MEMORANDUM OPINION, FINDINGS OF FACT, CONCLUSIONS OF LAW AND ORDER FOR JUDGMENT

VIETOR, Chief Judge.

This is an antitrust action brought by the United States of America ("government"). The government alleges that a long-term lease agreement entered into in 1982 by which defendant Archer-Daniels-Midland Company ("ADM") leased two corn wet milling plants, one in Iowa and one in New York, from defendant Nabisco Brands, Inc. ("Nabisco") is an acquisition that may substantially lessen competition in violation of section 7 of the Clayton Act, 15 U.S.C. § 18, and constitutes a contract, combination and conspiracy in unreasonable restraint of trade in violation of section 1 of the Sherman Act, 15 U.S.C. § 1. The line of commerce involved is the manufacture and sale of High Fructose Corn Syrup ("HFCS"), a nutrient sweetener made from corn that is used to sweeten soft drinks and food products.

Trial to the court was held only on the issue of liability.

JURISDICTION AND VENUE

This court has jurisdiction pursuant to section 4 of the Sherman Act, 15 U.S.C. § 4, and section 15 of the Clayton Act, 15 U.S.C. § 25. Venue in this district is proper. Clayton Act, § 12, 15 U.S.C. § 22; 28 U.S.C. § 139(c). (See paragraph 10 of Findings of Fact, infra.)

OVERVIEW

The legal standards for an antitrust evaluation of an acquisition or merger are substantially the same under both the Clayton and Sherman Acts. As provided in section 7 of the Clayton Act, an acquisition is proscribed "where in any line of commerce * * * the effect of such acquisition may be substantially to lessen competition or to tend to create a monopoly." This statutory language requires proof of a causal connection between the acquisition and a reasonable probability of future injury to the competitive process, which is measured by the likelihood of the acquisition creating ability in the acquiring party to exercise market power in the future. Market power is the ability of a firm profitably to increase prices above competitive levels and to maintain such higher prices for a significant period of time.

In applying this legal standard, a four-part analysis is used: (1) definition of the relevant product market; (2) determination of the relevant geographic market; (3) evaluation of statistical data on concentration in the defined relevant market; and (4) consideration of "other factors" pertinent to whether the acquisition creates a reasonable probability of the exercise of market power in the future. United States v. General Dynamics Corp., 415 U.S. 486, 94 S.Ct. 1186, 39 L.Ed.2d 530 (1974); Brown Shoe Co. v. United States, 370 U.S. 294, 82 S.Ct. 1502, 8 L.Ed.2d 510 (1952); FTC v. National Tea Co., 603 F.2d 694, 700 (8th Cir. 1979).

HISTORY OF THIS LITIGATION

This lawsuit has been active for many years. The complaint was filed December 14, 1982, six months after defendants entered into the lease.

In the early stages of the litigation, defendants moved for summary judgment contending that the lease is not an "acquisition" within the meaning of section 7 of the Clayton Act, and the government moved for partial summary judgment contending (For a that the lease is an acquisition (For a summary of the lease terms, see paragraphs 4-7 of Findings of Fact, infra.) I ruled that the lease is an acquisition. United States v. Archer-Daniels-Midland Co., 584 F. Supp. 1134 (S.D.Iowa 1984). Pursuant to 28 U.S.C. § 1292(b), I certified the question for immediate appeal, but in an unpublished order dated May 29, 1984, the United States Court of Appeals for the Eighth Circuit declined to accept the interlocutory appeal.

After a substantial period of extensive discovery, both sides filed cross motions for summary judgment on the relevant product market issue. The government, in its motion, requested a ruling that the relevant product market is limited to HFCS alone as alleged in the complaint. Defendants, in their motion, contended that undisputed facts on cross-elasticity of demand between HFCS and sugar precluded the government from carrying its burden to prove that HFCS is the only sweetener in the relevant product market. On May 29, 1987, I denied the government's motion and granted defendants' motion based on certain undisputed facts which led me to conclude that sugar and HFCS are in the same product market, and I ordered that the government's complaint be dismissed. (It is undisputed that if both sugar and HFCS are in the relevant product market, there is no antitrust violation.) A memorandum opinion explaining the rulings and order of dismissal was filed on August 6, 1987. United States v. Archer-Daniels-Midland Co., 695 F. Supp. 1000 (S.D.Iowa 1987).

The government appealed to the Eighth Circuit Court of Appeals from the order granting defendants' motion for summary judgment and dismissing the complaint, but did not appeal from the order denying the government's motion for summary judgment. The Eighth Circuit, in a 2-1 panel decision, held:

* * * that the district court erred in granting summary judgment that sugar is in the same relevant product market as HFCS. We, in turn, grant summary judgment to the appellant that sugar is not in the same relevant product market as HFCS. We therefore reverse and remand * * *.

On remand, after more discovery and a tour by counsel and me of the leased plant in Clinton, Iowa, and, for comparison purposes, a much newer ADM corn wet milling plant in Cedar Rapids, Iowa, the liability phase of this case was tried.

FINDINGS OF FACT THE PARTIES

(1) The plaintiff is the United States of America, which is represented by attorneys from the Antitrust Division of the Department of Justice.

(2) Defendant ADM is a corporation organized under the laws of Delaware and has its principal place of business in Decatur, Illinois. ADM is primarily an agricultural processing company engaged in the purchasing, transporting, storing, processing and merchandising of various agricultural commodities such as corn, wheat and soybeans. Currently, ADM sells its products throughout the United States and elsewhere.

(3) In 1981, Nabisco, Inc. and Standard Brands Incorporated ("Standard Brands") were merged to form defendant Nabisco, which is a corporation organized under the laws of Delawere and has its principal place of business in East Hanover, New Jersey. Nabisco has lately become a wholly-owned subsidiary of RJR Nabisco, Inc., a Delaware corporation. Nabisco manufactures and markets throughout the United States consumer-sized packages of branded grocery products, including cookies, crackers, cereals, confectioneries, snacks and other products.

THE LEASE

(4) In a lease agreement dated June 12, 1982, effective June 14, 1982, Nabisco granted to ADM for a noncancelable term of thirteen years the right to use the land, buildings and equipment of two plants for corn wet milling operations in Clinton, Iowa, and Montezuma, New York.

(5) ADM pays Nabisco rent each year under the lease, with annual net rental payments starting at $15,200,000 for the first year and then declining gradually thereafter to $9,500,000 in the thirteenth year. ADM has an option to renew the lease for one additional five-year term. The net rental payment for each renewal year is fixed at $15,000,000.

(6) ADM may not make any alterations, improvements or additions exceeding $1,000,000 in cost during any lease year without Nabisco's approval. Upon termination of the lease, all improvements and additions made at ADM's expense become Nabisco's sole property. ADM may not sublease or assign the lease without Nabisco's approval.

(7) ADM has an option to purchase the facilities, exercisable on June 30 in each of the lease years six through eleven (19881993) at the prices set forth in Schedule C to the lease. If purchased in year six, the acquisition price is $128,000,000. The annual option acquisition price then declines gradually as the facilities become older to $72,000,000 in year eleven. Thereafter, ADM possesses no option to purchase except as of the last day of the five-year renewal term. At that time, the purchase price is specified to be the fair market value of the leased assets, excluding any additions or improvements made by ADM at its sole expense. In addition, Nabisco has a one-time option, as of the last day of the intial lease term, to require ADM to purchase the plants for $55,000,000. If none of the above options is exercised, the entire property reverts to Nabisco's sole ownership upon termination of the lease.

(8) At the time of trial, ADM had not exercised any of its options to purchase the leased plants.

(9) The parties structured the lease as an "operating lease," and both companies were advised by their outside auditors that it is an operating lease, as distinct from a "capital lease" tantamount to an installment purchase arrangement.

INTERSTATE COMMERCE

(10) The business activities of ADM and Nabisco from 1982 to the present are within the flow of, and affect, interstate commerce. ADM and Nabisco transact business and are found within the Southern District of Iowa.

SWEETENERS

(11) Sweeteners are used to sweeten foods and beverages. There are basically two types of sweeteners, nutrient and nonnutrient, or artificial.

(12) Sucrose, commonly and hereinafter called "sugar," is a nutrient sweetener made from either sugar cane or sugar beets. Sugar is generally recognized as the "universal sweetener," because it is capable of performing the sweetening function across a wide range of foods and beverages. Refined sugar can be readily converted into liquid sugar, or invert sugar, for use as a liquid sweetener.

(13) In terms of sweetness value, sugar sets the standard and is assigned a numerical value of 100. The sweetness value assigned to other sweeteners is stated in terms of the percentage relationship to the sweetness value of sugar.

(14) Some other nutrient sweeteners are glucose (regular corn syrup), dextrose, fructose, HFCS, crystalline fructose, honey, maple syrup, edible molasses and sorghum syrup.

(15) In recent years, non-nutrient sweeteners have been developed and marketed. These are high intensity sweeteners, often hundreds of times sweeter than sugar, which have no nutrient value and respond to a rising demand of food and beverage consumers for low calorie or noncaloric products. A well known non-nutrient sweetener is aspartame, known in the United States by the brand name "Nutrasweet."

CORN WET MILLING

(17) Corn wet milling plants have existed since the 19th century, but prior to the 1970's they produced primarily corn starches and two corn sweeteners, dextrose and glucose corn syrup. Dextrose and glucose are both less sweet than sugar.

(18) The initial step in the corn wet milling process is called "steeping." Corn is soaked in a solution of water containing a small amount of sulphur dioxide that prevents germination or fermentation of the corn. This softens the corn kernel so that its component parts (germ, starch, gluten and hull) can be separated at various subsequent stages of the processing cycle.

(19) After steeping, the corn is then further processed to separate the starch from the other components of the kernel. The starch emerges as "starch slurry" and is the basic raw material from which finished products are manufactured, such as HFCS, corn syrup, dextrose, crystalline fructose, potable and fuel alcohol, and a variety of starches. The remaining components of the kernel produce various byproducts.

(20) To convert the starch slurry into corn sweeteners, enzymes are added. The enzyme "alpha amylase" acts to break down starch into glucose, and the enzyme "glucoamylase," sometimes called AMG, speeds the conversion of weak glucose solutions into dextrose. At particular points in this process, a corn wet milling plant is able to stop the chemical reactions of the enzymes in order to produce the particular form of glucose having the desired degree of sweetness and other characteristics. The resulting glucose products are often called "regular corn syrup" and have sweetening values ranging from approximately 25% to 67% of the sweetness of sugar.

(21) If the chemical reaction is allowed to proceed without interruption until fully completed, the slurry eventually becomes almost entirely dextrose, which is the sweetest form of glucose and is generally considered to be about 70% as sweet as sugar.

(22) The dextrose syrup resulting from the initial phase of the processing can be subjected to a second process, known as isomerization, to produce a sweeter product. This is accomplished by addition of an enzyme called "glucose isomerase," sometimes referred to as GI, that converts the dextrose into a sweeter syrup that contains 42% fructose and is commonly called "HFCS 42." HFCS 42, which is less sweet than sugar, is used in a wide range of food products, including canned fruits and vegetables, dairy products, baked products and many beverages.

(23) A technology called the second generation isomerization process can be used to separate HFCS 42 into two components, one of which contains approximately 90% fructose and is therefore identified as "HFCS 90." The other component, called "returns," is primarily dextrose that is returned to the beginning of the production line and is isomerized again to produce HFCS 42.

(24) The component consisting of HFCS 90 has a sweetening value of about 120% of sugar. A smaller quantity of HFCS 90, when used as an ingredient in foods or beverages, will produce the same level of sweetness than the larger amount of sugar requires to achieve an equivalent sweetness value.

(25) To produce a corn sweetener with a sweetening value approximately equivalent to the sweetening value of sugar, corn wet millers blend together appropriate quantities of HFCS 42 and HFCS 90 to produce a product containing approximately 55% fructose, which is therefore called "HFCS 55." HFCS 55 is sweeter than HFCS 42. Although HFCS 55 was originally intended to have the same sweetening value as sugar, the large soft drink companies, such as Coca-Cola and Pepsi Cola, have concluded that HFCS 55 is approximately 94% to 95% as sweet as sugar.

(26) Crystalline fructose is a relatively new product of corn wet milling. It is pure fructose which is crystallized from HFCS and sold in a pure white, free flowing granular form. Crystalline fructose has a relative sweetness value of up to 1.8 times that of sugar, although in liquids its sweetness value is approximately 117% that of sugar.

(27) The equipment used to convert the starch slurry into HFCS of different grades is complex, specialized and expensive.

(28) Corn wet milling plants vary considerably in design, equipment and product mix. Some corn wet millers specialize in particular products such as starches and have no facilities to make HFCS.

BASIC BACKGROUND OF PLANTS

(29) Clinton Sugar Refining Company was originally incorporated in 1906 as a subsidiary of National Candy Company. It built and, in 1906, began to operate a corn wet milling plant located in Clinton, Iowa. In 1956, Standard Brands purchased that wet milling plant and transferred it to its subsidiary, Clinton Corn Processing Company, Inc. ("CCP"). Thereafter, the plant was operated by CCP as a subsidiary of Standard Brands until 1981 and then as a subsidiary of Nabisco following the merger of Standard Brands and Nabisco, Inc. This plant is one of the two plants leased by Nabisco to ADM in June of 1982. (In the trial testimony as well as in these findings, the term "Nabisco" is sometimes used to refer to the corn wet milling facilities and/or operations of CCP during the period from 1956 up to the Lease Agreement in June of 1982). The long and productive career of this plant is traced in paragraphs 33-38, infra.

(30) Financed by its parent corporation, CCP completed construction in 1977 of a new corn wet milling plant in Montezuma, New York, which is the other plant leased in June of 1982 by Nabisco to ADM. The checkered career of this plant is traced in paragraphs 39-42, infra.

(31) At the time of the lease in June 1982, Nabisco's plant at Clinton was producing a variety of products with the corn wet milling process, including HFCS. In June 1982, Nabisco's plant at Montezuma was shut down, but had the capability of producing HFCS.

ADM'S OTHER WET MILLING PLANTS

(32) At the time of the lease in June 1982, ADM owned and operated two other corn wet milling plants, one in Cedar Rapids, Iowa, and the other in Decatur, Illinois. ADM had entered into corn wet milling operations back in 1971, and ADM began producing commercial quantities of HFCS in 1976. At the time of the lease in June 1982, ADM was producing commercial quantities of HFCS at both of its corn wet milling plants.

THE CLINTON PLANT

(33) After construction was completed in 1906, the plant at Clinton began producing corn syrup. Its initial capacity was 3,000 bushels of corn per day. Over the years, the plant "grew like Topsy," with brick multi-story buildings progressively added to the site in patchwork fashion, each housing different process functions and product lines. At the time of its acquisition by Standard Brands in 1956, the Clinton plant had a grind capacity of approximately 60,000 bushels of corn per day and was capable of making refined corn oil, dry starch products, dextrose, and regular corn syrup.

(34) In the mid-1960's, CCP, utilizing technology licensed from the Japanese Agency of Industrial Science and Technology, developed a commercial process of isomerization in which glucose isomerase was used to convert dextrose into fructose, a sweeter product. The first HFCS, containing 42% fructose, was produced and marketed in the United States by CCP in 1967. (It was not until the 1970's that other companies began producing HFCS.)

(35) The grind at the Clinton plant increased to approximately 106,000 bushels per day by 1972, by which time equipment had been added for the production of beverage alcohol and certain types of enzymes, as well as HFCS.

(36) At the time of the lease in June of 1982, the grind capacity of the Clinton plant was approximately 110 000 bushels per day.

(37) The Clinton plant was an inefficient, money-losing operation at the time it was leased to ADM. Immediately upon taking over the Clinton plant, ADM drastically reduced the number of employees and inaugerated its own lean, no-frills, no-nonsense, efficient style of management and operations with notable success. ADM substantially rebuilt and expanded the Clinton plant from top to bottom, largely using ADM proprietary process technology and equipment. As of the time of trial, ADM had made $206 million in capital expenditures at the plant and efficiency had been greatly improved. Processing costs per bushel were substantially cut, labor productivity was substantially increased, and the plant was integrated into ADM's multiplant network, with numerous functions previously performed at other plants consolidated at Clinton, resulting in cost savings.

(38) As of the time of the plant inspection by counsel and me in May 1990, the grind capacity of the Clinton plant had increased to 250,000 bushels per day. ADM has added new finishing capacity for the production of several new products that were not made prior to the lease, including fuel alcohol, degradable plastic resins and crystalline fructose. ADM has also substantially increased the Clinton plant's capacity for production of enzymes as well as the plant's finishing capacity to produce the same types of corn wet milling products that were being produced in June of 1982.

THE MONTEZUMA PLANT

(39) The Montezuma plant is in upstate New York. It was constructed by Standard Brands in the mid-1970's as a corn wet milling plant designed to produce HFCS. Construction was completed in 1977, and it was then ready to commence commercial production of HFCS. However, the plant was immediately placed in a "mothball" status and remained idle until 1978 because of a reduction in HFCS prices to a level that made production of HFCS at Montezuma uneconomical. After 1978, it operated on an intermittent basis.

(40) The Montezuma plant is poorly located. Corn costs are relatively high, quality of locally produced corn is poor, and the principal markets for HFCS as well as corn byproducts are a long distance away. The plant is located on a small country road far from interstate highways, and has no access to water transportation. During the intermittent periods that Montezuma was operated as a corn wet milling plant by Nabisco prior to the lease, it was a high cost and inefficient plant.

(41) Prior to the 1982 lease, the Montezuma plant was never profitable on any sustained basis. Corn wet milling operations at the plant were suspended again in December 1981, and employees were notified in January of 1982 that the suspension would continue for an indefinite period. Operations at Montezuma remained suspended at the time of the lease.

(42) After June of 1982, ADM reopened the Montezuma plant and operated it intermittently between 1984 and 1986. Thereafter, ADM concluded that Montezuma could not be operated economically, and the plant was shut down and remains shut down. Currently it is used as a warehouse and distribution point.

HFCS IN THE MARKET

(43) Since its commercial development, HFCS has been marketed as a functionallyequivalent lower-priced substitute for sugar in food and beverage applications. It has gradually replaced sugar as the sweetener in many uses for which the two products are functionally interchangeable. The replacement is virtually complete in products where the functional interchange is nearly perfect, but the replacement is less than complete in products where the functional interchange is noticeably less than perfect.

(44) Public acceptance of HFCS and growth of sales came gradually because food and beverage manufacturers were cautious about changing formulations in products that had customer acceptance and brand loyalty. It took some time for the HFCS industry to refine its technology to ensure consistent product quality. As product consistency improved, customer acceptance of HFCS increased. It took additional time for food and beverage manufacturers to revise product formulations and marketing strategies.

(45) HFCS 42 was initially marketed as a replacement for sugar in jams, jellies and certain baked goods, as well as some beverages.

(46) The largest single use of HFCS today is as a sweetner in soft drinks. Some HFCS 42 replaced sugar in soft drinks as early as the mid-1970's, but the soft drink producers were initially very cautious about revising their product formulas to permit the use of HFCS. The introduction of HFCS 55 in 1978 gave the industry a product nearly as sweet as sugar and led, gradually, to the acceptance of HFCS in the beverage industry.

(47) In 1978 the Coca-Cola Company approved the replacement of 25% of the sugar by HFCS in its non-cola beverages. In 1979 it approved the replacement of 50% of the sugar in its colas.

(48) At the time of the lease in 1982, Pepsi Cola still had not modified its formula to permit use of HFCS in its flagship brand.

(49) By November 1984, both Coca-Cola and Pepsi Cola modified their formulations to permit full substitution of HFCS for sugar.

(50) By the mid-1980's the market for HFCS showed signs of maturity. HFCS manufacturers demonstrated to food and beverage manufacturers that the HFCS supply was consistent and reliable, and that HFCS was functionally interchangeable with sugar in many applications. Because HFCS was substantially less costly, it became the sweetener of choice in many of the uses for which it is a functionally acceptable ingredient, especially in uses for which the price differential overcomes any quality difference, such as soft drinks.

(51) Since the mid-1980's, HFCS sales in the United States have continued to grow, although at a reduced rate compared to prior years.

(52) HFCS serves several functions as a food ingredient. It has a taste or sweetness profile very similar, although not identical, to sugar in many products.

(53) In addition to sweetness, HFCS brings at least two critical and related functional attributes to food and beverage formulations. HFCS provides bulk or volume to product formulations, and HFCS reacts with water and can serve to control or alter the activity of water in formulations.

(54) HFCS 55 is 77% solid materials by volume and HFCS 42 is 71% solids; thus, HFCS brings bulk, volume and substance to all product formulations into which it is incorporated.

(55) In certain applications, HFCS serves as a viscometric agent. In other words, it causes the water in the formula to flow more slowly and to become more syrupy. In soft drinks, this functional attribute is commonly described as "body" or "mouthfeel." In canned fruits HFCS contributes to the thickness and volume of the canning syrup.

(56) HFCS is also a humectant, absorbing and holding on to available water in certain applications. This characteristic makes HFCS essential to the manufacture of soft, chewy cookies.

(57) The water control properties of HFCS also allow it to be used as a preservative in some formulations, such as for jams and jellies, because it renders moisture unavailable to the bacteria that lead to spoilage. These properties also allow HFCS to retard staling in baked goods applications such as breads.

(58) HFCS is a reducing sugar, which means it promotes browning in baked products. Thus it can be used to provide characteristic coloring in baked goods such as breads, cookies and crackers.

(59) HFCS has nutritive or food value and serves as a source of calories and, therefore, energy.

(60) HFCS is composed of naturally occurring sugars and is produced through a process "Generally Recognized as Safe" by the United States Food and Drug Administration. It can therefore be used generally in foods and beverages without special labelling.

(61) HFCS is stable and will retain its sweetness and other functional attributes over time after it is incorporated in foods and beverages, although it may be a little less stable than sugar in some applications, such as canning.

(62) Today HFCS is almost the only sweetener used in colas and other regular or nutritively sweetened beverages, having supplanted invert sugar. At the time of the lease, however, and for a few years after, most soft drinks were sweetened with a combination of sugar and HFCS. In 1988, approximately 73% of HFCS sold in the United States was used in soft drinks is HFCS 55. The other major categories of use of HFCS are baked goods (5% of sales), canned and bottled fruits and vegetables and condiments (8% of sales), ice cream and other dairy products (3% of sales), confectionery (1% of sales), and miscellaneous foods (8% of sales).

RELEVANT PRODUCT MARKET

(63) As noted in the "History of this Litigation" section, supra, the United States Court of Appeals for the Eighth circuit granted summary judgment to the government in this case that sugar is not in the same relevant market as HFCS. This decision was based on a finding that although sugar and HFCS are functionally interchangeable, they are not reasonably interchangeable because of the price differential between the two products, the price of HFCS being 10%-30% lower than the price of sugar. ADM, 866 F.2d at 246-48.

(64) The Eighth Circuit, after noting that the price of sugar in the United States is kept artificially high by the government sugar program, stated: "If, at some time in the future, the government decides to eliminate the price supports, the price of sugar at that time may or may not constrain the price of HFCS, and we may have to take another look at this product market." ADM, 866 F.2d at 246.

(65) Sugar price supports have been controversial, and some observers expected that the price support program would not be renewed upon its expiration at the end of 1990. On November 28, 1990, however, the President signed the "Food, Agriculture, Conservation, and Trade Act of 1990." P.L. 101-624 (the "Act"). Title IX of the Act provides that the "price of each of the 1991 through 1995 crops of sugar beets and sugarcane * * * shall be supported." The Act provides for nonrecourse loan levels at "such level as the Secretary determines appropriate, but not less than 18 cents per pound for raw cane sugar." The price support level provided for in the Act is the same price support that has existed since 1985 under the expiring legislation.

(66) Glucose, or regular corn syrup, has been in use as a sweetener for almost two centuries. It is available in varying degrees of sweetness ranging from approximately 25% to 67% as sweet as sugar. It is used in baking, canning, confections, dairy, and other processed foods, usually in a blend with HFCS and sugar. Its price is lower than both sugar and HFCS, so food producers make price/quality tradeoff decisions regarding the extent of its use.

(67) Dextrose is about 70% as sweet as sugar. On a price/quality tradeoff basis, destrose can be substituted for sugar or HFCS in some products.

(68) In 1970, glucose accounted for over 70% of corn sweeteners consumed in the United States, dextrose accounted for 24% and HFCS amounted to only 4%. By 1980, HFCS had surpassed glucose, and the growth in consumption of HFCS continued throughout the 1980's. According to the United States Department of Agriculture, the growth in HFCS in the United States has been achieved, in part, through displacement of dextrose and glucose corn syrup. In at least one instance, however, a producer of table syrup replaced HFCS with glucose in order to respond to customer demand for a lower priced product. Overall, glucose and dextrose do not constrain the price of HFCS.

(69) Crystalline fructose, which is made from HFCS, is up to 1.8 times sweeter than sugar in dry form, and approximately 17% sweeter than sugar in a water solution. It is a dry, granular sweetener for use in a variety of soft drinks, dairy products, candies, and other food products as a partial or complete replacement of either sugar, HFCS or aspartame. It is much more expensive than HFCS and even more expensive than sugar. Crystalline fructose is in its infancy, and substantially all of its limited production is exported. It does not constrain the price of HFCS.

(70) I find that although glucose, dextrose and crystalline fructose have some degree of imperfect functional interchangeability with HFCS, they are not reasonably interchangeable with HFCS.

(71) Over the last twenty years, there has been increased consumer demand for diet and low calorie foods. In response, food and beverage manufacturers have introduced many new products that are advertised as low in calories or as special diet products. Artificial sweeteners have been developed and marketed for inclusion in these products because they have little or no caloric value. They are chemical products, not natural food products like sugar, HFCS and other corn sweeteners.

(72) The premier artificial sweetener is aspartame, which was developed in 1965. Patents were obtained in the United States and other countries. The aspartame patent in the United States, which is held by the Monsanto Company, expires on December 31, 1992, and patent protection has already expired in several other countries of the world.

(73) Aspartame is currently manufactured in the United States exclusively by the NutraSweet Company, a subsidiary of Monsanto, under the brand names "NutraSweet" and "Equal," a low calorie table top sweetener made with NutraSweet. Aspartame was developed by G.D. Searle Co. and the NutraSweet Company became a subsidiary of Monsanto in 1985 when Monsanto acquired Searle.

(74) Aspartame is approximately 200 times as sweet as sugar.

(75) Aspartame has been approved for use in a wide variety of products in almost 80 countries throughout the world.

(76) FDA approval for use of aspartame in dry products was granted in 1981, and approval for use of aspartame in carbonated beverages was granted in 1983.

(77) Aspartame is used in a wide variety of products, including those produced by the soft drink, baking, confectionery, dairy, and canning industries. Sales of aspartame to the carbonated beverage industry account for about 80 percent of NutraSweet's United States sales of aspartame.

(78) Per capita consumption of aspartame in the United States has grown from one pound (sugar equivalent basis) in 1982, the year of the lease, to 14 pounds in 1988. All artificial sweeteners accounted for approximately 6% of total United States sweetener consumption in 1980, and about 13% in 1988.

(79) NutraSweet's aspartame sales have increased substantially, and are expected to continue to grow steadily.

(80) Artificial sweeteners, including aspartame, provide sweetness only. They lack the multiplicity of functional characteristics that HFCS provides, such as bulk, texture, preservative qualities and color. They have no nutritional value.

(81) Artificial sweeteners, including aspartame, are not comparable to HFCS in terms of taste. The sweetness profile of artificial sweeteners generally is characterized by rapid onset of the sweetness sensation and very short duration, frequently accompanied by a detection of another taste. Although some consumers of soft drinks cannot detect a difference, a large portion of consumers do detect a difference in taste and mouth feel between a beverage sweetened with HFCS and one sweetened with aspartame.

(82) Artificial sweeteners, including aspartame, when used in food products produce qualities and characteristics different from those sweetened by HFCS and other nutrient sweeteners.

(83) The growing market for aspartame is because of the growing number of people who want low calorie food and beverage products. For instance, as Coca-Cola's Director of Sweetener and Ingredient Purchasing testified, "a customer that wants a low calorie drink will buy a Diet Coke. A customer that wants a nutritive-sweetened beverage will buy Coca-Cola Classic, and they are two separate product lines." He noted that diet and regular are different applications, resulting in different products with distinct marketing strategies and groups of customers. Manufacturers of food products also consider artificially sweetened products as belonging to a distinct category. Continental Baking is unable to use artificial sweeteners such as aspartame as a direct replacement for HFCS in its food products for a variety of reasons, including lack of bulking properties, instability, lack of preservative properties, and inability to ferment (for breads). Continental Baking might evaluate aspartame to develop a new product with specific properties, but not to replace HFCS or other nutritive sweeteners the company presently uses. Nabisco also considers its research into low calorie baked goods as new product development.

(84) Aspartame has been and remains considerably more expensive than HFCS on a sweetness equivalence basis. Neither purchasers nor sellers of HFCS consider the price of aspartame when buying or selling HFCS. When the United States patent on aspartame expires on December 31, 1992, however, the price of aspartame will probably drop significantly. That, of course, is a year away.

(85) It might be possible that at some future time a small percentage of the HFCS used to sweeten a beverage or food product could be replaced with aspartame without noticeably changing its taste or characteristics. Such blending has not yet been done and would not be economically advantageous unless and until the cost of aspartame on a sweetness equivalency basis went below the cost of HFCS.

(86) I find that aspartame and other artificial sweeteners have little, if any, functional interchangeability with HFCS, and they are not reasonably interchangeable with HFCS. (One cannot sweeten CocaCola Classic with aspartame, or Diet Coke with HFCS.)

(87) Aspartame and other artificial sweeteners do not constrain the price of HFCS.

RELEVANT GEOGRAPHIC MARKET

(88) Essentially all HFCS sold in the United States is produced in plants located in Canada and the United States. The Canadian plants are all owned and operated by CPC International (CPC), a United States company.

(89) Mexico does not have HFCS plants. Moreover, Mexican corn is not well suited for HFCS production and that makes construction and operation of a plant in Mexico problematic.

(90) Canada does not have a sugar price support program. Sugar prices in Canada are consequently tied to world sugar prices and since 1982 have been substantially below the price of sugar in the United States. Higher world sugar prices would increase HFCS demand in Canada, but what will happen in the future is speculative.

(91) Sugar prices in Mexico are regulated by the Mexican government, but are basically world prices. A limited amount of HFCS is exported to Mexico.

(92) Domestic HFCS producers sell essentially all of their HFCS to domestic customers. In no year have HFCS exports to all foreign countries, including Canada and Mexico, exceeded 2% of domestic production; in most years exports to foreign countries have been well below 1%. HFCS producers hope to export more in the future, but that is speculative. At least 90% of all HFCS imported into the United States comes from Canada.

(93) HFCS is relatively costly to transport and requires special handling. It will crystallize if not maintained at proper temperatures. These handling and shipping requirements prevent HFCS from being ocean shipped in significant quantities.

CONCLUSIONS OF LAW (Including Ultimate Fact Findings) RE: RELEVANT PRODUCT MARKET AND GEOGRAPHIC MARKET

Before proceeding to additional fact findings on other issues, it is appropriate to reach ultimate fact findings and conclusion of law on the issues of relevant product market and relevant geographic market.

A "line of commerce" under section 7 of the Clayton Act requires proof of a relevant product market. This is an essential element in the government's case because "[w]ithout a definition of that market there is no way to measure [the] * * * ability [of the challenged acquisition] to lessen or destroy competition." Walker Process Equipment Inc. v. Food Machinery Chemical Corp., 382 U.S. 172, 177, 86 S.Ct. 347, 350, 15 L.Ed.2d 247 (1965).

The Supreme Court has established two legal tests for determining the relevant product market. As set forth in Brown Shoe, 370 U.S. at 325, 82 S.Ct. at 1523, the "boundaries of a product market are determined by the reasonable interchangeability of use or the cross-elasticity of demand between the product itself and substitutes for it."

In determining what products belong in a particular product market, only those "commodities reasonably interchangeable by consumers for the same purposes make up that 'part of the trade or commerce,' monopolization of which may be illegal." United States v. E.I. DuPont de Nemours Co., 351 U.S. 377, 395, 76 S.Ct. 994, 1007, 100 L.Ed. 1264 (1959).

Products that are not "functionally interchangeable" are, of course, not "reasonably interchangeable."

Whether products that are "functionally interchangeable" are "reasonably interchangeable," involves a consideration of price, use and qualities. DuPont, 351 U.S. at 404, 76 S.Ct, at 1012. In other words, are the products economic substitutes for each other?

Defendants put into the record substantial evidence in an effort to show that sugar and HFCS are in the same relevant product market. The United States Court of Appeals, however, after noting that as a result of the government price support program there is a constant 10%-30% price differential between HFCS and sugar, granted summary judgment to the government in this case "that sugar is not in the same relevant product market as HFCS." ADM, 866 F.2d at 248. The court did say that if, in the future, the government decides to eliminate sugar price supports "we may have to take another look at this product market." ADM, 866 F.2d at 246. The government has not, however, decided to eliminate sugar price supports (see Finding of Fact No. 65, supra), so the summary judgment granted by the appeals court remains the law of this case and cannot be reexamined. See Bethea v. Levi Strauss and Company, 916 F.2d 453, 456-57 (8th Cir. 1990). Accordingly, I find as an ultimate fact and conclude that sugar is not in the same relevant product market as HFCS.

Based on the fact findings I have made that other nutrient and artificial sweeteners are not reasonably interchangeable with HFCS, I find as an ultimate fact and conclude that they are not in the same relevant product market as HFCS, and that HFCS is therefore the only product in the relevant product market.

The area "in which the seller operates and * * * the purchaser can practicably turn for supplies" is the relevant geographic market. Tampa Electric Co. v. Nashville Coal Co., 365 U.S. 320, 327, 81 S.Ct. 623, 627-28, 5 L.Ed.2d 580 (1961).

The purpose of defining a geographic market is to determine the area within which the merging parties may be able to exercise market power, i.e., "where * * * the effect of the merger on competition will be direct and immediate." United States v. Philadelphia Nat. Bank, 374 U.S. 321, 357, 83 S.Ct. 1715, 1738, 10 L.Ed.2d 915 (1963).

The relevant geographic market cannot be established with precision. United States v. Connecticut Nat. Bank, 418 U.S. 656, 669, 94 S.Ct. 2788, 2796, 41 L.Ed.2d 1016 (1974) (citing Philadelphia Nat. Bank, 374 U.S. at 360 n. 37, 83 S.Ct. at 1739-40 n. 7).

A properly defined geographic market serves to separate those firms that are important factors in the competitive analysis of a merger from those firms that are not. Tampa Electric Co., 365 U.S. at 327, 328, 81 S.Ct. at 628-29; Hospital Corp. of America v. FTC, 807 F.2d 1381, 1388 (7th Cir. 1986), cert. denied, 481 U.S. 1038, 107 S.Ct. 1975, 95 L.Ed.2d 815 (1987).

A given geographic area may be a relevant market notwithstanding: (1) sales by firms within that area to customers outside that area; or (2) sales by firms producing outside that area to customers within that area. Philadelphia National Bank, 374 U.S. at 359-61 nn. 36, 37, 83 S.Ct. at 1739-40 nn. 36, 37.

Based on the evidence and the foregoing legal principles, I find as an ultimate fact and conclude that the relevant geographic market for the HFCS line of commerce is the United States.

FINDINGS OF FACT RESUMED STATISTICAL CONCENTRATION

(94) A concentration ratio is a measure of concentration in an industry, such as the 2-firm concentration ratio (sum of the market shares of the two largest in a market), the 4-firm concentration ratio (the sum of the market shares of the four largest firms in a market), and the 8-firm concentration ratio (the sum of the market shares of the eight largest firms in a market). Another measure of concentration in an industry is the Herfindahl-Hirschman Index ("HHI"). The numerical value of the HHI for an entire market is calculated by summing the squares of the individual market shares of all the firms in such market. Markets with an HHI greater than 2000 are highly concentrated.

(95) The acquisition of Nabisco's corn wet milling business by ADM took place in a somewhat concentrated industry. The number of firms producing HFCS increased from one (CCP) in 1970 to eleven United States firms and one Canadian firm in 1981. But by the beginning of 1983 only nine firms were manufacturing HFCS in the United States and Canada.

(96) Data on production and capacity of HFCS for 1981, the last full year before the lease, and 1989, the last full year before the trial, establish on a production basis a total of HHI of 1745 for 1981 and a total HHI of 2050 for 1989, and establish on a capacity basis a total HHI of 1797 for 1981 and a total HHI of 2194 for 1989.

(97) The data also show the following changes in concentration measurements:

(98) The increase in the concentration after 1981 is attributable in part to the lease and in part to expansions of capacity and production by Cargill, Inc. ("Cargill") and CPC.

(99) Changes in the shares of total HFCS production held by individual companies moved dramatically in both directions.

(100) Measured by the HHI index used in the Merger Guidelines of the Department of Justice, concentration of HFCS production declined steeply from the early 1970's until it reached a level of approximately 2000 in 1978, which was four years prior to the lease. In the period from 1979 to 1989, the HHI measurement has been relatively flat, dipping slightly below 2000 in 1979-82, rising slightly above 2000 in 1983-85, and then trending back a little to barely above the 2000 level from 1986 to 1989. The following exhibit illustrates these trends:

(101) Although HFCS consumption has more than doubled since 1981, there were no new HFCS plants built after that year, although some plants have increased their HFCS production capacity. The number of firms producing HFCS has declined to seven. Those firms, which have a total of sixteen operating plants in the United States and three in Canada, are discussed below.

(102) ADM manufactures HFCS at its corn wet milling plants in Clinton, Cedar Rapids, and Decatur. ADM's closed Montezuma plant has HFCS manufacturing capacity, but there is scant likelihood it will reopen. In 1989 ADM's operating plants had the capacity to produce more pounds of HFCS than any other producer.

(103) Staley, which had the second largest HFCS production capacity in 1989, is headquartered in Decatur, Illinois, and is a subsidiary of Tate Lyle, p. 1.c., a United Kingdom corporation. Staley manufactures HFCS at plants located in Decatur, Illinois, Lafayette, Indiana, and Louden, Tennessee. In addition, it owns a nonoperational corn wet milling plant in Morrisville, Pennsylvania.

(104) Cargill, which had the third largest HFCS production capacity in 1989, is a privately owned agribusiness firm headquartered in Minneapolis, Minnesota. Cargill manufactures HFCS at corn wet milling plants located in Dayton, Ohio, Memphis, Tennessee, and Eddyville, Iowa.

(105) CPC, which had the fourth largest HFCS production capacity in 1989, is headquartered in Englewood Cliffs, New Jersey. CPC manufactures HFCS at corn wet milling plants in Argo, Illinois, Winston-Salem, North Carolina, and Stockton, California. It also has plants in London, Port Colborne and Cardinal, Canada.

(106) American Fructose, Inc. ("AFC"), which had the fifth largest HFCS production capacity in 1989, is a partially-owned subsidiary of American Maize Products Corp. ("AMPC"), headquartered in Stamford, Connecticut. AMPC is a corn wet miller but does not directly produce HFCS. AFC manufactures HFCS at plants located in Decatur, Alabama and Dimmitt, Texas. It acquired the Dimmitt plant in 1985 from Amstar, Inc.

(107) Hubinger, Inc. ("Hubinger"), which had the sixth largest HFCS production capacity in 1989, is headquartered in Keokuk, Iowa, and is a wholly owned subsidiary of H.J. Heinz Co., of Pittsburgh, Pennsylvania. Hubinger manufactures HFCS at a corn wet milling plant in Keokuk.

(108) Coors Biotechnology, Inc. ("Coors"), which had the smallest HFCS production capacity in 1989, is headquartered in Golden, Colorado, and is a wholly owned subsidiary of Adolph Coors Brewing Co. Coors manufactures HFCS at a plant located in Johnstown, Colorado, which it acquired from the Great Western Sugar Co. in 1981.

(109) By 1992, total industry HFCS capacity is expected to increase more than one billion pounds over 1989. Cargill's 1992 capacity will probably be up. Cargill, the seventh largest producer in 1980, will probably eclipse Staley as having the second largest HFCS capacity. ADM and a couple other companies will probably increase capacity.

(110) ADM's share of HFCS capacity in 1983, the year after the lease, was 33% This was 3.1% greater than ADM's 29.9% share of HFCS capacity immediately prior to the lease, and was below the 33.6% share of capacity that ADM realized in its year of entry into the HFCS business in 1976. By January of 1990, ADM's share of total HFCS capacity had declined to 29.4%; and, based on the results of the government's industry survey, ADM's capacity share is projected to decline further to 28.6% by January of 1992.

(111) The following shows the growth of HFCS production from 1981, the year before the lease, through 1989.

Total HFCS Production Percent of 1981 (Year (Dry Basis-Millions of Prior to Lease) Lbs) 1981 5,604 100% 1982 6,418 115% 1983 7,565 135%

Total HFCS Production Percent of 1981 (Year (Dry Basis-Millions of Prior to Lease) Lbs) 1984 8,701 155% 1985 10,593 189% 1986 11,029 197% 1987 11,500 205% 1988 12,073 215% 1989 12,203 218%

(112) In 1982, ADM's production of HFCS (excluding production at Nabisco's plants prior to the lease but including such production after the lease) actually accounted for 28.5% of total production. In 1983, ADM's first full year of operating the Clinton plant, ADM's share of total HFCS production rose to 30.7%; but by 1989, ADM's production of HFCS had declined to 28.9% of total production. Some 7.2% of total industry HFCS production in 1989 was attributable to ADM's operation of the Clinton plant.

OTHER FACTORS

(113) The existence of large, powerful buyers of a product mitigates against the ability of sellers to raise prices. Empirical studies have shown that the stronger and more concentrated the buyers' side of the market is, the less is any ability of sellers to elevate their prices. There is a significant concentration among buyers of HFCS. The purchase of HFCS is dominated by a few large, sophisticated and powerful buyers, such as Coca-Cola and Pepsi-Cola. Purchase decisions made by even one of these so-called "power buyers" can directly affect the market price for HFCS. Details of this buyer concentration and its significance follow.

(114) In the years since 1982, there has been a growing consolidation among HFCS buyers. They are becoming larger and more powerful, and they have more leverage over HFCS sellers. The consolidation has occurred both as a result of acquisitions and from the formation and continual enlargement of buying cooperatives.

(115) The largest 20 buyers of HFCS 55 from ADM represented more than 90 percent of its sales of HFCS 55 in 1990, compared to 64 percent in 1983. The largest 20 buyers of HFCS 42 from ADM represented more than 50 percent of its sales of HFCS 42 in 1990, compared to 39 percent in 1983. The largest 20 buyers of HFCS 42 and 55 combined represented 60 percent of ADM's total 1990 HFCS sales, compared to 40 percent in 1983.

(116) the consolidation among HFCS buyers is most pronounced in the beverage industry, which uses approximately 70 percent of HFCS output. There were approximately 2,000 soft drink bottlers in the United States ten years ago, and in 1990 the number was down to about half that. Pepsi had 450 bottlers ten years ago; today there are about 150. Consolidation has also occurred in other sweetener-using industries. For example, Phillip Morris now owns Kraft, General Foods and Miller Brewing, which formerly were separate companies that purchased sweeteners independently. In 1989, these three Philip Morris subsidiaries formed a buying group that purchases sweeteners jointly. There has also been substantial consolidation of buyers in the dairy and banking industries.

(117) The Coca-Cola Company has ten syrup manufacturing plants in the United States that produce sweetened soft drink syrup. This syrup, which is converted to finished product by addition of carbonated water, is sold to franchise bottlers as well as to companies such as McDonalds and Burger King and others that sell ready-to-drink soft drinks. Coca-Cola also manufactures unsweetened soft drink concentrate which is sold to franchise bottlers who add both carbonated water and sweetener to make a finished product.

(118) The Coca-Cola Company soft drink brands marketed by Coca-Cola and its franchise bottlers account for approximately 40 percent of the national carbonated soft drink market. Sweeteners are centrally purchased for Coca-Cola and the great majority of its franchise bottlers by Stanley Konzal, Coca-Cola Company Director of Sweetener and Ingredient Purchasing. In 1990, the total volume of HFCS purchased by Mr. Konzal for Coca-Cola and its franchisees was approximately 2.9 billion pounds dry basis, which represents approximately $500 to $600 million.

(119) Mr. Konzal purchased approximately 75 percent of the Coca-Cola system's sweetener needs in 1982. Thereafter, this percentage declined a little bit, then significantly increased. At the time of trial, Mr. Konzal purchased 85 to 87 percent of the system's sweetener needs in the United States. The increase is substantially due to growth in the Coca-Cola Sweetener Purchasing Co-op. The Co-op is composed of independent Coke franchise bottlers for whom the Coca-Cola Company purchases sweetener. It was formed in 1981 with one bottler. By 1990 the Co-op included 27 companies operating 110 bottling and canning plants in the United States. One of the Co-op members is Coca-Cola Enterprises, which was organized in 1985 or early 1986 and acquired 25 or 30 formerly independent Coca-Cola bottlers. Numerous Coca-Cola franchisees who formerly were direct customers of ADM now buy their sweetener through the Co-op.

(120) The volume of Pepsi-Cola's HFCS purchases has grown substantially over the last five years as Pepsi has acquired more. and more bottling operations. In 1990, Pepsi-owned bottling operations accounted for slightly over 50% of total United States sales of the Pepsi-Cola family of soft drink brands. Independent bottlers, who purchase their own sweetener directly or through co-ops, accounted for the remainder. Five years ago, when Pepsi owned only 15% of its United States bottling operations, its volume of annual HFCS purchases was much lower than its current purchases of 15 million hundredweight (1.5 billion pounds dry basis).

(121) Independent soft drink bottlers are increasingly forming cooperative buying groups.

(122) Wis-Pak is a Pepsi-Cola canning cooperative owned by 25 bottling companies with 50 franchises operating in nine states in the upper Mid-West. It purchases HFCS for its members, as well as for members of several other Pepsi-Cola bottling organizations. The intent of Wis-Pak is to increase its purchasing volume so as to be able to obtain lower HFCS prices. Wis-Pak purchased 275 million pounds of HFCS in 1989, and anticipates that its 1990 purchases will amount to nearly 400 million pounds.

(123) Wis-Pak is presently purchasing HFCS from ADM and three other suppliers. In 1990, ADM accounted for 75 percent of Wis-Pak's HFCS purchases. In 1989, ADM had only 30 percent. Purchases from ADM increased from 30 percent to 75 percent because ADM gave Wis-Pak the most favorable arrangement in terms of a guaranteed supply. ADM's price was better than some, but not all, of its competitors.

(124) Wis-Pak and several other, smaller co-ops of soft drink bottlers have also joined together to form a group known as Master Purchasing Cooperative ("MPC"). Among the members of MPC are Clinton's Ditch and Carolina Canners. Clinton's Ditch was formed by 18 bottlers and was the first Pepsi bottlers' co-op in the country. It has declined to 12 members due to consolidations. Carolina Canners was started by 35 Pepsi bottlers and is now a 120-million pound annual buyer of HFCS on behalf of its members and other small Pepsi bottlers located in the Carolinas. While MPC was intended to be a co-ops' coop for purposes of HFCS purchasing leverage, in practice Wis-Pak has taken over the purchasing on behalf of some of the other co-op members.

(125) There are other sizeable soft drink purchasing co-ops, as well as purchasing co-ops in other industries, such as baking and dairy, that are large buyers of HFCS. Another large buyer is Pepsi-Cola General Bottlers, Inc. of Rolling Meadows, Illinois, the largest independent bottler of PepsiCola in the United States.

(126) The power buyers and other large buyers use numerous tactics to obtain low prices for HFCS, including:

(a) Refusal to reveal the prices quoted by other suppliers and the price which a supplier must meet to obtain or retain business, creating uncertainty among suppliers.

(b) Swinging large volume back and forth among suppliers to show each supplier that it better quote a lower price to obtain and keep large volume sales.

(c) Delaying agreement to a contract and refusing to purchase product until a supplier accedes to acceptable terms.

(d) Holding out the threat of inducing a new entrant into HFCS production and assuring the new entrant adequate volume and returns.

(127) The power buyers and other large buyers, who purchase the vast majority of HFCS produced and marketed, are tough, experienced, sophisticated negotiators, and they consistently negotiate a price below the price announced or listed by the HFCS producer from whom they ultimately purchase.

(128) Since 1982, HFCS prices have fluctuated significantly from year to year, sometimes rising and sometimes falling.

(129) From 1981 to the most recent date for which United States Department of Agriculture ("USDA") information is available, announced prices for HFCS, even when unadjusted for inflation, have declined, as illustrated by the following data published by the USDA:

Year Annual Prices for 55 HFCS (Cents Per Pound — Annual Average) ------ ------------------------------ 1981 23.59 1982 18.81 1983 21.06 1984 22.69 1985 19.95 1986 19.96 1987 17.46 1988 18.68 1989 21.41

(130) The Midwest corn belt suffered a severe drought in 1988 and a good part of the corn crop was lost. This resulted in increased corn prices, which in turn caused the higher HFCS prices in 1989.

(131) HFCS prices declined from 1989 to 1990.

(132) Typically, HFCS manufacturers announce prices for the coming quarter two weeks or less before the beginning of the quarter. Buyers then attempt to negotiate discounts from such published or list prices.

(133) Most HFCS is sold pursuant to written contracts and the typical contract provides for quarterly pricing. Such contracts may and often do provide commitments to buy and sell for a year or longer but provide that prices are to be established separately for each quarter of the year.

(134) HFCS prices respond to changes in supply and demand. They are directly influenced by seasonal demand fluctuations. Prices are highest during periods of peak demand (second and third quarters) and lowest during periods when demand falls off (first and fourth quarters.) Quarterly pricing is a response to business risks and seasonal demand. Such pricing is a more efficient form of pricing than annual pricing. For example, economists and many regulators recognize the efficiency of such "peak load" pricing in the pricing regulation of utilities. In HFCS production, pricing beyond 90 days also impairs the producers' ability to hedge against fluctuations in by-product values.

(135) Capacity utilization also affects HFCS prices. Excess production capacity tends to lower HFCS prices. Relatively higher HFCS prices in 1982, 1983 and 1984 occurred during a period when capacity was tight. Capacity expansions by HFCS producers and consequent excess capacity caused prices to fall in 1985, 1986 and 1987.

(136) HFCS prices are also influenced by significant movement in sugar prices as well as the level of net corn costs.

(137) Although HFCS suppliers periodically announce the "list" prices at which they are willing to sell HFCS under contract or on a spot basis, most HFCS is actually sold at lower prices by all suppliers. Actual prices depart substantially from list prices and vary both among customers and suppliers.

(138) ADM's policy and practice on discounts off of list prices is to be competitive. Due to the provisions of the RobinsonPatman Act, ADM confines its granting of discounts to situations in which it reasonably believes that it is meeting a bona fide offer of a competitor. ADM will apply this "meeting competition" policy in any type of customer situation, be it a present customer, a former customer, or a prospective, new customer.

(139) ADM's "meet competition" policy has resulted in sizeable and persistent discounts, as measured by the difference between ADM's average list prices and its average actual sales prices.

(140) Although buyers grumble about the HFCS industry being a tough one with which to make negotiating headway, the fact is that small as well as large buyers are able to negotiate meaningful discounts from list price. (Cadbury Schweppes Beverages, according to the deposition testimony of its Director of World Sourcing, Allen P. Hagstrand, appears to be an exception. Mr. Hagstrand testified that he is not successful in trying to negotiate a price below list. That may be the result of his approach to purchasing; he does not play one supplier off against another because he is more interested in establishing a relationship with a buyer than getting the lowest price.)

(141) Discounts from list price are made available to customers in a variety of ways. Discounts are sometimes reflected directly on the customer invoice. Discounts are more commonly given in the form or rebates to customers. Discounts are also made available in negotiated freight rates or pick up allowances, or in giving a customer the benefit of a better price applicable in another pricing region.

(142) Although most contract negotiations are completed between the time of the quarterly price announcement and the beginning of the quarter, negotiations can sometimes take several weeks, and suppliers may, during negotiations, drop their prices several times.

(143) Actual transaction prices are maintained in confidence by sellers and buyers of HFCS. Such confidentiality makes coordinated pricing behavior difficult.

(144) HFCS supply contracts typically contain a "meet or release" clause that releases a buyer from its purchase obligations if the seller fails to meet a better price offered by a competitor during the course of a contract.

(145) Buyers continually play suppliers off against one another, cutting back or discontinuing purchases from sellers as a means of obtaining the lowest possible prices. As a result, purchaser-supplier relationships are highly unstable, with suppliers, including ADM, frequently losing business, regaining it, and losing it again on price grounds — sometimes within the same quarter year.

(146) In the early 1980's HFCS was sold under a variety of pricing mechanisms, including day to day, monthly, quarterly, and yearly.

(147) Before 1982, not much HFCS was sold on an annual contract basis. Most sales were priced on a spot basis or under 30 to 90 day contracts.

(148) By the time of trial, HFCS was sold under a variety of pricing arrangements, including on a spot basis, price date of shipment, quarterly pricing and annual pricing.

(149) Though seasonal pricing is the most efficient form of pricing for HCFS, buyers are able to obtain contracts that give them price protection beyond a quarter, including firm annual price contracts, tolling agreements and other arrangements if they are willing to compensate HFCS suppliers for the greater risks inherent in such arrangements or to accept some of the risks of more extended period pricing.

(150) ADM's policy on the duration of contracts for its corn products has always been "the shorter the better" to limit the risk.

(151) Despite ADM's preference for quarterly pricing in the sale of HFCS, a review of its twenty largest customers for HFCS (42 and 55 combined) revealed that, in ADM's 1990 fiscal year, customers accounting for over 43% of ADM's total HFCS sales were actually purchasing on a set price basis that extends for longer than a calendar quarter. In the case of HFCS 55, over two-thirds of its major customers are purchasing on longer-than-quarterly pricing terms.

(152) One of the most significant developments since 1982 has been the growth in the use of tolling agreements for HFCS. These agreements introduce variety and complexity into the negotiated relationships between HFCS purchasers and suppliers. Tolling agreements affect the duration of supply relationship, the pricing of the product, and the allocation of business risk among the parties. While tolling agreements are particularly successful purchasing strategies in the hands of power buyers such as Coca-Cola and Pepsi-Cola, their benefits are now also being passed on to smaller HFCS users through co-operative buying groups.

(153) A tolling agreement is an agreement whereby a corn wet miller will process a customer's corn into HFCS and charge the customer a tolling fee, with the sale of byproducts credited to the customer's account. Sometimes the wet miller will sell the byproducts for the customer, and sometimes the customer itself will sell the byproducts. Tolling agreements are typically for a duration of a year or longer. There are two HFCS producers that typically enter into multi-year tolling agreements.

(154) If managed properly, an HFCS tolling agreement can achieve better pricing for the customer than is available on the open market. Tolling agreements, however, have a high degree of risk.

(155) While ADM does not encourage tolling contracts, it has entered into several such contracts in the past two years, covering various volumes of HFCS. One ADM agreement gives the buyer a "market toll option" under which the buyer can choose whether to be priced on the basis of an agreed discount from the quarterly list price or on the basis of a toll. One tolling agreement was entered into by ADM because the buyer threatened to take away business unless ADM agreed to the toll.

(156) The HFCS business is characterized by large infrequent transactions, although there are a number of smaller "spot" transactions.

(157) There are significant differences in costs and product mix among firms in the HFCS industry, a factor that makes coordination among those in an industry more difficult.

(158) The HFCS industry is highly capital intensive, which results in a high ratio of fixed to variable costs, a strong incentive to keep capacity highly utilized. This feature of the HFCS industry mitigates against disciplined pricing strategies.

(159) The consumption of HFCS during the second and third quarters of each calendar year is in quantities that are near the total capacity of HFCS manufacturers. During the first and fourth quarters of the year HFCS consumption is substantially less than in the second and third quarters. Consequently, there is substantial excess capacity during the first and fourth quarters of the year. This seasonal excess capacity, combined with the high ratio of fixed to variable costs of production, mitigates against any successful coordinated pricing.

CONCLUSIONS OF LAW (Including Ultimate Fact Findings) RE: STATISTICAL CONCENTRATION AND OTHER FACTORS

Section 7 of the Clayton Act reflects a congressional determination that a merger which "produces a firm controlling an undue percentage share of the relevant market, and results in a significant increase in * * * concentration * * * is so inherently likely to lessen competition substantially that it must be enjoined in the absence of evidence clearly showing that the merger is not likely to have such anticompetitive effects." Philadelphia National Bank, 374 U.S. at 363, 83 S.Ct. at 1741.

The presumption that increased concentration is likely to lead to a reduction in competition is based on the rationale that a reduction in the number of competitors in a market makes it easier for competitors to coordinate pricing and other terms of sale, and thus more likely that they will do so. Hospital Corp. of America, 807 F.2d at 1387.

Common devices for assessing concentration levels are the HHI and the firm concentration ratios. See FTC v. PPG Industries, 798 F.2d 1500, 1503 (D.C. Cir. 1986); United States v. Rice Growers Ass'n of California, 1986-2 Trade Cas. (CCH) ¶ 67,288, at 61,466, 1986 WL 12562 (E.D.Cal. Jan. 31, 1986) (the HHI "is a common index used by economists to describe concentration levels in particular markets.")

Looking only at the HHI and firm concentration ratios drawn from the HFC production and capacity facts in this case, it can be concluded that the lease is "inherently likely to lessen competition substantially." Philadelphia National Bank, 374 U.S. at 363, 83 S.Ct. at 1741. The lease should not be enjoined, however, if there is other "evidence clearly showing that the [lease] is not likely to have such anticompetitive effects." Id. "[O]ther factors [must] * * * be considered in evaluating the probable effects of a merger in the relevant market * * * ." Brown Shoe, 370 U.S. at 344, 82 S.Ct. at 1534; United States v. Marine Bancorporation, 418 U.S. 602, 631, 94 S.Ct. 2856, 2874-75, 41 L.Ed.2d 978 (1974); United States v. Citizens and Southern Nat'l Bank, 422 U.S. 86, 120, 95 S.Ct. 2099, 2118-19, 45 L.Ed.2d 41 (1975). As the District of Columbia Circuit recently put it: "The Herfindahl-Hirschman Index cannot guarantee litigation victories." United States v. Baker Hughes, 908 F.2d 981, 983 (D.C. Cir. 1990).

Since the Supreme Court's 1974 decision in General Dynamics, 415 U.S. 486, 94 S.Ct. 1186, it has been clear that to overcome the government's statistical proof, the defendant need only introduce evidence sufficient to show "that the prima facie case inaccurately predicts the relevant transaction's probable effect on future competition." Baker Hughes, 908 F.2d at 991.

The Eighth Circuit recognizes the importance of "other factors" in evaluating a merger's likely competitive effects. National Tea Co., 603 F.2d 694. As explained in the Eighth Circuit's opinion:

In General Dynamics, the Court simply recognized that while market share is an important indicator of a firm's future competitive strength, other factors may discount its significance.

Because violators of the antitrust laws could purposely refrain from anticompetitive behavior while an antitrust suit is threatened or pending, too much weight should not be given to postacquisition events. General Dynamics, 415 U.S. at 504-05, 94 S.Ct. at 1197-98. It is uncontested, however, that evidence of postacquisition events is admissible and may be accorded some weight.

The HHI and firm concentration ratios that are sufficient to make out a prima facie case for the government have continued from 1982 to the time of trial. They went up after 1981 partly as a result of the lease. The HHI has remained relatively flat, barely above 2000.

The evidence fails to show, however, any form of coordinated pricing or price leadership in the HFCS industry with respect to actual transaction prices, and fails to show any likelihood of it. There is no evidence of a pattern of disciplined or coordinated pricing. Indeed, the evidence reveals precisely the opposite — a vigorous, competitive struggle for business by negotiated, competitive pricing. The court further notes that the HFCS industry since the lease has been one in which both costs and prices have fallen and output has increased. Competition has been intense and deep discounting and innovative contract arrangements have been the norm as suppliers have struggled to obtain and retain business. Market shares have fluctuated.

There are significant factors that are not, and cannot be, controlled by defendants that lead the court to conclude that the competitive events since the lease are legitimate rather than manipulated by ADM and other HFCS producers because of the pendency of this suit. These factors include the large and infrequent nature of HFCS transactions, the secrecy of transaction prices, and, especially, the negotiating power of the power buyers and large buyers.

The courts recognize the competitive significance of large and sophisticated buyers in the relevant market, including markets with only a few sellers. For example, the District of Columbia Circuit recently approved of the district court's treatment of large buyers as a significant other factor in the Baker Hughes case:

The second non-entry factor that the district court [properly] considered was the sophistication of HHUDR ["hard rock hydraulic drilling rigs"] consumers * * *. These products are hardly trinkets sold to small consumers who may possess imperfect information and limited bargaining power. HHUDR buyers closely examine available options and typically insist on receiving multiple, confidential bids * * *. This sophistication [the district] court found, was likely to promote competition, even in a highly concentrated market.

In another recent case, the District of Columbia Circuit emphasized that "[w]ell-established precedent and the United States Department of Justice Merger Guidelines recognize that the sophistication and bargaining power of buyers play a significant role in assessing the effects of a proposed transaction." The existence of large and sophisticated purchasers "make[s] any anticompetitive consequences very unlikely." FTC v. R.R. Donnelley Sons Co., 1990-2 Trade Cas. (CCH) ¶ 69,239, at 64,852, 64,855, 1990 WL 193674 (D.C. Cir. 1990).

The evidence in this case shows that the buying side of the HFCS industry is populated by very large and sophisticated purchasers and there is a continuing trend toward increasing concentration on the buying side, as large bottlers purchase formerly independent bottling franchises or bring them under their sweetener purchasing wings, and as smaller concerns band together in buying cooperatives to increase their purchasing leverage.

That this consolidation of buying power is an effective means of counteracting any potential market power that might be exercised by sellers is borne out by both economic theory and the facts. Buyers have successfully used a variety of tactics to obtain low prices from HFCS suppliers, including playing off suppliers against one another, swinging volume back and forth among suppliers, disciplining sellers by cutting them off entirely, successfully insisting on year long or multi-year tolling agreements, and holding out the threat of inducing a new entrant into HFCS production.

There is no question that the size and sophistication of buyers in the HFCS industry is a powerful "other factor" that strongly mitigates against the possibility of any attempt by HFCS suppliers to raise prices anticompetitively.

Evidence concerning the difficulty of successfull price collusion is an important "other factor" to be taken into account in evaluating the likely competitive effects of a merger. As the Seventh Circuit recently observed, "the acquisition of a competitor has no economic significance in itself; the worry is that it may enable the acquiring firm to cooperate (or cooperate better) with other leading competitors on reducing or limiting output, thereby pushing up the market price." Hospital Corp. of America, 807 F.2d at 1386.

The Merger Guidelines of the Department of Justice include a laundry list of factors that may "relate to the ease and profitability of collusion," and are therefore relevant "other factors." The Guidelines expressly recognize that whether deviations from a hypothetical collusive price arrangement can be detected and the "cheating" party can be punished by the other participating firms bears directly on whether anticompetitive behavior is probable. Merger Guidelines at section 3.42.

Similarly, the Guidelines recognize that certain market characteristics may affect the incentive to "cheat" on collusive pricing schemes and thus are important factors in evaluating the likelihood of anticompetitive conduct. Thus, for example, "[i]f orders for the relevant product are frequent, regular and small * * * collusion is more likely to succeed because the benefits of departing from the collusive agreement in any single transaction are likely to be small relative to the potential costs." Merger Guidelines at section 3.42.

This court finds and concludes that price collusion among HFCS suppliers would be very difficult to administer, and would provide incentives to cheat.

There is no dispute in the record that actual transaction prices for the sale of HFCS are regarded as highly confidential by both suppliers and purchasers. This factor makes the potential for industry collusion remote, because, as Professor Scherer testified: "[S]ecrecy is the antithesis of successful collusion."

The sale of HFCS is also characterized by relatively large, infrequent transactions. Most HFCS is sold pursuant to contracts having a duration of at least a quarter, and much is sold under arrangements extending a year or longer. Witnesses for both sides agreed that where transactions are large and relatively infrequent it is difficult to maintain a coordinated pricing scheme.

HFCS production is capital intensive. The high ratio of fixed to variable production costs creates strong economic incentives to produce at close to full capacity. This economic incentive to maintain full production, combined with seasonal excess capacity, works against the likelihood of any collusive price raising scheme which would require output restrictions.

The record reflects that there are significant differences in production costs and product mix among firms in the industry. These differences would make agreement on a single collusive price difficult to achieve.

The post-lease events, while viewed with caution, show the HFCS industry to be a highly competitive one. Substantial discounting is the norm and suppliers frequently gain, lose and regain business on price grounds. Sellers compete not only on the basis of price but on the basis of their willingness to offer tolling agreements (effectively a short term lease of a portion of a supplier's HFCS capacity) and other innovative arrangements. Since the lease, HFCS capacity has dramatically increased and prices, even unadjusted for inflation, have mainly fallen. Even unadjusted for inflation, prices are lower now than they were before the lease. Suppliers have reduced their costs, improved their labor productivity as well as the quality of their product, and introduced new products. This is not the performance expected of an industry behaving in a less than competitive fashion.

In sum, the defendants have made an adequate showing by their evidence of other factors sufficient to overcome the government's prima facie case based on statistical concentration, and the government has failed to prove that the lease entered into in 1982 between the defendants is likely to substantially lessen competition or tend to create a monopoly in the HFCS industry.

ORDER FOR JUDGMENT