Beatrice Foods Co

Text (OCR of the scan at left; may contain errors)

IN THE MATTER OF BEATRICE FOODS CO., ET AL.

DISMISSAL ORDER, OPINION, ETC., IN REGARD TO ALLGED VIOLATION OF SEC. 5 OF THE FEDERAL TRADE COMMISSION ACT AND SEC. 7 OF THE CLAYTON ACT Doket 9112. Complaint, June 1978-Final Order, May, 1983 This Final Order dismisses the complaint challenging Beatrice Foods Co. 's acquisition of Tropic ana Products Inc. The Commission found that the loss of actual competition in the ready to-serve orange juice industry was too little to establish a viola tion of the Clayton Act.

Appearances For the Commission: Katherine Boland, Layn R. Philips, Timothy B. Walthall and Ann Thalia Lingos. For the respondents: Edward L. Foote, John W Stack, David S. Acker, Jerome W Pope and Sidney Margolis, Winston Strawn Chicago, III. for respondent Beatrice Foods Co. and Robert R. Feagin Ill, Holland Knight, Tallahassee, Fla. for respondent Tropicana Products, Inc.

COMPLAINT The Federal Trade Commission, having reason to believe that the above named respondents, each subject to the jurisdiction of the Commission, have entered into a merger agreement and Beatrice has also entered into a stock purchase ageement, each of which, if effected would violate Section 7 of the Clayton Act, as amended, 15 U. C. 18, and Section 5 of the Federal Trade Commission Act, as amended, 15 C. 45, that said agreements constitute violations of Section 5 of the Federal Trade Commission Act, as amended, and that a proceeding in respect thereof would be in the public tnterest, hereby issues its complaint, pursuant to Section 11 of the Clayton Act, 15 U. C. 21, and Section 5(b), of the Federal Trade Commission Act, 15 U. 45(b), stating its charges as follows:

1. DEFINITION 1. For the purpose of this complaint, the following definition shall Complaint 101 F.T.C. apply: ready to serve orange juice is chilled single strength orange juice.

II. BEATRICE FOODS CO.

2. Beatrice Foods Co. (Beatrice), is a corporation organized, existing, and doing business under the law of the State of Delaware, with its principal place of business at 120 South Lasalle Street, Chicago, Illinois. (2) 3. Beatrice is engaged in the processing and distributing of food products and related dairy products, and in the manufacturing and distribution of various other products. Beatrice operates the nation third largest dairy system. In its fiscal year ending February 28, 1977 Beatrice had total net sales of $5 288 577 780 and net earnings of $182 566 209.

4. As a part of its food processing and distributing activities, Beatrice in the past has and continues to purchase bulk frozen orange juice concentrate for the purpose of reconstituting it into ready to serve orange juice. Beatrice has ready to serve orange juice for resale. III. TROPICAN A PRODUCTS, INC.

5. Tropicana Products, Inc. (Tropicana) is a corporation organized existing and doing business under the law ofthe State of Florida with its principal place of business at 100113th Avenue East, Bradenton Florida.

6. Tropicana is engaged in, among other things, processing and distributing citrus products. In its fiscal year ending August 31 1977 Tropicana had total net sales of $244 583 000 and net earnings of $22 461 000. In 1977, Tropicana ready to serve orange juice was the largest selling brand in the United States. Tropicana sells ready to serve orange juice to others, such as Beatrice, for resale to grocery stores.

IV. JURISDICTION 7. At all times relevant hereto, Beatrice and Tropicana have sold and shipped products in interstate commerce and engaged in commerce as "commerce" is defined in Section 1 of the Clayton Act, as amended, 15 U. C. 12, and each is a corporation whose business is in or affecting commerce as "commerce" is defined in Section 4 of the Federal Trade Commission Act, as amended, 15 U. C. 44. u...~~ 733 Complaint V. THE MERGER AGREEMENT AND STOCKJ?URCHASE AGREEMEN:r 8. On or about April 27, 1978, Beatrice and Tropicana entered into a merger agreement which provides inter alia for the merger of Tropicana into a subsidiary of Beatrice. Upon consummation of the merger, Tropicana wil become a wholly-owned subsidiary of Beatrice. (3) 9. On or about March 4, 1978, Beatrice entered into an agreement with two private foundations, the Aurora Foundation and Bible Alliance, Inc., under which Beatrice agreed to purchase from those foundations an amount of Tropicana common stock which accounts for approximately 15.4% of the currently outstanding shares of Tropicana common stock. The purchase of such shares is to occur on October 31, 1978, or on such earlier date as Beatrice may specify in writing.

VI. TRADE AND COMMERCE 10. A relevant line of commerce in which to assess the probable competitive effects of the merger is the processing, distribution and sale of ready to serve orange juice.

11. Relevant sections ofthe country in which to assess the probable competitive effects ofthe merger are the United States as a whole and submarkets thereof.

12. The retail distribution and sale of ready to serve orange juice is concentrated, with the top 4 firms accounting for approximately 58.6% of total sales in 1976.

VII. ACTUAL COMPETITION 13. Tropicana and Beatrice are and have been for some time actual competitors of each other in the processing, distribution and sale of ready to serve orange juice and actual competitors of others engaged in the processing, distribution and sale of ready to serve orange juice throughout the United States and submarkets thereof. 14. In 1976, Tropicana s market share was nearly twice the size of its nearest competitor, with Tropicana accounting for approximately 30.5% of ready to serve orange juice sales in dollar volume and 29. in unit volume (gallons). In 1976, Beatrice accounted for approximately 1 to 1.7% of ready to serve orange juice in unit volume. Inital Decision 101 F. VIII. EFFECTS; VIOLATIONS CHARGED 15. The effects of the proposed merger may be to substantially lessen competition or tend to create a monopoly in a relevant market in violation of Section 7 of the Clayton (4) Act, as amended, 15 U. 18, and Section 5 of the Federal Trade Commission Act, as amended 15 U. C. 45, in the following ways, among others: (a) substantial actual competition between Beatrice and Tropicana in the processing, distribution and sale of ready to serve orange juice in the United States and submarkets thereof wil be eliminated; (b) actual competition between competitors generally in the processing, distribution and sale of ready to serve orange juice may be lessened;

(c) Tropicana wil be eliminated as the major independent competitor in the processing, distribution and sale of ready to serve orange juice;

(d) the previously existing level of concentration in the processing, distribution and sale of ready to serve orange juice wil be increased and the possibilities for eventual deconcentration may be diminished; (e) mergers or acquisitions between other ready to serve orange juice processors, distributors and sellers may be fostered, thus causing a further substantial lessening of competition and tendency toward monopoly in the relevant market;

CD processors, distributors and sellers of ready to serve orange juice may be denied the benefits of free and open competition to their detriment and to the detriment of the general purchasing public and ultimate consumer.

INITIAL DECISION BY JAMES P. TIMONY, ADMINISTRATIVE LAW JUDGE NOVEMBER 21, 1980 PRELIMINARY STATEMENT The complaint in this matter, issued on June 20, 1978 alleged that Beatrice Foods Co. (Beatrice) entered into an agreement to purchase stock in Tropicana Products, Inc. (Tropicana) and that Beatrice and Tropicana entered, on April 27, 1978, a merger agreement providing for the merger of Tropic ana into a subsidiary of Beatrice, resulting in Tropicana becoming a wholly-owned subsidiary of Beatrice, and that these agreements violate Section 7 of the Clayton Act, 15 U . C. 18 BEATRICE FOODS CO.. ET AL. 737 733 Initial Decision and Section 5 of The Federal Trade Commission Act, 15 U. C. (2) 345. The effects of the merger are alleged in paragraph fifteen of the complaint to be in the processing, distribution and sale of ready-toserve orange juice in the following ways:

(a) substantial actual competition between Beatrice and Tropicana in the processing, distribution and sale of ready to serve orange juice in the United States and submarkets thereof will be eliminated; (b) actual competition between competitors generally in the processing, distribution and sale of ready to serve orange juice may be lessened;

(c) Tropicana wil be eliminated as the major independent competitor in the processing, distribution and sale of ready to serve orange juice;

(d) the previously existing level of concentration in the processing, distribution and sale of ready to serve orange juice wil be increased and the possibilities for eventual deconcentration may be diminished; (e) mergers or acquisitions between other ready to serve orange juice processors, distributors and sellers may be fostered, thus causing a further substantial lessening or competition and tendency toward monopoly in the relevant market;

(f) processors, distributors and sellers of ready to serve orange juice may be denied the benefits of free and open competition to their detriment and to the detriment of the general purchasing public and ultimate consumer.

Respondents generally denied the allegations of the complaint and raised issues involving (1) the relevant product market, (2) the relevant geographic market, (3) the type of the distribution systems used and the location and type of customers sold by Tropicana and Beatrice, (4) the barriers to entry in the ready to serve orange juice business.

Prior to the merger between Beatrice and Tropicana, the Federal Trade Commission sought a preliminary injunction against the proposed transaction. The application was denied. District (3) Court Judge George Hart determined that the two companies were not competitors in the processing, distribution and sale of ready-to-serve orange juice on a national basis in any sigoificant amount. As a result it did not appear likely that the plaintiff would succeed on the merits. FT v. Beatrice Foods Co. and Tropicana Products, Inc., (1978-2) Trade Reg. Rep. (CCH) TI 62 148 (D. C. 1978). After a motions panel of the Court of Appeals for the District of Columbia affrmed the denial, the Commission fied a motion for rehearing alleging that the district court judge s findings of fact were insuffcient to support his order. The panel vacated its previous order 738 FEDERAL TRAD COMl\ISSION DECISIONS Initial Decision 101 F. and remanded the record to the district court for specific findings of fact. FTCv. Beatrice Foods Co. and Tropicana Products, Inc., (1978-2) Trade Reg. Rep. (CCH) U 62 316 (D.C. Cir. 1978) (587 F.2d 1225 (1978)). Thereafter, the district court entered specific findings and conclusions in support of the order denying injunctive relief. Id. at 75 923- 26.

Judge Hart found that Beatrice and Tropicana were riot competitors and any competition, if present, was de minimis. Id. at 75 926. He further concluded that the United States was not the appropriate geographic market in which to test the actual competition. These findings are not binding here. A preliminary injunction does not dispose of litigation on the merits but is nothing more than tentative judgement of the litigation in order to preserve rights pending its final outcome. Wyrough Loser, Inc. v. Pelmor Laboratories, Inc. 376 F.2d 543 , 548 (3rd Cir. 1967); United States v. School D ,trict of Omaha 367 F. Supp. 179, 193 (D. Neb. 1973). Findings and conclusions accompanying a preliminary injunction are for purposes of just fying that particular order only and have no binding effect on any later jury or other fact finder. Nor do such findings estop either the parties or the fact finder from proceeding with the case on its merits. Westchester General Hospital v. Dept. of HEW, 464 F.Supp. 236 (M. Fla. 1979); Poev. Charlotte Memorial Hospital 374 F.Supp. 1302 1312 (W.D. N. C. 1974); Sierra Club v. Morton 348 F.Supp. 219 (N.D. Ca. 1972). Such a result is particularly important where, as here, the parties had extensive discovery only after the hearing on the preliminary injunction. Findings on the ultimate issues in a case can only be made after the parties have had opportunity to prepare and argue the case in detail.

After substantial discovery and pretrial motion practice, the administrative trial commenced in November, 1979, and was interrupted several times by the Commission s (4) appropriation problems! and by a somewhat languid and negligent fie search by Tropicana in response to discovery subpoenas, which belatedly produced numbers of documents material to the issues of this case.2 These documents were subpoenaed in May and September 1978 and finally produced in April and May 1980 well into respondents' defense hearings. Respondents offered over 500 of these documents as exhibits. This offer was rejected.

In their pretrial brief complaint counsel stated that the relevant product market in which to assess the probable competitive effects of the merger is the retail sale of ready to serve orange juice through I Orders dated March 14, 1980, April 30, 1980 and May 30, 1980. 2 Orders dated April 3, 1980, April 7, 1980, April 9, 1980, April 14, 1980, two orders dated May 21 1980. 3 Complaint counsel also offered many of the documents as exhibil 4 Order dated July 23, 1980.

..... . ... . ~.u~~ ~ ~...... .....

733 Initial Decision Because of the documents belatedly produced by rgrocery stores.5 spondents in the spring of1980, this market was expanded to include processing" and "distribution" of chiled orange juice in the retail market, complaint counsel having found documents allegedly showing Beatrice as a direct horizontal competitor of Tropicana in the processing and in the distribution of chiled orange juice for the retail market.

All objections or motions now pending, not decided by implication in this decision, are hereby denied. (5) The findings of fact include references to supporting evidentiary items in the record. These references are intended to serve as guides to the testimony and t\,e exhibits supporting the findings off act. They do not necessarily represent complete summaries of the evidence supporting each finding. The following abbreviations have been used: Tr. - References to the transcript are designated by the name of the witness and followed by the page number; CX - Complaint counsel' s exhibit, followed by its number and the referenced page(s);

RX - Respondent' s exhibit, followed by its number and the referenced pagers);

CPF - Complaint counsel's proposed finding; RPF - Respondent's proposed finding.

I. DEFINITIONS a. Chilled orange juice COJ) is chiled single strength orange juice made: (1) by squeezing fresh oranges, (2) by adding water to FCOJ, or (3) by thawing frozen single strength orange juice. b. Retail COJ market refers to COJ sold in grocery stores. c. Frozen concentrate orange juice FCOJ) is prepared by taking water out of fresh orange juice and freezing the residue. d. Bulk concentrate is frozen concentrated orange juice shipped in drums or tankers at temperatures below 32 F. (6) e. Canned orange juice (CSSOJ) is single strength orange juice sold in metallic containers.

"Trial Brief fied November 15, 1979 at p 52. ,; Order denying objection to complaint counsel's revised s1.tement ofissues, May 21, 1980. The complaint already alleged this market description. B'urthermore, to avoid prejudice to respondents, they were allowed to recall any witnesses who had already testified, for the purpose of additional examination based on this market contention. Order dated May 21 , 1980 at p. 2 7 Respondents argue that they were denied a fair hearing because of the admission into evidence of documents found in their fies. Without pointiog to any specific document, they make a general at.tack 00 ruings on admis.ability of documents. (RPF at p. 94) The thrust of the objection is that statements and opinions in documents prepared by Beatrice s advertising agency were admitted as admssions. The objection is denied.Owensv. Achinson, Topeka Santa Fe R.R. 393 F.2d 77, 79 (5th Cir. 1968),cat. denied 393 U.S. 855: Coxv. E.c;aShipping Co. 247 F.2d 629. 634 (5th Cir. 1957) j.g.

Initial Decision 101 F. f. Grocery stores are independent or chain grocery or convenience food stores.

Packaging COJ-when a milk plant buys COJ which has already been processed, puts it in containers and distributes it. h. Processing COJmaking COJ from fresh oranges, bulk FCOJ, or bulk frozen fresh orange juice.

i. Reconstituting COJ-making COJ from bulk FCOJ by adding water and sometimes pasturizing it or adding sugar or other preservatives.

Fresh COJjuice made by squeezing oranges. It is sometimes pasturized.

k. Orange drink-fruit juice with more water added than in juice which is freshly squeezed. Orange drink may be diluted by adding to orange juice water amounting to 50% up to 95% of the drink. II. FINDINGS OF FACT A. The Acquiring Firm: Beatrice Foods Co. 1. Beatrice Foods Co. ("Beatrice ) is a corporation doing business under the laws of Delaware, with its principal place of business at 120 South Lasalle Street, Chicago, Ilinois. (Complaint and Answers of Beatrice and Tropicana, U 2) 2. In fiscal 1978, Beatrice had total net sales of $6 313 888 000 and net earnings, after taxes, of $221 538 000. (CX 194A) In 1978 Beatrice was the nation s 31st largest industrial corporation in terms of net sales, 79th in total assets and 57th in terms of net income. (CX 194A; Dutt 3794) Beatrice sells more than 8 000 products. (CX 19B) 3. Beatrice is the nation s leading diversified food company. (CX 527M, no. 49; Dutt 3793; Karnes 4799-4800) Beatrice sells more than 000 retail dairy and grocery products, through supermarkets and grocery stores, using 1 400 food brokers. (CX 20D; CX 31D; Karnes 4798) 4. In 1978 Beatrice was the nation s 17th largest advertiser among all industrial corporations. (CX 192B) (7) 5. In 1978 Beatrice operated the nation s third largest dairy system. (Complaint and Beatrice Answer, U 3) Beatrice is the largest milk processor and distributor in the country and is more national in scope than any other dairy. (CX 509A-E; CX 1554; Karnes 4809) 6. Beatrice s dairy plants process and distribute COJ to grocery stores. (Complaint and Answers of Beatrice and Tropicana U 4) 7. At all times relevant hereto, Beatrice has sold and shipped products in interstate commerce and engaged in commerce as .tcommerce" is defined in Section 1 of the Clayton Act, as amended, 15 733 Initial Decision C. 12, and is a corporation wnose business is in or affecting com merce as "commerce" is defined in Section 4 of the Federal Trade Commission Act, as amended, 15 C. 44. (Complaint and Answers of Beatrice and Tropicana, n 1) B. The Acquired Company: Tropicana Products, Inc. 8. Tropicana Products, Inc. (Tropicana) was a corporation doing business under the laws of the Florida with its principal place of business at 1001 Thirteenth A venue, East Bradenton, Florida. (Complaint and Answers of Beatrice and Tropicana, n 5) 9. Tropicana is engaged in processing and distributing FCOJ, and other citrus juices and drinks (CX 40B-), and is the leading processor ofCOJ in the United States. (Complaint and Tropicana Answer, n 6; CX 40D) 10. In its fiscal year ended August 31, 1977, Tropicana had net sales of $244 583 000 and net earnings, after taxes, of $22 461 000. (Complaint and Answers of Beatrice and Tropicana, n 6) 11. In fiscal 1978 Tropicana sold 106 821 648 gallons of COJ in the United States for $206 649 295. (CX 376A; CX 560Z-17) 12. Tropicana s COJ is supported by intensive brand advertising expenditures and advertisements on national television. (CX 39, p. 6; ex 38E) Tropicana COJ is sold in 48 states, with a sales force and 57 food brokers. (CX 22F, I; CX 378A- 13. At all times relevant hereto, Tropicana sold and shipped products in interstate commerce and engaged in commerce as (!commerce" is defined in Section 1 of the Clayton Act, as (8) amended, 15 C. 12, and was a corporation whose business is in or affecting commerce as "commerce" is defined in Section 4 of the Federal Trade Commission, as amended, 15 U. C. 44. (Complaint and Answers of Beatrice and Tropicana, n 7) C. The Acquisition 14. Mr. Richard Voell, deputy chairman of the board and chief corporate offcer of Beatrice, negotiated the agreement-in-principle to proceed with the acquisition of Tropicana Products, Inc. (RX 2402A) 15. Mr. Richard Truelick, vice president of Beatrice, brought Tropicana to the attention of Mr. Voell sometime in the summer of 1977. (CX 231OA) 16. On September 14, 1977, Beatrice s senior management, other than Mr. Voell, exhibited an interest in acquiring Tropicana. (CX 318) In October of1977, Mr. Voell received a copy of an October 18, 1977 memorandum to James Dutt, Beatrice s chief operating offcer, which analyzed various possible stock exchanges between Beatrice and Tropicana. (CX 320A-G; CX 2310A) Sometime late in 1977, Mr. Voell Initial Decision 101 F. discussed the acquisition of Tropicana with Mr. Wallace Rasmussen then chairman of the board and chief executive offcer of Beatrice. (CX 231OA- 17. Mr. Wiliam Polidoro, a Beatrice dairy division offcial, contacted Tropicana regarding its possible acquisition in the fall of 1977. (Barnebey 4097-98) 18. At Mr. Truelick' s suggestion, Mr. Voell arranged to meet with Mr. Anthony Rossi, then Tropicana s chairman and chief executive offcer, on February 8, 1978. (CX 231OB; CX 321A-C; CX 322; CX 324) On February 28 1978, Messrs. Rasmussen, Voell and Truelick met in Bradenton, Florida with Mr. Rossi and Mr. Barnebey, then president of Tropic ana. (CX 319; RX 2402A- 19. On Saturday, March 4, 1978, Beatrice offcials reached an agreement-in-principle with Tropicana offcials concerning a merger between the two companies. (CX 327 A-C; CX 5280, no. 65; RX 2402B) On the evening of that day, at a special meeting of Beatrice s board of directors, the agreement-in-principle was approved. (eX 231OB; RX 2402B) 20. The first public announcement of the proposed merger came in the March 6 , 1978 edition ofthe Wall Street Journal. (RX 2402B) The proposed merger was mentioned at a Beatrice management meeting on March 7 or 8, 1978. (CX 528Z-21, no. 188) (9) 21. In April 1978, pursuant to the agreement with Beatrice Tropicana Products, Inc., was purchased by the entity T. , Inc., a wholly-owned subsidiary of Beatrice created solely for purposes ofthe acquisition. (Complaint and Answers of Beatrice and Tropicana, nn 8 9) Immediately following the purchase, T. , Inc., changed its name to Tropicana Products, Inc. Under the terms ofthe purchase arrangements, 52% of Tropicana s shares were converted into Beatrice preferred stock, and 48% were purchased outright by Beatrice for cash. Tropicana is now a wholly-owned subsidiary of Beatrice. (CX 22 Annex I, p. 1) D. Product Market a. The Retail COJ Market 22. The sale of orange juice to consumers is divided into three segments: FCOJ, COJ and CSSOJ. (CX 345A-Z-89; CX 347Z-, Z-7; ex 353A; CX 394; CX 400A; CX 410 I-K; CX 413 I-N; CX 414C-E; CX 415B; CX 423B- , L, R; CX 432A-C; CX 505C; CX 512A-P; CX 513L; CX 514A-R; CX 1650, p. 125; CX 2302, p. 19; Munkelt 748; Jessup 2877 2879; Barnebey 4008) 23. COJ is recognized by members of the industry and public offcials as a market separate from FCOJ and CSSOJ. (CX 6C; CX 13A-J; BEATRICE FOODS CO.. ET AL. 743 733 Initial Decision CX 147; CX 149B-C; CX 353A; CX 410K; CX 412H, J; CX 4131, M; CX 414E; CX 415B- , E; CX 432B; ex 505C; CX 512A-P; CX 513L; CX 514A-R; Munkelt 736-37) 24. Tropicana and Beatrice have recognized the existence of the COJ market in their fiings with the Securities and Exchange Commission ("SEC"). (CX 22F; CX 38E; CX 39C; CX 40B, D; CX 41B; CX 42B; CX 527Z-1) 25. The Florida Citrus Processors Association ("FCP A") is a trade association of Florida citrus processors. (CX 594J, p. 4) FCP A, formerly the Florida Canners Association, collects and disseminates to its members separate data on COJ, CSSOJ and FCOJ. (CX 594J, pp. 4- , 14, 66-7; Barnebey 3992) 26. The federal and state governments have promulgated different standards for FCOJ and CSSOJ, and COJ. (CX 399C) 27. The Florida Department of Citrus ("FDC"), a state agency financed by the citrus growers, has recognized separate markets for COJ, FCOJ and CSSOJ. (Hoffer 1813, 1816-17) 28. Separate advertising campaigns for COJ are used (10) by firms which sell COJ and FCOJ and/or CSSOJ. (CX 337 A-Z-21; CX 367; CX 527Z-) b. The Retail COJ Market is Distinct From the Institutional COJ Market 29. The COJ market is recognized by the industry as being divided into two submarkets: sales to grocery stores ("the retail market") and sales to institutional customers ("the institutional market"). (CX 505E; CX 1650, pp. 33-34; Munkelt 717-18; Lang 1049; Donovan 1297- 98; Miler 1481; Jessup 2882; Barnebey 4014) 30. Tropicana recognizes the institutional COJ market as separate from the retail COJ market. (CX 523, pp. 298-99; CX 1650, pp. 168-9) 31. Industry and government documents separately report on conditions in the COJ retail and in the institutional markets. (CX 6C; CX 54A; CX 416A-H; CX 418A-E; CX 421A-E; CX 505C, E; CX 506D-E; CX 593, pp. 388-89; CX 2303F-I; Munkelt 780-1; Barnebey 4013) 32. A.C. Nielsen reports relied on by Tropicana, Minute Maid, Kraft and H.P. Hood are limited to COJ sold to the retail market. (CX 252-253; CX 593, p. 389; CX sold; Munkelt 781, 784; Donovan 1461; Miler 1496-97, 1502, 1554; Hoffer 1825-26; Jessup 2902, 3098-99; Barnebey 3917) 33. Recognizing distinct institutional and retail markets, processors of COJ have established separate sales divisions to serve the customers in these two markets. (CX 506E; CX 1650, pp. 31- , 56; Munkelt 716-17; Goldman 996; Donovan 1293-94, 1297- , 1441; Miller 1481; Barnebey 4013, 4018) Initial Decision 101 F. 34. COJ is usually sold to the institutional market in small size containers (4 oz. and 6 oz.). (CX 3H; Goldman 961; Donovan 1298) Sales of COJ to the retail market are principally made in quart, one-half gallon and gallon sized containers. (Munkelt 711; Lang 1106; Donovan 1295, 1298; Miler 1480-1; Mirapaul 1576; Hoffer 1956) 35. Brand name recognition is important in sellng to grocery stores but unimportant in selling to restaurants. (Hoffer 1822) 36. Typically there are separate buyers for retail and institutional COJ. (CX 523, pp. 298-299; Donovan 1297- , 1323) Many institutional accounts are servced by specialized institutional wholesalers who do not sell to the retail trade. (CX 523, p. 299) (11) 37. Institutional customers usually want a commitment that COJ producers will supply them for a one year period. (Donovan 1443-5; Mirapaul1577 -80; Koch 4880) Such commitments hinder the shifting ofCOJ production from the institutional to the retail market. (Donovan 1443-5; MirapauI1577-80) 38. Retail buyers of COJ are concerned with whether consumers have been presold on the product. Institutional consumers of COJ do not know what brand they are drinking. (Donovan 1323; Mirapaul 1576) 39. There is no cross elasticity of demand between the COJ retail and institutional markets. A consumer does not choose between home consumption ofCOJ and institutional consumption ofCOJ based on the relative prices of COJ. (Munkelt 923; Miler 1495) Many factors other than the price ofCOJ go into a consumer s selection ofa particular airline, hospital, school or restaurant. 40. In pricing COJ for sale to the retail market, processors of COJ do not consider the price ofCOJ in the institutional market. (CX 1650 pp. 38, 40; Goldman 963; Lang 1061-63; Miler 1495; Mirapaul1580) c. The Retail COJ Market is Distinct From the Retail Markets for FCOJ and CSSOJ 41. Beatrice offcials recogoize the separateness of the retail COJ market. (CX 141F; CX 301B) 42. Tropicana s marketing documents recogoize that the COJ retail market is distinct from the FCOJ and CSSOJ retail markets. (CX 490Z-77) In its 1977 annual report, Tropicana separately analyzed its position in the retail COJ and FCOJ markets. (CX 39B) Similar separate analyses of the distinct markets for COJ and FCOJ are found in Tropicana s fiings with the SEC. (CX 40D; CX 41B; CX 42C) 43. H.P. Hood Company discontinued retail FCOJ production because it was so different from retail COJ production. (Donovan 1303) Hood never considered entry into CSSOJ. (Donovan 1315) 44. The Florida Department of Citrus ("FDC") recogoizes the dis- , p.

DD.t\IJ.'lI\JD I"VVL'o. \JV. , D.I .nli.

733 Initial Decision tinct markets for COJ, FCOJ and CSSOJ in its marketing activities to promote orange juice to consumers. (Hoffer 1822-24) FDC has employed separate advertising campaigns for CSSOJ, COJ and FCOJ. (Hoffer 1822-23) 45. Tropicana and Minute Maid recognize the (12) distinctness of the retail COJ and FCOJ markets by purchasing separate sales data with respect to each from the A.C. Nielsen Company. (Munkelt 749; J essu p 2950-52) 46. None of the leading sellers of COJ has ever produced CSSOJ. (CX 1650, p. 81; Donovan 1315; Miller 1505; Barnebey 4018) Tropica- , Kraft and Hood have never considered entry into CSSOJ. (Donovan 1315; Miller 1505; CX 1650, p. 81) 47. Sellers of COJ to grocery stores, including Tropicana and Beatrice, examine the prices of other brands of COJ but do not examine the prices of retail FCOJ, CSSOJ, or other citrus juices or beverages in setting their price for COJ. (CX 593, p. 286; CX 1650, p. 43; Munkelt 750-51, 759; Goldman 963-66, 1029; Lang 1059-63, 1111; Donovan 1314; Miler 1493-94; MirapauI1472-74; Jessup 3140; Parker 4322- 4327 4333; Koch 4937, 4940; CX 732B; CX 747A) 48. FCOJ producers do not consider the prevailing price of COJ or CSSOJ in setting their FCOJ prices. (Bock 1658) 49. Tropicana is the price leader in the COJ market. (CX 301B; CX 593, p. 423; Munkelt 760; Lang 1059; Donovan 1314; Miler 1507) Minute Maid is the price leader in the FCOJ market. (CX 301B) 50. In 1977 private label COJ amounted to about 19% of total COJ sales (CX 13G), and over 50% of FCOJ sales. (CX 353A) 51. Different production and distribution facilities are used for the three types of processed orange juice sold through retail outlets. (CX 527Z-29; CX 593, pp. 28-30, 35, 68-9; CX 1219C; CX 1650 , p. 82; Goldman 988; Miler 1505; Barnebey 4018) Tropicana s FCOJ is processed in a separate building at its Bradenton facilities. (CX 593 132) The production of retail pack FCOJ requires freezing equipment not used to produce COJ for the retail market. (CX 527Z-22; CX 593 pp. 29- , 68-9; Bock 1655) The fillng and packaging equipment used to produce COJ is different from that used in the production of FCOJ. (Bock 1654-- , 1660-1) A firm producing only COJ could not produce FCOJ on its equipment, and to enter the production of FCOJ would require an investment of several millon dollars, part of which cost would be represented by a can plant. (Donovan 1304; Miller 1505) 52. The storage and transportation facilities used for COJ sold to retail outlets differ from those used for FCOJ. COJ is kept chiled, but not frozen, whereas FCOJ must be kept frozen at all times until it reaches the consumer. Processors transport and store COJ while keeping it at a temperature of 35--5"F and they recommend to their , p.

746 FEDERAL TRADE; COMMSSION DECISIONS Initial Decision 101 F. customers that COJ be kept under refrigeration and stored at 35- 55T (Munkelt 714-15, (13) 799; Goldman 967, 1004-5, 1041; Lang 1052; Miller 1489-90; Mirapaul 1570; Jessup 3051; Barnebey 4078; Koch 4910; Piskac 5170) The only exception to COJ being kept under constant refrigeration by grocery stores occurs when glass containers occasionally are put out on an aisle display for a week or two during a promotion. After such a promotion, the COJ is returned to refrigerated storage. (Goldman 1004-5; Miller 1550) If COJ in glass containers is left out at room temperature, it will more quickly deteriorate in quality by losing flavor and changing color. (Goldman 1045; Miler 1488; Mirapaul 1570) 53. The production of CSSOJ requires canning facilities whereas the production of COJ does not. (CX 593, p. 68) The storage and transport of CSSOJ differs from COJ for the retail market. (CX 2302 p. 60) COJ requires cold storage facilities and refrigerated transport while CSSOJ does not. (Munkelt 714-16, 729; Goldman 967; Lang 1051-52; Donovan 1291, 1309) 54. FCOJ suppliers use no salesmen, do no advertising or promotion for either their private or packer labels and provide no marketing assistance to its customers. (CPF 230; CX 297 A; CX 593, pp. 436-37 444; CX 1650, pp. 87--8, 99; Bock 1650; Hoffer 1822, 1869, 1873 1978-0) 55. COJ is sold to diflerent grocery store buyers than is FCOJ or CSSOJ. (Munkelt 714-15, 798; Goldman 964-5; Lang 1052; Donovan 1300, 1303, 1315; Bock 1650; Jessup 2770, 2927, 2956) 56. COJ is stored in a different part of the grocery store than are FCOJ and CSSOJ. COJ is typically found in the dairy case although it is also found to a lesser extent in the produce section of the grocery store. (Munkelt 714-15, 798-99; Goldman 967; Lang 1052; Donovan 1459; Miller 1489; Jessup 2926-27; Koch 4910; Piskac 5170) FCOJ is found in the frozen food case of the grocery store and CSSOJ is found in the dry grocery section. (Munkelt 714-15; Parker 4325) 57. Total COJ industry sales have been increasing at about 20% in gallons. Total FCOJ sales have been increasing at about 10% (CX 899B-C) but in 1979 total FCOJ gallon sales decreased 6% while total COJ sales increased 20%. (CX 389K, N) Ofthe total domestic orange juice market, FCOJ has dropped from 77.2% in 1975 to 67.9% in 1979 and COJ climbed from 18.3% to 28% during that period. (CX 13A 149) The recent rapid increase in sales ofCOJ comes from new orange juice consumers and not from former FCOJ consumers. (Bock 1686; Hoffer 1926; Jessup 3056; Munkelt 814, 924) 58. When FCOJ is put on promotion, there is no perceptible effect on the sale of COJ in the retail market. (Goldman 965; Mirapaul 1572-73; Jessup 3056-57) (14) 733 Initial Decision 59. A study done by a research economist, employed by the Florida Department of Citrus, prepared in the course of his employment and not for use in litigation, concluded that the demand for COJ was inelastic and that there were no significant substitutes for it, including FCOJ and CSSOJ. (Tiley 2232, 2236-37, 2367, 2397) 60. Consumers perceive COJ, when compared to FCOJ, as a pure 100% fresh orange juice, prepared with little processing. (Munkelt 789-90; Hoffer 1956; Barnebey 4063) The extensive growth of COJ in recent years has been due to more women working and buying COJ as a convenience product. (Hoffer 1915-17) . 61. FCOJ has an indefinite shelflife if kept frozen, is more compact than COJ and can be more easily stored. (Donovan 1303; Bock 1652; Hoffer 1931; Tiley 2024-25, 2243-44; Barnebey 4063, 4084-6) Shelf life for COJ in paper cartons is about 20 to 30 days. (Munkelt 715; Donovan 1309; Barnebey 3911; Parker 4315; Piskac 5165; CX 62- Shelflife ofCOJ in glass is over one year. (Miler 1508; Barnebey 3911) 62. Consumers and industry members perceive CSSOJ as a product of inferior taste and quality. (Munkelt 790; Goldman 966; Lang 1060- 61; Donovan 1315; Miller 1494, 1505; Mirapaul 1573; Hoffer 1956; Barnebey 4018-19, 4021; Parker 4340, 4406; Bock 1652-54) d. The Retail COJ Market Includes Both COJ From Fresh Fruit and Reconstituted From Bulk Concentrate 63. COJ sold to grocery stores is processed directly from fresh oranges, bulk concentrate, bulk frozen single strength juice, or from a combination of fresh and frozen juice. (Jessup 879; Goldman 960; Mirapaul 1567; Barnebey 3983-85, 4009) 64. Retail COJ is packaged in glass, cartons or plastic, primarily of quart, half-gallon or gallon sizes. (Munkelt 710-711, 736-37; Goldman 960; Lang 1048, 1051; Donovan 1284-5; Miler 1480-81, 1496-97; Mirapaul 1567; Jessup 2876; Barnebey 3984-5) 65. During 1979, 70% of Tropicana s COJ was produced directly from fruit. (CX 2302, p. 107; Jessup 3015) Tropicana s source for its COJ varies during the year. In May it may be fresh squeezed oranges and in August it may be made from bulk concentrate. (CX 38M; CX 39C; CX 40D-E; CX 527Z-1) Throughout Tropicana s history, some of its COJ has always been (15) made from bulk concentrate. (CX 2302 p. 118) During 1978, Tropicana had a made-from-concentrate COJ on the market for six months. (Barnebey 4010; Munkelt 807) 66. H.P. Hood and Kraft process COJ for the retail market direc:tly from fresh oranges and from bulk concentrate. (CX 678B; CX 680A D; CX 759A; Donovan 1292, 1412; Miller 1483; Jessup 3044) H.P. Hood has processed COJ from single strength frozen blocks. (Donovan 1292) 67. Much ofthe same equipment used to produce COJ directly from , p.

Initial Decision 101 F. oranges is used to produce COJ reconstituted from bulk concentrate. (CX 382C, T- , Z-28 to Z-32; CX 420; CX 593, pp. 138-39) 68. Producers of COJ made solely from bulk concentrate perceive themselves to be processors that are in competition with Tropicana COJ from whatever source it was derived. (Munkelt 764, 805-6; Goldman 961, 963-64; Lang 1048) 69. Tropicana recognizes firms which process COJ only from bulk concentrate as competitors in the COJ market. (CX 752B; CX 759A; CX 951; CX 952; CX 1004B) 70. Consumers perceive little difference in COJ made from fresh oranges or from bulk concentrate. (Brick 5329; CX 2302, p. 105) In 1976 Hood switched from sellng COJ made directly from oranges to reconstituting from bulk frozen concentrate. Hood changed the labeling on their carton, as required by law, to show that the COJ was reconstituted. Hood's sales increased rapidly thereafter. (Donovan 1304-6, 1308) e. The Retail COJ Market Includes Glass, Paper Cartons and Plastic Containers 71. Members ofthe industry define the retail COJ market to include COJ sold in glass, paper cartons, and plastic containers. (CX 1650 10; Munkelt 710-11; 736-37; Goldman 960, 1041--2; Lang 1048; Donovan 1284-85; Miler 1480; Mirapaul1567; Bock 1654; HofTer 1815-16 1829-30; Tiley 2112; Jessup 2876; Barnebey 3983--4) 72. FDC, a state government agency financed by the citrus growers and an information center for the industry, includes COJ in glass, cartons and plastic containers under its definition of COJ as used in its offcial rules. (HofTer 1815-16) 73. FCPA, the industry trade association, defines COJ in its reports and internal membership division to include COJ packaged in glass, cartons and plastic containers. (CX 594J, (16) pp. 10-11; Barnebey 3991) 74. Tropicana, Johanna Farms, Home Juice Company, Pure Foods and Bodines sell COJ in glass, paper and plastic containers (CX 1650 pp. 94-95; Goldman 960-1; Lang 1051; Mirapaul 1567-68; Jessup 2781--2) Hood sells COJ in paper and plastic containers. (Donovan 1285) Kraft sells COJ under its Sealtest label in glass and paper containers and under its Kraft label in glass. (Miler 1481, 1483--4) Minute Maid sells COJ only in paper cartons but sold in glass in a test market in upstate New York from 1971-1976. (Munkelt 711, 763) Prior to the merger, Beatrice sold COJ in glass, paper and plastic. (CX 47) 75. Tropicana marketing reports recognize the close relationship of COJ packaged in cartons, glass and plastic. (CX 661D; CX 662D; CX 733 Initial Decision 665B, F; ex 666B; CX 671A; CX 739A; CX 741A; ex 951A; CX 987E; CX 988E; CX 1086C; CX lll1C; CX 1143B; CX 1145B; CX 1176; CX 1280D; CX 1342B) 76. In their pricing ofor:e container type ofCOJ, sellers take into consideration the prices of other types of containers of COJ. (M unkelt 759--0; Goldman 963, 970-71; Mirapaul 1574) In pricing its COJ Minute Maid which sells only cartons, takes into consideration the price of the Kraft brand ofCOJ which is sold only in glass. (Munkelt 711 759--0; Miller 1480-1) Beatrice s Mattoon, Ilinois, plant which processed COJ only in plastic containers examined the prices of competing COJ, including Tropicana, packaged in glass or paper cartons. (Parker 4322- , 4327) Beatrice s Dayton, Ohio plant which processed plastic gallons of COJ and sold glass quarts and half-gallons, examined the retail prices of Borden s paper half-gallons. (Koch 4936-0) f. The Retail COJ Market Includes COJ Sold Through Various Distribution Systems 77. Processors of COJ for the retail market sell to chain stores independent wholesale warehouses, individually owned grocery stores, dairies, and distributors who are independent routemen who own their own truck but no storage facilities. (CX 593, pp. 18, 86-87; CX 947G; CX 1070C; CX 1232; CX 1650, pp. 125-26; Lang 1067, 1094- 96; Donovan 1309; Jessup 2763-5) Their COJ arrives at chain stores by sale and delivery to: (a) the warehouse of the chain; (b) individual stores of the chain; or (c) dairies or distributors who deliver to the individual stores of the chains. (CX 593, p. 45; CX 661A; CX 741A; CX 1070C; CX 1188A; CX 1650, p. 102; Lang 1063-4; Donovan 1309 1354, 1359, 1366-8; Parker 4330) They also sell COJ to chain stores through dock pick up by the chain at the processor s facility or through dock pick up by distributors. (CX 593, p. 151-52; CX 1217B; Munkelt 836-37; Lang 1094-95; Donovan 1378; Mirapaul 1590-91; Davis 3170; Parker 4330-31, (17) 4383; Koch 4931) Sometimes a chain store will purchase COJ both for delivery at its warehouse and delivery to its stores. (CX 915F; CX lll1B; CX 1113A; Donovan 1354) Small chains and non-chain grocery stores receive-delivery ofCOJ from: (a) warehouses of wholesalers; (b) dairies; (c) independent routemen; or (d) processors. (CX 593, p. 346; CX 1232; Lang 1067--8, 1109; Jessup 2765, 2894, 2954) 78. In the past few years, processors have increasingly delivered COJ to the warehouses of chain retail stores. (CX 1005B; Lang 1064; Donovan 1311-12) 79. Home Juice originally delivered COJ directly to individual grocery stores but now relies on deliveries by its customers, including Initial Decision 101 F. chain store warehouses, independent distributors and dairies to reach grocery stores. (Lang 106 4, 1067, 1094-96, 1109) 80. Johanna Farms delivers COJ to chain store warehouses and makes store-door delivery to other grocery stores. (Goldman 972- 994) It also sells to independent distributors who deliver its products. (Goldman 974) In the past 10 years its sales ofCOJ to warehouses has increased. (Goldman 1019) 81. H.P. Hood also increasingly relies on sales to grocery stores through warehouse delivery and store-door delivery. (CX 505E; CX 752B; CX 760B; CX 1005B; ex 1006A; CX 1178B; Donovan 1309 1311-12) Hood began warehouse delivery in 1975 in trailerload quantities directly to chain store warehouses instead of to their individual stores. (CX 505E; CX 760B; CX 1178B) 82. Dairies such as Dean Foods, Knudsen, Sealtest and Western General Dairies process and sell COJ to grocery stores both by delivery to chain warehouses and by store-door delivery to individual grocery stores. (Goldman 972-73; Lang 1063-64, 1108; Donovan 1309 1354; Jessup 2866, 2919, 3045-7; Barnebey 4075-76) 83. In its early history Tropicana delivered its COJ in its own trucks in less than truckload lots. (CX 593, p. 10) Tropicana now distributes COJ by delivery to warehouses of chain stores, wholesale warehouses and dairies. (CX 593, pp. 44-5, 347; CX 919; CX 1650, pp. 125- , 130; Jessup 2764-6; Barnebey 4078-79) In New York City, Boston and Miami it sells to independent routemen who deliver store-door to retail grocery stores. (Barnebey 4044, 4047-48) These routemen own their trucks, take title to the COJ, and pick it up at the Tropicana terminal. (CX 593, p. 150; Barnebey 4047) 84. Beatrice milk plants deliver COJ for the retail market to individual units of chain grocery stores, to (18) independent groceries and to wholesale distributors. These milk plants also distribute COJ through independent routemen who pick it up at the dock ofthe milk plant or have it delivered to their locations. (Parker 4366, 4383; Granger 4433) 85. From 1973 to 1975 the Beatrice Dayton milk plant delivered milk to the warehouse ofa chain of75-0 convenience stores and tried to sell the chain COJ at a price lowered to reflect warehouse delivery. (Koch 4952-57) In 1977 the Beatrice milk plant in Greeley, Colorado delivered COJ to the distribution facility of a chain of 46 grocery stores. (CX 528V, no. 86) 86. Beatrice and Tropicana both provide in-store help to grocery stores sellng their brands of COJ. These in-store functions include checking product coding, shelf placement and rotation, and setting up displays and promotional programs. (CX 593, pp. 96-97; CX 665D- CX 666A; CX 672C-D; CX 768B; CX 1191B; CX 1258B, D-E; Jessup 733 Initial Decision 2756-57 2921 2925-26; Barnebey 4066; Koch 4913-14; Piskac 5181- 82) Tropicana uses brokers and salesmen and its territory managers to perform these functions. (CX 665D-E; CX 666A; CX 672C-D; CX 768B; CX 1191B; Jessup 2756-57; Barnebey 4066-7, 4080-1) Beatrice uses salesmen to perform these same in-store services. (Jessup 2927; Davis 3152, 3186; Parker 4390-91; Koch 4913-14; Piskac 5181- 82) 87. Representatives of Tropicana, Minute Maid and Home Juice while delivering COJ to warehouses, consider those who deliver on a store-door basis to be competitors. (CX 527Z-lO; CX 661A; CX 667H; CX 944; Munkelt 764-5, 784-85, 850; Lang 1058-59, 1063-66) 88. Dairies delivering COJ store-door compete with producers of COJ who deliver to warehouses, sometimes displacing them in the sale of COJ to grocery stores. For example, in Tampa and Orlando, Florida, Borden COJ which was delivered store-door had the effect of decreasing Tropicana s sales of COJ at Winn-Dixie grocery stores, in part because of the advantages of store-door delivery. (CX 1321B) Similarly, in New England, Tropicana in 1975 found it diffcult to compete for COJ sales with H.P. Hood's store-door delivered COJ because stores buying from Hood "get what they order, when they want it and there are very seldom any out of stocks on either size, in these stores." (CX 947-0) (Emphasis in original. 89. Tropicana s north central division manager, Mr. Jessup, considers dairies that reconstitute and sell COJ on a store-door basis to be s marketing competitors of Tropicana. (Jessup 2905-15) Mr. Jessup reports to Tropicana s management discuss this competition. (CX 681B; CX 682B; CX 685B; CX 686B; CX 687B; CX 688B; CX 690B; CX 691B; CX 692B; CX 695B-C; CX 696B; CX 697B; CX 698B; CX 699C; CX 701B; CX 702B; CX 703C; CX 704C; CX 705C; CX 706B; CX 707B; CX 708B; CX 709B; CX 710B; CX 711B; (19) CX 712A; CX 713B; CX 714B; CX 715B; CX 716B; CX 717B; CX 718B; CX 1110B; CX lll1B; CX 1112B; CX 1113B; CX 1114B; CX 1115B; CX 1116B; CX 1117B) Tropicana has lost some customers to such dairies engaged in storedoor delivery of COJ. (Jessup 2906) 90. Mr. Woerner, Tropicana s western division manager, considers dairies who delivered COJ on a store-door basis to be competitors of Tropicana. (CX 661A, D; CX 663; CX 670A; CX 672D; ex 666B) His weekly sales reports show sales lost by Tropicana and Minute Maid to dairies which were engaged in store-door delivery. (CX661A, Co CX 666B; CX 672D) These reports show that Tropicana lost sales to the 7-11 stores in Salt Lake City because ofCOJ packed by a Beatrice dairy and sold to that chain. (CX 661A; CX 663C) Another dairy in that market, Western General Dairies, was selling Dr. Lukes COJ at a low price in 1976, and Mr. Woerner told the Tropicana representa- 752 FEDERAL TRADE; COMMISSIQN DECISIONS Initial Decision 101 F. tive to keep a close watch and: "If we seem to be getting hurt, wil advise so we can take appropriate corrective action." (CX 6610) 91. Tropicana s internal marketing summaries submitted to its top offcer by its sales manager report "competitive activity" including prices ofCOJ that is store-door delivered by dairies. (CX 667H (Boarden); CX 669A-B (Borden and FarmbestJ; CX 678 (FarmbestJ; CX 679A and C 969C (Hood)) 92. A 1976 Tropicana study of COJ distribution and pricing in San Francisco listed Meadow Gold COJ and the COJ off our other dairies all sold on a store-door basis as part of the "competitive distribution in the stores contacted." (CX 807F; Dutt 3829-30) Minute Maid, which uses no direct store-door delivery, also considered Meadow Gold brand COJ to be a competitor, as well as the brands of other dairies which use store-door delivery for their COJ. (Munkelt 765, 785, 850) When setting the price for Meadow Gold COJ, Beatrice s dairy manager in Mattoon went to an IGA store and looked at the retail prices of Tropicana, Minute Maid and Kraft glass COJ because they were in the dairy case with his plastic COJ; those products are delivered to the warehouse while Meadow Gold COJ is delivered store-door. (Parker 4322-27) 93. There is a limited amount of shelf space available for COJ in grocery stores. (CX 593, p. 322; CX 6610; CX 670B; CX 737B; CX 779B; CX 947M, Q; CX 951A; CX 996A; Munkelt 767, 796; Lang 1058-59 1065-6; Jessup 2905--6" 2921-23; Davis 3150, 3169-70; Parker 4322 25) Producers of COJ want shelf space in grocery stores ("facings because loss of shelf space decreases their sales. (CX 947M; CX 996A; Jessup 2921-25) Tropicana has lost shelf space in grocery stores to processors who deliver COJ on a store-door basis. (Jessup 2922-23) 94. Minute Maid and Tropicana do not subscribe to the (20J market reporting service sold by Sellng Areas Marketing Inc. ("SAMI") for COJ because SAMI reports measure shipments from grocery warehouses to retail stores and fail to include COJ that is store-door delivered. (CX 1650, pp. 143-44; Munkelt 785; CX 593, pp. 229, 457-58) Minute Maid and Tropicana do purchase A.c. Nielsen market reports which include COJ reaching grocery stores through all methods of delivery including store-door. (CX 517F, U; CX 593, pp. 228-29; Munkelt 749; Munkelt 851) g. The Retail COJ Market Does Not Include Home Delivery 95. Home delivery ofCOJ is made by dairies or independent routemen who deliver dairy items and often other products along with COJ. (Complaint and Tropicana Answer n 6; Goldman 974-75; Mirapaul 1581; Karnes 4765-66) COJ purchased through home delivery is more expensive than COJ sold in retail outlets. (Goldman 974-75; Donovan , p.

""Lo'''''~v~ 733 Initial Decision 1320, 1322, 1335-36; Mirapaul 1581; Parker 4387; Piskac 5110;. ex 593, p. 90) 96. Processors ofCOJ for the retail market do not consider the home delivered price ofCOJ in setting their price. (CX 1650, p. 37; Munkelt 760; Lang 1063; Donovan 1322; Miler 1494; Mirapaul 1581) 97. Hotne delivery of COJ has been on the decline because of rising costs and today has almost disappeared from the market place. (CX 527J; CX 593, p. 90; Goldman 974; Lang 1063; Donovan 1320-22, 1334; Mirapaul1581; Davis 3180-2; Parker 4387-88; Koch 4935; Piskac 5170; Brick 5316) h. The Retail COJ Market Does Not Include Drinks and Other Juices 98. Fruit drinks and other fruit juices do not compete with COJ. (CX 301H; CX 520, pp. 56-57, 6tH6, 95; CX 1196; Munkelt 758; Goldman 966; Lang 1063; Miler 1495; Mirapaul 1573; Bock 1658-59; Hoffer 1920; CX 593, pp. 392-93) 99. Producers of COJ do not take the prices offruit drinks or other fruit juices into consideration in their pricing or marketing of COJ. (CX 667C, H; CX 1650, p. 38; Goldman 966; Lang 1063; Mirapaul1573) One reason that grapefruit juice prices are not considered in pricing COJ is the different raw material cost for the two juices. (CX 1650 38) Orange drink producers do not consider the price ofCOJ in pricing their drinks. (CX 520, pp. 56-57; CX 661E; CX 667H) Tropicana compares the price of Tropicana grapefruit juice only to that of other brands of grapefruit juice and the price of Tropicana (21) drink only to those of other drinks. (CX 675B; CX 688A; CX 689C; CX 730C; CX 748B; CX 750H; CX 7902F; CX 915A; CX 951B; CX 1126B; CX 1162A; CX 1177B) E. Geographic Market 100. There virtually are no imports ofCOJ into the United States. (CX 527E, no. 19) 101. The three largest sellers of COJ to retail grocery stores Tropicana, Minute Maid and Kraft, together sell COJ throughout all or almost all areas of the United States. (CX 46U, V- , 2-, Z-7; CX 345T-V; CX 354A; CX 388, tables 4 & 5; CX 526A-S; CX 527Z-7, no. 182; CX 560, 2-8, Z-99, Z-109, Z-114, Z-119, Z-121, Z-123, Z-125 Z-130, Z-135, Z-140, Z-142, Z-144, Z-149, Z-151, Z-156, Z-161; Munkelt 718, 760, 763-4; Miler 1482, 1492; Barnebey 4021, 4033) 102. The brands of the three largest sellers of COJ sold in grocery stores are recognized by industry offcials as "national brands" with differences in market penetration in parts of the country. (CX 354A; , p.

Initial Decision 101 F. CX 1650, p. 69-70; Munkelt 760; Lang 1058, 1141; Miller 1490-91; Mirapaul 1575; Barnebey 4069) 103. Industry offcials examine and report on the national sale of COJ through grocery stores. (CX 13A-J; CX 38F; CX 39B; CX 386 101; CX 394, pp. 38-0; CX 396M; CX 400A-U; CX 401A-K; CX 402A- K; CX 403A-K; CX 404A-K; ex 405A-K; CX 406A-K; CX 407A- Munkelt 749, 785-6) Minute Maid uses national Nielsen data on the retail COJ market in presentations to its customers and prospective customers. (Munkelt 749) Tropicana marketing documents characterize the movement ofCOJ through grocery stores in the United States as "Ready to Serve Orange Juice Sales Glass, Cartons and Plastic - Total U.s." (CX 13B; CX 394, p. 25; CX 560 p. 11) 104. Tropicana tries to sell its COJ throughout the United States. (Barnebey 4040) It has an internal sales organization and broker system which covers all of the United States. (CX 527Z-9, no. 188; CX 527Z-35, no. 302; Barnebey 4022- , 4038; CX 1654) In 1977 Tropicana had 57 brokers working in 36 states and Washington, D.C. (CX 375) 105. Tropicana has a single f.o.b. Bradenton price-plus freightfor its COJ sold in truckloads to all types of customers throughout the United States. (Barnebey 4068) Tropicana also sells COJ in less than truckload amounts on a delivered price basis in several pricing zones. (Barnebey 4067, 4069-4070; Jessup 2791) 106. Most promotional and advertising programs at (22) Tropicana are centrally planned by its marketing stafr at Bradenton. (CX 1650 p. 70-71; Jessup 2784, 3076-77) All pricing decisions at Tropicana are made in Bradenton, Florida. (CX 593, p. 26; Jessup 3054) 107. Tropicana and Minute Maid sometimes advertise their COJ nationally, using media which serve the entire country. (CX 38E; CX , p. 6; CX 916E; CX 936A-C; CX 1250C- , J-K; CX 1650, pp. 71-73; Munkelt 791; Barnebey 4069) 108. Some chain grocery stores use national buyers for COJ. (CX 1339C; CX 1650, pp. 57-58) Such buyers are called "national accounts" by Tropicana. (CX 1650, pp. 57-58) Among firms considered to be national accounts by Tropicana are Safeway, Kroger, A&P Winn-Dixie and Topco Associates. (Barnebey 4041-42) 109. Tropicana has a national accounts sales manager who calls on chain store accounts. (Barnebey 4041-42) 110. Tropicana sells COJ from its plant in Bradenton, Florida which ships to every state except Alaska and Hawaii. (CX 378A-E) Tropicana has been shipping COJ to California, Washington and Oregon since the early 1960's and has served the mountain states from Florida for more than five years. (CX 1650, pp. 61-62; Barnebey 4026) 111. Minute Maid sells COJ to grocery stores throughout the coun- BEATRICE l''UUU CU.. ET AI,. 100 733 Initial Decision try, all of which is reconstituted at seven facilities including four milk plants, from bulk FCOJ shipped from Florida. (Munkelt 718, 721, 724) 112. Kraft processes its COJ for sale throughout the country to grocery stores under the Kraft brand at a plant in Florida and through a west coast co-packing arrangement. (CX 1650, p. 74; Miler 1482) The Florida plant primarily serves all areas of the United States except the west coast and Hawaii, but has shipped COJ to California; the west coast and Hawaii generally are served by the co-packing arrangement. (CX 1355A; CX 1650, p. 74; Miler 1483) Approximately 90% of Kraft brand COJ is packed in Florida and shipped to the warehouses of corporate foodchains. (Miler 1508-9) 113. Hood Dairy Company primarily sells COJ in New England and to grocery stores in Buffalo, Chicago, and parts of the southeastern United States. It sells in ten states. (CX 345T-U; CX 388 tables 4 & 5; CX 389 tables 4 & 5; CX 505C; CX 506C; CX 1650, p. 145; Munkelt 764-5; Donovan 1296) 114. Almost all COJ sold throughout the United States is made from orange juice or frozen concentrate from Florida. (CX 22F; CX 40D; CX 41B; CX 42C; CX 527D, no. 18; CX 1650, p. 12; Munkelt 723, 835; Goldman 961 , 1001, 1032; Lang 1101; Miler (23) 1539; Mirapaul1569; Barnebey 3950, 3990, 3999, 4001) At times of short supply, however processors have relied on imports. (Barnebey 4001) California oranges are used for blending purposes or when the Florida supply is disrupted. (Munkelt 723; Goldman 1043-4; Barnebey 3950-51) COJ made entirely from concentrate of California or Arizona oranges has a low sugar content and is sour. (Munkelt 723; Goldman 1044; Bock 1665; Barnebey 3930, 3950) Texas oranges are too sweet for COJ. (Goldman 1044) California and Texas oranges have a low juice content, thereby making their use for COJ processing very expensive. (Barnebey 3950- 51) 115. In 1977 Beatrice processed COJ was sold to grocery stores in 23 states (CX 46E-Z-):

Alabama Kentucky California Louisiana Colorado Montana Florida Nebraska Georgia Nevada Hawaii New Mexico Idaho Ohio Illnois Oklahoma Indiana Oregon Iowa South Dakota Kansas Utah Wyoming Initial Decision 101 F. 116. Beatrice has dairy distribution facilities in Detroit; Baltimore; Alexandria, Virginia; New York; Los Angeles; Nashvile; Fort Worth; and Pittsburgh which could be used for the distribution of COJ. (CX 526V- 117. In 1977 Beatrice and Tropicana directly competed in the sale of COJ to the retail market on all or part of twenty-one states. (Compare CX 46 with CX 526I-M and Zl-25) F. Market Structure a. Market Shares 118. In 1975 gallon sales of COJ at retail through grocery stores totalled 132.5 milion gallons (CX 560Z-17) The retail dollar sales of COJ in that year totalled $250 000 000. (CX 560Z-24) 119. In 1976 the gallon sales ofCOJ at retail through grocery stores totalled 160.9 million gallons. (CX 423L; CX 560Z-17) The total retail dollar sales of COJ in that year totalled $310 200 000. (CX 423L; CX 560Z-24) (24) 120. In 1977 gallon sales of COJ at retail through grocery stores totalled 188.9 milion gallons. (CX 426F; CX 560Z-17) The retail dollar sales of COJ in that year totalled $411 500 000. (CX 560Z-24) 121. In 1978 the gallon sales ofCOJ at retail through grocery stores totalled 204.1 milion gallons. (CX 560Z-17) The retail dollar sales of COJ in that year totalled $545 200 000. (CX 560Z-24) 122. In 1975 the leading brands of COJ sold in grocery stores had the following market shares (CX 560Z-18):

Market Share Market Share Brand (in gallons) (in dollars) Tropicana 28. 30. Kraft 13. 15. Minute Maid Hood 55. 57.

123. In 1978, the four leading brands of COJ sold in grocery stores had the following market share (CX 560Z-18; Z-25): Brand Market Share Market Share (in dollars) (in gallons) Tropicana Kraft 29.9 30.11.5 Minute Maid 15. 15.4 10.4 Hood 4.4 4. 60.2 61. 124. In 1978, private label gallon sales of COJ in retail groceries totalled 18.7% ofthe market in gallons and 17.4% in dollars. In 1975 this share of the market was about the same. (CX 560Z-18) .. , .. .. .. .. , , 733 Initial Decision 125. Sales ofCOJ to grocery stores has increased from 132.5 millon gallons in 1975 to 204. 1 millon gallons in 1978. (CX 560Z-17) Tropicas share of the market has been about the same during that time. (CX 560Z-18) Industry FCOJ sales volume has remained stable. (Hoff er 1928-1930) 126. In the 1977-78 growing season 256.3 milion gallons of COJ were processed in the United States from Florida oranges. (CX 594J pp. 24-25; CX 594K, pp. 14A, 1C) Of that amount, 134.4 milion gallons of COJ were produced directly from fresh oranges while 121.9 millon gallons of COJ were reconstituted. (CX 594J, pp. 24-25; CX 594K, pp. 14A, 1C) (25) 127. Almost all of the COJ processed in the United States comes from Florida oranges. (CX 1650, p. 12) In the 1977-78 season, Tropicana processed about 40% (109.6 milion gallons). (CX 594J, p. 15-16 19; CX 5941) Tropicana, Minute Maid, Kraft and Hood together processed in Florida more than 171.4 milion gallons ofCOJ, more than 60% of the total processed in the United States. (CX 594J, p. 15-16, 19, 22-23; CX 5941) 128. In 1976 and 1977, Tropicana shipped 55% ofthe COJ processed in Florida. (CX 527E; CX 527R; CX 1650, p. 24; Barnebey 3993, 3999) 129. Minute Maid division of the Coca-Cola Company started processing COJ and sellng it to grocery stores in 1971, and in 1979 had nationwide distribution. (CX 298; CX 593, p. 253; Munkelt 718, 761) 130. Some smail processors of COJ have gone bankrupt in recent years, including Orange Blossom, Glacier Grove and Polar-Vim. (CX 527Z-1l; ex 527Z-4; CX 680H; CX 1094C; Goldman 1032, 1035-36; Davis 3164) b. Barriers to Entry 131. Ex-Cell-O Corporation makes carton-making machinery and has about 50% of the market. These machines make gable-topped paper milk or juice cartons. (Brick 5287-93) A small 30 quart per minute Ex-Cell-O machine can be leased for $150 per month. (Brick 5325) A half-pint packaging machine can be purchased for about $120 000. (Polidoro 5375) 132. A glass packaging line for COJ costs $250,000 to $600 000. (CX 1650, p. 96) The cost of a carton packaging line (case packers, box makers and pasturization equipment used with the Ex-Cell-O machine) is about $250 000. (Barnebey 4046-7) Separate packaging lines and Ex-Cell-O machines are required for quarts, half-gallons and gallons, and each size carton. (Brick 5325-26) Refrigerated warehouse space must be built or leased. (CX 527Z-56) 133. Blow molding is a process whereby polyethylene or polystyrene is melted and formed into a container by blowing air into the thin wall Initial Decision 101 F.TC. ofthe plastic. (Donovan 1295) It may be used in packaging of milk and COJ (CX 1081B; CX 1084B; RX 17H; RX 18D; Donovan 1285, 1295-96) Equipment to produce blow molded plastic gallon containers costs about $500 000 to $750 000. (Munkelt 826; Donovan 1329; Dutt 3833) 134. Several hundred dairies in the United States process COJ. (Brick 5294; finding 261) (26) 135. Entry into the processing ofCOJ by dairies is usually quick and inexpensive. The plant manager simply orders frozen concentrate. The dairy already has all the necessary equipment and method of distribution. (Parker 4312-13; Koch 4873-74; Piskac 5167; Goldman 1022-23) 136. To obtain shelf space in a grocery store, national brands ofCOJ spend milions of dollars on advertising and promotion. (CX 572Z-; Koch 4955-56; CX 345N) In 1974, Tropicana spent $4.2 millon on advertising and promotion. (CX 527Z-) 137. Grocery stores wil stock nationally advertised COJ, even at a higher price than unadvertised brands, because of consumer acceptance of the advertised brand. (Koch 4955-56; CX 661F) 138. Regional and local processors sometimes advertise their brand of COJ in newspapers and on local radio and television, as well as bilboards, and in-store promotions. (CX 660B; CX 662C-D; CX 778- 784) 139. Regional and local processors and private labels use lower prices and promotional allowances to sell their COJ. (Goldman 964 1010-1111; Mirapaul 1576; Lang 1133-34; CX 883A) 140. A typical grocery store handles about five brands of COJ including two national brands and sometimes a private label. (CX 764B; CX 769A; CX 1147B; CX 1650, p. 124; Lang 1058; Mirapaul 1575; Jessup 2981-82) 141. There are forty plants in Florida supplying bulk frozen orange juice concentrate. (Barnebey 3899-3900) Oranges, the raw material for COJ, are in limited supply, however, and new entrants have sometimes had diffculty obtaining an assured supply of quality frozen concentrate. (MirapauI1582; Bock 1657, 1682; Dutt 3779-80) 142. It takes five to seven years from the time an orange tree is planted until oranges can be harvested. (Tiley 2365) 143. In recent years the number of producing orange trees has declined slightly, but with newly planted, disease resistant trees and better production methods future supply should continue to increase. (CX 22G- 144. A freeze in January 1977 caused a reduction in the number of Florida oranges available for orange juice production. (CX 22G) (27) .

733 Initial Decision G. Beatrice and Tropicana Before the Merger a. Tropicana 145. Tropicana principally sells COJ to grocery chain store warehouses and wholesale grocery warehouses. (Barnebey 3886) Tropicana also sells COJ to dairies (CX 41B) 146. About 7% or 8% of Tropic ana s COJ business is private label packed for Safeway, Kroger, Pet, Food Fair, Topco and A&P. (Jessup 2761; Barnebey 3972, 4041) 147. About 82% of Tropic ana s COJ business is in the states bordering the Atlantic Ocean and Gulf of Mexico, plus Cleveland, Detroit Chicago, Los Angeles and San Francisco. (Barnebey 3891-92) 148. In 1977 Tropicana s brands ofCOJ were offered to consumers through grocery stores in 242 ofthe 282 standard metropolitan statistical areas (SMSAs) determined by the Bureau of the Census. (CX 47; CX 5261) 149. In 1977, national COJ sales through food stores, in milions of gallons, were as follows (CX per Cent 13H):Gallons Tropicana 8.93 glass 2.82 9. carton 6.11 19. 11. Kraft 3.43 Minute Maid 4.19 13. HoodPrivate label1.21 3. 5.87 19. All other glass 8.4 All other cartons 2.07 6. All other plastic 2.46 2.58 Total market 30.74 99. 150. In the mid 1950's, Tropicana s principal customers were dairies. By the mid 1960's the principal customers were warehouses of chain grocery stores and grocery (28) wholesalers. (Barnebey 3887) 151. In the mid 1960's only Tropicana and Kraft were sellng COJ in glass. (Barnebey 3888) 152. Now, in addition to Tropicana and Kraft, COJ is packed in glass by other Florida processors: Citrus World, Adams Packing, Southern Fruit, Tree Sweet, Sealsweet and Ben Hil Griffn. (Barnebey 3889) 153. Tropicana is the price leader in the sale of COJ. Other COJ processers examine the prices of the Tropicana COJ in determining the price at which they wil sell their COJ. (Argeros CX 593, p. 423; Goldman 964 1029; Lang 1059; Donovan 1314) Tropicana makes its 8 About 700 000 gallons of the private label tota! were processd by Tropicana. (Bamebey 3972) Ths is more than 2% of the market.

Initial Decision 101 F. own pricing decisions on COJ based on its costs rather than on the prices charged by its competitors. (Barnebey CX 1650, pp. 35-37, 41) 154. In 1977 Tropicana sold COJ in paper cartons, and glass bottles of various sizes, and some plastic individual serving containers. (CX , pp. 4-7; CX 334B- 155. Tropicana processes COJ at its plant in Bradenton, Florida and ships it by train to its warehouses in Kearney, New Jerseyand Hammond, Indiana. (CX 3821) Tropicana also leases warehouses in Los Angeles, Seattle, Portland and Salt Lake City which it supplies by truck. (CX 527Z-25) 156. In 1977, Tropicana COJ was sold through grocery stores in all of the states of the United States except Hawaii, and Alaska. (CX 378A- 157. Tropicana owns 200 refrigerated rail cars. Every five days a unit train of at least 60 Tropicana rail cars, over one mile long, leaves the Bradenton Tropicana plant nonstop to the Kearney warehouse. Tropicana owns about 800 vehicles for transporting oranges and orange juice. (CX 39, p. 14) b. Beatrice 158. In 1977 Beatrice s dairy division processed, distributed and sold three labels of reconstituted COJ. (CX 57A-Z-22; CX 58A-Z-22; CX 59A-Z-22) These brands included "Meadow Gold " the dairy division s primary brand name; "Sanitary Dairy" brand ofCOJ marketed by the Minden, Louisiana plant (CX 47Z-8; CX 521, p. 9; CX 528Z-18; Lang 1057); and "Dixie Sunshine," a brand of COJ which was being test marketed by Beatrice just prior to the acquisition of Tropicana. (CX 47A-Z-204; CX 528) 159. In 1977 Beatrice dairy plants bought and resold prepackaged COJ from processors such as Tropicana, Home Juice, (29) Early Bird and Minute Maid. (CX 47) In the fiscal year ending February, 1978 Beatrice sold COJ through the following 37 fluid plants: 160. The Beatrice milk plant in Beckley, West Virginia sold 45 100 gallons of COJ for $126 000, processed by Tropicana and sold under the Tropicana label to the following types of customers: institutions, 35%; independent groceries, 38%; individual units of chain stores 27%. (CX 47F) 161. The Beatrice milk plant in Billngs, Montana sold 10 504 gallons of COJ for $24 478. About half of the total was purchased from Early Bird Juice Co. of Spokane, Washington and resold under the Early Bird label. The other half was purchased from another reconstituting Beatrice plant and was resold under the Meadow Gold label. 733 Initial Decision Sales of all this juice were 20% to institutions, 50% to independent and 30% to individual units of chainstores. (CX 47Z-Z; CXgreries, 47Z-157; CX 57-D) 162. The Beatrke milk plant in Boise, Idaho sold 52 700 gallons of COJ for $107 100. Of this total, 93% for 49 200 gallons and $99 600 was juice reconstituted at the plant, and 7% for 3 500 gallons and 500 was purchased in packaged form and resold by the plant. All of the reconstituted juice was sold under the Meadow Gold label. The resold juice was sold under the Early Bird label, processed by Early Bird, Spokane. The reconstituted juice was sold to the following types of customers: (a) independent groceries, 25%; (b) individual chainstore units, 75%. The resold juice was sold to the following types of customers: (a) independent groceries, 85%; (b) individual chainstore units, 15%. (CX 47Z-3) 163. The Beatrice milk plant in Champaign, Ilinois had sales of 258 gallons ofCOJ for $167 837. None of this juice was reconstituted at the plant and all was purchased in packaged form and resold by the plant. Eighty percent of the packaged juice was sold under the Vita-Fresh brand name, processed by Central States Processing, and 20% was sold under the brand name of Meadow Gold, processed by the Beatrice dairy plant in Mattoon, Ilinois. Institutions purchased 30% ofthe packaged juice, home delivery customers purchased 50%, independent groceries purchased 7%, and chain stores or individual units thereof, purchased 13%. (CX 47Z-) 164. The Beatrice milk plant in Clarksburg, West Virginia sold 29,549 gallons of COJ for $88 909. No juke was reconstituted at the plant, and all was purchased in packaged form and resold by the plant. All resold juice was sold under (30) the Tropicana label, processed by Tropicana. This resold juice was sold to the following types of customers: (a) institutions, 45% for 13 297 gallons and $40 009; (b) home delivery, 5% for 1 477 gallons and $4 445; (c) independent groceries, 50% for 14 775 gallons and $44 455. (CX 47Z-11) 165. The Beatrice milk plant in Dayton, Ohio sold 81 170 gallons of COJ for $169 759. Sixteen percent was reconstituted, 84% was purchased in package form and re-sold. The reconstituted juice has been sold under the Meadow Gold label and the plant has been reconstituting juice in the years 1977 and 1978. Reconstituted juice customers were 10% for institutions, 80% for independent groceries and 10% for other restaurants and cocktail lounges. The packaged juice in 1978 was Tropicana with approximately 5% going to institutions, 80% to independent groceries, 2.5% to chain stores' individual units and 10% to restaurants and cocktail lounges. (CX 47Z-3) 166. The Beatrice milk plant in Denver, Colorado sold COJ under the Meadow Gold label obtained from Beatrice s Greeley, Colorado Initial Decision 101 F. plant. For 1978, this plant sold 112 812 gallons of COJ for $245 562. Its customer mix is 14% institutional, 16% home delivery, 50% independent groceries, and 20% individual units of chainstores. (CX 47Z- 35) 167. The Beatrice milk plant in Durham, North Carolina did not reconstitute any orange juice. All was purchased and resold. Except for approximately 2% of 1978 purchases, all COJ purchas"d and resold was Tropicana. The 2% was purchased from Biltmore. COJ sales were 10 465 gallons at $25 061. The customer types were: institutions (20%); home delivery (10%); independent groceries (50%); IGA and co-op stores (20%). (CX 47Z-8) 168. The Beatrice milk plant in Eugene, Oregon sold 8,644 gallons ofCOJ for $16 800, purchased for resale from Dairy Gold of Eugene and sold under the Meadow Gold label. All sales were to independent groceries. (CX 47Z-0) 169. The Beatrice milk plant in Fort Wayne, Indiana never reconstituted orange juice. The re-sold juice which it carries, has been supplied by Benhil-Griffn or its predecessor, Dunlop, under that label. Approximately 5% of its orange juice sales have come from Tropicana, and have been sold under that label. In 1978, the plant sold 602 gallons of COJ for $44,620. The customer mix for this plant was: home delivery (35%), independent groceries (30%), and grocery wholesalers (35%). (CX 47Z-1) 170. The Beatrice milk plant in Gadsden, Alabama sold 115 691 gallons of COJ for $225 601. Of this, 87 394 gallons at $207 496 was reconstituted at the plant under the Meadow Gold label. The remainder of 28 297 gallons at $48 105 was Tropicana (31) that was purchased and resold. The customer type was the same for reconstituted and resold: institutional-25%; home delivery-15%; and independent groceries--O%. (CX 47Z-3) 171. The Beatrice milk plant in Grand Island, Nebraska has not reconstituted any COJ. All COJ is obtained from the Beatrice Lincoln plant. The sales volume was: 5 196 gallons at $10 810. In 1978, 30% of COJ sales was to institutions, 40% to home delivery and 30% to independent groceries. (CX 47Z-5) 172. The Beatrice milk plant in Grand Junction, Colorado sold 54,402 gallons of COJ for $128,800. Seventy percent was of Minute Maid and 30% of Meadow Gold obtained by intercompany transfer from Meadow Gold, Greeley, Colorado. Sales for these years were broken down into 20% to independent stores, 20% to distributors, and 60% to individual units of chain stores. (CX 47Z-6) 173. The Beatrice milk plant in Great Falls, Montana sold the following amounts ofCOJ: 16 134 gallons for $36 908. The plant has never reconstituted juice and has obtained its entire supply for resale BEATRICE FOODS CO., ET AL. 7(j 733 Initial Decision from Early Bird of Spokane, Washington. Sales were broken down to 30% institutional, 60% to individual units of chains to res, 5% to home delivery, and 5% to independent groceries. (CX 47Z-7) 174. The Beatrice milk plant in Greeley, Colorado sold 91 396 gallons of COJ of which 74 284 gallons for $142 334 was reconstituted and sold under the Meadow Gold label. The plant resold 17 112 gallons ofCOJ in 1978 for $39 621, 75% under the Oak Farm label and 25% under the Minute Maid label. Reconstituted juice was sold 60% to institutional, 20% to independent stores, and 20% to chainstores or individual units thereof. The plant also reconstituted COJ and sold large amounts on intercompany transfer to plants at Colorado Springs, Denver, Grand Junction, Topeka and elsewhere. (CX 47Z-8; CX 47Z-174; CX 57Z-7; CX 57Z-; CX 57R; CX 57I; CX 521, p. 11 175. The Beatrice plant in Honolulu, Hawaii has sold nothing but reconstituted COJ marketed under the Meadow Gold label. Twentyfive percent of the COJ has gone to independent groceries with the remainder going to individual units of chain stores. Sales in 1978 were 538 gallons of COJ for $127 317. (CX 47Z-50) 176. The Beatrice milk plant in Huntsvile, Alabama sold 237 341 gallons of pure juice of all types, representing $586 284 in sales, of which approximately 90% was COJ. None ofthe juice was reconstituted at the plant, and all was purchased in packaged form and resold by the plant. Eight-three percent was purchased from Tropicana and sold under the Grove Queen label and 17% was purchased from Cal- Tex and sold under the Vita Fresh label. Six percent of the juice was sold to institutions, (32) 19% to home delivery customers, 67% to independent groceries, 8% to chain stores or individual units thereof. (CX 47Z-51) 177. The Beatrice milk plant in Lima, Ohio sold 48 313 gallons of COJ for $114 501 all of which was purchased for resale. A total of 000 gallons was purchased from Ohio Pure Juice of Columbus, Ohio and sold under the Ohio Pure Juice label. The remainder was bought from Tropicana and sold under its label. This resold juice was sold to the following types of customers: institutional, 15%; home delivery, 5%; independent groceries, 75%; individual units of chain stores, 5%. (CX 47Z-57) 178. The Beatrice milk plant in Lincoln, Nebraska sold 40 800 gallons ofCOJ for $74 600, reconstituted at the plant and sold under the Meadow Gold label. This reconstituted juice was sold to the following types of customers: (a) institutions, 60%; (b) home delivery, 25%; (c) independent groceries, 10%; and (d) individual units of chain stores 5%. (CX 47Z-59) 179. The Beatrice milk plants in Louisvile, Kentucky sold 56 731 gallons ofCOJ for $116 708, 42 799 gallons at $74 132 of reconstituted Initial Decision 101 F. and 13 932 gallons at $42 574 of resold. The reconstituted was all Meadow Gold label. The resold was 90% Home Juice and 10% Tropicana. In 1978, sales of COJ were 2% institutional, 25% home delivery, and 73% to independent groceries, restaurants and the like. (CX 47Z-2) 180. The Beatrice milk plant in Mattoon, Ilinois sold 68 880 gallons of COJ for $75,641, reconstituted at the plant, and sold under the Meadow Gold label to independent groceries. (CX 47Z-4) 181. The Beatrice milk plant in Missoula, Montana has not reconstituted COJ. All was purchased and resold. In 1978, 32 610 gallons were sold for $84,485. Of this, 65% was packaged under the Meadow Gold label by the Early Bird Juice Co. of Spokane, Washington. The remainder was also purchased from the Early Bird Juice Co. but sold under the Early Bird label. The customer types in 1978 were: institutions (5%), home delivery (5%), independent groceries (70%), and individual units of chain stores (20%). (CX 47Z-9) 182. The Beatrice milk plant in Muncie, Indiana sold 24 032 gallons of COJ for $43 680. Five hundred gallons were purchased from Orchard Grove and sold under its label, with the rest being purchased from Ben Hil Griffn of Plymouth, Indiana for sale under its label. Eighty percent of sales were to independent groceries, 20% to institutional customers. (CX 47Z-70) 183. The Beatrice milk plant in New Bremen, Ohio sold 25 546 gallons of COJ for $58,099, all purchased in package form (33) and resold by the plant, under the Everfresh brand name, processed by Home Juice Company. Five percent ofthe COJ was sold to institutional customers, 75% to home delivery customers, 15% to independent groceries, 5% to chain store individual units. (CX 47Z-71) 184. The Beatrice milk plant in Opelika, Alabama sold 115 692 gallons of COJ for $309 570 purchased in packaged form and resold under the following labels: Vita Fresh, processed by Cal-Tex Citrus 60% (for 69,415 gallons and $185 742); Tropicana, processed by Tropicana, 40% (for 46 277 gallons and $123 825). This resold juice was sold to the following types of customers; (a) institutions, 25%; (b) home delivery, 25%; (c) independent groceries, 30%; and (d) individual units of chain stores, 20%. (CX 47Z-76) 185. The Beatrice milk plant in Orange City, Florida resold 277 011 gallons ofCOJ for $506 653 and packaged 138 505 gallons for $253 324 ofCOJ. The packaged juice was sold under the Meadow Gold label and obtained from Ardmore Farms. The resold juice was purchased from Tropicana and sold under the Tropicana label. Of the packaged juice 10% was sold to institutional customers, 60% to independent groceries, and 30% to individual units of chain stores. The resold juice was 733 Initial Decision sold to independent groceries for 61i% ofsaies with 35% to individual . units of chain stores. (CX 47Z-93) 186. The Beatrice milk plant in Ottumwa, Iowa sold 18 410 gallons of COJ for $43,020. All was purchased in packaged form and resold by the plant under the Vita-Fresh label, processed by Central States Processing, Columbia, Mo. This resold juice was sold to the following types of customers: institutions, 5%, home delivery, 15%; independent groceries, 57%; individual units of chain stores, 23%. (CX 47Z- 95) 187. The Beatrice milk plant in Radford, Virginia sold 85 713 gallons of COJ for $225 381 all purchased in package form and resold by the plant. The COJ was purchased in package form from Biltmore Dairy Farms for $23 311 and sold under the Meadow Gold brand name, or from Tropicana and sold under the Tropicana brand name. The Radford plant purchased $159 086 of Tropicana brand orange juice. Ten percent of the COJ sold to institutional customers, 20% to home delivery customers, 60% to independent grocery stores and 10% to chain store independent units. (CX 47Z-108) 188. The Beatrice milk plant in Reno, Nevada sold 20 334 gallons ofCOJ for $35 966; 13 000 gallons, representing $19 000 in sales, were reconstituted at the plant, and 7 334 gallons for $16 966 was purchased in packaged form and resold by the plant. All of the COJ was sold under the Meadow Gold brand name. The pre-packaged juice was purchased from Edward' s Marketing Company. Fifteen percent was sold to institutional (34) customers, 5% sold to home delivery customers, 10% to independent groceries, 70% to chain store individual units. (CX 47Z-113) 189. The Beatrice milk plant in St. Joseph does not reconstitute single strength orange juice. All COJ is purchased and resold. Ninetyfive percent of this is obtained from Tropicana and 5% is Vita-Fresh from Central States. In 1978, orange juice sales were 7 591 gallons at $18 020. The customer types were: institutions (5%), home delivery (10-15%), independent groceries (70-0%) and individual units of chain stores (5%). (CX 47Z-117) 190. The Beatrice milk plant in Salt Lake City, Utah sold 109 000 gallons of COJ for $297 600, purchased in packaged form and resold by the plant under the Meadow Gold label, processed by Edwards Marketing. This resold juice was sold to the following types of customers: (a) institutions, 1 %; (b) independent groceries, 60%; (c) individual units of chain stores, 39%. (CX 47Z-119) 191. The Beatrice milk plant in San Jose, California sold 128 713 gallons of COJ for $266 709. No juice was reconstituted at the plant and all was purchased in packaged form and resold by the plant under the Meadow Gold label, processed by Edwards Marketing. This resold Initial Decision 101 F. juice was sold to the following types of customers; (a) institutions 15%; (b) home delivery, 5%; (c) independent groceries, 20%; and (d) individual chain store units, 60%. (CX 47Z-122; CX 57Q). 192. The Beatrice milk plant in Topeka, Kansas did not reconstitute COJ. In 1978, all COJ was Meadow gold obtained by intercompany transfer from Greeley, Colorado. Sales for the plant are as follows: 1978, 8 203 gallons and $22 501. (CX 47Z-133) 193. The Beatrice milk plant in Tulsa, Oklahoma sold 170,412 gallons ofCOJ for $344 776, 50 049 gallons of which was reconstituted at the Tulsa plant and sold for $95 052. All of the juice reconstituted at the plant was sold under the Meadow Gold label. The juice which was purchased and resold by the plant was sold 90% under the Mr. Pure label, processed by the Home Juice Company, and 10% of the juice which was purchased and resold was sold under the Everfresh label also processed at the Home Juice Company. Ninety percent of the reconstituted juice was sold to home delivery customers, 5% of the reconstituted juice was sold to independent grocery stores and 5% was sold to chain store units. Ofthejuice which was purchased and resold 30% was sold to institutional customers, 15% to home delivery customers, 30% to independent groceries, and 25% to chain store units. (CX 47Z-134) 194. The Beatrice milk plant in Tuscaloosa, Alabama (35) sold 27 299 gallons of juice for $72 222, of which 95% was COJ. All was purchased in packaged form and resold by the plant. The resold juice was sold under the following labels: Vita-Fresh, processed by Vita- Fresh, 20%; and Tropicana, processed by Tropicana, 80%. This resold juice was sold to the following types of customers: (a) institutions 20%; (b) home delivery, 20%; (c) independent groceries, 30%; and (d) individual chain store units, 30%. (CX 47Z-138) 195. The Beatrice milk plant in Westervile does not reconstitute juice. All orange juice that is distributed is purchased from outside sources. In 1978, total COJ were 81 015 gallons at $139 047. Of this 75% was Meadow Gold obtained from the Louisville plant; 15% was Tropicana and 10% was Vita-Fresh. Five percent of sales were to institutions, 85% to home delivery, 5% to independent groceries and 5% to individual units of chain stores. (CX 4 7Z-14 7) 196. The Beatrice milk plant in Zanesville, Ohio sold 20 402 gallons ofCOJ at $40 931. Of this, 16 591 gallons at $27 948 was reconstituted under the Meadow Gold label. The remainder (3 811 gallons at $12,- 983) was purchased and resold Tropicana. For the reconstituted sales the customer types were as follows: institutions (10%); home delivery (10%); independent groceries (80%). The customer type for the resold juice was: 97% home delivery and 3% independent groceries. (CX 47Z-150) 733 Initial Decision For the fiscal year ending February 28, 1978 Beatrice sold CQJ through the following six additional plants which were not milk producers:

197. The Beatrice dairy plant in Colorado Springs, Colorado sold 035 gallons of COJ for $130 716, all purchased in packaged form and sold under the Meadow Gold label, processed by Meadow Gold Greeley, Colorado. This resold juice was sold to the following types of customers: (a) institutions, 15%; (b) home delivery, 40%; (c) independent groceries, 20%; (d) individual chain store units, 25%. (CX 47Z- 22) 198. The Beatrice dairy plant in Dothan, Alabama has not reconstituted any orange juice. Sales in 1978 were 28 266 gallons ofCOJ at $65 229. Of this, 70% was Vita-Fresh obtained from Meadow Gold Opelika; 20% was Meadow Gold obtained from Orange City; 10% was Tropicana obtained from Orange City. Five percent of sales was to institutions and 95% was to independent groceries. (CX 47Z-36) 199. The Beatrice dairy plant in Meinerz, Louisiana had sales of $306 118, purchased by Meinerz Creamery in packaged form from Ben Hil Griffn and sold under the Sun-Blossom brand (36) label. All single strength orange juice was sold to other dairy companies. (CX 47Z-6) 200. The Beatrice dairy plant in Minden, Louisiana sold 169 753 gallons of COJ for $288 688, all of which was reconstituted at the plant. This reconstituted juice was sold under the Sanitary Dairy label. The juice was sold to the following types of customers: institutional, 5%; home delivery, 15%; independent groceries, 40%; individual units of chain stores, 40%. (CX 47Z-8) 201. The Beatrice dairy plant in Pocatello, Idaho sold 24 329 gallons of COJ for $56 574 purchased in packaged form and resold by the plant. The resold juice was sold under the Meadow Gold label, processed by Meadow Gold, Boise, Idaho. This resold juice was sold to the following types of customers: independent groceries, 50%; individual units of chain stores, 50%. (CX 47Z-106) 202. The Beatrice dairy plant in Wichita, Kansas sold 5 453 gallons ofCOJ for $14 722, all of which was purchased for resale from Vita- Fresh Company of Columbia, Missouri and sold under the Vita-Fresh label. Eighty percent of sales were to home delivery and 20% to independent groceries. (CX 47Z-149) 203. In 1977 Beatrice had the following 37 milk plants in the United States (CX 47; CX 241-243):

Beckley, WV (CX 47F) Billings, MT (CX 4 7Z Boise, 10 (CX 4 7Z Champaign, IL ICX 47Z-6) Initial Decision 101 F. Clarksburg, WV (CX 47Z 11) Dayton. OH (CX 47Z 33) Denver (Oxford-Englewood), CO (CX 47Z 35) Durham, NC (CX 47Z-38) Eugene, OR (CX 47Z-40) Fort Wayne, IN (CX 47Z-41) Gadsden, AL (CX 47Z-43) Grand Island, NE (CX47Z-45) Grand Junction, CO (CX 4 7Z -46) Great Falls, MT (CX 4 7Z -4 7) Greeley, CO (CX 47Z-48) Honolulu, HI (CX 47Z 50) Huntsvile, AL (CX 47Z 51) Lima (CX 47Z 57) Lincoln, NE (CX 47Z 59) Louisville, KY (CX 47Z-62) Mattoon, IL (CX 47Z-64) Missoula, MT (CX 4 7Z -69) Muncie, IN (CX 47Z 70) New Bremen, OH (CX 47Z 711) Opelika. AL (CX 47Z 76) (37) Orange City, FL (CX 47Z 93) Ottumwa. IA (CX 47Z 95) Radford, VA (CX 47Z 108) Reno, NV (CX 47Z 113) St. Joseph, MO (CX 47Z 117) Salt Lake City, UT (CX 47Z 119) San Jose, CA (CX 47Z 122) Topeka. KS (CX 47Z 133) Tulsa, OK (CX 47Z 134) Tuscaloosa, AL (CX 47Z 138) Westerville, OH (CX 47Z 147) Zanesvile, OH (CX 47Z 150) 204. In 1977, Beatrice s Meadow Gold brand of COJ was sold to grocery stores in the following 18 SMSAs where Tropicana did not so sell its COJ (CX 47; CX 526I- , Z-l-Z-): Anniston, Indiana Billngs, Montana Boise City, Idaho Eugene, Oregon Evansvile, Indiana Fort Collns, Colorado Greeley, Colorado Honolulu, Hawaii Lawton, Oklahoma Lincoln, Nebraska Omaha, Nebraska Owenboro, Kentucky Provo-Drem. Utah Reno, Nevada Salinas-Seaside-Monterey, Caliornia BEATRICE FOODS CO., ET AI,. 769 733 Initial Decision Santa Cruz, California Sioux City, Iowa Springfield, !Wnois 205. In 1977 Beatrice s brands of COJ were sold and delivered by Beatrice milk plants to individual stores of supermarket chain stores in the United States. (CX 528) These included stores of chains such as Safeway, Kroger, Albertson s, IGA, King Soopers, Winn-Dixie, Pantry Pride, Super Valu, Dilon, Alled, Humpty-Dumpty, Piggly-Wiggly, Star Market and Jewel and A&P. (CX 47Z-3 , Z-22, Z-35, Z-50 Z-8, Z-9, Z-73, Z-93, Z-134, Z-195; CX 526Z- to Z-12; CX 526Z-6 to Z-12; Parker 4323, 4330) 206. In fiscal 1978, Beatrice dairy plants bought and resold 643 511 gallons of packaged Tropicana COJ. (CX 47) 207. Prior to its acquisition of Tropic ana, Beatrice offered its brands ofCOJ in carton and plastic containers and had just started distributing the glass Dixie Sunshine private label. (CX 528Z-29; CX 180F) (38) 208. In the fiscal year ending February 28, 1978, the following amounts of Meadow Gold Brand COJ processed by Beatrice were sold by the following plants to grocery stores: Gallons Sold to Grocerv Stores Dollar SalesPlant 202 365 (CX47Z-2; ex 57-Billngs 4Boise 49200 99Champaign 600713 (CX47Z-(CX47Z-6; ex 57T) 3170 6 Colorado Springs 26 116 58Dayton 21,729822 (CX47Z-33)(CX47Z-22) 10389 171 (CX47Z-35)Dathan 4862 11Gadsden 124,498440 (CX47Z-36;CX57H)(CX47Z-43)Grand Island52436 243 (CX47Z-45. 171; CX57X) 1,559 3 Grand Junction 640 (CX47Z-46) 16321 38 933 (CX 47Z- , Z-174: CX 57Z-Greeley 29714 56Honolulu 53,538 127 317 (CX57Z-16) 192 (CX47Z-59)Uncoln 6 120 11 116 (CX47Z-62)Louisvile 31 243 54Matton 68,880 75 574 (CX47Z-106:CX57G)Pocatello 24329 56 641 (CX43; CX47Z-64:CX57D)Reno 279 (CX47Z-134)Tulsa 10.400 15 200 (CX47Z-113: CX57Z- 27567 52Westerville 428 (CX47Z-147:CX57Z-18) 6076 10 Zanesvile 358 (CX 47Z-150; CX 57Z- 13,273 22Total 518.363 $1 028 981 209. Grocery stores mark up dairy products 25% for resale to consumers. (Parker 4376; Clayton 6049) Hence the (39) retail shelf value of the COJ processed by Beatrice and sold through grocery stores in fiscal 1978 was $1 286 226.

Initial Decision 101 F. 210. In fiscal 1978, Beatrice processed and sold 135,802 gallons of COJ to grocery stores under the Sanitary Dairy brand name from its plant at Minden, Louisiana. The sales price of that COJ was $230 950 and it had a retail sales value of $288 688. (CX 47 8; CX 57S) 211. In fiscal 1978, Beatrice processed and sold to grocery stores in the United States a total of 654,165 gallons of COJ for $1 259 931. (findings 208 and 210) The retail sales value of that COJ was $1,574,- 914. (findings 208-10) 212. In fiscal 1978 the following amounts of COJ not processed by Beatrice were sold to grocery stores under the Meadow Gold brand by the following Beatrice fluid milk plants:

Gallons Sold to Groceries Dollar Sales $ 16,800 (CX 47Z-40) Eugene 8,644 49,423 (CX47Z-69) Missouli 19,077 992 (CX47Z-93) Orange City 124 655 227 200 (CX47Z-108; CX57Z-12) Radford 52896 139 573 (CX47Z-113) 13 Reno 5,867 624 (CX47Z-119;CX57L) Salt Lake 107 910 294 367 (CX47Z-122;CX57Q) San Jose 102,970 213 979 Total 422 019 $954 The retail value of the Meadow Gold brand COJ sold through grocery stores in 1977 that was not processed by Beatrice was $1 193 724. (finding 209) 213. In fiscal 1978 the total amount of Beatrice Meadow Gold or Sanitary Dairy brand COJ sold to grocery stores was 1.08 milion gallons with a retail value of $2.77 milion. Beatrice label COJ had a market share of .57% of the retail COJ market measured in gallons or .67% measured in dollars. (findings 120, 208-12) 214. In fiscal 1978, the total amount of COJ processed by Beatrice sold to grocery stores was 654 165 gallons with a retail value of $1,574 914. (finding 211) Beatrice processed .35% of the retail COJ market measured in gallons or .38% measured in dollars. (finding 121) (40) 215. In fiscal 1978 the following amounts ofCOJ processed by Beatrice were sold by the following dairy plants: Gallons Processed Billngs 252 (CX47Z-2; CX57- Boise 200 (CX47Z- Champaign 851 (CX 47Z-6; CX 57T) Colorado Springs 035 (CX47Z-22) Denver 112 812 (CX47Z-35) Dayton 12.987 (CX 47Z-33) Dothan 653 (CX47Z-36) 733 Initial Decision Gallons Plant Processed 394 (CX47Z-43)Gadsden 87 196 (CX47Z-45: CX 57X)Grand Island 5Grand Junction 16 320 (CX47Z-46) 284 (CX348;CX57Z-Greeley 74 (CX 47Z-50: CX 57Z- 16)Honolulu 53538 800 (CX47Z-59)Lincoln 40 799 (CX47Z-62; CX57N)Louisville 42 (CX47Z-64;CX57D)Mattoon 68.880753 (CX47Z-68;CX575)Minden 169 (CX47Z-106;CX57G)Pocatello 24,329 299 (CX 57Z.Reno 18 049 (CX57Z-TulsaTopeka508.203 (CX47Z-147;CX57Z.18)Westervile 60.761 (CX47-133;CX571) 453 (CX 57V) ZanesvilleWitch ita 5 591 (CX47Z-15D;CX57Z- 002,439 (41)Total 1 216. Beatrice s market share in the total processing of COJ (wherever sold) in fiscal 1978 was about .39%. (findings 127, 219) 217. COJ processed by Beatrice is sold to retail grocery stores at price low enough to allow sale to consumers at a price per serving which is 25% to 35% lower than national brands ofCOJ like Tropica- , Kraft and Minute Maid. Beatrice s production ofCOJ is primarily in cartons and blow molded plastic containers. (CX 738A; CX 739A; CX 740B; CX 753E; CX 755C; CX 770A; CX 922B; Parker 4324; Koch 4876) One gallon blow molded containers of COJ offer consumers a lower per serving cost than do half-gallon cartons of COJ. (Jessup 2900; Parker 4325-26; CX 1086C) Last year in Mattoon, Ilinois, Beatrice s one gallon plastic container ofCOJ sold to consumers for $2.49 compared to a per gallon price for glass or carton half-gallon containers of Tropicana, Minute Maid and Kraft of $3.38 to $3.78. (Parker 4325-26) c. Competition Between Beatrice and Tropicana 218. Beatrice dairy plants sell COJ by reconstituting bulk frozen concentrated orange juice and selling it under its own labels and/or resellng COJ processed by others and sold under Beatrice s or the processor s labels. (CX 47A-Z-204) 219. Tropicana recognized as competitors dairies which sell their own brand ofCOJ in the retail market. (CX 527Z-10, no. 191; CX 593 220. In 1976 through 1978, Tropicana s western division sales manager reported that Tropicana s sales to 7/lls had been hurt by 772 FEDERAL TRAHE COMMISSION DECISIONS Initial Decision 101 F. the COJ packed in plastic jugs sold by the Beatrice milk plant in Salt Lake City. (CX 661A; CX 663C; CX 770A; Dutt 3825) The Beatrice dairy in Salt Lake City was packing a private label COJ under the Meadow Gold label, processed by Edwards Marketing. (CX 57L) Tropicana s vice president for sales reported that the 7/11 store managers preferred Meadow Gold COJ because of the store-door delivery. (CX 808C) 221. Reports of Tropic ana sales personnel recogoized as competitors Beatrice milk plants distributing COJ in Alabama, California, Flori- , Ohio and Oklahoma. (eX 744A; CX 1301B; CX 922B; CX 753E; CX 807F; CX 808C, H; CX 593, p. 350) 222. Beatrice dairy plants delivered COJ to the warehouses of King Soopers grocery chain store in Denver (CX 528V) and IGA in Durham. (CX 47Z-38) The Beatrice dairy plant in Champaigo has had a chain store warehouse pick up COJ (42) from its loading dock. (Parker 4384) 223. In 1977, King Soopers was a chain of approximately 46 retail no. 83; CXgrocery stores located in eastern Colorado. (CX 528U, 1107B) Beatrice s Greeley, Colorado milk plant delivered Meadow Gold COJ to King Sooper s dairy warehouse in Denver for distribution to individual grocery stores by King Soopers. (CX 528V, no. 86) In 1977, Meadow Gold COJ was sold in 34 Denver area King Soopers grocery stores. (CX 528V, no. 85) 224. In 1977, Tropicana sold Tropicana brand COJ to King Soopers grocery stores and delivered COJ to the distribution warehouse in Denver. (CX 528V, X) 225. In 1977, Beatrice s San Jose milk plant sold Meadow Gold COJ to Fry s Food Stores, a chain of25 stores, in San Francisco. (CX 528W no. 90; CX 236E) Also in 1977, Tropicana sold Tropicana COJ to Fry Food Stores in San Francisco. (CX 528W, no. 89) 226. In 1977, Beatrice s Meadow Gold and Tropicana s brand ofCOJ were both sold to grocery stores in the following 48 SMSAs (CX 47; CX 526I- , Z-1-Z-5):

Akron, Ohio Lynchburg, Virginia Atlanta, Georgia MeJbourne Titusville- Birmingham, Alabama Cocoa, Florida Bloomington-Normal, Illnois Miami, Florida Bradenton, Florida Oklahoma City, Canton, Ohio Oklahoma Champaign-Urbana-Rautol Orlando, Florida Ilinois Panama City, Florida Chattanooga, Tennessee Pensacola, Florida Cincinnati, Ohio Peoria, Illnois Colorado Springs, Colorado Pueblo, Colorado Columbus, Ohio Roanoke, Virginia Dayton, Ohio St. Louis, Missouri Daytona Beach, Florida Salt Lake Citv-Oaden. V.l. J.J.Uv.J. J.' ........ v... .J~ 733 Initial Decision Denver, Colorado San Jose, California Fayetteville-Springdale, Arkansas Sarasota, Florida Fort Lauderdale, Florida Springfield, Missouri Fort Meyers, Florida T aUahassee, Florida Fort Smith, Arkansas Tampa-St. Petersburg, Gadsden, Alabama Florida Gainesville, Florida Terre Haute, Indiana Hamilton-Middletown, Ohio Topeka, Kansas HuntsviHe, Alabama Tulsa, Oklahoma Indianapolis. Indiana West Palm Beach-Boca Lakeland-Winterhaven, Florida Raton. Florida (43) Lexington, Kentucky Louisville, Kentucky H. Beatrice s Plans To Enter the Market a. Beatrice s Intent to Expand COJ Sales 227. In 1975, Beatrice was processing COJ at dairy plants in Gadsden, Honolulu, and Louisville. (CX 49A-B; CX 47Z-9; CX 47Z-5; CX 47Z-2) 228. In 1976, Beatrice urged its plant managers to process COJ for sales to schools because: "With capabilities of packaging orange juice . . . in various plants across the country, we are at a definite advantage." (CX 54) 229. Also in 1976, Beatrice requested Peter S. Goldman, an offcer in a successful COJ processor, Johanna Farms, Inc., for an assessment of the feasibilty of processing COJ, through ajoint venture, using the Beatrice Deland (Orange City) dairy plant. After an inspection ofthe plant, Mr. Goldman rejected the idea on the grounds that an economical operation would have entailed a "very considerable investment. (CX 529A) 230. Realizing that the COJ market had more than doubled from 1971 to 1976 (77 milion gallons in 1971, and 161 milion gallons in 1976), Beatrice urged its plant managers to follow the example of the Beatrice milk plant in Tulsa which had successfully become a processor of COJ. (CX 3) 231. By February 28, 1978, COJ was processed by the following 12 Beatrice dairy plants (CX 47):

Boise, Idaho Louisville, Kentucky Dayton, Ohio Mattoon, Ilinois Gadsden, Alabama Minden, Louisiana Greeley, Colorado Reno, Nevada Honolulu, Hawaii T uJsa, Oklahoma Lincoln, Nebraska ZanesviJJe, Ohio b. Orange City Plant 232. In early 1977 Beatrice s Orange City milk plant, near Deland, Initial Decision 101 F. Florida had been losing money for two years because of the increase of transportation costs. (Polidoro 537&-77; 5381) The plant added a $120,000 packaging machine to increase production of paper cartons which it could use to package COJ as well as milk. (Polidoro 5370) (44) 233. Since 1976 the plant had been packing and distributing COJ processed by Ardmore Farms of Deland, Florida. The processor squeezed the COJ from fresh oranges, pasturized it and delivered it to the Orange City plant in bulk tankers. (CX 62B; CX 94A; CX 125B; CX 47Z-189) The Orange City plant then packaged the COJ directly from the bulk tankers in gallon plastic containers, gave it a 21 day shelf life code, and distributed it under the Meadow Gold label. (CX 94A) 234. The Orange City plant also distributed Tropicana label COJ in glass containers which it had resold since 1970, and which amounted to 67% of its COJ sales in 1976 and 1977 (CX 47Z-93-94; CX 47Z-189) 235. By February 14, 1977, the Orange City plant manager had decided to expand sales by reselling COJ packed on his new machine to other Beatrice dairies. He saw a market for freshly squeezed COJ (as different from COJ made from concentrate). (CX 63B) c. Dixie Sunshine 236. The Beatrice dairy marketing board was a committee of eight Beatrice dairy division executives, including the director of advertising and the four regional marketing directors, and a representative of Beatrice s advertising agency. The committee was chaired by the dairy division director of marketing. Top managers of Beatrice, such as James Dutt, then corporate executive vice president, and Wiliam Polidoro, dairy division executive vice president, attended meetings of the board. (CX 61; CX 117; CX 180) Mr. Dutt created the dairy marketing board. (CX 528Z-11) 237. One of the main purposes of the dairy marketing board was to generate and implement ideas for new products. (CX 528Z-10) It also provided marketing assistance to dairy plants (CX 528C); assessed each Beatrice plant for advertising of new products (CX 6IB; CX 528Z-12, Z-22); approved names for new products (CX 5281); issued guidelines to dairy division plants concerning the advertising of dairy products (CX 528C); authorized advertising for new products (CX 528Z 12); and approved all advertising for dairy products. (CX 528Z-21) 238. On March 1, 1977, the dairy marketing board was briefed on the Orange City plant's plans to market fresh COJ (not made from concentrate). They also discussed the growing marketing opportunities for COJ packaged in paper containers and distributed by dairies. (CX 12A) 239. On April 4, 1977, the dairy marketing board (45) discussed the Orange City plant' s packaging ofCOJ in paper and plastic and buying ...

U.Ln..l.luu.L .l' '-'-''U U'-. , -'.l 733 Initial Decision a private label COJ in glass from Tropicana. (CX 12B) 240. On May 4, 1977, the dairy marketing board considered the name "Dixie Sunshine" as the Beatrice COJ to be distributed by the Orange City milk plant. (CX 12C) 241. By May 31, 1977, the brand name Dixie Sunshine had been adopted for the COJ to be processed by Ardmore Farms and packed by the Orange City plant. (CX 94A) For years Beatrice had used the label "Meadow Gold" for milk and for reconstituted COJ, and the dairynewmarketingname.board felt(Polidorothat this freshly squeezed5373)COJ needed a 242. By June 20, 1977, the dairy marketing board offcials were deciding to expand the Dixie Sunshine program to include COJ packed in glass as well as paper and plastic, and had approached Tropicana to obtain a private label packed in glass quarts and halfgallons. (CX 101) By August 16, 1977, the dairy marketing board agreed that either Tropicana or Ardmore Farms should process the Dixie Sunshine glass containers. (CX 12C) In December 1977 Tropicana agreed to process the Dixie Sunshine COJ in glass. (French 5684) 243. In a meeting on July 12, 1977, the dairy marketing board decided that Dixie Sunshine COJ should be sold in markets as far as it could reasonably be shipped. (CX 12C) They were looking at the national market for COJ. (CX 110) 244. In the meeting on July 12, 1977 the dairy marketing board discussed competing with established brands ofCOJ like Tropicana by sellng Dixie Sunshine at a lower price. (CX 12C) 245. By July of 1977, the dairy marketing board had considered expanding the sale of Dixie Sunshine by supplying 30-35 Beatrice milk plants that were already processing or could process COJ from frozen concentrate. They felt they could have national distribution in 18 months. At that time they looked at the introduction of Dixie Sunshine in Florida as a test market. (CX 121; CX 122; CX 124; Rosenberg 1172-73) 246. The planned production of Dixie Sunshine COJ in October 1977 was thwarted because of a shortage of fresh orange juice caused by the freeze in January 1977. Neither Ardmore Farms nor Tropicana could supply the Orange City plant for a short while. (French 5682) 247. In a national meeting of the Beatrice dairy division on November 15, 1977, held in Colorado Springs, the Dixie Sunshine COJ was introduced and promoted to plant managers from all Beatrice dairy plants by various Beatrice executives (46) including Wallace Rasmussen, Beatrice s chairman. (CX 140G 248. In November 1977, the dairy marketing board appointed a task force" composed of three Beatrice regional marketing directors to develop a marketing plan and pricing structure for Dixie Sunshine. They were to study the "wholesale prices of competitive products g.

776 FEDERAL TRADE COMMISION DECISIONS Initial Decision 101 F. (Tropicana, Minute Maid, Kraft, Hood, and private labels)," in Ohio and Alabama. (CX 143C; CX 107) In July of 1977, members of the board had studied COJ retail prices in Chicago and New York. (CX 109; CX 151) 249. In November 1977, Richard Voell, Beatrice chief corporate offcer who negotiated the acquisition of Tropic ana, became aware of the Dixie Sunshine program. (CX 2310C) 250. The Orange City plant was stil losing money and in the fall of1977 the decision was made to close the plant. The plant was closed on February 28, 1978. (Polidoro 5383) 251. Dixie Sunshine was being packaged in paper cartons by the Orange City plant by November 17, 1977. (CX 12C) By February 20 1978, Tropicana was processing Dixie Sunshine COJ. (CX 168) 252. At the meeting on March 8, 1978, the dairy marketing board concluded that the Dixie Sunshine program would not be backed by advertising and that the paper packaging would cease. (CX 12E) The program stopped shortly thereafter. (French 5686-7) I. Beatrice s Ability To Expand Its COJ Operation 253. The reconstituting of COJ by dairy milk plants has been responsible for much ofthe great growth ofthat market in recent years. (CX 593, p. 341; CX 9Ila; CX 1005C; CX 1084B; Jessup 2897) 254. Dairies process and package COJ on the same machinery used for the processing of milk. (CX 3D; CX 14D; CX 528S; RX 5D; RX 17H; RX 18D; Munkelt 724-25; Donovan 1292, 1300; Davis 3153; Dutt 3753 , 3831; Parker 4313- , 4320, 4393; Koch 4912-13; Piskac 5167; Brick 5292, 5295, 5314, 5318-19, 5325, 5335; Polidoro 5401-D2) 255. A milk plant uses the same storage facilities for COJ and for milk. (Donovan 1293, 1300; Koch 4875) 256. The distribution system of milk plants and dairies can be used for the distribution ofCOJ to the retail market. (CX 663D; CX 770B; CX 995B; CX 1089B; CX Il60C; CX 1228; CX 1650, pp. 104-105; Lang 1067--9; Donovan 1300; (47) Jessup 2897-98; Parker 4393; Brick 5335) 257. Milk plants sometimes process COJ as an alternative to purchasing packaged juice from a processor such as Tropicana when the price of the packaged product gets too high. (CX 892A-E; CX 1221A- 258. Tropicana, in the early years of its existence, relied almost exclusively on dairies for the distribution ofits COJ. (CX 1650, p. 104; Barnebey 3887) COJ, like milk, has a short shelf life, and the only dependable means of refrigerated distribution was by the dairies. (CX 1650, p. 105) Tropicana stil uses dairies for some of the distribution of its COJ. (e. CX 1221B; CX 1223; CX 1233A-G; CX 1234; CX 1338A; CX 1342A- , pp.

733 Initial Decision 259. COJ, with other dairy products mark;'ted by milk plants, IS marketed to consumers through the dairy case or, to a lesser extent through the produce section of the grocery store. (CX 14F; CX 593 112, 116; CX 665C; CX 1251; Munkelt 714-15, 798-99; Lang 1052; Donovan 1310, 1313) Dairies sellng COJ have an advantage in obtaining shelf space in the dairy case, having existing ties with the dairy case buyers. (CX 593, p. 325; Lang 1069-70; Donovan 1300, 1310) 260. Dairy products and COJ are subject to similar advertising and promotional techniques. (Donovan 1313) The reputation of a dairy dairy products is of great assistance in the sale ofCOJ to grocery store customers. (CX 6F; CX 660B; CX 786B; Donovan 1313-16; Parker 4395) 261. In the past few years the number of milk plants in the United States has been declining. (CX 593, p. 349; Goldman 975; Donovan 1319; Mirapaul 1582--3; Davis 3182; Parker 4335-36; Karnes 4818; Koch 4899-4900; Greiner 5075; Brick 5305, 5332) The trend has been toward larger, more effcient dairies. In 1972 there were 1287 milk plants with 20 or more employees in the United States and in 1977 there were 907. (CX 598A- 262. Beatrice uses blow mold containers in its milk and COJ packaging at a number of its milk plants, and produces its own blow molded containers for use in its plants in Lincoln, Nebraska; Topeka, Kansas; Champaigo, Ilinois; Denver, Colorado; Salt Lake City, Utah and Alabama. (Piskac 5171-5174) 263. Beatrice sells milk in a broader geographic area than any other dairy in the United States. (Karnes 4809) Beatrice also operates more milk plants than any other dairy in the United States. (Karnes 4810) 264. In 1977, prior to its acquisition of Tropic ana, Beatrice operated milk plants in 23 states, all of which plants (48) were capable of processing and selling COJ to the retail market (CX 526T; CX 528S no. 76):

Alabama Missouri California Montana Colorado Nebraska Florida Nevada Georgia Ohio Hawaii Oklahoma Idaho Oregon Ilinois Texas Indiana Utah Iowa Virginia Kansas West Virginia Kentucky Initial Decision 101 F. 265. Beatrice had dairy plants or distribution facilities in the following additional states (CX 526V, no. 75):

Arizona New Yark Maryland North Carolina Michigan Pennsylvania Minnesota Tennessee Mississippi Wisconsin New Jersey Wyoming 266. In 1976, Beatrice operated eight refrigerated storage warehouses in New England, four in Florida, three in Chicago, one in Denver, three in Los Angeles, and one each in Scranton, Pennsylvania; Detroit, Michigan; Lincoln, Nebraska; Kansas City, Kansas; Denver, Colorado and the other in Allentown, Pennsylvania. (CX 18C; CX 19D; CX 280) 267. Small dairies can encounter diffculty in processing and sellng COJ. They may have their equipment already being used to full capacity producing milk, or may lack an assured supply of bulk FCOJ or the ability to produce a quality product, and they may not have a recognized brand name. (CX 915C; CX 1650, pp. 108-9; Mirapaul 1582; Jessup 3064-5; Davis 3165) Some dairies do not have suffcient refrigerated warehouse space and trucks available to make chain store warehouse deliveries although they do have the distribution system to deliver store-door. (CX 1650, pp. 105-106; Parker 4321) 268. In 1978, the Beatrice Greeley plant reconstituted and transferred COJ to other Beatrice plants or facilities in Pueblo, Colorado Springs, Grand Junction, and Denver, Colorado; Salt Lake City, Utah; Topeka and Wichita, Kansas; Lincoln, Nebraska; St. Joseph, Missouri; and Ottumwa, Iowa. (CX 521, pp. 9-11; Piskac 5102-03) 269. Beatrice s Mattoon plant reconstitutes COJ and (49) supplies COJ to Beatrice s milk plant in Champaign. (Parker 4313, 4388) Beatrice s Dayton plant reconstitutes and supplies COJ to the Westerville, Lima and New Bremen, Ohio plants. (Koch 4912, 4918, 4974) 270. Because of the perishability of their products and transportation costs, most dairies, delivering store-door, sell to customers in a radius of 150 miles from their plant, although some wil try to deliver store-door up to 400 miles from the dairy. (CX 204K; CX 218J; CX 202K; Brick 5294; Polidoro 5379) Delivering to warehouses, however COJ processors, including dairies, can effectively sell up to 500 miles from their plant. (Munkelt 718-20, 854; Goldman 961-63 1010; Lang 1056, 1131; Donovan 1296-97) 271. Johanna Farms delivers COJ store-door and recently to chain store warehouses. (Goldman 974, 1019) Some chain store warehouses pick up direct COJ from the processor. (Goldman 994, 1019) 733 Initial Decision 272. In 10 years Hood' s delivery ofCOJ to the warehouses ofgrocerchains has gone from almost nothing to about 40% of its sales. (Donovan 1309, 1312) 273. In 1971, about 15% of the COJ produced by Home Juice was delivered to warehouses. By 1979, warehouse delivery accounted for approximately 50% of Home Juice s COJ shipments. (Lang 1063-4) 274. Beatrice has semitrailers and van trailers necessary to deliver COJ to warehouses. (RX 121A, Parker 4308, 4321, 4389-90; Koch 4872, 4898, 4926) For example, Beatrice s Lincoln plant owns 13 trucks ofthe 40 foot size. (Piskac 5184) These trucks are used to haul its products including COJ to its branches and its distributors. (Piskac 5184, 5188-9, 5194-96, 5201-05, 5239-40) 275. Beatrice delivers milk and other dairy products by its refrigerated truck fleet to the customer s store or warehouse, and to a p. 2;lesser extent to the consumer s home. (CX 16B; CX 17B; CX 18, CX 19B; CX 20D; CX 21C; CX 22, p. 38; Dutt 3798; Koch 4871 , 4935) 276. The Beatrice Champaign milk plant serves a warehouse account through pick up at its dock. (Parker 4384; Polidoro 5403, 5427) From 1973 to 1975, Beatrice s Dayton plant delivered dairy products to a warehouse which serviced 75-0 convenience store outlets. (Koch 4952-53) 277. Beatrice s Mattoon plant manager would like to sell COJ via warehouse delivery. (Parker 4413, 4384) Beatrice milk plants could save money on transportation costs by having to make only one stop rather than incurring the costs of many (50) stops at individual stores. (Parker 4384-85; Koch 4954-55) 278. In 1978 Beatrice operated a large refrigerated truck fleet based at its New Bremen, Ohio plant. (Granger 4531-32; RX 121H) The fleet has been used to transport COJ between the processing plants and distribution centers. (Koch 4927) 279. Beatrice has 1 400 food brokers selling its products in the United States. Brokers are an effective method of obtaining access to shelf space in grocery stores. (CX 18B) 280. In 1978, Beatrice spent $150 000 000 on advertising and projected spending from $170 000 000 to $175 000 000 in 1979. (CX 192B; Dutt 3795-96) Its advertising expenses ranked 17th among all industrial corporations in the United States. (CX 192B) 281. The production, distribution and marketing of COJ and yogurt are similar. Both products are processed in dairy plants and require refrigerated storage and delivery facilities due to short shelflife; they are distributed to grocery stores and sold to the dairy buyer and placed in the dairy case of grocery stores. (Dutt 3814-18) 282. Dannon yogurt was sold only in the New York metropolitan area when it was acquired by Beatrice in 1959. (CX 29D; CX 527Z-6; , pp.

Initial Decision 101 F. Dutt 3814; Karnes 4757) Dannon is now sold coast to coast by Beatrice s dairy division. (Dutt 3814-15; Karnes 4757) 283. Beatrice distributes Dannon Yogurt coast to coast using a store-door delivery system. (CX 2E; Dutt 3817; Karnes 4757-58) Johanna Farms and other firms distribute yogurt and COJ to warehouses of chain grocery store customers. (Goldman 972-73; Dutt 3816 18; Karnes 4760) 284. Prior to its acquisition of Tropicana, Beatrice processed and distributed COJ in Australia by its subsidiary, Patra Holdings Pty. Ltd. ("Patra ). (CX 527- , 70, 71, 74, 76, 81, 82) 285. Under Beatrice s ownership, Patra s COJ processing operations have been expanded. (CX 24F; CX 26F; CX 291; CX 463A- 286. Beatrice executives learned production and marketing skils through Patra which could be used in this country. (CX 144A; CX 436-1) 287. A major corporate objective of Beatrice is to expand its products with local distribution into national distribution. (CX 2B; 527"0"; CX 1553; Dutt 3796, 3816; Karnes 4802--4) (51) 288. Milk plants which process COJ achieve more effcient distribution oftheir milk through combined deliveries of milk and COJ. (Goldman 1021-22; Dutt 3754, 3830-31) Dairies have added products to their milk trucks in order to reduce distribution costs. (Goldman 1022; Dutt 3754; Parker 4391) 289. Milk is a low margin product compared to most food items. (CX 1250B; Karnes 4751) COJ is a high margin product compared to other food products. (Karnes 4790) 290. Because of the higher profitabilty of COJ sold to the retail market compared to milk, and the excess capacity in almost all of Beatrice s milk plants, Beatrice had an economic incentive to expand its retail COJ business. (CX 14C; CX 29D; CX 202--X 295; CX 528Z- 20; Dutt 3754; Karnes 4790) 291. Dairy companies such as Hood, Johanna Farms, Dean, Knudsen and Foremost-McKesson have successfully entered the market by sellng COJ to grocery stores along with their other dairy products. (CX 14B; CX 593, pp. 270, 272-73; CX 807F; CX 1086C; CX 1650 92- , 100-101; Lang 1132; Jessup 2799; Barnebey 4049- , 4053; Brick 5312-13) 292. Johanna Farms delivers COJ along with yogurt to chain grocery store warehouses and delivers COJ along with all its dairy products including milk store-door to other grocery stores. (Goldman 972- , 994, 1021-22) BEATRICE FOODS CO.. ET AL. 781 733 Initial Decision J. Entrenchment 293. A processor of fresh COJ shipping from Florida to other states incurs higher freight costs than a milk plant which buys bulk concentrate and reconstitutes COJ near the grocery store customers. (CX 1005B; CX 1007B; CX 1009A; CX 1650, pp. 169-71; Munkelt 824-29; Goldman 1024-25; Mirapaul 158) 294. Freight cost is relatively greater for glass containers than other COJ containers due to their heavier weight. (Miller 1513-14) The share of the retail COJ market packed in glass has been decreasing. (CX 915B; Munkelt 827; Lang 1102; Miler 1513-14; Hoffer 1952; Jessup 3114; Barnebey 4005) 295. In 1971, Minute Maid started processing and sellng COJ and was sellng COJ nationally by 1979. Minute Maid gradually established COJ processing facilities in Florida, California, and New Jersey and entered into co-packing arrangements with dairies in Boston Massachusetts; Madison, Wisconsin; Indianapolis, Indiana; and Burkburnett, Texas (near Wichita Falls, Texas). (CX 1005A; Munkelt 718- 724; Barnebey (52) 3954, 4091-92; Brick 5313-14) Tropicana recognizes that co-packing arrangements with regional dairies gives Minute Maid a great cost savings. (CX 738b) Each of these seven locations processes bulk FCOJ obtained from Minute Maid' s Auburndale, Florida facility. (Munkelt 721) 296. In 1977, Kraft entered into a co-packing arrangement with Sunkist to serve West Coast customers. (CX 992B; Miler 1484) Under this arrangement, the COJ is produced under the Kraft brand and specifications, and Kraft markets it. (Miler 1484-85) Kraft entered into the co-packing arrangement to lower its distribution costs on the West Coast. (Miler 1510) Kraft's Sealtest dairy in St. Louis began reconstituting COJ about three years ago. (Jessup 3145) 297. Prior to 1976, Hood processed COJ only at its Florida plant. (Donovan 1305) Now, 75% of its COJ for the retail market is processed at Hood's dairy plants in the northeast. (Donovan 1291- , 1294-95) 298. Some dairies far from Florida have reconstituted COJ and sold it under their own label rather than paying the high cost oftransportation for Tropicana. (CX 740B; CX 743A) Tropicana in 1977-78 had diffculty selling COJ in Salt Lake City, Utah, due to the cost of transporting and competition from other processors. (CX 661C-D; 663D-E; Barnebey 3928-29) Tropicana tried to sell COJ to the Beatrice dairy in Salt Lake but the dairy reconstituted COJ itself at a cost Jess than Tropicana s price. (CX 663D; CX 664E; CX 665F; CX 770B; Barnebey 3931) 299. There is an industry tren9 toward increased processing ofCOJ Initial Decision 101 F. from bulk FCOJ at dairies outside Florida. (CX 1650, pp. 27-29; Tilley 2290-91; Jessup 2897-98; Barnebey 3997- , 4006) 300. Prior to its acquisition by Beatrice, Tropicana was interested in expanding its sales ofCOJ throughout the United States, particularly in the middle west and west where it is relatively the weakest. (Barnebey 4040) 301. Tropicana has recently considered shipping bulk FCOJ to a California dairy, and processing and distributing it, under a different label. (CX 811A- 302. Tropicana has been interested in acquiring a processor ofCOJ in California. (CX 528Z-26) Because of increased freight rates Tropicana purchased a plant site in California in 1977 with the idea of reconstituting COJ there. (Barnebey 3972-73) The plant was not built because of high costs involved. (Barnebey 3973) It is likely that in the next ten years Tropicana wil have a plant reconstituting COJ in California. The glass packed fresh COJ wil continue to be (53) shipped from Florida. (CX 1650, p. 169-71) DISCUSSION The following discussion summarizes and supplements the findings of fact and presents conclusions of law.

A. Introduction In the early days oranges were sold by food stores to consumers who took them home, squeezed the fruit and drank the juice. In the 1930' the canning process for orange juice was developed. In the mid 1940' the process of concentrating orange juice was developed, and in the early 1950's chiled orange juice (COJ) was introduced. (Barnebey 3904) COJ is processed almost always entirely from oranges grown in Florida. (findings 100, 114) Oranges from other states are too sour or sweet or have a low juice content. (finding 114) Processors obtain the juice either by squeezing fresh oranges or blending water with frozen concentrated orange juice (called "reconstituting ) or by melting frozen natural strength orange juice. (finding 63) They package the COJ for resale in grocery stores in glass bottles, paper cartons or plastic jugs. (finding 64) They distribute the COJ by delivery to chain grocery stores or the chain s warehouse, to smaller independent grocery stores, and to wholesalers. (finding 77) Tropicana was founded by Anthony T. Rossi who was the chairman of the board and chief executive offcer from its inception through its acquisition by Beatrice. In 28 years the company grew entirely through internal growth to be the largest citrus processor in the world. (CX 39, p. 2) In fiscal 1977, Tropicana had net sales of$244,583,- _.. , 733 Initial Decision 000 and earnings after taxes of $22 461 000. (finding 10) Tropicana the leading processor of COJ with over 30% of the retail market (finding 149), distributes primarily to the warehouse of grocery chain stores and to grocery wholesalers from warehouses across the country which it supplies by rail or trucks. (findings 145, 150, 155) Every five days, a train of Tropicana refrigerated rail cars, over one mile long, leaves the Bradenton plant for the warehouse in New Jersey. (finding 157) Although it sells in every state but Hawaii and Alaska (findings 110, 148), Tropicana is strongest on the east coast and markets closest to its plant in Florida. (finding 147) Transportation costs have limited its effectiveness in markets in the far west. (CX 661D) Tropicana is the price leader in the sale of COJ. (CX 153) (54) Before the merger Beatrice was the 37th largest industrial corporation in the country, with over $6 bilion in sales in fiscal 1978. (finding 2) Beatrice operates the nation s third largest dairy, with 37 milk plants and many other plants producing dairy products. (finding 5; CX 47) It operates more milk plants and has the widest geographic sales area of any dairy in the country, and distributes products in 35 states. (findings 5, 263, 264) In 1975, Beatrice processed COJ at three dairy plants. (finding 227) Realizing that the COJ market was expanding rapidly, and that COJ is a high margin product compared to milk, Beatrice increased its processing of COJ and by the time of the merger it processed COJ at 12 plants (findings 230, 231, 289) and had about .35% of the market measured in gallons ofCOJ sold to grocery stores in the United States. (finding 214)9 Beatrice also distributed COJ processed by other firms including Tropicana. (findings 206, 212; CX 47) In addition just before the merger, Beatrice was completing plans to enter the retail COJ market with a strong, national effort to process, distribute and sell a COJ product which was to be called "Dixie Sunshine." (findings 237- 52) Three COJ processors sell nationally: Tropicana, Minute Maid division of The Coca-Cola Company and Kraft, Inc. H.P. Hood, Inc., a New England dairy, is the fourth largest seller, processing and sellng in the northeastern and southeastern United States. (findings 101 , 113) Tropicana distributes fresh and reconstituted COJ from its plant in Florida to the nation (findings 65, 110), and packages its COJ for the retail market in glass and paper containers. (finding 154) Minute Maid processes COJ by shipping bulk concentrate from Florida to seven processing plants across the country, and by packaging the reconstituted COJ in paper cartons. (finding 111) Kraft packs fresh and reconstituted COJ in glass bottles at its plant in Florida and 9 I accept the Niel en surveys in the record as more accurate evidence of the market than the te t:mony of respondents' paid expert. (CSC reply at pp. 78-4) Initial Decision 101 F. distributes to most of the country. It serves the west coast and Hawaii by a co-packing arrangement with a plant in California. (findings 66 112) Hood packs fresh and reconstituted COJ in paper and plastic at its plant in Florida and its milk plants in New England. (findings , 74, 113) Hundreds of dairies, including Beatrice, also process and sell COJ to the retail market by mixing bulk concentrate with water, using the same equipment and packages they use in packing milk, and delivering to their milk customers at (55) individual grocery stores and to grocery chain warehouses. (findings 134, 253-261) These local and regional sellers of COJ depend primarily on a lower price to obtain shelf space in grocery stores. For example, before the merger, Beatrice processed and sold COJ to grocery stores at a price which allowed consumers to buy it at a per serving price which was 25% to 35% lower than national brands of COJ like Tropicana. (finding 217) In the last ten years COJ sold through grocery stores has been a fast growing industry and now, of the entire orange juice market, COJ accounts for about 28% and frozen concentrated orange juice (FCOJ) accounts for 68%. (finding 57)10 Consumers increasingly are buying COJ as a convenience product. It is ready-to-serve and requires no additional preparation. (Barnebey 4063) Consumers feel they are getting a purer product when they buy COJ, rather than FCOJ. (finding 60) This characteristic of consumers, preferring the quick drink of COJ rather than taking the time to mix FCOJ (which is cheaper lighter andean be easily stored), is somewhat diffcult to understand. It may be due, in large measure, to the advertising and promotion of COJ by the larger processors, such as Tropicana and Minute Maid. (CX 141F)11 The use of heavy advertising to enter new COJ markets is seen in the recent entry of Minute Maid COJ. Minute Maid, a division of The Coca-Cola Company, had 25 000 acres of orange groves in Florida and started sellng COJ in 1960, but, because of a freeze in Florida, withdrew from the market in 1961. (Munkelt 761) In 1970 Minute Maid again started processing and sellng COJ, opening one processing facility at a time, either by buying milk plants or by co-packing arrangements with COJ processors. Minute Maid COJ is now processed from bulk concentrate shipped from its Florida plant to its seven processing plants across the country, where it is reconstituted and packed in paper cartons. Minute Maid entered each new market with a program of high advertising and promotional expenditures. (Munkelt 792) In 1975, Minute Maid had 6.2% of the national retail COJ 10 Caned orange juice accounts for about 4% of the industry. II One result of this advertising and promotion by the larger processors is that, while private label sales are 50% of al) FCOJ sold at retail, they amount to amy 19% of COJ retail sales. (finding 50) 733 Initial Decision market (CX 392, p. 38), and by 1979, it had 19.4% and was sellng nationally. (CX 560Z-18; Munkelt 718) Most of this market share was taken from local and regional COJ (56) processors. (Munkelt 788-9)12 During this time Tropicana was holding its national market share having 28.9% in 1975 and 30.5% in the summer of 1979. (finding 122; CX 560Z-18)13 Tropicana responded to each incursion by Minute Maid into a new market by retaliatory advertising and promotional programs. (Munkelt 805) B. Relevant Product Market To test whether an acquisition may substantially lessen competition, the area of effective competition must be determined by reference to a product market. Brown Shoe Co. v. United States 370 U. 294, 324 (1962). The outer boundaries of a product market are determined by reasonable interchangeabilty of use or the cross-elasticity Id. at 325. of demand between the product itself and substitutes for it. Within the broad market, well-defined submarkets may exist which constitute product markets for antitrust purposes, and the boundaries of such subsmarkets may be determined by looking at: industry or public recognition of the submarket as a separate economic entity, the product' s peculiar characteristics and uses, unique production facilities, distinct customers, distinct prices, sensitivity to price changes, and specialized vendors. (ld. at 325) Not all of these criteria must be met before a relevant submarket is v. Aluminum Co. found. Brown Shoe, 370 U. S. at 326; United States of America 377 U.S. 271, 276-77 (1964).

The evidence here overwhelmingly shows a separate relevant product market of COJ sold to the retail market. The industry and government agencies, the public and the respondents themselves have recogoized that market. (findings 23-39, 41-45, 71-75) COJ has peculiar characteristics which differentiate it from FCOJ and CSSOJ. It is ready to serve whereas FCOJ must be (57) thawed and have water added. (CX 593, pp. 28-29) COJ does not have the quality problems of CSSOJ. (findings 60, 62) COJ has distinct prices (findings 47, 48, 76 99) and customers. (Barnebey 4063; Hoffer 1929) It has unique production, packaging, storage and distribution facilities (findings 51- , 56) and specialized vendors and display areas in retail grocery stores. (findings 36, 55, 56) While the record contains scattered proof relating to effects of the acquisition on the submarket of processing COJ (findings 126, 128 12 Kraft, the other national processr and seller ofCOJ to the retaij market also lost national market share during this period. (findings 122. 123) Kraft depends on gcneraJ advertsing of its dairy products and does not advertse its COJbrand to the consuming public. (Munkelt 805) 13 Tropicana also had over 2% of the market in its sales of private label COJ. (finding 149) Initial Decision 101 F. 130) and on the sub market of distributing COJ (finding 159), most of the evidence is centered on the market of processing, distributing and sellng COJ to retail food stores and that is the relevant market in this case.

C. Geographic Market This merger had its most evident effect in the geographic area where both the acquired and the acquiring companies sold COJ to the same kind of customers. That is where the impact of the merger wil be most clearly felt and most easily measured. In addition to the effects on horizontal competition between Tropicana and Beatrice however, the complaint alleges, in paragraph "fteen, unlawful effects from the merger on the COJ retail market: that competition may be lessened among competitors generally; 14 that Tropicana may be elimithat concentration may benated as an independent competitor;!5 increased and the chance for deconcentration diminished;!6 and that mergers among other processors may be encouraged,1 The market facts below show that these issues must be examined on a national market basis.!8 (58) Tropicana sold COJ to retail grocery stores in 48 states (finding 156) in 242 of the 282 metropolitan areas in the United States. (finding 148) Most of its sales are on the east coast, in the states bordering the Gulf of Mexico, and in large metropolitan areas. (finding 147) Nevertheless, Tropicana ships its COJ to the west coast from its plant in Florida (finding 110), and advertises and tries to sell throughout the United States. (findings 104, 107) These facts alone are suffcient to show that the nation is the relevant geographic market. United States v. Marine Bancorporation, Inc. 418 U. S. 602, 621 (1974); Jim Walter Corp. v. Federal Trade Commission, 5 Trade Reg. Rep. 535 (5th Cir., Sept. 12, 1980), at p. 76 878-79 (625 F.2d 676). Beatrice, with a very small share of the national market, sold COJ in 23 states (finding 264) through 43 dairy plants (findings 160-202), and had plans to sell on an expanding basis, and eventually to sell nationally. (findings 243-5, 247-48) Furthermore, the market is increasingly acquiring a national character. Kennecott Copper Corp. Federal Trade Commission, 467 F.2d 67, 71 (10th Cir. 1972). The three largest brands of COJ, with a growing share of the market (findings 122, 123), sell nationally. (finding 101) 1. Subparagraph (b) and (0 "Subparagraph (c) Subparagraph (d) 11 Subparagraph (e) lB For example, as proof that the merger diminishes the chance for dcconcentration, complaint cOUIl allege that Beatrice s market share understated its future competitive impact because of the phtnned national expansion of Dixie Sunshine. Assuming that the allegation has merit. the geographic area to test that eJlpansion would nC1esrily be rational g.

BEATRICE FOODS CO., ET AL. 787 733 Initial Decision These facts compel a finding that the nation as a whole constitutes the relevant market. Beatrice Foods Co. 86 FTC 1 , 60 (1975), affd 540 2d 303 (7th Cir. 1976). That freight costs are a significant factor (finding 293) giving an advantage to a seller with a plant located closer to the customer, does not foreclose firms from selling nationwide. Id,19 Further, the complaint in this case alleges injury from the merger to other processors of COJ. Hundreds of dairies process COJ in this country (findings 134, 261), and many of them compete against Tropicana in the areas of the country where Beatrice does not now sell. (e. finding 113) The effects ofthe merger should be weighed on a national scale. (59) D. Competitive Effects After determining the relevant market, the next step is to ascertain whether the probable effects of the acquisition may be substantially to lessen competition in the market. Statistics reflecting market shares and concentration are the primary index of this effect. United States v. Philadelphia National Bank, 374 U.S. 321, 326-6 (1963). Where concentration in a market is already great, an acquisition which results in even small increases of market share wil be presumptively unlawful. United States v. General Dynamics 415 U. 486, 497 (1974).

The test for market effect of a horizontal merger was recently stated in Hublein, Inc. 3 Trade Reg. Rep. (CCH) n 21 763 (Docket 8904 Commission order issued October 7, 1980) (96 F. C. 385). In Hublein the merger involved market shares of 17.9% and .79% in the all wine market. The Commission found, on consideration ofthe conventional elements of horizontal merger theory in a rapidly growing industry, that the increase in concentration resulting from the merger was not likely to produce significant anticompetitive effects. The Commission found that Hublein s market share of .79% in the all wine market overstated the competition between it and the acquired company because: (1) the two sold markedly different products at different ends of the price scale of wines; (2) Hublein had no special marketing skils and numerous other firms with a small share of the market were equally important future competitive factors; and (3) a three year increase in four firm concentration went from 47% in 1968 to 56. in 1971 after the merger, and then decreased to 54.2% in 1976. Applying the Hublein test for market effect of horizontal mergers 19 In Federal Trade Commi.. ionv. Proctor Gamble 386 U.S. 568, 571 (1967), a national market was found even though it was not practical to ship the product more than 300 miles from the plant because of high shipping cost and the acquired !inn was the only company having plants throughout the country. 2\ Other market ratios in theHubleincasc were 15.6% and .23% in table wine /lnd 21% and .54% in dessrt wine. The Commssion did not consider these submarkets because these market shares do not substantially differ from the market share of the all wine market.Hublein lit p. 21 939. ), 788 FEDERAL TRADE COMMISSON DECISIONS Initial Decision 101 F. to this case, I find that the first two factors are inconclusive. The Tropicana COJ and the COJ processed and sold by Beatrice are not disparate products to the same extent that the wines in Hublein sold by the acquiring and acquired firms differed in price, quality and sweetness. The main difference between the COJ processed and sold by Beatrice and Tropicana is in price with Beatrice selling at retail for 25%-35% lower than Tropicana. (finding 217) The quality ofthe two products does not vary significantly. (findings 60, 70) Although Tropicana s COJ in glass has a longer shelflife than Beatrice s reconstituted COJ packaged in paper or plastic (finding 61), consumers perceive little difference in (60) Tropicana s COJ packed in paper cartons (which amounts to almost 70% of Tropic ana s sales ofCOJ)21 and Beatrice COJ. (finding 70) The Commission also found that Hublein had no special marketing skills. In this case, while there is no showing that Beatrice has unique innovative or marketing skils, there is proof in the record that Beatrice had a great potential capacity for marketing COJ and has skills obtained from a successful subsidiary processing COJ in Australia. (findings 253-92) The factor which clearly distinguishes this case from Hublein however, is the trend toward concentration in the COJ industry. The top four companies in the national COJ market controlled 55.8% of the gallon sales in 1975. (finding 122) By 1978 the four )eading brands sold in grocery stores had 60.2%, (finding 123) and a few months after the merger, in June-July 1979, they had 64.2%. (CX 560Z-18)22 Further the top two COJ firms had more than 50% of the gallons sold in that period (id. perhaps an even greater indicator in predicting "interdependent anticompetitive behavior. Hublein at p. 21 939. Beatrice s small market share of .35% does not automatically mean that the competitive effect of the merger was de minimis. In Federal Trade Commission v. Pepsico, Inc. 477 F.2d 24 (2d Cir. 1973), a horizontal merger case for preliminary relief, the court of appeals upheld the finding that there was a reasonable probability of competitive injury resulting from an acquisition by Pepsico, with 16.3% of the market, of a company with .3% of the market, in a highly concentrat- , growing (61) industry. United States v. Crowell, Collier MacMillan, Inc. 361 F.supp. 983 (S. NY 1973) does not help respondents. 21 (finding 149) 22 The legality ofa merger under Section 7 is to be teted by whether at the time of the suit there is a reasonable probability that the acquisition win lessen competition.United Statesv. E. l riu Pont rk Nemour. & Co., 353 U. 586 589 (1957); United States v. Penn-Olin Co. 378 U,S. 158, 168 (1964). However, post-acquisition evidence is admssible to confirm trends in the relevant market perceived at the time of the merger.United Statesv. Fa/stoff Brewing Corp- 383 F. Supp. 1020, 1027 (n.R.I. 1974). Adding Tropicana s private label sales of more than 2% of the market (fnding 149), the four firm concentration ratio now approaches being "highly concentrated" by the FTC standard. Federal Trae Commission v. Pepsico Inc. 177 F-2d 24, 27 n.5 (2d Cir. 1973).

733 Initial Decision Crowell, Collier, with .6% ofthe market, acquired the leading firm in that market. The district court refused to find a horizontal violation even though the four leading firms' concentration was 69. 6% and the acquired firm had 41.9%. There was, however, no trend toward concentration nor any other anticompetitive factors. In United States Aluminum Co. of America, 377 U. S. 271 (1964) and Stanley Works Federal Trade Commission 469 F.2d 498 (2nd Cir. 1972), cert. denied 412 U.S. 928 (1973), the facts were closer to this case. In Alcoa the percentage increases were 1.3% plus 27. 8% and in Stanley, 1 % plus 22-24%. Both cases involved the largest firms in their markets just as Tropicana is the largest COJ producer. In Alcoa the top two firms controlled 50% of the market and there was a trend toward concentration. In Stanley Works the top four firm concentration was 50% and there was evidence the acquisition would turn the concentrated market into a rigid market with limited price competition and lead to greater concentration.

Here, the trend toward concentration was primarily caused by the rapid expansion of Minute Maid, which had gone from 7.3% in 1975 to 19.4% in the summer of 1979, and the increased advertising expenditures by Tropicana stirred up by Minute Maid's market invasions. (finding 122; CX 560Z-18; Munkelt 805) This market share has been taken primarily from the local and regional COJ processors (Munkelt 788-9), thus depriving the market of brands which compete by lowering prices (finding 217) and substituting competition by advertising and product differentiation. (Munkelt 792, 805; Lang 1065) While the share of the market by Beatrice prior to the merger was extremely small (finding 214), I feel that the .35% of the retail market substantially understated competition between Beatrice and Tropicana in sellng COJ to grocery stores because of Beatrice s recent growing interest in processing COJ (findings 227-231), and its unusual potential capacity for marketing that product. (findings 253-266, 274 284-6) Furthermore, this small market share must be considered in the context of being added to Tropicana s more than 30%. In a concentrated relevant market even this small increase in market share is presumptively unlawful. United Statesv. General Dynamics 415 U. at 497. Respondents have not successfully rebutted that presumption. Respondents argue that Beatrice and Tropicana did not compete for chain store customers because Beatrice delivers only store-door while Tropicana delivers to the warehouses of chain stores. However Tropicana and other processors who deliver COJ to chain warehouses recognize as competitors dairies delivering COJ store-door. (findings 87-92) Some chains prefer store-door delivery and Tropicana has lost shelf space to dairies (62) delivering store-door. (findings 88, 93) For Initial Decision 101 F. example, in Salt Lake City Beatrice beat Tropicana for sales of COJ to a chain store by this extra service. (finding 220) The record shows that Beatrice and Tropicana regarded each other as competitors in the processing, sale and distribution ofCOJ to retail food stores. (findings 76, 221; Koch 4954) The record shows instances where Beatrice milk plants and Tropicana sold or tried to sell COJ to the same chain food stores (findings 220, 222-25) and tl1ey vied for sales to grocery stores in at least 48 metropolitan areas. (226) Respondents also argue that Beatrice could not deliver to the warehouses of chains because of union contracts with their truck drivers. No such executed contracts were put in evidence, nor was there any explanation why they could not be renegotiated if such contracts do exist. Furthermore, some Beatrice milk plants have serviced chain food stores by delivery to the warehouse or through pick up by the chain at the dairy (finding 276) and other Beatrice dairies were fully capable of doing so (findings 274, 277, 278) just as other dairies have expanded their COJ sales by delivering to chain store warehouses. (findings 272, 273, 292) E. Entry Barriers The strongest argument made by respondents to show that the merger does not tend substantially to lessen competition is that the barriers to entering the processing and distribution ofCOJ to grocery stores are relatively low for dairy companies. (findings 134-35)23 Dairies, the most likely potential entrants into the processing of COJ, typically undersell the national brands of COJ by offering a lower price. (finding 217) There are hundreds of these dairies waiting on the edge of the COJ market to supplement their gross income and to defray their (63) distribution costs by processing and distributing COJ. They just add tap water to FCOJ, and use the same production equipment, distribution facilities and personnel they use to sell milk. (findings 134-35, 228, 253-60) The only additional cost is for bulk concentrate. The dairy merely calls a concentrate supplier in Florida, flushes the milk from the production equipment with boiling water and is quickly into the business of processing and distributing COJ. (Parker 4309, 4313, 4317) Unfortunately, however, local and regional dairy processors ofCOJ are becoming an increasingly less important factor in grocery stores across the country. (findings 122-23) Supplies of bulk FCOJ are not always available to the prospective new entrant into the business of 2J Respondents also argued that chain stores are potential entrant. by private label CW- Such new competition would be in the market fot sellng COJ at ret.jJ but there is no evidence that additional private labels would have a substantial effect on the proessing ofCOJ- Respondents also argue that firms which couccntrate FCOJ in Florida are potential entrants- These firms lack an assured supply of oranges, equipment. sales force and established brand names, and are Iltllikely entrant. into the business of processing COJ. (Bock 1656-4) 733 Initial Decision processing and distributing COJ to food stores (findings 141 , 246) and other barriers are there. (finding 267) The national COJ processors with their powerful advertising budgets, are winning an ever growing market share. (findings 122-23) A small COJ processor testified about this (Lang 1065):

As a regional packer, again we are competing with nationally-known, well advertised well-promoted competitive products who by and large almost automatically are represented in every store and for a local brand, such as ourselves, we are fighting for what is left of that space in the dairy case. It' s a very diffcult job to compete with these nationally known brands for shelf space in view of the fact that our company cannot afford to advertise anywhere near as significantly as our competitors do. Absence of high entry barriers, moreover, cannot be depended upon to ensure effectively competitive conditions. Ekco Products Co., 65 C. 1163 , 1208 (1964), affd 347 F.2d 745 (7th Cir. 1965): A merger may violate Section 7 even though there do not appear to be formidable barriers to entry into the market affected by the acquisition; the existence of potential competition does not justify or excuse elimination of actual competition. In such a case where the merger s effects on competition are those proscribed by Section 7, its illegality cannot be overcome by a showing of ease of entry. The reasons that low entry barriers are not a defense to a Section 7 violation are because (1) entry by significant new (64) competitors is likely to be at best a long term affair and (2) even an entry-discouraging low price is not likely to be as low as if there were actual competition. Id. at 1208.

F. Conglomerate Theories Complaint counsel argue that the complaint alleges not only unlawful effects from the merger on actual competition but also unlawful future effects in that the "possibilties for eventual deconcentration may be diminished." They argue that the merger will entrench Tropicana as the leading processor of COJ sold to food stores and already has eliminated Beatrice s planned role as a deconcentrator through the introduction of the Dixie Sunshine brand of COJ. Respondents argue that complaint counsel, in pointing to the future eflects of the merger, were trying to prove a violation based upon theories which were not pled in the complaint. They argue that actual and potential competition are not interchangeable concepts. In addition to the traditional horizontal theory of the complaint the Commission in Hublein also decided the case under conglomerate theories, holding that Hublein was not a unique potential deconcentrator of the market and that the merger did not entrench the acquired company s market position. The Commission looked at these Initial Decision 101 F. conglomerate theories both as an independent basis of violation and also as an aggravating factor in the alleged horizontal violation. Hublein at p. 21 942.

Section 7's incipiency standard requires not merely an appraisal of the immediate impact of the merger upon competition, but also a prediction of its impact upon competitive conditions in the future. 362 United States v. Philadelphia National Bank 374 U.S. 321, (1963); Brown Shoe Co. v. United States 370 U.S. 294, 32:r n.38 (1962). Therefore, even in a horizontal merger case, the future effects of the merger must be weighed. Hublein, supra; Stanley Works v. Federal 8, 505 (1972), cert. denied, 412 Trade Commission 469 F.2d 498, 502 n. S. 928 (1973).

G. Dixie Sunshine In the middle 1970's Beatrice grew interested in the processing of COJ, since that market was expanding quickly and was a complement to the processing and distribution of milk. (findings 228-30, 253-0 288-90) From 1975 to 1978 Beatrice expanded its COJ processing from three to twelve dairy plants. (findings 227, 231) (65) In early 1977, the highest executives in the Beatrice dairy division became interested in expanding the COJ processing business more rapidly. (finding 236, 237) This plan included a new label COJ to be sold by Beatrice-Dixie Sunshine.

The test market stage for Dixie Sunshine included distributing COJ processed and packed in glass bottles under that label by Tropicana, and packaging fresh orange juice supplied by Admore Farms in paper cartons at the Beatrice milk plant in Orange City, Florida. (finding 241 242) The Beatrice dairy division executives soon started thinking about a national market for Dixie Sunshine COJ, with COJ to be reconstituted by Beatrice at 30 to 35 milk plants with national distribution in 18 months. (findings 243, 245, 248) Just as Dixie Sunshine was coming on the market, Beatrice bought Tropicana. The merger was approved by the Beatrice board of directors on March 4, 1978. (finding 19) At a meeting on March 8, 1978 Beatrice dairy division executives stopped the Dixie Sunshine program. (finding 252) Beatrice had a unique ability to enter the business of processing and distributing COJ on a national level. Dairies process and package COJ on the same machinery used for processing milk. (findings 254, 262) Milk plants use the same storage facilities and distribution system for milk and COJ and generally find the products compatible. (findings 255, 256, 258-0) Beatrice has more milk plants and sells in a wider geographic area than any other dairy in the United States. The likelihood of Beatrice expanding the business of processing, 733 Initial Decision distributing and sellng COJ was eliminated by the merger, thus reducing the chance for deconcentration in the market. H. Entrenchment A large acquiring firm can confer on an acquired firm competitive advantages over competitors in the acquired firm s market and those advantages may substantially reshape the competitive structure of the industry by raising entry barriers and by dissuading smaller firms from aggressively competing. Hublein at p. 21 948. Here, complaint counsel argue that the acquisition conferred on Tropicana advantages in advertising and distribution which will entrench it as the market leader. There is no substantial evidence in the record that Beatrice brought to Tropicana a significant competitive advantage in advertising. Although Beatrice is a large advertiser (finding (66) 280), there was no proof of advertising effciencies not available to other firms in the retail COJ market. Hublein, at p. 949-50.

I find, however, that the acquisition did confer on Tropicana advantages in distribution which wil lead to greater expansion of Tropicas sales of COJ to the retail market in areas of the country where shipping costs have prevented it from domination, and wil undoubtedly increase the already high concentration in the market. Tropicana has been the leader in the retail COJ market for many years and now has over 30% of an increasingly concentrated market. (findings 122, 123; CX 560Z-18) Tropicana is the price leader in sellng COJ and in making pricing decisions Tropicana considers only its own costs rather than prices charged by its competitors. (finding 153)24 Tropicana s only obvious marketing disability before the merger involved diffculty in sellng to markets far from Florida because of increased freight rates incurred in shipping COJ to those markets. (finding 293) This deficit should be alleviated by the merger. Before the merger, Tropicana was interested in expanding its sales of COJ to food stores in California and other areas in the west and midwest. (finding 300) Tropicana could have done this by shipping bulk COJ to dairies in those areas, which would process it and distribute it under a different label. (finding 301) It was likely, but for the merger, that Tropicana would have a plant reconstituting COJ in :u This aversion to price competition in sellng COJ to grocery stores is coupled with an aversion to price competition by grocery stores in selling COJ. In 1978, Mr. Barnebey, the president of Tropic ana, wrote to a dairy customer who had complaiaed about a chain store selling Tropicana COJ as a 10S8 leader. Decrying this price competition, Mr. Barnebey said that "Because Tropicana represents such a largf! portion of the total chilled juice sales, we have tred to maintain the most orderly market possible." Mr. Barnebey related in the letter how TIopicana. in order to avoid such retail price competition, had resisted direct sales to gro ry warehouses for several years aftr Kraft had startd the practice.(CX 802A) Initial -Decision 101 F.T. California in the next ten years. (finding 302)25 The merger solved this problem for Tropicana. Beatrice (67) already has milk plants in San Jose, California (finding 191) and throughout the west and midwest. (RX 1750, RX 1752) Tropicana now has the ability to lower its costs in California, as well as the other markets now serviced by Beatrice dairy plants. The merger has entrenched Tropicana as the market leader in the processing, distribution and sale of COJ to grocery stores. CONCLUSION Beatrice violated Section 7 of the Clayton Act and Section 5 of the Federal Trade Commission Act 26 by acquiring Tropicana, because the effect of the acquisition may be to substantially lessen competition in the processing, distribution and sale ofCOJ to retail food stores in the United States.

RELIEF Only complete divestiture can return Tropicana to a position as an inde.pendent competitor in this concentrated market. United States EJ :In Pont de Nemours Co. 366 U.S. 316, 326-31 (1961) In addition, "- 10 year ban on future acquisitions is now customary when Sedi,,", 7 is violated. Kaiser Aluminum Chemical Corp., (1976-79 Transfer Binder) Trade Reg. Rep. TI 21 578 at p. 21 697 (appeal pending).

Complaint counsel argue that Beatrice should be required to divest Tropicana through a public stock offering or by a pro rata distribution of stock to Beatrice shareholders, and not by a sale to another corporation. The basis of their proposal is that, because of the size of Tropic ana, only a large corporation could buy it and the "deep pocket" thus afforded would further entrench Tropicana as the market leader in the COJ market. I have found, however, that Tropicana has been entrenched by the merger because of the enhancement of its processing and marketing of reconstituted COJ through the Beatrice milk plants, not because of a deep pocket theory. While it may be desirable to have an independent market leader (68) which is not affliated with any conglomerate corporation, an equally forcible argument can be made that ownership by a large, well-financed non-dairy company may be a more realistic way to have Tropicana battle competitors such as The Coca-Cola Company, Kraft, and, hopefully, Beatrice. Ligget Meyers, Inc. 87 F. C. 1074, 1141 aff'd, Ligget Meyers, Inc. v. Federal Trade Commission 567 F.2d 1273 (4th Cir. 1977) :! Much of Tropicana s COJ is already processd by reconstituting bulk concentrate. (finrlng 65) :! No separate proof or argument was made under the Section 5 count but any vio!atioo of Section 7 is a violation of Section 5.Federal Trade Commision v. Brown Shoe Coo, 384 U.s. 316, 321-22 (1966) BEATRICE F()QDS CO . ET AL. 795 733 Initial Decision I also believe that Beatrice should divest the profits made by Tropicana during the ilegal tenure. The notice order attached to the complaint in this case, in paragraph 2, stated that, should the Commission conclude that there had been a violation oflaw, it might order . . . divestiture by Beatrice of Tropicana and of any profits derived therefrom so as to create a viable, independent entity engaged in the processing, distribution and sale of ready to serve orange juice. While that language appears to include divestiture of profits as part of divestiture of Tropic ana, it certainly put the respondents on notice that the profits made by Tropicana during Beatrice s ownership of the company might be divested. Divesting the profits of Tropicana would merely raise the price to be received by Beatrice from the sale of Tropicana,27 and I believe they should be divested separately, to the United States Treasury. The purpose of this divestiture of profits is not only to avoid unjust enrichment but "to prevent ilegal practices in the future Federal Trade Commission v. Ruberoid Co. 343 U. 470 473 (1952), by avoiding a continuing violation of Section 7 by the possible use of the profits obtained through the illegal acquisition to further restrain trade even after Tropicana is divested. The Commission has the authority to compel such affrmative acts of compliance. Ecko Products Co., supra 65 F. C. at pp. 1212-16. (69) ORDER It is ordered That Respondent Beatrice Foods Co. (hereinafter Beatrice ) shall:

A. Within one (1) year from the date on which this order becomes final, divest all its interest in the assets and business of Tropicana Products, Inc. (hereinafter "Tropicana ) which Beatrice acquired in August 1978, together with all additions and improvements. B. Within three (3) months from the date on which this Order becomes final, submit to the Federal Trade Commission for its approv- , the plan of divestiture referred to in paragraph A ofthis part, and make such changes in the plan and submit such additional information with respect to it as the Commission may require. The plan shall be desigoed to reconstitute Tropicana with all its pre-acquisition assets and business (and any new business in which it has engaged subsequent to the new acquisition) as a viable, independent, corporate and competitive entity, and shall include divestiture to the United 27 Ligget Meyers, Inc., supra at p. 1141. 2. This provision is not meant to punish Beatrice.Cf United States PaperCToft Corp. 393 F,Supp- 415, 425-26 (W. Pa. 1975) Initial Decision 101 F. States Treasury of all of the profits made by Tropicana while that company was owned by Beatrice.

II.

It is further ordered That following the divestiture of Tropicana required by Part I ofthis Order, no employee, offcer (70) or director of Beatrice shall at the same time be an employee, offcer or director of Tropicana.

II.

It is further ordered, That pending divestiture as required by Part I of this Order, Beatrice shall cause Tropicana s business to be conducted in the normal manner; no plans for expansion or improvement of its plants or business shall be halted or interrupted; and Beatrice shall not permit any changes in Tropicana s business, corporate or financial structure, or otherwise, which would impair Tropicana volume of business, profitabilty or ability to survive following divestiture.

IV.

It is further ordered, That for a period often (10) years from the date this Order becomes final, Beatrice shall not acquire, or acquire and thereafter hold, directly or indirectly, through subsidiaries or otherwise, without the prior approval of the Federal Trade Commission the whole or any part of the stock, share capital, assets, any interest in or any interest of. any business, corporate or noncorporate, engaged in processing, distributing, and selling ready to serve orange juice. (71) It is further ordered, That on the first anniversary date of the effective date of this Order and on each anniversary date thereafter until the expiration of the prohibitions in Paragraph IV of this Order Beatrice shall submit a report in writing to the Federal Trade Commission listing all acquisitions, mergers and agreements to acquire or merge made by Beatrice; the date of each such acquisition, merger or agreement; the products involved and such additional information as may from time to time be required.

733 Opinion VI.

It is further ordered That for ten (10) years following the divestiture required by Part I of this Order, Tropicana shall not agree to merge with or be acquired by any individual, partnership or corporation without the prior approval of the Federal Trade Commission; and neither the offcers nor the directors of Tropicana shall recommend the merger or acquisition of Tropicana to Tropicana s shareholders without such prior approval.

OPINION OF THE COMMISSION By PERTSCHUK Commissioner:

The question before the Commission is whether the 1978 acquisition by Beatrice Foods Co. (Beatrice) of Tropicana Products, Inc. (Tropicana) violated Section 7 of the Clayton Act, 15 VB. C. , and Section 5 of the Federal Trade Commission Act, 15 C. 45. The complaint, issued on June 20, 1978 following denial of an application for a preliminary injunction against the proposed merger FTC v. Beatrice Foods Co. and Tropicana Products, Inc. 587 F.2d 1225 (1978), alleged that the effects of the merger may be to substantially lessen competition or tend to create a monopoly in the processing, distribution and sale of ready-to-serve orange juice in the United States. The complaint alleged that the acquisition eliminated actual competition between Beatrice and Tropicana and between competitors generally, and that it might foster other mergers (2) between competitors, causing a further loss of competition in the processing, distribution and sale of ready-to-serve orange juice. The complaint further charged that the previously existing level of concentration in the relevant market wil be increased, and the possibilities for eventual deconcentration decreased, as a result ofthe acquisition. The administrative law judge (ALJ) found the acquisition unlawful based on a horizontal theory of violation. In addition, stating that future effects must also be assessed in a horizontal merger case, he found the acquisition unlawful based on the "conglomerate" theories that it eliminated significant potential competition by Beatrice in the relevant market and entrenched Tropicana as the market leader. The ALJ ordered Beatrice to divest itself completely both of Tropicana and of the profits made by Tropicana during the time it was owned by Beatrice.

We disagree with the ALJ' s conclusion that the acquisition violates the antitrust laws and order that the complaint be dismissed. We find the loss of actual competition resulting from the merger to be too little Opiniun fin FTC. to establish a violation of Section 7. We find the loss of potential competition not to have been pled in the complaint, and therefore we are constrained from reaching the merits of that allegation. Were we able to decide that question, however, we would be unlikely to find a substantial loss of potential competition resulting from the acquisition. Finally, we find that the proof of entrenchment of Tropicana is insuffcient to warrant dissolution of the merger on that theory of violation. (3) I. THE MERGING FIRMS AND THE INDUSTRY Beatrice is the nation s leading diversified food manufacturer, producing and sellng to consumers more than 8 000 products, LD.F. 2 including over 4 000 dairy and grocery products. LD.F. 3,1 In fiscal 1978, Beatrice had total net sales of $6 313 880 000, making it the 31st largest industrial corporation in America in net sales. LD.F. 2. In the same year, Beatrice was the nation s third largest dairy and the largest milk producer and distributor, with 37 milk plants serving 35 states. LD.F. 5; LD. 54. In conjunction with its milk business, Beatrice sold ready-to-serve chiled orange juice (COJ) to grocery stores and institutional customers, such as hospitals and restaurants. (4) LD. Using its dairy plants, Beatrice reconstituted COJ from bulk concentrate, LD.F. 158, and also distributed COJ processed by other orange juice producers. I. F. 159. In comparison to many other of its product lines, Beatrice s sales of its COJ products were a small part of its business prior to the merger, generating only $2.4 milion in fiscal 1978 on the total sale of 1.2 milion gallons of COJ. I. F. 210- 12. Although its actual ranking was unknown, Beatrice had a tiny fraction of the relevant market, but had been increasing its production and sale of ready-to-serve orange juice in the years just before the merger. LD. 54.

Tropicana, the acquired firm, is a major processor and distributor of orange juice products, and the leading processor of COJ in the 1 The fonowing abbreviations wil be used:

tD.F. - Initial Decision Finding of Fact N0. Tr.1.D -- Tran&riptInitial Decisionof TestimonyPage No Page No. Complaint Connscl's Exhibit No ex - CPFRX - Respondent's- Complaint Counsel'sExhibit ProposedNo. Finding of Fact No. CRRPF - -Respondent'sComplaint Counsel'sProposedReplyFindingto Respondent'of Fact Nos Proposed Findings of Fact RR - Respondent's Reply to Complaint Counsel's Proposed I-'indings of Io'act CAB - Complaint Counsel's Appeal Brief Page No TAB - Tropicana s Appea! Brief Page No.

BAB - Beatrice s Appea! Brief Page No.

TRB - Tropicana s Reply Appeal Brief Page No. BRB - Beatrice s Reply Appeal Brief Page No. 2 For precioo JefiI1itioJ's of chiled orange juice and other technical terms that wil he u d herein, su.ch as bulk concentnlte, proce ed and reconstituted COJ, and frozen concentrate orange juice (ICOJ),see LO- at 5-. 733 Opinion United States. LD.F. 9. In 1977, Tropicana had total net sales for all its products of $244 583 000. LD.F. 10. In fiscal 1978 it sold 106 821,- 648 gallons of COJ for $206 649 295. LD. F. 11. Approximately 90 percent of the oranges used in producing various orange juice products in the United States comes from Florida, the rest coming from the Southwest, Mexico, and Brazil. LD.F. 127; CX 1650, pp. 12-13. Processors like Tropicana either turn the oranges directly into COJ, frozen concentrate orange juice ("FCOJ"), bulk concentrate (frozen juice in bulk form, which is the (5) raw material for reconstituted COJ), or canned single strength orange juice CSSOJ"). LD. F. 22.

The chain of production and sale in this industry includes the citrus growers who supply the oranges; the processors, such as Tropicana, who own or buy the oranges directly from the growers and process them for use in various forms;' firms (usually dairies), such as Beatrice, who purchase "bulk concentrate" and toreconstitute" it. into orange juice by adding water to it, much as a person adds water to FCOJ; wholesale distributors, who include the (1) processors who sell their product directly to the retail or institutional customer, (2) the dairies, who in addition to reconstituting orange juice from bulk concentrate, resell juice already processed into COJ, which is supplied to them by the processors, and (3) others such as independent warehouses and food brokers; and the end-line industry customers, the grocery stores and institutions, who sell the orange juice to consumers. Processors, like Tropicana, Minute Maid, and Kraft, usually distribute their product directly to central retail warehouses and institutions, and less frequently through independent distributors such as the dairies. Dairies which sell orange juice, such as Beatrice typically distribute their product directly to the grocery or institution. (6) There is incomplete vertical integration in the industry, in that processors own little citrus acreage themselves' and do not own any retail outlets. However, processors like Tropicana, Minute Maid and Kraft, who have direct access to oranges and the facilities for turning them into orange juice products, are more vertically integrated than dairy reconstituters ofCOJ, who are dependent on producers of bulk concentrate for supply.

In sales of COJ to the retail segment of the market, shelf space is the prize, and there is intensive competition for it. To win shelfspace, suppliers of orange juice must convince store owners that their brands 3 For the most par, they obtain supply from the grwers. 4 In 1977, Trupkana owned 690 acres of citnm grves, which it usd primarily for experimental purposes CX527G; it purchased nearly all of its s1.apply of oranges from independent growers. CX527F. 5 Grocery i3torcsl.ually cary about five brands, including major brands such as Tropicall or Minute Maidand a few regional or private label products I.D-F. 140. Opinion 101 F.

will sell. To do that, they must offer a competitive price or cultivate consumer loyalty to their brands, relying heavily on consumer advertising and promotion. I.D.F. 136-139. Tropicana and the other leading firms engage in substantial brand advertising to consumers. I.D.F. 12 136-37; LD. 55. Beatrice was not a strong advertiser in the COJ market prior to the merger, but it is a very large advertiser generally, ranking 17th in the nation in 1978 among all industrial corporations. 4. I..F.

Beatrice acquired Tropicana at a time of significant and sustained growth in the orange juice market. Historically, frozen orange juice (FCOJ) has been, and remains, the leading orange juice (7) product but COJ has made rapid gains in the last several years. From 1975 to 1979, FCOJ dropped from 77 percent in 1975 to 68%, while COJ climbed from 18 to 28 percent at an annual growth rate of20%, twice that of FCOJ. LD.F. 57. Although COJ costs more than FCOJ, it is easier to use, apparently a major reason for its growing popularity with consumers. LD. 55. The recent growth of the COJ segment of the market has attracted, and been assisted by, new entry, most notably that of Minute Maid. Id. Prior to Minute Maid's entry, Tropicana and Kraft had been the only two COJ firms who sold COJ throughout most of the country. Minute Maid has provided significant competition to these two leading firms, as well as to smaller local and regional rivals. Stil, despite this growth and new entry, the ALJ found the level of concentration and loss of competition from the merger substantial enough to warrant a finding that the merger violated Section 7. wil address the question of concentration and the other relevant antitrust considerations of the merger in the discussion that follows. II, THE RELEVANT PRODUCT MARKET The market definition questions in this case are complicated. Concerning the product market, the complaint alleges that a relevant line of commerce in which to assess the legality of the merger is "the processing, distribution and sale of ready-to-serve orange juice," Paragraph 10. The ALJ held that the evidence "overwhelmingly shows a separate line of commerce for COJ (8) sold to the retail market, I.D. , justifying the exclusion of orange juice sales to institutions from consideration in assessing the competitive impact of the merger. LD.F. 29--0. We agree with the ALJ' s determination to exclude institutional sales, and note that respondents have not seriously challenged it on appeaJ.

(, Minute Maid entered the Co.l hU5ipes: from its position as the Icadipg processor of FCOJ. It owns a bulk concentrate processing plant in Florida, which supplies seven regional reconstituting plants located throughout the country.

7 Respondent Tropicana did cha!lenge the factual basis ofIDF 39 concerning a lack of cros.'I-clastlcity of demand between the retail and i!1stit.utional market. , TAB 32, hut it did not question the fundament,,! soundness theof (footnotecollt' ), , .

LI.l.nJ.J.H -' J. ..OJ .....

733 Opinion A. The Relev-;nt Produr;ts The ALJ found, as the complaint alleged, that COJ sold to retail is a separate relevant product market, distinguishable from the sale of FCOJ and CSSOJ. The ALJ made his determination on the basis of the criteria for product market definition enunciated in Brown Shoe Co. v. United States 370 U.S. 294, 325 (1962). Specifically, he cited evidence that COJ is recognized as a separate market by industry and consumers, that it is ready to serve whereas FCOJ must be prepared that it is far superior in quality to CSSOJ, and that it has qistinct prices, customers, and facilities for production, distribution, and instore display. LD. 56-57. He stressed that it is not essential to satisfy all of the Brown Shoe criteria in order to establish a relevant submarketo Brown Shoe 370 U.s. at 326 United States v. Aluminum Co. of America ("Alcoa-Rome 377 U.S. 271, 276-277 (1964). On appeal respondents do not question the ALJ' s findings on industry and public recognition of a separate COJ market or on the peculiar usage characteristics and quality of COJ. They argue, (9) however that there is suffcient price interaction and cross-elasticity of demand between COJ and FCOJ to warrant a finding that they belong in the same relevant product market. Respondents are correct to identify cross-elasticity of demand as the most important factor product market definition. See FTC Statement Concerning Horizontal Mergers at 12 and the revised Department of Justice Merger Guidelines at Ila), both issued on June 14, 1982. However, respondents contentions about price elasticity are not borne out by the evidence. Industry representatives, including a Tropicana offcial, testified that promotional sales ofFCOJ do not perceptibly influence the sale of COJ in the retail market. LD.F. 58. Further industry testimony indicated that sellers of COJ to retail customers pay no attention to the prices of FCOJ and CCSOJ in setting their own COJ prices. LD.F. 47. In addition, a national statistical study of price and consumer purchasing patterns for orange juice products between 1972 and 1979 concluded that there are no statistically significant substitutes for COJ, including FCOJ and CCSOJ, and that consumer demand for COJ is relatively price inelastic.9 Dr. Tiley attributed his findings of price (10) inelasticity for COJ to the peculiar characteristic of COJ buyers to get into the "habit" of drinking COJ; unlike FCOJ, he observed that many consumers wil develop a stable attachment to COJ and contin- ALJ' s finding of a separate market ofCOJ saes to the retail segment. Respondent Beatrice s appeal briefs did Dot address thiaissue at all.

Crosslasticity of supply represents another important factor, as the FTC Statement and the DOJ Guidelines also recognze. Respondents alleged below that there was cross-(laaticity of supply between COJ and FCOJ which also justified inclusion of both in one market. However, respondents did not make this argument on appeal 9CX 514 B, R; Tr. 2236-7, 2243, 2397. The study was done by Dr. Daniel Tiley, then a research economist with the Florida Deparment of Citrus and an adjunct professor at the University of Florida. Opinion 101 F.

ue to buy it regardless of price fluctuations in COJ and the availability of interchangeable products such as FCOJ. Tr. 224c1 2243.1 Under cross-examination, Dr. Tiley conceded there is some price sensitivity between COJ and FCOJ, Tr. 2383- , but read in the context of the study itself and his entire testimony, this statement does not undercut his basic thesis that the cross-elasticity of demand between the two products is 10w.11 Beyond reference to this thread of testimony, TAB 33-34, respondents have not attacked the validity of Dr. Tiley s statistical findings, nor provided any other reason why we should not give them substantial weight.

Respondents complain that the Initial Decision relies on the Tiley study while completely overlooking two studies of the New York City market which, they claim, provide significant evidence of substantial switching or substitution ofCOJ and FCOJ. RX 1, 1P; TAB 33. We believe, however, that these studies of a single local market, an area which is nowhere suggested to be representative of (11) the COJ market as a whole, are substantially less probative on the question of cross-elasticity of demand than the national Tiley study and industry testimony on COJ pricing practices already cited.1 Respondents argue that if the COJ market is going to include COJ sold in both cartons and glass containers, as the ALJ says it should, LD.F. 71-76, then it logically must include FCOJ too, since FCOJ and carton COJ are closer together in price than are carton and glass COJ. TAB 24-25; CX 560T (in camera). Alternatively they argue that if FCOJ is excluded, then a separate submarket must be found for COJ sold in glass, because of its higher price. TAB 25.1 These arguments place too much weight on absolute price differences as opposed to price elasticities that, under the FTC Merger Statement and DOJ Merger Guidelines, should bear most decisively on the inclusion or exclusion of various types or packages of orange juice products in the relevant product market. Despite a price spread between COJ and FCOJ that is narrower than that between glass and carton COJ, evidence of other distinguishing factors, particularly a relatively low (12) cross-elasticity of demand, warrants a market division between these two products. Further, despite the greater price disparity be- 10 He admitted, however, that his study did not consider the pOl'ibility that a reguar buyer of COJ m.ight occasionally buy FCOJ when it is on le, and then resrne the habit ofhuying COJ. Tr. 2391. Even so, alowing for this possibility does not neces. arily refute Dr- Tiley s general findings of a "purchasehabit rBsponae" and substantial price insensitivit.y on the part of COJ buyers 11 Moreover, due to the nature of the data base usd, the study may have tended to overstate the degree or any price sensitivity and crosslasticity of demand. CPr' 127 , citing Tr. 2094-. 12 Obviously FCOJ aod, perhaps, even CCSOJ, are to some extent interchangeable with CQ.J in the sense that their end use is the same and their price ranges, while substantially below those of COJ, are nut dramatically so. It would be surprising if there were no interchangeability at all among such closely-related types of products as these. But it is riot at all iIlCoilsistent to say that products can be 30mewhat interchangeable but, in light of prevailing commercial attitudes and realities, that their division into separate markets whi provide a more useful staing point for antitrust analysis.See, e. Sullvan Handbook of the Law of Antitrust 607-08 (1977). U Such a finding would exclude Beatrice from the relevant product market, since it sold litte ifany COJ in gla9S. 733 Opinion tween glass and carton containers of COJ, and other alleged differences in quality and handling between these forms of packaging, RPF 22- , the record indicates both that the industry recognizes them as being in the same market, and that there is a close competitive relationship between them, in which COJ producers typically consider the price of all types of containers in pricing their own products. LD. 71- 75; Tr. 711, 759-60, 963, 970-71, 1480-1, 1492. The existence of such price sensitivity is persuasive evidence that carton and glass COJ belong in the same product market.

Respondent Tropicana relies on Alcoa-Rome and A.G. Spaulding Bros. v. FTC 301 F.2d 585 (3d Cir. 1962) in contending that price disparity between glass and carton COJ is a suffcient basis for placing them in separate submarkets here. In A lcoa-Rome however, the prices of insulated aluminum and copper conductors did not respond to each another, and the price difference of 50 to 65 percentl1 was the single most important factor " overriding other clear aspects of competitive overlap which, ordinarily, would have been enough to place the products in the same market. 377 U.s. at 275-276. The evidence here, which shows price and promotional interaction between sales of COJ in glass and carton (13) containers, clearly distinguishes this case from the facts in Alcoa-Rome,15 A.G. Spaulding Bros. also is unpersuasive authority for respondent Tropicana s position. There the court grouped several different types of athletic equipment into a single market. The very same factors that justify a separate COJ market in this case industry recognition and a regular competitive relationship between products justified the finding of an all-inclusive industry market in that case. The court did find distinct product classes within that broad market divided into high and low-price categories of athletic goods, but it did so on the basis of evidence showing that there were high and low-price groupings of customers. In this case there is little evidence that price stands out as the determinative factor in the selection of either a certain type of orange juice or type of orange juice container. In short, applying principles of cross-elasticity of demand recognized by the Commission and the Justice Department as a central consideration in product market definition, we find no internal contradiction between the inclusion of glass-packed COJ in (14) the rele- 14 This is substantially greater than the 20 to 50% difference between glass and carton COJ alleged here. TAB 34- Complai1't coun 1 have argued that this alleg-ation overstates the seluaJ price difference. CR 26-27. 15 On the average, carton-packed COJ undoubtedly does have II price advantage over glasspacked COJ and appears tv be the trend in the industry, largely for that reason. Nevertheless, sellers ofCOJ in glass, most notably Tropicana and to II lessr extent Kraft, have competed effedivcJy in this package form on the basis ofa premium image, superior qualiy and shelf life, promotional advantages, and brand identification. Tropicana and other sellers orcOJ in glass are thus far better oITthan thc fabricators ofinsuJated copper conductorin Alco-Rume who were competitively crippled by their price dis8advantage. 377 U.S. at 276 . .

Opinion 101 F.

vant product market and the exclusion of FCOJ from that market. Of course, any attempt to define a single relevant market wil involve some measure of arbitrary line-drawing. Market definition is only the start of the analysis, and the definition of a "relevant market" should not preclude consideration of any competitive pressures provided by products outside that definition. See the FTC Merger Statement at 5-, and the DOJ Merger Guidelines at TIII(C)(l)(b). Nor if a broad market is selected, are we precluded from recognizing that the products within that broad market may be something less than perfect substitutes. In this case, a fair reading of the evidence is that competing producers ofFCOJ may well have some impact on the price and production of COJ, though clearly not as much impact as do the actual producers of COJ.

As long as these effects are properly recognized, the question of market definition takes on less importance. We recognize that there are arguments for taking these effects into account by defining a broad product market and using the resulting smaller market share figures (as Commissioner Douglas urges in his concurring opinion). We are more inclined to take them into account by using the narrow- , COJ market figures, but considering the presence ofFCOJ producers in evaluating the significance of those (15) shares. However, given the result we reach it is unnecessary to resolve this issue here. As we discuss below, even using the narrower market definition we do not find this acquisition likely to be anticompetitive. B. The Relevant Level of Distribution As previously noted, the complaint alleges that a relevant line of commerce in this matter is the processing, distribution and sale of COJ. The ALJ found that the relevant market is the processing, distribution and sale of COJ to retail food stores.1 He essentially defined this market to include processed, reconstituted, and relabeled COJ distributed directly or indirectly to the retail level by the owner ofthe COJ brand label.B LD.F. 210-218. This definition groups together processors, reconstituters, and private relabelers as horizontal competitors to the extent they all sell their own brands ofCOJ to the retail level.9 It derives from the nature ofthe evidence in the record, which 16 The ALJ found that respondents' contention below that fresh and reconstituted COJ were separate Buhmarkets was also unconvincing. J.D..!. 63-70. We agree with these fimiings, and note that respondents have given short shrift to this issue on appeal J7 This finding is premised, of course, on his prior finding that sales of COJ to the retail level are a segregable market.

16 As noted, Tropicana, like other processors, seUs most of its own brand of processed COJ directly to retailers, Bnd the rest to retail through distributors. Beatrice sells iL" own braorl ofreconstittlted and relabeled COJ to retail in addition to distributing pre-processed CO,T under the processor s label. 10-1". 159. '9 ProCf!!'Ors, as previously discussed, make COJ and/or bulk concentrate straight from the oranges. Rcconstituters, such as dairies, use their milk production plants to make their own COJ from bulk concentrate bought from a processor. Private relabelers, often dairies as well, buy ready-to-serve COJ from a processor and reseU it to retail 733 Opinion centers on brand sales to retail, (0. 57, imdfrom (16) evidence. that competition for shelf space and sales to consumers are primarily based on brand promotion. If this market definition is accepted, it would necessarily include all direct and indirect COJ sales to retail by Tropicana under the Tropicana label, and all COJ sales to retail by Beatrice under Beatrice-owned labels 2o whether or not the Beatrice product was bulk concentrate reconstituted into COJ by Beatrice or pre-processed COJ bought from a processor and resold by Beatrice under its label.21 This (17) definition seems basically to comport with complaint counsel's view of the relevant line of commerce. Respondent Tropicana contends that this line of commerce is overly broad, arguing that reconstituters and relabelers like Beatrice do not belong in a market with processors because they are functionally different from processors and operate at an entirely different level namely, distribution. TAB 12-17. It asserts they are not true competitors because reconstituters and relabelers lack the same control over supply and play no part, as do processors, in the conversion of oranges into COJ and bulk concentrate. TAB 16. It further argues that no competitively meaningful distinction can be made among dairy sales ofreconstituted COJ, relabeled COJ, and COJ distributed by dairies under a processor s label. For these reasons, respondent Tropicana concludes that dairies are not in competition with processors because they act essentially as distributors, even when they reconstitute or relabel the product. TAB 16.

We find this argument unconvincing. In contending that processors and dairies which reconstitute or relabel their own brand of COJ are not horizontal competitors, it is too little to say that processors enjoy an advantage over dairies in their access to orange supply. This would prove only the existence of unequal degrees of access or vertical integration, not the non-existence of competition between such entities at a mid-level between processing (18) and pure distribution. Conse- 2C The main Beatrce label was Mcadow GoJd; othera were Sanitary Dairy and Dixie Sunsine. LD.F. 158. l The Initial Decision is somewhat ambiguous on this definition, however. Whle it contains findings on the s.es volume and market share of al COJ wid by Beatrce to reWJ under a Beatrice label, whether reconstitute or relabeled; I. F. 212-213, and gives great weight in its potential competition analysis to Beatrce s Dixie Sunshine Project, which originate as a private relabeling program, I.D.F. 236-2, I.D. 65 see discussion at pp. 59-0, infra, the Beatrice market shar it refers to in discussing the competitive effects of the merger represents only the amount ofCOJ reconstituted by Beatrce. The difference between these two market shares, however, is only 0. (0.57% versus 0.35%). We adopt the 0.57% figue. Neither side nor the ALJ haa said that the mere distribution ofCOJ is in and ofit.Jfa relevant line ofcoroerce. Ths level wowd include COJ distributed to retail without regard to brand ownership; for example, Beatrce s share oflle distrbution market would include COJ which it sold to retail under the Tropicana and other proce!lrs labels, while Trpicana s share would include that porton ofits CO products which it distrbute itslf directly to retail, the vast percentage of its total sales. Hlldred of dairies compete with each other as distrbutors ofCOJ pro.e98d and labeled by others, and Beatrice and Tropicna ocClioPlllly compete directly as wholesale distrbutors ofTropicana-Jabeled juice. Nevertheless, while the distribution level theoretically could be a relevant line of commerce, neither the complaint nor the pares focused on it, and, not surprisingly, the record conWins no evidence of concentration ratioll, market shares, or possible competitive effect. from the acquisition at thilllevcJ. Fuher, given the ease of eJ:try and muJtitllde affirms operating as distributors in thill indu.try, it is extremely doubtfw that the merger could have had any mgnificart anticompetitive effects at this level. Opinion 101 F.

quently, this assertion is legally significant only insofar as it raises the possibility that access to supply may be a barrier to entry into the processing or reconstituting of COJ, and does not lead to the conclusion that processors do not compete with reconstituters. Furthermore, respondent Tropicana s argument fails to recognize the fact that processors regard many dairies which reconstitute COJ, such as Johanna Farms, Home Juice, Ohio Pure Foods, Sealed Sweet, and Vita Pak, to be competitors in the production and sale of COJ. LD.F. 87-92. In addition, respondents' argument overlooks evidence of commercial realities in the COJ industry which lend support to a finding that processors, reconstituters, and relabelers ofCOJ sold to retail may be appropriately grouped in a single product market. It is true, as they contend, that dairies which distribute COJ supplied to them by processors are customers of those processors and in that sense are not in a horizontal relationship (19) with them. RPF 102--3. Both complaint counsel and the AU recognize this reality. CR 44; LD.F. 159. Indeed, until recently at least, the role of the dairies in the COJ industry traditionally had been as distributors of processors' brands of COJ as an adjunct to their regular dairy lines. The existence of a vertical relationship between a processor and dairy does not mean however, that they do not become competitors when the dairy assumes COJ reconstituting or relabeling functions. In this respect, we believe the central question is whether the transformation of dairy distributors into reconstituters or relabelers of COJ sold by them under their own brand names has independent economic significance and places them in a horizontal competitive relationship with processors in the sale of COJ to the retail market. Despite important functional differences between processors and dairies in the COJ market, which may serve to operate as significant barriers to large-scale COJ production by dairies, there is substantial evidence in the record to indicate that dairies, such as Beatrice, can operate as independent competitive forces in the sale of their own brand of COJ to the retail market. In the first place, they have the capacity, as reconstituters, to generate and add to the production of COJ that (20) is made available to the retail market. They obtain Moreover, it appears that Kraft, regarded by Tropicana and others as Olle ofthe three leading processrs of COJ, actually operates as a relabeler in a small percentage of its .sles to the retail market. Kraft has a ccrpack.!1g agreement with Sunk.st for aoout 10% of its product, under which Sunkist processes Hod packages the juice for Kraf, and Kraft then distributes it under its own labe!. I.D.F. 112- Re:;pondents never explain why these relabeled product 531es by Kraft are considered to be in the relevant !.ne of commerce, but not al! other brands ofreCoDstituted or relabeled COJ products sold to the retail market. 2: Il should be noted that there are no vertical allegations in the complaint arising from this seller.buyer relationship.

24 Indeed, the ALJ found that increased recollstituting by dairies has been all importt contributor to the rapid growth of COJ production in recent years. LD.F. 253. One dairy, Hood, which recorutitutes COJ from bllk concentrate, is the fourth largest producer orga! in the country. In addition, several other dairies which make recollstituted COJ are recognized as signficant competitors to processors and others in the sale of COJ to reta customers. While adequate supply of concentrate may not always he assured, it is significant to note that there (footnote copt' , p.

---.~_u_- - 733 Opinion bulk concentrate, which they reconstitute into COJ, from suppliers who are not necessarily also processors of COJ; in fact, the leading processors ofCOJ, Tropicana and Minute Maid, sell no bulk concentrate at all to dairies. Processors such as Tropicana and Minute Maid thus are in no position to restrict the output of bulk concentrate needed by dairy reconstituters. In view of these factors, and despite other possible limitations on the extent oftheir ability to reconstitute COJ, such as inadequacies in plant capacity, supply, and marketing resources, dairies can be the practical equivalent of processors in terms of the production of COJ for sale to the retail market. Second, brands of COJ reconstituted or relabeled by dairies tend to represent a lower-price alternative to the more established national brands sold under the Tropicana, Minute Maid, and Kraft (21) labels. I.D. 55; LD.F. 217, 257, 298.25 Indeed, as even respondents recognize dairies commonly cease distributing COJ under the processor s more expensive label, and make the relatively easy switch to reconstituting or relabeling their own product, when it is necessary for them to offer more competitive prices or promotional allowances to retailers in order to get shelf space. TAB 12. Price thus works both as a variable which helps to determine whether a relationship between a processor and a dairy is going to be vertical or horizontal, and as a means of creating competition between processor and dairy-owned brands for shelf space. Competition for shelf space, based on significant price variations among brands owned by processors and dairies, can result in a greater range of prices, and price savings for consumers. believe this fact alone confers a degree of economic significance in the COJ market to the reconstituting and relabeling activities of dairies. Third, dairies which produce and sell their own brands market and advertise them to retail customers and consumers in the same manner-if not to the same degree-that processors promote the (22) brands they sell.26 They are typically wholly responsible for the promotion of their brands, deriving little if any marketing assistance from their suppliers of bulk concentrate or processed COJ. CX 593 are nwneroW! suppliers of bulk concentrate (40 in Florida), LD.F- 141, such that dairies are not dependent on the major COJ processors for bulk concentrate supplies. Respondents argue that they are in horizontal competition with suppliers of bulk concentrate to these dairies, hut not with the dairies them!!lves when they sell COJ to the ret81 market. Whle such horizontal competition between processors and bulk concentrate suppliers for dairy customer orange juice accoUtts obviously exists, it does not follow that there cannot be horizontal competition between processors and dairies at a lower level in the distribution chain, namely, the sale of COJ to retail establishments Respondent Tropicana argues that dairy distribution of juice sold under another processor s label cannot be distingushed from saes of COJ reconstituted or rBlabeled by dairies Utder their own label. TAB 15-16. To the contrary, the lower price at which dairies can sell reconstituted or relabeled juice, and their substantially greater involvement in the promotion oftheir own brand products, represent signficant competitive distinctions frornju.ie which they merely distribute for another firm, under that finn s label 2f Again, this is in marked contrast to the more passive promotional role dairies play when they merely distribute juice bought from and sold Utder another company s label. Io the latter case, they typically have little responsibilty for or direct investment in the marketing and advertising of the brands distrbuted, those being primarily the fUtctions of the processor which owns the brand. 808 FEDERAL TRADE; COMMISSION DECISIONS Opinion 101 F.

117- 340-1. Through in-trade promotions and consumer advertising designed to gain consumer recognition and build consumer demand for their brands, they compete with processors for shelf space. CX 663A. Insofar as dairies produce, package, price, market, and advertise their own brands ofCOJ, and thereby expand the universe of products available to grocery outlets and consumers, they provide competition and operate on the same plane as processors in the sale ofCOJ to retail. The fact that they are less vertically integrated than processors- me step removed from the point of supply and original manufacture ofCOJ and bulk concentrate-is oflittle consequence in defining the relevant line of commerce under the terms of the complaint, and does not negate the commercial reality oftheir horizontal relationship to processors and others in the sale of COJ. Attempts to draw bright lines between firms engaged in the processing ofCOJ on the one hand, and the reconstituting and relabeling of COJ on the other, cannot obscure the fact that those lines are blurred in reality, and that the battle for retail shelf space among (23) processors and dairies which own brands is a relevant area of competition in the COJ industry.

In view of the above, it is not surprising that the evidence overwhelmingly indicates that processors, including Tropicana, see dairies as competitors when they engage in the reconstituting ofCOJ for sale to retail under their own brand names. LD.F. 87- , 219; CPF 181-88 369; CX 593, p. 348; CX 527Z-10; CX 620, p. 34-35, 90, 96; CX 661A; CX 6610; CX 662D; CX 892B; CX 953; Tr. 764-5, 786, 1058 1065 2906, 2922- , 2983-4.2 Dairies which fall into this group are the large regional reconstitutors, such as Hood and Home Juice, and assorted smaller reconstituters, such as Beatrice.28 Given the prevailing industry recognition that processors and reconstituters do compete for sales to retail outlets, respondents ' argument that they do not compete because they tend to use different delivery methods is overstated and not (24) persuasive.29 Where processors and dairies are sellng their products in the same geographic region, they are competing, and see themselves as competing, for the retail shelf space regardless of the method of delivery they use. CPF 183. But see,Tr. 4892 (Bealrice witness testified that Beatrice competes not with Tropicana, but with dairies that buy and distribute Tropica.na) 28 It is less certin that processrs view dairies as competitors in the sale ofCOJ to retail when they relabel COJ bought from a processor; Tropicana doe3 appear to consider such relabeled juice to be competition at the retail level, however. See, e. ex 593, p. 348. :! The major processors typically distribute to independent and chain warehoustl buyers, or individual chain stores, while dairies tend to deliver COJ straight to the grocery door. I.D.F. 77-84. Processrs and dairies do utili2:e both types of delivery, however. LD.F. 77, 79. , 84-5, 222-223, 258, and the recent trend of dairies has been to make warehouse rather than direct, "store-oor" delivery. LO.F- 78-1, 272-273, 292- Greater distributional and cost effciencies can be achieved through delivery to centra! warehouse Joc!3:tions. tD.F- 27U. 10 Indeed, brands of dairies delivered store-door have displaced warehousedelivered brands on the grocery shelf or at least have adversely affected their sales in many instances. l.D.F. 88-93 Respondent Beatrice assert that the Commission s decision in Beatrice Foods Co. ("Sexton ), 81 F. C. 481 (1972) (footnote cont' , p.

BEATRICE FOODS CO., ET AL. 809 733 Opinion Lastly, there is the question of whether retail chains which sell their own brands of COJ are in the relevant product market. In addition to their sale of processor and dairy-owned brands to consumers many retail chains, such as A&P, Safeway, and Kroger, buy juice from processors and sell it under their own brand label, which they market advertise, and otherwise promote without substantial assistance from 340-1. Theirthe processor. LD.F. 122- , 139, 146, 149; CX 593 private label brands do compete (25) with brands owned by processors and dairies for sales to consumers. With very few exceptions, however retailers are not vertically integrated backward into the processing and reconstituting ofCOJ, and the ALJ found that their private label sales would not have a substantial effect on competition at these levels. LD. 62. Furthermore, retailers with their own private labels are dependent for their juice supplies on those processors with whom they compete for consumer sales; they are thus subject to any potential output restrictions imposed by their processor-suppliers, which serves to limit the competitive significance of their horizontal relationship with processors at the retail level. For these reasons, we conclude that the relevant market does not include retailers. At the same time, it does not follow that retail COJ brand share cannot be used as a measure of concentration in the relevant product market to the extent it reflects reasonably accurately concentration and market shares in that market see discussion at pp. 40-1 infra. We affrm the ALJ' s conclusion that the relevant line of commerce in this matter is the processing, reconstituting, and relabeling ofCOJ for sale to the retail market. (26) Ill. THE RELEVANT GEOGRAPHIC MARKET The complaint alleged that the relevant section of the country for assessing the competitive effects of this merger is the United States as a whole and submarkets of it. Complaint counsel argue that the most appropriate geographic market is the nation, and offered no proof below of any relevant regional markets. Respondents contend the industry is divided into regional markets, citing the federal district court's finding in the preliminary injunction action of a ninestate area ranging from Colorado to Ohio as the relevant geographic market for assessing the merger. BAB 28. Respondents did not offer is persuasive precedent for their position that the existence of different methods of delivery establishes distinct lines of commerce between COJ sold by processors and dairies. BAB 31-32; ERB 38. We disagree- The Initial Decision and Commission opinion in that matter did note signficant differencesin two methods of distrbution in the dry grocery institutional wholesaling industry. 81 F. C. at 496-98, 521. However, contrary to what respondents assert, or at least imply, BAB 32, neither the ALJ nor the Commission held that these different methods of delivery constituted separate and distinct lines of commerce within the institutional market. 81 F- C. at 504, 521 See, e. , Proctor Cumbie 63 F. C. 1465 (1963),rev 358 F.2d 74 (6th Cir 1966),rev 386 U.S. 568 (1967); General Foods 69 F. C. 380 (1966),atrd 386 F.2d 936 (3rd Cir. 1967). 810 FEDERAL TRAD.E COMMISSION DECISIONS Opinion 101 F.

proof ofthat or any other regional market definition in this proceeding, however, beyond minimal testimonial evidence. TAB 36-38. The ALJ agreed with complaint counsel, finding the nation to be the relevant geographic market. He based his conclusion on a finding that the three largest firms in the industry, Tropicana included, sell nationally, and on the need to analyze the merger on a national scale in view of the complaint's allegations that the competitive effects of the merger were industrywide. I.D. 57-58.

Tropicana, Minute Maid, and Kraft, the top three firms, sell their product to grocery stores throughout all or most of the country. I.D. 101. Tropicana distributes to the entire continental U.s. from its plant in Florida, but most of its sales (82%) are in the eastern half of the country. It offers a single cost-based F. B. Bradenton, Florida price, plus freight, throughout the U.S. I.D.F. 105; TAB 39. Accordingly, the distributional costs and price of Tropicana rise in direct (27) proportion to shipping distances, the primary reason most of its business is in the east, close to its single plant in Bradenton, Florida. Tropicana s single price, plus freight, is the price available to all of its retail customers, wherever located, except when Tropicana offers promotional discounts in local markets around the country. CPF 206-7. Tropicana has sales representatives in every state but Alaska and Hawaii, CPF 206, and advertises in the national as well as the local media. I.D.F. 212. Minute Maid produces COJ by first processing bulk concentrate at its facility in Florida and then shipping it to its seven reconstituting plants located around the country. I.D. 54. Minute Maid sells COJ in almost all parts ofthe country. Kraft processes and ships most of its COJ from its plant in Florida, concentrating its business in the eastern and central portions ofthe country. It reaches the west coast through a co-packing arrangement with Sunkist in California, whereby Sunkist processes the juice and provides it to Kraft for distribution under Kraft labels. I.D. 54. These three national sellers of COJ generally base their prices to retail customers on processing and shipping costs, CX 593, p. 423but monitor and respond to prices and other competitive conditions prevailing in various locales and regions. See, e. TAB 19-20. They examine each other s prices most closely through sales data purchased from the A.C. Nielsen survey service. CX 560; Tr. 760, 1507. In addition, they use sales data and results from different regions in seeking sales in other areas of the country. CR 15. (28) Almost all ofthe other hundreds ofCOJ processors and reconstituters sell to retail customers in various parts of the country ranging in size from large multi-state areas to sections of a single state.32 By The president of Beatrice stated that he thought one and perhaps a few Florida COJ processors besides Tropicana shipped their product to California- ex 1650, p- 74. BEATRICE FOODS CQ., ET AL. 811 733 Opinion 1978, Beatrice had 12 milk plants reconstituting COJ in 11 states and distributing it to retail outlets in over 20 states, although its total market share was quite small. I.D.F. 231, 264. Its COJ operations are based mainly in the central and west central regions of the country, although they reach the west coast and the south as well. Several dairies are large regional reconstituters ofCOJ which compete with local firms and usually one or more of the three national companies in their areas of operation. Hood, the fourth largest producer in the nation, operates primarily in the northeast and some in the southeast. I.D.F. 113. Ohio Pure Foods, Johanna Farms, and Home Juice are big in the midwest and plains states. Foremost- McKesson and Knudson are among those who are prominent in the west. With the exception ofthese and other large regional reconstituters, dairies which sell COJ tend to operate in only one or a few states. The scope of a dairy s COJ sales can be limited by little available capacity for reconstituting and packaging COJ, and by the advertising and marketing costs of competing for shelf space against well-known national and regional brand names. Perhaps more importantly, the distance which dairies can ship COJ from their individual plants is inhibited by transportation costs, (29) method of delivery, and product perishability. Such distances generally range from 150 to 500 miles. I.D.F. 270.

Like the leading firms, local and regional dairies in the industry get nearly all of their bulk concentrate supply from oranges grown and processed in Florida. The availability and price of COJ sold to the retail level throughout the nation thus is directly affected by conditions of supply in Florida. The local and regional dairies, in monitoring the prices of competitors in their areas, pay the most attention to the prices ofthe national brands with which they compete. Tr. 964-5 1029, 1059, 1314, 1575; CPF 223, 242. In order to vie effectively for shelf space against the greater brand acceptance, loyalty, and advertising capacity of the national sellers, smaller competitors everywhere are forced to price their products substantially below those of the national brands. CX 560 (local and regional brands usually found to cost less than national brands in regional price surveys); I.D.F. 139 217 257; I.D. 55, 63. In this way, the three national sellers ofCOJ- Tropicana, Kraft, and Minute Maid-influence the terms of competition not only among themselves but also with those sellers with whom they compete only in local and regional areas across the country. As this description indicates, the COJ processing and reconstituting industry shows many of the characteristics of a highly regionalized geographic market. The vast majority of firms in the industry operate locally and regionally, constrained from expansion by a variety of factors, most notably marketing costs and (30) restricted shipping Opinion 101 F.

distances from individual plants. They are not significant competitive factors in the sale of COJ to retail beyond their localized regions and, practically speaking, cannot readily expand into new areas in response to increases in price in those areas.34 Thus, there could well be regional markets which might also be considered in judging the competitive effects of this merger-though, of course, the ability of national firms to expand into particular regional markets would have to be recognized in assessing any possible regional effects. Unfortunately, we are unable to identify any relevant regional markets on the basis of this record. As previously indicated, complaint counsel made no attempt to prove a relevant regional market as an alternative to a national market. Respondent Beatrice, in asserting that the market is regional, did little more than rely on the Nine- State area found by the district court and contends that assessment of the competitive effects ofthis horizontal merger has to be limited to this alleged area of (31) pre-merger competitive overlap between Beatrice and Tropicana. We reject this view as contrary to existing authority, which recognizes that the competitive effects of a merger between competitors can sometimes be felt beyond the immediate areas where they directly competed with each other. See, e. , Jim Walter Corp. 625 F. 2d 676 (5th Cir 1980); Brown Shoe 370 U.S. at 336-37; RSR Corp. v. FTC, 602 F. 2d 1317 (9th Cir. 1979); Spaulding, 301 F.2d at 606; US. v. Bethelem Steel 168 F.Supp. 576, 600 (S. 1958).35 Further, there is little evidentiary proof in the record that this Nine-State region comprises a reasonably well-defined, distinct group of COJ processors and reconstituters. The fact that it may be a general area of overlap of the two merging firms proves little if anything in defining the relevant geographic market. The other regional markets proposed are internally inconsistent with respondent Beatrice s own erroneous "overlap" test of definition in that they are not stated to be areas of competitive overlap, and otherwise suffer from a similar lack of proof. Respondent Tropicana suggests there are 7 to 12 regional markets, none of which corre- 1; By way ofiJJustration, it is useful to mention the unsuccessful attempt by Hood, the largest regional recon. stitutcr in the country, to expand beyond its New England base to the midwest. The record indicates the expansion effort failed due to advertising, marketing, and shipping barriers. It also bears mentioning that while Minute Maid sells throughout most of the U.s, its reconstituting centers can only distribute COJ regionally in light of the limited distances COJ can be transported. Only Tropicana, and to a les. cr extent Kraft, have managed to overcome the barriers to shipping CQ,11ong distances from individual plants on close to a national scale. And, as the ALJ found, transportation costs have limited Tropicana s growth in the tar west. I.D. 53.

J4 See discussion of geographic market consideratiuns in the DOJ Merger Guidelines. 35 Respondent. rely heavily on US. v. Philadelphia National Bank 374 U-S- 321, 357 (1963), in which the Supreme Court said that the relevant inquiry should be where "within the area of competitive overlap, the effect ofthe merger on competition will be direct and immediak." That decision does not support respondent Beatrice proposition. In the first place, the geographic market found in that case did not reflect perfect competitive overlap between the merging banks and other banks included in the market. Secondly, a horizontal merger can have a direct and immediate" effectoutside the precise area of competition between the merging firms. 733 Opinion sponds to Beatrice s nine-state area. (32) The 12-market estimate comes from Kraft' s 12 sales areas, as described at the trial below by an offcial for the company. These areas were designated by Kraft only for its own sales management purposes, and there is no other evidence in the record to indicate they are separate markets. CAB 33. The 7-market estimate, also proposed by respondent Tropicana, corresponds only to Minute Maid' s areas of distribution from its regional reconstituting centers. Id. In sum, there is insuffcient proof in the record to establish these areas as relevant geographic market under the accepted methods of definition. We conclude that the record fails to delineate any relevant regional markets which, on the basis of regional shipping patterns and other factors, could be reasonably assumed to exist in the COJ industry.

We are thus left to decide whether the record wil support the ALJ' finding of a national market in the absence of proof of relevant regional sections of the country within which to test the merger. In determining whether such a broader market exists, it is necessary to consider the abilty of firms in the industry to overcome shipping limitations in attempting to reach more distant markets, the extent to which prices found in various regions are interdependent with one another, and other pertinent factors.

Upon review of the record, we believe there is enough evidence to support the finding of a national market, although it is quite possible that regional markets could also exist which would reflect, perhaps more precisely, the commercial realities of the COJ industry and the competitive effects of the merger. In the first place, the area of actual and potential competitive (33) confrontation was itself not insubstantial; it appears that before the merger Beatrice and Tropicana actually or potentially overlapped in nearly 55 cities in 21 states in several diverse regions of the country comprising a substantial majority of the national population.36 Evidence that merging firms were serving most ofthe national buying population in the relevant product market prior to the merger has been taken into account in defining the We believe thegeographic market as the entire country.37 nationwide sales of COJ by Tropicana, coupled with Beatrice s geographically broad reconstituting and distributional system, constitutes one Ildicia of a national geographic market in this case. 36 CAB 29. Prior to the merger, both Beatrice Rnd Tropicana sold their COJ products to the retail market in the far west, rocky mountain, southwest, midwest, and BOuthem regions of the country. I.D.F. 1W, U5. According to 1970 census figures, the population within the area generally or potentially oorved by Beatrce through it.s vast m.lk distribution network, and with which Tropicana s natiowide sales substantialy overlapped, represented nearly 80% of the entire UB. population. ex 2306 A-Z-2. 31 RSR Corp. U. FT 602 F.2d 1317, 88 F. C. 873, 883 (1976). 38 Beatrce de minimis market share, smaller than that of the large regional dairy reconstituters such as Hood, belies its geographically wide area of distribution, which spans much of the country and is much larger than that ofthooo flrs.

Opinion 101 FTC.

Another one is the fact that the other leading firms in the COJ industry, Kraft and Minute Maid, along with Tropicana, also sell COJ to the retail market throughout all of most of the nation. LD.F. 101. Like Tropicana, Kraft concentrates its COJ sales from its Florida plant in the southeast, but has been able to ship product from that plant to the midwest and west central U.S. (34) beyond the usual upper limit of 500 miles. As already mentioned, it also reaches the west coast through a co-packing arrangement with Sunkist. Minute Maid sells nationally through its 7 reconstituting plants, which receive their supply of bulk concentrate from Minute Maid' s processing plant in Florida, which is not subject to shipping restrictions. There are thus three firms in the industry, controllng a majority of total industry sales, which have developed means to distribute COJ great distances to all or most of the nation despite the associated high transportation costs. They represent alternate sources of COJ supply to the retail market throughout the country and, generally speaking, have the capacity to increase or expand sales into most areas of the country in response to price rises or other competitive openings. There is a fourth company, Beatrice, which through its geographically diverse distribution network also represented an actual or potential alternate source of supply to much of the nation prior to the merger. Transportation barriers to distant shipping, a reality for most of the firms in the industry, have not been an insurmountable obstacle to these particular firms. While we reiterate that the regional shipping restrictions which apply to the great majority of firms in the industry strongly suggest the existence of regional markets as well the wide geographic, or even nationwide, scope of sales by all the leading firms and Beatrice constitutes evidence of a national market. (35) Pricing patterns in this industry also lend some support to a national definition of the geographic market. A common pricing factor taken into account by retail customers in different sections of the country is the uniform F. B. price, plus freight, which Tropicana charges wherever it sells its COJ product. LD.F. 105. The ALJ found that other COJ processors and reconstitutors around the country typically regard this price as a ceiling, and accordingly set their own prices beneath it. ! Such smaller firms, including Beatrice, tend to charge less than the price of Tropic ana and the other national brands because that is the only way they can compete against the strong The ALJ also found that these firms centrally plan, price and advertse saes of COJ to retail OIl a national basis. I.D.F. 104, 106, 107, 109 .0 Tropicana, as well as Minute Maid and Kraf, discounts price in variou,, setions of the country as local competitive circumstances dictate 41 The ability of a firm to impose a price ceiling in a distant area has been cited as evidence of regional interdependence indicative of a broad geographic ma ket. See Bethlehem Steel Corp. 168 F-Supp. at 600. .......... . , .

733 Opinion consumer loyalty attached to these brands. Conversely, the national firms centrally monitor local prices through regional sales reports and Nielsen data, and are in a position to move shipments around in response to localized price changes and promotions. We believe the external influence that Tropicana s uniform pricing policies (and those of the other two leading firms) have on local pricing decisions along with their practice of adjusting their centrally set prices where necessary to respond to local competitive conditions, demonstrates some regional interdependence mildly supportive of a determination of a national market in this industry. (36) We find the record capable of sustaining the ALJ' s finding that the relevant section ofthe country in this matter is the nation as a whole. The three largest firms in the industry, including one of the merging partners, compete on a nationwide level.42 While freight costs limit the distance to which individual plants can readily ship COJ, the constraint is only relative, and, as we have seen, certain firms do send their product into regions far distant from their plants, where they are able to respond to local competitive conditions. In particular, both ofthe merging firms here serve, or have the capacity to serve, all or almost all sections of the country and most of its buying population and overlapped before the merger in diverse regions of the country. Our finding that the competitive effects of the merger can be weighed in a national market is not inconsistent with our belief that regional markets, albeit undefined, probably exist as well. It indicates only that the relevant line of commerce in this case is both regional and national in character, and ilustrates once again the often unavoidable imprecision of geographic market drawing in antitrust analysis. Yet the imprecision is not so great here as to prevent a determination that the U.S. as a whole is an appropriate geographic market for an examination of the competitive effects of the merger to which we wil now turn. (37) IV. COMPETITIVE EFFECTS OF THE MERGER Before we discuss the competitive impact of the merger, we need to resolve a threshold dispute between the parties over the appropriate data base which should be used to estimate market shares and concentration levels in the COJ industry. Over respondents' objection the ALJ relied on Nielsen retail brand surveys in making these determinations, finding that they provided the most accurate evidence of the ,? In the matter ofthe Procter Gamble Co. 63 F. C- at 1561, where a national market was stipulated but nonetheless questioned by the Commssion because of the prevalence of regional shipping patterns, only one finn in the industry was a national producer. Here there are three, and a fourth that had a preexisting distrbutional capacity for nearly nationwide Jes Opinion 101 F.

market available. I.D. 54. Below we examine the merits of respondents' attack on the reliabilty of the Nielsen evidence. The essence of respondents' objection is that the Nielsen surveys cannot be relied upon because they seriously understate the actual size of the relevant market by concentrating inordinately on the major chains in big city markets while missing grocery stores outside urban areas. Respondents argue that this alleged defect has the effect of significantly overstating industry concentration and the market shares of Tropicana and the other leading firms, which make most of their sales in the big cities. Respondents attempt to establish this allegation by questioning the validity of the Nielsen methodology itself, by asserting that it is not relied upon by the industry to measure market shares, and by citing the industry sales data of Dr. Jay Gould, an economist retained by respondents, which allegedly show that the market is much larger than the Nielsen estimates. (38) The Nielsen Retail Index, established in 1934, CX 517C, has become a household name in American business. It is widely used by a variety of industries to monitor business performance by measuring sales and market shares for its subscribers and their competitors. CX 517 F, J. The accuracy of Nielsen s findings depends, of course, on the reliability of its sample as an indicator of the structure and size ofthe market. Nielsen s national store sample is derived from regularly updated S. Census data and its own detailed accounts of large and small stores by geographic area. CX 517 R. Selection of stores for the sample is determined mathematically to minimize bias. CX 517 R-S. Of particular relevance to the present proceeding, Nielsen indicates that it designs its grocery store sample to include the extremely important universe characteristics of geographic location of stores and their sales volume. CX 517 S. Its 1600-store grocery sample "is designed to achieve maximum geographic dispersion, containing stores located in Id. At the same timesome 600 counties in the 48 contiguous states. 44 it attempts to take a realistic picture of actual overall sales activity by including more stores with large sales volumes than with small ones. 45 (39) Respondents contend that this methodology produces an unrepresentative picture of the market because it misses hundreds of brands sold in smaller stores not reflected in the sample, and because Meadow Gold, which is sold by Beatrice in over twenty states, showed up in only one store in a special Nielsen survey commissioned for 1 The ALJ made this finding without discussion, Some analysis in support of the finding would have been helpful to the Commission in its review of the issue 44 Xielsen maintains that such dispersion makes its sample of value to regional as well as national sellers. CX 5175.

;; Such disproportionality "creates the equivalent, from the standpoipt of accuracy, of a much larger sample, CX 517 T, serving to assign the correct proportional weight to the relative sales of large and smaH stores in the market. CR 83-4 g., , , p. 733 Opinion respondents. BAB 23; RX 255. Yet the logic of the Nielsen methodology seems self-evident and likely to enhance the projectability of its findings to the entire retail market. Insofar as the national and major regional brands dominate the market and sell in the highvolume stores, while the smaller brands like Beatrice s are found less in the big stores and more in the smaller ones, Nielsen is not "missing" smaller brands, it is simply reflecting their proportionately smaller presence in the market.

Testimony from the representatives of all the leading firms and other industry offcials clearly establishes that Nielsen is regarded as the most reliable and most relied upon measure of the (40) market. CPF 247-255. Respondents contend, however, that it measures only market trends, not brand shares. This assertion is not supported by the record. The witnesses for every major firm, including Tropicana testified that they rely on Nielsen-often exclusively-for market share as well as market trend information. CX 593, p. 24, 94-95 388-9; CX 1650, pp. 135, 139-40; Tr. 780, 782, 1318, 1496-97, 2903. This testimony corroborates Nielsen s own statement and reports, see CX 517 F , which indicate that it provides market share as well as various other kinds of data to its users. Despite the strong evidence in support of Nielsen, respondents urge us to find, on the basis of Dr. Gould's estimates, that the Nielsen data materially understate the size ofthe market to (41) Tropicana s legal detriment.'9 For fiscal 1978, Nielsen found that there were 204. milion gallons of COJ sold for $545.2 milion at retail. 50 This is substantially less than Dr. Gould's estimate, which for calendar year 1977 was $780 milion. RX 2401. Dr. Gould's $780 milion estimate includes sales of other types of citrus juice, not just COJ. CR 78; RX 2401, pp. 205-210; Oral Arg. Tr. at 41. Further, his estimates for total 46 The above-mentiuned Nielsen survey indeed pruves that Meadow Gold was ubserved while it does not, un its face, establish respundent Beatrice s prupusitiun that it was seen in only one sturein t.he natiunaJ survey. RX 255; seealsoCR59.

.) The witness fur the Florida Departent of Citrs, which uss Nielsen, said it "represents the entire retail food market. . . certnJy the best fur retailing." Tr. 1827-28. The sales manager fur Tropicana testified that Nielsen is accurate tu a 95% confidence level and is representative of retail stures thruughout t.he u. s. ex 593, p. 194 255-7. He furher testified that it covers even small " venience" stores, and that. he could not name one t.ype of retail food outlet that. it did not. include in its survey. Iri- at 181 (in camera).A spokesman for Minute Maid also vowed that Nielsen was reliable and produced results remarkably close to Minut.e Maid's own sales data. Tr. 783. Finally, Dr. Gould, respondents' expert, testified that. he bad a high regard for Nielsen methodology. ex 2313 Respondent Beatrice implies that the reliability of Nielsen is unproven because no one from Nielsen testified in support of it. We reject such an inference, concluding that Nielsen s reliability has been established by its wide use and acceptance in t.he industry.

tS We have relied on Nielsen grocery store ssmplesin prior proceedings to obtain market share and concentration statistics. See, e- , FT v. Proctor Gamble 63 F, C. at. 1536. .9 Acceptance of respondenl ' estimat.es of Tropicana s market shar over complaint counsel's would leave Tropicana with a substantially smaller market share and considerably lessen the level of concent.ration in the market., BAB25 F. 121; ex 5602-24.

51 RX 2401 contains Dr. Gould' s testimony in the federal district cour proceeding concerning this acquisition. He was deposed in the administrative proceeding below see ex 2313, but did not testify at the trial. Opinion 101 F.

production ofCOJ, which fall in the $600-50 milion range, include sales to institutional (e. restaurants, hospitals, schools) as well as to retail customers. RX 2401, pp. 206-210. If institutional sales are subtracted 52 Dr. Gould' s estimates would not necessarily be inconsistent with the retail sales of $545 milion estimated by Nielsen. Further the Nielsen estimates are corroborated by the calculations of other citrus industry experts. See, e. Tiley, CR 79-81 in camera. In short Dr. Gould' s testimony, when broken down, does not refute the reliability of the Nielsen data. (42) We conclude the record evidence strongly supports the use of the Nielsen surveys as a reliable measure of the relevant market in this case. While Nielsen observes market activity only at the retail level it appears here to provide a reasonably accurate reflection of concentration and market share in the processing and reconstituting ofCOJ for sale to the retail level. The universe of retail sales of COJ should be about the same size as that ofCOJ sold to the retail market, except for unsold inventory. The percentage of retail sales that a reconstituter like Beatrice realizes should reasonably reflect its market share at the producing level, since it sells all of its brand directly to the retailer. Further, the retail brand share of processors like Tropicana should conservatively reflect their share of the relevant market, since a small percentage of the product they process is distributed to retail by other firms under those firms' labels, and under our definition of the relevant product market would be attributed to the other firm market share.

A. Horizontal Effects This case involves the acquisition ofthe largest firm in the relevant market by the nation s largest diversified food processor, which had a very small presence in the COJ market at the time of the acquisition. A key precedent for analysis of such a merger is the Commission s decision in Heublein Inc. 96 F. C. 385 (1980). Under that case it must be shown that the market is concentrated and that the market share of the smaller merging firm materially understates its true competitive significance and potential at the time ofthe acquisition. The ALJ found that these criteria were (43) met. We disagree, and reverse the ALJ's finding of a Section 7 violation on a horizontal theory.

Using Nielsen data, the ALJ found that for 1978, the year of the merger, Tropicana s market share was 29.9%, LD.F. 123, the highest in the COJ industry, and Beatrice s was .57%, LD.F. 213, a neglible Whle the record doe not indicate the tota annual amount of such institutional sal,"!!, we can reasonably infer from the great number and size of the institutional purchasrs of COJ that they are in the scores of millons. 733 Opinion percentage.53 At the time of the acquisition, four-firm concentration was 60.2%, reflecting "moderately high" concentration under existing authority54 and the Justice Department Merger Guidelines. Two-firm concentration was 45.4%, which is quite high and a source of greater concern than the four-firm ratio, at least under the view that two-firm ratios may be more useful than four-firm figures in assessing the likelihood of anticompetitive interdependent behavior.56 However, while these concentration levels are high enough to trigger antitrust concern, we stil must examine other relevant. characteristics of the market to determine whether the merger is likely to be anticompetitive. This additional analysis is particularly necessary here, given the somewhat imprecise character of the relevant product and geographic (44) markets that are capable of being drawn on this record. As noted by our Merger Statement at pp. 5- and the DOJ Guidelines at TIII(C)(1)(b), market share data should be given somewhat less weight than usual when the market definition is ailbigous or when products outside the relevant market may stil be reasonably close substitutes.

Horizontal violations have been found in a few cases involving only a minimal increase in concentration, but rarely in one with an increase as tiny as that in this proceeding.57 In assessing the competitive consequences of mergers which increase concentration to such slight degree the courts have focused on the following qualitative criteria in addition to market share and concentration statistics: (1) merger and concentration trends; (2) the existence of barriers to entry; (3) evidence of interdependent pricing and other potentially collusive behavior; (4) the extent of actual competition between the merging firms; (5) the degree of vertical integration among competing firms; and (6) the degree to which the (45) market share ofthe smaller firm understates its true competitive importance in the relevant market.

6J Market share and concentration statistics used herein reflect gallon sales, not doJjar sales. " See Heublein 96 F. C. at 577, ll. 9.

"Under those gudelines, this 4-firm ratio would fall within the range of moderate concentration (1000 to 1800) on the Herfindahl.Hirschman Index.

06 Id. at 577. The disparity between the market shares of the top 2 firms, Tropicana and Minute Maid (15.5%), or the top 3 if we include Kraft (10.4%), and the rest ofthe firms in the industry, is quite dramatic. Ths is a factor which the Commission recently said requires "particular attention" in market share and market performance analysis. See FTC Merger Statement at 3.

57 The ALJ relies onFederal Trde Commision u. Pepsico, Inc. 477 F.2d 24 (2d Cir. 1973), which involved an action for a preliminary injunction against the acquisition ofa firm with 0.3 to 1% of the relevant market. While the cour of appeals upheld the finding of a reasonable probabilty that the merger was anticompetitive, it did so on the basis of an incomplete record and incomplete ana.ysis of the likely competitive effect of the merger. Id. at 27. As such, only limited weight can be given to the decision. To the extent it is authority, moreover, the decision is factually distingushable from the present matter in that there was evidence of much higher concentration and a pre-merger trend in the soft drnk industry- Id. at 26. r'inally, it remains the only decision to date we are aware of which has held that a merger involving a market share as low 9) Beatrice s may be a horizonta violation of Section 7.

See our application of horizontal merger case law inHeublein 96 F. C. at 578-2; see also Fl Merger Statement at 3- and DOJ Merger Guidelines at TI II (B)-(C). Opinion 101 F.

On the surface the present acquisition resembles somewhat past acquisitions which were held unlawful by reference to the above criteria. Like Alcoa-Rome and Stanley Works v. FTC, 469 F.2d 498 (2d Cir. 1972), cert. denied 412 U.S. 928 (1973), the present case involves the leading firm in an industry which was fairly heavily concentrated at the time ofthe merger.59 In addition, the ALJ found that Tropicana was a "price leader" in the COJ market, LD.F. 153, a factor that received considerable weight in the finding of violation in Stanley Works. On closer analysis, however, the two cases are not analogous. In Stanley Works, price competition was already limited in the relevant market and there was evidence Stanley would bring to that market its policy of minimizing price competition that it had displayed in other product Iines.6o Under these circumstances the Second Circuit concluded that the acquisition of the market leader by a company like Stanley posed a dire threat to competition in an already competitively weak industry. We do not find the same circumstances here. The ALJ made no findings that Beatrice has operated in other markets as a (46) price leader in the same anticompetitive fashion as Stanley. Also, there is little evidence of such limited price competition in the COJ industry. While Tropicana and the other national sellers are "price leaders" insofar as they establish price ceilngs, there is nonetheless evidence of meaningful price competition in this industry. The record shows that the rest ofthe industry competes by underpricing Tropicana and the other large firms, indeed, that they must do so in order to compete effectively against the brand name recognition and consumer loyalty attached to the national brands. LD.F. 139 LD. 62--3. It is not surprising, therefore, that the ALJ found the COJ prices of a dairy like Beatrice to be on the average 25-35 percent below those of national brands like Tropicana. LD.F. 217. Given these findings and the lack of evidence in the record that Beatrice through its acquisition of Tropicana is likely to suppress price competition in the COJ market, we do not find the imminent threat to competition resulting from the merger that was central to the Second Circuit' holding in Stanley Works.

Barriers to large-scale entry in this industry are high but do not appear to have unduly impaired competition in the COJ market. Large-scale entry requires access to large and reasonably certain supply, substantial plant and distributional capacity, a big advertising budget and/or established brand name, and a large, aggressive sales force. LD.F. 267. A few firms, most notably (47) Minute Maid and the big regional processors, have overcome these barriers to varying de- 59 Indeed the market share ofTropic8.na is even higher than that of the larger firm in each of those cases- Also finn concentration in the COJ industry at the time of the acquisition was greater than thatStanleyin Works hut less than that inAlcoa.

00 78 F. C. 1023, 1067-1074 (1971).

-_. u_- ~ + .u 733 Opinion grees and become significant competitors.

In addition respondents contend that Minute Maid represents just the beginning of a new wave of entrants who will be trying to capitalize on the growing demand for COJ over FCOJ. They argue that FCOJ packers and even grocery chain stores are likely entrants. It is not clear, though, that such entities, even if they have the incentive to enter, have the necessary capacity for entry on a national or even broad regional scale. Minute Maid's growth, while extraordinary, illustrated what it takes to become a significant, potentially destabilizing force in the national COJ market While large-scale entry appears extremely diffcult, the record indicates that small-scale entry by dairies into the COJ market is easy and fairly common. LD.F. 131, 134-35. The ease and prevalence of such entry by dairies on the local and regional level, who compete by underpricing the big firms, serve to maintain a reasonably competitive marketplace offering consumers (48) a fairly wide range of COJ quality and price.63 In our judgment this reality, along with the economic incentives for at least modest entry and expansion created by the great growth in demand for COJ, checks at least somewhat the risk to competition otherwised posed by the high barriers to major entry.

Next we must decide whether the market shares of Tropicana and Beatrice reasonably reflect the degree of competition between them that would be sacrificed by the merger, and whether Beatrice minimis share accurately reflects its true competitive significance in the COJ market. The ALJ seemed to indicate that the evidence that they regarded each other as competitors and directly competed with each other in various locales see LD.F. 218-226, is suffcient to establish that Beatrice s .57% share of the national market reasonably reflects the degree of actual competition between them. This is open to question. While the two firms generally overlapped in several regions, their respective operations were concentrated in different regions- Tropicana in the east, Beatrice in the midwest and west. While the record does show actual competitive confrontation in some areas and the potential for direct (49) competition in many others, it will 61 Minute Maid, of course, has been an exceptionally successful entrant as a national seller ofCOJ , using heavy advertising to increase its market share from 7.3 percent in 1975 to19-1'0 in 1979, the 2nd highest. LD. 55. 62 While Minute Maid' s entry did not reduce overall industry cuneentration before the merger, it may still have benefited consumers and price competition by accelerating the trend from glad to lower cost paper orange juice containers- There is some evidence that Tropicana, which once sold CO.l only in glass bottles, has increased its sales of COJ in paper cartons, perhaps in response to the Minute Maid challenge. 6J Thto AL found such entry responsible for much of the recent growth in COJ saes, LD.F- 253, and furher found that ind\L try tmnds are moving in the direction ofim:reaserl rtoconstituting ofCOJ by dairies outside Florida to avoid the high freight costs associated with juice processtod in and shipped from Florida. I.D.F. 293-99. Somewhat contradictorily, however, he also found that local COJ processors are becoming a ltoss important factor in the COJ industry, based not on evidence of less competition by such finns but on a moderate increase in concentration prior to the acquisition. r.D. 63; I.D.F. 122-23. We question whether the latttor finding can be madto on the basis ofthe concentration data aJonto.

Opinion 101 F.

Ifnot support a finding of heavy competitive overlap of their sales.6' anything, it suggests that overlap in the national market is far from substantial and that Beatrice s .57% market share may well overstate the head-on competition that was eliminated by the merger. The ALJ further found that Beatrice was a more significant competitive force in the market than its de minimis market .share would suggest. This finding must be examined by reference toAlcoa-Rome as interpreted in Heublein. In Alcoa-Rome, the Supreme Court said that Rome s small market share significantly understated its competitive potential, based on evidence that Rome was an unusually aggressive competitor and technologically-skiled innovator in the relevant product market. In Heublein the Commission read Alcoa- Rome to require proof that the firm is a truly special or unique small competitor-with a clear capacity to destabilze the market-as measured by such qualities as originality in product research and development and boldness in price competition. 96 F. C. at 581. Absent these qualities, we said there, evidence that the firm was outstanding in other areas, such as advertising and (50) marketing, would not be enough to find that a firm was a special small competitor within the meaning of A lcoa. I d.

In support of his finding, the ALJ said that while Beatrice was not an innovative competitor, it possessed "great potential capacity" and a "unique ability" for marketing COJ nationally through its vast system of milk plants-the largest in the country. LD. 60, 65. Combined with evidence of Beatrice s growing interest and expansion in the reconstituting and marketing of COJ prior to the merger see discussion at pp. 58-0 infra the ALJ believed this was enough to support a finding that Beatrice s small market share substantially understated its competitive significance and made it, in effect, a "special small competitor" within the meaning of Alcoa and Heublein. LD. 61.

We agree with the ALJ that great skil and resources in advertising and marketing appear necessary to gain sizeable market share in the COJ industry. Tropicana and Kraft, the traditional leaders in the market, have won consumer loyalty largely by virtue of established brand name recognition and sustained promotion of their products. The other large firm, Minute Maid, soared from zero to over 19% of the market in less than a decade on the strength of massive advertis- 64 We do not mean to imply that evidence ofsl.bstantial competitive overlap is lleteBSry to a finding of horizonta violation. However, as we said inHeublein the extent ofhead-o!' competitionin a merger cas involving a firm with a very small market share is relevant in evaluating whether that share realsticaly represeuts the firm actual competitive position and signficance in the industry. 96 F. C. at 581, n. 15. See a/so DOJ Merger Guidelines at n II (C)(1)(c) 65 Specifically, we sad "it would distortAlcoa definition of the special small competitor to make advertising or marketing, oxcept in special circumstances not present here, a distinguishing characteristic. Heu.blein, 96 at 581- BEATRICE FOODS CO., ET AL. .823 733 Opinion ing and marketing expenditures which had the effect of extending preexisting loyalty to the Minute Maid name to its COJ product. LD. 55. These three national sellers have the capacity to (51) utilize and rely on their well-known names and superior advertising resources as a means of maintaining their market strength to one degree or another against numerous competitors who have to compete on price rather than promotion because they lack a comparable ability to win sales through heavy advertising and marketing.

In short, large-scale entry and expansion in the national COJ market has not been achieved, and appears virtually unattainable, without the capacity to advertise, market and promote one s product on a large scale.66 Beatrice, as the largest diversified food processor in the nation and second largest advertiser of consumer food, has demonstrated that capacity in many ofits other product lines. LD.F. 279-83 287. Although it showed no special or unique marketing skills or commitment in its COJ line prior to the merger, one may reasonably assume that if in fact it had intended to expand significantly in the COJ market other than by a large acquisition, it possessed the requisite advertising and marketing resources, as well as distributional means, to achieve its ends. Indeed, in view of its track record in other product areas and status as the only "conglomerate" among the small" firms in the national COJ market at the time of the acquisition, it is (52) probable that Beatrice s ready capacity for such expansion was unique. Accordingly, this possibly could be seen as one of those exceptional cases mentioned in Heublein where the possession of a unique advertising and marketing capacity makes a company a special small competitor" within the meaning of Alcoa.6 To the extent such capacity is essential to enable a firm to become a destabilizing force in the relevant market, it is arguably the functional equivalent of product innovation or aggressive pricing as a catalyst for competition.

Stil, the record does not show that Beatrice had begun or firmly committed itselfto actually utilizing its full advertising and marketing capacity in the relevant product line prior to the merger, and we are thus reluctant to classify it on this basis as a special small competitor in the COJ market at the time of the acquisition. We believe that A lcoa-Rome requires primarily a look at factors other than a 00 Of course, a firm must also possess the capacity to proce9S and distrbute on the !!ile scale. Ths requiree the capacity to operate seveml geographically varied processing and distrbution ccnters, which Minute Maid hils, or the unusual capacity to ship long distances, which Tropicana haa shown. It is no doubt tre that many ofthe dairies which recoiltitute COJ lack these capacities for growth, in addition to lacking the requisite advertsing and marketing capacity.

67 It is important to emphasize that unlke the market inAlcoa-Rome and perhaps the market in Heublein technological innovation ia not a critical competitive factor here. Despite differences in quality and shelf life among competing COJ product. which reflect how the product is processd and packaged, the potential for technological advancement in the makinl1 and "hiuninE" of COJ is finite. Opinion 101 F.

firm s mere possible plans for expansion, in assessing its true market sigoificance in a horizontal merger analysis. Otherwise the legal theories of horizontal and actual potential competition, which rest on different analytical criteria, would begin to lose their individuality. The absence offactors which would allow us to conclude that Beatrice was a special small competitor within the (53) meaning of Alcoa and Heublein, combined with Beatrice s extremely small market share argues strongly against a finding that the merger is unlawful on a horizontal theory of violation.

Lastly, concentration trends are available as an index of the competitive effect the Beatrice-Tropicana merger is likely to have on the COJ market. The ALJ found this factor dispositive in his horizontal analysis ofthe merger. Unlike the Commission s findings in Heublein the ALJ found here a strong trend toward concentration prior to and continuing after the merger. Four-firm concentration rose from 55.8 percent in 1975 to 60.2 percent in 1978, the year of the merger, and to 64.2 percent for a two-month period shortly after the merger. LD. 122-23; LD. 60. In contrast, there was no similar pre-merger trend in Heublein and a sharp post-merger increase in concentration in that case was tempered by a subsequent decline. We said that the net increase there from 47.9% to 54.2% in an eight year period was insuffcient to elevate to the level of a violation the increase in concentration resulting from this merger. Heublein 96 F. C. at 582. We reach the same conclusion here. While the COJ market was somewhat more concentrated than the Heublein all-wine market at the time of the present acquisition, and while the increase in concentration over a several year period was slightly higher in this case than in that one 68 we cannot say that the difference is great enough particularly in view of recent trends discussed below, to support the ALJ' s finding that (54) the trend toward concentration in the COJ market, by itself, is what "clearly distinguishes " this case from Heublein. LD. 60.

Furthermore, updated Nielsen data indicate there has been a trend toward less concentration in the COJ market since the merger, including a recent drop of several points in Tropicana s market share. 68 8.4% here between 1975 and 1979, and 6.3% there for a 7 year period. 69 In distingushing the two cases, the ALJ seemed to focus mote on a 2.5% decrease in concentration inHe, blein over the later half of the period in question, rather than on the overall net increase of6.3% for the entire period. Looking at the whole trend, rather than just the part, the distioCtiOll appears to be olle without much of a difference.

70 We grant respondents' motion for limited reopening of the record, fied November 3, 1981, to receive the new Nielsen evidence. Post-acquisition evidence of concentration trends call be relevant and the Nielsen data, 8S the record shows, is highly reliable. Also, the post-acquisition evidence in this case largely corroborates the evidence of record supporting our finding that concentration tret'ds are not dispositive of the anticompetitive effect of this particular merger. Were post-acquisition evidence dramatic"Uy different from the record evidence, we would be much Jess inclined to admit it unit and unless it hold been subjected to croBBxamination by the party adversely affected by it.

733 Opinion After rising to 62.9% in 1979, 4-firm concentration dropped slightly to 61.3% in 1980 and appears from the bimonthly data to have slid even further in early 1981, although yearly figures are not available.71 Finally, the new Nielsen figures show an increase from 20. in 1980 to around 24% in the first half of 1981 in the "all other category ofCOJ , which includes, inter alia the sales of the local and regional producers. This post-acquisition evidence, while not dramat- , (55) does undercut a finding of horizontal violation based on concentration trends in the relevant market.

In summary, the record does not support the ALJ' s view that the merger substantially lessens horizontal competition in the relevant COJ market. We emphasize that in a case in which one of the merging firms has only .57% ofthe relevant market, the party challenging the merger is under a particularly heavy burden to show that notwithstanding the de minimis increase in concentration, the merger is anticompetitive and a violation of Section 7. That showing was not made here.

B. Potential Effects The more plausible theory of violation in this proceeding is that the merger had the probable effect of eliminating Beatrice as a special potential expander and deconcentrator in the national COJ market. Complaint counsel alleged, and the ALJ agreed, that the merger could be held unlawful on this theory. We disagree, finding the record insuffcient to support this conclusion. Before we can even reach the merits of this issue, however, it is necessary to deal with respondent Beatrice s threshold argument that complaint counsel's allegations of Beatrice s plans to expand are based on a "potential competition theory of liability which was not pled by the complaint and which therefore, cannot be used to establish a finding of ilegal conduct. In the particular circumstances of this case respondent Beatrice argument presents a very close question, but on (56) balance we hold that no potential competition theory ofliability was pled in the complaint and that it is therefore inappropriate to consider evidence under this theory. We accordingly decline to rely on this theory in assessing the legality of the acquisition. We recognize that a respectable argument can be made in support of the proposition that an implied potential competition count was pled in the case. Although the complaint does not expressly allege that the merger would eliminate potential competition by Beatrice the absence of such language does not necessarily compel a reversal 11 Two-firm concentration did rise from 45.4% in 1978 to 49.3% in 1980, appearing to reflect an increase in Minute Maid' s market share to 20.8%. However it declined considerably in the first half of 1981 , mainly lis Ii result of Tropicana s market share dropping to around 24% durng that time. , Opinion 101 F.

of the Initial Decision on this ground. The Commission s rules simply require that complaints provide "a clear and concise factual statement . . . of the type of acts or practices alleged to be in violation of the law "72 and need not plead the evidence upon which complaint counsel wil rely. In construing such complaints, including the one here which uses legal terms such as "actual competition " it is appropriate to look to the case law interpreting the term, particularly contemporaneous case law that might shed light on what questions the Commission, in exercising its law enforcement discretion, intended to authorize complaint counsel to litigate. Under such contemporaneous precedent, a Commission complaint that pled only actual competitive effects has not automatically foreclosed inquiry into the loss of potential competition as well, particularly in the case of the elimination (57) ofa potential expander, as Beatrice was alleged to be here. In such cases, tp.e Commission has recognized that there is " clear line between actual and potential competition theories Retail Credit Co. 92 F. C. 1 , 152 n. 43 (1978), rev d on other grounds, 616 2d 63 (9th Cir. 1980), and--f particular importance to the instant case-has held potential expander evidence relevant to a horizontal as well as a potential competition theory of liability. Stanley Works, 78 F. C. at 1064; Pillsbury Co. 93 F. C. 966 (1979). On the other hand, while the line between these two theories has not always been "pristine" clear, generally speaking they have become suffciently distinct and well-established that a complaint pleading one theory, unless amended in accordance with Commission Rule 3.15(a)(1), would not ordinarily allow proof primarily relevant only to the other.73 The complaint here not only does not specifically allege any effect on potential competition, it clearly speaks in terms of an effect on "'actual (58) competition. See Paragraphs 13, 15 (a)- (b).74 Given the reasonably clear distinction between the two theories the Commission, with its own antitrust expertise, presumably would have distinguished between them and understood the legal significance of using only the term "actual competition" in pleading the anticompetitive effects of this acquisition. Considering the Commission s conscious choice of this particular complaint language, com- 12 Commssion Rules of Practice, Section 3. 11(b)(2). 13 This became particularly true following the Commisson a decision in Heublein, which clarified that it is necegsry to examine the issue of potential expansion within the analytical framework of potential competition antitrut theury, under the criteria for conglomerate merger euforcement established in United States v. Marine Bancrrporation 418 U.S. 602 (1974).

74 The complaint allegation that 8S a result of the acquisition the previously existing level of concentration .. wil be increased and the possibilitiesfor eventua deconcentration may he diminished" 15(d), provides at best oblique support for adjudication of a potentia. expander theory. Such "eventual deconcentration" language doe not amount to a specific allegation that thecquiring finn was a potential expander; indeed,imilar language has, been used in Supreme Court and Commission deci jong that did not rely upon potential expBnder evidence or analysis.See e. , United States v. Philadeiphia Nat Z Bank 374 U.S. at 365 n.42; Liggett Myers, Inc_ 81 F. 1074, 1165(1976).

.

733 Opinion plaint counsel were authorized by the Commission to try this case on a horizontal, but not a potential competition, theory of violation. Notwithstanding some precedent to the contrary, any other un.derstanding would have represented a strained interpretation of the reasonably clear meaning and intent of the "actual competition" allegations in the complaint.

As such, when complaint counsel advised the ALJ and respondents counsel, at the very first prehearing conference, that it intended to prove a violation based on the unpled theory of the loss of potential as well as actual competition, the obligation arose under Rule 3. (a)(l) to fie a motion for amendment of the complaint which should have been certified by the ALJ to the Commission. (59) In its dual role as prosecutor and judge, the Commission has a special obligation, different from that of ordinary courts of law, to maintain eflective control over the purpose, construction, and adjudication of the complaints it issues. Rule 3.15 (a)(l) was placed on the books to facilitate the Commission s exercise of such control. To allow new theories to be added, provided only that the respondent has adequate notice and an opportunity to litigate the issues, would defeat the very purpose of this important safeguard in our rules, and undermine the Commission s control over its prosecutorial discretion. Fortunately, the constraint on our ability to reach a potential competition theory in this case is immaterial, since the merger does not appear from the record to have substantially lessened potential competition in the COJ industry anyway. While the evidence indicates that Beatrice was interested in expanding its production of COJ for sale to the retail market, and probably would have become a larger competitor in the industry, it falls short of establishing a probabilty that Beatrice would have gained suffcient market share from the leading firms to deconcentrate the industry or otherwse contribute sigoificantly to a more competitive national market. Though Beatrice increased (60) the number of its plants reconstituting COJ from three to twelve between 1975 and 1978, LD.F. 227, 231, this does not seem to have represented a major corporate commitment to COJ-backed up expanded plant capacity for COJ and big promotional campaigosthat could have resulted in a shake-up of the national market or capture of substantial market share from the market leaders. In addition, it does not appear that Beatrice s new Dixie Sunshine brand, test marketed briefly before the merger and abandoned shortly thereafter, would have become the major, nationally sold product that would have vaulted Beatrice into the big time in this industry. Despite some evidence cited by complaint counsel to the contrary, there were formidable commercial obstacles to the accomplishment of longrange, nationwide distribution of this product from its planned pro- Opinion 101 F.

cessing center in Orange City, Florida. Dixie Sunshine was a "fresh" juice, and Beatrice lacked the supply and special freezing facilties necessary to large-scale production and distribution of a fresh COJ product from the Orange City plant, which also happened to be a financially unstable operation with an uncertain future. Further there is very little evidence that Dixie Sunshine ultimately would have been reconstituted and sold nationally through Beatrice s milk plant system, as complaint counsel claimed, or that Beatrice s plant managers were even interested in distributing the new product. The weight of the evidence, viewed in light ofthese commercial realities, is that absent the merger, Dixie Sunshine would have been sold primarily (61) as a fresh product, packaged in Florida and limited in distribution principally to the southeast. C. Entrenchment Effects The ALJ found that the merger had the anticompetitive effect (as alleged in paragraph 15 (b)-(d) of the complaint) of entrenching Tropicana as the market leader. While he rejected complaint counsel' s entrenchment theory that the acquisition conferred special advertising advantages on Tropicana through access to Beatrice considerable advertising assets, he agreed that it did confer special advantages in distribution through access to Beatrice s milk plant network that would strengthen Tropicana s market position in the west and thereby entrench its position nationally. LD. 65-6. Underlying this conclusion were the ALJ' s findings that Tropicana was searching for a more economically effcient way than shipments from Florida to penetrate the western market, that the industry trend of reconstituting at plants closer to the retail destination was the most economical way to do it, and that access to Beatrice s plants assured the attainment of its western expansion plans. I. F. 293-302; LD. 66-7. While entrenchment can be an independent basis of antitrust liability,75 the record does not support the ALJ' s conclusion that the distributional advantages resulting from the merger were likely to entrench Tropicana in violation of Section 7. (62) First, we affrm the ALJ' s findings that the record proves no unique advertising effciencies to the merged firm that are a direct consequence of the acquisition. With respect to the alleged competitive advantage in distribution, we agree with the ALJ that the acquisition would appear to enable Tropicana to expand substantially in the west by using Beatrice s dairy plants as distribution or reconstituting centers. However, the record fails to establish that Tropicana had plans to integrate the Beatrice milk plant network for this purpose following the merger. There is little evidence that Tropicana was prepared 75 Heu.blein 93 F, C. at 592, citing FTC v. Procter Gambl Co. 386 US 568 (1967) 733 Concurring Opinion to abandon its historical commitment to marketing a fresh glass product even though the industry trend was toward lower-cost reconstituting and Tropicana had made some movement itself in that direction.'6 A continuing primary commitment by Tropicana to its fresh glass product and its premium image could explain the absence-at least so far--fintegration ofthe reconstituting facilities of Beatrice plants into Tropicana s COJ business. Moreover, even if Tropicana eventually did use Beatrice s distributional resources for the purpose of making a reconstituted Tropicana product and achieving greater inroads in the west, the record does not support a finding that that would necessarily result in a substantial lessening of competition in the national COJ processing and reconstituting market. In view ofthe growth in demand for COJ, the ongoing expansion by Minute (63) Maid as well as major regional reconstituters in response to that demand 77 and the general case of entry (at least on a small scale), it is at least questionable whether even full utilization of Beatrice distribution system by Tropicana would have the effect of sigoificant- Iy raising barriers to entry or expansion or otherwise substantially decreasing competition in the national COJ processing and reconstituting market. Finally, the post-acquisition stabilzation and recent decline of Tropicana s market share do not bear out complaint counsel's prediction, made in support of its entrenchment allegation CAB 56, that the merger would increase both Tropicana s dominant market share and industry concentration.

In Heublein we said that because "adverse competitive effects from entrenchment' can be rather elusive, it is particularly important that a factual basis be carefully constructed. Id. at 593. That predicate has not been established here.

CONCURRING OPINION OF COMMISSIONER DOUGLAsl I concur in the result that the majority reach in this case. However I am troubled by portions of the majority opinion, including the product and geographic market discussions. This concurring opinion summarizes the most important of those concerns. Relevant Product Market As the majority opinion recogoizes, cross-elasticities of demand and supply are important determinants of the relevant product market. 76 In fact, Tropicana s commitment to its fresh product would appear to have beet! reaffrmed by jt. completion of a new $16 millon glass factory in197R Tr. 3784, 3915-16. 77 Along with Minute Maid, such signficant reconstituters as Joharma Farms, Home Juice and Ohio Pure Foods all grew after the merger and anticipate more expansion. RR 30; Tr. 1603, 1607. Furher, economist Tiley conservatively estimated industry growth at 50% over the next 5 years. Th. 2348--49. ! Chairman James C. Miler, III joins in this concurring opinion. Concurring Opinion 101 F. A universe can be characterized as a separate product market ifwithin an appropriate period of time-a small change in the price of products within the universe does not induce significant changes in the quantities of products outside the universe that are demanded or supplied.2 Of course, it may sometimes be diffcult to develope the empirical data needed to measure these factors in particular cases. Nevertheless, they provide the theoretical basis for product market definition. (2) In this case, the majority opinion concludes that the relevant product market includes chiled orange juice (COJ), but does not include frozen concentrated orange juice (FCOJ). I disagree with this conclusion. The majority opinion also recognizes that including or excluding FCOJ in or from the relevant product market does not affect the outcome of the case.' Nevertheless, I believe that accurate market definition should be an important part of merger cases, and that the record evidence on cross-elasticities of demand and supply is suffciently clear in this case to establish that FCOJ should be included in the relevant product market.

Cross-Elasticity of Supply The record evidence suggests that the cross-elasticity of supply between FCOJ and COJ is high. Since COJ can be produced from fresh oranges, from bulk concentrate (FCOJ) or from bulk frozen single strength juice 6 any firm that can produce or purchase bulk concentrate can sell both FCOJ (3) and COJ. Moreover, much of the equipment used to produce COJ directly from oranges can also be used to 2 FT Statement On Horizontal Mergers (June 14, 1982), reprinted in42 ATRR Special Supplement (June 1 1982) (hereinafter cited asFT Statement;, at 3-15; Justin€ Department Merger Guidelines (June 14, 1982), reprinted in42 ATRR Special Supplement (June 17, 1982) (hereinafter cited DOJas Guidelines), at (8-) - (Si5). The Justice Department suggests that, as a first approximation, the lest of the viabilty ofa prospective market should be whether a 5% price iocreasc for products in that. market would induclT-within one year-a significdnt percentage of buyers of products already included in the market to shift to other products not yet included in the market. DO Guideliues. supra at S- J Majority opinion at. 25.

'ld.at15 5The respondents did not pre this argument on appeal. However, that does not foreclose the Commssion from considering the n"cord evidence on this importat issue sua sponte. 6 I.D.F- 63. The majority opiniorl treats COJ made from fresh orMlges)md COJ made from concentrate as part of the same product market. Majority opinion at 14 n. 16. 7 For example, Minute Maid began producing COJ after becoming the leading proces.or of l"COJ. Majority opinion at 7 n- 6. It should aloo be noted that the hundreds of dairies operating nationwide can easily switch from fluid milk production to COJ production.&e note 29 and accompanying text, infra. The packaging equipment needed to produce COJ and FCOJ differs, because FCO,J is typicaly sold in small csrdboard cans, while COJ is sold in cartns, plastic containers, or glass contanfJrs- I.D.F. 51. However, the equipment needed to manufacture the different containers is apparently widely available and not particularly expcosive. Moreover, producers can purchase the finished conlainers from specialized manufacturer&--or set up co-packing arrangements with firms that already have the neces;ry equipment-if they wish to do so- For example, Kraft presently has a co-packing agreement with Sunkist, under which Sunkist processes and packages approximately 10% of Krafl' s juice, and Kraft then distribute it under its own label Majority opinion at If! n. 22.

BEATRICE FOODS CO., ET AL. 831 733 Concurring Opinion produce COJ from bulk concentrate." Furthermore, there are a number of processors in Florida, including grower cooperatives, that currently have all the equipment needed to process oranges into COJ or FCOJ and can change their capacity from one to the other as a timetion ofthe relative profitability of each form." As a consequence, the evidence indicates that FCOJ producers would encounter relatively little diffculty in shifting to COJ production in response to higher and more profitable COJ prices. It would not be as easy for COJ producers to switch to the production of FCOJ, (4) because different equipment is required. However, the crucial issue in treating COJ as a separate market is whether an increase in COJ prices and profitability could be expected to induce FCOJ producers to divert any productive capacity from FCOJ to COJ. The record evidence indicates that they could do so easily.

Cross-Elasticity of Demand Intuitively, one would expect demand for FCOJ and COJ to be highly correlated because of their highly similar taste and use characteristics. Most COJ is produced from bulk concentrate and therefore tastes exactly like the orange juice that consumers themselves prepare from frozen concentrate. Moreover, as the majority opinion apparently suggests, COJ made from fresh oranges and COJ made from bulk concentrate should both be included in the relevant product market, in part because consumers apparently perceive little or no difference in taste between them)O It therefore seems ilogical to treat COJ made from fresh oranges and COJ made from frozen concentrate as part of the same market while excluding frozen concentrate itself from that market. The only apparent differences between COJ and FCOJ relate to storage and preparation. COJ can be stored for twenty to thirty days in a refrigerator, while FCOJ occupies less space and can be stored indefinitely. In addition, COJ is ready to serve, while FCOJ must be thawed and diluted. These differences hardly seem suffcient to justify placing COJ and FCOJ in separate markets. (5) The pricing evidence in the record does not refute this intuitive judgment. As the majority opinion recognizes, FCOJ and COJ prices do not differ dramatically. As of August 1, 1979, the average nationwide price of a six ounce serving of FCOJ was only 14 percent lower than the average price of an equivalent serving of chiled orange juice packaged in cartons or plastic.l1 That represents a considerable re- 6 LD.F. 67. Tropicana, for example, makes 70% of its COJ directly from frut, and the remaining 30% from bulk concentrate- LD.F. 65. Similarly, Hood and Kraft make COJ from both fresh oranges and FCOJ. I.D.F. 66. 9 ex 594I,J,KO,R; Tr. 902-04, 1287-89, 1418- , 1543, 1668-71, 3899, 3904-7. 10 Majority opinion at 14 n. citing LD.F. 63-70. 11 (;"X fifi(r' (In nmprn). Over two-thirds of all COJ is sold in plastic containers or carons. ex (in560z--wmera). Concurring Opinion 101 F. duction from the 32 percent price disparity recorded in 1970.1 Moreover, during the 1971 - 1979 time period, the ratio oftotal ready-toserve orange juice sales to total frozen concentrate sales increased from less than 25 percent (in gallons) to over 50 percent (in gallons).3 This evidence is consistent with the inference that price competition between COJ and FCOJ has strengthened considerably since 1971.4 The record does not contain any evidence (6) on current FCOJ and COJ prices with which we could determine whether this trend has continued, but there is no reason to believe that it has weakened or ended.

One study submitted by complaint counsel (the Tilley study) provides more direct evidence on the issue of demand cross-elasticity. The study s principal objective was to determine the degree to which COJ and FCOJ consumption can be explained by habit and inventory effects. As a corollary to the effort, the study produced a number of demand cross-elasticity estimates. The Tiley study represents a professional and sophisticated statistical effort, and I should like to emphasize that well formulated statistical evidence can be of subs tantially greater importance in delineating relevant markets than the anecdotal type of "industry or public recognition" evidence suggested in Brown Shoe Co. v. United States.l5 However, the results of the Tiley study raise several concerns that limit its utility in determining whether COJ and FCOJ lie in the same or in separate markets. The first concern relates to the data upon which Dr. Tiley relied. The price and quantity data were based upon beverage purchase reports from a consumer panel (7) However, Douglas Hoffer, the Director of Marketing for the Florida Department of Citrus (and formerly its Director of Market Research), testified that he did not believe that the data were particularly reliable,1 Possibly as a consequence of his testimony, complaint counsel declined to introduce the data themselves into evidence as a basis for establishing prices and price trends. 18 Moreover, apparently as a result of these reliability problems, the Department of Citrus later stopped purchashing the '2 CX 560T (in camera).

CX560W (in camera).

!4 It is true t.hat the price of.ICOJ remains considerably lower-approximately 29 percent lower-than the price ofCOJ in glascontaioers. ex 560T (in mmera). However, 8S the majority opinion points out, the disparty between the price of COJ in glass containers and the price of COJ in plastic containers or carns is actually greater than the disparity between the latter and the price ofFCOJ. Neverthclc8S, the majority opinion treats COJ sold in glass containers and COJ sold in plastic containers or cartons as part of the same product market. Majority opinion at 11-14.

370 U.S. 294, 325 (1962).

'6 Tr. 1990 referring toCX400 through CX407. !' Tr. 1886-7, 1927. The Administrative Law Judge later characterized this opinion as "very persuasve." Tr. 222l.

IB Tr. 1888. Complaint counsel did offer the data for the purpose of showing that they were categorized on the basis of beverage type--rather than "for the truth ofthe price-trend data eontaitJed therein and the AdmDistrative Law Judge accepted them for that limited purpo!i only. Tr. 1881:9 .. . g., ..~~u'- ~ '~H' 733 Concurring Opinion data,19 The Tiley study results may therefore have been affected by inaccuracies in the data base.

Second, the study results may have been affected by multicollinearity problems, which can cause substantial errors in the regression estimates that a study produces.2o If several independent variables tend to move together, it is diffcult to determine whether the value ofa given coeffcient for anyone of those variables actually measures (8) the effect of that variable (e. the cross-elasticity of demand between FCOJ and COJ), or instead simply reflects the strong influence of other related independent variables. Dr. Tiley agreed that multicollnearity might have affected his regression estimates and that if he had limited the number of substitutes in his study to chiled, frozen, and canned juice, his study might have established a greater degree of substitutability among the three. Third, the study results may have been affected by serial correlation. The system of equations Dr. Tiley employed includes the quantity of each product consumed in the immediately preceding period as an independent variable. Since, for each time period, quantity serves as the dependent variable while prices serve as the independent variables, including the immediately preceding quantity as an additional independent variable makes each quantity observation a function not only of current prices but also of all previous price observations. This technique may introduce serial correlation into the dependent variable, further affecting the accuracy of the regression estimates. The study results (9) do indicate that both FCOJ and COJ consumption in any given period were strongly influenced by the quantity of each product consumed in the immediately preceding period. Apart from these systematic diffculties, the cross-elasticity coeffcients generated by the study are themselves not inconsistent with treating COJ and FCOJ as part ofthe same market. For example, in the first relevant equation, which regresses per capita consumption of COJ on the prices of FCOJ and other types of orange juices and orange drinks, the study found a cross-elasticity of demand of - 0633 between FCOJ and COJ. Because of its high standard error, this value is not significantly different from zero. Zero would be the expected value if the demands for the two products were completely independent from each other. However, the fact that the statistical Tr. 2382.

20 J. Kmenta Element. of Econometrics 380--91 (1971). 21 Tr. 2144--7.

22 ex 5141. The study also generated some anomalous results. For example, it indicates that the crosslasticity of demand between the price of canned single strength orange juice (CSSOJ) and per capita corummption of FCUJ is negative and significantly different from zero at the 90% level.ld. This suggests that CSSOJ and FCOJ are actually complements; that is, when the price of CSSOJ increass, FCOJ consllption declines. This result does not make parlcularly good sense, and suggests that the study may sufer from some specification problems. 23 CX514I.

Concurring Opinion 101 F. test failed to reject the null hypothesis does not allow one to conclude that the null hypothesis is true and that the (10) cross-elasticity of demand must be zero.24 In other words, the statistical finding is consistent with-but does not prove-the hypothesis that demand for the two products is independent.

In the second relevant equation, which regresses per capita consumption of FCOJ on the prices of COJ and other types of orange juices and drinks, the study found a cross-elasticity of 0.2267 between FCOJ and COJ. This result is significantly different from zero at the 90% confidence level26 Therefore, as Dr. Tiley testified, one can have fairly high level of confidence, above 90 percent, that COJ is in fact a substitute for FCOJ. (11) Moreover, the true value of the short-run cross-elasticity measure may actually be as high as 0.4346, and in any event no lower than 0188.

It is diffcult to determine precisely how to evaluate the Tiley study results. The cross-elasticity estimates the study produced are not particularly high, but the data reliability, multicollinearity, and serial correlation concerns described above limit the utility ofthe estimates. Moreover, the estimates themselves focus upon short-run effects, and although the study did not produce any long-run cross-elasticity estimates, it does indicate that the long-run price elasticities of demand for FCOJ (- 8589) and COJ (- 8474) are nearly identical, and considerably higher.29 These conflicting considerations suggest that on balance, the Tiley study-although a professional effort-cannot be relied upon to determine the degree to which FCOJ and COJ can be treated as substitutes. (12) By contrast, two New York studies submitted by the respondents suggest a substantial degree of substitutability or cross-elasticity between COJ and FCOJ. As the majority opinion points out, the New York studies are regional rather than national in scope. However See, e. R Wonnacott and T. Wonnacott Econometrics 64-6. (1970). 2. ex 514l.

l6 The absolute value of the ratio of the observed value (here, 0.2267) to its standard error (here, 0. 1225) determines the de ce to which one can be confident that the value of the cocf1cient actually differs from zero When the number of observations exceeds 30, as is the case here, then a ratio of J .697 pennits one to be 90% confident that the value actually differs from zero. Since the ratio or the observed value (0. 2267) to its standard error (0- 1225) exceeds 1.697 , one can be over 90% sure that the actual value ofthe variable is positive and different from zero I Tr. 2384.

!8These figures represent the bounds ofthe 90% confidence interval. That interval can be created by respectively adding and subtracting the product of the standard error (here, 0. 1225) and 1.697 to and from the ob rved value (here, Q.2267) 29 ex 514K. The "long run" for FCOJ and for eOJ were estimated to be 8 months add 5 months, respectively 30 Because the New York studies were pn' pared by the same firm that colleCled the data that Dr. Tilley relied upon and, presumably, relied upon the same type of data, their utilty may be similarly limited. /;.

BEATRICE FOODS CQ" ET AL. 835 733 Concurring Opinion since COJ accounts for a greater proportion of orange juice sales in New York than in other parts of the country, one might actually expect the New York studies to show a lower degree of substitutability of FCOJ for COJ there than elsewhere. That did not prove to be the case. One of the studies found a substantial level of substitution between Minute Maid and Tropicana COJ, on the one hand, and a variety of brands of FCOJ, on the other.3! The second studyindicated that Minute Maid FCOJ buyers devote 27 percent of their purchases (in fluid ounces) to brands of COJ; that Minute Maid COJ buyers devote 39 percent oftheir purchases to brands ofFCOJ; that Tropicana FCOJ buyers devote 23 percent of their purchases to brands of COJ; and that Tropicana COJ buyers devote 45 percent of their purchases to brands of FCOJ. In short, the New York studies suggest that there may be a substantial degree of substitutability between FCOJ and COJ. (13) Industry Perceptions The record evidence concerning industry perceptions appears to be rather inconclusive. As the majority opinion recognizes, industry perceptions can sometimes help to delineate the relevant product and geographic markets. However, in this case the perceptions of different industry participants conflict substantially,33 and it is diffcult to determine which should be relied upon and which should be ignored. Moreover, the fact that some industry firms consider COJ and FCOJ to differ from one another to some degree does not mean that they should be treated as separate products. The crucial issue is whether the price of one could be increased significantly without inducing increases in the quantities of the other that are demanded and supplied. The best way to resolve conflicting industry perceptions is to consider the more objective and reliable evidence provided by crosselasticities of supply and demand.

Conclusion The record evidence indicates that the cross-elasticity of supply between COJ and FCOJ is relatively high, and that the cross-elasticity of demand between the two versions of orange juice may be at least somewhat significant. As a result, any effort to raise COJ prices to a sigoificant degree could be expected to induce FCOJ producers to increase (14) COJ production substantially, and to induce at least some COJ consumers to switch to FCOJ. I would therefore treat FCOJ and COJ as part of the same product market. 31 RX I-p, l-q.

32 RX 1-391 :11 Compare, e. ex 6C, ex 147, ex 149B-C with Tr. 4328, 4479, 4877 , 5107, RX 120-T. . . .

836 FEDERAL TRADE COMMJSSION DECISIONS Concurring Opinion 101 F. Relevant Geographic Market The majority opinion concludes that the relevant geographic market in this case is the nation as a whole. However, the opinion also states that "it is quite possible that regional markets could also exist which would reflect, perhaps more precisely, the commercial realities of the COJ industry and the competitive effects of the merger " and that "regional shipping restrictions. . . strongly suggest the existence of regional markets as well. . . 34 Together, these statements suggest that the relevant markets in this case can be both regional and national simultaneously. I disagree with this approach. I believe that carefully delineating the single most accurate geographic market possible should be a prerequisite to evaluating the competitive effects of any given merger. I would therefore simply conclude that in this case, the nation as a whole represents the most competitively significant market within which to assess the effects of the acquisition at issue.

The relevant geographic market can be delineated most accurately by calculating the degree to which-within an appropriate period of time-price changes in a given area (15) will induce changes in the quantities of the relevant product supplied from and demanded in other areas. This suggests that geographic markets-like product markets-can be visualized as a series of concentric circles radiating outward. As the diameter of the circle expands, the cross-elasticities of demand and supply between the relevant product within the circle and the relevant product outside the circle can be expected to decline. In this case, the record evidence establishes that the three largest producers, Tropicana, Minute Maid, and Kraft, all sell their products on a nationwide basis, and monitor and respond to prices and other competitive conditions that prevail in different parts of the country. Moreover, they compete directly with large numbers of local and regional producers. Furthermore, although transportation costs apparently represent a significant proportion of product value, they do not regionalize competition to any significant extent. Approximately 90 percent of the product-in the form of fresh oranges, chilled juice or bulk concentrate-must be shipped from Florida.35 It is therefore highly unlikely that any national, regional, or local producer would have a significant freight advantage over any other.36 (16) As a result of these market characteristics, any effort to increase prices significantly above the competitive level in a given part of the country can 34 Majority opinion at 32, 34.

3. The remainio.g ten percent comes from the Southwest, Brazil, and Mexico- Majority opinion at 4. 36 Since bulk concentrate is more concentrated than fresh oranges or chiled juice, it may be mOfe economical to ship on a per pound basis. IIowevtJf, it must be kept frozen, unlike fresh oranges or chiled juice, and that may conversely increase it. shipping costs relative to the other two product forms 733 Concurring Opinion be expected to induce increased snipmentg of the product from other parts of the country relatively quickly. I would therefore concluile that the relevant geographic market is the nation as a whole, and dispense with the suggestions in the majority opinion that the relevant geographic markets in this case may be both national and regional simultaneously.

Competitive Effects Ifthe product market is defined to include both chiled orange juice and frozen concentrate, and ifthe geographic market is defined to be nationwide, then it is highly unlikely that Beatrice s acquisition of Tropicana had any significant anticompetitive effects. Market share data strongly support this conclusion. In 1978, Tropicana and Beatrice respectively accounted for approximately 11 percent and 0.0016 percent of sales in the relevant market. These shares are considerably lower than the shares that Tropicana and Beatrice respectively accounted for in the (17) chiled orange juice sector alone, according to the majority opinion.38 Industrywde concentration levels in the larger FCOJ-COJ market probably are also considerably lower, because Tropicana s share of sales in the larger market is only one-third as large as its share in the COJ sector alone.39 The majority opinion concludes that-in conjunction with other evidence-the market share and concentration levels in the COJ sector establish that the merger is unlikely to have any anticompetitive effects. The much lower share and concentration levels in the more relevant product market that includes both COJ and FCOJ make it substantially less likely that the merger would have any anticompetitive effects. Other relevant economic factors strengthen this determination even further. As the majority opinion points out, small-scale entry into the chiled orange juice sector is both easy and common.40 I would also conclude-unlike the majority opinion-that large-scale entry into the larger, more relevant market would not be particularly difficult. Any dairy-including the very large regional dairies with wellestablished trademarks located throughout the country-can quickly and inexpensively shift from (18) fluid milk production to chiled orange juice production, because their fluid milk plants already possess all the production and distribution equipJIent needed to produce / TheBe estimates ate D&essarly approximate; they were derived as follows. In (hlcal 1978, Tropicana. and Beatrice respectively accounted for $166.8 milion (I.D.F. 123) and $2.4 millon (Majority opinion at 4) in sales of chilled orange juice to the retail sector. Tota chilled orange juice sales in that sector amoLlIlted to $545.2 million. F. 121. Tota! non-institutional sales of FCOJ amounted to approximately $1.05 billon in flSlaJ 1978.Federal Trae Commission. v. Beatrice Food Co. 587 F.2d 1225, 1233 (D.C. Cir. 1978). Tota non-institutionnl sales ofFCOJ and 001 therefore together amounted to approximately $1-5 hilion, yielding the market shares cited above. :J Majority opinion at 43 The record dons not indudc market share data for any of the four largest firms in the FCOJ-COJ market except Tropicana (which apparently makes almost no FCO sales). 40 Majority opinion at 46.

Concurring Opinion 101 F. chiled orange juice from bulk concentrate.4! Large-scale new entry into the industry therefore would probably not be particularly difficult. In fact, a number of strong regional firms have recently entered the orange juice industry on a substantial scale. Conclusion The record evidence establishes that the merger would be unlikely to have any anticompetitive effects in the nationwide market for chiled and frozen concentrated orange juice. I therefore concur that the complaint in this matter should be dismissed. I would like to emphasize that my conclusion in this regard would be the same if I were to accept the narrower product market definition in the majority opinion. Moreover, it would almost certainly be the same if we were today considering a complaint proposal rather than a final disposition. It is interesting to note that the complaint in this matter alleged almost precisely the same facts that the majority opinion now relies upon to dismiss the proceeding. In particular, the complaint alleged that (19) the relevant product and geographic markets consisted of the processing, distribution and sale of ready to serve orange juice . . . . (in) the United States as a whole and submarkets thereof. . . . The complaint also alleged that the four largest firms in the market accounted for 58.6% of total sales in that market in 1976, that Tropicana and Beatrice respectively accounted for 29% and 1% - 1.7% of total unit volume (in gallons) in 1976, and that Tropicana share was nearly twice that of its nearest competitor.43 The Commission relied upon these allegations to issue the complaint in this matter. The majority opinion now relies upon essentially the same market definitions, determines that concentration levels and the market shares of Tropic ana and Beatrice within those alleged markets are roughly the same, and nevertheless concludes that the acquisition did not injure competition. This change in perspective ilustrates the Commission s growing understanding that economic characteristics other than market shares and concentration levels are crucially important to determining whether any particular acquisition is likely to injure competition. I welcome the majority opinion s recognition of that fact, and I look forward to applying these principles both to future cases and to future proposals for complaints alleging violations of Section 7 of the Clayton Act.

.1 I.D.F. 135. The only conceivable constraint to shift ofthis sorl WOQld be access to bulk fro en conc€utra.te. 42 These firms inc!llde Johana Farms on the Atlantic COBst, the Ohio Pure Foods Co. and the Home Juice Co. in the Midwest, and Knudsen Dairy on the West Coast. Tr. 994, 1018-19, 10290, 1094, 1591, 1603, 1607, 4053--6. Beatrice Foods Coo Docket No. 9112 (Complaint) (Issued on June 29, 1978), at 3. .

733 Final Order FINAL ORDER This matter has been heard by the Commission upon the appeal of respondents from the initial decision and upon briefs and oral argument in support of and in opposition to their appeal. For the reasons stated in the accompanying Opinion, the Commission has determined to sustain respondents' appeal. All motions which have not yet been disposed of in the accompanying Opinion or by prior orders of the Commission are denied. Accordingly, It is ordered That the complaint is dismissed. Complaint 101 FTC.