STATE OF NEW YORK

TAX APPEALS TRIBUNAL
________________________________________________

                     In the Matter of the Petition

                                         of

        TROPICANA PRODUCTS SALES, INC.                                                              DECISION
                                                                                                                                    DTA NOS. 815253
for Redetermination of a Deficiency or for Refund of                                                      AND 815564
Corporation Franchise Tax under Article 9-A of the
Tax Law for the Fiscal Years Ended July 31, 1989
through July 31, 1993.
________________________________________________

Petitioner Tropicana Products Sales, Inc., 1001 13^th Avenue East, Bradenton, Florida 34208-0338, filed an exception to the determination of the Administrative Law Judge issued on November 25, 1998. Petitioner appeared by Morrison & Foerster (Hollis L. Hyans, Esq., of counsel) and Horwood, Marcus & Berk (Fred O. Marcus, Esq., of counsel). The Division of Taxation appeared by Barbara G. Billet, Esq. (James P. Connolly, Esq., of counsel).

Petitioner filed a brief in support of its exception. The Division of Taxation filed a brief in opposition and petitioner filed a reply brief. Oral argument, at petitioner's request, was heard on December 10, 1999 in New York City.

After reviewing the entire record in this matter, the Tax Appeals Tribunal renders the following decision.

ISSUES

I. Whether the Division of Taxation properly determined that petitioner, Tropicana Products Sales, Inc., was required to file a combined franchise tax report which included Progress Services, Inc. to avoid distortion of petitioner's New York activities, business, income or capital.

II. Whether petitioner might have filed a separate franchise tax report on the ground that a combined report with its parent corporation was not required in order to avoid distortion of petitioner's New York activities, business, income or capital.

III. Whether the Division of Taxation, which required petitioner to file a combined New York State franchise tax report which included Tropicana Products, Inc. and Progress Services, Inc., violated the Due Process Clause and the Commerce Clause of the United States Constitution.

FINDINGS OF FACT

We find the facts as determined by the Administrative Law Judge except for finding of fact "47" which has been modified. The Administrative Law Judge's findings of fact and the modified finding of fact are set forth below.

1. Tropicana Products, Inc. ("TPI") is a Delaware corporation with its principal offices in Bradenton, Florida.

2. The petitioner herein, Tropicana Products Sales, Inc. ("TPS"), also a Delaware corporation, is a wholly-owned subsidiary of TPI. Its principal offices are located at TPI's Bradenton, Florida facility.

3. Progress Services, Inc. ("PSI"), a Florida corporation, is a wholly-owned subsidiary of TPI. Its principal offices and production facilities are located at TPI's Bradenton, Florida facility.

4. For the years at issue,⁽¹⁾ TPI, TPS, PSI and the other members of the TPI federal affiliated group filed a consolidated Federal income tax return. During the years at issue, TPS filed combined New York returns that included the income and factors of TPI and certain of its subsidiaries ("TPS/TPI group"), but excluded the income and factors of PSI.

5. Pursuant to an examination of the books and records of TPS for the fiscal years ended July 31, 1989 through July 31, 1991, the Division of Taxation ("Division") issued a Notice of Deficiency dated April 11, 1994 to petitioner in the amount of $368,674.00, plus penalty and interest, for a total amount due of $495,294.49 for these years. The Division determined that PSI was engaged in a single unitary business with the TPS/TPI group and that in order to properly reflect the income of TPS, the income and factors of PSI had to be combined with the income and factors of the TPS/TPI group. The Division does not contend that TPI or PSI were required to file New York separate franchise tax returns for the years at issue.

6. On or about April 30, 1996, petitioner filed amended New York general business corporation franchise tax returns (forms CT-3) and amended metropolitan transportation business tax surcharge returns (forms CT-3M/4M) for the fiscal years ended July 31, 1989 through July 31, 1991 and requested a refund of the taxes paid for those years. The theory upon which petitioner filed the amended returns was its contention that TPS and TPI should be allowed to file on a separate basis on the ground that a combined return with TPI was not required in order to avoid distortion of the New York activities, business, income or capital of TPS. The Division neither allowed nor denied TPS's refund claims.

7. On or about December 16, 1996, or more than six months after the date on which TPS filed the amended returns, TPS filed a petition with the Division of Tax Appeals seeking the amounts asserted to be due to TPS per its amended returns (this petition was in addition to the petition filed on or about July 30, 1996 contesting the amounts asserted to be due from TPS in the Notice of Deficiency).

In this matter, the Division waived reliance on the 30-day rule (20 NYCRR 6-2.4) as a basis for denial of the refunds requested in the amended returns. In exchange, TPS agreed that the Division would be allowed to conduct a field audit to determine whether TPI and TPS should be filing on a combined basis. On October 20, 1997, the Division issued a letter affirming the deemed denial of TPS's refund claims.

8. For purposes of this proceeding, TPS does not, for the years at issue, contest that it, TPI and PSI constitute a unitary business involving the production and sale of juice, juice-based beverages and cattle feed made from orange by-products.

9. For purposes of this proceeding, TPS does not, for the years at issue, contest that it had "substantial intercorporate transactions" with TPI within the meaning of 20 NYCRR 6-2.3(b) nor does TPI contest that it had "substantial intercorporate transactions" with PSI within the meaning of 20 NYCRR 6-2.3(b).

TPI's BUSINESS

10. TPI is engaged in the production, distribution and sale of not-from-concentrate chilled orange juice, grapefruit juice and apple juice, juices made from concentrate and a variety of other unchilled blended fruit juice beverages (they shall be collectively referred to as the "juice product"). TPI is the world's leading producer of chilled orange juice and holds a dominant share of the chilled orange juice market in the northeastern United States. All of TPI's juice product for domestic consumption is sold to TPS. TPI makes all foreign sales of its juice product through Seagram Export Foreign Sales Corporation Limited, a foreign corporation within the meaning of section 922 of the Internal Revenue Code.

11. TPI owns and maintains its headquarters and primary production facility in Bradenton, Florida and also owns and operates a second production facility at Ft. Pierce, Florida.

12. TPI is the largest processor of Florida oranges. Oranges are the largest single raw material purchased by TPI. Approximately 95 percent of the oranges purchased by TPI are purchased from Florida growers under contracts which range in length from one to twenty years.

13. TPI processes whole oranges and grapefruits into not-from-concentrate chilled juices at each of its Bradenton and Ft. Pierce facilities. Juice processed at the Bradenton facility is also packaged there. Juice processed at the Ft. Pierce facility is shipped in bulk to the Bradenton facility for packaging. Juice processed at the Bradenton facility for shipment to TPS's City of Industry, California facility is packaged by TPI in barrels, in concentrate form, at the Bradenton facility. TPI's processing facilities are located close to the source of its raw materials supply because such a location minimizes transportation costs and eliminates potential spoilage.

14. TPI also engages in product research and development. TPI's research and development activities include formulating new products, participating in product design and identifying customer needs. TPS assists in formulating new products by providing input regarding product design and customer needs. TPI incurs all the expenses and bears all the risks associated with its product development, process development and other research and development activities.

15. TPI owns the "Tropicana" brand name and other product brand names. The "Tropicana" brand name is considered to be one of the most powerful brand names in the food and beverage industry. The book value of this and the other intangible assets owned by TPI was approximately $412,000,000.00 in fiscal year 1991.

16. TPI provides a number of services centrally to TPS and PSI, including pension administration, payroll, centralized accounting, tax return preparation, information systems and other corporate programs.

17. TPI bears virtually all of the risks arising in connection with its juice production business. TPI bears the market risks for sales of its juice products. TPI is solely responsible for in-store loss or product spoilage.

TPS's BUSINESS

18. TPS is primarily engaged in the wholesale sale and distribution of juice product manufactured by TPI in Bradenton, Florida. During the years at issue, there was no written contract in relation to the sale and transfer of juice product between TPI and TPS and there were no invoices prepared. TPS provides at least three distinct services for TPI: sales and distribution, transportation and marketing.

19. TPS sells its product directly and indirectly to five types of customers located throughout the United States and Canada: large grocery chains, food service, convenience stores, mass merchandisers and the military.

TPS sells juice products to large retail grocers on a nationwide basis; these sales to large grocery chains account for approximately two-thirds of TPS sales. TPS ships product directly to these grocery chain customers who generally maintain their own refrigerated warehouses and distribute the product to their own store locations. For retail grocers who do not have their own warehouse facilities, TPS either distributes the juice products to wholesalers or the juice products are picked up by the wholesalers at TPS's distribution facility in Jersey City. These sales are supported by TPS with brand and trade marketing such as newspaper coupons, grocery store ads, special display cases and television air time.

TPS's food service customers include schools, hospitals and restaurants. Sales to these customers are generally made through large independent wholesalers. As with its grocery customers, TPS either distributes its products directly to food service operators or indirectly through wholesalers. For food service operator customers in the New York City metropolitan area, TPS provides a direct store delivery option from its Whitestone, New York distribution facility.

TPS sells to convenience stores through smaller local distributors, brokers or routemen in the densely populated northeastern United States. In the New York City metropolitan area, TPS sells directly to independent routemen who distribute products daily to smaller customers using their own refrigerated trucks.

20. TPS also sells juice products to mass merchandisers like Sam's, Wal-Mart and BJ's. Although mass merchandisers are the largest single type of purchaser of juice products from TPS by volume, sales to such customers do not account for the largest percentage of TPS's sales. TPS distributes its products to mass merchandisers either directly or indirectly through wholesalers.

21. Sales to military bases and installations account for the remainder of TPS's sales of juice products. It distributes its products to the military directly or indirectly through wholesalers.

22. In addition to its principal location in Bradenton, Florida and its City of Industry, California plant, TPS also maintained regional offices in Braintree, Massachusetts; New York, New York; Chicago, Illinois; Ft. Lauderdale, Florida and Los Angeles, California. It also operated a distribution facility in Kearney, New Jersey until the summer of 1990; it opened a distribution facility in Jersey City, New Jersey during the summer of 1990.

23. TPS employed approximately 55 employees at its Bradenton, Florida offices. These employees were engaged in sales, management and support activities and included managers, analysts, coordinators and clerical staff.

Until it ceased operations, the Kearney, New Jersey distribution facility employed approximately 100 people. They included order takers, dispatchers, logistical personnel and product handlers. The Kearney distribution facility serviced the northeastern United States, including New York. When it commenced operations, the Jersey City facility employed a similar number of individuals in similar positions.

TPS employed an additional 60 to 70 individuals in the New York City metropolitan area. These individuals were primarily engaged in direct sales. Terrence S. Schulke, TPS's vice president for sales employed at the Whitestone, New York facility, stated that he was aware of the fact that certain managers, originally employed by TPI, had been transferred to the employ of TPS.

24. Product sold to customers located in the northeastern United States, including New York (more than a third of TPS's U.S. sales are in the northeast, with most of those sales to the metropolitan New York area), was shipped by TPS via rail from Bradenton, Florida to New Jersey in its unit train, i.e., specialized refrigerated rail cars owned and operated by TPS. This mode of transportation produces large cost savings. Product from TPI's Florida facilities sold to customers in the remainder of the United States is shipped form Bradenton, Florida via rail or common carrier. TPS performs much of the direct shipment of products to wholesalers for which purpose it owns a fleet of trucks. TPS bears the costs associated with the transportation of juice products from Bradenton to customer locations.

25. After arrival at TPS's Jersey City distribution facility, packages of Tropicana juice products are unloaded and stored for sale to large distributors and major consumers. Some of the juice product is transported by tractor trailer to TPS's other distribution facility in Whitestone, New York for ultimate sale in the New York City metropolitan area.

26. TPS performs both brand and trade marketing. Brand marketing consists of general advertising related to the promotion of the Tropicana name such as national television ads, event sponsorships (such as the Tropicana Bowl) and manufacturer's coupons. Trade marketing activities include funding special retail displays, grocery store advertisements and product placement (such as on store shelves and in aisle displays). TPS incurs all expenses arising in connection with its brand and trade marketing activities. Other marketing expenditures incurred by TPS include hiring advertising agencies, media production costs, syndicated market research, specific retail store couponing and air time for ads.

27. TPS does not perform any research and development; all such activities are performed and funded by TPI in Bradenton. All product liability claims, credit and collection costs and interest rate risks are borne by TPI, not by TPS. Although TPS holds inventory at its Jersey City and Whitestone distribution facilities, inventory loss is guaranteed by TPI.

28. TPS does not own any of the intangible assets related to the Tropicana products it sells. The corporate brand name and other associated trade names and trade marks (Pure Premium, Twister, Grovestand, Homestyle, Season's Best, etc.) are all owned by TPI.

29. Employees of TPS handle customer complaints.

30. TPS purchased a bottling and packaging plant in City of Industry, California in January 1990. The plant performs all the packaging and bottling of Tropicana juice product sold in California and several surrounding states and also produces juice product. The plant also packages and bottles juice for third parties.

31. TPS represents that the transfer price for juice product is the price that results in TPS earning a profit equal to one percent of TPS's gross receipts from sales to third parties. Under the formula, 100 percent of TPS's costs (for transportation, sales and distribution and marketing) are reimbursed by TPI. As previously noted, there was no written contract between TPI and TPS covering the sale and transfer price of juice products during the years at issue (see, Finding of Fact "18"). The Division agrees that the columnar spreadsheets attached to the consolidated Federal income tax returns are consistent with this pricing methodology.

32. During the audit relating to TPS's claim for refund (based upon its contention that TPS and TPI should be permitted to file on a separate basis), in response to a question as to how TPS had developed its pricing methodology, its representative indicated that he had no knowledge of how or when the pricing formula had been developed or who had developed the pricing formula, except that it was developed prior to 1979. The Division's auditors requested proof of arm's-length pricing such as a pricing study. In response TPS's representatives referred to a pricing study which was in the process of being prepared. On several occasions, the auditors requested a copy of this pricing study, but it was not until October 6, 1997 (approximately two weeks before the hearing held herein) that they were supplied with a copy of a report, dated October 6, 1997, prepared by Price Waterhouse LLP in support of TPS's claim that the transactions between TPS and TPI were at arm's length. A revised version of the report was introduced into evidence at the hearing by TPS. The pricing study was prepared in contemplation of litigation. Price Waterhouse LLP is the public accounting firm for TPI's parent, Joseph E. Seagram & Sons, Inc. After reviewing the pricing study, the Division, on October 20, 1997, issued a letter in which it affirmed its deemed denial of TPS's refund claims.

PSI'S BUSINESS

33. PSI was incorporated in 1980 and is engaged in the processing of waste orange peel into cattle feed and in the sale thereof. Henry Marchman, who is employed by the Tax Department of Joseph E. Seagrams and Sons, Inc. and who formerly was the Tax Manager at TPS from 1979 until 1996, stated that PSI was created to "carve out its feed operations into a more readily identified entity."

34. PSI owns a facility which houses certain production equipment which it uses to produce the cattle feed. The facility is located within TPI's Bradenton, Florida facility on real property leased from TPI and is known as the "feed mill". The feed mill owned by PSI commenced operation in 1973.

35. PSI operates exclusively in Florida. It does not sell anything to or buy anything from TPS.

36. PSI acquires the orange peels which it utilizes in its cattle feed production operation from TPI at no cost. Prior to the commencement of PSI's operation, TPI disposed of its waste orange peels, incurring a cost to do so. TPI buys oranges from a number of Florida growers, sometimes under long-term contracts. After they have been transported to the Bradenton plant, the oranges are graded and then are moved on a conveyor belt to the extraction room where the juice is extracted. The wet peel arrives at PSI's production facility directly from TPI on a conveyor. The manner in which the system is constructed prevents PSI from accepting peels from anyone other than TPI. Upon arrival, lime is added to the wet peel to neutralize acid contained therein and to cause the release of oils. This oil is distilled into a product called D'limonent. D'limonent is used as an industrial cleaner. Approximately ten percent of PSI's income is attributable to the sale of D'limonent.

The wet peel next enters a hammer mill where it is chopped up. Molasses is then added to cause the oil and water contained in the peel to be released. The wet peel is then pressed, dried and pelletized if for export or left in chunks if for domestic consumption.

37. PSI sells its cattle feed product to traders and agricultural brokers. During the years at issue, approximately 85 percent of PSI's product was sold for export; 15 percent was sold domestically. All of PSI's domestic customers were located in Florida. No feed was sold to customers in New York.

The price at which PSI sold its cattle feed product was determined by reference to the Amsterdam-Rotterdam futures market. The price fluctuated considerably and was largely dependent upon factors in the European market, including the number of animals, the grain situation, the corn situation and exchange rates.

38. During the years at issue, PSI produced between 135,000 and 150,000 tons of cattle feed per year. 675,000 to 750,000 tons of wet peel were required to produce these 135,000 to 150,000 tons of cattle feed. A document prepared by the former Tax Manager of TPS, Henry Marchman, set forth the exact tonnage of feed sold by PSI during the years at issue. This document which was introduced into evidence at the hearing indicates that the tax consequence which would result from requiring PSI to file a combined report would be $307,548.00 for the three years at issue. Mr. Marchman stated that dividing the entry to taxable income amount (this figure is the amount of income which has been added to the combined report as a result of the Division's proposed adjustment and was calculated to be $10,269,920.00 for the fiscal year ended July 1989, $9,624,089.00 for the fiscal year ended July 1990 and $11,274,701.00 for the fiscal year ended July 1991) by five (the number of wet tons required to yield a ton of dry feed), then dividing by the total tons of feed sold, results in a $14.00 per ton adjustment.

The market for PSI's wet peel is very limited and is restricted only to Florida. Because of the prohibitive transportation costs and the fact that the wet peel is highly perishable (it becomes rancid within 48 hours), the wet peel could not have been shipped beyond the immediate area of PSI's facility.

39. PSI directly employed the persons who were involved in the production of the cattle feed. Employees of TPI, including Penelope Durham, Manager of Citrus Commodities (during the years at issue, her title was Senior Agricultural Economist) performed a number of business functions for PSI pursuant to agreements between TPI and PSI. All nonproduction activities of PSI are performed by TPI, including freight service, payroll services, customer billing, contract negotiating and credit checking. These services were rendered pursuant to an agreement dated December 27, 1984 which required PSI to pay TPI the sum of $18,750.00 per month for these services. In 1990 and 1991, TPI charged PSI $125,000.00 per year for the services which it performed for PSI. Ms. Durham's annual salary during these years was approximately $35,000.00. Approximately 50 percent of Ms. Durham's time was spent on selling the cattle feed on behalf of PSI. By lease agreements which were also dated December 27, 1984, PSI leased certain delivery equipment used in connection with the distribution of its feed products (the rental charges were the actual billing to the final feed customer for freight) and also leased the feed mill and feed warehouse (the rent under this agreement was $1,100.00 per month, or $13,200.00 per year).

At the hearing, no evidence was introduced as to how the prices charged by TPI under the service or lease agreements were arrived at or whether such prices charged by TPI were arm's length. Ms. Durham stated that she did not know how the charges to PSI were computed and she did not know whether such charges remained constant during the years 1989 through 1991.

40. One of PSI's major expenses in its production process was heating. A letter to the Division's representative, James P. Connolly, Esq., from one of petitioner's representatives, Fred O. Marcus, Esq., dated October 9, 1997, stated that energy costs associated with running PSI's production equipment are allocated to PSI, i.e., electricity charged is based upon the volume of production (tons of feed produced). At hearing, no evidence was presented as to how the allocation was calculated or whether the cost allocated was arm's length.

41. Juice Bowl Products, Inc. ("Juice Bowl"), an unrelated processor of citrus juice, also produced wet peel. Juice Bowl did not have a feed mill and, as a result, sold its wet peel to a third party, Sun Pac.

The price paid by Sun Pac to Juice Bowl for the wet peel was negotiated by the two parties. The average price paid by Sun Pac to Juice Bowl for the year 1990 was $5.80 per ton of wet peel. Juice Bowl incurred a cost of $3.70 per ton for shipping the wet peel to Sun Pac's processing facility, since the agreement between the parties required Juice Bowl to bear the cost of transportation. The net price received by Juice Bowl for its wet peel was $2.10 per ton in 1990, which is calculated by subtracting the shipping costs incurred by Juice Bowl from the price per ton paid by Sun Pac for the wet peel. The price Juice Bowl received for wet peel remained relatively constant during the years at issue. Richard Miller, controller of Juice Bowl, indicated that most processors of oranges have peels operations. He stated that as Sun Pac's volume of production went up, its production costs would go down.

Mr. Miller stated that there would be certain advantages for a peels processor to be connected to an orange processing plant, as PSI is connected to TPI's plant, because it would save on handling and transportation costs. Juice Bowl's proceeds from its sale of wet peels to Sun Pac were reduced by $1.00 per wet ton due to Sun Pac's cost of handling of the peels. He indicated that, while it would be expensive to start up a peel processing plant, he felt that if Juice Bowl elected to do so, it could make a profit similar to that of Sun Pac.

42. Penelope Durham, TPI's Manager of Citrus Commodities (see, Finding of Fact "39") and Richard Miller, an employee of Juice Bowl, each testified that the price for peel fluctuates. Ms. Durham stated that during the years at issue, the market for wet peel was very limited and the market price was between $5.00 and $10.00 per ton of wet peel. Ms. Durham and Mr. Miller stated that, in some years, sellers of wet peel receive nothing for it. Ms. Durham acknowledged, however, that PSI has never had to destroy the peels for lack of a market. In one instance, Juice Bowl gave the owner of a cattle pasture wet peel for free in exchange for the privilege of disposing of wastewater on the owner's land.

43. At the hearing, the Division introduced the testimony of Richard Mayer, the supervising auditor for the field audit of TPS. With respect to prior audits, TPS had filed on a combined basis with TPI. Mr. Mayer stated that while performing the field audit which gave rise to the issuance of the Notice of Deficiency in this matter, no representative of TPS ever raised the issue that TPS and TPI should be filing on a separate basis.

44. With respect to the audit relating to the claims for refund, Richard Mayer, in reviewing the consolidated spreadsheets which set forth the sales figures of TPS, TPI and PSI, noted that "[w]hile TPS had 100 times the sales activity that PSI had, their bottom lines were essentially the same."

45. On the Tropicana group's consolidated Federal income tax returns (forms 1120) for the years at issue, the following information concerning TPI, TPS and PSI was set forth:

46. Of the Tropicana group's total operating expenses for the audit period, TPS incurred approximately 70% ($630 million divided by $893 million). Of the group's total taxable income for the audit period, TPS reported approximately 17% of that total ($27 million divided by $163.6 million). PSI is shown to have had .9% of the group's total operating expenses ($830,000 divided by $893 million) while it reported 19% of the group's total taxable income ($31.1 million divided by $163.6 million). TPI had approximately 29% of the group's total operating expenses for the audit period while reporting 64% of its total taxable income.

THE PRICE WATERHOUSE REPORT

We modify finding of fact "47" of the Administrative Law Judge's determination to read as follows:

47. TPS submitted into evidence a report entitled "Analysis of Intercompany Pricing Between Tropicana Products, Inc. and Tropicana Products Sales, Inc." ("the report"), prepared by the international accounting firm of Price Waterhouse LLP which was explained through the testimony of Larry L. Dildine, a partner in the firm. Mr. Dildine was retained to perform this pricing report on October 1, 1996, approximately six months after TPS filed the refund claims which are at issue in this proceeding. Mr. Dildine was qualified as an expert in economics with expertise in transfer pricing. Mr. Dildine had appeared as an expert witness in regard to section 482 of the Internal Revenue Code ("IRC § 482") in only a single case (on behalf of Altama Delta). This case involved the application of the 1968 regulations promulgated under IRC § 482, not the 1994 regulations at issue in this proceeding. Mr. Dildine testified that he has written articles on IRC § 482; however, his curriculum vitae did not list any of these articles and, upon cross examination, he could not recall the titles of any of the articles.

The report summarized the results of a transfer pricing study performed by Price Waterhouse under the supervision of Mr. Dildine. The purpose of the study was to analyze whether the relationships between TPS and TPI during the years at issue met the arm's-length standards of IRC § 482; the report concluded that the intercompany pricing between TPI and TPS was at arm's length.⁽²⁾

48. The pricing formula is not fully described in the report. It states that TPS is compensated for the functions which it performs for TPI "by a markup on its costs." Upon cross examination, Mr. Dildine stated that TPS is guaranteed reimbursement for all of its expenses, plus one percent. Pursuant to the written stipulation entered into by the parties on October 22, 1997, TPS represented that the transfer price for juice product is the price that results in TPS earning a profit equal to one percent of TPS's gross receipts from sales to third parties (under the formula 100 percent of TPS's costs were reimbursed by TPI).

When this error was brought to Mr. Dildine's attention during cross examination, he contended that it was insignificant. However, his report states that: "TPI bears the market risk for its product. Through the pricing arrangement with TPS, TPI's profit varies with fluctuation in third party sales while TPS's profit is unaffected by fluctuations in sales."

49. The report indicates that, pursuant to IRC § 482, the arm's-length principle is the standard in determining appropriate transfer prices for transactions among related entities, i.e., transactions between related parties would produce the same result if they were undertaken by unrelated parties dealing at arm's length. Treas Reg § 1.482 affirms the arm's length standard and specifies various methods of analysis to determine transfer prices. Mr. Dildine stated that all of the methods are aimed at finding comparable transactions or comparable prices or results to the transactions between the related parties at issue.

The most specific method employed under the regulations is the comparable uncontrolled price method or "CUP". This method is applicable if the same goods are traded between unrelated parties under the same circumstances. If so, then the price charged in the uncontrolled transaction would govern the price charged in the controlled transaction.

A second method which requires similarity, but not exactness in the transaction, is the comparable uncontrolled transaction or "CUT". Mr. Dildine stated that, under this method, you look at the markup on costs for a manufacturer or service provider or to the resale margin if the taxpayer is selling or reselling a product. These methods are referred to in the regulations as cost plus and resale methods. Both, however, require a comparable uncontrolled transaction.

The third method, usually applied when no clearly comparable transactions can be located, is called the comparable profits method or "CPM". Mr. Dildine stated that this method generally employs a statistical analysis to deal with variations among profit comparisons.

50. In performing his transfer analysis, Mr. Dildine first conducted a functional analysis of TPI and TPS to determine each company's functions, risks and intangible property. He visited TPI's Bradenton, Florida facilities and TPS's Whitestone, New York facility, interviewed personnel and examined public information regarding the marketplace for the products of the companies. Mr. Dildine concluded that the pricing methodology between TPI and TPS was best analyzed by evaluating TPS in comparison with other wholesale distributors of groceries and related products.

Mr. Dildine then performed an economic analysis of the transactions between TPI and TPS during the years at issue, i.e., he analyzed "the sale of finished beverage products from TPI to its related company, TPS, in which TPS has principally the functions of transportation, marketing and distribution." Mr. Dildine stated that he could find no comparable uncontrolled prices because TPI does not sell its product to any other company at the first stage for distribution in the United States market and TPS does not buy product in any significant quantity from any other supplier. Utilization of a CUP analysis was, therefore, not possible in the opinion of Mr. Dildine. When asked if he considered a CUT, Mr. Dildine replied that he looked for transactions where a company made sales of similar products under similar conditions and did find the results for two other companies that distribute Tropicana products in a region of the country, and he used that information as part of a supplementary analysis (see, Finding of Fact "66").

51. Mr. Dildine categorized TPS as a company with relatively limited risks and functions that are similar to other wholesale distributors of food products so, using the CPM method, he chose to test TPS against the profitability of other companies which do similar things, but have unrelated suppliers. He acknowledged that comparability is a critical factor in performing a proper CPM analysis. Two such CPM analyses were performed.

In selecting comparables for his first CPM test, Mr. Dildine selected for comparison companies that purchased grocery products from unrelated suppliers and compete as distributors of these products. He identified such companies using Standard & Poor's Compustat PC Plus and Disclosure's Compact SEC databases. He searched those databases for companies that were considered wholesale distributors of groceries and related products under Standard Industrial Classification ("SIC") Code No. 514. The Standard Industrial Classification System categorizes companies based on their economic activities; the companies either classify themselves or are classified by the Securities and Exchange Commission ("SEC").

The search revealed 49 public companies which were classified on one or the other of the two databases as being primarily in the business of wholesale distribution of groceries and related products. Some of these companies were eliminated from consideration. In a few cases, it was found that a company was related to its supplier, such as being a subsidiary of a larger company, and those were eliminated. Other companies were eliminated when their business description suggested that they engaged, in a significant manner, in processing or manufacturing or developing products. Some companies were misclassified and were, therefore, eliminated. Companies that had, as a major part of their business, retail grocery sales were also eliminated. A total of 14 companies were eventually identified as being comparables. Of the companies selected, four were located on the East coast (Atlantic Beverage Co., Inc. of Maryland; Krantor Corp. of New York; Richfood Holdings, Inc. of Virginia; and Sun City Industries of Florida).

The first test compared TPS's profitability to that of the 14 companies deemed comparable by Mr. Dildine, using return on sales ("ROS") as the profit level indicator. The ROS ratio is the ratio of operating profits to net sales (sales after returns and allowances). An ROS ratio was calculated on the sample of companies by using information provided from SEC filed reports and by applying that ratio for each one of the years at issue.

A median ROS and an interquartile range of ROS were computed. The "median" is the value of the ratio above and below which no more than 50 percent of the sample values are located. The "interquartile range" is the middle 50 percent of the ratio observations in the sample, containing the 25 percent of the observations immediately above and the 25 percent of the observations immediately below the median. Guidelines for determining the arm's length range are discussed in Treas Reg § 1.482-1(e)(2).

52. According to TPS's profit and loss statements, its average ROS over the three-year period was 0.85%. For the 14 companies deemed comparable by Mr. Dildine, the median ROS for the years at issue was 1.2% and the interquartile range was 0.7% to 1.8%. Because the ROS for TPS fell within the interquartile range for the sample companies, Mr. Dildine concluded that intercompany transactions between TPI and TPS during the years at issue were conducted at arm's length under the standards of IRC § 482.

53. The report states that "TPS is a wholesale distributor which also performs additional functions: transportation and marketing." In explaining the different approach taken in the second CPM analysis, the report states, "Again, we perform a CPM analysis, but this time, we account for the fact that TPS performs slightly different functions from what a typical wholesale food distributor may perform. To do so, we found comparables for each of the following functions: distribution, transportation, and marketing." At the hearing, Mr. Dildine stated that "TPS has more than the usual amount of responsibility for transportation and advertising."

54. The Tropicana group's Federal returns set forth the following expenses for TPS during the years at issue:

For each of the years at issue, TPS's transportation and marketing expenses represented more than 50 percent of its total operating expenses for the year.

55. The report states, at pages 10 and 11 thereof, the following:

There are three unique facts which distinguish TPS from other national food distributors. First, a majority of the products distributed by TPS are not-from-concentrate fresh orange juice, which has a relatively short shelf life (63 days from production) compared to other grocery products. For this reason, there is a very high inventory turnover rate, and risk of spoilage and product loss. Fresh product generally must be stored in refrigerated warehouses, transported in refrigerated trucks ("reefers") or trains, and finally stored in the refrigerated section of retail supermarkets, stores, or restaurants.

Second, there is a very high concentration of sales to customers in the northeastern U.S., with virtually all production taking place in Florida. While TPS distributes products to markets throughout the U.S., 35% of all U.S. sales are in the Northeast, with much of those sales to the metro New York area. The Tropicana brand is much stronger on the East coast of the U.S. than elsewhere in the country. While Tropicana has dominant market share on the East coast, it is weaker in other parts of the U.S.

Third, orange juice is a relatively heavy product which is expensive to transport, but which must be processed close to the source of the oranges to insure freshness and consistency. For these reasons, TPS ships products to the Northeast by rail in the TPS-owned unit train which results in large cost savings.

56. Of the companies which the report deemed as being comparable to TPS for purposes of performing the first CPM test, some were described as performing solely distribution-type activities while others also performed marketing functions. However, one company, Nash Finch Company of Minneapolis, Minnesota is described, in part, in the report as follows: "owns and operates supermarkets, warehouse stores and combination general merchandise/food stores; grows, packs, ships and markets fresh fruits and vegetables; . . . .; owns vineyards and orchards for the production of table grapes, tree fruit, kiwi and citrus." Certified Grocers of California, Ltd. was described in the report as operating dairy and bakery facilities and providing data processing, finance and insurance services to its patrons. Fleming Companies, Inc. operates a dairy facility and owns and operates supermarkets and a bakery. Roundy's, Inc. owns and operates retail warehouse food stores and operates a dairy and ice cream plant.

57. At hearing, when queried on why it was proper to use a company such as Vestro Natural Foods as a comparable despite the fact that its form 10-K⁽⁴⁾ stated that it performed baking, i.e., a manufacturing function, Mr. Dildine stated that he had investigated the data on Vestro and found that about 75 percent of its business was distribution at wholesale of natural foods and about 25 percent was baked goods, so the preponderance of its business was similar to TPS.

58. The report indicated that of the companies from the original 49 classified in SIC 514, some were eliminated because they "were in a startup situation." Mr. Dildine stated that he would reject a company which "has the characteristics of a startup, which means it has very low sales, and often would show a loss."

59. Mr. Dildine stated that a company which performed a buyout, i.e., it acquired some additional businesses, would not be an appropriate comparable.

60. The report's description of the companies used as comparables does not provide any information as to how long any of these companies were in business or whether they had undergone any recent structural changes.

61. The form 10-K filed by Krantor Corporation for the year 1991 states as follows:

Comparative results from 1990 to 1989 are not very meaningful as 1989 was not a full year of operation, and was the start up year of the Company. The 1989 loss of $297,679 ($.11 per share) is attributable to developing the Company's customer base, marketing the Company to the industry, as well as finding funding sources for the Company's activities. 1990 was an extension of that development; thus the losses continued until the Company developed its current base of accounts.

62. The form 10-K filed by Vestro Foods, Inc. indicates that it began its program of acquiring specialty food companies during 1987 as a result of an equity infusion of $5,000,000.00 by a group of institutional investors. Effective August 1, 1987, a wholly-owned subsidiary of Vestro acquired the assets of Heidi's Pastry, Inc. (this subsidiary is the foundation of the company's Fine Baked Products core). Effective January 1, 1988, Westbrae Natural Foods, Inc. merged with and into a wholly-owned subsidiary of Vestro. This subsidiary was the first in the company's Natural Foods core. Effective September 29, 1989, Vestro acquired Little Bear Organic Foods, Inc., a national marketer of organic snack foods. Effective May 1, 1990, Vestro acquired Jan Holzmeister Cheesecake, Ltd., a leading supplier of New York style cheesecake and other cakes and pies to the institutional foodservice market. The financial statements contained in the 1990 form 10-K indicate that the results for the years 1987 through 1990 were affected by these acquisitions.

The form 10-K states that operating expenses of Vestro grew by $1,829,000.00 from the prior year, primarily due to Westbrae Natural Foods, Inc. (acquired in 1988):

where a new management team was installed after the operation was moved from Northern California to the Los Angeles area. The Company's strategy has been to invest in substantial management resources for each operating company with the capability to grow it to significantly higher levels. The Company's experience is that this investment raises expense levels in the first year after acquisition prior to commensurate increase in sales and gross profit.

63. With respect to Atlantic Beverage Company, Inc., Mr. Dildine stated that because the company had, according to its financial statement filed with the SEC, acquired some additional businesses during the year 1991, that year was not included because 1991 was not comparable to results in previous years. Despite these statements, however, the report did utilize Atlantic's ROS (return on sales) for the year 1991 in its calculations.

64. Two of the comparables, Certified Grocers of California, Ltd. and Roundy's, Inc., are described in the report as being cooperatives which distribute food products and nonfood items to supermarkets. During cross examination, Mr. Dildine testified that he did not know whether a third comparable, United Grocers, Inc., was a co-op.

Mr. Dildine stated that a purchasing co-op "would be one that is related to its customers. It buys on their behalf." When asked if a co-op necessarily maximizes profit, he replied "[t]hat depends on what the arrangement is." He stated that he did nothing to determine whether any kind of cooperative agreement existed between the co-op and its shareholders because "we have no access to agreements between those parties. That would be confidential information."

In response to a question as to whether it would be possible for a purchasing co-op to agree to some allowances or rebates in regard to pricing, Mr. Dildine replied, "Again, there are very many different arrangements for compensation of those kinds of businesses."

The report indicates that Mr. Dildine rejected the use of a company (KFC Natl Purch Co-op), which was described in the report as "coop makes purchases of equip & food for KFC & Taco Bell." According to the report, the reason for its rejection as a comparable for purposes of the first CPM study was that it was a co-op. At the hearing, Mr. Dildine testified that this entry was in error; the reason for its rejection was because of its relationship to the supplier.

65. The report indicates that, "[t]o confirm the arm's length nature of the intercompany transactions between TPI and TPS," a second CPM test was performed, "but this time, we account for the fact that TPS performs slightly different functions from what a typical wholesale food distributor may perform." For purposes of this analysis, Mr. Dildine attempted to evaluate TPS as a service provider for each of the services which TPS provided for TPI, to wit: distribution, transportation and marketing services.

The second CPM test took TPS's costs for providing each of the three services, multiplied by a markup which was derived from groups of what Mr. Dildine considered to be comparable businesses performing each of these services and arrived at what the report labeled as "the minimum arm's length profit that should be earned by TPS for these three services." The report refers to the product of the costs for the services times the markup as the "arm's length profit." Mr. Dildine stated that second analysis was "somewhat unorthodox."

66. As to the distribution function, the report indicates that Mr. Dildine searched for transactions between TPI or TPS and unrelated parties for use as comparables to the intercompany transactions to be analyzed. The report acknowledges that TPS does sell Tropicana product to unrelated distributors which provide similar services as those provided by TPS to TPI. Mr. Dildine was able to obtain financial information for two of these distributors. The names of these companies were provided by TPS.

The first distributor, B.K. Miller, is an unrelated company which purchases Tropicana products in large quantities, breaks down the products into smaller quantities and resells the products to wholesalers and other distributors in the Baltimore-Washington, D.C. area. B.K. Miller also sells Tropicana products to wholesalers whose sales volume is not large enough to purchase directly from TPS in the minimum quantities required. B.K. Miller's financial statements were obtained from TPS. These financial statements were provided to TPS for two of the years at issue, 1989 and 1990. Despite the fact that it was a private company and, therefore, was not required to file a report with the SEC, B.K. Miller provided the financial statements to TPS as part of its distribution relationship. No financial statements were provide to TPS for 1991. Mr. Dildine stated that he had done nothing to verify that the financial statements were, in fact, certified, audited financial statements. He also stated that his report did not utilize privately-held companies "because that information, when it can be obtained, is unaudited information, and not generally as reliable as information from public companies for which you would have reports and financial reports. That's common practice."

The second distributor, Atlantic Beverage Co., Inc., is also an unrelated distributor which purchases product from TPS and sell to other distributors in the metro Baltimore and Washington, D.C. areas.

67. Based on data for the 1989-1990 period, Mr. Dildine computed a median markup on total costs and an interquartile range of markup on total costs for B.K. Miller and Atlantic Beverage Co., Inc. The median markup on total costs for the two companies was 5.0% and the interquartile range of markup on total costs was 3.1% to 7.0%. Mr. Dildine was unable to obtain data for 1991.

68. For the transportation portion of the second CPM test, Mr. Dildine attempted to develop a sample of independent companies in the category of freight forwarders; he searched for companies classified in SIC 4731, Transportation of Freight and Cargo, which includes agents in arranging transportation for freight and cargo. It also includes freight forwarders, which undertake the transportation of goods from the shippers to receivers for a charge covering the entire transportation, making use of the services of other transportation establishments to effect delivery. The report notes that "[t[his description matches very closely the transportation function performed by TPS." At the hearing, upon cross examination, Mr. Dildine acknowledged that a freight forwarder arranges transportation for customers, but normally does not provide the actual transportation. He further agreed that TPS did, in fact, own trucks and rail cars used to transport Tropicana product. To the question of whether TPS would be less comparable to a freight forwarder by virtue of the fact that it owned transportation assets, Mr. Dildine replied that "[t]here would be a difference." When asked what he did to verify what transportation assets TPS has, in order to make a comparison with a freight forwarder, Mr. Dildine's response was "nothing specifically."

Using Standard & Poor's Compustat PC Plus and Disclosure's Compact SEC databases, Mr. Dildine's search produced an initial sample of 17 companies in the SIC 4731 category. He eliminated several companies from the initial sample (for the reasons that they were controlled subsidiaries, were foreign based, were in a startup situation, were engaged in significantly different lines of business or had insufficient data) and was left with a sample of ten companies. For the years at issue, Mr. Dildine then computed a median markup on total costs and an interquartile range of markup on total costs for the freight forwarders which he deemed comparable. The median markup on total costs for the ten freight forwarders was 2.6% and the interquartile range of markup on total costs was 0.3% to 5.6%.

69. For the marketing services function, Mr. Dildine searched on CD-ROM databases containing financial information and business descriptions for U.S. companies, Standard & Poor's Compustat PC Plus and Disclosure's Compact SEC. He searched for companies classified in SIC 7311, Advertising Agencies, which includes companies engaged in preparing advertising and placing such advertising in various media for clients on a fee or contract basis. An initial sample of 27 companies was found. Companies were eliminated for the same reasons as the transportation sample (see, Finding of Fact "68"). The remaining seven companies were deemed comparable.

For the years at issue, a median markup on total costs (8.3%) and an interquartile range of markup on total costs (7.6% to 11.2%) was computed for the advertising agencies which Mr. Dildine determined to be comparable.

70. Mr. Dildine stated that the calculations in the second CPM analysis were performed to determine the amount it would have cost TPI to compensate unrelated companies (based on a markup of total costs) on an arm's-length basis for the distribution, transportation and marketing services which it received from TPS during the years at issue. According to Mr. Dildine's calculations, TPI would have had to pay an unrelated marketing company at least 7.6% over its costs, that number being the lower limit of the interquartile range for the sample found for advertising agencies.

The report found that TPI would have had to pay an unrelated transportation company at least 0.3% over its costs, the lower limit of the interquartile range for the sample freight forwarders.

For the distribution companies, since there were only two companies deemed comparable in his analysis, Mr. Dildine stated that he could not calculate an interquartile range; therefore, he took the average (or median) between the two companies. The median was calculated to be 5.0% over costs.

The results of Mr. Dildine's computation of markup on total costs for the companies for 1989 -1991, utilized in the second CPM analysis are set forth in the following table:

71. The appropriate return for the various functions performed by TPS were calculated by Mr. Dildine and are summarized in the following table:⁽⁵⁾

72. In order to determine whether TPS reported a profit which was at least as high as the minimum arm's-length constructive profits of the distribution, transportation and marketing companies which he deemed to have been comparable, Mr. Dildine applied the arm's-length markup on total costs of the comparables to TPS's costs for its own distribution, transportation and marketing activities during the years at issue. As set forth in the table (see, Finding of Fact "71"), by applying the arm's-length markup on total costs computed for the comparable companies to TPS's actual costs, TPS's average profit for the years at issue would have been $9.7 million. The actual profit reported by TPS for this period was $10.1 million. Because TPS's actual profits exceeded the profits which it would have earned using the arm's-length markup on total costs for the sample companies, Mr. Dildine concluded that, under the second CPM analysis, the prices between TPI and TPS during the years at issue were at arm's length and that the second analysis confirmed the results of the first CPM analysis.

73. When questioned as to the source of the cost figures for "brand marketing" and "transportation" for TPS, as set forth in Mr. Dildine's report (these figures are contained in the table in Finding of Fact "71"), he stated that they came directly from the Federal tax return of TPS. As to the "distribution" cost figure, Mr. Dildine stated that it was from a residual category, calculated by subtracting the transportation and advertising expenses from the total operating expenses in TPS's financial report. It is apparent that this method of calculating distribution costs results in such expenses noted on the Federal return as wages and salaries, depreciation and rent being treated as "distribution" costs in the second CPM test.

74. The "Executive Summary" portion of the report describes the profit level indicator used in the second CPM test as markup on total costs ("MOTC"). The table in Appendix A of the report which sets forth the financial information for the two companies used in the distribution portion of the test, i.e., B.K. Miller Co., Inc. and Atlantic Beverage Co., Inc., states that markup on total expenses was calculated as operating income divided by total expenses. In fact, the markup calculation shown in the table was computed by dividing operating income by operating expenses, not total expenses, meaning that the test excludes cost of goods sold.

THE DETERMINATION OF THE ADMINISTRATIVE LAW JUDGE

The Administrative Law Judge framed the issue in this matter as whether the Division properly required petitioner to file on a combined basis with its parent (TPI) and sister company (PSI) in order to properly reflect petitioner's tax liability under Article 9-A of the Tax Law. The regulations require three conditions prior to allowing or demanding combination: a stock ownership test; a unitary business test; and a distortion of income test. The first two were conceded and the third is at the very heart of this action. Given the substantial intercorporate transactions between petitioner and TPI there is a presumption of distortion when the taxpayer files on a separate basis. This presumption can be rebutted by a taxpayer by a showing that the transactions were at arm's length (Matter of Standard Mfg. Co., Tax Appeals Tribunal, February 6, 1992). The Tribunal went further in Matter of Silver King Broadcasting of N.J. (Tax Appeals Tribunal, May 9, 1996) when it stated that it was not enough for the Division of Taxation to merely identify possible areas of distortion, but it must identify with particularity the activities or transactions which give rise to the distortion and how such activities produce distortion.

The Tax Appeals Tribunal has relied upon the principles of IRC § 482 and the regulations promulgated thereunder to determine whether a taxpayer has rebutted the presumption of distortion which resulted from substantial intercorporate transactions (Matter of Sears Roebuck & Co., Tax Appeals Tribunal, April 28, 1994). The regulations promulgated under IRC § 482 require that any method of testing arm's length transactions be reliable and take into account the degree of comparability between the controlled transaction (between related parties) and those of uncontrolled comparables and the quality of the data and assumptions used in the analysis. The regulations dictate that uncontrolled comparables must be similar to the controlled transaction and, if they differ materially, adjustments must be made to improve the reliability of the results, if possible. The Administrative Law Judge examined the report submitted by petitioner in light of the regulations and determined that the tests employed in the report were flawed and, therefore, insufficient to rebut the presumption of distortion.

The first test performed was a comparable profits method which used return on sales as the profit indicator. The comparable companies chosen were wholesale distributors of groceries and related products which were also involved in growing, manufacturing and other forms of production. In addition, companies in the start-up stages of operation and those with recent acquisitions (which presumably had an effect on sales) were included in the comparables. No adjustments were made by petitioner as prescribed by the section 482 regulations even though there were material differences between petitioner and the comparables. The first test also utilized purchasing co-ops which are related to its customers and may not seek to maximize profits and only four of the comparables operated in an East coast market. Finally, the factor the Administrative Law Judge found most damning was that the first comparative profit method focused on petitioner's distribution function only, ignoring the significant transportation and marketing functions. Although petitioner's report established an interquartile range for the return on sales ratios of the comparables and that TPI's return on sales was within this range (equating to arm's length transactions), the Administrative Law Judge found that the report ignored the functional comparability and violated the section 482 regulation (Treas. Reg. § 1-4825[b][4][ii]) which noted that closer functional comparability normally is required under a financial ratio like return on sales.

The second comparable profits method used evaluated petitioner as a provider of distribution, marketing and transportation services to TPI. Here, petitioner determined the cost for providing these services and multiplied it by a markup derived from comparable businesses to arrive at a minimum arm's length profit on the performance of these services. The comparables chosen for the distribution function operated in a different market than petitioner and the distribution costs of petitioner were derived from total operating expenses using a residual figure, that which was remaining after subtracting costs for transportation and advertising expenses. The remaining figure included wages, salaries, depreciation and rent. The transportation portion of the test used freight forwarders as comparables which did not provide transportation, only arranged for it. This was distinguishable from petitioner which owned transportation vehicles and transported goods. Finally, the marketing function was tested against comparables consisting of advertising agencies. Petitioner computed a median markup on total costs and an interquartile range of markup on total costs for advertising.

Petitioner then used the comparable values to estimate an arm's length markup on total costs. This was applied to petitioner's actual costs and resulted in an average profit figure which was lower than the actual profit reported. Petitioner claimed this was proof that its pricing with TPI was at arm's length.

The Administrative Law Judge found that the second test utilized two profit indicators which flipped between operating and total expenses in the denominator of the markup calculation for the distribution, transportation and advertising functions. The Administrative Law Judge determined that this was an additional flaw in the second test which merited dismissing it. The Administrative Law Judge also determined that it was unacceptable that petitioner chose to compare itself to three separate types of businesses with functions similar to those found in petitioner's business operation instead of businesses which also contained the three functions. The Administrative Law Judge perceived this division to distort the results obtained and failed to support petitioner's conclusions drawn from its analysis.

Regarding petitioner's argument that PSI should not have been included in a combined report with it based on the argument that TPI and PSI also have pricing policies which result in arm's length transactions, the Administrative Law Judge found that there was a market, albeit limited, for the wet peel which was given to PSI which indicated that the transaction with TPI was not at arm's length. In addition, the services performed by TPI for PSI were estimated to be $18,750.00 a month, or $250,000.00 a year. Yet in 1990 and 1991, the amount charged was only $125,000.00. No reason for the lower amount was given. Leases for equipment and rental agreements between PSI and TPI were not accompanied by any evidence that they were negotiated or coincidentally arrived at arm's length prices. In sum, as stated by the Administrative Law Judge:

PSI received wet peel from TPI at no cost. It received from TPI considerable administrative services, certain equipment, use of facilities and energy to run production equipment. [The wet peel had a market price of between $5.00 and $10.00 a ton.] Clearly, TPI's transfer of the peel at no charge cannot be found to be at arm's length. With respect to the services, equipment and facilities provided by TPI to PSI, there has been no evidence presented from which it can be reasonably concluded that the charges to PSI were consistent with what would have been charged by TPI if it had been dealing at arm's length with an uncontrolled party (Determination, conclusion of law "J").

Finally, the Administrative Law Judge rejected petitioner's contention that requiring combined filing in this matter would be a violation of the Due Process Clause and the Commerce Clause of the United States Constitution because New York would be taxing income of TPI and PSI that was earned outside of its borders. Relying on Mobil Oil Corp. v. Commissioner of Taxes of Vermont (445 US 425, 63 L Ed 2d 510), the Administrative Law Judge found that a corporations's income may be fairly apportioned among the states for tax purposes by formulas which utilize in-state aspects of interstate affairs, and that there need be only minimal connections between the interstate activities and the taxing state and a rational relationship between the income attributed to the state and the intrastate values of the enterprise. The Administrative Law Judge found that petitioner exploited New York markets for substantial profit and, in return, New York's taxation bears a relationship to the benefits and opportunities it bestowed on the unitary group.

The Administrative Law Judge dismissed petitioner's Commerce Clause argument, stating that no reason was given why Florida should have sole authority to tax the unitary group given the fact that part of the group's business was conducted in other states. Citing Container Corp. of America v. Franchise Tax Bd. (463 US 159, 77 L Ed 2d 545), the Administrative Law Judge noted that the Constitution does not prohibit a state from implementing a formula which apportions income of a unitary business merely because it results in taxation of some of the income that did not have its source in the taxing state. A taxpayer must show by clear and cogent evidence that the state tax results in extraterritorial values being taxed in order to take advantage of Commerce Clause protection (Exxon Corp. v. Wisconsin Dept. of Revenue, 447 US 207, 65 L Ed 2d 66). Since petitioner did not submit such evidence herein, its Commerce Clause argument failed.

ARGUMENTS ON EXCEPTION

Petitioner argues that, based on the Price Waterhouse report and the testimony of its expert witness, it has established that the intercompany pricing between petitioner and its parent was at arm's length and in accord with the principles of IRC § 482 and the regulations thereunder. Petitioner contends that the proof of arm's length pricing rebuts the presumption of distortion of its New York income arising from the substantial intercorporate transactions and demonstrates that it should not be required to file on a combined basis with its parent.

Petitioner takes issue with the Administrative Law Judge's delineation of perceived flaws with the comparables chosen by Price Waterhouse for use in its report, arguing that the Administrative Law Judge had overlooked other attributes which made the comparables even better than initially portrayed. In addition, petitioner believes the Administrative Law Judge went beyond the scope of section 482 and the regulations in analyzing the ancillary characteristics of the comparables chosen by Price Waterhouse, which it claims properly compared the economically significant functions of the controlled and uncontrolled taxpayers (Treas. Reg. § 1.482-1[d][3][i]). In addition, petitioner claims that the Treasury is willing to tolerate functional diversity between comparables (Treas. Reg. § 1.482-5[c][2][ii]), and that the degree of functional comparability required to obtain a reliable result under the comparable profits method is less than the degree of functional comparability required under other pricing methodologies. Petitioner also takes issue with the Administrative Law Judge's criticism of using a comparable which is a "start-up" or newly acquired company (Krantor). Petitioner infers from the Administrative Law Judge's criticism of Krantor that any comparable not showing a profit must be excluded from consideration. Petitioner argues that such a theory "stands Section 482 regulations on its head."

Petitioner disagrees with the Administrative Law Judge's criticism that petitioner's expert made no adjustments to account for any of the analytical deficiencies the Administrative Law Judge identified. Petitioner does not believe that there were any material differences between the comparables and itself that would materially affect the measure of an arm's length result under the chosen pricing method. Petitioner, rather, claims that the Administrative Law Judge presumed that every perceivable difference was material and that adjustments should have been made. In the alternative, petitioner states that the expert made adjustments by applying an interquartile range as prescribed by Treasury Regulation § 1.482-(e)(2)(iii)(B) and (C).

Petitioner argues that the Administrative Law Judge erred in placing too much emphasis on the difference in geographic markets between the controlled and uncontrolled companies, citing the regulation which says that it is but one of eight factors to weigh (see, Treas. Reg. § 1.482-1[d][3][iv]). Petitioner notes that there is no evidence that the other geographic markets were any different than the East coast market where petitioner operated. Also, petitioner believes the interquartile range accounted for the difference in the markets (Matter of Silver King Broadcasting of N.J., supra; Matter of Campbell Sales Co., Tax Appeals Tribunal, December 2, 1993).

Petitioner also argues that the Administrative Law Judge's determination erred in finding the failure to account for petitioner's other activities as a critical flaw. Petitioner notes that the Administrative Law Judge earlier criticized the report for not accounting for ancillary activities of the uncontrolled comparables but later criticizes the report for not accounting for the fact that petitioner is not a "pure distributor." Petitioner maintains that the report did not ignore the ancillary functions but placed its focus on the economically significant activities of the controlled and uncontrolled parties.

The supplemental analysis, performed to confirm the results reached on the first comparative profits methodology, looked at the three components of petitioner's business, evaluating each function separately, i.e., distribution, transportation and marketing.

Petitioner takes issue with the Administrative Law Judge's criticism of using the Washington, D.C./Baltimore market and its computation of distribution costs by using a residual rather than the actual expense data. Petitioner believed the Washington, D.C./Baltimore market was adequate and that the residual was a reasonable approximation of actual expenses. Petitioner constructed a table to show how a reallocation of the distribution expenses by 10% to marketing and transportation actually enhances the results found by the Price Waterhouse report, adding further support to its conclusions.

Petitioner contends that it does, in fact, operate like the freight forwarders it has been compared to in the report and not the "in-house" transportation company portrayed by the Administrative Law Judge.

Petitioner also disagrees with the Administrative Law Judge's criticism that the profit markup for distribution was computed using the ratio of operating income to operating expenses while the markup for transportation and marketing ratios used operating income to total expenses. The latter ratio does not include cost of goods sold because they traditionally do not buy or sell goods. Therefore, petitioner contends that the denominators are actually the same.

Finally, petitioner argues that the Administrative Law Judge was not fair in his criticism of the way in which the report utilized a comparison of the distribution function and then confirmed its accuracy using a comparison of the three separate functions of distribution, transportation and marketing.

Petitioner maintains that, although this Tribunal is not bound to accept the uncontroverted testimony of petitioner's witness, and that transfer pricing is not an exact science, it should accept the conclusions of the Price Waterhouse report because it presents the only reasonable conclusion.

Regarding petitioner's claim that PSI should not be forced to file on a combined basis with TPI, petitioner takes issue with the determination's conclusion that the wet peel had value during the years in issue and points to the evidence it introduced to prove this fact. In addition, the service fees paid to TPI by PSI were so insignificant that they should not rise to the level of substantial intercompany transactions. Further, petitioner asks how the service fees between TPI and PSI contributed to the distortion of petitioner's income from sources within New York. Finally, petitioner argues that the $14.00 per wet ton price arrived at in the determination is unreasonable especially in light of the Administrative Law Judge's willingness to accept such a high figure to compensate for TPI's forgiveness of service fees and rental receipts during the years in issue, a conclusion which petitioner claims would yield a value for the fees and receipts of $2,968,440.00, if taken to its logical conclusion. Petitioner states that this underscores the error in the methodology and the lack of anything in the record to support such a result.

Petitioner contends that the determination was wrong in its conclusion that requiring combined reporting with TPI and PSI was not violative of the Commerce Clause and Due Process Clause of the United States Constitution because the determination did not identify a rational relationship between petitioner's income from sources in New York and the out-of-state activities of TPI and PSI. Further, the determination did not explain how the required combination would not result in the taxation by New York of income earned beyond the State's borders.

The Division concurs with the conclusions of the Administrative Law Judge. Particularly, the Division contends that the Price Waterhouse report is insufficient evidence that the intercompany pricing was arm's length because it does not adhere to the section 482 regulations and is internally contradictory. The Division argues that the use of purchasing co-ops was inconsistent with the 1994 regulations; that the use of comparables in a start-up mode or which had recently made a major acquisition was inconsistent with the section 482 regulations; and that the report ignored all but the principal functions of the controlled and uncontrolled parties.

The Division also contends that petitioner should be required to include PSI in the TPS-TPI combined group because they are all part of the same unitary business and not including PSI would distort petitioner's income and activities in New York. The Division maintains that petitioner has not established that the transactions between PSI and TPI were at arm's length or that by requiring combined reporting with PSI the Division is creating more distortion than it seeks to cure.

Petitioner argued in response that its comparable profits method analysis produced a reliable measure of arm's length transactions and that the report properly chose and classified comparables, including purchasing cooperatives and grocery distributors (Krantor and Vestro). In addition, petitioner believes that the companies chosen for its report were functionally comparable to petitioner and were chosen consistent with the Treasury Regulations. Any adjustments deemed necessary were taken into account by means of an interquartile range, which increased the reliability of the transfer pricing study and which statistical method is expressly suggested by the regulations (Treas. Reg. § 1.482-1[e][2][B]).

Petitioner defends its "unorthodox" second transfer pricing study, stating that it is conceptually sound, taking into account the other aspects of petitioner's business which were missed by the first test. Petitioner defended its use of freight forwarders as similar to petitioner in its transportation function.

Finally, petitioner maintains that accepting the Division's position would "raise the bar" for petitioner in the area of required combination and the interpretation of Treasury Regulations regarding section 482. It cites this Tribunal's cases in Matter of Silver King Broadcasting of N.J. (supra), Matter of New York Times Co. (Tax Appeals Tribunal, August 10, 1995) and Matter of Campbell Sales Co. (supra) as the accepted standard and contends that accepting the Administrative Law Judge's conclusions would effectively change our precedent and create an unattainable standard. In addition, petitioner cites Bausch & Lomb v. Commissioner (92 T.C. 525, affd 933 F2d 1084) in support of its interpretation and application of section 482 and the regulations thereunder.

OPINION

In Matter of Campbell Sales Co. (supra), we very succinctly set forth how the presumption of distortion arises in cases like this:

We are presented with the issue of whether petitioner has overcome the presumption existing in this case that its filing of a separate corporate franchise tax report under Article 9-A of the Tax Law results in an improper reflection of its New York tax liability.

Tax Law § 211(4), which is the governing statute in this dispute, provides in part:

[i]n the discretion of the [Division], any taxpayer, . . . substantially all the capital stock of which is owned or controlled either directly or indirectly by one or more other corporations . . . may be required or permitted to make a report on a combined basis covering any such other corporations . . .; provided, further, that no combined report covering any corporation not a taxpayer shall be required unless the [Division] deems such a report necessary, because of inter-company transactions or some agreement, understanding, arrangement or transaction referred to in [Tax Law § 211(5)], in order to properly reflect the tax liability under this article (emphasis added).

We have held where a taxpayer corporation and related corporations are found to meet the requirements⁽⁶⁾ set forth in the Division's regulations interpreting section 211(4), a rebuttable presumption arises that the taxpayer corporation's New York income will not be properly reflected without reporting on a combined basis (Matter of Standard Mfg. Co., supra). This is commonly referred to in the case law and regulations as "a presumption of distortion." If this presumption arises, the burden is on the taxpayer to overcome this presumption by establishing that separate reporting results in the proper reflection of income (Tax Law § 211[4]; Matter of Standard Mfg. Co., supra). Because petitioner has conceded that the requirements necessary to give rise to the presumption are present during the years at issue, we must only address whether petitioner has provided evidence sufficient to overcome this presumption.

Ultimately, the issue of whether a petitioner has rebutted the presumption of distortion is a question of fact, and petitioner bears the burden of proof. We conclude that petitioner has not met its burden herein and affirm the determination of the Administrative Law Judge.

Petitioner has attempted to rebut the presumption of distortion with the testimony and report of its expert, Lawrence Dildene, a partner at Price Waterhouse. Mr. Dildene was qualified as an expert in economics and transfer pricing, however, he had never testified in a proceeding which dealt with the regulations before us in this case, could not relate a single article he had written regarding IRC § 482 (nor did his curriculum vitae list any), and worked for the same firm which was employed by Tropicana's parent (Joseph E. Seagram and Sons, Inc.) as its public accounting firm. With these shortcomings in mind, we shall review his report.

Mr. Dildene's report was dated October 6, 1997, three years after the Notice of Deficiency was issued (April 11, 1994) in this matter, and was provided to the Division for the first time on the same day. It was the first time a transfer pricing study had been performed for TPS. Mr. Dildene was hired to do the study on October 1, 1996, some six months after the refund claims were filed. Prepared specifically for this litigation, the report sought to determine if the transfer pricing between TPI and TPS was arm's length.

We pause here to review the facts surrounding the pricing policy. There was no written contract or other evidence of how or when the pricing policy was developed, except that it was implemented prior to 1979. Hence, pursuant to the policy as represented by TPS, the transfer price for juice product is the price that results in TPS earning a profit equal to 1% of TPS's gross receipts from sales to third parties. Under the formula, 100% of TPS's expenses are reimbursed by TPI. In addition, TPI provided a number of services centrally to TPS, including pension, administration, payroll, accounting and tax return preparation.

Also, there was no evidence that the policy was created with the intention of being at arm's length and we are not convinced that such a policy could be created prior to 1979 and still retain its arm's length character for more than a decade, without modification, revision or replacement to account for fluctuations in the economy and changes within the corporations involved. Therefore, close scrutiny of the section 482 analysis is warranted to confirm the arm's length nature of the pricing policy and that the report is not merely an effort to defend the policy.

Internal Revenue Code § 482 is simple enough in scope:  it allows the Secretary of the Treasury to apportion income among organizations to reflect clearly the income of those organizations, where the organizations are owned or controlled directly by the same interests. The regulations provide the mechanics for determining whether the income is clearly reflected, including methods of determining the arm's length nature of prices charged between such related entities. The provisions of Tax Law § 211(4) invoke the importance of inter-company transactions where the Division seeks to have one or more companies report on a combined basis, which is why this Tribunal adopted the methodologies and analyzes set forth in the section 482 regulations in Matter of Campbell Sales Co. (supra [where we applied the comparable uncontrolled price ("CUP") method and determined that Campbell's services and charges were indistinguishable from those provided by independent brokers, thus, successfully rebutting the presumption of distortion]).

In this matter, petitioner's expert and the report underlying his opinion relied on an analysis rooted in section 482 as well. The economic analysis performed was targeted to determine the appropriate compensation to TPS for its services provided to TPI in the distribution and sale of Tropicana Products. It is noted that petitioner's expert report stated that petitioner had minimal risks in its arrangement with TPI because it was guaranteed a markup on "all its expenses incurred." As noted above, the actual terms of the arrangement between TPI and TPS guaranteed a profit equal to one percent of gross receipts. Petitioner's expert dismissed this difference as inconsequential. Although this figure could be expressed as a percentage of expenses, it is not an accurate portrayal of the pricing policy, which is the focus of the Dildene report. In fact, the report never describes the pricing policy in detail.

One of the grounds for petitioner's exception to the determination of the Administrative Law Judge was that the Judge raised the bar for establishing arm's length pricing by rejecting the expert's report because of numerous inconsistencies, both internal and external, without any controverting expert evidence. We disagree with this argument because a section 482 standard is no different than any other legal standard applied to the facts of a matter before an Administrative Law Judge. The Administrative Law Judge may, as he did in this case, look at the facts as presented, and decide whether petitioner has met its burden. In this matter, we believe the Administrative Law Judge did just that and properly analyzed the Dildene report and testimony against the section 482 standard. Interestingly, petitioner's only comment on cases before us where the Division did not produce a controverting expert, was to acknowledge that we have no obligation to accept uncritically the uncontroverted testimony of a petitioner's expert testimony (Matter of Medtronic, Tax Appeals Tribunal, September 23, 1993).

Even so, petitioner was compelled to argue that in the absence of expert assistance and "the disjointed results of half-informed questions asked by the Division's counsel" (Petitioner's brief, pp. 48-49) only petitioner's expert and his report remain as a basis for our decision on arm's length transfer pricing. We note that petitioner's remark in its brief, while supportive of its position, is rather disingenuous. After all, it spent a year preparing a pricing report to justify its arm's length transfer pricing policy at hearing, but it did not provide the Division with a copy until 10 days before hearing. However, it is neither the quality of the Division's cross-examination nor the failure of the Division to present expert testimony which is the cause of petitioner's dilemma herein.

The essence of the basis for rejection of the report by the Administrative Law Judge is the selection of comparables. Clearly, if comparables are not carefully chosen, no adjustments or statistical methodologies will form a basis for accepting the expert's report as indicative of arm's length pricing. What is and is not "comparable" is subject to interpretation. The courts have cautioned the Internal Revenue Service against interpreting the term "comparable" as "identical" (see, Bausch & Lomb v. Commissioner, 933 F2d 1084, 1091). Based on the Administrative Law Judge's analysis and our review of the report, the companies chosen were not comparable. Further, we believe the Administrative Law Judge's analysis of the comparables was accurate and, therefore, confirm his conclusion that the report does not rise to the level required to rebut the presumption of distortion.

Treasury Regulation § 1.482-1(d) discusses the concept of comparability generally, noting that the comparability of transactions and circumstances must be evaluated considering all factors which could affect prices or profits including functions, contractual terms, risks, economic conditions and property or services (Treas. Reg. § 1.482-1[d][1]). As petitioner has cautioned and we stated above, the uncontrolled transaction need not be identical, only sufficiently similar so that it can render a reliable measure of an arm's length result (Treas. Reg. § 1.482-1[d][2]). However, if there is a material difference between the controlled and uncontrolled transactions, adjustments must be made to improve the reliability of the results. If adjustments for material differences cannot be made (material differences are those that affect the measure of an arm's length result), the uncontrolled transaction can still be used but the reliability of the analysis is reduced. Petitioner appears to argue that any company with a similar function is sufficient, regardless of the surrounding circumstances and additional functions performed, and that use of a statistical methodology (like an interquartile range) will eliminate material differences. We believe petitioner is in error. If several comparables are so flawed that they can be termed materially different, then they will necessarily skew the median (the chosen measure of central tendency) and the interquartile range (the chosen measure of dispersion or variance). As it was, petitioner's calculated interquartile range, or range of arm's length profitability under the comparative profits method, only narrowly included petitioner and elimination of one or more of the 14 chosen comparables would have changed both the median and interquartile range.

Specifically, in the first comparable profits method analysis, Mr. Dildene's report sought to test the arm's length analysis of the return TPS earns for providing its services to TPI. In this test, TPS's profitability was compared to independent companies. Profitability was tested by two methods: overall profitability (operating profit relative to sales) and a profitability of the three functions of transportation, distribution and marketing. The first test looked for comparables which had functions similar to petitioner. The second test looked for comparables which performed only segments of petitioner's business, to wit: distribution, transportation and marketing.

For the first CPM test, petitioner searched several databases for companies which distributed groceries and related products. This resulted in 49 responses, of which only 14 were selected for comparison. The others were eliminated because they were controlled, were foreign-based, had significant retail operations, had manufacturing operations, had significant intangibles, were in a startup situation, were involved in significantly different lines of business or had insufficient data. The CPM determines if the pricing was arm's length by comparing profit level indicators. In this matter, the return on sales ratio was chosen by petitioner because it can be effective when the companies compared have similar business characteristics, such as functions, sales volume, risk, and intangible assets. The return on sales compares the net sales to operating income. Petitioner calculated the return on sales ratio for the 14 comparables and then immediately applied the statistical adjustment provided for in the regulations to eliminate the outliers in the top and bottom 25%, leaving an interquartile range within which petitioner's return on sales fell. Based on this calculation, petitioner believes it has established an arm's length pricing policy between TPI and TPS.

But the regulations caution that the extent and reliability of any adjustments will affect the relative reliability of the analysis (Treas. Reg. § 1.482-1[d][2]). We cannot overlook the deficiencies found by the Administrative Law Judge in several of the comparables and accept petitioner's argument that a statistical methodology will cure the ills resulting from its failure to make adjustments for material differences. Comparables should be included in the arm's length range when "the information on the controlled transaction and the uncontrolled comparables is sufficiently complete that it is likely that all the material differences have been identified, each such difference has a definite and reasonably ascertainable effect on price or profit, and an adjustment is made to eliminate the effect of each such difference" (Treas. Reg. § 1.482-1[e][2][iii][A]).

Petitioner was under a duty to come up with comparables which have, or could be brought to, a similar level of comparability and reliability. "[U]ncontrolled comparables that have a significantly lower level of comparability and reliability will not be used in establishing the arm's length range" (Treas. Reg. § 1.482-1[e][2][ii]). How then can petitioner justify using a startup business like Krantor Corporation which began business in 1989, sustained losses in 1989 and 1990, and had a three-year average return on sales of 0.0%? Clearly, sufficient information should have suggested an adjustment to petitioner for this "comparable" and, if not, it should not have been used at all. The fact that a statistical method is used to "adjust the range" does not counteract the skew of the data. Without Krantor, the first quartile becomes .8 instead of .7, the third quartile becomes 1.9 instead of 1.8 and the median becomes 1.3. TPS's ROS, .85, becomes perilously close to being outside the range of arm's length. At the very least, one could conclude, as did the Administrative Law Judge, that the reliability of the economic analysis was quite impaired. In a similar manner, Vestro Foods, Inc. also was a poor choice as a comparable. Vestro was engaged in the acquisition, ownership, and operation of specialty food businesses. During the years 1987, 1988, 1989 and 1990, Vestro acquired at least four companies and one business. These acquired companies comprise the operating core businesses of Vestro, according to its Form 10-K, filed with the SEC on April 15, 1991. Therefore, as a comparable, it appears that Vestro is best described as a startup corporation (like Krantor) during the years in issue here. Although such startup companies can be used as comparables (Treas. Reg. § 1.482-5[c][2][i]), the stage of a business in the business and product cycle must be given weight in determining the degree of comparability between the uncontrolled and controlled parties. In fact, operating profit may be affected by differences in business experience, such as whether the business is in a startup phase. This was not taken into account by Mr. Dildene.

If Vestro had been eliminated, the skewness of the interquartile range would have shifted again. This time the first quartile would have been 0.8, the third quartile 1.9 and the median 1.35.

The choice of purchasing co-ops was also ill-advised without any in-depth analysis of the agreements which existed between the co-op and its shareholders. Although petitioner claims this is inconsequential, petitioner has the burden of rebutting the presumption and simply asserting that there is little likelihood of these particular co-ops not maximizing profits or how their relationships with shareholders might have affected sales or operating income does not remove the cloud over these corporations (or meet their burden of proof). This is especially true since Mr. Dildene's report curtly described the reason for the rejection of "KFC Natl Purch Coop" as "coop makes purchases of equip & food for KFC & Taco Bell/coop" (Petitioner's Ex. 7, Appendix C). The fact that no adjustment was made for the cooperative element despite the fact that another cooperative was rejected, underscores the need for petitioner to have made an offer of proof as to why these comparables should have been used and accepted.

In addition, petitioner did not address the differences in geographic market which presented itself in the case of many of the comparables, even though the regulations specifically provide that different geographic markets can affect the prices charged (Treas. Reg. § 1.482-1[d][3][iv][A]). Since the geographic markets of many of the comparables were different than petitioner, their reliability is weakened. Petitioner bridled at this conclusion by the Administrative Law Judge, contending that there was no evidence that the difference in geographic markets had an effect on comparability. Petitioner misses the mark. It bears the burden of proof since the regulations specifically note geographic markets can affect comparability (Treas. Reg. § 1.482-3[b][2][ii][B][4]). It was incumbent on petitioner to address the issue.

The Administrative Law Judge found that the functional comparability was an overriding factor, which was not sufficiently weighted by petitioner. Citing the regulations, the Administrative Law Judge noted that the financial ratio chosen for comparison, return on sales, was sensitive to functional differences in the comparables (Treas. Reg. § 1.482-5[b][4][ii]). The key phrase of the regulation cited is that "closer functional comparability normally is required under a financial ratio than under the rate of return on capital employed to achieve a similarly reliable measure of an arm's length result" (Treas. Reg. § 1.482-5[b][4][ii]). Since the comparables were not functionally similar, the financial ratio used for comparison was less reliable.

We conclude that the Administrative Law Judge was correct in his analysis of the reliability of the CPM analysis done by petitioner's expert and we are not persuaded by its arguments on exception. We believe that petitioner has not met its burden given the inconsistencies and shortcomings in its analysis. Although there is no requirement that the Division produce an expert to refute the evidence offered by petitioner, given the potential for complexity in cases which involve transfer pricing issues, the Division appears without an expert at its own peril. However, in this matter, the shortfalls of petitioner's analysis were apparent without the assistance of an expert and even though the expert's testimony and report were uncontroverted, we are not bound to accept the expert's ultimate conclusion when the record does not support it, as it did not here (see, Matter of Medtronic, supra).

The second CPM analysis performed by Mr. Dildene was similarly flawed. This test was designed to confirm the results of the first test. The profit level indicator utilized here was "return on total costs" (choice of which was never explained) which compared the return petitioner received on total costs for each of its three main functions (distribution, transportation and marketing) with the same return on total costs for comparable companies that operated in the same areas. Of course, this methodology completely ignored the fact that petitioner realized the benefits of its business synergy, functional integration and economies of scale where all three functions operated together within the same entity towards the same goal. The comparables only represented one of the functions performed by petitioner and created an artificial relationship. Even petitioner conceded that its analysis of the various functions of the business as independent entities was "unorthodox," but the expert never addressed why such a comparison should be considered valid.

We think the Administrative Law Judge stated it succinctly when it was noted that petitioner ignored the other functions of petitioner's business in the first CPM analysis but then chose to recognize the functions of marketing and transportation in its second analysis, comparing petitioner's three main functions to companies in those businesses to establish that the results of the first CPM were correct. Comparing component functions of petitioner's business to a comparable's chief function does not appear to us to embody the intent of IRC § 482 when it describes the procedure for establishing comparability.

Factors such as geographic markets, the use of unaudited financial statements, calculation of distribution costs as a residue of total operating expenses less transportation and advertising expenses, and questionable comparables for petitioner's transportation function (freight forwarders who did not own transportation equipment like petitioner) all contributed to the Administrative Law Judge's rejection of the second analysis. Given the unorthodox nature of the second test and the failure to make any adjustments for these problems, the reliability of the comparables and the results of the test were severely weakened.

With regard to the Division's decision to combine PSI with TPI, we find that the Administrative Law Judge fully and adequately dealt with the issue and affirm the conclusion to permit the Division to require combination. PSI received raw material with value at no cost and paid a reduced sum to TPI for freight service, payroll service, customer billing, contract negotiation and credit checking. There was no evidence in the record why PSI received a discount on the amount charged. In addition, there were various lease agreements between TPI and PSI for delivery equipment, a feed mill and a warehouse. No evidence was presented which established that the $1,100.00 monthly rent for the mill and warehouse was at arm's length or if the allocation of energy costs was at arm's length. Given petitioner's failure to meet its burden of showing that the pricing involved in the wet peel operation was at arm's length, we conclude that the Administrative Law Judge's determination was correct.

Finally, we have reviewed petitioner's argument that requiring petitioner to file on a combined basis with TPI and PSI is in violation of the Due Process Clause and Commerce Clause of the United States Constitution. This is the identical argument petitioner made before the Administrative Law Judge below and we believe the Administrative Law Judge adequately and correctly decided the issue and we affirm on that basis.

Accordingly, it is ORDERED, ADJUDGED and DECREED that:

1. The exception of Tropicana Product Sales, Inc. is denied;

2. The determination of the Administrative Law Judge is affirmed;

3. The refund applications of Tropicana Products Sales, Inc. are denied;

4. The petition of Tropicana Products Sales, Inc. is denied; and

5. The Notice of Deficiency, dated April 11, 1994, is sustained.

DATED: Troy, New York
               June 12, 2000

                                                                                                                       /s/Donald C. DeWitt          
                                                                                                                       Donald C. DeWitt
                                                                                                                       President



                                                                                                                      /s/Carroll R. Jenkins          
                                                                                                                      Carroll R. Jenkins
                                                                                                                      Commissioner

                                                                                                                     /s/Joseph W. Pinto, Jr.       
                                                                                                                     Joseph W. Pinto, Jr.
                                                                                                                     Commissioner

1. While the petition in this proceeding indicated that the periods at issue were fiscal years ended July 31, 1989 through July 31, 1993, petitioner, at the hearing, withdrew its claims for refund for fiscal years ended July 31, 1992 and July 31, 1993, as audits for these periods had not yet been conducted. Accordingly, the periods at issue are fiscal years ended July 31, 1989 through July 31, 1991, only.

2. Finding of fact "47" was changed to properly reflect the record in that the Administrative Law Judge found that the witness was an expert in economics and transfer pricing.

3. The amount set forth for "Transportation" was taken from the "other deductions" schedule under the category of "distribution."

4. A form 10-K is an annual report required to be filed with the SEC pursuant to section 13 or 15(d) of the Securities Exchange Act of 1934.

5. Table 3 in the report sets forth this table and notes that all figures are in $US thousands. In TPS's financial statement, found in Appendix A of the report, which Mr. Dildine testified is the source of these figures (see, Finding of Fact "73"), the indication is that all data is in $US millions.

6. The requirements necessary to give rise to the presumption of distortion in this case are as follows:

1) "substantially all of the capital stock of the taxpayer and substantially all of the capital stock of the other corporations which are to be included in the combined report are owned or controlled, either directly or indirectly, by the same interests" (20 NYCRR former 6-2.2[a][3]);

2) "the corporations are in substance parts of a unitary business conducted by the entire group of corporations" (20 NYCRR former 6-2.3[a][1]); and

3) "there are substantial intercorporate transactions among the corporations" (20 NYCRR former 6-2.3[a][2]).

[Back to Top][What's New?][Home]