that petitioner's accountant, Mr. Price, was willfully neglectful when he disregarded Ms. Sheffer's findings that a cost sharing payment was due. Ms. Sheffer, a tax senior, determined that petitioner was required to make a cost sharing payment, but Mr. Price disagreed with Ms. Sheffer's conclusion because Mr. Price concluded that the fee paid by petitioner to Ro-Search related entirely to processes performed in the Darien plant. Respondent argues that Mr. Price's disregard for Ms. Sheffer's findings evidences willful neglect. Respondent further argues that Mr. Fleming's reliance on Mr. Price's conclusions shows willful neglect on the part of Mr. Fleming.

Mr. Price testified that it was his opinion that no cost sharing payment was required to be made by ADPR to ADC. Without deciding whether Mr. Price was negligent in not ordering further research, it is clear that his failure to advise ADPR to make such a payment was not willful, but merely a mistaken conclusion from a complex statute.

Furthermore, we consider the statute clear that the fraud or willful neglect must be on the part of the electing corporation to cause a revocation of the election. Generally, fraud or willful neglect by a corporation must be by the corporate officers or controlling shareholders. The record here shows that the officers of ADPR and also ADC properly relied on the advice of their accountants in not having ADPR make a cost sharing payment. The record shows that representatives of petitioner's accounting firm spent considerable time researching the relevant issues and based their conclusions on this research. Their research included an analysis of cases, statutes, regulations, and secondary authority pertinent to the cost sharing election. Though Mr. Price was incorrect in his analysis, he spent considerable time in preparation of ADC's and ADPR's returns and thoroughly researched the cost sharing issue. Mr. Price disregarded Ms. Sheffer's conclusion on the basis that Ms. Sheffer did not have full knowledge of the relationship between ADC and ADPR. Mr. Price advised the officers of petitioner and ADPR that ADPR was not required to make cost sharing payments for product area research or intangibles, and the officers relied on this advice. We, therefore, conclude that ADPR's failure to make timely cost sharing payments was not due to willful neglect.

Under the cost sharing method of reporting income, section 936(h)(5)(C)(i)(I) requires ADPR to pay a share of the cost (if any) of the product area research paid by the affiliated group. The amount of the payment under section 936(h)(5)(C)(i)(IV)(a) required to be made by ADPR is not income to ADC, but reduces the amount of deductions otherwise allowable to ADC. See sec. 1.936-6(a)(5), Q&A-1, Income Tax Regs.

In general terms, section 936(h)(5)(C)(i)(I) states that the amount of the payment shall be not less than 110 percent of the cost of the product area research paid by the affiliated group, which is then multiplied by a fraction, the numerator of which is “possession sales” and the denominator of which is “total sales”. When a cost sharing payment is not timely made, an additional amount is added to the payment pursuant to section 936(h)(5)(C)(i)(III)(a).

Section 936(h)(5)(C)(i)(I)(c) provides that “possession sales” are:

the aggregate sales * * * for the taxable year to persons who are not members of the affiliated group by members of the affiliated group of products produced, in whole or in part, by the electing corporation in the possession which are in the same product area as is used for determining the amount of product area research, and of services rendered, in whole or in part, in the possession in such product area to persons who are not members of the affiliated group.

Here, all of the uppers made by ADPR and transferred to ADC were sold as part of the total boot to persons who were not affiliated with either ADC or ADPR. Therefore, under this provision, “possession sales” of ADPR were its total sales of uppers. The problem is determining the amount of these sales since all of ADPR's production was for components of the boot. In that respect section 1.936-6(a)(2), Q&A-2, Income Tax Regs., provides:

Question 2: For purposes of the numerator of the cost sharing fraction, how are possession sales computed where the possession product is a component product or an end-product form?

Answer 2: (i) The sales price of the component product or end-product form is determined as follows. With respect to a component product, an independent sales price from comparable uncontrolled transactions must be used if such price can be determined in accordance with §1.482-2(e)(2). If an independent sales price of the component product from comparable uncontrolled transactions cannot be determined, then the sales price of the component product shall be deemed to be equal to the transfer price, determined under the appropriate section 482 method, which the possessions corporation uses under the cost sharing method in computing the income it derives from the active conduct of a trade or business in the possession with respect to the component product. * * *

Therefore, in this case, it is necessary to determine the appropriate transfer price of uppers transferred from ADPR to ADC under section 482 before we can determine the appropriate amount of ADPR's share of the payment by the affiliated group for product area research. This amount, in accordance with the statute and regulations, also determines the “arm's-length” price for the uppers which should have been paid by ADC to ADPR. As the regulation states, the two are the same.

gives the Commissioner broad authority to allocate gross income, deductions, credits, or allowances among commonly controlled organizations if the Commissioner determines that such an allocation is necessary to prevent the evasion of taxes or clearly to reflect the income of such entities. The purpose of section 482 is to place a controlled taxpayer on a tax parity with an unrelated taxpayer by preventing an artificial shifting of incomes among controlled taxpayers. Section 482 is designed to accomplish this purpose by providing for a determination of the true taxable income of the controlled taxpayer according to the standard of an uncontrolled taxpayer. Seagate Tech., Inc. & Consol. Subs. v. Commissioner, 102 T.C. 149, 163 (1994); Sundstrand Corp. v. Commissioner, 96 T.C. 226, 352-353 (1991); Paccar, Inc. v. Commissioner, 85 T.C. 754, 783 (1985), affd. 849 F.2d 393 (9th Cir. 1988).

The Commissioner's determination as set forth in the notice of deficiency is presumptively correct. The Commissioner also has broad discretion in her application of section 482, and her determination will be upheld unless the taxpayer shows it to be arbitrary, capricious, or unreasonable. Seagate Tech., Inc. & Consol. Subs. v. Commissioner, supra at 164; Sundstrand Corp. v. Commissioner, supra at 353; Paccar, Inc. v. Commissioner, supra at 787. The focus is on the reasonableness of the result, and not on the details of the methodology used. Seagate Tech., Inc. & Consol. Subs. v. Commissioner, supra; Bausch & Lomb, Inc. v. Commissioner, 92 T.C. 525, 581-582 (1989), affd. 933 F.2d 1084 (2d Cir. 1991); see also Eli Lilly & Co. v. United States, 178 Ct. Cl. 666, 372 F.2d 990, 997 (1967).

Petitioner contends that respondent's determinations in the notice of deficiency and as argued at trial are arbitrary and unreasonable since respondent reallocated the profit split between ADC and ADPR by attributing to ADPR approximately 7.58 percent of the profit of the affiliated group when ADPR incurred over 50 percent of the costs of manufacturing of the affiliated group. Petitioner argues that using a proper selling price for the uppers transferred by ADPR to ADC results in a profit for ADPR greatly in excess of that determined by respondent, thereby establishing the arbitrariness and unreasonableness of respondent's determinations.

Respondent determined in her notice of deficiency that in arriving at a reasonable arm's-length price for the uppers manufactured by ADPR, the cost for raw materials should not be included in the costs to which a percentage markup is applied. Respondent applied a 15-percent profit margin for the years 1985 through 1987 to ADPR's costs other than raw material costs. Respondent has now applied a 7.5-percent operating profit margin first to costs without including materials costs and, in the alternative, if a cost sharing election is in effect, to all costs, including raw materials costs.

Petitioner initially contends that the burden of proof should shift to respondent because respondent's determinations in the notice of deficiency were based on a contract manufacturing theory, a theory that respondent has now abandoned. Respondent argues that not only did she not employ the contract manufacturing theory in the notice of deficiency, but that the burden of proof is on respondent only as to “increases in deficiency”, and since there is no increased deficiency for any year in the instant case, respondent should not be required to bear the burden of proof on any issue.

It is well settled that generally the taxpayer bears the burden of proof to show error in the deficiency determined by the Commissioner in the notice of deficiency. Rule 142(a); Welch v. Helvering, 290 U.S. 111 (1933). The fact that the Commissioner relies on alternative theories at trial, supported by methodology different from that used in the notice of deficiency, does not necessarily place the burden on the Commissioner or cause her determinations to be arbitrary, capricious, or unreasonable. See Sundstrand Corp. v. Commissioner, 96 T.C. at 354-355. Since respondent has not claimed a deficiency in excess of that determined in the notice of deficiency, the burden of proof has not shifted to her merely because of revisions made at trial to the detailed section 482 reallocations contained in the notice of deficiency. Seagate Tech., Inc. & Consol. Subs. v. Commissioner, supra at 168-172. Considering all the facts in this case, we conclude that the burden remains on petitioner to show that respondent's determinations are arbitrary or unreasonable, and to show the amount of a reasonable transfer price for the uppers.

In arriving at a selling price for its uppers, ADC applied a profit percentage to ADPR's actual cost of manufacturing the uppers including materials costs. This profit factor was calculated by estimating the total profit on each contract and dividing it between petitioner and ADPR on the basis of the manufacturing costs incurred by each in producing the boots covered by the contract. The transfer price of the uppers for each contract differed depending on the estimated profit. If the costs incurred by ADPR increased, ADC would adjust the purchase price based on the increased costs.

This method of determining the transfer price is not one of the methods described by the regulations.

Respondent now relies on her expert's testimony to determine a transfer price rather than on the method used in the notice of deficiency. The method used by respondent's expert is not a method described by the regulations.

Section 936(h)(5) requires that where a cost sharing election has been made, the electing corporation determine its intercompany pricing under the appropriate section 482 method. Sec. 936(h)(5)(C)(i)(IV)(b); sec. 1.936-6(b)(1), Q&A-12-17, Income Tax Regs. The appropriate methods for determining the price of transferred tangible property under the 1968 section 482 regulations, the regulations in effect during petitioner's fiscal years 1985 through 1987, in order of priority, were: (1) The comparable uncontrolled price method; (2) the resale price method; and (3) the cost-plus method. Sec. 1.482-2(e), Income Tax Regs. Only if none of the above methods is viable may a fourth method be used. Sec. 1.482-2(e)(1)(iii), Income Tax Regs. The parties agree that neither the comparable uncontrolled price method nor the resale price method is applicable to the facts here present, and we agree with their conclusion. The cost-plus method is therefore the preferred method under the regulations, unless we find that the cost-plus method is likewise inappropriate. The cost-plus method using gross profit margins is preferable to a method utilizing operating profits if there are appropriate comparables available. Seagate Tech., Inc. & Consol. Subs. v. Commissioner, 102 T.C. at 188-192.

Under the cost-plus method, the arm's-length price of a controlled sale of property is computed by adding to the cost of producing such property an amount that is equal to such cost multiplied by the appropriate gross profit percentage, plus or minus any adjustments as provided for in the regulations. Sec. 1.482-2(e)(4)(i), Income Tax Regs. The appropriate gross profit percentage is equal to the gross profit percentage earned by the seller or another party on the uncontrolled sale or sales of property which are most similar to the controlled sale in question. Sec. 1.482-2(e)(4)(iii), Income Tax Regs.

The most important factors to be considered in determining the similarity of the uncontrolled sale or sales are: (1) The type of property involved in the sales; (2) the functions performed by the seller with respect to the property sold; (3) the effect of any intangible property used by the seller in connection with the property sold; and (4) the geographic market in which the functions are performed by the seller. Sec. 1.482-2(e)(4)(iii)(a)-(d), Income Tax Regs.

Whenever possible, gross profit percentages should be derived from uncontrolled sales made by the seller involved in the controlled sale. However, evidence of an appropriate gross profit percentage may be derived from similar uncontrolled sales by other sellers whether or not such sellers are members of the controlled group. Sec. 1.482-2(e)(4)(iv), Income Tax Regs.

Dr. Dildine, petitioner's expert witness, testified that the appropriate method of analyzing the related party transactions between ADPR and petitioner was the cost-plus method. He performed a functional analysis of ADPR and ADC and determined that the transfer prices charged by ADPR to ADC for finished uppers, which resulted in an average annual gross profit margin of 23 percent for ADPR for the years at issue, were equivalent to arm's-length and reasonable.

In his analysis, Dr. Dildine found that the cost-plus method was appropriate in cases involving the manufacture, assembly, or other production of goods sold to a related party. In utilizing the cost-plus method, Dr. Dildine looked for functional comparables, with a preference for products in the same product category or general industry.

Dr. Dildine testified that there were no perfect comparables to ADPR, but concluded that the military footwear industry, in which only four companies were active during the years in issue, provided the closest comparables. Dr. Dildine recognized that petitioner's profits were higher than the profits of the footwear industry as a whole, but found that petitioner's profits were comparable to the profits of the other companies that sold combat boots to the DOD.

Dr. Dildine determined that the two military footwear companies for which financial data was available, Wellco and McRae, were the best comparables from which to obtain a gross profit percentage markup to be applied to ADPR's costs. Sales of military boots to the DOD comprised approximately 95 percent of the revenues of McRae's footwear and apparel business segment over the 1985 to 1987 period, while approximately 80 percent of Wellco's revenues were from the sale of combat boots to the DOD over that same period. Wellco's remaining revenues were from sales of military-style boots for export and for the U.S. civilian market, and from service fees generated by its mold distribution program under its agreement with the other three manufacturers of combat boots for the DOD. Segment data for Wellco was not available. During the 1985 to 1987 period, Wellco had manufacturing plants in North Carolina and Puerto Rico, and McRae maintained a manufacturing plant in North Carolina.

In order to appropriately compare ADPR's transactions with ADC's and the uncontrolled transactions between Wellco and McRae and the DOD, Dr. Dildine considered the relative levels of operating expenses and cost of goods sold as they related to the costs of production at each entity. Dr. Dildine included in the costs of production the total cost of goods sold, which included each company's determination of its materials costs, purchases, labor costs, and factory overhead.

Since no separate cost of goods sold or operating expense data was available for McRae's boot and apparel manufacturing segments, Dr. Dildine allocated the reported cost of goods sold for all of McRae's segments according to the ratio of net revenues for McRae's footwear and apparel business segment to net revenues for all segments. Thus, Dr. Dildine's analysis did not separate McRae's combat boot segment from its other segments.

As a check on the cost-plus method, Dr. Dildine performed a variation on the cost-plus method that used a measure of markup on total costs with respect to all three companies. Dr. Dildine found that the total cost markup was a reasonable variation on the cost-plus method.

Under the cost-plus method, Dr. Dildine found that Wellco's cost of goods sold markup ratios were 31.9 percent in 1985, 32.0 percent in 1986, and 38.1 percent in 1987, with a weighted average over the period of 34.1 percent. He found that McRae's footwear segment cost of goods sold markup ratios were 28.7 percent in 1985, 32.6 percent in 1986, 43.3 percent in 1987, and a weighted average of 35.2 percent over the period 1985 to 1987.

Dr. Dildine determined that a profit percentage of 35 percent was justified from the experience of Wellco and McRae, and that the profit of 23 percent used in computing ADPR's transfer sales prices, which was significantly less than this amount, was reasonable.

Dr. Dildine checked the reasonableness of his results with the published markups of other possession corporations under section 936. Data was available from the IRS, which published a study of tax returns of these corporations. Dr. Dildine found that the closest comparable group, leather manufacturers, maintained an average cost-plus markup of 27 percent. However, Dr. Dildine acknowledged that he did not know whether these companies owned any significant intangibles.

Dr. Dildine also compared ADPR's transactions with ADC's under what he referred to as “the profit split method”. Under section 1.482-2(e)(1)(iii), Income Tax Regs., if none of the three methods described under section 1.482-2(e)(1)(iii), Income Tax Regs., which we have discussed above, are appropriate under the facts present in the particular case, some other appropriate method of pricing is necessary. In Eli Lilly & Co. v. Commissioner, 84 T.C. at 1147-1153, we approved a profit-split approach and cited a number of other cases in which such a method had been used. Dr. Dildine testified that, under a profit-split formula, receipt by ADPR of 54 percent of the total profit generated by contracts for which the uppers were cut and stitched in Puerto Rico for the years at issue was fair and reasonable and, therefore, the transfer prices charged by ADPR were at arm's length. While this computation is some support to the cost-plus computation of Dr. Dildine, it is not otherwise relevant in this case, since the equivalent of arm's-length prices can be determined under the cost-plus method described in the regulations.

Dr. Dildine concluded that the adjustments proposed in the notice of deficiency, which allocated approximately 92 percent of total profits to ADC and only 8 percent to ADPR, were unreasonable.

Dr. Thomas Horst (Dr. Horst), the expert witness offered by respondent, testified that computations which he made supported respondent's determination. Dr. Horst testified that his computations applied a variation of the “cost-plus” method by using operating margins instead of gross markups to apply to ADPR's costs to determine the appropriate transfer price.

Dr. Horst based his computations on the assumption that the cutting and stitching operations performed by ADPR were operations that could have been performed by any footwear manufacturer. Dr. Horst testified that although the cutting and stitching operations employed by ADPR used certain intangibles, those intangibles were not unique to ADPR and, in fact, any shoe manufacturer had these same intangibles. Dr. Horst testified that Wellco and McRae were not close comparables of ADPR since both companies owned significant intangibles due to their participation in the Ro-Search agreement.

Dr. Horst testified that in his opinion it was appropriate in determining a comparable to ADPR to look to companies subject to foreign competition that were engaged in manufacturing activities similar to those of ADPR. Dr. Horst used as comparables independent U.S. footwear manufacturers which sold to unrelated parties commercial type footwear partially in the retail marketplace. Dr. Horst testified that he sought the following characteristics for close comparables with ADPR, companies that: (1) Manufactured, rather than imported, footwear; (2) manufactured footwear from leather, rather than fabric or synthetic materials; (3) did not have extensive retail operations; and (4) did not own any significant intangibles. Using these criteria, Dr. Horst concluded that Justin Industries, Inc. (Justin Industries), the B.B. Walker Co. (B.B. Walker), and the shoe operations of Official Industries, Inc. (Official Industries), were close comparables of ADPR.

Dr. Horst described Justin Industries as a company that designed, manufactured, and distributed men's and women's western style, safety, work and sports boots, shoes and belts, primarily for sale in the United States. Dr. Horst determined that Justin Industries owned no patents, trademarks, or licenses that were material in relation to the company's operations as a whole based on an analysis of Justin Industries' forms 10-K. Justin Industries provided sales, operating profits, and identifiable assets with respect to its footwear segment, but did not publish more detailed financial statement data.

B.B. Walker was primarily engaged in footwear manufacturing and distribution. Its products were mainly men's outdoor footwear, which included service industrial and ranch and western boot types, as well as a variety of styles of men's work shoes and safety-toe footwear. B.B. Walker operated a retail store division with four factory outlet stores offering brand-name merchandise at discount prices. Based on information contained in B.B. Walker's financial statements, Dr. Horst was of the opinion that B.B. Walker owned no trademarks that were material to its operations, nor did it have any material patents, licenses, franchises or concessions which were vital to the company. The financial data available pertained to the company as a whole and not to a specific segment (i.e. their cutting and stitching operations).

Official Industries' division that manufactured men's and women's shoes was Geo. B. Keith Co. (Geo. B. Keith). Geo. B. Keith operated two divisions: Klev-Bro, which manufactured medium-priced, high fashion women's shoes using synthetic materials, and Walk-Over, which manufactured medium to mediumhigh priced men's dress and casual shoes, nearly all of which had leather uppers. Approximately 63 percent of Walk-Over shoes were sold under the “Walk-Over” brand names, and 37 percent used customers' labels. Dr. Horst testified that shoes sold under the Walk-Over label were produced for the company's own inventories from which customers' orders were filled, as opposed to being produced to fill a specific customer order.

Dr. Horst testified that for the period 1985-87, Justin Industries reported cumulative operating profits equal to 7.5 percent of cumulative sales over that same period, and its operating profit did not vary greatly from year to year during that period. Dr. Horst testified that B.B. Walker's and Official Industries' average operating profit margins were 2.5 percent and 0.2 percent of sales, respectively.

Dr. Horst testified that it was more appropriate to apply a markup rate based on a commercial footwear manufacturer's operating profit margins to ADPR's total costs than to apply a markup rate based on a commercial manufacturer's gross profit margins to ADPR's cost of sales. Dr. Horst based his conclusion on his belief that the financial data available for the comparative companies included data based on sales and distribution functions that resulted in these companies' incurring operating expenses that ADPR did not incur, such as sales and marketing costs.

Expert testimony is weighed by the Court not only on the basis of the expert's qualifications, but also on the reasonableness of the conclusions of the expert based on all other credible evidence in the record. Sundstrand Corp. v. Commissioner, 96 T.C. at 359; Estate of Newhouse v. Commissioner, 94 T.C. 193, 217 (1990). We are not bound by the opinion of any expert witness, but will consider that opinion based on our analysis of the evidence in the record and applying our judgment in consideration of that evidence. Sundstrand Corp. v. Commissioner, 96 T.C. at 359;

Estate of Newhouse v. Commissioner, supra; Chiu v. Commissioner, 84 T.C. 722, 734 (1985). While we may choose to accept the opinion of one expert in its entirety, we may also be selective in the use of any portion of that opinion. Sundstrand Corp. v. Commissioner, 96 T.C. at 359 and the cases cited therein.

Instead of using gross profit margins, respondent's expert applied operating profit margins. We agree with petitioner that Dr. Horst did not apply the cost-plus method described in section 1.482-2(e)(4), Income Tax Regs. We reject as unreasonable his method for determining the appropriate transfer prices. See Seagate Tech., Inc. & Consol. Subs. v. Commissioner, supra at 191.

Dr. Horst's comparison of ADPR to manufacturers outside the combat boot industry is without merit. Dr. Horst did not know whether Justin Industries incurred expenses relating to the research and development of new styles, whereas the combat boot industry is a relatively static market.

We are also not convinced that Dr. Horst's assumption that Justin Industries owned no significant intangibles is accurate. His reliance on this conclusion is based on Justin Industries' financial statements, though even Dr. Horst admitted that this was not always an accurate determination of whether a company owns significant intangibles. For instance, as Dr. Horst admitted, it would not be unlikely for Justin Industries to have contracts that would be considered intangible property.

We find that respondent's proposed allocation is arbitrary, capricious, and unreasonable.

Based on this record, we conclude that we should look to the combat boot industry and not the footwear industry as a whole to find close comparables of ADPR. The reasons for this conclusion include: (1) The limitation on competition because solicitations for the combat boots contained “set-asides” for small businesses and preference to plants located in labor surplus areas, thus deterring larger manufacturers from entering the combat boot market; (2) it was unnecessary for combat boot manufacturers to research, develop, or market products, since the design of the boots and specifications for their manufacture were determined by the Government, resulting in combat boot manufacturing companies' not incurring costs for research and for marketing of the type incurred by other footwear manufacturers; (3) during the years here at issue, there was an unusually large solicitation for contracts for combat boots due to an error in the Government's inventory system resulting in an increased demand which inflated the combat boot industry's profits; (4) most combat boot manufacturers, including petitioner, Wellco, and McRae, submitted bids for lesser quantities than they could manufacture in order to maintain higher profit margins; and (5) the introduction of the new boot increased profits. All of these factors contributed to higher gross profit margins for combat boot manufacture and resulted in the combat boot industry's not being closely comparable to the footwear industry as a whole. In fact, the fashion footwear business and the combat boot business operated in two entirely separate marketplaces.

Dr. Dildine's use of Wellco and McRae in comparing the transactions between petitioner and ADPR was based on similar uncontrolled sales. Wellco and McRae are not strictly comparable to ADPR since they incurred expenses that ADPR did not incur. However, in our view, comparisons of companies outside the combat boot industry with ADPR are without merit.

Wellco's combat boot business constituted approximately 80 percent of its total business, while McRae's combat boot business was approximately 95 percent of its total business. Based on the profit margins of companies manufacturing commercial footwear, it is reasonable to conclude that the other businesses of Wellco and McRae had gross profit margins lower than those of their combat boot division. Therefore the use of gross profit margins of these companies is reasonable for comparison purposes.

Although the profit margins of the entire operation of both Wellco and McRae were higher than the profit margin of ADC, we find no basis in the record to conclude that ADPR in arm's-length transactions would have attained an overall gross profit margin greater than that of its parent. Gross profit margins for ADC averaged 21.01 percent, while ADPR's profit margin as computed by petitioner averaged 22.95 percent.

ADC incurred a greater risk in the sale of the completed combat boot than ADPR incurred in the manufacture of the leather uppers, since ADC guaranteed the purchase of ADPR's product. We would expect that a manufacturer operating at arm's length could expect to earn a greater return on the sale of the completed boot to the DOD than the return the manufacturer of a component of the boot, the uppers, obtained, due to the greater risk incurred in the operation. See Seagate Tech., Inc. & Consol. Subs. v. Commissioner, 102 T.C at 193-195. However, certainly the profit margin of the component manufacturer would at arm's length be related to that of the company that finished the product. We find that petitioner has shown that a markup on costs including materials costs of ADPR to arrive at transfer prices for the uppers is a proper manner of arriving at arm's-length charges. We conclude that ADPR was entitled to an arm's-length transfer price related to ADC's profit margin, but that the average gross profit margin of 23 percent as claimed by petitioner for each of the years here in issue is excessive.

Petitioner's analysis does not account for differences between Wellco and McRae, and ADPR. However, there are differences. Wellco and McRae each performed all of the manufacturing operations necessary to make a combat boot, while ADPR only manufactured the leather upper portion of the boot. Section 1.482-2(e)(4)(v), Income Tax Regs., provides as follows:

Where the most similar sale or sales from which the appropriate gross profit percentage is derived differ in any material respect from the controlled sale, the arm's length price which is computed by applying such percentage must be adjusted to reflect such differences to the extent such differences would warrant an adjustment of price in uncontrolled transactions. The differences referred to in this subdivision are those differences which have a definite and reasonably ascertainable effect on price. Since we find that the transfer price computed by petitioner was not at arm's length, it is necessary for us to make our best estimate of the appropriate transfer prices on the basis of all facts available in the record. Seagate Tech., Inc. & Consol. Subs. v. Commissioner, supra at 195; Sundstrand Corp. v. Commissioner, 96 T.C. at 375. The regulations require an adjustment of the arm's-length price of other companies for material differences between such companies and the entity to which the comparison is being made.

We find that while ADPR did incur risks with respect to its production of the leather uppers, ADC virtually guaranteed the purchase of ADPR's product. ADPR, however, was responsible for utilizing the leather in an efficient manner. ADPR's manufacturing process could have been performed by any number of leather and footwear manufacturers. ADC, meanwhile, maintained the Government contracts and established the agreement with Ro-Search for the leasing of the sole molds.

ADPR did, however, incur a large portion of the costs of manufacturing the combat boots produced. It is clear from the record that the cost of the leather was the most significant cost incurred in the manufacture of the combat boot. Therefore, on the boots whose uppers were produced by ADPR, ADPR had a substantial materials costs and a significant risk with respect to those costs.

We further find no evidence in the record for ADPR's gross profit margin to be significantly higher in its fiscal year 1985 than in its fiscal years 1986 and 1987. ADPR's reported gross profit margin in 1985 was over 10 percent higher than in the 2 subsequent years, even though the record showed that profits were driven high as a result of the introduction of the new boot. Neither Wellco's nor McRae's gross profit margins were significantly higher in 1985, and, in fact, were lower than in each of the subsequent years.

Based on this evidence, we conclude that ADPR is entitled to a gross profit margin in each of its fiscal years 1985, 1986, and 1987 in an amount equal to the average gross profit margin of ADC for its fiscal years 1986 and 1987, of approximately 19.2 percent.

This takes into account the lower risks involved for the subsidiary, as well as the fact that petitioner procured the Government contracts.

Now that we have determined how the total sales of ADPR should be computed on the equivalent of an arm's-length basis, the amount of the payment for the use of the molds under the Ro-Search agreement can be determined under the regulations. This amount is not less than 110 percent of the cost of the product area research (which is 110 percent of ADC's payment to Ro-Search) times a fraction, the numerator of which is ADPR's sales computed as we have held, and the denominator of which is total sales of the affiliated group. This latter figure can be computed from the record. Therefore, the proper payment by ADPR for product area research of ADC is merely a matter of computation. Likewise, the increased amount payable under section 936(h)(5)(C)(i)(III)(a) can be computed from the record.

Petitioner in its return claimed an amount for location savings. Respondent concedes some amount of location savings for ADPR is appropriate if we find ADPR made a cost sharing election. Location savings represent net savings in the cost of producing the product in a foreign country rather than the United States. Where location savings exist, the full benefit of those savings is allocated to the possessions corporation and added to the profit that the possessions corporation is permitted to earn under section 482. See Sundstrand Corp v. Commissioner, supra at 319.

Petitioner's accounting firm, Ernst & Young, determined location savings for the fiscal years at issue as follows:

+-------------------------------------------------------------------------+ ¦Location savings ¦1985 ¦1986 ¦1987 ¦ +-----------------------------------+-----------+------------+------------¦ ¦Labor rate savings ¦$113,349.85¦$142,594.36 ¦$152,098.54 ¦ +-----------------------------------+-----------+------------+------------¦ ¦Job training credit ¦23,840.00 ¦34,800.00 ¦11,313.00 ¦ +-----------------------------------+-----------+------------+------------¦ ¦Employee benefit savings ¦66,986.06 ¦66,284.77 ¦74,089.31 ¦ +-----------------------------------+-----------+------------+------------¦ ¦Local income tax rate differential ¦14,857.19 ¦2,012.84 ¦9,290.89 ¦ +-----------------------------------+-----------+------------+------------¦ ¦Land & building rental differential¦12,986.00 ¦12,986.00 ¦12,986.00 ¦ +-----------------------------------+-----------+------------+------------¦ ¦Freight into Puerto Rico¹ ¦65,053.00 ¦113,638.00 ¦126,456.00 ¦ +-----------------------------------+-----------+------------+------------¦ ¦Freight out of Puerto Rico ¦(65,053.00)¦(113,638.00)¦(126,456.00)¦ +-----------------------------------+-----------+------------+------------¦ ¦ ¦ ¦ ¦ ¦ +-------------------------------------------------------------------------+

Respondent concedes that labor rate savings, job training credit, and the land and building differential are properly includable in location savings. Respondent argues that employee benefit savings, freight costs, and State and local taxes are not properly includable in determining cost savings.

Respondent's expert testified briefly to the appropriate amount of cost savings for ADPR. Petitioner merely relied on the computation of its accountant without further evidence. Petitioner has failed to prove its claimed location savings. Therefore, we determine the location savings adjustment to be the amounts for each year as conceded by respondent in her brief. In concluding as we do, we are aware that the amounts conceded by respondent are less than the amounts claimed by petitioner with respect to the items respondent concedes to be properly includable in cost savings. However, on the basis of this record, we conclude that petitioner has not proven the amounts it claims.

The final issue is whether interest income should be allocated to petitioner from ADPR under the provisions of section 482 and, if so, the amounts to be allocated. Respondent argues that interest should be added to petitioner's income since the transactions between ADC and ADPR were not at arm's length under sections 1.482-1(d)(4) and 1.482-2(a)(1), Income Tax Regs. Petitioner contends that since respondent's notice of deficiency reflects an arbitrary and capricious allocation of income to ADC, the issue is moot.

Section 1.482-1(d)(4), Income Tax Regs., provides as follows:

If the members of a group of controlled taxpayers engage in transactions with one another, the district director may distribute, apportion, or allocate income, deductions, credits, or allowances to reflect the true taxable income of the individual members under the standards set forth in this section and in section 1.482-2 notwithstanding the fact that the ultimate income anticipated from a series of transactions may not be realized or is realized in a later period. * * *

Section 1.482-2(a)(1), Income Tax Regs., provides that where one member of a group of controlled entities makes a loan or advance directly or indirectly to another member of the group and charges no interest, the Commissioner may make appropriate allocations to reflect an arm's-length rate of interest for the use of such loan or advance. Respondent argues that the funds transferred to ADPR by ADC as sales proceeds in excess of an arm's-length purchase price were in effect a loan to ADPR, and, hence, interest should be imputed to ADC on the loan.

Petitioner has offered no evidence to show that interest should not be imputed to it on those portions of the sales proceeds which are greater than arm's length under the determination we have made of an appropriate transfer price. Unquestionably, there would be imputed interest on a direct loan from ADC to ADPR which was made without interest. See Latham Park Manor, Inc. v. Commissioner, 69 T.C. 199, 209-216 (1977), affd. without published opinion 618 F.2d 100 (4th Cir. 1980). Petitioner makes no claim that the transfer of excess sales price to ADPR should be characterized as a capital contribution. Whether an excess payment by a parent corporation to its subsidiary is in effect a loan is a question of fact. The excessive transfer price for uppers from ADC to ADPR has been determined by respondent to be a loan, and petitioner has not shown otherwise. We, therefore, conclude that under the facts here present imputed interest on the amount of sales proceeds transferred by ADC to ADPR in excess of the amounts we have determined to represent arm's-length transfer prices is in effect a loan to ADPR from ADC on which imputed interest is appropriate.

Decision will be entered under Rule 155. 104 T.C. No. 22, 104 T.C. 424, Tax Ct. Rep. (CCH) 50,586, Tax Ct. Rep. Dec. (RIA) 104.22

¹ This expense has already been deducted by Puerto Rico.

+-----------------------------------+ ¦Markup on cost of goods sold for ¦ +-----------------------------------¦ ¦ADPR, McRae, and Wellco ¦ +-----------------------------------¦ ¦1985 to 1987 ¦ +-----------------------------------¦ ¦(millions of dollars; ¦ +-----------------------------------¦ ¦numbers may not add due to ¦ ¦rounding) ¦ +-----------------------------------¦ +-----------------------------------+

Cost of Gross Gross profit Company Year Sales goods sold profit margin McRae 1987 $20.9 $14.6 $ 6.3 43.3% 1986 16.5 12.4 4.1 32.6 1985 17.3 13.4 3.9 28.7 Total 54.7 40.4 14.2 35.2 Wellco 1987 19.1 13.8 5.3 38.1 1986 17.2 13.1 4.2 32.0 1985 15.6 11.8 3.8 31.9 Total 51.9 38.7 13.2 34.1 ADPR 1987 7.9 6.6 1.3 20.3 1986 5.7 4.7 1.0 21.2 1985 3.9 3.0 1.0 31.6 Total 17.6 14.3 3.3 23.0

+--------------------------------------------------+ ¦ ¦ ¦ ¦Cost of ¦Gross ¦Gross profit¦ +-------+-----+-----+----------+------+------------¦ ¦Company¦Year ¦Sales¦goods sold¦profit¦margin ¦ +-------+-----+-----+----------+------+------------¦ ¦McRae ¦1987 ¦$20.9¦$14.6 ¦$ 6.3 ¦43.3% ¦ +-------+-----+-----+----------+------+------------¦ ¦ ¦1986 ¦16.5 ¦12.4 ¦4.1 ¦32.6 ¦ +-------+-----+-----+----------+------+------------¦ ¦ ¦1985 ¦17.3 ¦13.4 ¦3.9 ¦28.7 ¦ +-------+-----+-----+----------+------+------------¦ ¦ ¦Total¦54.7 ¦40.4 ¦14.2 ¦35.2 ¦ +-------+-----+-----+----------+------+------------¦ +-------+-----+-----+----------+------+------------¦ ¦Wellco ¦1987 ¦19.3 ¦14.1 ¦5.2 ¦36.6 ¦ +-------+-----+-----+----------+------+------------¦ ¦ ¦1986 ¦17.3 ¦13.3 ¦4.1 ¦30.5 ¦ +-------+-----+-----+----------+------+------------¦ ¦ ¦1985 ¦15.4 ¦12.0 ¦3.6 ¦29.6 ¦ +-------+-----+-----+----------+------+------------¦ ¦ ¦Total¦52.0 ¦39.4 ¦12.8 ¦32,4 ¦ +--------------------------------------------------+

These subsequent findings are substantially the same as the findings in Dr. Dildine's original report.

+----------------------+ ¦¦1985¦1986¦1987¦Total ¦ +----------------------+

ADPR gross profit as computed by petitioner Net sales $3,943,178 $5,716,563 $7,946,439 $17,606,180 Cost of goods sold 2,996,316 4,716,341 6,607,645 14,320,302 Gross profit 946,862 1,000,222 1,338,794 3,285,878 Gross profit margin 31.60% 21.21% 20.26% 22.95% ADC gross profit Net sales 16,910,021 15,636,913 17,792,655 50,339,589 Cost of goods sold 13,551,493 13,300,939 14,746,418 41,598,850 Gross profit 3,358,528 2,335,974 3,046,237 8,740,739 Gross profit margin 24.78% 17.56% 20.66% 21.01%

+-------------------------------------------------------+ ¦ ¦1986 ¦1987 ¦Total ¦ +-------------------+-----------+-----------+-----------¦ ¦Sales ¦$15,636,913¦$17,792,655¦$33,429,568¦ +-------------------+-----------+-----------+-----------¦ ¦Cost of goods sold ¦13,300,939 ¦14,746,418 ¦28,047,357 ¦ +-------------------+-----------+-----------+-----------¦ ¦Gross profit ¦2,335,974 ¦3,046,237 ¦5,382,211 ¦ +-------------------+-----------+-----------+-----------¦ ¦Gross profit margin¦17.56% ¦20.66% ¦19.19% ¦ +-------------------------------------------------------+