Coca-Cola Company

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

In THE MATTER OF THE COCA-COLA COMPANY, ET AL.

ORDER, OPINION, ETC., IN REGARD TO ALLEGED VIOLATION OF SECTION 5 OF THE FEDERAL TRADE COMMISSION ACT Docket 8855. Complaint, July 15, 1971 — Final Order, April 7, 1978 This order, among other things, requires an Atlanta, Ga. soft drink manufacturer and three parent bottlers to cease imposing in any manner territorial limitations or class of customer restrictions on its licensed Coca-Cola or allied product bottlers, in connection with the sale or distribution of soft drink products sold in other than refillable containers. The firm is additionally required to provide protection for confidential information submitted by its bottlers. Appearances For the Commission: Raymond L. Hays, Martin A. Rosen, Duncan J. Farmer, Michael J. Bloom and Jeffrey F. Shaw. For the respondents: Gordon B. Spivack, Harry G. Sklarsky and Thomas D. Brislin, Lord, Day & Lord, New York City. J. Guy Beatty, Jr., Miller, Martin, Hitching, Tipton, Lenihan & Waterhouse, Chattanooga, Tenn. for Coca-Cola Bottling Co. (Thomas), Inc. and Coca-Cola Bottling Works (Thomas), Inc. For the intervenors: Willis B. Snell, William M. Hames, Michael L. Denger, Stephen S. Cowen and Robert W. Clark, III, Sutherland, Astill & Brennan, Washington, D.C. for The Coca-Cola Bottlers’ Association, The Scioto Coca-Cola Bottling Company, Roddy Manufacturing Company, Elberton Coca-Cola Bottling Company, Westminster Coca- Cola Bottling Company, Inc. and Texas Coca-Cola Bottling Company. Thomas Matthews, Wald, Harkrader & Ross, Washington, D.C. for Coca-Cola Bottling Company of Los Angeles, Quaker State Coca-Cola Bottling Co. and Detroit Coca-Cola Bottling Company. Bartlett H. McGuire, Davis, Polk & Wardwell, New York City for The Coca-Cola Bottling Company of New York. Theodore Kleinman, Zuckert, Scoutt & Rasenberger, Washington, D.C. for Associated Coca-Cola Bottling Co.

John P. Gaither, Witt, Gaither, Richardson, Henniss & Whitaker, Chattanooga, Tenn. for Consolidated Coca-Cola Bottling Co. Robert N. Nanovic, Nanovic & McKinley, Jim Thorpe, Pa. for Coatesville Coca-Cola Bottling Works, Inc.

William H. Sanders, Blackwell, Sanders, Matheny, Weary & Lombardi, Kansas City, Mo. for The Coca-Cola Bottling Company of Mid-America.

Complaint 91 F.TE..

Joe A. Walters, O’Conner & Hannan, Minneapolis, Minn. for Coca- Cola Bottling Midwest, Inc.

COMPLAINT The Federal Trade Commission, having reason to believe that the parties named in the caption hereof, each of which is hereby. made and is sometimes hereinafter referred to as respondent(s), have violated the provisions of Section 5 of the Federal Trade Commission Act, and it appearing to the Commission that a proceeding by it in respect thereof would be in the public interest, hereby issues its complaint, stating its charges in this respect as follows:

PARAGRAPH 1. For the purposes of this complaint, the following definitions shall apply:

(a) Bottler - any individual, partnership, corporation, association, or other business or legal entity which purchases respondents’ concentrate for use in the manufacture and sale, primarily at wholesale, of respondents’ pre-mix or post-mix syrups or soft drink products, or who purchases respondents’ pre-mix or post-mix syrups or soft drink products for resale, primarily at wholesale; (b) Central warehousing - a method of distribution in which soft drink products are received at a storage facility and either resold or delivered to retail outlets or wholesalers; (c) Concentrate - the basic soft drink ingredient sold to bottlers by respondent, usually as a syrup, and which is combined with water and other ingredients for packaging in bottles or cans for sale and distribution as soft drink products, or is used to make post-mix and pre-mix syrups; [2] (d) Consignment - a form of distribution in which the consignor retains title, dominion, bears all risks of loss and delivers his products to the consignee who is indistinguishable from a salesman or agent, (e) Place of business ~ the location of any facilities available to a bottler without regard to customers or geographic area for production or service in the conduct of business operations, to include but not limited to business headquarters, branch sales offices, warehouses and garages, but specifically excluding the plant at which a bottler combines concentrate with water, and possibly other ingredients, for the packaging of soft drink products;

(f) Post-mix syrup — soft drink concentrate which is used in fountain dispensing or vending equipment and is usually sold by bottlers in steel tanks. A typical post-mix system draws one ounce of syrup from a tank, usually having about a five-gallon capacity, and mixes it at the point of sale with five ounces of carbonated water to produce approximately 600 six-ounce finished soft drink servings per tank; THE COCA-COLA CO., ET AL. 519 517 Complaint (g) Pre-mix syrup — although essentially the same syrup as post-mix, a pre-mix system differs from a post-mix system in that it draws from a tank, usually having about a five-gallon capacity, a finished serving of soft drink product containing both syrup and carbonated water, “pre-mixed,” to produce a 100 six-ounce soft drink servings per tank; and (h) Soft drink products — nonalcoholic beverages and colas, carbonated and uncarbonated, flavored and non-flavored, sold in bottles and cans, or through pre-mix and post-mix systems or the like. Par. 2. Respondent The Coca-Cola Company, sometimes hereinafter referred to as Coca-Cola, is a corporation organized, existing and conducting its business under and pursuant to the laws of the State of Delaware. It maintains its office and principal place of business at 310 North Ave., N.W., Atlanta, Georgia. The Coca-Cola Company and subsidiaries had net sales of $1,185,808,864 (approximately 45 percent of which is accountable to foreign operations), and assets of $802,100,548 in 1968. In 1968, Coca-Cola made sales to over 900 domestic bottlers located throughout the United States. [3] Respondent Coca-Cola Bottling Co. (Thomas), Inc., sometimes hereinafter referred to as Thomas Company, is a corporation organized, existing and conducting its business under and pursuant to the laws of the State of Delaware. It maintains its office and principal place of business at 1600 American Bank Building, Chattanooga, Tennessee. In 1968, Thomas Company made sales to over 196 bottlers located principally in Indiana, Maryland, Mississippi, New Jersey, New York, Ohio, Pennsylvania, Virginia and West Virginia. Respondent Coca-Cola Bottling Works (Thomas), Inc., sometimes hereinafter referred to as Thomas Works, is a corporation organized, existing and conducting its business under and pursuant to the laws of the State of Delaware. It maintains its office and principal place of business at 1600 American Bank Building, Chattanooga, Tennessee. In 1968, Thomas Works made sales to over 65 bottlers located principally in the States of Kentucky and Tennessee.

Respondent Coca-Cola Bottling Works 8rd, Inc., sometimes hereinafter referred to as Works 3rd, is a corporation organized, existing and conducting its business under and pursuant to the laws of the State of Delaware. It maintains its office and principal place of business at 1600 American Bank Building, Chattanooga, Tennessee. In 1968, Works 3rd made sales to over 25 bottlers located principally in the States of Pennsylvania and New Jersey.

Par. 8. Respondent Coca-Cola, through its Coca-Cola U.S.A. division, is engaged principally in the manufacture and sale of soft drink products and concentrate which it sells to over 900 bottlers who Complaint 91 F.T.Co purchase the concentrate under a license to produce and sell soft drink products under such trade names of respondent Coca-Cola as “Coca- Cola” (“Coke”), “TAB,” “Sprite,” “Fresca,” “Fanta” and “Simba.” Bottlers combine the concentrate with water and other ingredients and package the mixture in bottles and cans for resale as soft drink products to retailers. In addition to manufacturing and selling soft drink products and concentrate to its bottlers, Coca-Cola operates bottling plants in 27 areas of the United States and sells soft drink products to retailers. [4] Respondent Thomas Company has operated for many years as a parent bottler under an agreement with Coca-Cola by which Thomas Company was granted certain rights from Coca-Cola with respect to the sale of Coca-Cola soft drink products in certain designated territories. Thomas Company is engaged principally in the purchase of concentrate from Coca-Cola for resale by Thomas Company to numerous bottlers which have obtained licenses from it to bottle and resell certain specified trade name soft drink products of Coca-Cola. Respondent Thomas Works has operated for many years as a parent bottler under an agreement with Coca-Cola by which Thomas Works was granted certain rights from Coca-Cola with respect to the sale of Coca-Cola soft drink products in certain designated territories. Thomas Works is engaged principally in the purchase of concentrate from Coca-Cola for resale by Thomas Company to numerous bottlers which have obtained licenses from it to bottle and resell certain specified trade name soft drink products of Coca-Cola. Respondent Works 8rd has operated for many years as.a parent bottler under an agreement with Coca-Cola by which Works 3rd was granted certain rights from Coca-Cola with respect to the sale of Coca- Cola soft drink products in certain designated territories. Works 3rd is engaged principally in the purchase of concentrate from Coca-Cola for resale by Works 3rd to numerous bottlers which have obtained licenses from it to bottle and resell certain specified trade name soft drink products of Coca-Cola.

Par. 4. Respondents are engaged in “commerce” within the meaning of the Federal Trade Commission Act (15 U.S.C. 44) in that a continuous flow of interstate commerce in concentrate and soft drink products exists between their headquarters and production facilities and the numerous bottlers located throughout the United States which purchase their products.

Par. 5. In the course and conduct of their businesses, respondents, except to the extent limited by the acts, practices and methods of competition hereinafter alleged, have been and are now in competition with other corporations, firms, partnerships and persons engaged in THE COCA-COLA CO., ET AL. 521 517 . Complaint the manufacture, processing, distribution and sale of concentrate and soft drink products in commerce.

Par. 6. Respondents have hindered, frustrated, lessened and eliminated competition in the distribution and sale of pre-mix and postmix syrups and soft drink products [5] sold under their trade names by restricting their bottlers from selling outside of a designated geographical area. This restriction is set forth in the agreement between respondents and their bottlers.

A typical license between respondent Coca-Cola and its bottlers provides that as to a specifically described geographic territory: ... COMPANY agrees to furnish to BOTTLER, and only to furnish for the territory herein referred to, sufficient syrup for bottling purposes to meet the requirements of BOTTLER in the territory herein described.

* * * * * * * ... COMPANY does hereby select BOTTLER as its sole and exclusive customer and licensee for the purpose of bottling the Bottlers’ bottle syrup, COCA-COLA, in the territory described.

[BOTTLER agrees]. . . not to use trade-marks COCA-COLA or COKE, nor bottle nor vend said product except in the territory herein referred to. This limitation, however, is not to prevent BOTTLER from acquiring similar rights for other territory. * * * * * * * [BOTTLER agrees]. . . not to use said distinctive [COCA-COLA] bottle for any other purpose than the bottling of COCA-COLA, and not in any territory except as herein referred to.

A typical license between respondents Coca-Cola Bottling Co. (Thomas), Inc. and Coca-Cola Bottling Works (Thomas), Inc. and the bottlers of each provides in part that licensor, wishing to assign to the bottler certain rights as to a specifically described geographic territory which has been received by approved transfer from The Coca-Cola Company, agrees:

. .. to obtain and furnish to party of the second part [bottler] and only to obtain, for the territory herein referred to, sufficient syrup for bottling purposes to meet the requirements of party of the second part in the territory herein described, provided [6] party of the first [licensor] can obtain the delivery to it of such syrup from The Coca- Cola Company under the contract existing between party of the first part and The Coca- Cola Company.

* * * * * * * [To select bottler]. . . as its sole and exclusive customer and licensee for the purpose of bottling Bottlers’ Coca-Cola syrup, and using the name Coca-Cola thereon in the territory herein described.

Complaint 91 F.T.C.

In consideration therefor, bottler agrees:

- Not to use the name Coca-Cola nor bottle nor vend said product except in the territory herein referred to without the written consent of party of the first part and The Coca-Cola Company. This limitation, however, is not to prevent party to the second part from obtaining such rights from parties authorized to use the name Coca-Cola and to bottle and vend said product.

* * * * * * * ... To order, for the purpose of bottling Coca-Cola, the distinctive bottle, and none other, adopted or that may be adopted by party of the first part; to use said distinctive bottle and none other, in bottling Coca-Cola, and not to use said distinctive bottle for any other purpose than the bottling of Coca-Cola, and not in any territory except as herein referred to without the written consent of party to the first part and The Coca-Cola Company.

The license restrictions between Coca-Cola Bottling Works 3rd, Inc. and its bottlers are substantially similar to that of Coca-Cola, Coca- Cola Bottling Co. (Thomas), Inc. and Coca-Cola Bottling Works (Thomas) Inc. Coca-Cola is a party to the agreement between Coca- Cola Bottling Co. (Thomas), Inc., Coca-Cola Bottling Works (Thomas), Ine. and Coca-Cola Bottling Works 8rd, Inc. and their bottlers. Par. 7. The aforesaid agreements used by respondents have had, and may continue to have, the following effects: (a) Competition between and among respondents’ bottlers in the distribution and sale of “Coca-Cola” (“Coke”), “TAB,” “Sprite,” “Fresca,” “Fanta” and “Simba” brands of soft drink products has been eliminated; [7] (b) Competition between and among Coca-Cola’s bottling operations and its bottlers in the distribution and sale of Coca-Cola soft drink products at the wholesale level has been eliminated; (c) Competition between and among Coca-Cola’s bottling operations and bottlers licensed by Coca-Cola Bottling Co. (Thomas), Inc., Coca- Cola Bottling Works (Thomas), Inc. and Coca-Cola Bottling Works 8rd, Inc. in the sale and distribution of Coca-Cola’s soft drink products at the wholesale level has been eliminated;

(d) Competition between and among bottlers licensed by Coca-Cola and bottlers licensed by Coca-Cola Bottling Co. (Thomas), Inc., Coca- Cola Bottling Works (Thomas), Inc. and Coca-Cola Bottling Works 8rd, Inc. in the sale and distribution of Coca-Cola soft drink products at the wholesale level has been eliminated;

(e) Innumerable retailers and other customers have been deprived of the right to purchase “Coca-Cola” (“Coke”), “TAB,” “Sprite,” “Fresca,” “Fanta” and “Simba” brands of soft drink products from the bottler of their choice at competitive prices; and THE COCA-COLA CO., ET AL. 523 517 Initial Decision (f) Consumers of “Coca-Cola” (“Coke”), “TAB,” “Sprite,” “Fresca,” “Fanta,” and “Simba” brands of soft drink products have been deprived of the opportunity of obtaining such products in an unrestricted market and at competitive prices. Par. 8. Respondents’ contracts, agreements, acts, practices and methods of competition aforesaid have had, and may continue to have, the effect of lessening competition in the advertising, merchandising, distribution, offering for sale and sale of pre-mix and post-mix syrups and soft drink products; deprive, and may continue to deprive, the public of the benefits of competition in the purchase of pre-mix, postmix and soft drink products; and constitute unfair methods of competition and unfair acts or practices, in commerce, in violation of Section 5 of the Federal Trade Commission Act. Initial Decision By Joseph P. DUFRESNE, ADMINISTRATIVE LAw JUDGE October 8, 1975 PRELIMINARY STATEMENT In a complaint dated July 15, 1971, the Commission charged respondents with violation of Section 5 of the Federal Trade Commission Act (15 U.S.C. 45). The crux of the charges was that the territorial exclusivity provisions in trademark licensing contracts The Coca-Cola Company (hereinafter “Coca-Cola”) enters with its bottlers impose an illegal restraint on competition. The provision limits the geographical territory in which a bottler may manufacture and sell “Coca-Cola” products. .

It was alleged that the illegal effects of the provision were that competition had been eliminated:

(a) between independent bottlers of “Coca-Cola” products; (b) at the wholesale level between independent bottlers of “Coca- Cola” products and “Coca-Cola’s” own bottlers; (c) at the wholesale level between “Coca-Cola’s” own bottling operations and bottlers licensed by the other respondents; and (d) at the wholesale level between bottlers licensed by “Coca-Cola” and those licensed by the other respondents. It also was alleged that retailers and other customers had been deprived of the right to purchase “Coca-Cola” products from the bottler of their choice at competitive prices. Lastly, it was alleged that consumers of “Coca-Cola” products have been deprived of the opportunity of obtaining “Coca-Cola” products in an unrestricted market and at competitive prices. [4] Initial Decision 91 F.T.C.

In their answers, “Coca-Cola” and the three other respondents (Coca-Cola Bottling Co. (Thomas), Inc. (hereinafter Thomas Company), Coca-Cola Bottling Works (Thomas), Inc. (hereinafter Thomas Works), and Coca-Cola Bottling Works 8rd, Inc. (hereinafter Works 3rd)) admitted that they are “in commerce” within the meaning of the Federal Trade Commission Act. They. denied that the territorial exclusivity provisions have any illegal anticompetitive effects. As an affirmative defense, respondents stated that in The Coca-Cola Bottling Co. v. The Coca-Cola Co., 269 F. 796 (D. Del. 1920), the District Court held in an opinion which has not been rescinded, vacated or. modified, that the restraints included in the contracts between respondents and the bottlers are lawful and valid under federal and state antitrust laws and not in unreasonable restraint of trade. They also asserted the defense that the Commission had failed to include indispensable parties, the bottlers, who were the other parties to the contracts, and that to the extent the Commission’s proceeding sought to eliminate or abridge the respondents’ and bottlers’ contractual property rights, the proceeding would impose a penalty or forfeiture barred by the statute of limitations (28 U.S.C. 2462). Lastly, in summary, they asserted (1) that a Commission order would place them in the position of violating either that order or the District Court’s order, depending on which they complied with; (2) that the effect of a Commission order would be unfair to consumers and small businessmen bottlers and anticémpetitive by increasing prices and accelerating the demise of small grocers; (8) that respondents’ constitutional right to due process would be violated; and (4) that there is no public interest in the proceeding.

Because of the time which has elapsed since the complaint issued, it is appropriate to set forth a listing of significant events which have taken place in this proceeding since July 15, 1971. These were: [5] 1. On July 30, 1971, a Motion for Consolidation of this case with Dkts. 8853-8859, which pertain to exclusive territory provisions in the bottler licensing contracts used for other national brands of soft drinks, was filed by complaint counsel. That Motion was denied by the administrative law judge on September 28, 1971. 2. On August 27, 1971, September 8, 1971, and December 14, 1971, motions to dismiss the complaint for non-joinder of indispensable parties (i.e., all of the licensee bottlers) were filed by “Coca-Cola,” the other respondents, and the intervenors participating in the proceedings at that time. These motions were denied by the administrative law judge on January 7 and 12, 1972.

3. On April 21, 1972, the respondents filed a complaint for injunctive relief and a declaratory judgment against the Commission THE COCA-COLA CO., ET AL. 525 517 Initial Decision in the United States District Court for the Northern District of Georgia. The Court dismissed the complaint on May 22, 1972. That dismissal was appealed to the United States Court of Appeals for the Fifth Circuit on June 6, 1972. The appellate court affirmed the decision of the District Court on February 20, 1973. On May 8, 1978, respondents sought certiorari to the Supreme Court; however, the Court denied the writ on October 9, 1973 (Coca-Cola Co. v. Federal Trade Commission, 342 F. Supp. 670 (N.D. Ga. 1972), aff'd, 475 F.2d 299 (5th Cir. 1978), cert. denied, 414 U.S. 877 (1978)). 4, On July 31, 1972, former complaint counsel filed a motion for partial summary decision. That motion was denied by the administrative law judge on April 5, 1978.

5. On March 15, 1974, intervenors Association, Scioto, Elberton, Roddy, Westminster, and Texas filed a motion to require complaint counsel to file an environmental impact statement and to stay the proceedings until such a statement is filed. The administrative law judge denied that motion on April 2, 1974. The intervenors then filed a motion to reconsider or to certify the motion to the Commission on April 10, 1974. That motion was denied by the administrative law judge on May 30, 1974. [6] 6. On July 30, 1974, complaint counsel filed a motion to authorize disclosure of documents containing nonpublic business information to be used at the hearings. On October 3, 1974, the administrative law judge authorized the disclosure.

7. On August 12, 1974, respondents and intervenors filed a motion for issuance of a subpoena duces tecum to the Secretary of the Commission to have him produce materials in the Commission’s nonpublic files. Briefs in support and opposition were filed on December 9 and 10, 1974. The subpoena was quashed by order of the Commission dated March 4, 1975.

It also should be pointed out that in addition to the intervenors identified above a number of petitions to intervene were filed by a variety of other, small and large Coca-Cola bottlers beginning on September 8, 1971, with the last filed on March 11, 1975. All such petitions were granted. As a result, there were 14 individual bottler intervenors plus the Coca-Cola Bottlers’ Association. In early 1971, 99 percent of the domestic bottlers of Coca-Cola, accounting for about 815 productive facilities, were members of the Association. The adjudicative hearings were held in Washington, D.C., from May 5 through June 11, 1975. The record was closed for the reception of evidence on July 7, 1975.

The findings of fact following are based on a review of the allegations made in the complaint, respondents’ answers, stipulations Initial Decision 91 FIC.

entered by counsel, the evidentiary record and upon a reading of the transcript record of the testimony and consideration of the demeanor of the witnesses at the hearings. In addition, the proposed findings of fact, conclusions and orders, together with reasons and briefs in support thereof filed by both sides have been given careful consideration. To the extent not adopted by this decision in the form proposed or in substance, they are rejected as not supported by the record or as immaterial. [7] For the convenience of the Commission and other readers of this initial decision, the findings of fact include references to supporting evidentiary items in the record. Such references are intended to serve as guides to the testimony, evidence and exhibits supporting the findings of fact. They do not necessarily represent complete summaries of the evidence considered in arriving at such findings. The following abbreviations have been used:

CX - Commission’s Exhibit, followed by number of exhibit being referenced.

RX — Respondents’ Exhibit, followed by number of exhibit being referenced.

Tr. — Transcript preceded by the name of the witness and followed by the page number.

FinpinGcs OF Fact The Coca-Cola Company 1. In 1886 Mr. John S. Pemberton invented a formula for a soft drink syrup and first prepared and sold a soft drink made pursuant to that formula under the trademark Coca-Cola (Stipulation No. 1, RX 1A, Tr. 555; Smith Tr. 677).

2. In 1891 Mr. Asa G. Candler acquired Mr. Pemberton’s rights to the secret formula for Coca-Cola and to the trademark Coca-Cola. In 1892 Mr. Candler formed The Coca-Cola Company, a Georgia corporation, predecessor in interest to the respondent “Coca-Cola” (Smith, Tr. 678; Stipulation No. 1, RX 1A-B, Tr. 555). [8] 3. Respondent, “Coca-Cola,” is a corporation organized, existing and conducting its business under and pursuant to the laws of the State of Delaware. “Coca-Cola” maintains an office and its principal place of business at 310 North Ave., NW., Atlanta, Georgia. “Coca- Cola” and its subsidiaries in 1968 had consolidated net sales of $1,185,808,864 and consolidated assets of $802,100,548. Approximately one-third and more of the sales are accountable to foreign operations. In 1968 “Coca-Cola” sold or shipped syrup or concentrates. to THE COCA-COLA CO., ET AL. 527 517 Initial Decision approximately 900 domestic bottlers located throughout the United States. (Admitted in Answer; Stipulation No. 2, CX 1248A-B, Tr: 465.) When the complaint issued in July 1971, there were 726 bottlers operating 804 bottling plants (Smith, Tr. 687). Respondents’ Business 4. “Coca-Cola,” through its Coca-Cola U.S.A. division, is engaged principally in the manufacture and sale of syrup (a mixture of ingredients in liquid form which, when properly mixed with pure carbonated water, becomes the finished soft drink product (Smith, Tr. 569)) and concentrates (a mixture of basic ingredients to which sugar must be added by the bottler to prepare a syrup (Smith, Tr. 570)). The syrup and concentrate are used in the processing and sale of soft drinks under one or more of the trade names Coca-Cola or Coke, TAB, Sprite, Fresca, Fanta, Simba, Santiba, and Mr. PiBB. (Admitted in Answer; Smith, Tr. 584-586.) 5. Coca-Cola is a cola type drink. TAB is a low calorie cola drink. Fresca is a citrus base, low calorie drink. Sprite is a lemon-lime drink. Fanta is a trademark which is applied to a line of flavored drinks including orange, root beer, gingerale, and strawberry. Simba is no longer on the market. Santiba is a trademark which is applied to a line of mixers such as carbonated water, gingerale and Tom Collins mix. Mr. PiBB, a relatively new product, is a pepper type drink. Products other than Coca-Cola are referred to as “allied products” (Smith, Tr. 584-586). [9] 6. Beginning in 1955, “Coca-Cola” from time to time authorized its licensed bottlers to bottle and sell Coca-Cola and allied products (as they were introduced) in an increasingly wide variety of sizes and types of containers. Coca-Cola and allied products currently are offered in returnable bottles, non-returnable bottles, and cans. Package sizes include 6 1/2-ounce, 10-ounce, 12-ounce, 16-ounce, 32-ounce, 48-ounce, and 64-ounce containers. It is being test marketed in a 72-ounce size (Smith, Tr. 714).

7. “Coca-Cola’s” Foods Division manufactures other beverage products, including Maryland Club coffee, Hi-C fruit drinks, Minute Maid orange juice, Snow Crop orange juice, and other citrus products (Smith, Tr. 578, 580). Its Temco subsidiary produces private label spray and freeze dried instant coffee (Smith, Tr. 578) and its Aqua-Chem, Inc., subsidiary manufactures steam boilers and water purification equipment (Smith, Tr. 578).

8. In July 1971, “Coca-Cola” had 8 subsidiaries which operated 27 bottling plants in various cities of the United States and sold soft drink products to retailers in the areas thereof (Smith, Tr. 687-688). Initial Decision 91 F.T.C.

9. The bottler subsidiaries of “Coca-Cola” today are: The Coca-Cola Company of New England (Dedham, Mass.); The Coca-Cola Bottling Company of Baltimore (Maryland); The Coca-Cola Bottling Company of Ohio (Columbus, Ohio); The Coca-Cola Bottling Company of Michigan (Lansing, Mich.); The Coca-Cola Bottling Company of Wisconsin (Milwaukee, Wis.); The Coca-Cola Bottling Company of Chicago (Illinois); The Coca-Cola Bottling Company of California (San Francisco); and The Pacific Coca-Cola Bottling Company (Bellevue, Wash.). In 1971, these companies served areas containing approximately 14 percent of the population of the United States (Ogden, Tr. 828, 844). [10] 10. Most of the subsidiary bottling plants were acquired years ago to assure the availability of Coca-Cola in territories when bottlers in those territories had to sell out or did not have the financial resources and independent firms with money and know-how were unavailable in those areas (Susong, Tr. 918, 922). The policy of the company for over a decade has been that such a subsidiary bottler is to be sold when the company acquires one so that the number of subsidiaries remains stable (Susong, Tr. 914, 921).

11. Another subsidiary, Canners of Coca-Cola Bottlers, Inc. (hereinafter Canners), constructed canning plants in the 1960’s to provide Coca-Cola and allied products in cans to bottlers under agency agreements in areas where there was inadequate canning capacity (Ogden, Tr. 825, 837).

12. Canners’ plants are located at Nashua, New Hampshire; Baltimore, Maryland; College Park, Georgia; Columbus, Ohio; Alsip, Illinois; and San Leandro, California (Ogden, Tr. 824). As independent bottlers in some areas desire to invest in them, the company has sold canning plants, e.g.,.in Salt Lake City, Utah; Plymouth, Florida; Greensboro, North Carolina; Houston, Texas; and Phoenix, Arizona (Ogden, Tr. 825-826).

13. Total dollar sales of soft drink syrup by “Coca-Cola” to bottlers in the United States for the indicated years were as follows: 1964 $178,961,477 1965 184,959,054 1966 211,629,426 1967 2271 844,768 1968 246,962,024 THE COCA-COLA CO., ET AL. 529 517 Initial Decision (CX 4F). [11] 14. Total dollar sales and total assets of “Coca-Cola” for the indicated years were as follows:

Sales Assets 1968 $1,185,808,864 802,100,548 1969 1,365,443,068 802,100,548 1970 1,606,401,160 1,005,777,214 1971 1,728,827 ,572 1,107,873,472 1972 1,876,192,397 1,231,612,887 (CX 51 - CX 54).

15. In July of 1971, “Coca-Cola” produced its full line of soft drink syrups and concentrates in plants at the following locations which supplied all of the licensed Coca-Cola and allied product bottlers in the United States:

Atlanta, Georgia Baltimore, Maryland Chicago, Illinois Columbus, Ohio Dallas, Texas Ewa Beach, Hawaii Houston, Texas Kearny, New Jersey Los Angeles, California Nashua, New Hampshire New Orleans, Louisiana Plymouth, Florida Portland, Oregon St. Louis, Missouri San Francisco, California (Ogden, Tr. 811-815). [12] The Other Respondents 16. In 1899 “Coca-Cola” granted to Mr. B. F. Thomas and Mr. Joseph B. Whitehead the exclusive right to bottle Coca-Cola and to sell Coca-Cola in bottles throughout the United States with the exception of the six New England States and the States of Mississippi and Texas where local jobbers were already bottling Coca-Cola on a very small scale, with the permission of “Coca-Cola” but without a contractual agreement. The sale of fountain syrup was specifically reserved to Initial Decision 91 F.T.C.

“Coca-Cola” (Smith, Tr. 679-680; Stipulation No. 1, RX 1B, Tr. 555; CX 4A), 17. Lacking sufficient capital or time to build sufficient plants expeditiously to bottle Coca-Cola and to create consumer demand throughout the area covered in the 1899 contract with “Coca-Cola,” Messrs. Thomas and Whitehead divided their territory between them. Mr. Thomas’ interest was termed The Coca-Cola Bottling Company (Stipulation No. 1, RX 1C, Tr. 555).

18. The Coca-Cola Bottling Company was soon succeeded in interest by Coca-Cola Bottling Co. (Thomas), Inc. Coca-Cola Bottling Co. (Thomas), Inc., transferred part of its territory to a sister company, Coca-Cola Bottling Works (Thomas), Inc. It also licensed Coca-Cola Bottling Works 8rd, Inc., to act as licensor with respect to certain designated territory (CX 4B; Tr. 488).

19. The Coca-Cola Bottling Company, under Mr. Whitehead and his new partner John T. Lupton, also formed several Parent Bottling Companies to whom the right to act as licensor with respect to certain territory was conveyed (CX 4B, Tr. 488). [13] 20. Beginning in 1900, Coca-Cola Bottling Company (Thomas) and The Coca-Cola Bottling Company (Whitehead) initiated a program aimed at granting exclusive licenses to local bottling companies throughout their respective territories. By 1920, they and the Parent Companies which they had formed had licensed over 880 local bottling companies to produce and sell Coca-Cola in bottles (Smith, Tr. 680; Stipulation No. 1, RX 1D, Tr. 555).

21. Respondent Thomas Company is a corporation organized, existing and conducting its business under and pursuant to the laws of The State of Delaware. It maintains its office and principal place of business at 1600 American Bank Building, Chattanooga, Tennessee 37402. In 1968, Thomas Company made sales to 122 bottlers located principally in Alabama, Delaware, Indiana, Maryland, Mississippi, New Jersey, New York, North Carolina, Ohio, Pennsylvania, Virginia, and West Virginia. These bottlers had contracts with 45 sub-bottlers in Illinois, Maryland, Mississippi, Missouri, New Jersey, New York, Ohio, Pennsylvania, Virginia, and West Virginia (Stipulation No. 2, CX 1243B, Tr. 465, 467; Susong, Tr. 491).

22. Respondent Thomas Works is a corporation organized, existing and conducting its business under and pursuant to the laws of the State of Delaware. It maintains its office and principal place of business at 1600 American Bank Building, Chattanooga, Tennessee. In 1968, Thomas Works made sales to 38 bottlers located principally in Alabama, Georgia, Kentucky, and Tennessee. These bottlers had THE COCA-COLA CO., ET AL. 531 517 Initial Decision contracts with 18 sub-bottlers in Alabama and Kentucky (Stipulation No. 2, CX 1248B-C, Tr. 466, 467; Susong, Tr. 491). [14] 28. Respondent Works 8rd is a corporation organized, existing and conducting its business under and pursuant to the laws of the State of Delaware. It maintains its office and principal place of business at 1600 American Bank Building, Chattanooga, Tennessee. In 1968, Works 3rd made sales to five bottlers located principally in Pennsylvania and New Jersey. (Admitted in Answer; Stipulation No. 2, CX 1248C, Tr. 466.) 24. Thomas Company and Thomas Works contract with bottlers pursuant to agreement with “Coca-Cola” under which the bottlers bottle and sell Coca-Cola soft drinks in their territories. Works 3rd contracts with Thomas Company for the same purpose. These three respondents engage principally in the purchase of Coca-Cola syrups for sale to numerous bottlers they have licensed in the various States of the United States. (Admitted in Answer; Stipulation No. 2, CX 1243D- E, Tr. 465-467.) 25. Since 1928, “Coca-Cola” has acquired all of the original Parent Bottling Companies with the exception of the three Thomas Companies. In doing so, “Coca-Cola” has acquired all of their rights and assumed all of their obligations. “Coca-Cola” presently deals directly with those bottlers not under contract to the three Thomas Companies in connection with bottled Coca-Cola and deals directly with all bottlers, including the Thomas bottlers, in connection with the bottling of all allied products (Susong, Tr. 538).

Interstate Commerce 26. Each of the respondents is engaged in “commerce” within the meaning of the Federal Trade Commission Act (15 U.S.C. 44). (Admitted in Answer.) There is a regular and continuous flow in interstate commerce of syrups and concentrates produced by “Coca- Cola,” between syrup and concentrate manufacturing facilities located in various states, and production facilities of “Coca-Cola” located in other states (Ogden, Tr. 812-17). [15] History of the Soft Drink Industry 27. The bottling of flavored carbonated soft drinks did not begin in the United States on a significant scale until the latter half of the nineteenth century. Prior to 1850 such soft drinks had been served almost exclusively at soda fountains for immediate consumption. During the late 1800’s a growing number of extract or syrup manufacturers entered the business. These persons and companies developed and introduced many new proprietary flavors, including Initial Decision 91 FT.C.

Hires, Vernor’s, Clicquot Club, Dr Pepper, Coca-Cola, and Pepsi-Cola (Stipulation No. 8, CX 1244A, Tr. 470).

28. Around the turn of the twentieth century syrup companies were largely small operations, typically owned by pharmacists or their families (Stipulation No. 3, Tr. 471). In order to provide the necessary inducement for local entrepreneurs to supply the capital required and to make the necessary effort to promote consumer acceptance of new bottled soft drink products, soft drink licensors included exclusive territorial provisions in trademark licenses (Stip. No. 8, CX 1244B-C, Tr. 471-72).

29. Territorial restrictions encouraged greater development of marketing and distribution efforts since exclusive licensees knew that their licensors and other licensees could not obtain a “free ride” on their efforts; they made possible the licensor’s maintenance of quality control, thereby insuring uniform application of his common law trademark; they facilitated the licensor’s production planning by enabling greater accuracy in calculating the forthcoming demand for syrup in a territory; they reduced the selling cost of the product by avoiding duplication of sales effort in a territory; and they encouraged the bottler to develop the potential of his territory to the fullest, thereby maximizing sales of the trademarked product (Stip. No. 8, CX 1244C, Tr. 471-72). [16] 30. The system of exclusive territorial licenses consistently has been widely employed in the manufacture and distribution of bottled soft drinks. There are over 50 syrup companies who have licensed local bottlers, 36 of them nationwide. These companies market more than 150 different soft drink brands through 7,500 agreements with local bottlers. These agreements for the local production and sale of trademarked products are unique when compared with the traditional organizational structure of American manufacturing and marketing (Stip. No. 8, CX 1244D, Tr. 472-78).

31. One unique feature of the soft drink trademark licensing system is that a nationally advertised product is manufactured locally by independent businessmen who are required to make substantial and continuing investments in plant, equipment, packaging and warehouse space (Clements, Tr. 4029-31). No other industry could be identified where a single national brand owner sells an ingredient to hundreds of independent licensees who manufacture a finished product from that ingredient and others under a trademark license (Clements, Tr. 4082). 32. The soft drink industry is also unique in that it sells a refreshment product which is an “impulse item” (Clement, Tr. 4029), whose most important characteristic is a distinct taste (Clements, Tr. 2989-90, 4032). Constant sampling is necessary to maintain demand for THE COCA-COLA CO., ET AL. 533 517 Initial Decision a brand: (Clements, Tr. 4088), and total availability of a brand at a multiplicity of outlets is essential to provide constant sampling necessary to successful marketing of that brand (Clements, Tr. 4029; Reid, Tr. 3554). The soft drink industry is also different from other industries in the broad range of flavors and package sizes and types required to be made available to satisfy consumer demand (Millard, Tr. 2319-20, 2821), in the need for frequent local store-door service | (Millard, Tr. 2319, 2824-25), the importance of in-store merchandising (Millard, Tr. 2324-25), and the requirement of a store-door delivery system to sustain the use of a returnable container (Clements, Tr. 4029-30). Soft drinks are the only major product still available in food stores in returnable containers (Ellis, Tr. 94445, 1022). [17] 38. The syrup company manufactures and sells syrup or flavoring concentrates to the bottlers. Syrup companies also participate in many of the advertising and promotion programs made in connection with their trademarked products, provide advice and technical assistance on production, quality control, management and sales problems, and engage in development and test marketing of new products and containers (Stip. No. 3, CX 1244D, Tr. 478). 34. The bottlers are the manufacturers and distributors of trademarked soft drinks produced from the licensor’s syrup or concentrate and manufactured according to his quality standards and specifications. The bottler decides on the plant and equipment to be used, the volume of production by size and type of container and product mix, as well as the price to be charged and the manner in which he can maximize his market penetration and secure the widest possible distribution of trademarked soft drinks throughout his territory (Stip. No. 3, CX 1244E, Tr. 473-74). 35. In most local markets, there are independent producers who manufacture and sell soft drinks under their own labels. Generally, these local and regional brands are not franchised. In recent years, however, several of these companies have expanded from their local bases and have begun to compete in regional and national markets. For example, Shasta, which just over 10 years ago was marketed exclusively on the West Coast, is now available in all states. Frank’s is currently marketed in many areas on the East Coast, and Faygo has branched out into almost a dozen eastern and midwestern states. In the process of expansion, these and other local and regional brands have utilized various techniques—distant shipments from existing plants, the establishment of new production facilities in other parts of the country, contract producers and in a few cases, territorial license agreements (Stip. No. 3, CX 144E-F, Tr. 474). [18] 36. Since the early 1960’s, private label soft drink brands have Initial Decision 91 E.T.C.

become a substantial competitive force in the soft drink industry. These are sold by a retailer under the retailer’s own trade name or under a trade name of a wholesaling or other organization from which the retailer purchases, e.g., 7-Eleven, Yukon Club (A&P), and Gayla (Topco). Private label soft drinks are manufactured either by the retailers or wholesalers themselves or for them by contract bottlers or canners and are generally sold for home consumption in non-returnable bottles and cans. Hundreds of new private labels have entered the market, many in the last five years. In the Washington, D.C., metropolitan area for example, over 15 private label brands are being marketed. Retailers are able to allocate scarce shelf space and prime store locations to their own private label soft drinks and are able to promote them vigorously with both point-of-sale and local media advertising (Stip. No. 8, CX 1244F, Tr. 474-75). 87. Flavored carbonated soft drinks are only one of a number of other commercial beverages available to the American consumer. Coffee is and has been America’s most popular commercial beverage. In 1950, soft drinks represented an estimated 10 percent of the total commercial beverage market; by 1970, soft drinks had more than doubled their market share, accounting for an estimated 21 percent. Soft drink consumption has been climbing rapidly, growing at an annual rate of approximately 6 percent for the last 2 decades. This increase enabled flavored carbonated soft drinks to become the nation’s second most popular commercial beverage in 1969. Such soft drinks may in the near future overtake coffee as America’s number one commercial beverage (Stip. No. 8, CX 1244F-G, Tr. 475-76). 38. Soft drink consumption patterns exhibit both a high degree of regional and seasonal variation. Consumption of soft drinks is higher in summer months and in areas with warm and humid climates, such as _the South (Stip. No. 3, CX 1244G, Tr. 476). [19] 89. Consumer taste, flavor preferences, and willingness to try new . products also have an impact on soft drink consumption patterns. On a national basis, cola has been the most popular soft drink flavor since 1920 when it surpassed gingerale. Since 1947, various studies have shown that the cola flavor accounts for approximately 60-65 percent of all soft drinks sold. While the share accounted for by cola drinks may have dipped slightly in recent years, according to some surveys, cola is America’s most popular flavor, followed by lemon-lime, orange, gingerale, and root beer. As with soft drink consumption, flavor preferences exhibit wide regional variation, with cola more popular in the South than in other areas of the country (Stip: No. 3, CX 1244G-H, Tr. 476).

40. The emergence of low-calorie soft drinks in the last 20 years THE COCA-COLA CO., ET AL. 535 517 Initial Decision also has been responsible for creating a number of new soft drink consumers. Before the introduction of low-calorie soft drinks, consumers voiced two primary reasons for avoiding soft drinks—caloric content and restricted diet. With the advent of the dietetic drinks, these two barriers were overcome. Although “No-Cal” and “Diet-Rite” were originally marketed in 1958, diet-flavored soft drinks did not become a significant market factor until the early 1960's. Following. the cyclamate ban in October 1969, which virtually removed lowcalorie soft drinks from the market, soft drink companies marketed saccharin substitutes, which by the end of 1970 had recaptured about 10 percent of total soft drink sales (Stip. No. 3, CX 1244H-I, Tr. 477). 41. The last 10 years have also seen the introduction of a variety of completely new soft drink categories including “isotonic” and “thirstquenching” soft drinks, such as “Gatorade.” Since 1960, a score of new trade names and flavors also have been introduced. The increase in per capita soft drink consumption has thus in part been brought about by the ability of the soft drink industry to develop new soft drink products, brands and flavors, and the willingness of the consumer to accept them (Stip. No. 8, CX 1244I, Tr. 477-78). [20] 42. Other factors contributing to the growth in per capita consumption of soft drinks are the general increase in disposable personal income and the increasing tendency of Americans to congregate in urban areas. Two other contributions to the growth in per capita soft drink consumption have been the major shift from small size (6-8 oz.) over to medium size containers (10-16 oz.), particularly in the home market, but also in vending machines dispensing packaged products, and the ability of the industry to make readily available at virtually every location soft drinks in a wide variety of brands, flavors, and packages (Stip. No. 8, CX 12441, Tr. 478). 48. In 1971, soft drinks were available in more than 99 percent of the over 200,000 retail food outlets in the United States. In addition, soft drinks are available in virtually every service station, restaurant, sports stadium, and theater in the country. The total number of retail outlets dispensing or selling soft drinks is well over a million (Stip. No. 8, CX 1244J, Tr. 478-79).

44. In recent years, there has also been a rapid proliferation of vending machines, which has produced an even more intensive coverage of the market. In 1971, there were nearly 1,400,000 vending machines dispensing soft drinks in bottles, cans, and cups (Stip. No. 3, CX 1244J, Tr. 479).

45. Fountain-dispensed soft drinks are available in well over 100,000 retail outlets. The rapid expansion of drive-in eating and Initial Decision 91 FE.

drinking places in recent years has also spurred fountain sales of soft drinks (Stip. No. 8, CX 1244J, Tr. 479).

46. The wide availability of flavored carbonated soft drinks is in part attributable to the high degree of market penetration achieved by bottlers of national brands employing store-door delivery in welldefined areas (Stip. No. 8, CX 1244J, Tr. 479). [21] 47. In addition to increased availability of soft drinks generally, there has been a proliferation of brands, flavors, and containers for the consumer to select from. At a given outlet, it is not uncommon for the consumer to have a choice of 25-30 flavors, both in dietetic and regular form (Stip. No. 8, CX 1244J—K, Tr. 479).

48, Since the turn of the century, soft drink consumption has undergone significant changes. In 1900, 70 percent of soft drinks were consumed on the premises of the vendors, and 30 percent were consumed in homes. Most of the flavored carbonated soft drinks consumed on the premises were dispensed from soda fountains; soft drinks consumed at home were for the most part delivered direct from the bottling plant. With the emergence of supermarkets in the 1930’s, the development of the six-bottle carton “Home Package” for Coca- Cola in 1922, the distribution of Coca-Cola in coin-operated vending machines in 1985, the patterns of soft drink distribution and consumption began to change. By the mid-twentieth century, the modern self-service supermarket had become a widespread reality. In the mid-1950’s, soft drinks sold for home consumption first surpassed those sold for consumption on the premises. Nationally, food stores account for between 85 percent to 90 percent of the sales for home consumption, or just above 50 percent of total soft drink sales. Soft drinks purchased for home consumption are typically purchased in medium-to-large-size containers (10 oz. and above) and multi-container cartons (Stip. No. 8, CX 1244K-L, Tr. 480). 49. With regard to consumption on the premises, the market is relatively evenly divided between vending machines (which account for approximately 18 percent of total flavored carbonated soft drink sales) and service sales at counters and tables (also approximately 18 percent). Canned and bottled soft drinks compete with bulk soft drinks dispensed at soda fountains and in cup vending machines (Stip. No. 3, CX 1244L, Tr. 480). [22] 50. In the United States, production and consumption of flavored, carbonated soft drinks have increased dramatically over the years, as evidenced by the following table:

Year Total Value $ Total Cases Per Capita (Wholesale) (192 ounce) (8 oz. containers) 1849 760,000 1,520,000 16 1859 1,415,000 2,830,000 2.2 THE COCA-COLA CO., ET AL.

Initial Decision 4,222,000 4,742,000 14,354,000 23,269,000 43,508,000 135,341,000 214,322,238 361,690,917 411,699,200 553,879,040 526,185,088 580,351,000 629,681,100 584,994,000 617,168,600 745,676,000 835,157,300 860,959,300 876,582,600 989,442,500 1,019,295,000 1,089,513,000 1,166,605,000 1,252,276,000 1,308,000,000 8,444,000 9,484,000 26,098,180 38,781,660 62,154,280 169,176,200 272,428,486 482,995,676 550,400,000 740,480,000 708,456,000 773,801,600 812,491,700 781,242,500 771,460,800 901,664,000 1,009,863,700 1,012,893,300 1,001,751,474 1,048,825,000 1,182,550,000 1,177,852,000 1,176,674,000 1,264,925,000 1,321,214,000 6.4 9.9 12.2 16.2 38.4 53.1 88.6 100.1 133.6 126.2 138.6 147.1 182.9 182.3 150.9 164.4 162.0 158.0 162.7 174.0 177.5 .

174.2 184.2 188.9 Initial Decision 91 F.T.C.

1957 1,847,241,500 1,360,850,000 189.2 1958 1,427,463,500 1,359,489,000 186.4 1959 1,633,015,900 1,484,560,000 199.8 1960 1,698,025,600 1,476,544,000 192.0 1961 1,829,083,200 1,524,236,000 198.3 1962 2,001,016,800 1,667,514,000 213.4 1963 2,341,189,000 1,800,915,000 227.4 1964 2,533,167,000 1,948,590,000 242.9 1965 2,785 ,567,000 2,104,282,000 259.1 1966 3,175,980,000 2,352,587,000 287.0 1967 3,458,632,000 2,470,452,000 298.1 1968 4,165,552,000 2,777,035,000 831.6 [23] 1969 4,369,664,000 2,913,110,000 344.4 1970 4,799,784,000 3,096,635,000 362.8 1971 5,346,960,000 3,353,615,000 388.1 1972 5,683,820,000 3,541,417,000 406.4 1973 6,223,156,500 3,771,610,000 429.6 (CX 1261) Bottlers and How They Operate 51. (Note: Complaint counsel called all of the Coca-Cola bottlers (17) in a “corridor” consisting generally of the area from southern Virginia to upstate New York. In addition, counsel for respondents called bottler witnesses, both for Coca-Cola and other brands whose territories are located in widely separated parts of the United States. These bottler witnesses typify all licensed bottlers in that they cover the spectrum of firms classifiable as urban, suburban, rural, single plant, multi-plant, large, small, first line bottlers, sub-bottler, marketing bottler (.e., one with no production facilities), multi-territory, THE COCA-COLA CO., ET AL. 539 517 Initial Decision single territory, privately held, publicly held, “Coca-Cola” subsidiary, etc.) 52. Most bottlers of Coca-Cola are small, independent manufacturers. Their size can be measured by (1) the population in a bottler’s territory (Wimberly, Tr. 871-72), (2) the bottler’s annual case sales of soft drinks (Wimberly, Tr. 872; Susong, Tr. 920; Sales, Tr. 1285), or (8) the number of bottler’s employees (RX 36 (p. 2)). 53. Bottlers of Coca-Cola whose territories contain less than 250,000 people include Hartwell, Georgia (35,000 (Rooks, Tr. 1360)); Spirit Lake, Iowa (175,000 (LaDoux, Tr. 1458)); Montross, Virginia (89,000 (Carver, Tr. 1619)); Jamestown, North Dakota (112,000 (Bernabucci, Tr. 1956)); Coatsville, Pennsylvania (240,000 (Filoromo, Tr. 2168)); Dover, Delaware (85,000 (Massey, Tr. 2236)); Westminster, Maryland (128,000 (Roadcap, Tr. 2482-33)); and Ada, Oklahoma (82,000 (Crabtree, Tr. 2662)). Territories between 250,000 and 500,000 people [24] include LaCrosse, Wisconsin (300,000 (Bernabucci, Tr. 1965)); Washington, Pennsylvania (380,000 (Cameron, Tr. 2048-44)); and Stockton, California (290,000 (DeLap, Tr. 2564)). Territories between 500,000 and 1,000,000 people include The Coca-Cola Bottling Company of the Peninsula in Belmont, California (700,000 (Sheldon, Tr. 2607)). 54. In 1974, there were 111 independently owned bottlers of Coca- Cola which had sales below 250,000 equivalent cases of 24 eight ounce bottles; 98 independently owned bottlers of Coca-Cola which had equivalent case sales between 250,000 and 500,000; 87 which had case sales between 500,000 and 1,000,000; and 47 which had between 1,000,000 and 2,000,000 cases (Susong, Tr. 914-15). Of these 343 bottlers with sales below two million cases, there was no common ownership or control (Susong, Tr. 915).

55. The Small Business Administration classifies any manufacturing business with less than 250 employees as a small business (RX 36 (p. 2)). The bottler system for “Coca-Cola” has a large number of small plants, with 85 percent having less than 100 employees (RX 35D). In the census of bottlers of Coca-Cola completed in 1974, of the 567 bottlers who responded, representing about 75 percent of all domestic bottlers of Coca-Cola (RX 85B), 193 had less than 25 employees, 168 had between 25 and 49 employees, 121 had 50-99 employees, 46 had 100-199 employees, and only 39 had more than 200 employees (RX 35P- 35Y). Of these plants, 14 were owned by subsidiaries of “Coca-Cola.” One of the 14 had less than 50 employees; one had between 50 and 100; five had between 101 and 200; and seven had over 200 (Bernabucci, Tr. 1954).

56. Bottlers are often closely held, family corporations (Brown, Tr. 1658; Burks, Tr. 3000; Ippolito, Tr. 3227). It is not unusual to find all Initial Decision 91 F.T.C.

members of the family, including wives, grandparents, children and other relatives working in some capacity in the business. For example, the president of the Northern Neck Coca-Cola Bottling Company, Montross, Virginia, testified that his son is vice-president of the company and his daughter is secretary (Carver, Tr. 1590). The president of the Coatesville Coca-Cola Bottling Works, Coatesville, Pennsylvania, testified that his three brothers [25] work as sales manager, delivery driver and service manager, respectively, and that one of his sons is planning to go into the business, joining two nephews who are already full-time employees (Filoromo, Tr. 2167). The president of the Dr Pepper bottler in Galveston, Texas, testified that his wife is the company bookkeeper; his son is sales manager and truck checker; and his brother-in-law works in the company’s office (Ippolito, Tr. 3229).

57. The bottler of Coca-Cola in Jamestown, North Dakota, testified that his father bought the company in 1988 when it was bankrupt (Bernabucci, Tr. 1955-56). In the Coatesville, Pennsylvania Coca-Cola Bottling Works, the first dividends were not paid until 1985. The founder’s salary was so meager that he operated a grocery store with his wife and seven children because, as his son testified, “We had quite a task just keeping bread and butter on the table” (Filoromo, Tr. 2160). Bottlers of Coca-Cola and other national brands initially had to develop the demand for their respective products by sneaking bottles of their products into cases of more accepted “rainbow” flavors (Stokes, Tr. 1072; Cameron, Tr. 2034; Ippolito, Tr. 3228). 58. Bottlers build demand for Coca-Cola and other national. brands by generating good will in the community, and have become identified with their products in their communities. For example, the bottler of Coca-Cola in Hartwell, Georgia, has his route salesmen operate a check cashing service for small stores located in rural areas (Rooks, Tr. 1878A).

59. Not all bottlers are small, family-owned corporations. Coca- Cola New York, a publicly held corporation, operates one of the largest independent soft drink businesses in the world with net sales in 1971 of $160,642,000. Coca-Cola New York operates 900 route delivery trucks, serves 70,000 accounts and almost 25 million people in New York, New Jersey, Pennsylvania, Connecticut, and Massachusetts. In addition to its soft drink operations, Coca-Cola New York has three subsidiary corporations: Mogen David Wine Corporation; Franzia Wine Company; and Igloo Corporation (Millard, Tr. 2248, 2318, 2828; CX 1220). [26] 60. The “Coca-Cola” and Thomas Companies licensing agreements have always and do now grant to the licensee the exclusive right to manufacture, distribute and sell trademarked soft drink products in THE COCA-COLA CO., ET AL. 541 517 Initial Decision bottles and cans in a defined geographical area. The contracts limit the licensee to the sale of his products only for ultimate resale within the licensee’s exclusive territory (Susong, Tr. 520; Smith, Tr. 588, 642). 61. License agreements in effect between “Coca-Cola” and its bottlers contain, and have contained the following typical provisions: . company agrees to furnish to bottler, and only to furnish to the territory herein. referred to, sufficient syrup for bottling purposes to meet the requirement of the bottler” in the territory herein described.

* * * * * * * . .. company does hereby select bottler as its sole and exclusive customer and licensee for the purpose of bottling the bottler’s bottle syrup, Coca-Cola, in the territory described.

* * * * * * * [Bottler agrees]. . . not to use the trademarks Coca-Cola or Coke, nor bottle nor vend said product except in the territory herein referred to. This limitation, however, is not to prevent bottler from acquiring similar rights for other territory. * * * * * * * [Bottler agrees]. . . not to use said distinctive bottle for any other purpose than the bottling of Coca-Cola and not in any territory except as herein referred to. (Susong, Tr. 506-508; CXs 10, 34, 40, 4446, 48.) [27] 62. There are other similar agreements in effect between the Thomas Companies and their bottlers. All such Thomas bottler agreements contain, and have contained, the following typical provisions:

[Licensor agrees]. . . to obtain and furnish to party of the second part [bottler] and only to obtain, for the territory herein referred to, sufficient syrup for bottling purposes to meet the requirements of the party of the second part in the territory herein described, provided party of the first part [licensor] can obtain the delivery to it of such syrup from The Coca-Cola Company under the contract existing between party of the first part and The Coca-Cola Company.

* * * * * * * [Licensor agrees to select bottler]. . . as its sole and exclusive customer and licensee for” the purpose of bottling Bottlers’ Coca-Cola syrup, and using the name Coca-Cola thereon in the territory herein described.

* * * * * * * [Bottler agrees]. . . Not to use the name Coca-Cola nor bottle nor vend said product except in the territory herein referred to without the written consent of party of the first part and The Coca-Cola Company. This limitation, however, is not to prevent party Initial Decision 91 K.-C.

of the second part from obtaining such rights from parties authorized to use the name Coca-Cola and to bottle and vend said product. (28] * * * * * * * [Bottler agrees] . . . To order, for the purpose of bottling Coca-Cola, the distinctive bottle, and none other, adopted or that may be adopted by party of the first part; to use said distinctive bottle and none other, in bottling Coca-Cola, and not to use said distinctive bottle for any other purpose than the bottling of Coca-Cola, and not in any territory except as herein referred to without the written consent of party of the first part and The Coca-Cola Company.

(CXs 11-15, 20-28, 26, 29, 31-33.) 63. Bottlers do not ship their products to a food chain if they believe all, or part, of the product is going to be transshipped outside the bottler’s territory. If they find that product which has been delivered to a chain or other customer has been transshipped outside the bottler’s territory, the bottler undertakes to get the chain to stop such transshipment in the future (Smith, Tr. 595). 64. When bottlers have notified “Coca-Cola” that their exclusive territories have been violated, it has been the policy of “Coca-Cola” to bring the subject to the attention of the bottler from which the product was alleged to have come (Smith, Tr. 664).

65. Usually when the matter is brought to the attention of the trespassing bottler, such bottler does what he can to stop the practice (Smith, Tr. 665).

Bottlers sell only in their own territory for ultimate resale in the territory (Carver, Tr. 1606; Ellis, Tr. 965; Brendle, Tr. 1764; Navarre, Tr. 1522, 1526).

66. Bottlers are free to, and they do, independently of “Coca-Cola” establish their prices for their soft drink products in their defined geographic territories (Smith, Tr. 688-89; Stokes, Tr. 1128). They also are free to purchase “Coca-Cola” approved bottles and crowns, other supplies, vending machines, coolers, vehicles, cases, bottling equipment, ete., from whomsoever they choose and “Coca-Cola” receives no royalty, compensation, or commission (Smith, Tr. 653-54, 690-92). Obtaining approval is not difficult (Smith, Tr. 65). [29] 67. In deciding on their pricing and packaging, bottlers are aware that other Coca-Cola bottlers will not violate their territory or compete in price or in packaging (Stokes, Tr. 1123; Navarre, Tr. 1522-23; Massey, Tr. 2235; Filoromo, Tr. 2153).

68. Bottlers compete within their territories against national brands of soft drinks such as Pepsi-Cola, Seven-Up, Royal Crown, Canada Dry, Dr Pepper, and Shasta, and against regional brands such as Rock Creek, Faygo, and Vernor’s private label brands (Stokes, Tr. THE COCA-COLA CO., ET AL. 543 517 Initial Decision 1109, 1112-1118; Ellis, Tr. 958-61; Massey, Tr. 2231-32; Filoromo, Tr. 2150-53).

69. Bottlers may and do choose whether they will engage in their own or participate in cooperative promotional efforts with “Coca-Cola” and other bottlers (Smith, Tr. 798-800; Stokes, Tr. 1121; Christian, Tr. 1825; Brown, Tr. 1681-82).

70. Bottlers may, and some do, produce brands other than those of “Coca-Cola.” Many of them do not produce all of the “Coca-Cola” brands and do not package in all of the sizes (bettles) authorized by “Coca-Cola” (Sales, Tr. 1202, 1210; Navarre, Tr. 1495; Massey, Tr. 2224; Filoromo, Tr. 2124-25, 2178). If a bottler of Coca-Cola does not carry one or more of its allied products, “Coca-Cola” is free to license another bottler to manufacture, distribute and sell that soft drink brand in the territory of the bottler who declined to handle the brand (Smith, Tr. 675) although to date “Coca-Cola” has chosen not to do so (Smith, Tr. 675).

71. In 1971, 488 of the 726 licensed bottlers of “Coca-Cola” and the Thomas Companies were licensed to sell trademarked soft drink products other than those of “Coca-Cola” (Smith, Tr. 689). 72. Bottlers use a store-door delivery system and do not deliver to chain store warehouses (Brown, Tr. 1678; Navarre, Tr. 1516-17; Massey, Tr. 2228, 2282, 2240; Millard, Tr. 2300). [30] 78. The current average cost in the United States to deliver Coca- Cola and allied products in cans directly to individual supermarkets and to merchandise the product under the prevailing store-door delivery system used by bottlers is approximately the same ($.06 more per case) as the average cost of delivery and merchandising Coca-Cola through the food store warehouse delivery system (Cowart, Tr. 3361- 62; RX 128). Cans are the easiest package to deliver through the warehouse system because cans are a compact, low cubage container (Cowart, Tr. 8362). This is indicated by the fact that cans are the predominant container delivered through the warehouse delivery system. Eighty percent of Shasta soft drinks (Meyers, Tr. 1787) and most private labels (Smith, Tr. 735) are sold in cans. 74, The difference between costs of the store-door delivery system and the warehouse delivery system in different parts of the country for delivering a case of cans would vary from three to five cents (Cowart, Tr. 3488). This difference is about 1 percent to 1 1/2 percent of the current wholesale price of a case of cans which, for example, is $4.77 in San Francisco (Cowart, Tr. 3425, 3488). The average cost in the United States to deliver Coca-Cola in 32-0z. non-returnable bottles through warehouses would be 9.6 cents more per case than the cost under the current store-door delivery system (Cowart, Tr. 3488-39). Initial Decision 91 F.T.C.

The reason the costs of the store-door delivery system and the warehouse delivery system are approximately the same for nonreturnable containers is that the same functions are performed under each system (Cowart, Tr. 3826-29, 3882-34; RX 121; RX 122). 75. The bottlers obtain canned trademarked soft drink products in the following ways. (1) Some bottlers have their own canning facilities. (2) Some bottlers have canned products produced for them by other bottlers under an agency canning agreement. (3) Some bottlers have formed canning cooperatives to produce canned products for their members. (4) Some bottlers obtain canned product from Canners for Coca-Cola Bottlers, Inc., a subsidiary of “Coca-Cola.” Others obtain canned products from independent contract canners approved by “Coca-Cola” (Ogden, Tr. 834, 836, 837). [31] 76. A typical bottling line includes: (1) a system for cleansing the container, (2) a water purification system, (3) a carbonation system, (4) a system which mixes syrup in the correct ratio with water and carbonation supplied by supporting systems, (5) a filling system whereby the mixture is transferred to bottles in the correct amount, (6) a closure system for final capping of the finished product, and (7) ancillary systems which case and transfer finished products in accordance with the requirements of the particular operation (Smith, Tr. 586-87).

77. Bottlers’ facilities are inspected by “Coca-Cola” on the average of three to four times a year (Sales, Tr. 1239-40; Alford, Tr. 2491; Hornsby, Tr. 3153-54). These inspections are generally unscheduled and unannounced (Susong, Tr. 921; Rooks, Tr. 1879; Bernabucci, Tr. 1980; Millard, Tr. 2326). They have included water analysis, bacteriological checks on water and processing equipment, measure of syrup content and degree of carbonation, bottle washing solution checks, and sanitation monitoring (Ellis, Tr. 978; RX 33A—J). Quality control inspectors also visit retail outlets and purchase products for analysis (Susong, Tr. 921; Ellis, Tr. 974-75). In addition, each production facility is required to submit monthly samples of every product to a quality control laboratory of “Coca-Cola” (Rooks, Tr. 1379-80; Alford, Tr. 2491-92; Hornsby, Tr. 3153). Many production facilities have their own laboratories, chemists, and quality control personnel (Ellis, Tr. 975-76; Millard, Tr. 2325-26; Alford, Tr. 2480-81; Crabtree, Tr. 2669). Inspections conducted by “Coca-Cola” were characterized as “more stringent” than governmental inspection by state authorities and the Food and Drug Administration (LaDoux, Tr. 1468). On occasion “Coca- Cola” has notified bottlers of Coca-Cola that it would cancel their contracts if substandard quality conditions were not corrected (Susong, Tr. 911-12).

THE COCA-COLA CO., ET AL. 545 517 Initial Decision 78. Bottlers of Coca-Cola rotate the stock on retail shelves to insure that the product does not get old and that the consumer purchases a fresh-tasting drink (Smith, Tr. 698; Ellis, Tr. 979; Sales, Tr. 1241, 1310- 11; Rooks, Tr. 1873; LaDoux, Tr. 1465; Carver, Tr. 1632; Christian, Tr. 1857; Millard, Tr. 2324; Roadeap, Tr. 2454; Burks, Tr. 3008). Route salesmen check the date code marked on the product (Ellis, Tr. 979; Sales, Tr. 1241; Rooks, Tr. 1878; Carver, Tr. 1632). Flavored. [32] carbonated soft drinks do not contain preservatives (Smith, Tr. 698; Ellis, Tr. 978). Most people familiar with the soft drink industry agree that soft drink products have a definite shelf life, a span of time beyond which the taste begins to noticeably deteriorate (Smith, Tr. 698). This deterioration affects taste but does not present a danger to health (Smith, Tr. 698; Ellis, Tr. 978). Deterioration results from the process of oxidation, which varies depending on exposure to heat and light (Ellis, Tr. 978-79). Estimates of this span are variously given for bottled products as two to four weeks (Stokes, Tr. 1111-17), three to eight weeks (Sales, Tr. 1240-41), sixty days (Carver, Tr. 1632), a few weeks (Ellis, Tr. 979), a month or so (Cameron, Tr. 2087), and four to eight weeks for bottles and 25 days for cans if stored in a cool, dark place (Rooks, Tr. 1881). The shelf life of cans depends on whether the cans are made of steel or aluminum. Aluminum cans retain taste quality a little longer than steel cans and do not transmit a metallic flavor to the drink (Stokes, Tr. 1116-17; Sales, Tr. 1240-41, 1843-44; Millard, Tr. 2300). The taste of flavored carbonated soft drinks becomes “unbearable” in eight to ten weeks in steel cans (Sales, Tr. 1343-44).

79. If despite stock rotation old product is discovered at a retail outlet, route salesmen are instructed to replace the product and bring the old product back to the plant for disposal (Ellis, Tr. 979; Sales, Tr. 1811; Rooks, Tr. 1881; Carver, Tr. 1632; Roadcap, Tr. 2454). 80. The plant inspection and the submission of product samples (Cobetto, Tr. 2833-34; Strachan, Tr. 2878-79; Ippolito, Tr. 3245; Hurst, Tr. 3453; Clements, Tr. 3989) and stock rotation by bottlers’ route salesmen (Cobetto, Tr. 2884; Strachan, Tr. 2893; Ippolito, Tr. 3248) are employed by other national brand soft drink licensors and their bottlers.

81. The distributional system of every licensed bottler of Coca-Cola is predicated upon store-door delivery and the direct servicing of all accounts (Smith, Tr. 655--56). [33] 82. Coca-Cola is distributed widely by bottlers and is available for purchase by consumers, often through vending machines owned by bottlers or others, through such outlets as: Initial Decision 91 F.T.C.

Supermarkets Train Stations/Depots Convenience Stores Bus Stations/Depots Small Grocers Bars and Taverns Other Food Stores Restaurants Drug Stores Luncheonettes Service Stations ~ Drive-Ins Department Stores Cafeterias .

Discount Stores Snack or Specialty Shops Variety Stores Offices Sports Arenas Industrial Plants Civic Centers and Schools Auditoriums Hospitals and Nursing Care Golf Courses Homes Amusement Parks Prisons, and Retail Establish- Bowling Alleys ments Ice and Roller Rinks not listed above (e.g., beauty Hotels & Motels salons, barber shops, liquor Airports stores, home delivery vendors and movie theaters).

(Stipulation No. 5, RX 2-Z41-Z42.) 83. While chain store accounts usually represent from 20 to 30 percent of a bottler’s business, they constitute only a minute fraction of the bottler’s delivery stops (Smith, Tr. 660). 84. Licensed bottlers of “Coca-Cola” service many unprofitable accounts in order to have the product available in as many outlets as possible (Smith, Tr. 700). (See Findings 187-148, 174-178 below.) [34] 85. The bottler of Coca-Cola in Annapolis, Maryland, obtained a 3.7 percent return on investment after taxes in 1974 (Brendle, Tr. 1789), which return was greater than that obtained in 1978 (Brendle, Tr. 1786). The Coatesville Coca-Cola Bottling Works, Inc. has been profitable since its inception in 1917 (Filoromo, Tr. 2196), but did not pay a dividend until 1935 (Filoromo, Tr. 2160), and still pays “meager” dividends today (Filoromo, Tr. 2199). The Filoromo family, which owns Coatesville Bottling Works, works cn small salaries and makes a return on invested capital of 2 1/2 -3 percent after taxes (Filoromo, Tr. 2197). The James E. Crass Coca-Cola Bottling Plants, Inc., with plants in Washington, D.C., Richmond, Virginia, Frederick, Maryland, and elsewhere, pays dividends of 2 — 3 percent on book value, which percentage would be much lower if based on market value of the investments (Stokes, Tr. 1076). The bottler of Coca-Cola in Hartwell, Georgia, made an after-tax profit of $46,000 (Rooks, Tr. 1406) and its THE COCA-COLA CO., ET AL. 547 517 Initial Decision two owners receive monthly salaries of $1,000 and $1,250, respectively (Rooks, 1407). The bottler of Coca-Cola in Spirit Lake, Iowa, made a profit during 14 of the last 17 years, which profit level has never been in excess of four percent of sales (LaDoux, Tr. 1478-79). The Spirit Lake bottler was profitable last year, obtaining a two percent return on sales (LaDoux, Tr. 1478-79), barely profitable the year before, and unprofitable two years ago (LaDoux, Tr. 1475). The bottler of Coca- Cola in Newport News has been profitable since 1914 and obtains a return on investment below five percent (Brown, Tr. 1701). The Coca- Cola Bottling Company of New York, Inc., a publicly held bottler, has a current return on equity of 6.4 percent (Millard, Tr. 2357-58). The bottler of Coca-Cola in Ada, Oklahoma has obtained a 3 — 4 percent return on book value for the last few years (Crabtree, Tr. 2684) and achieves a 10 cent per case profit before taxes (Crabtree, Tr. 2684). The bottler of Coca-Cola in Jamestown, North Dakota, receives a 7 - 8. percent return after taxes on the replacement value of his investments (Bernabucci, Tr. 1973-74). The salaries taken by bottlers and their rates of return on invested capital and sales are reasonable. [35] 86. Effective interbrand competition has also kept the profits of bottlers of other national brands at a reasonable level. The Pepsi-Cola bottler in Albany, New York, obtains a 4 1/2 percent return on the market value of his investments (Strachan, Tr. 2873). The Dr Pepper- Pepsi-Cola bottler in Dyersburg, Tennessee obtains a five percent before tax return on the replacement value of the company’s assets (Burks, Tr. 3047), takes less than $20,000 in salary (Burks, Tr. 3047), has paid only $16,000 in dividends since 1965 (Burks, Tr. 3046-47), and makes a per case profit of only 18 to 14 cents before taxes (Burks, Tr. 3048). The Dr Pepper bottler in Galveston, Texas, makes a profit of $40,000 on sales of $1,600,000 (Ippolito, Tr. 3271-72), obtains a five percent return on the market value of his investments (Ippolito, Tr. 3268), makes a profit of 15 to 20 cents per case before taxes (Ippolito, Tr. 8267), and has not paid any dividends in the last decade (Ippolito, Tr. 8267).

87. When family-owned bottlers of Coca-Cola have decided that they no longer wish to remain in the bottling business, they have been and are free to sell their exclusive trademark license to bottle and sell Coca-Cola in a given territory. “Under the existing system, if it becomes inefficient or unprofitable for a bottler to continue to serve his area, or when the members of a family decide that they no longer wish to remain in the bottling business, they can sell their franchise, as well as their plant and equipment” (Smith, Tr. 614). “Usually a bottler can afford to pay his neighboring bottler a generous price for his territory” and “the bottler who sells his plant is usually assured that he Initial Decision 91 F.

will receive a fair and equitable compensation for the investment has made in developing his territory" (Smith, Tr. 616). Bottlers can acquire another bottler without the approval of "Coca-Cola" through a stock acquisition (Smith, Tr. 628). "Coca-Cola " if asked, has recom­ mended that bottlers merge or consolidate where appropriate (Smith Tr. 615-16). However, bottlers are independent businessmen who make their own independent decisions and who frequently a~t contrary to the advice of "Coca-Cola" (Smith, Tr. 615-16). (36) Competition in the Soft Drink Industry Generally 88. There is intense competition in the sale of flavored carbonated soft drinks which stems from the fact that there is a large number of brands available to the consumer in local markets. In 1971, a Neilsen Survey showed that there were 135 different brands of cola flavored soft drinks marketed in food stores (Smith, Tr. 705). In the Washing­ ton, D. , metropolitan area alone, there are over 30 brands of colas being marketed (Sales, Tr. 1243-1251). In addition to the cola brands more than 20 other brands of flavors such as root beer, orange gingerale, and lemon-lime were being sold in the Washington, D. market (Sales, Tr. 1243, 1255; CX 372; CX 373). In the territory of the Newport News, Virginia bottler of Coca-Cola, between 30 and 40 different brands of orange and grape soft drinks were being marketed (Brown, Tr. 1666). Over 176 different brands of flavored carbonated soft drinks were sold in the territory of The Coca-Cola Bottling Company of New York, Inc. (Millard, Tr. 2347). 89. The consumer has a substantial number of brands available from which to choose in large urban areas, small towns, and rural communities alike. For example, over 40 different brands in the areas around Elmira (RX 78A-E), and Syracuse, New York (RX 83A-G); over 35 different brands in the areas around Richmond, Virginia (CX 364A-C; CX 367A-B), Utica (RX 84A-D), Watertown (RX 85A-D), and Binghamton, New York (RX 76A-C); over 30 different brands in the areas around Philadelphia (RX 73A-D), Fredericksburg, Virginia (CX 365; CX 368A-B), Petersburg, Virginia (CX 366; CX 369A-B), Frederick, Maryland (CX 362A-B), Westminster, Maryland (RX 53A­ C), Cortland (RX 77A-C), Finger Lakes (RX 79A-D), Gloversville (RX 80A-C), and Oneonta, New York (RX 82A-D); and over 20 different brands in Camden, New Jersey (RX 73~F), Albany (RX 74A-C), Glens Falls (RX 75A-C) and Hudson, New York (RX 81A-B). Many brands are marketed in a number of different flavors. At a given retail outlet choice of 25--30 it is not uncommon for the consumer to also have a , ... THE COCA-COLA CO., ET AL. 549 517 Initial Decision flavors, both in dietetic and regular form (Stip. No. , CX 1244K, Tr. 479). (37) 90. The degree of competition in the industry both nationally and in over two dozen local markets was described by various witnesses in the business as follows: "very active" (Smith, Tr. 707); "highly com etitive Ellis Tr 986 fierce intense " and "tremendous (Sales, Tr. 1242, 1332); "very aggressive" (LaDoux, Tr. 1458); "bloody, fierce" (Navarre, Tr. 1536); "extensive" (Brown, Tr. 1697); "strenuous (Brendle, Tr. 1781); "quite intense" (Levin, Tr. 1912); - severe competitive situation" (Filoromo, Tr. 2179); "vicious, intense, heated most competitive business I have ever seen" (Millard, Tr. 2351-53); high(ly J competitive" (DeLap, Tr. 2576); "very, very competitive" and very, very intense" (Crabtree, Tr. 2673, 2679); "very intense (Cobetto, Tr. 2827); "fierce" (Strachan, Tr. 2885-86); "brutal" (Connellee, Tr. 2965); "terrific aggressive" (Ippolito, Tr. 3233-34, 3264-65); more competitors. . . than anywhere in the supermarket" (Reid, Tr. 3560); "the most competitive industry, very intensive, very fierce and very vigorous" (Clements, Tr. 4024).

91. The testimony of witnesses familiar with the soft drink industry was that such competition was increasing. (Smith, Tr. 707-08; Ellis, Tr. 986; Sales, Tr. 1243; Navarre, Tr. 1536; Brendle, Tr. 1781; Levin, Tr. 1912-13; Cameron, Tr. 2057; Filoromo, Tr. 2180; Alford, Tr. 2508; Delap, Tr. 2576; Sheldon, Tr. 2615; Strachan, Tr. 2886; Reid, Tr. 3560.

Brand Competition 92. "Coca-Cola" and allied products sold by bottlers of Coca-Cola compete against a wide variety of national brand flavored carbonated soft drinks. Among the national brand products with which bottlers of Coca-Cola and allied products testified they compete in various localmarkets are Pepsi-Cola (Ogden, Tr. 839; Stokes, Tr. 1170; LaDoux, Tr. 1456; Carver, Tr. 1610; Levin, Tr. 1899) and other products of the Pepsi-Cola Company such as Diet Pepsi (Brown, Tr. 1684; Alford, Tr. 2501), Patio flavors (Millard, Tr. 2344) and Mountain Dew (Rooks Tr. 1388; Brendle, Tr. 1774); Seven-Up (Smith, Tr. 705; Stokes, Tr. 1170; Rooks, Tr. 1388; Brown, Tr. 1684; (38) Brendle, Tr. 1774; Christian, Tr. 1834) and other products of the Seven-Up Company such as Howdy Cola (RX 73B) and Diet Seven-Up (Millard Tr. 2344; Alford, Tr. 2501); Dr Pepper (Rooks, Tr. 1388; LaDoux, Tr. 1456; Carver, 1610) and Sugar Free Dr Pepper (Millard, Tr. 2344); Canada Dry (Rooks, Tr. 1388; Brendle, Tr. 1774; Christian, Tr. 1834; Cameron, Tr. 2057; Filoromo, Tr. 2151; Massey, Tr. 2231) and other products of - the Canada Dry Corporation such as Jamaica Cola (Sales, Tr. 1251) and Wink (Carver, ... .

Initial Decision 91 F.

Tr. 1611); Royal Crown (Ogden, Tr. 839; Stokes, Tr. 1170; Navarre, Tr. 1514-15; Christian, Tr. 1834; Levin, Tr. 1899) and other products of the Royal Crown Cola Co. such as Par- Pak (Carver, Tr. 1611), Diet Rite (Stokes, Tr. 1146; Rooks, Tr. 1388; Alford, Tr. 2501) and Nehi flavors (Stokes, Tr. 1146; Alford, Tr. 2501-02); Schweppes (Ellis~ Tr. 985; Brown, Tr. 1684); Squirt (Ellis,. Tr. 985); Nugrape (Wimberly, Tr. 875); London Dry (Rooks, Tr. 1388); YooHoo (Rooks, Tr. iass; Levin, Tr. 1899) and Brownie (Brown, Tr. 1684) chocolate sodas; Dixi Cola (Sales Tr. 1250); Climax gingerale (Carver, Tr. 1610; Brown, Tr. 1684); Lipton (Sales, Tr. 1252) and Nestea (Crabtree, Tr. 2676) canned iced teas; Frostie (Stokes, Tr. 1146; Alford, Tr. 2502), Dad's (Stokes, Tr. 1146), Hires (Levin, Tr. 1899), Ma s Old Fashioned (Filoromo, Tr. 2152) and A& W (Levin, Tr. 1899) root beers; Orange Crush (Levin, Tr. 1899; Bernabucci, Tr. 1983; Massey, Tr. 2230); No Cal (Stokes, Tr. 1146; Filoromo, Tr. 2151; Millard, Tr. 2344); Weight Watchers (Levin, Tr. 1899); Tru-Ade (Stokes, Tr. 1146; Carver, Tr. 1610); Double Cola (Sales Tr. 1248; DeLap, Tr. 2576; Crabtree, Tr. 2676); Triple Cola (Stokes, Tr. 1147); Cliquot Club (Sales, Tr. 1250); White Rock (Stokes, Tr. 1147; Millard, Tr. 2344); Snow Peak (Stokes, Tr. 1147); C&C Cola (Stokes, Tr. 1147; Millard, Tr. 2344); Wild West Sasparilla (Brown, Tr. 1684); and Shasta (Wimberly, Tr. 875; LaDoux, Tr. 1456; Carver, Tr. 1611). Bottlers of Pepsi-Cola (Strachan, Tr. 2883-84), Dr Pepper (Burks, Tr. 3035-36; Ippolito, Tr. 3243 3256 3265), and Seven-Up and Royal Crown (Cobetto, Tr. 2825) as well as witnesses selling Dr Pepper (Clements Tr. 3987, 4015); Shasta (Meyers, Tr. 1724, 1727, 1733); and Lipton (Reid Tr. 3562-63) and Nestea (Hurst, Tr. 3455-56) canned iced teas testified that their respective products compete with Coca-Cola and allied products as well as other national brand flavored carbonated soft drinks. (39) 93. In most local markets, bottlers of Coca-Cola and other national brands compete with local and regional brands of flavored carbonated soft drinks manufactured and sold by independent producers under their own labels (Stip. No. , CX 1244E, Tr. 474). The brands of these local-and regional manufacturers, such as Blair House and Rock Creek in Washington, D. , Suburban Club in Baltimore, Frank' Philadelphia, Graf' s in Milwaukee, and Faygo and Vernor s in Detroit have been strong competitors in specific markets for decades (Stip. No. , CX 1244E, Tr. 474).

94. Other local and regional brands of flavored carbonated soft drinks with which Coca-Cola and allied products compete include Canfield' s in Chicago (Ogden, Tr. 839; Meyers, Tr. 1732); Frank' s in Pennsylvania, Maryland, Delaware, Virginia, and the District of Columbia (EI'is, Tr. 984; Stokes, Tr. 1146; Sales, Tr. 1249; Carver, Tr. , ... THE COCA-COLA CO., ET AL. 551 517 Initial Decision 1611; Levin, Tr. 1899; Massey, Tr. 2231; Roadcap, Tr. 2423); Hoffman in Maryland, Virginia, Pennsylvania, and New York (Ellis, Tr. 985; Stokes, Tr. 1146; Meyers, Tr. 1733; Christian, Tr. 1899; Millard, Tr. 2344); Suburban Club in Maryland and Virginia (Ellis, Tr. 985; Stokes Tr. 1146; Carver, Tr. 1610); Rock Creek (Sales, Tr. 1249; Brendle, Tr. 1780) and Blair House (Sales, Tr. 1249; Brendle, Tr. 1780) in the greater Washington, D. , area; Green Spot, A-Treat, Reading Beverages, and Crystal Beverages in the Reading, Pennsylvania, area (Levin, Tr. 1899); Ritz in Florida (Meyers, Tr. 1732); Mission in San Antonio and the Southwest (Meyers, Tr. 1733; Alford, Tr. 2505); Checkers in Louisville (Meyers, Tr. 1733); Texas Beverages in San Antonio (Roadcap, Tr. 2482); Belfast in the San Francisco are~ (Sheldon, Tr. 2614); White Rock in New York 'and New Jersey (Millard, Tr. 2344); Regent in Pittsburgh (RX 2Z34); and Variety Club in Toledo (RX 2Z10). Various regional brands were described as a "very strong competitor" (Suburban Club) (Ellis, Tr. 985), "tough competitors (regional brands collectively) (Meyers, Tr. 1732), and "tremendous competitors" (Faygo and Shasta) (Cameron, Tr. 2058). In fact, the , bottler of Coca-Cola in San Antonio had The Coca-Cola Company develop a red cream soda Fanta Red " in order to compete with the strong market performance in the area of "Big Red " a local red cream soda brand (Alford, Tr. 2482, 2502-04). Big Red retails at the same price as Coca-Cola and has 10 percent of the home market (Alford, Tr. 2502-03). (40) 95. Shasta, which is produced by Consolidated Foods Corp. and which 10 years ago was confined to the West Coast, has now become a national brand marketed in all 50 states (Smith, Tr. 707; Meyers, Tr. 1710-14; Stip. No. , CX 1244E, Tr. 474). Shasta, in 1971, on the basis of estimates by Neilsen, was the fourth ranked brand in food stores in San Francisco, San Jose, Fresno, Seattle, Spokane, Tacoma, Portland Baton Rouge, Las Vegas, Denver, Reno, Phoenix, Kansas City, and St. , Louis (RX 2Z-2Z1). 96. Physical case sales of Shasta brand flavored carbonated soft drinks nationally and in the New Y ork- Virginia corridor area in the years indicated were as follows:

Year - Va. Corridor Uniwd Slaws 1967 493 000 645 000 1970 008 000 676 000 1972 629 000 992 000 (Stip. No. , RX 102Z16).

97. Shasta soft drink products are priced below the prices of .

Initial Decision 91 F.

national soft drink brands and above the prices of private label soft drink products (Meyers, Tr. 1751).

also compete 98. Bottlers of Coca-Cola and other national brands , havewith private label soft drinks which, since the early 1960' become a substantial competitive force in the soft drink industry. Private label soft drinks are sold by a retailer under the retailer s own wholesaling or othertrade name or under a trade name of a organization from which the retailer purchases Eleven, Yukon labels are Club (A&P), and Gayla (Topco) soft drinks. Private manufactured either by the retailer or wholesalers themselves or for such firms by contract bottlers or canners. Private labels are generally sold for home consumption in non-returnable bottles and cans (Stip. No. , CX 1244F, Tr. 474-75). Private label soft drinks are sold by food chains, independent grocers, drug stores (Sales, Tr. 1247-48; Brown Tr. 1247;Tr. 1683; Connellee, Tr. 2965)" convenience stores (Sales, Navarre, Tr. 1520; Connellee, Tr. 2965), dairy stores (Sales, Tr. 1247), (41) restaurant chains (Sales, Tr. 1246-47), and others. In fact, many food chains now sell more than one private label (Ellis, Tr. 983; Sales Tr. Tr. 1243-45; Navarre, Tr. 1518-19; Brendle, Tr. 1775; Filoromo, 2151-52; Hornsby, Tr. 3176). Nationally, literally hundreds of new private label soft drinks have entered the market, many in the last five years (Stip. No. , CX 1244F, Tr. 475). In the East, sales of private label flavored carbonated soft drinks by contract canners grew "dramatical­ " between 1~64 and 1971 in part because more food chains went into the sale of private labels (Hornsby, Tr. 3173). 99. Coca-Cola and allied products and other national brand flavored carbonated soft drinks compete with a wide variety of private labels in virtually every market (Smith, Tr. 705; Sales, Tr. 1244-48; LaDoux, Tr. 1456; Navarre, Tr. 1518-20; Millard, Tr. 2346-48; Roadcap, Tr. 2424-25; Alford, 'Pr. 2504-05; DeLap, Tr. 2576; Strachan, Tr. 2885; Clements, Tr. 4015).

100. In smaller communities there are typically between four and ten private label soft drinks being marketed which compete with Coca- Cola and allied products. For example, in Richmond, Virginia, Coca- " Giant Cola and allied products compete with A&P's "Yukon Club Foods Glee " Safeway s "Cragmont " Grand Union s "Penguin Colonial Stores Zesty," Food Fair s "Pantry Pride" and- Hy Tyme and 7-11 convenience store s " 11" brand (Stokes, Tr. 1114; CX 364A; the bottler of Coca-Cola competes ex 366). In Annapolis, Maryland, with Giant Foods Giant" and "Glee" brand private labels, Safeway Cragmont," A&P's "Yukon Club " Pantry Pride s "Pantry Pride" and Fyne Taste " and Acme s "Bala Club" and "Ideal" brands (Brendle Tr. 1775). In Wilmington, Delaware, Coca-Cola and allied products ...

THE COCA-COLA CO., ET AL. 553 517 Initial Decision compete with Acme s "Ideal Super Saver " and "Bala Club " Penn Fruit's "Gayla " Pantry Pride s "Pantry Pride" and "Hy Tyme " A&P' brand" "Bond Street" and "Yukon Club" Sho Rite s "Sho Rite brand, Pathmark' s "Pathmark" private label, and " 11" (Navarre, Tr. 1518-19; RX 32). (42) 101. Bottlers of Coca-Cola in Charlottesville, Virginia (Christian Tr. 1839-41); Jamestown, North Dakota (Bernabucci, Tr. 19~:-84); Westminster, Maryland (Roadcap, Tr. 2424-25); Coatesville, Pennsyl­ vania (Filoromo, Tr. 2151-52); Montross, Virginia (Carver, Tr. 1611 1614-16); and Dover, Delaware (Massey, Tr. 2234), testified that they compete with a number of private label brands. 102. Private label brands are a substantial competitive force. Nationally, Neilsen estimates that 20 percent of flavored carbonated soft drink sales in food chains are private label brands and that percent of the sales of canned soft drinks are private labels (Smith, Tr. 705-06). The same source also estimates that private labels account for approximately 40 percent of all flavored carbonated soft drinks sold in food stores in the Washington, D. , area (Sales, Tr. 1275) and approximately 30 percent in the New York metropolitan area (Millard Tr. 2348).

Competition in Prices 103. There is price competition between Coca-Cola and allied products and other brands of flavored carbonated soft drinks (Smith Tr. 742, 744; Sales, Tr. 1238, 1258-59; Rooks, Tr. 1362; Navarre, Tr. 1537-38). This pricing competition was described as "intense" (Levin Tr. 1912; Reid, Tr. 3562) and "very tough" (Smith, Tr. 742). There is also very intense price competition from regional brands. For example C&C Cola increased its market share in food stores in metropolitan New York from not being traceable in 1970 and 1972 to 3.6 percent of the market in 1975 (Millard, Tr. 2358). However, as a result of the exclusive territory provisions, competition between licensed Coca-Cola bottlers has been eliminated in their pricing of soft drinks, in the packaging of products, the sizes of containers and in the services they provide, such as warehouse delivery and pick-up at the bottling plant (Smith, Tr. 672-77). (43) 104. Because of keen interbrand price competition, bottlers of Coca-Cola attempt to price Coca-Cola and allied products at a level equal to or below major national brand competitors (Alford, Tr. 2483). In May 1975, Coca-Cola was priced below Pepsi-Cola in such markets as Washington, D.C. (Sales, Tr. 1238), Humbolt, Iowa (LaDoux, Tr. 1459), Westminster, Maryland (Roadcap, Tr. 2443), Montross, Virginia (Carver, Tr. 1630), Charlottesville, Virginia, (Christian, Tr. 1854-56), Initial Decision 91 F. C. ... ­ Newport News, Virginia (Brown, Tr. 1671-A), and Coatesville Pennsylvania (Filoromo, Tr. 2180). When Coca-Cola is priced below Pepsi-Cola and other competing brands, as happened in Charlottesville between November 1974 and May 1975 , sales of Coca-Cola improve substantially at the expense of the other brands (Christian, Tr. 1854­ 56). ,These other national brands view competitive pricing as important (Burks, Tr. 3046; Clements, Tr. 4024), and the Dr Pepper bottler in Dyersburg, Tennessee, indicated that he could not afford to .sell at a retail price of even one two or three cents higher on a six pack than Coca-Cola and other national brands for an extended period of time (Burks, Tr. 3046). The bottler of Coca-Cola in San Antonio testified that his sales will be adversely affected if his prices are two or three cents higher on a six pack (Alford, Tr. 2483), and the licensor of Nestea canned iced tea testified that even a one cent difference on a six pack may affect consumer choice (Hurst, Tr. 3457). 105. During late 1973 and 1974, there was a substantial increase in the price of sugar (Brown, Tr. 1694; Cameron, Tr. 2056; Sheldon, Tr. 2616; Clements, Tr. 4017), which is one of the principal ingredients in flavored carbonated soft drinks, and there also were increases in container and other costs (Brown, Tr. 1694). These increases in the cost of ingredients and containers brought about a substantial increase in the prices of flavored carbonated soft drinks (Brown, Tr. 1693-94; Sheldon, Tr. 2616; Clements, Tr. 4017). In 1974, home market sales of flavored carbonated soft drinks declined for the first time (Clements Tr. 4016) as the substantial increase in the price of flavored carbonated soft drinks resulted in consumers purchasing powdered mixes and other types of beverages (Clements, Tr. 4016-17). (44) 106. The prices charged by bottlers of Coca-Cola and allied products are determined by the prices of competing brands, costs of ingredients, and containers (Ellis, Tr. 965; Sales, Tr. 1259- , 1324; Carver, Tr. 1617, 1640; Millard, Tr. 2310). Such competition also controls the prices charged by bottlers of Dr Pepper (Clements, Tr. 4024) and other national brand soft drinks (Cobetto, Tr. 2836). When costs of ingredients and containers are relatively stable, bottlers Coca-Cola have been able to maintain their price levels. For example the wholesale prices charged by Washington Coca-:-Cola Bottling Company did not rise from April 1, 1971, until April 1, 1974 (Sales, Tr. 1260).

107. Because of competition from other brands when the cost sugar forced the price of sweetened flavored carbonated soft drinks to increase, many bottlers of Coca-Cola did not correspondingly increase the price for diet flavored carbonated soft drinks which do not contain sugar (Carver, Tr. 1625; Cameron, Tr. 2091-92; Roadcap, 1'r. 2448-49). ..

THE COCA-COLA CO., ET AL. 555 517 Initial Decision The result was a 30 to 50 cent per case differential between diet and sugar sweetened flavored carbonated soft drinks (Carver, Tr. 1625; Roadcap, Tr. 2448-49). Because of a reduction in the price of sugar in 1975, bottlers of Coca-Cola began to reduce their wholesale soft drink prices (Rooks, Tr. 1385; Cameron, Tr. 2056; Sheldon, Tr. 2617; RX 68). 108. When prices charged by bottlers of Coca-Cola are above those charged by major competitors such as Pepsi-Cola, sales of Coca;Cola decline. In early 1973, the effective price of Pepsi-Cola fell below Coca- Cola in New York City. The result was that the home mark~t share average of Coca-Cola over Pepsi-Cola declined from 18 versus 11 percent to 14 versus 12 percent. This forced The Coca-Cola Bottling Company of New York to increase pricing and promotion expenditures from $7-8 million to $17 million to recapture its position (Millard, Tr. 2355-57). In Hartwell, Georgia, loss of market share forced the bottler of Coca-Cola to meet the quantity discounts offered by his Pepsi-Cola competitor (Rooks, Tr. 1391-92). Similarly, when as a result of increased sugar prices Coca-Cola Bottling Company of the Peninsula in California found itself to be premium priced over Pepsi-Cola (45) and other national brands for the first quarter of 1975, the price differential led to a considerable slowdown in sales of Coca-Cola. The result was that Coca-Cola "priced (itself) out of the market place (Sheldon, Tr. 2616-17). When Coca-Cola Bottling Company of the Peninsula reduced its price in March 1975 (Sheldon, Tr. 2617; RX 68), sales of Coca-Cola significantly increased (Sheldon, Tr. 2620). 109. The wholesale prices charged by bottlers of Coca-Cola are competitive with the wholesale prices charged by their national brand competitors in the same local markets. On July 15, 1971, for example the regular wholesale prices for a case of 24 12-ounce cans of Coca-Cola and Pepsi-Cola were $3.00 each in Annapolis (CX 453A; RX 1031); $3.00 each in Charlottesville (CX 431; RX 103X); $3.00 each in Montross Virginia (CX 387B; RX 103Z1); $3.00 each in Westminster, Maryland (CX 991; RX 103Z1); $3.25 each in Albany, New York (CX 929; RX 103T); $3.10 each in Elmira, New York (CX 929; RX 103S); $3.25 each in Glens Falls, New York (CX 929; RX 103R); $3.25 each in Syracuse New York (CX 929; RX 103Z2); $3.10 each in Binghamton, New York (CX 929; RX 1030); $3.00 each in Philadelphia (CX 694; RX 103V); and $3.10 each in Watertown, New York (CX 929; RX 103N). 110. The most economical packages sold by bottlers of Coca-Cola and other national brands in almost every market are the larger size returnable bottles, namely the 16-ounce returnable (Smith, Tr. 770; Stokes, Tr. 1140-41; Sales, Tr. 1261, 1338-39; LaDoux, Tr. 1459, 1461; Brendle, Tr. 1773; Christian, Tr. 1822-23, 1863; Levin, Tr. 1925, 1940; Filoromo, Tr. 2148; DeLap, Tr. 2579; Connellee, Tr. 2971) and the 32­ Initial Decision 91 F.

ounce returnable/resealable (Stokes, Tr. 1141; LaDoux, Tr. 1459; Navarre, Tr. 1540; Carver, Tr. 1631; Christian, Tr. 1861; Ippolito, Tr. 3249-50). Even bottlers who do not distribute returnables (Strachan Tr. 2870-72) or who sell only a small percentage of returnables (Millard, Tr. 2305) testified that the returnable bottle with a reasonable trippage was the "most economical" package for the consumer. (46) 111. The returnable bottle is the most economical package to the consumer because the higher cost of the returnable container itself is spread over a large number of trips which the bottle makes before it is lost, destroyed or is no longer usable (Ellis, Tr. 997). For example, a 16­ ounce returnable bottle which costs 12 cents, and which makes 18-20 trips before it is lost or destroyed, averages out to a container cost of only a fraction of a cent per trip (LaDoux, Tr. 1461-62; Alford, Tr. 2488; Clements, Tr. 3996). On the other hand, when a consumer buys a non-returnable bottle, the full cost of the container must be recovered in the purchase price of the beverage (Clements, Tr. 3996 4042; RX 16). 112. Container costs are a substantial part (often over 50 percent) of the total cost of flavored carbonated soft drinks sold in non­ returnable containers (Hornsby, Tr. 3177). In May 1975, a case of 24 empty 12-ounce conventional steel cans cost $1.44 or 6 cents per can (LaDoux, Tr. 1461-62). The cost of aluminum cans is almost identical conventional steel cans (Sales, Tr. 1342-43). 113. Nationwide, approximately 55 percent of the sales of Coca- Cola in bottles and cans on a volume basis is sold in returnable bottles (Smith, Tr. 661, 777-78; Teasley, Tr. 3633, 3653 , 3755). The percentage of soft drinks sold in returnable bottles varies in different areas of the country (Teasley, Tr. 3758-59, 3777-78; RX 7). Fifty percent of the sales of Coca-Cola in bottles and cans in Richmond are packaged in returnable bottles (Stokes, Tr. 1115 , 1167); 30 percent in Washington C. (Stokes, Tr. 1167); 65 percent in Hartwell, Georgia (Rooks, Tr. 1384); 70 percent in Spirit Lake, Iowa (LaDoux, Tr. 1462) and the State of Iowa generally (LaDoux, Tr. 1463); 30 percent in Wilmington Delaware (Navarre, Tr. 1541-42); 25 percent in Havre de Grace Maryland (Navarre, Tr. 1542); 75 percent in Charleston, West Virginia (Navarre, Tr. 1542); 54 percent in Miami (Navarre, Tr. 1542); 74 percent in Montross, Virginia (Carver, Tr. 1633); 60 percent in Charlottesville, Virginia (Christian, Tr. 1859); 40 percent in Reading, Pennsylvania (Levin, Tr. 1916); 65 percent in Washington, Pennsylvan­ ia (Cameron, Tr. 2042); 20 percent in Coatesville, Pennsylvania (Filoromo, Tr. 2172); 47.9 percent in Westminster, Maryland (Roadcap, Tr. 2448); 41 percent in Dover, Delaware (Massey, Tr. 2226); 51 percent in San Antonio, Texas (Alford, Tr. 2487); 45 percent in . Stockton ... .

THE COCA-COLA CO., ET AL. 557 517 Initial Decision California (DeLap, Tr. 2567); 55-57 percent in Palo Alto, Burlingame and San Mateo, California (Sheldon, Tr. 2610); 60 percent in Jamestown, North Dakota (Bernabucci, Tr. 1982) and 70 percent in Ada, Oklahoma (Crabtree, Tr. 2670). (47) 114. There has recently been a resurgence of the use of returnable bottles and it appears that the share of the total soft drink market accounted for by returnable bottle sales has stabilized (Smith, Tr. 904 609 , 661-62; Strachan, Tr. 2872; Teasley, Tr. 3640, 3645-46). The percentage of Coca-Cola sold in returnable bottles has risen in Richmond, Virginia (Stokes, Tr. 1179-80), is increasing in Miami Florida, and Wilmington, Delaware (Navarre, Tr. 1542), and reverted from 45 percent returnable/55 percent non-returnable to 55 percent returnable/45 percent non-returnable in the territory of The Coca-Cola Bottling Company of the Peninsula in California (Sheldon, Tr. 2610). The principal reasons that the share of the market may have stabilized are adjustments in the deposit structure and the increasing segmenta­ tion of the market to the point where economy oriented purchasers are buying returnable bottles and convenience buyers are purchasing convenience packages rather than buying returnables and discarding them (Teasley, Tr. 3640). The introduction of the 32-ounce returnable also helped arrest the decline in returnable bottle sales (Smith, Tr. 662). In the immediate future, there are no market forces or trends in consumer preferences which are likely to bring about a substantial change in the share of soft drink volume accounted for by sales returnable bottles (Teasley, Tr. 3650-51).

115. Other national brands have also emphasized the low price per ounce returnable bottle (Strachan, Tr. 2871, 2915-16). Nationally, 65 percent of Dr Pepper s volume is sold in returnable bottles (Clements Tr. 3994). Eighty-five (85) percent of the volume of the Dr Pepper­ Pepsi-Cola Bottling Company in Dyersburg, Tennessee is sold in returnable bottles (Burks, Tr. 3030, 3096). Sixty (60) percent of the volume of Dr Pepper in the State of Texas is sold in returnables (Ippolito, Tr. 3252-53 3287). In Herminie, Pennsylvania, and Wheeling, West Virginia, the Seven-Up/Royal Crown bottler has approximately 80 percent and 50 . percent, respectively, of his sales volume in returnables (Cobetto, Tr. 2818-19 2837).

116. In July 1971, the average retail price of Coca-Cola in the United States in 16-ounce returnable bottles, according to Neilsen sources, was lower than the average price per ounce at which Coca- Cola in the . 6 1/2-ounce returnable bottle was sold at retail in 1900 (Smith, Tr. 716). (48) 117. Neilsen reported that in the period December 1970-January 1971 the retail price per ounce of Coca-Cola in 16-ounce returnable Initial Decision 91 F.

bottles nationwide was on the average of one percent less than the price per ounce on private label cans sold in. chains and large independent food stores (RX 8). In June-July 1971, the natipnwide average retail price per ounce for Coca-Cola in 16-ounce returnable bottles was the same as the average price per ounce for private label soft drinks in all containers combined (RX 9). 118. hi many local markets, Coca-Cola in 16 and 32-ounce returnable bottles is cheaper than private labels. In Montross, Virginia a 32-ounce returnable bottle of Coca-Cola in May ~975 retailed for 33 cents; 32 ounces of Rich Food's private label in 28-ounce non­ returnable bottles retailed for 38 cents; 32 ounces of Safeway Cragmont retailed for 48 cents; 32 ounces of A&P's Yukon Club retailed for 53 cents; and 32 ounces of Shasta retailed for 59 cents (Carver, Tr. 1634-35). In Newport News, Virginia, in May 1975, the 16­ ounce returnable bottle of Coca-Cola sold at retail for a little over a penny an ounce (six 16-ounce bottles for $1.08 or 96 ounces for $1.08) (Brown, Tr. 1680), while a six-pack of private label soft drinks in cans (72 ounces) retailed for between 99 cents and $1. , or over 1.33 cents per ounce (Brown, Tr. 1693). In Jamestown, North Dakota, 32-ounce returnable bottles of Coca-Cola retailed at 1.2 cents per ounce and private label cans retailed at 1.4 cents per ounce (Bernabucci, Tr. 1981­ 82). In San Antonio, Coca-Cola in 16 and 32-ounce returnable bottles retailed at regular everyday prices at about one cent per ounce whereas private label soft drinks in cans retailed at approximately 18­ 20 cents each or 1.5 to 1.7 cents per ounce (Alford, Tr. 2492). And in Ada, Oklahoma, 32-ounce returnable bottles of Coca-Cola retailed at 33 cents each or three for a dollar, which is about a penny an ounce (Crabtree, Tr. 2671); 12-ounce cans of private label soft drinks retailed at 18-19 cents each at 1.5 cents per ounce (Crabtree, Tr. 2674); and 64­ ounce non-returnable bottles of the private labels retailed at 69 cents or just over one cent per ounce (Crabtree, Tr. 2690). (49) 119. The 16 and 32-ounce returnable bottles provide direct price per ounce competition to private labels (Ellis, Tr. 995, 1021-22; Brown, Tr. 1661; Strachan, Tr. 2870). Because national brands sold in returnables provide the strongest price competition for private labels, food chains dislike handling national brands in economical returnable bottles (Ellis Tr. 1021-22; Strachan, Tr. 2867A 2870). In areas where national brands are predominantly in returnable bottles, private label market penetration is weak; in areas where national brands have been forced into non-returnable containers, private labels are strong (Strachan, Tr. 2883-85). In markets where 32-ounce returnables are sold, chains often keep them at locations in the beverage section which are physically distant from their private labels (Burks, Tr. 3032-33). _.

THE COCA-COLA CO., ET AL. 559 517 Initial Decision 120. Without exclusive territories the use of the returnable bottle by bottlers of Coca-Cola and other national brands would be substantially reduced, if not eliminated (Smith, Tr. 734--736; Ellis, Tr. 995; Rooks, Tr. 1400, 1431; Bernabucci, Tr. 1994; Roadcap, Tr. 2466; Crabtree, Tr. 2689; Strachan, Tr. 2903; Clements, Tr. 4038--39). If the chain stores converted to a system of warehouse delivery, the chain stores would eliminate returnable bottles entirely because the returna­ ble bottle is incompatible with warehouse delivery (Smith, Tr. 73&-::36; Ellis, Tr. 995; Rooks, Tr. 1401; Cobetto, Tr. 2838--39; Strachan, Tr. 2870 2903). In Los Angeles, where national brand soft drinks are delivered to warehouses, no returnable bottles are handled through the warehouse system (Hurst, Tr. 3496).

121. Returnable bottle usage would also decline because bottlers would be reluctant to invest in returnable bottles when they had no assurance that they would be able to recapture their large investment in returnable bottles for reuse (Smith, Tr. 700; Rooks, Tr. 1404; Cobetto, Tr. 2813; Clements, Tr. 4046).

122. If bottlers lose their high volume accounts or such accounts shifted to predominantly nonreturnable containers, the cost of providing returnable bottles to the remaining low-volume accounts will necessarily increase to cover fixed costs (Alford, Tr. 2540; DeLap, Tr. 2590), thereby reducing the price per ounce advantage and economical appeal of returnable bottles to consumers, and consequently reducing demand for returnable packages (Smith, (50) Tr. 735-36; Ellis, Tr. 1020; Roadcap, Tr. 2466; Connellee, Tr. 2989, 2994). Once returnable bottles lose their economy image, and the price per ounce differential with non-returnables narrows, consumers will stop purchasing returna­ bles (Cameron, Tr. 2066--68). Without the economy appeal of the returnable, trippage (i. the number of fillings of a returnable bottle) will decline and the returnable bottle will die (Smith, Tr. 734-36). 123. The retail price per ounce differential between Coca-Cola in returnable bottles and non-returnable containers varies in specific markets. For example, in July 1971, in Baltimore it cost the consumer approximatley 30 percent more per ounce to buy Coca-Cola in 16-ounce non-returnable bottles than in 16-ounce returnable bottles and percent more per ounce in 12-ounce cans than in 16-ounce returnable bottles (Ellis, Tr. 982). In Wilmington, Delaware, the retail price of 32­ ounce returnable bottles of Coca-Cola is four for $1.69 or 1.32 cents per ounce; the prevailing retail price for cans is six for $1.49 or 2.06 cents per ounce, 36 percent more expensive to the consumer on a per ounce basis (Navarre, Tr. 1541). In May 1975, in Montross, Virginia, the 16­ ounce returnable bottle retailed in supermarkets at 1.08 cents per ounce (Brown, Tr. 1680); cans retail at 2 cents per ounce or Initial Decision 91 F.

approximately twice as much (Brown, Tr. 1680, 1692). In Jamestown North Dakota, the May 1975 retail price per ounce of Coca-Cola in 32­ ounce returnable bottles was 1.2 cents; the price per ounce in 32-ounce non-returnables was 1.5 cents, and in cans, 2.2 cents (Bernabucci, Tr. 1981). In May 1975, Coca-Cola in 16-ounce returnable bottles was on a per ounce basis, 29 percent cheaper than Coca-Cola in 16-ounce nonreturnables, 27 percent cheaper than Coca-Cola in 32-ounce nonreturnables, 16 percent cheaper than Coca-Cola in 64-ounce nonreturnables; and 61 percent cheaper than Coca-Cola in 12-ounce cans in Reading, Pennsylvania (Levin, Tr. 1925). In San -Antonio, Texas, the May 1975 retail price per ounce for Coca-Cola in 16 and 32-ounce returnable bottles was about a penny. Coca-Cola in 48 and 64-ounce non-returnable bottles retailed at about 1.5 cents per ounce, or 50 percent more expensive than it is in a 16-ounce returnable bottle, and Coca-Cola in cans retailed at 1.9 cents per ounce, or 90 percent more expensive (Alford, Tr. 2488, 2551).

124. Large size returnable bottles provide the same price per ounce advantages to other national brands as they do to Coca-Cola. Dr Pepper in Dallas sells 16 and 32-ounce returnables at the same price per ounce that prevailed 70 years ago (Clements, Tr. 3994-95). From 1971 until 1974, the Dr Pepper bottler in Dyersburg, Tennessee, was selling Dr Pepper in 16-ounce, and later in 32-ounce, returnable bottles on a per ounce basis at approximately the same price that soft drinks were being sold for in the 1930's (Burks, Tr. 3030). (51) 125. Other brands in large size returnables also sell at a retail price per ounce comparable to private labels. For example, Dr Pepper in 32­ ounce returnablelresealable bottles in Dyersburg, Tennessee, in June 1975 retailed for between 41 and 43 cents per bottle (Burks, Tr. 3031). The private label of the largest chain in Dyersburg, Kroger s "Big K in 28-ounce non-returnable bottles sold for 43 cents in June 1975, or for more than Dr Pepper on a per ounce basis (Burks, Tr. 3032). And in Galveston, Texas, in June 1975, Dr Pepper in 32-ounce returnables was cheaper on a price per ounce basis than private label cans (Ippolito, Tr. 3255-56).

126. Other brands of soft drinks sold in returnable bottles are less expensive on a per ounce basis than the same brands sold in non­ returnable bottles and cans. In June 1975, in Galveston, Dr Pepper in 32-ounce returnablelresealable bottles retailed for 33 cents or about one cent per ounce; in 32-ounce non-returnables the price was 49 cents of 1 1/2 cents per ounce; in cans the price was si~ for $1.40 or almost 2 cents per ounce (Ippolito, Tr. 3255-56). In Dallas, the retail per ounce price of Dr Pepper in cans was 1.75 cents versus less than 1 cent in returnable bottles (Clements, Tr. 3994-96). In Dyersburg, Tennessee, a _.

THE COCA-COLA CO., ET AL. 561 517 Initial Decision 32-ounce returnable/resealable bottle of Dr Pepper retailed for between 41 and 43 cents; a 28-ounce non-returnable bottle of Dr Pepper retailed for 49 cents and a six-pack of 12-ounce cans of Dr. Pepper retailed for between $1.37 and $1.39. The retail price per ounce of Dr Pepper in 32-ounce returnable bottles was 1.3 cents, and for cans the per-ounce price was nearly 2 cents (Burks, Tr. 3031-34). 127. Bottlers of Coca-Cola and other brands compete by offering price promotions. Promotions are normally offered in the form of allowances or cents-off the wholesale price per case (Sales, Tr. 1333) or by providing one case free with a given number of cases purchased (Stokes, Tr. 1181-82; Millard, Tr. 2389). Many industry witnesses indicated they (52) compete through price promotions in their respective markets (Ogden, Tr. 840; Sales, Tr. 1237; Navarre, Tr. 1503­ 05; Carver, Tr. 1604; Christian, Tr. 1854; Filoromo, Tr. 2192-93; Millard, Tr. 2357-58; Alford, Tr. 2510-11; DeLap, Tr 2576-77; Strachan, Tr. 2886; Burks, Tr. 3043, 3088; Hurst, Tr. 3456; Reid, Tr. 3562-63; Clements, Tr. 3995 4025).

128. Price promotions are increasing substantially year after year (Navarre, Tr. 1503; Bernabucci, Tr. 1986; Millard, Tr. 2357; Strachan Tr. 2886). For example, the Delaware Coca-Cola Bottling Company in Wilmington, Delaware, has increased its discount or promotional budget, that is the amount of the reduction of normal wholesale prices which it allocates for cents-off promotions, from $70 000 two years ago to $550 000 today (Navarre, Tr. 1503). The Coca-Cola bottler in Jamestown, North Dakota and in LaCrosse, Wisconsin indicated that 10 years ago price promotions were "rarely offered" but that currently his expenditures for promotional allowances ran $50 000 a year in Jamestown and $60 000 in LaCrosse (Bernabucci, Tr. 1986). The bottler of Coca-Cola in Stockton, California, testified that to keep up with competition "we are running into more and more promotions and promotional discounts" (DeLap, Tr. 2576). The amount of money the bottler of Coca-Cola in Stockton spent on cents-off discounts increased from $12 734.42 in 1967 to $104 505.85 in 1974 (DeLap, Tr. 2577). The bottler of Coca-Cola in San Antonio testified that "10 years ago we never did any price promotions" and "now it is a fact of life" and that if we don t price promote * * * then we are dead" (Alford, Tr. 2508). The Seven-Up bottler in Herminie, Pennsylvania, indicated that he has been "promoting heavily" with cents-off promotions the last four or five years, and that this type of "competition is very keen" (Cobetto Tr. 2829). (53) 129. Food chains do not always pass on promotions in the form of lower retail prices to the consumer. The bottler of Coca-Cola in Wilmington, Delaware, indicated that the chains pass on the promo­ .

Initial Decision 91 F.

tions 80-90 percent of the time (Navarre, Tr. 1504). Safeway in Washington, D. , refused to pass on a 52 cents per case promotion for Coca-Cola during November 1971 and instead increased its margin to 37 percent (Sales, Tr. 1286-87, 1290; RX 23). In November 1974, Dart Drug received a 60 cents per case wholesale price reduction from the Washington, D.C. bottler of Coca-Cola for a two-week period but reduced its retail price to the consumer only during one of the two weeks (Sales, Tr. 1288; RX 24). In Dyersburg, Tennessee, Kroger, the largest food chain in the area, received one case free with the purchase of a case of 32-ounce returnable bottles of Bubble-Up, effectively reducing the wholesale price 50 percent, but did not reduce the price to the consumer below its normal retail price for a carton of 32-ounce returnable reseal able bottles (Burks, Tr. 3041-42). In other markets food chains have refused to pass on promotions to the consumer (Stokes, Tr. 1181-82; Millard, Tr. 2380; Strachan, Tr. 2890; RX 18). During promotions, food chains try to purchase extra stock which they store in their back rooms and then sell at the normal retail price after the promotions (Massey, Tr. 2194; Millard, Tr. 2376). And, in New York City, food chains will not pass on promotions of one case free with ten cases; the bottler of Coca-Cola has to give one case free with five in order to have the wholesale savings passed on (Millard, Tr. 2376--77). The fact that outlets other than food chains receive and pass promotions influences the food chains' willingness to pass on soft drink promotions (Ellis, Tr. 994, 1021; Sales, Tr. 1319). 130. Promotions for Coca-Cola, Pepsi-Cola, Royal Crown, Seven- Up, Dr Pepper, canned iced teas, and private label flavored carbonated soft drinks affect sales. When bottlers of Coca-Cola are running promotions on Coca-Cola, sales volume increases noticeably (Carver Tr. 1630-31; Levin, Tr. 1914; Alford, Tr. 2510, 2578). According to bottlers of other brands, promotions of Coca-Cola take sales away from Pepsi-Cola (Strachan, Tr. 2886--87, 2889-90), Seven-Up (Crabtree, Tr. 26~2), and Nestea (Hurst, Tr. 3456) and Lipton (Reid, Tr. 3562-63) canned iced teas. When Pepsi-Cola (Stokes, Tr. 1123; Carver, Tr. 1631; Levin, Tr. 1914; Crabtree, Tr. 2682), Royal Crown (Carver, Tr. 1631), 2682), (54) Dr PepperSeven-Up (Bernabucci, Tr. 1972; Crabtree, Tr. (Alford, Tr. 2510), private labels (Brown, Tr. 1697), Nestea canned iced tea (Hurst, Tr. 3456), and other national brand competitors (Brown, Tr. 1697; Alford, Tr. 2578) run promotions of their own, sales of Coca-Cola are adversely affected: Correspondingly, when Seven-Up (Cobetto, Tr. 2832), Pepsi-Cola (Strachan, Tr. 2886), Nestea (Hurst, Tr. 3456) and other brands run promotions, their sales volume increases. 131. Bottlers of Coca-Cola and other national brands charge a Tr. 659; uniform price to both large and small accounts (Smith, THE COCA-COLA CO., ET AL. 563 517 Initial Decision Navarre, Tr. 1577; Brown, Tr. 1673; Brendle, Tr. 1784; Alford, Tr. 2520; Clements, Tr. 4040, 4060-61). Uniform pricing allows the smaller accounts to be serviced (Smith, Tr. 659; Navarre, Tr. 1577; Alford, Tr. 2520; Clements, Tr. 4042-44).

Competition in Fountain Drinks 132. Coca-Cola and other flavored carbonated soft drinks sold in bottles and cans compete with post-mix soft drink products dispensed at soda fountains and in cup vending machines (Stip. No. , CX 12441 Tr. 481). Bottlers of other national brands face interbrand competition from post-mix Coca-Cola. The interbrand competition among Coca- Cola, Dr Pepper and other post-mix soft drinks has been characterized as "very tough" (Smith, Tr. 672; Clements, Tr. 4017 4022). Bottlers of Seven-Up and Royal Crown (Cobetto, Tr. 2827-28) and Dr Pepper (Ippolito, Tr. 3262-64) have lost bottle and can accounts to interbrand post-mix competition from "Coca-Cola." Post-mix Dr Pepper has taken bottle and can accounts from "Coca-Cola" (Burks, Tr. 3042-43). 133. Coca-Cola sold by licensed bottlers in bottles, cans and pre-mix containers is subject to vigorous intrabrand competition from post-mix Coca-Cola sold by independent wholesalers. The president of "Coca- Cola" (Smith, Tr. 672, 687, 706-07, 744) and bottlers testified that Coca- Cola sold in bottles, cans and pre-mix containers competes against post-mix Coca-Cola sold by independent wholesalers or jobbers (Ellis Tr. 990-91; 1aDoux, Tr. 1457; Carver, Tr. 1627, 1640; (55) Millard, Tr. 2348-50; Crabtree, Tr. 2677-78). There usually are a number of postmix wholesalers selling in a given area, including some bottlers of Coca-Cola who also serve as post-mix wholesalers (Sales, Tr. 1257; 1aDoux, Tr. 1457).

Competition in Powdered and Noncarbonated Drinks 134. Flavored carbonated soft drinks also face competition from powdered soft drink mixes in pricing, advertising, and shelf space. Price competition from powdered soft drink mixes is evidenced by the low per-ounce price of powdered mix brands (Sales, Tr. 1253). The Coca-Cola Bottling Company of New York analyzed monthly the prices of powdered soft drink mixes in making decisions on the pricing of Coca-Cola and allied products (Millard, Tr. 2322). The surge in sugar costs heightened price competition by increasing the per-ounce price differential between powdered soft drink mixes and sweetened flavored carbonated soft drinks (1aDoux, Tr. 1457). During the summer months media advertising of powdered soft drink mixes intensifies (Carver, Tr. 1639-40). Advertisements stress the lower per­ ...

Initial Decision 91 F.

ounce price of powdered mixes vis-a-vis. flavored carbonated soft drinks (Levin, Tr. 1939; Alford, Tr. 250&-06). In 1974, home market sales of powdered soft drink mixes increased substantially while sales of flavored carbonated soft drinks declined (Clements, Tr. 4016-17 4059-60).

135. Coca-Cola ahd other national brand flavored carbonated soft drink products also compete to some degree with noncarbonated drinks, such as Hi-C (Sales, Tr. 1257; Strachan, Tr~ 2885; Hurst, Tr. 3455), Hawaiian Punch (Rooks, Tr. 1388; Strachan, Tr. 2885; Burks, Tr. 3037), Gatorade (Millard, Tr. 2282), Tru-Ade (Burks, Tr. 2965), fruit juices (Rooks, Tr. 1388; Cobetto, Tr. 2827), and other beverages which satisfy the consumer s desire for liquid refreshment (Smith, Tr. 705; Rooks, Tr. 1389; Alford, Tr. 2483). (56) 136. Flavored carbonated soft drinks "compete with every other liquid a person consumes" (Smith, Tr. 705). There is data which indicates that a human being will consume a finite quantity of liquid over a given period of time (Smith, Tr. 705). Annually, per capita consumption of liquids has remained at about 120 gallons over the last two decades (Hurst, Tr. 3453-54). Within the 120 gallons, there have been fluctuations as consumption of different types of beverages increases and decreases (Hurst, Tr. 3454-55). In this broad sense, Coca- Cola and other flavored carbonated soft drinks compete with every other beverage (Smith, Tr. 705; Navarre, Tr. 1533; Meyers, Tr. 1732; Strachan, Tr. 2861).

Competition in Product Availability 137. Bottlers of Coca-Cola and other national brand flavored carbonated soft drinks compete to have their brands available in a large number of outlets (Sales, Tr. 1344; Navarre, Tr. 1538; Burks, Tr. 3044). This competition was described as "quite vigorous" (Reid, Tr. 3563) and as "constant fighting in the marketplace to make ourselves available" (Ippolito, Tr. 3267). In the soft drink industry, "(i)t seems like everyone would like to be everywhere, on every street corner " (Reid, Tr. 3563.

138. To provide availability, bottlers of Coca-Cola and other national brands have not imposed minimum delivery requirements. Some deliver one case of product on a weekly basis (Sales, Tr. 1232; Christian, Tr. 1869; Filoromo, Tr. 2191-92; Millard, Tr. 2374; Burks, Tr. 3013). This contrasts with warehouse delivered soft drink products such as Shasta, which has minimum delivery requirements at its Baltimore, Maryland, and Union, New Jersey, plants of 500 cases of 12­ ounce cans and 660 and 550 cases, respectively, of 28-ounce non­ returnable bottles (CX 1262). (57) THE COCA-COLA CO., ET AL. 565 517 Initial Decision 139. Bottlers of different brands compete to place vending machines (Sales, Tr. 1333; Brendle, Tr. 1781-82; Christian, Tr. 1852-53; Ippolito, Tr. 3267; Clements, Tr. 4026). The president of the Dr Pepper Company testified that competition for vending machine locations is a very, very intensive battle all the time" because the bottler makes an extreme effort to obtain a prime location but has no assurance that he is going to be able to hold that location from some competitor (Clements, Tr. 4026). In Washington, D. , the bottler of Coca-Cola competes with Pepsi-Cola by placing vending machines in unproductive outlets as part of its image building process (Sales, Tr. 1333). 140. Bottlers also compete to supply soft drinks and services at special events (e. school picnics, football games, etc.) in order to create consumer demand through sampling of the soft drink product as well as to build goodwill (Stokes, Tr. 1078; LaDoux, Tr. 1460; Levin Tr. 1916; DeLap, Tr. 2583; Strachan, Tr. 2895; Burks, Tr. 3018-19; Clements, Tr. 3992-93).

Competition in Merchandising 141. Effective merchandising of soft drinks requires adequate shelf space in food stores. The president of the Dr Pepper Company testified that shelf space determines "whether you live or die" in a retail account (Clements, Tr. 4000). Bottlers testified that adequate shelf space is "a means of survival" (Sheldon, Tr. 2620); "probably one of the biggest keys to sales success in a supermarket" (Burks, Tr. 3019-20); critical" (LaDoux, Tr. 1465); and a "constant problem" (Alford, Tr. 2513). L. T. Christian summarized the relationship between shelf space and sales when he testified that "(t)he larger the shelf space, the more sales I will make" (Tr. 1858).

142. Bottlers compete in home market accounts food stores, to obtain shelf and display space. They also compete in the placement of point-of-sale advertising, stock rotation, and in providing frequent service (Millard, Tr. 2324-25). (58) 143. Competition for shelf space in food stores was described as follows; "(b )loody, fierce, makes the Battle of the Bulge look like a Sunday school picnic" (Navarre, Tr. 1536); "every day battle (LaDoux, Tr. 1459); "ferocious" (Ippolito, Tr. 3266); "intensive (Ogden, Tr. 840); "constant day in and day out battle" (Millard Tr. 2325). In discussing all forms of competition among soft drink products, the president of the Dr Pepper Company described shelf space competition as "probably one of the toughest ones" (Clements Tr. 4025). To obtain more and better shelf space, bottlers of Coca-Cola conduct "space to sales" studies for food stores to persuade food store management to allocate more space to their respective products , .. Initial Decision 91 F.

(Navarre, Tr. 1540). Point-of-sale material is provided to customers free of charge (Levin, Tr. 1915), and is placed by route salesmen who periodically check the material (Rooks, Tr. 1372). 144. Bottlers also compete to place carton coolers in food stores and to set up special racks or gondolas which help sales of their respective products (Ogden, Tr. 840; Rooks, Tr. 1363; Millard, Tr. 2325; Strachan, Tr. 2892). Displays increase sales substantially (Sales, Tr. 1221; Reid, Tr. 3556-57; Clements, Tr. 4001). Bottlers also compete by providing refrigeration units free or at a nominal charge in which to store soft drinks in cartons. They are called visi-coolers or carton coolers and cost $600 to $1100 each (DeLap, Tr. 2574-75; Strachan, Tr. 2868; Connellee, Tr. 2968-69).

Competition in Providing Service 145. Another important aspect of merchandising competition among soft drink bottlers is in service they provide resellers (Sales, Tr. 1267; Brown, Tr. 168~; Levin, Tr. 1915; Ippolito, Tr. 3244). Competition in the area of service was described as "the name of the game" (LaDoux, Tr. 1464), and "a key to success" (Burks, Tr. 3008). The president of The Coca-Cola Bottling Company (59) of New York Inc. indicated the soft drink industry "was enormously dependent" on frequency of service (Millard, Tr. 2319). As one bottler of Coca-Cola stated (iJf we fall down on service in supermarkets, Pepsi starts getting the business from us and vice versa. It is a very competitive business" (Brown, Tr. 168~).

146. Service competition embraces a number of separate competi­ tive elements. First, service competition involves frequency of service in delivery of product by a route salesman. Frequency of service is necessary because, on a container basis, soft drinks are the fastest moving item in a food store (Hurst, Tr. 3511). Supermarkets and other accounts are frequently served more than one time per week (Crabtree Tr. 2664; Ippolito, Tr. 3242-43; Hurst, 3465-66; Reid, Tr. 3567). Bottlers frequently provide service to food chains on weekends, including Sundays (Crabtree, Tr. 2664; Strachan, Tr. 2892). 147. Service competition also emphasizes such tasks as cleaning and filling beverage coolers and sorting empty bottles (Christian, Tr. 1856; Levin, Tr. 1915; DeLap, Tr. 2581; Burks, Tr. 3008-09). 148. Bottlers also engage in service competition with respect to keeping vending machines supplied with soft drinks and in providing prompt, usually free, repairs for vending machines which become inoperative (Smith, Tr. 697; Carver, Tr. 1633; Levin, Tr. 1915; Connellee, Tr. 2968). Some bottlers, such as the bottler of Coca-Cola in THE COCA-COLA CO., ET AL. 567 517 Initial Decision Stockton, California, provide 24-hour-a-day repair service even on weekends and holidays (DeLap, Tr. 2581). (60) Competition in Packaging 149. Soft drink bottlers compete in packaging (Ellis, Tr. 986-89 1054-55; Carver, Tr. 1631; Levin, Tr. 1913-14; Roadcap, Tr. 2427; DeLap, Tr. 2580; Crabtree, Tr. 2681; Cobetto, Tr. 2829-30; Ippolito lfr. 3266; Clements, Tr. 4025).

150. Pressure from competitive packages forced The Coca-Cola Company in 1955 to abandon its single package (the 6 1/2-ounce returnable bottle) philosophy and add other packages (Smith, Tr. 714). Today, Coca-Cola in the United States is being sold in 42 different package types, sizes, and configurations (e. 6-packs, 8-packs) (Ellis Tr. 1054). No bottler of Coca-Cola sells all packages (Ellis, Tr. 1054). The local competitive situation varies immensely and it primarily determines which packages, sizes, and configurations are used by a particular bottler of Coca-Cola (Ellis, Tr. 1054; Roadcap, Tr. 2427). For example, in South Carolina, because of a state tax bottlers use 36­ ounce rather than 32-ounce returnable bottles to offer the best price to the consumer (Ellis, Tr. 1055).

151. Each national brand soft drink bottler must offer a wide variety of packages in order to compete (DeLap, Tr. 2580). For example, The Coca-Cola Bottling Company of New York, Inc. produces and sells a total of 82 package sizes and types (Millard, Tr. 2320). The bottler of Coca-Cola in San Antonio testified that "we will not be out-packaged" (Alford, Tr. 2483), and that he responded to the introduction of 64-ounce non-returnable bottles of Pepsi-Cola in his territory by obtaining two weeks later 64-ounce non-returnable bottles of Coca-Cola from the bottler of Coca-Cola 600 miles away in Las Cruces, New Mexico. He continued to transport 64-ounce non-returna­ bles 600 miles and sell them at the same price as Pepsi-Cola, at profit, for a period of four months until he was able to obtain equipment to produce that package size (Alford, Tr. 2483-84 2511-12). (61) 152. Pepsi-Cola recently took two percent of the home market volume away from Coca-Cola in Ada, Oklahoma by offering a package- the 64-ounce returnable bottle-which the bottler of Coca- Cola did not have available (Crabtree, Tr. 2681-82). The Havre de Grace Pepsi bottler called package competition from Coca-Cola brutal" (Connellee, Tr. 2965-66). Within two weeks after he intro­ duced a 16-ounce returnable/resealable bottle in Havre de Grace, the Coca-Cola bottler introduced it in that size (Connellee, Tr. 2966). 153. The Dr Pepper bottler in Galveston indicated there was so .._ Initial Decision 91 F.

much packaging competition that " w)e have got a packaging revolution going on right now in the (soft) drink business" (Ippolito Tr. 3266). The Dr Pepper bottler in Dyersburg, Tennessee whose territory embraces parts of the territories of four bottlers of Coca- Cola, testified that three of his competitors had 32-ounce returnable bottles of Coca-Cola on the market within two weeks after introduced the package (Burks, Tr. 3046).

Competition with New Entries 154. Over the last two decades, there has been vigorous and increasing competition from the entry of new types and brands of soft drink products (Stip. No. , CX 1244H-12441, Tr. 477-78; Brendle, Tr. 1781; Roadcap, Tr. 2450). After losing market position, The Coca-Cola Company was forced to abandon its single product philosophy around 1960 and to introduce a line of flavors and various allied products (Smith, Tr. 714-15). (62) 155. A number of entirely new types of soft drink products have entered the market, including low-calorie soft drinks (Stip. No. , CX 1244H, Tr. 477). Low-calorie drinks compete with sugar sweetened soft drinks (Millard, Tr. 2359-60). Although No-Cal and Diet-Rite were originally marketed in 1953, diet flavored soft drinks did not become a significant market factor until the early 1960's. Following the cyclamate ban in October 1969, which virtually removed low-calorie soft drinks from the market, soft drink companies began marketing saccharin substitutes which by the end of 1970 accounted for about 10 percent of total soft drink sales (Stip. No. , CX 12441; Tr. 477). Substantial numbers of low-calorie flavored carbonated soft drinks are being marketed today, including TAB (Smith, Tr. 585), diet TAB flavors (Smith, Tr. 644), Fresca (Smith, Tr. 585), diet-Pepsi (Brown, Tr. 1684), diet Royal Crown, Diet-Rite (Stokes, Tr. 1146), No-Cal (Stokes Tr~ 1146), Weight Watchers (Levin, Tr. 1899), diet Shasta (Meyers, Tr. 1714-15), Sugar Free Dr Pepper (Millard, Tr. 2344), diet Seven­ (Millard, Tr. 2344), Nestea sugar free canned iced tea (Hurst, Tr. 3450), large numbers of low-calorie private labels (Hornsby, Tr. 3158-59), and others.

156. Another new category of soft drinks which h~s been intro­ duced in the last decade is the "isotonic" or "thirst quenching" soft drink, such as Gatorade (Stip. No. , CX 12441; Tr. 477). 157. Yet another new type of soft drink is the iced tea sold by bottlers in cans (Hurst, Tr. 3451, 3478-79) and bottles (Hurst, Tr. 3525­ 26). Iced tea flavored soft drinks are manufactured by bottlers (Hurst Tr. 3474, 3478-79), distributed by bottlers (Hurst, Tr. 3477), and sold in the carbonated beverage section of supermarkets (Hurst, Tr. 3456; .. .

THE COCA-COLA CO., ET AL. 569 517 Initial Decision Reid, Tr. 3553- , 3604) and in soft drink vending machines (Burks, Tr. 3007; Reid, Tr. 3563). Tea was first test marketed in cans in 1968 (Hurst, Tr. 3458). (63) 158. Entry of new firms and brands into the soft drink industry is easy. There are numerous flavor houses from which a company entering the soft drink business can purchase syrups or concentrates (Meyers, Tr. 1734). There are also a large number of facilities available for the manufacture of soft drinks in bottles and cans which can be purchased, leased, or which will produce flavored carbonated soft drinks on a contract basis (Meyers, Tr. 1734; Hornsby, Tr. 3148). Competition among contract bottlers or canners is very tough (Hornsby, Tr. 3148 3159- 3179). There is no problem in obtaining an adequate supply of cans or bottles in which to package a new brand of soft drinks (Meyers, Tr. 1734). Personnel with experience are available in the industry (Meyers, Tr. 1734). Many new companies have entered the packaged soft drink business in the last 10 years, such as A& W Root Beer (Meyers, Tr. 1735; Connellee, Tr. 2964). 159. Many brands of soft drinks have been able to enter new markets and obtain immediate distribution in such markets at virtually no expense by entering into exclusive territorial license agreements with established bottlers already manufacturing and distributing other national brand soft drinks (Clements, Tr. 4063). By this "piggyback­ ing" on the products of an established national brand bottler, a brand attempting to enter a market capitalizes on the bottler s existing production facilities, vehicles, vending machines, sales force, and good will in a market and can obtain substantial distribution in a market in very short time (Bernabucci, Tr. 2005; Millard, Tr. 2338-40; Clements, Tr. 4063).

160. By entering into exclusive territorial license agreements with established national brand bottlers and expanding the number of its bottlers from 395 in 1961 to 512 in 1971, Dr Pepper Co. has been able to enter a substantial number of new markets and expand the geographic areas in which Dr Pepper is available from those containing 114 million people to areas with 198 million people or almost 98 percent of the population (Clements, Tr. 3983 398~7). During this period, Dr Pepper s national share of the flavored carbonated soft drink market grew from 2 to 2 1/2 percent to nearly 4 percent, and is about 5 percent today (Clements, Tr. 3981). In 1971, about 70 percent of the bottlers of Dr Pepper were licensed to sell other brands (Clements, Tr. (64) 3983­ 84). During the 1961 to 1971 period, 70 percent of Dr Pepper s growth came from the multibrand plants (Clements, Tr. 3985), and Dr Pepper grew at a rate 2 to 3 times the rate of the industry (Clements, Tr. 3981).

..

Initial Decision 91 F.

161. Similarly, Nestea canned iced tea was able to obtain distribu­ tion in areas serving approximately 90 percent of the population (H burst, Tr. 3483) in three years by entering into exclusive territorial licenses (Hurst, Tr. 3473-74) with 135 established national brand bottlers, including 55 to 60 bottlers of Pepsi-Cola, 45 to 50 bottlers of Seven-Up, and 30 bottlers of Coca-Cola (Hurst, Tr. 3477-78). In three years of licensing bottlers, Nestea has attained this distribution with the commitment of less than 20 employees (Hurst, fr. 3484) and no investment in capital equipment (Hurst, Tr. 3484-85). In the same way, Lipton canned iced tea was also able to obtain distribution in areas serving 90 percent of the population within a period of four years (Reid, Tr. 3570-71) by entering into exclusive licenses with about 200 established national brand bottlers of Coca-Cola, Pepsi-Cola, Sevenand Royal Crown (Reid, Tr. 3574). Lipton was able to obtain this distribution with the commitment of only six people on a full-time basis (Reid, Tr. 3574) and no capital investment in land, production facilities, transportation vehicles or packaging supplies (Reid, Tr. 3572­ 73).

162. Other new brands, such as Welch's Sparkling Grape Soda which "piggybacked" into the New York area by affiliating with The Coc~-Cola Bottling Company of New York (Millard, Tr. 2342-43; RX 50A-50X), Bubble-Up, which entered into exclusive territorial license agreements with bottlers including the Dr Pepper bottler of Dyersburg Tennessee (Burks, Tr. 3006-07; RX 100A-100F); Frostie root beer (RX 93A-93C; RX 115A-115C); Mason s Root Beer (RX 105A-105F); Nugrape (RX 116A-116D); and Suncrest (RX 117A-117D) have similarly been able to enter new local markets. (65) Market Shares 163. On a unit basis, sales of Coca-Cola as a percentage of total domestic food store sales of flavored carbonated soft drinks are estimated by Neilsen, the source generally relied on by the soft drink industry (Stip. No. , RX 2D), to have declined from 41.2 percent of total domestic food store sales of flavored carbonated soft drinks in 1950 to 24.4 percent in 1965 (Stip. No. , RX 2C). During that same period, sales of Pepsi-Cola increased from 15.5 percent in 1950 to 23. percent in 1960 and then declined to 19.3 percent in 1965 (Stip. No. RX 2P). In 1950, the percentage of total domestic food store sales of flavored carbonated soft drinks accounted for by Royal Crown was 5. percent; by 1965, the percentage had declined to 4.8 percent (Stip. No. , RX 2U). During the period from 1950 to 1965, the percentage of total domestic food store sales accounted for by Seven-Up ranged from 5. .. .

THE COCA-COLA CO., ET AL. 571 517 Initial Decision percent in 1950 to 10.9 percent in 1960 and back to 7.8 percent in 1965 (Stip. No. , RX 2S).

164. Beginning in 1955, Neilsen began to report data on a statistical case basis, which converts soft drinks sold to the equivalent of 24 8­ ounce containers (Stip. No. , RX 2B). On a statistical case basis between 1960 and 1971 the percentage of total domestic food store sales of flavored carbonated soft drinks accounted for by Coca-Cola declined from 22.3 percent to 20.8 percent (Stip. No. , RX 2C). During the same period, sales of Pepsi-Cola decreased from 24.6 percent to 19. percent (Stip. No. , RX 2P). The regional market share for Coca-Cola also fluctuated substantially over a period of time (Stip. No. , RX 2D­ 2E).

165. The importance of any particular package sold through food stores varies from bottler to bottler and from region to region. The sales of Coca-Cola as a percentage of total sales of that package also vary from bottler to bottler and region to region. During October and November 1971, for example, cans ranged from 18.0 percent of the physical case sales of flavored carbonated soft drinks in food stores in Coca-Cola " Region 7 (Southeast U. ) to 42.2 percent in Region 2 (metro New York area). (66) Sales of Coca-Cola as a percentage of total physical case sales of carbonated soft drinks in cans ranged from a low of 6.9 percent in Region 5 (metro Chicago area) to a high of 20. percent in Region 8 (Southwest U. ) (Stip. No. , RX 21). The same variations occur for non-returnable and returnable bottles (Stip. No. RX 2J-2K).

166. The largest selling brands of flavored carbonated soft drinks sold in various markets change over relatively brief periods of time. During the period from 1969 to 1974, Coca-Cola and Pepsi-Cola were both at times the leading brand in Detroit, Pontiac, Lansing, Flint, and Wyandotte, Michigan; Las Vegas Nevada; Kansas City, Missouri; Milwaukee, and Green Bay, Wisconsin; and Waterbury, Connecticut. Coca-Cola and Seven-Up were at different times the leading brand in Seattle Washington. Pepsi-Cola, Coca-Cola and Seven-Up were at different times the leading brand in Portland, Oregon (Stip. No. , RX 2Z2-2Z38).

Environmental Considerations Returnable v. Non-Returnable 167. From the standpoint of raw materials consumed in the manufacture of soft drink containers; the energy used in the processing of raw materials, the manufacture of containers, the filling of containers, and the transportation associated therewith; water use; solid waste generation; air pollutants; water-borne wastes; and energy effluents, the returnable bottle is the preferred soft drink container ...

Initial Decision 91 F.

even at trippage rates below ten (Nuss, Tr. 3799-3802; RX 126Q; RX 127F -127K). At approximately four trips, the returnable bottle has the same composite impact on the environment as the conventional steel can (Nuss, Tr. 3801; RX 126Z20-126Z23). A returnable bottle with trippage as low as two still has roughly the same impact as a non­ returnable bottle (Nuss, Tr. 3801; RX 126Z20-126Z23). A returnable bottle with a trippage between four and five has the same composite environmental impact as does an aluminum can which is recycled at an 80 percent rate (Nuss, Tr. 3802; RX 126Z20-126Z23). (67) 168. Nationally, on the basis of 1974 data, the average trippage of returnable bottles of Coca-Cola is approximately 14 (Teasley, Tr. 3645). Tlw Effect of Eliminating Exclusive Territories 169. If territorial exclusivity were eliminated, the chain stores would request bids for warehouse delivery from bottlers with the equipment and capacity to service their needs, thus resulting in shortterm wholesale price competition for warehouse delivery business (Smith, Tr. 659, 729-30, 739; Ellis, Tr. 991-93; Stokes, Tr. 1152-53; Navarre, Tr. 1568; Levin, Tr. 1929; Bernabucci, Tr. 1988-89; Millard Tr. 2375; Roadcap, Tr. 2464--66).

170. A short-term or temporary wholesale price reduction might result from wholesale price competition for warehouse delivery of non­ returnable containers, but only long enough to force the small bottlers out of business and reduce competition (Smith, Tr. 659; Stokes, Tr. 1151-52; Navarre, Tr. 1571; Levin, Tr. 1936; Strachan, Tr. 2900; Hurst Tr. 3502).

171. Even with lower wholesale prices for soft drinks, there is no assurance that the chain stores would pass this reduction on to the consumer in the form of lower retail prices (Smith, Tr. 739; Ellis, Tr. 1001-02, 1022; Stokes, Tr. 1153; Sales, Tr. 1286; Navarre, Tr. 1571-72; Millard, Tr. 2377; Alford, Tr. 2525--27 2543; Strachan, Tr. 2899; Hurst Tr. 3488--89; Clements, Tr. 4040; RX 23, RX 24). The president of The Coca-Cola Company testified that it was "speculative" to predict lower retail prices will result from lower wholesale prices (Smith, Tr. 731 739, 740). In the early 1960' , during an experiment with warehouse delivery of cans by bottlers of Coca-Cola in the San Francisco area to Safeway, Lucky, and Purity, Coca-Cola and Pepsi-Cola retailed at the same price in chain stores irrespective of the fact that the wholesale price for Coca-Cola in cans delivered to the warehouse was lower than the price of Pepsi-Cola in cans delivered directly to retail stores (DeLap, Tr. 2572; Sheldon, Tr. 2626, 2631). In another instance, Alpha Beta chain stores in parts of California in May 1975, received Coca-Cola in cans through a Los Angeles (68) warehouse, but there was no retail ...

THE COCA-COLA CO., ET AL. 573 517 Initial Decision price difference between warehouse delivered Coca-Cola in cans and store-door delivered canned Pepsi-Cola (DeLap, Tr. 2585; Sheldon, Tr~ 2632-34). Also, there was no difference between the retail price for Coca-Cola in cans in Alpha-Beta Stores and the price for stores receiving Coca-Cola in cans by route delivery (DeLap, Tr. 2589). 172. In June 1975, Delaware Punch and Dad's Root Beer were delivered to a Houston warehouse and sold in supermarkets in Galveston, Texas (Ippolito, Tr. 3239--41). Despite a lower wholesale price, which for Delaware Punch was approximately 90 cents pe~ case lower (Ippolito, Tr. 3242), these products were priced by the chain stores at the same retail price as the store-door delivered national brands such as Coca-Cola and Pepsi-Cola (Ippolito, Tr. 3241). The Dr Pepper Company was forced by the chain stores to reduce the wholesale price of Dr Pepper by a minimum of $.50 before the chain stores would accept Dr Pepper for warehouse delivery in Los Angeles. Nevertheless, Dr Pepper was sold in chain stores at the same retail price as store-door delivered Coca-Cola (Clements, Tr. 3997- , 404fr­ 47).

173. Chains take lower margins on private label soft drinks to maintain the price differential between private labels and national brands and to thereby indicate to consumers that all of their private label products are cheaper than national brand products (Alford, Tr. 2527). With respect to food products generally, including products delivered through chain warehouses, the National Commission on Food Marketing found that the advertised brand retail price was about percent higher than the private label product with which it competed (RX 106F). For example, food chains keep a 30 percent differential between Minute Maid frozen concentrate orange juice and their private labels to establish the "bargain" value of their private labels (Smith, Tr. 730-31). And, in Baltimore, chains refused to sell Coca-Cola in flat-top cans on a loose pack basis because the low price (69) would cut into their private label business (Ellis, Tr. 1001-02). Even with lower wholesale prices for Coca-Cola and other national brand soft drinks delivered to warehouses, chain supermarkets are not likely to reduce their retail prices for national brands because of their usual practice of maintaining the price spread between their private label brands and national soft drink brands (Smith, Tr. 730-31; Ellis, Tr. 1001-03; Navarre, Tr. 1571-72).

174. With the elimination of exclusive territories and a switch from store-door to warehouse delivery to chain supermarkets, bottlers of Coca-Cola products and of other brands of soft drinks would be unable to engage in effective merchandising activities and merchandising functions would be substantially curtailed or eliminated (Ellis, Tr. Initial Decision 91 F. C.~ 1005; Sales, Tr. 1317-18; LaDoux, Tr. 1474; Christian, Tr. 1864; Bernabucci, Tr. 1992; Strachan, Tr. 2901; Connellee, Tr. 2982). 175. Bottlers of Coca-Cola charge the same price to all their customers; but if bottlers lost chain store outlets which are the high volume and high profit accounts to warehouse delivery, they would be obliged to cut back service to small accounts or to raise prices, either of which would reduce volume (Smith, Tr. 659- , 736). 176. Without exclusive territories, the price of soft drink products would have to increase in the smaller non-chain store accounts still serviced by bottlers (Smith, Tr. 659, 736; Stokes, Tr. 1152; Levin, Tr. 1929; Filoromo, Tr. 2187-88; Sheldon, Tr. 2637; Connellee, Tr. 2980; Hurst, Tr. 3500-01; Clements, Tr. 4036).

177. Without exclusive territories, bottlers would reduce the frequency of service to those small outlets which they continued to serve (Ellis, Tr. 1000, 1003-04; Rooks, Tr. 1401; Brendle, Tr. 1784; Bernabucci, Tr. 1991-92; Alford, Tr. 2520; Ippolito, Tr. 3272). 178. Without exclusive territories other small accounts would not be served at all (Stokes, Tr. 1077- , 1152; Sales, Tr. 1278--79; LaDoux Tr. 1473; Levin, Tr. 1932-34; Filoromo, Tr. 2188; Alford, Tr. 2520-21; Burks, Tr. 3054-55; Clements, Tr. 4038, 4074-77) because the higher incremental cost of servicing those accounts by route trucks would be prohibitive without the large volume outlets lost to warehouse delivery (Smith, Tr. 659-60; Rooks, Tr. 1404-05; Filoromo, Tr. 2188; Clements Tr. 4036). (70) 179. Without exclusive territories, a substantial number of soft drink brands and flavors would be eliminated in local markets (Smith Tr. 730-32; Sales, Tr. 1282, 1317-18; Navarre, Tr. 1550--51; Brendle, Tr. 1785; Levin, Tr. 1932-33; Hurst, Tr. 3491 3503; Clements, Tr. 4048--50). 180. The chains already want fewer brands and flavors and would cut out slower moving brands if they had warehouse delivery and could better determine a brand's sales (Ellis, Tr. 1007; Navarre, Tr. 1550--51; Clements, Tr. 4048--50). Many chains have refused to handle slower moving products such as TAB flavors in New York (Millard, Tr. 2377­ 78); Crass flavors in Washington (Sales, Tr. 1232); Suncrest and Nugrape in Galveston (Ippolito, Tr. 3259); and K-S Canning Compa­ s own brands (Hornsby, Tr. 3169-70). The Kroger chain in Dyersburg, Tennessee (Ippolito, Tr. 3256--57) and the- Handy Andy chain in San Antonio (Alford, Tr. 2542) both threw out Shasta when they introduced their own private labels. As many national brand bottlers go out of business, the lesser known secondary brands which had "piggybacked" into local markets by licensing such bottlers would be forced out of local markets (Hurst, Tr. 3490-91). Moreover, many bottlers seeking to survive in a warehouse delivery environment would THE COCA-COLA CO., ET AL. 575 517 Initial Decision 0,-.. cut their slower moving brands (Millard, Tr. 2370-71; Reid, Tr. 3582; Clements, Tr. 4048).

181. Even better known brands such as Seven-Up and Dr Pepper might not survive in many local markets (Sales, Tr. 1220, 1317; Navarre, Tr. 1550-51). Smaller trademark licensors, such as the Dr Pepper Company (Clements, Tr. 4048-50) and Thomas J. Lipton Co. (Hurst, Tr. 3491) would be placed in economic peril as availability of their products in many markets was reduced or eliminated entirely. The Nestle Company, for example, estimates that 60 percent of the N estea bottlers would go out of business without exclusive territories for national brands (Hurst, Tr. 3500). Many of the weaker allied products of large licensors, such as The Coca-Cola Company s Fanta (Hurst, Tr. 3491-92) and Fresca (Smith, Tr. 610), would also be forced out of local markets (Hurst, Tr. 3491-92).

182. The long term result of eliminating territories would be that . the strongest three or four national brands in each local market and private labels would survive (Sales, Tr. 1220) and ease of entry would diminish. The president of Dr Pepper Company said with respect entry into new local markets "no one else will ever do it again (Clements, Tr. 4050-51). (71) 183. Nationally hundreds of bottlers of Coca-Cola and other brands would go out of business if exclusive territories were determined to be unlawful (Smith, Tr. 615, 732-33; Clements, Tr. 4037-38). The number of bottlers would be reduced to a fraction of the number that would otherwise exist under the present system (Smith, Tr. 615). Small bottlers of Coca-Cola and other national brands testified that they would go out of business in a short period of time if territorial exclusivity were declared illegal (Rooks, Tr. 1399-1402, 1405-06, 1412 1418-19; LaDoux, Tr. 1472; Carver, Tr. 1551, 1637-38; Brown, Tr. 1699­ 1700; Brendle, Tr. 1782-84; Christian, Tr. 1863, 1865-66, 1873, 1876; Levin, Tr. 1928-29; Bernabucci, Tr. 1992-93, 1995; Cameron, Tr. 2074 2098-99; Filoromo, Tr. 2185-86;. Massey, Tr. 2238; Roadcap, Tr. 2457­60).184. Large bottlers and other witnesses agreed that such small bottlers and some large bottlers would be forced out of business territories were declared illegal (Smith, Tr. 614-15, 732-34, 807-m~; Ellis, Tr. 1008-09, 1011, 1041, 1045; Sales, Tr. 1284, 1317; Navarre, Tr. 1570-71; Alford, Tr. 2515, 2517-19; Hurst, Tr. 3490, 3499-3500; Reid Tr. 3581-82; Clements Tr. 4035-38, 4048-50). The fate of the small bottler was variously described as "doomed would go down the drain " (Sales, Tr. 1284, 1317); "simply be a matter of time completely out of business" (Rooks, Tr. 1402, 1412); "disappear rather immediately" (Navarre, Tr. 1551); "total disaster" (Brendle, Tr. 1782); Initial Decision 91 F. C. disastrous" (Sheldon, Tr. 2636); "doomsday" (Connellee, Tr. 2976); doomed just a matter of time" (Burks, Tr. 3048). 185. Without exclusive territories, large bottlers of Coca-Cola and other national brands would take small bottlers' chain store business because chain store warehouses are located mainly in territories large bottlers (Rooks, Tr. 1399; LaDoux, Tr. 1472; Brendle, Tr. 1782-84; Levin, Tr. 1928--29; Filoromo, Tr. 2186-2187; Massey TF. 2238; Roadcap, Tr. 2458-60; DeLap, Tr. 2590; Sheldon, Tr. 2636; Crabtree, Tr. 2684; Burks, Tr. 3092-93; Clements, Tr. 4036). The loss- of the chain store accounts, (72) comprising so large a part of the bottler s volume, would put small bottlers out of business (Smith, Tr. 732-33; LaDoux, Tr. 1472; Brendle, Tr. 1782-1783; Christian, Tr. 1873; Levin, Tr. 1929; Bernabuc­ , Tr. 1992-93; Filoromo, Tr. 2186-87; Alford, Tr. 2517-18). Because soft drinks are a high volume, low mark-up product, volume is critical. For example, the bottler of Coca-Cola in San Antonio made a study that showed that a loss of 15-20 percent of its volume would put it at a break-even point due to its fixed costs (Alford, Tr. 2518). 186. Small bottlers of Coca-Cola are located close to large bottlers. For example, the Westminster, Maryland bottler, surrounded by bottlers of Coca-Cola in Washington, D. , Baltimore, Maryland, and Wilmington, Delaware, testified that "(iJt is sort of like the flea and the elephant" (Roadcap, Tr. 2459). This situation applies to many small bottlers of Coca-Cola. For example, there are at least six large bottlers of Coca-Cola within 150 miles of the Hartwell, Georgia bottler of Coca- Cola (Rooks, Tr. 1359-60). The Spirit Lake, Iowa bottler is located within 90-120 miles of the Des Moines, Iowa bottler of Coca-Cola as well as within 90-120 miles of a plant owned by the Minneapolis bottler of Coca-Cola (LaDoux, Tr. 1448-49). The Annapolis bottler of Coca- Cola is surrounded by the Baltimore territory on one side and the Washington territory on the other side (Brendle, Tr. 1773). The bottler of Coca-Cola in LaCrosse, Wisconsin is surrounded on three sides by large bottlers with their own canning lines (Bernabucci, Tr. 1976-77). The bottler of Coca-Cola in Coatesville, Pennsylvania is located near the Philadelphia bottler owned by Associated Coca-Cola Bottling Company and "within arm s reach" of the bottler in Baltimore (Filoromo, Tr. 2185). Family-owned bottlers of Coca-Cola within 75 miles of The Coca-Cola Bottling Company of New York, Inc., include Pittston, Pottsville, Reading, Coatesville, Williamsport, York, and Lancaster, Pennsylvania, and North Hampton and Pittsfield, Massa­ chusetts (Millard, Tr. 2362). There are 13 bottlers of Coca-Cola with annual case sales ranging from 100 000 to 700 000, located within 150 miles of the Coca-Cola Bottling Company of San Antonio, Texas, which had case sales of 5 500 000 in 1974 (Alford, Tr. 2480, 2514). The Coca­ ...

THE COCA-COLA CO., ET AL. 577 517 . Initial Deci~ion Cola Bottling Company of the (73) Peninsula is completely surrounded by the Coca-Cola Bottling Company of California (Sheldon, Tr. 2636). The bottler of Coca-Cola in Ada, Oklahoma is within 120 miles of the large bottler of Coca-Cola in Tulsa and 32 miles from a distribution center of the OklaJ~oma City bottler of Coca-Cola (Crabtree, Tr. 2683). 187. Some large bottlers, because of their proximity to food chain warehouses and because of their greater financial resources, would be in a financial position to underprice small bottlers for a short period on cans and non-returnable bottles, and thereby take away enough of the small bottler s volume to put him out of business (Ellis, Tr. 1018-19; Carver, Tr. 1613, 1687-38; Levin, Tr. 1936-38; Bernabucci, Tr. 1988; Filoromo, Tr. 2185-87; Roadcap, Tr. 2458-59; Alford, Tr. 2515; Connellee, Tr. 2976-78; Burks, Tr. 3052 3091-93; Ippolito, Tr. 3275; Reid, Tr. 3582).

188. A small bottler could not supply chain store warehouses by allowing "backhauling (i. the buyer takes delivery at the supplier point of shipment) from his bottling plant, because the chain store truck servicing the few stores in that territory would not have enough room to pick up a significant supply on a backhaul (Rooks, Tr. 1417; Christian, Tr. 1843; Roadcap, Tr. 2458-59). An empty tractor-trailer can only accommodate 2 000 cases of 12-ounce cans (Meyers, Tr. 3175). Large bottlers could still deliver to a nearby chain warehouse at a lower cost than the chain transport could pick up on a backhaul. As the former president of the Coca-Cola Bottling Company of Baltimore testified: "It is not free to take a great big transport into a bottling plant in Westminster, pull up to their dock, wait to be loaded, you know, loaded on the dock, and the time that is wasted there, it is not free. I am quite sure that we could deliver from Baltimore to the A&P warehouse cheaper than they could pick up from Westminster on their trucks and deliver. And I think that will be the deciding factor. " (Ellis Tr. 1015.) (74) 189. Because price is the determining factor in selling to the chains and because the per case cost of delivering a transport full of Coca- Cola from a plant in Baltimore to a chain store warehouse would only be 2-3 cents a case, the Baltimore plant could take away chain business from any bottler allowing plant pick-up simply by dropping the per case price by a nickle (Ellis, Tr.

190. Small bottlers could not survive by merging their plants with 1018-19).other small bottlers or by entering into cooperative arrangements (Ellis, Tr. 1045-46; Rooks, Tr. 1418-19; Navarre, Tr. 1570-71; Levin, Tr. 1938-39; Filoromo, Tr. 2139-41, 2214; Roadcap, Tr. 2467). All that would result is "just a bunch of small bottling'plants together that will all go out of business " (Ellis, Tr. 1045) and thereby obtaining a "larger , Initial Decision 91 F.

percentage of nothing" (Rooks, Tr. 1409). Many small bottlers, such as Westminster, Maryland, are surrounded by large bottlers and there­ fore have no one with which to form a cooperative (Roadcap, Tr. 2467). Small bottlers in a region do not together have the financial resources to withstand the underpricing of large bottlers. For example, the combined sales of the small bottlers in the Philadelphia area "don come close" to the sales of the Philadelphia Coc~-Cola Bottling Company (Levin, Tr. 1938-39). As the Coatesville, Pennsylvania bottler of Coca-Cola also testified * we could not together merge a large enough factor to even be competitive with - Philadelphia and Baltimore (Filoromo, Tr. 2214).

191. Because it would cost $5 million to $10 million to build the facilities necessary to compete for chain warehouse business, small bottlers acting jointly would still not have the necessary resources to compete with large bottlers (Rooks, Tr. 1419). Joint buying of items such as glass bottles would produce no savings because the glass producers do not provide volume discounts (Filoromo, Tr. 2141). Because of costs arising from shipping distances, a cooperative would not achieve the savings on non-returnables necessary to compete with large bottlers (Ellis, Tr. 1045-46), since the chain store warehouses are in closer proximity to the larger bottlers' plants (Rooks, Tr. 1415-16). (75) 192. Small bottlers would not survive by being acquired by large bottlers (Smith, Tr. 787; Rooks, Tr. 1409- , Brown, Tr. 1703; Alford Tr. 2559; Crabtree, Tr. 2688-89). There is no incentive for a large bottler to purchase a small bottler since the large bottler instead could sit around and simply take his (the small bottler s) business" (Rooks Tr. 1409). The large bottler would not be interested in using the small bottler as a distributor (Rooks, Tr. 1410-11, 1420-21; Brown, Tr. 1703) because many large bottlers already have distribution centers close to the small bottlers' territories (Crabtree, Tr. 2688); such use would entail purchasing a very expensive manufacturing facility in order to convert it to the distribution warehouse (Crabtree, Tr. 2688); and there would be a decreased need for local distribution in light of the demise of the returnable bottle (Crabtree, Tr. 2688-89). Acquiring the small bottler for use as a marketing bottler is "a theoretical option of no economic value" (Smith, Tr. 788).

193. Many large bottlers would be forced out of business since they do not have the resources to compete against bottling plants owned by the largest bottlers and by subsidiaries of the syrup companies (Ellis Tr. 1039-40; Stokes, Tr. 1162; Sales, Tr. 1279-80; Navarre, Tr. 1543-46 1570; Cameron, Tr. 2061-62; Roadcap, Tr. 2460; Clements, Tr. 4051). The Coca-Cola Company and other large national soft drink companies ....

THE COCA-COLA CO., ET AL. 579 517 Initial Decision would survive because of their resources and diversification (Sales, Tr. 1280, 1282; Navarre, Tr. 1546; Levin, Tr. 1936-37). The natural consequence of the demise of more and more bottlers is that ultimately the bottling business would become concentrated in the hands of a very few bottlers for each brand (Smith, Tr. 733, 738). The very few bottling companies would be owned by large companies, publicly held corpora. Tr. 1281-82). (76)tons, and food chains (Clements, Tr. 4051; Sales, 194. The economic impact on local communities would be substan. tial with the elimination of exclusive sales territories. Many people employed by bottlers of Coca-Cola and other national brands which would be forced to curtail services or close their doors would be out of jobs, as has happened in the case of small local bakeries and dairies (Sales, Tr. 126~67, 1294; Levin, Tr. 1924-25). A large number of the approximately 52 000 employees of bottlers of Coca.Cola (RX 35C), many of them long-term employees (Carver, Tr. 1620; Filoromo, Tr. curtailed or2212-13), would be discharged as route delivery is eliminated (Smith, Tr. 733; Sales, Tr. 1317; Clements, Tr. 4044). 195. Contributions to the economic health of many communities by soft drink bottling companies through payment of taxes, purchase of supplies, and borrowing of money would also decline (Cameron, Tr. 2062; Crabtree, Tr. 2686; Cobetto, Tr. 2840-41; Ippolito, Tr. 3271-72). Many small towns, of which Coatesville, Pennsylvania (Filoromo, Tr. 2212-13), Ada, Oklahoma (Crabtree, Tr. 2686), Herminie, Pennsylvania (Cobetto, Tr. 2840-41), Washington, Pennsylvania (Cameron, Tr. 2062), and Montross, Virginia (Carver, Tr. 1620), are typical, would find one of their larger employers out of business.

DISCUSSION The Legal Theory on Which the Decision Is Not Based Complaint counsel specifically disavowed reliance upon a per se case (see Northerntheory in offering evidence at the hearing of this Pacific Railway Co. v. Uniwd Staws 356 U.S. 1 (1958)), preferring rather that the acts alleged to violate Section 5 of the Federal Trade Commission Act be judged according to the rule of reason (Complaint Counsel' s Proposed Findings pp. 7-8). Complaint counsel' s disavowal is not binding on me; however, in my view, complaint counsel took the proper course, and it is appropriate to explain why the per se approach is inappropriate and the rule of reason should be applied. (77) If the bottlers had agreed as to their exclusive territories, this case would involve horizontal territorial market allocations those agreed upon by competitors at the same level of the market structure Staw8 v. TCYpCoand such action would be illegal per Be. Uniwd . ,. . . . ... Initial Decision 91 F.

Associates, Inc. 405 U.S. 596 , 608 (1972). But there is no evidence in this record of an agreement or understanding among the bottlers to divide markets. The evidence shows instead that the exclusive territorial provisions of their contracts were vertically and unilaterally imposed, that is Coca-Cola" acted alone in imposing the restrictions and it alone determined the territories. Further, there is no evidence that the territories were agreed upon by bottlers who then had "Coca- Cola impose " them. See United States v. Sealy Inc. 388 U.S. 350 352-354 (1967).

The fact that "Coca-Cola" has acquired some - bottlers does not transform the system into a horizontal agreement in illegal restraint of trade. See White Motar Co. v. United States 372 U.S. 253 at 258 (1963) where White in its agreements with its dealers retained the exclusive right to sell . trucks to Federal and State governments or any department or political subdivision thereof " and the Supreme Court declined to base its decision on a per se theory. Vertical territorial allocations such as those disclosed on this record are not illegal per se. After declining to intimate its view as to the legality of a vertical arrangement by a manufacturer restricting the territory of his distributors and dealers, the Supreme Court in White Motar, supra 372 U.S. at 263, distinguished between horizontal and vertical restraints: (78) Horizontal territioriallimitations, like group boycotts or concerted refusals by traders to deal with other traders are naked restraints of trade with no purpose except stifling of competition. A vertical territorial limitation mayor may not have that purpose or effect. . . . They may be too dangerous to sanction or they may be allowable protections against aggressive competitors or the only practicable means a small company has for breaking into or staying in business and within the rule of reason. I t also is well established that a territorial representation agreement between a supplier and his customer, which gives the latter exclusive rights in a given area, does not inevitably violate the Sherman Act. See Schwing A/otar Co. v. Hudson Sales Corp. 138 F. Supp. 899 902 (D. Md. 1956), aiid 239 F.2d 176 (4th Cir. 1956), cert. denied 355 U.S. 823 (1957); Paa:kard Motor Car Co. v. Webste' r Motor Car Co. 243 F.2d 418 420 (D. C. Cir. 1957), cert. denied 355 U. S. 822 (1957); The Coca-Cola Bottling Co. v. The Coca-Cola Co. 269 Fed. 796 (D. Del 1920). The Supreme Court held in United States v. Arrwld, Schwinn Co. 388 S. 365, at 376 (1967), referring to United States v. Colgate Co. 250 S. 300 (1919), that if equivalent, competitive products are readily available in the market, an exclusive representation agreement between supplier and customer is not in and of itself unlawful. The record in this proceeding amply demonstrates that there are numerous equivalent brands of soft drinks available both nationally and more y. , . . . THE COCA-COLA CO., ET AL. 581 517 Initial Decision importantly in the markets comprising bottler s territories (Findings 92-102).

The Court did state in Schwinn that "the Sherman Act was violated when the manufacturer imposed upon the dealers to whom the manufacturer sold goods, restrictions as to the customers to whom the dealers could resell: "(IJt is unreasonable, without more, for a manufacturer to seek to restrict and confine areas or persons with which an article may be traded after the manufacturer has parted with dominion over it." 388 U.S. at 379. But the Court added that " (79) the vertically imposed distribution restraints absent price fixing and in the presence of adequate sourtes of alternative products to meet the needs of the unfranchised-may not be held to be per se violations of the Sherman Act." 388 U.S. at 381.

I do not believe the "unreasonable without more" language Schwinn does or that it was intended to apply to restrictions imposed incidental to the grant of a trademark license such as the one "Coca- Cola" and the other national bottlers use. In fact, in Schwinn, supra 388 U. S. at 379 n. , the Supreme Court expressly declined to consider whether a patentee has any greater rights to reserve control over the destiny of the product manufactured or the conditions of its resale. And more recently a U. S. District Court Judge held in Jack Winter Inc. v. Koratron Co., Inc. 375 F. Supp. 1, 63 n.87 (N. D. Cal. 1974), that the question of a trademark licensor rights fall within this reservation. Cf. p. 31, Report of Ad Hoc Committee on Franchising, FTC Staff Report at 31 (June 2 1969) (RX 109Z5). Coca-Cola" has not parted with dominion over the product it sells and it does not sell its bottlers a finished product for resale. Rather, it sells (a) the right to use its trademark and (b) a syrup or concentrate made from a secret formula. The bottler manufactures a finished product according to the licensor Coca-Cola " -specifications using its own facilities and supplies. In addition, in the agreem~nts with its bottlers Coca-Cola" reserves the right to make quality inspections and these reasons the "unreasonable without more per se- language in Schwinn is not controlling in this case. (80) There is no doubt that if a dealer in or.a distributor of a product is insulated from sales by contiguous or other dealers observing the territorial restraints a manufacturer has imposed, the restraint raises questions which have arisen for many years under the antitrust laws and Section 5 of the Federal Trade Commission Act. See Dr. Miles Medical Co. John D. Park Sons Co. 220 U.S. 373, 404-408 (1911); Snap- On Tools Corp. v. FTC 321 F.2d 825 (7th Cir. 1963); Sandura Co. v. FTC 339 F.2d 847 (6th Cir. 1964). A restraint is illegal per se if it has such a pernicious effect on competition and is so lacking of any , .

Initial Decision 91 F.

redeeming virtue that it is conclusively presumed to be unreasonable without elaborate inquiry as to the precise harm caused. (See Northern Pacific Ry. Co., supra 356 U. S. at 5.) Agreements to fix prices, group boycotts, tying arrangements and horizontal divisions of markets by competitors, for several examples, none of which are present here have been held to be per se violations. A unilaterally imposed territorial sales restriction which forms a part of the scheme or agreement to fix prices is violative per se of the antitrust laws. Schwinn, supra 388 U. at 373. The facts brought to light here do not reveal an effect on competition so pernicious as to warrant per se treatment. The Commission s view is succinctly stated in Holiday Magic, Inc. Dkt. 8834, (84 F. C. 748 at 1057) 3 TRR ~20 757 at 20 623 (1974): Here, as in Adolf Coors Co. (Dkt. 8845, (83 F. C. 32) 3 TRR ~20 403 (1973) J imposition of exclusive territories was accompanied by price fixing, and that combination renders the use of exclusive territories illegal per se. " But there is no evidence in this case that there was any scheme or agreement on the part of "Coca-Cola" or effect from its actions which fixed prices (Findings 34 and 66). What is more Coca- Cola" did not and does not (1) tie its licenses to the purchase of other products the company might supply (Finding 66), (2) preclude bottlers from manufacturing other brands of soft drinks (Finding 70), or (3) otherwise limit the business activities of its bottler licensees. (81) Respondents' bottlers not only were and are free to sell in accordance with their own judgment as to what the price was and is to , but also are free to conduct their business as they see fit. It is worthy of mention that no bottler witness who testified said that the territorial limitations infringed on his freedom to operate his business as he thinks is best.

Since the evidence in this matter contains no persuasive evidence suggesting that a per se approach should be used, it has been decided under the rule of reason. That is the basis on which the litigative hearings were held and that is the basis on which this initial decision has been prepared.

The Rule of Reason The rule of reason is the alternative to the per se theory for deciding whether a trade restraint is legal. The rule of reason concept stems from Standard Oil Co. v. Uniwd Staws 221 U. S. 1 , 62 (1911). Chicago Board of Trade v. Uniwd Staws 246 U.S. 231 (1918), which was brought under the Sherman Act, is, and again was very recently recognized to be, the landmark case defining the parameters of the rule of reason. (See American Motor Inns, Inc. v. Holiday Inns, Inc. , . . , . . ... . THE COCA-COLA CO., ET AL. 583 517 Initial Decision 365 F. Supp. 1073 (D. J. 1973), afi'd in part, rev d in part 1975-2. Trade Cases, ~60 390, at 66 721 (3rd Cir. 1975). In contrast with its absence here Chicago Board of Trade involved a form of horizontal price fixing by reason of the Board' s rule holding a price firm from the close of business one day to the start of business the next. In the decision, however, Justice Brandeis explained the ground rules for determining whether restraint on trade reasonable (246 U. S. at 238): (82) Every agreement concerning trade, every regulation of trade, restrains. To bind, to restrain, is of their very essence. The true test of legality is whether the restraint imposed is such as merely regulates and perhaps thereby promotes competition, or whether it is such as may suppress or even destroy competition. To determine that question the court must ordinarily consider the facts peculiar to the business to which the restraint is applied; its condition before and after the restraint was imposed; the nature of the restraint, and its effect, actual or probable. The history of the restraint, the evil believed to exist, the reason for adopting the particular remedy, the purpose or end sought to be obtained, are all relevant facts. It should be noted that although Chicago Board of Trade involved charges brought under the Sherman Act, it was established long ago that charges under that Act are cognizable under Section 5 of the Federal Trade Commission Act (15 U. C. 45), FTC v. CerMnt Instituu 333 U.S. 683 (1948). "The Federal Trade Commission Act may be construed in pari materia with the Sherman and Clayton Acts. 'This construction allows for using cases decided under any of the antitrust laws in cases brought by the Commission. Atlantic Refining Co. Federal Trade Commission 344 F.2d 599, 606 (C.A. 6), cert. denied 382 S. 939. ArMrican Cyanamid Co. , et al, V. Federal Trade Commission 363 F.2d 757, 770 (6th Cir. 1966).

Consequently, a violation of Section 5 may be found where the "basic policies" of the Sherman Act or the Clayton Act are violated. Beyond this, if the evidence brought forth at the adjudicative hearing warrants it, even though an actual violation of those statutes is not found, Section 5 of the FTC Act still may have been violated. Golden Grain Macaroni CO. V. FTC 472 F.2d 882, 886 (9th Cir. 1972), citing FTC V. Brown Shoe Co. 384 U.S. 316 (1966). However, the Commission has made it clear that, in its view . (NJot every method of competition which involves some restriction on competition is an unfair method of competition ' under the FTC Act" Sarulura, supra 339 F.2d at 849. (83) In making the determination as to whether a method of competition is unfair . the Federal Trade Commission does not arrogate excessive power to itself if, in measuring a practice against the elusive but congressionally mandated standard of fairness, it, like a court of Initial Decision 91 F.

equity, considers public values beyond simply those enshrined in the letter or encompassed in the spirit of the antitrust laws. FTC Sperry Huuhinson Co. (S&H), 405 U.S. 233, 244 (1972). The quoted language in S&H cuts both ways. That is, the Commission may find or may not find a violation by considering public values beyond those enshrined" in the antitrust laws. Thus, the Commission is not limited necessarily to consideration of the criteria which Justice Brandeis listed in deciding a case under the rule of reason and has every right to act as a court of equity. For this purpose, the Commission has described guidelines as to what it considers in determining whether a restraint on trade should be held to be violative of Section 5. These guidelines are:

(1) whether the practice, without necessarily having been previously considered unlawful, offends public policy as it has been established by statutes, the common law, or otherwise-whether, in other words, it is within at least the penumbra of some commonlaw, statutory, or other established concept of unfairness; (2) whether it is immoral, unethical, oppressive, or unscrupulous; (3) whether it causes substantial injury to consumers (or competitors or other businessmen). FTC v. Sperry Hutchinson Co. (S&H), 405 U. S. 233, 244-45 n.5 (1972), quoted from Statement of Basis and Purpose of Trade Regulation Rule 408 (Unfair or Deceptive Advertising and Labeling of Cigarettes in Relation to the Health Hazards of SmokingJ, 29 F.R. 8324 8355. (84) Clearly, territorial exclusivity is within the "penumbra" of a concept of unfairness because it does impose a restraint on trade in that a businessman s geographical area of competition is limited. Equally clear is the fact that under the Commission s guidelines territorial exclusivity is legal if it is not immoral, unethical, oppressive or unscrupulous and does not cause substantial injury to consumers competitors or other businessmen. In my view, the territorial exclusivitysubstantialCoca-Cola bottlersinjuryenjoy is nonetoofanyone.these and does not cause In keeping with the concept of its sitting as a court of equity, the Commission declined in S&H (supra) to appraise the allegedly traderestraining conduct in that case on the basis of Schwinn. In its S&H opinion, and in harmony with the view that there is a difference. between the sale or transfer of a bicycle to a dealer merely for resale and the sale of syrup or concentrate to a bottler f()r manufacturing another product, the Commission said:

(WJe do not believe it appropriate to decide the broad competitive questions presented in the record on the narrow and technical basis of a restraint on alienation. The circumstances here are much different from that where products are transferred to a dealer for resale. They are complicated by the nature of the trading stamp scheme. It is essential in this matter, we believe, . . . to determine whether or not there has been or may be an impairment of competition. Thus we intend to look at the substan~ of the ...

THE COCA-COLA CO., ET AL. 585 517 Initial Decision allegedly illegal practice rather than to decide the case by application of a technical formula. S&H 405 U.S. at 247-248 n.

(85) As in S&H the circumstances in this case also are complicated by many factors, including the investments bottlers make, the fact that most of them are relatively small, local businessmen, the nature of Coca-Cola " trademark licensing, and its quality-control practices involving inspections, product testing and related activities. Is tM Territorial Restraint Reasonable? The fact that the territorial restraint is not horizontally agreed upon or coupled with price fixing does not obviate the possibility that the restraint is illegal. This, because it might unreasonably restrain trade in some other way.

The territorial restriction here allows bottlers to enjoy area exclusivity while prohibiting them from shipping Coca-Cola outside their territories and from selling to resellers who may do so. As such, it does eliminate intrabrand competition between Coca-Cola bottlers but the restriction is not necessarily unreasonable for that reason. Snap- Tools, supra 321 F.2d at 831-832; Sandura, supra 339 F.2d at 858-S59; American Motor Inns, supra 1975-2 Trade Cases ~60 390 at 66 720­ 722. Application of the rule of reason was not confined to intrabrand competition in Schwinn, supra 365 U.S. at 382. Consequently, consideration must be given to still other factors. The initial question to answer is whether the restraint is legal in its purpose. Chicago Board of Trade, supra 246 U.S. at 238. Then, if (1) the purpose is to further legitimate business goals rather than to hinder competition, (2) the restraint is no broader than is reasonably necessary, and (3) it promotes competition by merely regulating it, the restraint is legal. See S. v. Addyston Pipe Swel Co. 85 Fed. 271 (6th Cir. 1898), afi'd 175 U. S. 211 (1899); S. v. Columbia Pictures Corp. 189 F. Supp. 153 at 178 (S. Y. 1960); American Motor Inns supra 1975-2 Trade Cases ~60 390 at 66 722.-66 723. (86) The evidence shows that the territorial exclusivity provision was adopted by "Coca-Cola" for the legitimate purpose of broadening the distribution of Coca-Cola by persuading local businessmen to invest the capital, time and ability required to manufacture and sell Coca-Cola in bottles (Findings 28--31). The vehicle-round-trip-in-one-day measure­ ment "Coca-Cola" used for setting the territorial boundaries (Smith Tr. 681) was both reasonable and practical and no broader than necessary to provide the bottler licensee with a territory he could serve and develop. Likewise, the evidence shows that exclusive bottler territories continue to serve "Coca-Cola " legitimate business needs by Initial Decision 91 F. C,. encouraging a high level of continuing capital investment by bottlers and maintenance of the intensive merchandising and servicing efforts which preserves the widespread availability and high sales of Coca- Cola products (Findings 137-140). Territorial exclusivity also encourag­ es use of returnable bottles (Findings 120-121) which are preferable from the standpoint of economy and environmental impact (Findings 110-126, The territorial167-168).provision reasonably regulates competition because it enables the bottler licensee to focus on his own territory and the strong interbrand competition he faces within that marketing area. With a protected territory the bottler need not be concerned over his fellow Coca-Cola bottler licensees being the beneficiary of his efforts to promote and sell Coca-Cola in his territory and his investments in returnable bottles.

The arrangement thus concentrates the bottlers' competitive efforts in the direction of other brands rather than on other bottlers of Coca- Cola (i. interbrand instead of intrabrand competition). This differ­ ence in the interests of bottlers of different brands should generate a stronger competitive effort than would be the case if those efforts were diluted by also being directed toward other Coca-Cola bottlers. The result is that territorial exclusivity leads to greater rather than lessened competition in the soft drink industry. (87) Another element to consider in determining whether a restraint is reasonable is the power of the imposer of the restraint in the relevant market. Columbia Pictures Corp., supra 189 F. Supp. at 178. The price setter s market power is determined by his freedom from concern over prices charged by his competitors in the market in which he sells the product. S. v. E. 1. dupont de Nerrwurs Co. 118 F. Supp. 41 (D. Del. 1953) aff' 351 U.S. 377, 391-392 (1956). But there is more than adequate evidence in this record to show that in the national soft drink market "Coca-Cola" and other respondents, as well as other national brand licensors and the bottlers in their regional and local markets are very m~h concerned with interbrand prices (Findings 103-109, 127­ 128). The evidence also shows that market shares of "Coca-Cola nationally are declining but that locally or regionally in various bottlers' territories the market shares fluctuate in direct response to interbrand competition (Findings 163-166). It also is worthy of mention that bottlers of Coca-Cola face strong intrabrand competition from post-mix Coca-Cola (Finding 133).

Is There an Adverse Effect on Consum€rs Competitms or Other Busirt€ssm€n? Chicago Board of Trade, supra also indicates that whether . . .... . THE COCA-COLA CO., ET AL. 587 517 Initial Decision competition is or may be promoted or curtailed by the restraint is factor to be considered in determining its legality (246 U.S. at 238). As indicated above, the evidence here shows that focusing the bottlers attention on their own territorial markets stimulates their competitive effort. There is keen interbrand pricing and also packaging competi­ tion (Findings 103-109, 149-153) and there are many brands of soft drinks available (Findings 92-102). In the last few years in particular many new brands of soft drinks have successfully been introduced into the territorial markets of bottlers (Findings 154-162). (88) The bottlers also compete intensely in having their brands available at a multitude of outlets and in obtaining both desirable shelf-space and display locations in food stores (Findings 137-140, 141-144). and is worth repeating that the prices of Coca-Cola and allied products are determined by the bottlers individually and that those prices are sensitive to the prices of other brands and types of soft drinks (Findings 66, 103-109, 127-131).

What if Exclusive Territories Were Held To Be Rlegal? Declaring "Coca-Cola" territorial exclusivity provisions to be illegal would adversely affect competition because it would lead to the business failure of many small and some large bottlers as well as to the accelerated growth of large bottlers (Findings 183-186, 193). This result would be in direct conflict with the purpose of the Congress in enacting and in agencies administering the antitrust laws " . to perpetuate and preserve, for its own sake and in spite of possible cost an organization of industry in small units which can effectively compete with each other. S. v. Aluminum Co. of Am€rica 148 F. 416, 429 (2d Cir. 1945). See also Brown Shoe Co. v. 370 U.S. 294 333 , 344 (1962); ef. Fashion Originators ' Guild v. FTC 312 U.S. 457 467 (1941); ef. Federal Trade Commission v. Fred Meyer Inc., et al. 390 S. 341, 349-351 (1968).

With the failure of small bottlers would go the jobs of the owners and of the thousands of their employees. The contributions to the economies of the areas in which small bottlers and their employees now earn their living would certainly diminish substantially and would disappear completely where the bottler was forced out of business(Findings 194-195). I do not believe that enforcement of the antitrust laws and of the Federal Trade Commission Act was or is intended by the Congress or the Commission to bring about such a result. For the reasons referred to above, such a holding would be harmful to competition and on the basis of extensive evidence in this record would not lead to any . .. ... , Initial Decision 91 F.

countervailing advantage to the interest of the public in price, quality or availability. (89) In fact, the probable result would be an increase in prices of soft drinks to consumers because of the elimination of small bottlers and the resulting enhanced market power of those bottlers who would survive along with the eventual reduction in the number of brands which would be marketed (Findings 170, 176, 179). As to these, there is much testimony in the record to the effect that there would be even greater use of cans and non-returnable bottles (Findings 120-121). These are more expensive than returnable bottles and greater use of them probably would lead to higher prices (Finding 110). In addition industry witnesses testified that the reduction in the number of small bottlers and the resulting growth of large bottlers would lead to a reduction in the number of different brands of soft drinks available (Findings 179-182).

The evidence is convincing that elimination of territorial exclusivity would increase the use of non-returnable bottles and cans (Findings 120-122). This would be contrary to the policy of the Congress in enacting the National Environmental Policy Act (42 V. C. 4331(a) (1972)), as follows:

it is the continuing policy of the Federal Government, in cooperation with State and local governments, and other concerned public and private organizations, to use all practica~le means and measures, . . . in a manner calculated to foster and promote the general welfare, to create and maintain conditions under which man and nature can exist in productive harmony, and fulfill the social, economic and other requirements present and future generations of Americans. A decision which would lead to even greater use of environmentally undesirable, disposable bottles and cans would not be in harmony with the quote above. The decision in this matter is one means for contributing to accomplishment of the objectives of the legislation. (90) CONCLUSIONS Respondents are engaged in commerce and their acts and practices are both "in commerce" and "affect commerce" within the meaning of the Federal Trade Commission Act; hence the Commission has jurisdiction over respondents and their acts and practices. The territorial exclusivity provisions in the contracts respondents have with their bottlers do in fact restrict intrabrand competition between bottlers. However because of the unique nature of the soft drink industry, the restrictions are not adverse to the interest of the public and they do further the legitimate business goals of respon­ ...

THE COCA-COLA CO., ET AL. 589 517 Dissenting Statement dents; hence, they are not unreasonable restraints on trade. This because there is substantial and effective interbrand competition in each Coca-Cola bottler s territory in (1) pricing, (2) maintaining product quality, (3) packaging, (4) providing servi~ to retailers and to members of the public, and (5) merchandising. The territorial exclusivity provisions in the contracts respondents have with their bottlers are no more restrictive than is necessary to . persuade bottlers to make and to continue to make the sizable capital investments necessary to operate successfully, to hire and to continue to hire employees to operate the equipment and to contribute to the economies of the communities in which they are located. Elimination of the territorial exclusivity provisions would have an adverse effect on competition and be contrary to the objectives of the antitrust and environmental impact laws because (1) many small bottlers would be forced out of business with the adverse effects mentioned above, and (2) there would be greater use of non-returnable bottles and cans. Greater use of non-returnable bottles and cans would result in higher pric~s per ounce of soft drink and greater detriment to the environment. (91) The adverse effect on intrabrand competition of the territorial exclusivity provisions in respondents' contracts with their bottlers is outweighed by the beneficial effects which those provisions have on interbrand competition in national and local soft drink markets. ORDER It is ordered That the complaint in this matter be, and it is hereby, dismissed.

DISSENTING STATEMENT OF COMMISSIONER CLANTON In light of the Supreme Court' s recent decision in Continental T. V. Inc. v. GTE Sylvania, Inc. 433 U.S. 36 (1977), this case presents important questions about the proper application of the rule of reason in judging the legality of vertical distributional restraints. For the reasons set forth below, I dissent from the majority s ruling and recommend that the case be remanded for further analysis of the effect of exclusive territorial restrictions on interbrand competition in the soft drink industry. (2) Absent evidence of horizontal collusion the reasonableness of Coca-Cola s territorial restrictions, in my view 1 I adhere to the abbreviations used by the Commission, with the addition of "Maj. Op, " to refer to the Majority Opinion and "Oral Arg. II" to refer to the second oral argument conducted on July 28, 1976. 2 As the majority correctly concludes, there is no evidence that the territorial restrictions were imposed by the bottlers; nor is there any evidence that the restraints have furthered collusive schemes among bottlers or syrup manufacturers.

Dissenting Statement 91 F. can be properly assessed only after examining the competitive state of the soft drink market and the extent to which respondents and their bottlers are capable of exercising market power within this interbrand environment. Although the majority opinion thoroughly analyzes the record and from it presents the best case possible under the circumstances for abrogating Coke s territorial restrictions, I believe the record is insufficient to make a confident judgment about the overall competitive effects of the restraints. ­ The majority concludes that the territorial restraints imposed on respondents' bottlers are unreasonable and in violation of Section 5 of the Federal Trade Commission Act.3 In reaching its decision, the Commission relies on several factors: (1) the elimination of intrabrand competition at the wholesale level;4 (2) the failure of respondents to provide a convincing justification for the territorial restraints; (3) the adverse effects of piggybacking of brands by bottlers upon the level of interbrand competition; and (4) the prospects for lower wholesale prices due to the interaction between intrabrand and interbrand price competition if the restrictions are removed. Although the Commission finds these restraints to be unreasonable for both refillable and nonrefillable containers, it nevertheless acknowledges that only partial relief is warranted. Accordingly, the Commission s order would permit the continuation of exclusive territories for returnable bottles. Further, recognizing that Coke ha3 a legitimate interest in maintain­ ing the quality of their products at the point of sale, the Commission order allows the respondent syrup manufacturers to require inter alia that bottlers take steps to assure that stocks are rotated at retail outlets. (3 I. SYLVANIA Before examining the facts of this case, it is essential to review the guidance provided by Sylvania for assessing the legality of vertical restrictions under Section 5 of the Federal Trade Commission Act. While not providing definitive standards for applying the rule of reason, the Court cited with approval the classic formulation of the rule in Chicago Board of Trade6 and gave some additional clues for sorting out the complexities of vertical restraints given "their 3 It is unclear, however, whether the Commission s order would bar the imposition of exclusive territories by independent bottlers upon sub-bottlers.

4 The order would appear to prohibit any limitation on the sales territories of respondents' bottlers although the Commission seems to recognize that other territorial restraints, such as location clauses, may be less restrictive of intrabrand competition. (Maj, Op. at 46 n,41) 5 Although returnables account for approximately 40 percent of flavored carbonated beverages industrywide (RPF 352), they represent 55 percent of Coke s sales (RPF 348), Chicago Board of T'NMk v. United States,246 D,S. 231 , 238 (1918), THE COCA-COLA CO., ET AL. 591 517 Dissenting Statement 0.- . potential for a simultaneous reduction of intrabrand competition and " 7 In overturning the per se rulestimulation of interbrand competition. of Schwinn the Sylvania Court quite clearly indicated that the nature of the restriction-be it territorial, customer or location-is relevant but not controlling.s Nor does it appear that the Court intended limit the legality of vertical restrictions to certain classes of firms such as new entrants or failing or faltering firms.9 More importantly, the effect of theseCourt emphasized that the critical factor is the restraints on interbrand competition, noting that such competition " the primary concern of antitrust law. " 10 (4) discussion of the Equally instructive is the Court's extensive economic benefits of vertical restraints. Observing that such restric­ tions may enhance marketing efficiencies and, therefore, the ability of the manufacturer to compete with other brands, the Court cited l1 Restrictions imposedadvantages to both new and established firms. necessaryby new entrants may help to induce distributors to make investments, while established manufacturers may seek to expand output by encouraging dealers to engage in greater promotional activities or to provide more services. Because of the "free-rider effect, the Court recognized that dealers might be dissuaded from making expenditures for such investments, advertising, or services. Significantly, the Court concluded that "there is substantial scholarly of suchand judicial authority supporting (the) . . . economic utility" restrictions and "relatively little authority to the contrary. " 13 Since Section 5 certainly reflects, at least in part, the policies of the Sherman Act, judicial consideration of territorial restrictions under Section 1 of the Sherman Act necessarily serves to advance the inquiry here. Of course, it is well-established that acts or practices which fall short of violating the Sherman Act may nevertheless traverse the more it is well' toexpansive area of illegality defined by Section 5. But, remember that: Just as the "rule of reason" has been read into the v. United States 221 Sherman Act by Standard Oil Co. of New Jersey S. 1, 62, 31 S.Ct. 502, 55 L.Ed. 619, 646 (1911), to allow some competitive practices which restrain competition in some degree, not 7 483 8 " (W)e agree with the implicit judgment &hwinnin that apeT Be rule based on the nature of the restriction is, in general, undesirable.V.S,Although distinctionsatcan be drawn51-52.among the frequently used restrictions, we are inclined to view them as differences of degree and form. (Citations omitted.) We are unable to perceive significant social gain from channeling transactions into one form or another, Id. at 58 n.29. Id.

10 Id. at 52 n.19, II Id. at 54-57.

12 Id. at 55, 13 Id, at 57-58. It seems clear, of course, that the Court in highlighting the benefits of vertical restrictions did not intend to carve out any classes of transactions which should ,be treated aspeT Be legal. Nor do I support such an approach whether it be by judicial or legislative action. .. .

Dissenting Statement 91 F. every method of competition is an "unfair method of competition under the F. C. Act. Sandura Co. v. 339 F.2d 847, 849 (6th Cir. 1964).

Recognizing, therefore, that we are measuring respondents' method of distribution against the relatively more-encompassing standard of Section 5, I turn to an examination of the competitive effects of the practice. (5) II. INTRABRAND COMPETITION At the outset, it is important to recognize what is and is not at issue in this case. There is no disagreement that Coke s system of exclusive territories-in contrast to other types of vertical restrictions, such as location clauses-completely inhibits intrabrand competition.14 There is also apparently no dispute that the territorial restrictions imposed here were justified at their inception. (See Maj. Op. at 10 n.12; 23.) And it is not contended that the restrictions have furthered anticompetitive ends by limiting output or facilitating horizontal collusion among syrup manufacturers or bottlers. Rather, the Commission stresses that the restraints are no longer justified primarily because of changed conditions affecting distribution.

Much is made of the fact that improvements in transportation and central warehousing by food chains have rendered territorial boundar­ ies obsolete, thereby creating inefficiencies and higher prices. Clearly, market conditions today are different than those which existed at the turn of the century. (6) Yet to the extent that modern delivery and distribution systems foster inefficiencies, Coca-Cola has an incentive to bring about adjustments which would enhance its bottlers' ability to compete with other brands (at least by means short of inducing more intrabrand competition). This incentive is perhaps enhanced by the fact that Coke s economic leverage vis-a-vis its bottlers is limited syrup prices to bottlers may rise only as the price of sugar rises. (CX 9A- , 11A- , 13A-B) Hence, Coke s syrup revenues vary proportion­ ately with the sales volume generated by its bottlers. Where inefficiencies exist, which might make some bottlers less competitive Coke could be expected to take steps to urge bottler consolidations or other adjustments necessary to enable its bottlers to contend with current competitive realities on an interbrand level. In fact, the record shows that such changes have occurred with the support of Coca-Cola, including 107 consolidations and mergers between 1968 and 1971, 14 temporary marketing bottler agreements by which a bottler who discontinues production continues to distribute Coke produced by a neighboring Coke bottler (Tr. 650-51 , CX 1245A­ 14 This assumes, of course, that post-mix (or fountain) sales, which are not sold via an exclusive territorial system are excluded from consideration, an assumption not accepted by respondents, THE COCA-COLA CO., ET AL. 593 517 Dissenting Statement , CX 1246A-J), and several agency arrangements (Maj. Op. at 22 n. 22). Conversely, attainment of scale economies, notwithstanding distributional limitations may be facilitated for other bottlers by piggy-backing" several brands and engaging in contract bottling or canning. The record, however, does not afford a thorough evaluation the trade-offs involved in maximizing operating efficiencies, and provides little insight with respect to the optimal size of a bottler marketing area.

As for central warehousing, respondent concedes that chains prefer warehouse delivery or plant pick-up rather than direct delivery to their stores. (RPF 88) Nevertheless, the record does not establish a clear cost advantage for central warehouse over store-door delivery. (Maj. Op. at 25 n. 25.) It is at least possible, however, that if this cost advantage were significant, there would be even greater pressure for such service 15 as well as greater market penetration by brands such as Shasta and others which do not utilize a store-door delivery system. (7) Of course, it should be kept in mind that, by their very nature, vertical restrictions such as those employed here place constraints on natural market forces. Territorial boundaries, even when first imposed may not perfectly reflect the parameters of economic markets which would exist if competition were totally unrestrained. That is not to say that considerations relating to the nature of the restraint and its effect on intrabrand competition are irrelevant to the inquiry. But, while legitimate restrictions may become unjustified due to subsequent changes in the market, the significance of the lost intrabrand competition can best be evaluated, as Sylvania indicates, in the context of the reasons advanced for imposing such restraints and their effect on interbrand competition.

III. INTERBRAND COMPETITION A. PROCOMPETITIVE EFFECTS Respondents offer several justifications for imposing exclusive territories including: (1) inducement of bottlers to make. necessary capital investments, (2) more extensive market penetration and increased availability of Coke, (3) greater advertising and promotional efforts by bottlers, (4) quality control, particularly at the distribution level, and (5) preservation of returnable bottles.17 With respect to the 15 In Los Angeles, the chain stores forced bottlers to provide them with warehouse delivery of nonreturnable packages. Alpha Beta, for example, refused store-door deliveryof Coke at wholesale prices below the price paid by Alpha Beta to pick up canned Coke at Coke s Los Angeles bottler. (Tr. 3486, 2585-86, 2633-M) 16 See 433 U.S. at 54-55, 17 Exempting returnables from its order, apparently as an actof discretion in crafting an appropriate remedy, the Commission does not address the possibility that territorial restrictions in the ofcasereturnable containers may be (Continued) ... _ Dissenting Statement 91 F. first four justifications, the majority claims that the strength of the Coke brand provides sufficient incentives for investment, and that any decrease in the current level of bottler advertising will be offset by increases in manufacturer and retailer advertising. Acknowledging that territorial restrictions have increased the availability and market penetration of Coke, the Commission expresses the view that market forces should determine the proper combination of price and availabili­ ty. In addition, the Commission concludes that the quality of Coke can be maintained at the distribution level through sp~t checks facilitated by placing bottler identification marks on containers. And, as noted earlier, Coca-Cola may require bottlers to follow specific monitoring and inventory rotation policies. (8) With respect to capital formation, the majority concedes that exclusive territories provide bottlers with some additional measure of certainty regarding their ability to recover their investments. How­ ever, the Commission is unable to conclude that a free market would otherwise render the bottlers incapable of operating at a profit. Granted, Coke bottlers today need less incentive to make the requisite investment of capital and labor to maintain their competitive position than would a newcomer attempting to enter the market from scratch.I8 Yet the record indicates that a substantial investment is needed to maintain a successful bottling operation. 19 More relevant to our discussion is (1) whether adequate incentives exist for Coke bottlers to maintain the kind of distribution and quality control system utilized here in the absence of territorial restraints, and (2), assuming restrictions are necessary, whether they can be justified in light of their impact on interbrand competition. Since, in my view, the record supports the need for investment incentives to maintain the present marketing system, a point which the majority implicitly concedes as to returnable containers, careful scrutiny of the interbrand effects is required. To say that the current level of capital investment might be unwise in a different competitive milieu (Maj. Op. at 29) simply begs the question of whether such investments enhance or impair interb­ rand competition.

justified under the rule of reason and Section 5. Just as I am unable to concur with the majority s decision to prohibit territorial restrictions for nonreturnables, I am reluctant on this record to sanction exclusive territories for refillable containers, 18 Respondents do not contend that they will be unable to attract bottlers in the absence of territorial restrictions; they do assert, however, that bottlers will have substantially less incentive to make the kind of investments necessary to maintain the current marketing scheme utilizing store-door delivery, 19 The bottling of flavored carbonated soft drinks is capital intensive, with an investment of $1.00 required for every $1.52 in sales volume, (Tr. 1532) In addition to land and bottling equipment, bottlers must purchase and maintain route trucks and a returnable bottle float. The Coke bottler in Washington, D.C., for example, operates more than 180 route trucks costing $14 000 apiece, and seven tractor-trailer trucks costing $100 000 apiece. (Tr. 1.229, 1911-12, 2047­ 2050 2456) The Washington bottler, who currently has $2.250,000 to $2,500,000 invested in returnable glass bottles decided against introducing a 32 oz. returnable bottle in view of the competitive situationand the $1 million investment in glass bottles that would be needed, (Tr. 1262, 1314-15) .... .

THE COCA-COLA CO., ET AL. 595 517 . Dissenting Statement Adopting a similar approach, the majority does not challenge the assertions that Coke s territorial restrictions have benefitted interbrand competition by increasing the availability of Coke and fostering greater market penetration than would otherwise be the case. Instead the Commission suggests that the degree of availability and market penetration currently enjoyed by Coke does not reflect variations in the cost of serving different customers and claims that increased intrabrand competition would allocate market resources more effi­ ciently. (Maj. Op. at 32) Such an analysis falls short of the kind of inquiry demanded by Sylvania concerning the procompetitive aspects of the challenged restrictions and their overall effect on interbrand competition. (9 J Exclusive territories, it is argued, facilitate interbrand competition by enhancing availability and output and inducing greater demand for soft drink products. Testimony of record lends support to the view that widespread availability is a significant merchandising factor in this industry. (IDF 137-140) For example, the president of Dr. Pepper testified:

I will have to start with the fact that soft drinks are an impulse item, and that soft drinks are consumed many times a day by an individual, and in many places. And because of those customs or habits, total availability is essential to provide the opportunity for those people to enjoy. (Tr. 4029) Dr. Pepper tried distribution of their product via warehouse delivery in Indianapolis and Los Angeles. After finding that Dr. Pepper was not available in numerous small outlets, the company decided that the only way to achieve widespread availability was to use a route delivery system. (Tr. 3996-4000) Similar marketing decisions were made by Lipton and Nestle. (RPF 95-96) The central feature of this marketing scheme, maximization of availability, is made possible by level pricing, which means that all retail customers in a territory are charged the same price and that some Coca-Cola is provided at less than its actual cost and some is priced above. The record provides no basis, however, for determining whether these cost differentials reflect legitimate promotional consid­ erations or represent a form of price discrimination practiced by firms having a significant degree of market power. (See discussion infra 21-22) Furthermore, cost differentials may not warrant the burden of establishing a host of different price schedules.21 Nor does the record 20 That is not to deny, of course, that more recent entrants to the soft drink business, such as Shasta, have achieved a degree of success without the use of exclusive territories. Such entry, though, has been limited primarily to warehouse delivery of nonreturnables and may, in fact, have been facilitated by increases in demand for soft drink products generated by the very marketing schemes at issue in this case. 21 For example, under the current stortHloor delivery system, the expense associated with allocating distribution costs among the numerous outlets served by a bottler may not justify an expanded price list, Dissenting Statement 91 F. C. Ortell us whether consumers would suffer more from reduced competi­ tion over availability than they would benefit from the presence of variable pricing and the imposition of minimum delivery and pick-up volume. (10) On the issue of advertising, the record provides no prognosis of the expected level of interbrand competition if bottlers .are given less incentive to advertise Coke to ultimate consumers by virtue of the free-rider" effect. Coke and its bottlers spent more than $50 million on national cooperative media advertising in 1974 -and about half of these expenditures were made by bottlers. (Tr. 652, 685-86) The majority questions the utility to consumers of much of this advertising, some of which is admittedly of the brand enhancement or image type and suggests that much Coca-Cola advertising is unnecessary in view of Coke s "widely recognized name.

These judgments concerning the quality of Coke s promotional efforts were not based upon a systematic and thorough review of Coke advertising, since such a task would have complicated this proceeding enormously. Yet, even if the Commission could say with any confidence that Coke s advertising conveyed little information of use to consum­ ers, prohibiting exclusive territories might only result in "a shift to less efficient methods of obtaining the same promotional effects. Sylva­ nia 433 S. at 56 n. 25. While the Commission concedes that it will be in Coke s interest to assume more of the brand-enhancement burden in the future, the record evidence gives no clue as to the relative efficiency of manufacturer advertising versus bottler advertising. But the very difficulty of making such fine-tuned assessments only underscores the importance of focusing on broader competitive considerations in weighing the reasonableness of advertising-related or any other justifications.22 For example, promotion of the Coke brand may have resulted in such a high degree of product differentiation and conferred upon respondents and their bottlers such a substantial amount of market power that the Commission would be justified in striking down the restrictions irrespective of any countervailing benefits. As noted infra however, the record in its present state does not warrant such a conclusion. (11) With respect to the quality justifications proffered by Coke, there appears to be no dispute that territorial restrictions encourage individual bottlers to produce a high quality product and, through proper rotation of stock, ensure that rigorous quality standards are met until final delivery to the customer. Monitoring the quality of 22 The majority conclusion (Maj. Op. at 35) that advertising-related considerations may justify an intrabrand restraint only in the case of new or faltering finns suggests that the Commission has adopted Justice White' concurrence inSylvania 433 V,S. at 63--65, It is clear, however, that a majority of the Court would not so limit application of a rule of reason anaiysis, 433 V.S. at 47 n. , 53-54 n. , 59. THE COCA-COLA CO., ET AL. 597 517 Dissenting Statement Coke at the retail level is facilitated because a Coke bottler knows he will be held accountable, on both a public relations and product liability basis, for all Coca-Cola sold within his district. However, applying the underlying rationale of the Commission decision in Adolph Coors Co. 83 F. C. 174 (1973), the Commission finds Coke s territorial restric­ tions to be unreasonable due to the availability of less anticompetitive means" to the same end, namely, the establishment of reasonable quality control standards, including inventory rotation policies, and a requirement that bottlers place their own identifying mark .on each bottle or can. The Commission concludes, however, that territorial restrictions are not "reasonably necessary to ensure the taste uniformity or the purity of these products." (Maj. Op. at 38) Undoubtedly alternate means are available for ensuring some degree of quality control, though at what cost and with what measure of effectiveness are not ascertainable from the record. That other, if less satisfactory, strategies exist does not by itself render the use exclusive territories to maintain product quality unreasonable. Such a determination requires further consideration of interbrand competitive effects.

To conclude, as the majority seems to imply, that the availability suitable marketing approaches less restrictive of intrabrand competi­ tion is sufficient to find a violation, places the Commission in the difficult position of having to make its own judgment about the commercial merits of various strategies without regard to their impact on interbrand competition. While such a test may have some support in the case law 23 it has never been sanctioned by the Supreme Court. Moreover, in rejecting adoption of a "no less restrictive alternative rule in a recent decision, the Third Circuit highlighted the problems inherent in such a standard: (12) Entrepreneurs. . . would then be made guarantors that the imaginations of lawyers could not conjure up some method of achieving the business purpose in question which would result in a somewhat lesser restriction of trade. And courts would be secondguessing business judgments as to what arrangements would or would not provide adequate" protection for legitimate commercial interests. American Motor Inns, Inc. Holiday Inns, Inc. 521 F.2d 1230, 1249-1250 (3d Cir. 1975). Likewise, the Commission s opinion implicitly second-guesses Coke belief that obstructions to intrabrand competition are needed maintain the high quality of its product. Yet the record affords no real 23 White Moor Co, v, United States 372 U,S, 253, 271 (1963) (Brennan, J., concurring); Copper Liqunr, 1m, AM/ph Coors Co., 506 F.2d 934, 947 (5th Giro 1975); Siegel v. Chicken Delight, Inc. 448 F,2d 43, 51 (9th Giro 1971), cert. denied 405 U.S. 955 (1972); Sandura Co. V. C" 339 F,2d 847 856 (6th Cir, 1964). 24 The court distinguishedSiegel and Cooper Liqum as involving restraints which wereper se illegal. Id, at 1.249, Similarly, the Commission s opinion indoors upon which the majority relies, found Coors' territorial and warehouse delivery restrictions illegalBe,per Coors 83 F. C. at 194-95 (1973), Dissenting Statement 91 F. basis for analyzing the costs or benefits associated with "less restrictive" approaches to support the majority s assessment. While the examination of the business reasons advanced in support of a restraint is certainly an appropriate, indeed necessary, part of the inquiry, that assessment cannot be divorced from interbrand considerations, particu­ larly where, as here, the evidence supports the redemptive value of quality-based and other justifications.

On balance, I am unable to conclude from the record before us that the restrictions imposed here are not reasonably related to the marketing strategy employed by Coke: enlargement of market coverage through level pricing, store-door delivery to numerous retail outlets, and frequent service to maintain quality. And, I cannot conclude on the basis of this record that such strategy has so little redeeming competitive virtue that we may dispense with further consideration of interbrand effects. Like any marketing approach Coke s strategy involves a trade-off between different combinations of price and nonprice competition, but one s preference for one kind competition over another should not automatically condemn that strategy.25 In short, the reasonableness of the restraints cannot be judged in isolation from their overall impact on competition. That is the critical factor at issue here. (13) Notwithstanding the Court' s express recognition in Sylvania of justifications similar to those advanced here, the Commission appears to be taking the position that the kind of competition fostered by the vertical restrictions in this case is inherently less desirable than the competition which would otherwise occur in the absence of such restraints, at least where well-established, successful firms are concerned. Indeed, the majority decision comes close to establishing per se standard of illegality where territorial restraints are imposed by leading firms in an industry, a result seemingly inconsistent with Sylvania.

B. ANTI COMPETITIVE EFFECTS This conclusion is reinforced by the Commission s discussion of the adverse effects of the restrictions on interbrand price competition and its finding that these effects are so substantially adverse as to place upon respondents the burden of showing that the net effect of the restraint is procompetitive. These ill-effects, it is alleged, are evidenced by price disparities among brands sold in adjacent territories and testimony that wholesale prices to some customers would be lower if 25 It may be true that the procompetitive effectstheseof restrictions might be greater were they imposed by a new entrant or other finn seeking to challenge the industry leaders. Nevetheless, I am not persuaded by either legal or economic reasons that their use should be so limited. .

THE COCA-COLA CO., ET AL. 599 517 Dissenting Statement the restrictions were removed. (Maj. Op. at 47-51) That, in my view, is wholly insufficient as a basis for voiding such restraints. The absence of price uniformity among territories, without additional data such as pricing trends (including their magnitude and duration) or differences in profit margins, is not by itself conclusive and may only suggest that competition is not static and that its intensity and form are likely to vary over time.26 Indeed, price uniformity among brands and throughout various territories might give rise to even greater alarm. As for the prospect of lower wholesale prices, the evidence at most (14) respect to suggests that greater price competition will occur with chain stores sales, which represent approximately 20-30 percent of Coke s volume nationwide (RPF 325).27 There is no evidence that lower prices and greater availability overall will occur upon removal of Coke s territorial restrictions.

That some measure of additional price competition might result in certain situations is not unique to this type of restraint or this case, as the Supreme Court implicitly recognized in Sylvania. The majority's analysis, regrettably, provides no framework for assessing the impact of the territorial restraints upon the overall competitive structure of . the soft drink industry. Concluding that competition might function differently were the restrictions lifted doesn tell us whether interbrand competition is better or worse off because of these restrictions. The more relevant question, which cannot be answered by recourse to this record, is whether the system of exclusive territories has enabled Coke and its bottlers to obtain an unreasonable degree of market power or exploit such power unreasonably. The Commission correctly asserts that proof of monopoly power or unrestricted market power, as argued by respondents, is an unneces­ sary prerequisite to a finding that a particular restraint is unreason­ able (Maj. Op. at 18). Citing Sylvania for the proposition that a less 26 The Commission expresses concern for the "competitive dynamics" of the soft drink industry, (Maj. Op. at 19), but its decision is premised upon a static analysis of prices. Thus, the record contains comparative territorial pricing data for Coke, Pepsi, Shasta, and Green Spot for a single day, July 15, 1971. (Because this data reflects list prices and does not account for promotional discounts, the Commission has little infonnation regarding transaction prices.) While complaint counsel introduced no evidence to demonstrate the variability of these prices over time, there is some anecdotal evidence of the responsiveness of Coke to price adjustments of other brands, and vice versa. (IDF 103-109) 27 The Coke executive quoted by the Commission (Maj, Op. at 50) to support its belief that prices would fall to principal customers explained in the next breath that the invalidation of territorial restrictions "would mean that we might be faced with either selling to a chain store in a warehouse delivered situation at a low price, and at a higher price to our smaller stores that we were still giving store-door delivery to." ( Tr. 993;see also Tr. 1000--01.) Other industry witnesses also expressed the view that wholesale prices to non-chain stores would increase in the absence of territorial restrictions, (Tr, 659 993, 1001 , 1QO3..W, 1152, 1401, 1472, 1572, 1929, 1992, 2066, 2187-88, 2462, 2637, 2686, 2898 2980, 3054--55, ~1, 3583-84, 4036, These expected price differences not only will reflect lower bottler costs associated with warehouse delivery, but they also allegedly will result from bottler efforts to skim off the high volume chain store accounts. Respondents argue that bottlers who successfully capture those accounts will have less incentive to continue service to smaller, less profitable outlets. Conversely, other bottlers will be forced to raise prices to smaller a~unts to compensate. for the lost chain business. (Ans,Br, at 54-ii8) While this scenario may not prove correct, the record provides no firm basis for concluding otherwise, Dissenting Statement 91 F. C. 0,­ sweeping restraint than Coke s exclusive territories might be unrea­ sonable for a company with a small market share standing in the shadows of (15) a dominant firm 28 the Commission adopts the view that an inquiry into the market power of Coke is unessential. (Maj. Op. at 19)29 Nevertheless, without relying on such a showing, the majority opinion displays the limited, though incomplete, evidence of market power contained in the record. (See Maj. Op. at 30 n.3Q, 34-35, 39-40 , 44 nAO)3o (16) That the Commission should dispute the relevancy of market power is somewhat peculiar. The analysis and approval of vertical restrictions in the context of consignment transactions in Schwinn illustrates the significance attached by the Court to Schwinn s declining market share and the availability of other competitive bicycles despite Schwinn claim of product excellence. 31 Similarly, in Sylvania the Court made clear that the level of interbrand competition confronting the manufacturer was a crucial consideration in analyzing a vertical restraint, taking note of market share data available in that case although it had been tried on a per se theory.32 Obviously, the greater 28 Sylvania had 5 percent of national television sales and 15 percent of television sales in the Sacramento area. Sylvania 433 U,S, at 39 n, 29 Complaint counsel share this view, as expressed in their Reply Brief: Contrary to the assertions of respondents. , , ' market power' considerations, either by way of presence or absence, are not controlling as to the issues herein. The Commission s complaint does not rest upon allegations of market power resident in respondents. Its allegations which charge violation of Section 5 of the Federal Trade Commission Act are predicated solely upon the elimination of intrabrand competition. No allegations with respect to ' substantiality' or ' section of the country' appear in the text of the complaint, and there is no basis for inferring that complaint counsel have the obligation of proving either of these elements as a part of their pn71U1 facie showing, Exclusive territoriallicenses are unlawful because they eliminate intrabrand competition." (Rep,Br, at 16) 30 Although it is perhaps difficult to argue that "the most widely recognized name in American commerce" does not possess some degree of market power, we cannot take judicial notice of Coke s market power. To be sure, there is some market share data contained in the record, although complaint counsel and respondents agreed not to admit it for purposes of establishing any relevant product market, (Stipulation No. 5; RX 2A). Assuming delineation of a proper product market or submarket, these shares are not insignificant, Such data, however, allows calculations only of a two-firm concentration ratio representing the combined share of Coke and Pepsi. While complaint counsel and respondents agreed to a stipulation setting forth the top four brands in 36 metropolitan areas of the country, only numerical rankings, and not market shares, are given. (Stipulation No. 5; RX 2Z2-2Z38) Moreover, as the Commission points out (Maj. Op. 39-41 n.37), the market share figures found in the record and ~lied upon by the law judge represent sales to food stores rather than the complete universe of Coca-Cola and Pepsi-Cola sales, and are unable to provide an accurate indication of Coke s syrup sales to bottlers or bottlers' sales of finished soft drink. Further observations should be made, however, with respect to this data. First, Coke s share of ~.8 percent for statistical case sales of flavored carbonated soft drinks to food stores in 1971 conceals a large measure of variability from region to region. For example, although Coke accounted for 33.2 percent and 33.1 percent of sales in the Southeast and the Southwest, respectively, it was responsible for only 13,1 percent of sales in the metropolitan Chicago region, 14,6 percent of sales in New England, and 14,7 percent of sales in the Pacific region. (Stipulation No. 5; RX 2D­ 2E) Second, Coke s market share is ironically largest for returnables, the sole container permitted by the Commission to be sold via exclusive territories. Thus, measured on a physical case basis, Coke s sales represented 32. 1 percent of all returnable bollles sold in 1971, 17,3 percent of all nonreturnables, and 12.6 percent of all cans, (Stipulation No. 2; RX 2H) 31 United Statesv, Arnold, &hwinn Co. 388 V,S. 365, 381--S2 (1967), Schwinn s market share had fallen from 22,5 percent in 1951 to 12.8 percent in 1961.&hwinn, ld. at 368, 32 Sylvania 433 U.S. at 52 n,19, In a pre-Sylvania case American Motor Inns, I~, v. Holiday Inns, I~.,521 F.2d 1230 (3d Cir, 1975), the Third Circuit rejected the district court's finding that a restriction on a Holiday Inn franchisee s ownership of non-Holiday Inn motels was unreasonable. Among other things, the court of appeals noted that the lower court's conclusions did not demonstrate "that considerations like the number or size of the firms in the (Continued) ,, _ THE COCA-COLA CO., ET AL. 601 ' 517 Dissenting Statement the market power of respondents, the less they need fear competition from their rivals and the easier it is for that market power to be transferred through the distribution chain. (17) The majority opinion also downplays the significance of product market delineation and assessment of the overall level of interbrand competition currently prevailing in that market.33 Rather than employing the kind of product market analysis which it uses elsewhere to minimize competition between bottlers and independent syrup jobbers (Maj. Op. at 19-21), the Commission notes that the complaint and trial focused upon a market which included virtually every liquid a person might consume, other than alcoholic beverages, and that testimony of industry witnesses concentrated on another market, viz flavored carbonated soft drink beverages. (Maj. Op. at 19 n.21) Although a broader market may be more favorable to respondents as the majority suggests, the record unfortunately does not provide a satisfactory basis for assessing Coke s position in that or any other market.

The lack of an adequate record basis for establishing an appropriate product market carries over to analysis of the present level interbrand competition. Admitting only that the market is not "devoid of interbrand competition " (Maj. Op. at 19) the Commission views its task as limited to arguing the following propositions: (1) that interbrand competition is not as vigorous as one might expect due piggybacking; (2) that industrywide resort to territorial restrictions exacerbates the adverse effect on interbrand competition that results from an individual Coke bottler being unable to compete with other brands beyond his territory; and (3) that enhanced competition among Coke bottlers will result in greater interbrand competition due to the sensitivity of other brands to the price of Coke. (Maj. Op. at 51-52) (18) The majority contends that piggybacking, rather than enhancing competition, reflects even greater concentration at the bottler level. Obviously, piggybacking of different brands by the same bottler leads to less competition for the patronage of retailers than would otherwise occur if each brand were sold by a separate bottler. If the choice is between a market of single brand bottlers and a market with the same industry or (Holiday Inn s) market share played any part in the Court's decision on this question." Id. at 1247-48 (citations omitted) And, as the Commission itself indicated in Coors, supra: It is where the manufacturer of a branded item possesses substantial market power - the power to set prices irrespective of interbrand competition. that vertical territorial restrictions are especially pernicious, for they eliminate the possibility of interbrand competition which in an imperfect market is a critical supplement to competition between and among different brands," (83 F. C. at 196) 33 One commentator has urged that in adjudicating a vertical territorial restriction under the rule of reason, the factfinder should consider such factors as (1) the percentage of business controlled, (2) the strength of the remaining competition, and (3) whether the action stems from business requirements or purpose to monopolize, J, von Kalinowski, The Per Se Doctrine - An E11!€rging Philcwphy of Antitru.straw, 11 V. A. L,Rev. 569, 589 (1964). Cf. United Statesv, Columbia Steel Co. 334 S, 495, 527 (1948), Dissenting Statement 91 F. number of bottlers (and equivalent bottler concentration) offering more soft drink products, our choice would obviously be for the latter. Yet, reliance by new entrants on existing channels of distribution may also reflect high entry barriers into bottling owing in part to the major brands' ability to successfully differentiate their products.' At the same time, the freedom of bottlers to carry other brands may signal greater competition among syrup manufacturers for the custom of bottlers. Additionally, there is some evidence of record that, by lowering entry barriers, piggybacking has enabled certain brands to enter the market which would otherwise be unable to afford -the steep capital requirements of bottling. (IDF 159-162) Thus, piggybacking is a complex factor in this industry and, like vertical restraints, capable of having procompetitive as well as anticompetitive consequences. Given the ambiguity surrounding the proper product market, and the absence of useful market share data, the competitive effects of piggybacking cannot be adequately assessed.

The Commission also expresses concern over the fact that territorial restrictions prevent, for example, a Coke bottler from competing with a Pepsi bottler in the territory of another Coke bottler, and likewise inhibit a Pepsi bottler from competing with Coke in the territory of another Pepsi bottler. (Maj. Op. at 45-49) But the fact that Coke or Pepsi bottlers are not potential interbrand competitors in (19 J markets adjacent to their territories is an inevitable consequence of exclusive territorial restraints.35 By underscoring this effect, the implication is that all exclusive territories, whatever their net effect on competition are beyond the pale. Further, while the widespread use of territorial restrictions in the industry may facilitate parallel pricing policies by reducing the number of bottlers competing for any particular account the significance of this phenomenon is difficult to evaluate without first examining the extent to which interbrand competition is foreclosed by respondents and other industry members. . The Commission s final argument is that intrabrand competition among Coke bottlers will serve to lower wholesale prices and, because of the price sensitivity (or cross-elasticity of demand) between Coke and rival products, wholesale prices for other flavored carbonated soft drinks, powdered mixes, and noncarbonated drinks are also likely to decline. (Maj. Op. at 50-52) As previously indicated, the record supports the majority s forecast of lower wholesale prices only with respect to chain warehouse sales; other outlets may well face higher prices for Coke. More importantly, the "ripple-effect" on interbrand J4 In sanctioning the use of Schwinn s distribution scheme in the context of consignment transactions, the Supreme Court placed particular emphasis on the fact that Schwinn s distributors and retailers handled other brands of bicycles as well as Schwinn s. &hwinn 388 V,S. at 381. 35 For rejection of a similar argument see Sandura, supra, 339 F.2d at 854. ... .

THE COCA-COLA CO., ET AL. 603 517 Dissenting Statement price competition which the majority asserts will follow from increased intrabrand competition may signify nothing more than that the level of interbrand competition is already substantial. While the record contains evidence that competing brands are price sensitive to coke there is also evidence that prices of Coca-Cola products are responsive to the pricing policies of competitors. (IDF 104-109) Although price differentials between major and lesser known soft drink brands may, at times, exist-a characteristic which reflects differences in quality, service and promotional effort-more information concerning the size frequency and duration of such differentials is needed to properly assess their significance. But whatever competitive repercussions may flow from the Commission s order lifting the territorial restrictions the net effect of such action is impossible to discern without fully examining the parameters .of existing competition. (20) C. NEED FOR FURTHER INQUIRY Unfortunately, the record does not provide adequate insight into these parameters. We are informed that interbrand competition is less vigorous than it might otherwise be without territorial restrictions, but never told whether interbrand competition is or is not presently sufficient to "provide a significant check on the exploitation of intrabrand market power because of the, ability of consumers to substitute a different brand of the same product. Sylvania 433 U. at 52 n.19.36 Similarly, while the Commission rejects the justifications offered by respondents, it does not examine the extent to which the alleged efficiencies generated by Coke s territorial scheme "promote interbrand competition. Sylvania 433 U. S. at 54. The limited record evidence regarding Coke s market power or the intensity of interbrand competition does not, of course, imply that Coke or the other major brands lack substantial market power. Although a rule of reason standard was applied, this case was never tried on a theory which called upon complaint counsel to produce evidence on the issue of market power. In fact, at one point in their appeal brief complaint counsel argue that no degree of interbrand competition-no matter how intense-could legally justify the total elimination of intrabrand competition. App.Br. at 58. Elsewhere, it appears that complaint counsel viewed their burden as one of establishing a prima facie case that intrabrand competition had been eliminated, upon which the burden of proof shifted to respondent to establish that the benefits to interbrand competition outweighed the 36 See also American Motar Inns, supra, 521 F.2d at 1247, where the court of appeals observed that the district court' s analysis did not "take into account whether the competition eliminated by the clause is significant in the context of the total competition extant in the industry, , , , Dissenting Statement 91 F. loss of intrabrand competition. (App.Br. at 9, Oral Arg. II at 21-22) Since vertical restraints invariably reduce intrabrand competition, the latter formulation would normally place the ' entire burden of persuasion upon respondent. 37 (21) Given the somewhat confusing state of the law post-Schwinn and pre-Sylvania, it is understandable that some uncertainty would creep into a case challenging vertical distributional restraints, irrespective of whether a per se or rule of reason standard was being applied. In my view, the rationale underlying Sylvania whether applied in a Sherman Act or Section 5 case, requires the Government to show that nonprice vertical restraints are unreasonable, taking into account both intraand interbrand effects on competition. Although one commentator has suggested otherwise 38 I do not agree that such a burden is likely to be insurmountable in the vast majority of cases. The indicia for measuring market power are familiar concepts which do not present unmanageable problems of proof in a rule of reason case. As I have noted, the majority opinion suggests that the lack of continuing business justification for the restrictions is sufficient to establish liability, but some scrutiny is also given to the interbrand effects of the restraints. Indeed, the Commission concludes that p1-irna fac.ie violation has been made out from evidence showing a substantial lessening of both intrabrand and interbrand price competi­ tion. It is further contended that respondents have not overcome this showing by evidence that the territorial restraints have had an overriding procompetitive effect. The rule of reason does not preclude adoption of procedural devices such as a standard of presumptive illegality or even a per se rule (as Sylvania acknowledges) where the circumstances warrant. However, the record in its present state does not justify application of any such device. For these reasons, I believe that the public interest justifies a remand of this case in order to examine more specifically the effect of these restraints on interbrandcompetition.The scattered bits of evidence relating to bottler profitability, concentration levels and pricing patterns contained in this record are incomplete and inconclusive for measuring the competitive health of the soft drink industry.39 Knowledge of Coke s market share over time (22) as well as the level and trend of concentration in the industry 37 Although Sylvania does not specifically address the question of proof, one commentator has suggested that the division of burden urged by complaint counsel here does not accord with the Court' s emphasis on the benefits of vertical restraints. Note, The Supreme Court 1976 Term 91 Harv. Rev. I ~ n.M (1977). In any event, it is doubtful whether adoption of such a rule would be appropriate,See Sandura, supra.; S1U!p-On Tools Corp. v. C., 321 F, 825 (7th Cir. 1963).

38 Note TIw Supreme Court 1976 Term 91 Harv. Rev, I, (1977). 39 For example, although we are told that bottlers earned profits of 2--7 percent on investment (IDF 85), Coke profitability at both the syrup level and where it is integrated forward into bottling is not disclosed. THE COCA-COLA CO., ET AL. 605 517 Dissenting Statement 0,would be particularly helpful in assessing the extent to which Coke either individually or in combination with other firms, exercises market power in the appropriate product market.4O So too would information concerning the nature and extent of inroads made by new entrants. Such evidence clearly provides one of the best means for measuring the competitive vigor of the market. Moreover, since the majority relies heavily on testimonial evidence that removal of the restraints will lower Coke s wholesale prices to supermarket chains, a remand would also present an opportunity to develop more evidence regarding the sensitivity of Coke s market share to the prices of other brands, and vice versa. To the extent that Coke loses sales when a competitor cuts prices, brand loyalty (and the pricing independence that it confers) may not be as great as the familiarity of the Coke brand might suggest.41 Further, more rigorous analysis of profitability at the manufacturer and bottler levels could provide additional insight into the extent of product differentiation and market power in the flavored carbonated soft drink market. Coke s use of "level-pricing," which in reality may be a form of price discrimination, may also portend the presence of substantial market power.42 Further inquiry might show, for example, whether "level pricing" represents the traditional effort of a firm with market power to exploit (23) different demand elasticities among its customers.43 On the other hand, the existence of joint or common costs for different customers suggests that price may deviate from the marginal cost of serving a particular customer without raising an inference of market power.44 Also, the frequency and size of price promotions and discounts, which the record reveals to occur with some regularity in this industry, may indicate whether price discrimination is having a harmful or benign effect on competition. In short, the evidence of price discrimination displayed here can not alone suffice as a suitable proxy for measuring market power.

Another avenue worth exploring would be to compare the marketing strategies of independent Coke bottlers with those of Coke s subsidiar­ ies. If the arguments raised by the bottlers are valid, one might expect to find Coke s distributors employing similar policies. Of course, even legitimate reasons for imposing the restraints may be insufficient to 40 If the case were remanded, it might be rlesil'able to amend the complaint to plead other product markets, such as flavored carbonated beverages, which are encompassed within the larger soft drink market. 41 Note Restricted Chamwls of Distribution Under the Sherman Act,75 Harv. Rev. 795, 833 (1962). 42 R. Posner Antitrust Law: An Econom-k Perspective at 63 (1976); F. Scherer Industrial Market Stnwture and Eco-TWm-k Perfo-rma1lce 43 Uniform pricing onatsales that involve253-272different costs might also facilitaten970),collusion among competing bottlers because of the ease of monitoring compliance and detecting cheating. Again, there is no basis on this record to do more than speculate upon the opportunity for collusion, 44 Pos1wr, supra note 36, Dissenting Statement 91 F. C. ...-~ save them in the face of substantial adverse effects on interbrand competition.

Given the more recent development of the nonreturnable, particular attention should be given to the reaction of the market to such new packaging and the effect of exclusive territories on the growth these new containers. Quite clearly, there has been more entry in this area by Shasta, Frank' , Faygo, and the private brands than there has been with respect to returnable bottles. Indeed, the record contains uncontradicted evidence that competition compelled Coca-Cola to abandon its exclusive reliance on a single size returnable bottle in the mid-1950s and expand into nonreturnable cans and bottles (Tr. 714 1344). Since the majority ironically limits relief to this form of packaging, further analysis of this development is warranted in a new hearing.45 (24) By emphasizing the importance of market power analysis, I am not suggesting there is some magic formula for determining which restraints are legal and which are i~legal. The examples highlighted above are only suggestive of the kinds of evidence that might be useful to such an analysis. They are not meant to be exhaustive, nor would proof as to each be required before the restraints could be declared unlawful. Moreover, I do not contend that liability will attach in vertical restraint cases such as this one only upon some showing that one or more competitors in an industry exercise substantial market power. Indeed, there may be instances where intrabrand restrictions are so patently unrelated to legitimate justifications that more intensive market scrutiny proves unnecessary. The problem with this case, however, is that it did not proceed on a theory requiring complaint counsel to put on evidence of interbrand effects. Given the state of the record, I cannot conclude whether the territorial restrictions should be approved or condemned without a further look at the state of interbrand competition.46 I am, of course reluctant to urge remand of a case which already contains a voluminous record and has been in litigation for several years. Nevertheless, in view of the intervening Supreme Court precedent, I believe the public interest would be served by a remand focusing on interbrand competitive considerations.

45 The relative market performance and profitability of fountain sales for which exclusive territories are not granted might also assist the inquiry.

46 In finding the record inadequate, I have not ignored respondents' considerable efforts to paint a rather glowing picture of the competitive vitality of the soft drink market, Yet, evidence that competition is flourishing, such as the variety of flavors and package sizes available, does not point unambiguously to the conclusion that interbrand competition as a whole is healthy. That the record is unsatisfactory, of course, normally would not entitle complaint counsel to another bite at the apple. But where significant changes in the law have occurred, as they have here, I believe complaint counsel should be given an opportunity to make their case in light of these changes. . .,.. .

THE COCA-COLA CO., ET AL. 607 517 Opinion OPINION By DOLE Commissiomr:

The basic question on this appeal is whether territorial restrictions which eliminate competition among the independent bottlers of Coca- Cola and allied soft drink products are unfair within the meaning of Section 5 of the Federal Trade Commission Act. I. INTRODUCTION Respondent Coca-Cola requires little introduction. It is a diversified corporation with interests ranging from steam boilers to orange juice. In 1968 it had consolidated net sales in excess of $1.1 billion and consolidated assets exceeding $802 million. Pertinent to the issues. raised in the complaint in this proceeding are the operations of its Coca-Cola USA division. It is this division which manufactures and sells the soft drink syrups and concentrates used in the (2) processing of finished flavored carbonated soft drinks sold under one or more of the trade names licensed by respondents to the bottlers.1 In 1968 its syrup sales to bottlers exceeded $246 million. Around the turn of this century, The Coca-Cola Company sold its right to bottle Coca-Cola and licensed the "Coca-Cola" trademark, in perpetuity, to private investors who, as independent businessmen operated their own bottling facilities within assigned territories.2 At the time, The Coca-Cola Company itself produced no bottled soft drinks, and although it does today in certain areas of the country, its entry into the business of bottling the products which bear its trademarks results from the reacquisition of the bottling rights which had been previously granted to local bottlers. Today it operates bottling plants which serve exclusive territories3 (3) encompassing about 14 percent of the population of the U.S. (RPF 44; Tr. 828, 844). I In addition to Coca-Cola syrup, The Coca-Cola Company manufactures the key syrup and concentrate ingredients for several other soft drink products. These products, including Sprite, Fresca, Fanta, TAB, and Mr. PiBB are collectively referred to as "allied products." The first of these, Sprite and Fanta, were introduced in the early 19608. (Tr. 518-19 692)0 2 The Thomas Company respondents are the successors in interest of Job. Thomas, one of the original purchasers of Coca-Cola bottling rights, whose exclusive. territory covered states in the South, Southeast, and northward alonp' the eastern seaboard to New York. Respondent Thomas Company granted exclusive bottling licenses to numerous independent bottlers in Alabama, Delaware, Indiana, Maryland, Mississippi, New Jersey, New York, North Carolina, Ohio, Pennsylvania, Virginia, and West Virginia. The Thomas Works respondent licensed bottlers in Alabama Georgia, Kentucky, and Tennessee. Respondent Works 3rd, Inc., granted exclusive bottling licenses to bottlers located principally in Pennsylvania and New Jersey. 3 A subsidiary of The Coca-Cola Company, Canners for Coca-Cola Bottlers, Inc., as its name implies, operates canning plants which produce canned soft drinks for the bottlers, In 1974, 42 percent of the canned product of this subsidiary was produced for the bottling subsidiaries of respondent Coca-Cola, 38 percent was produced for the independent bottlers, and 20 percent was produced for sales overseas. (Tr. 846). 4 The following abbreviations are used for citations: ID - Initiai Decision of the Administrative Law Judge; IDF - Initial Decision Finding;

(Qmtill/ud) Opinion 91 F. C:..-~­ The rest of respondents' bottlers are relatively independent business­ men who conduct their commercial affairs as they see fit, subject to three key limitations:

First, when The Coca-Cola Company decided to sell the rights to bottle its product, it agreed to sell to its bottlers a continuous supply of the necessary soft drink syrups, but it refused to yield the secret Coca- Cola syrup formula which would have enabled the bottlers to produce the syrup themselves. Later, when the allied products were introduced it adopted a similar policy. As a result, respondent The Coca-Cola Company is the bottlers' only source of vital Coca-Cola and allied product syrups or concentrates used in the preparation of the finished soft drinks.

Second, respondent Coca-Cola has retained the right to establish quality standards for the products which carry its trademarks and to insist that the bottlers maintain those standards. Failure on the part of a bottler to meet the quality standards it has established may trigger one of the few contingencies justifying the forfeiture of a bottler bottling rights. (Tr. 778). (4) Third, respondents have imposed, by contract, and have enforced, in practice, the territorial restrictions which prevent these independent soft drink bottlers from competing with one another in the sale of bottled, canned, and pre-mixed Coca-Cola and the allied soft drink products made from the syrups and concentrate ingredients produced by The Coca-Cola Company.6 It is this latter interference with the Tr, - Transcript of Testimony;

CX - Commission Exhibit;

RX - Respondents' Exhibit;

App. Br. - Complaint Counsel's Appeal Brief; Ans. Br. - Respondents' Answering Brief;

Rep. Br. - Complaint Counsel's Reply Briefs; CPF - Complaint Counsel's Proposed Findings of Fact; RPF - Respondents' Proposed Findings of Fact; IFF - Intervenors' Proposed Findings of Fact. Unless otherwise indicated. "respondent" in the singular refers to The Coca-Cola Company. 5 The bottler purchases these ingredients from respondent Coca-Cola, and if he was originally licensed to bottle Coca-Cola by one of the Thomas Company respondents, the Thomas Company receives a copy of the purchase order and a commission on the sale. (Tr. 631, 817-18, 855). 8 Respondents make no attempt to understate their firmness in enforcing these restrictions. (Tr. 669). As a consequence, border disputes involving sales of bottled and canned Coca-Cola and allied products by one bottler into the territory of an~ther are rare and usually insignificant. (RPF 47-54, IDF 68-65). According to the testimony of Mr. J. Lucian Smith, President of The Coca-Cola Company, respondents have a system to detect unusually large syrup orders by a bottler which may indicate extra-territorial sales. Moreover, respondents candidly submit that: If an instance of transshipment is brought to the attention of The Coca-Cola Company, it willattempt to contact the bottler from whose territory the product was alleged to have come, and almost always "the bottler does what he can to stop the practice." (RPF 54). Should a bottler refuse to heed such a warning, his supply of syrup or concentrates may be rationed. Thus: , .. in Taft,California, when it was clear that a bottler was purchasing extra quantities of Coca-Cola syrup for the purpose of selIingCoca-Cola in cans outside his territory. (sic) The Coca-Cola Company sold the bottler only enough syrup to meet existing demands and likely growth in demand within his territory, (RPF 54). Respondents have, for the better part of this century, successfully confined their bottlers geographically and prevented intrabrand competition among the bottlers in the sale of Coca-Cola and allied products i1fbottles and cans. .... .

THE COCA-COLA CO., ET AL. 609 517 Opinion bottlers' geographic markets which resulted in the complaint now before us. In essence, this complaint alleges that these territorial (5) restrictions injure competition among the bottlers and deprive retailers and consumers of the benefits of open competition in the sale of Coca- Cola and the allied products packaged in bottles and cans. After a lengthy trial which delved in detail into the day-to-day business of bottling soft drinks the administrative law judge issuedhis initial decision in which he concluded that territorial restrictions are, in the context of the soft drink industry, procompetitive. Accordingly, he entered an order dismissing the complaint, and counsel supporting the complaint have appealed.

In addition to complaint counsel and the named co-respondents there are 14 independent Coke bottlers and the Coca-Cola Bottlers Associa­ tion taking part in these proceedings. In 1971 this association included 99 percent of the domestic bottlers of Coca~Cola. At various times during the pretrial, these bottlers and their association were granted leave to intervene with rights of full participation before the administrative law judge. The intervenors filed briefs on appeal and were afforded time to present oral argument before the Commission. Also participating at the oral argument and on brief were Consumers Union Consumer Federation of America, and National Consumer Congress. The consumer organizations were, by order entered March 2 1976, granted leave to appear, amici curiae, and the respondents and intervenors were authorized to file additional briefs in response toamICI.We have carefully reviewed the arguments advanced in briefs and at oral argument in light of the record and the initial decision and have concluded, for the reasons stated below, that the territorial restraints respondents impose on their independent bottlers are unreasonable and in violation of Section 5 of the Federal Trade Commission Act. Our order will lift the restrictions which place limitations on the sale Coca-Cola and allied products packaged in pre-mix containers, or in nonrefillable, nonreusable bottles and cans. For reasons discussed in detail later in this opinion, we find it unnecessary to disturb. the exclusive territorial relationships with respect to the sale of these products packaged in returnable, refillable bottles. The Commissi~m has also given careful consideration (6) to the arguments respondents and the bottler intervenors advocating geographic market segmentation as a legitimate method of protecting "small" bottlers from intrabrand competition. ' We have reviewed, in-depth, the 7 Each respondent is engaged in commerce as "commerce" is defined in Section 5 of the Federal Trade Commission Act, and the acts and practices challenged in this proceeding occur in the course of such commerce. (CPF 668-681; CX 59-72; Tr,812--17, 664--65; RPF 50-54).

. . _.. ... Opinion 91 F.

evidence and the precedents cited in support of this contention, and have concluded that this argument is without merit. Accordingly, we hereby vacate the judge s order dismissing the complaint and his findings of fact and conclusions 8 and substitute in their place the findings and conclusions noted in this opinion. II. SCOPE OF REVIEW A. CLASSIFYING THE RESTRAINTS In their briefs on appeal, both amici and complaint counsel contend that these restrictions are unlawful; complaint counsel believe that the trial record as a whole will, upon de novo review by the Commission demonstrate that the challenged practice constitutes an unreasonable vertical restraint of trade. They also take an alternati~e position: that the restraints are per se illegal horizontal market division agreements. (App. Br. p. 10). The consumer organizations, appearing amici curiae urge that the practices be declared per se illegal horizontal and vertical restraints on the distribution of Coca-Cola and the allied products lOunder the Supreme Court' s decisions in Schwinn9 and Topco. While the appeal in this matter was pending, however, the Supreme Court in (7) Continental T. V. , Inc. v. GTE Sylvania, Inc. 433 U.S. 36, 1977-:­ Trade Cases, ~61 488(1977), overruled the vertical per se rule stated in Schwinn but it did not rule out the application of a per se standard in appropriate vertical restraint cases. The court noted that in overruling Schwinn:

we do not foreclose the possibility that particular applications of vertical restrictions might justify per se prohibition under Northern Plre. R. Co. But we do make it clear that departure from the rule of reason standard must be based upon demonstrable economic effect rather than - as in Schwinn - upon formalistic line drawing (Id. at 71 902).

In the aftermath of GTE, Topco-type market division agreements among competitors clearly remain per se illegal (GTE, supra at 71 901 fn. 28), while supplier-imposed vertical territorial restrictions must generally be policed under the rule of reason unless it can demonstrated that, in a particular situation, they typically have or are likely to have a "pernicious effect on competition" and that they "lack . any redeeming virtue. . (Id. at 71 902). Under the court' s most 8 A comparison of respondents' proposed findings and briefs with the initial decision shows that respondents and the judge were of like mind to an extraordinary degree on all key disputed issues. We have carefully considered each of these findings in light of our ownnovode review of the entire record and have determined that the judge erred in the legal and factual conclusions which he drew from the evidence. For example compare IDF 183-187 with RPF 326­ 329; IDF 188, 189 with RPF 333; IDF 190, 191 with RPF 336; IDF 192 with RPF 337; IDF 193 with RPF 339; IDF 194, 195 with RPF 341. (But seeText at 65-77infra). S. v, Arnold, &hwinn Co. 388 D.S, 365 (1967). 10 S, v. Topeo Associates, Inc. 405 D.S, 596 (1972). THE COCA-COLA CO., ET AL. 611 517 Opinion recent pronouncement, then, the first step in evaluating these restraints is to classify them as horizontal or vertical. The Topco Theories The Coca-Cola Company has over the years, by acquisition integrated forward into the bottling business. Thus complaint counsel assert that the territorial restraints on the distribution of the Cor.a- Cola brand soft drink were vertical when respondent was simply a supplier of soft drink ingredients, but now that it has acquired bottling facilities, the restraints are horizontal. In addition, when the Coca-Cola Company, wh~le operating its bottling subsidiaries, introduced its allied product lines under licensing agreements which granted exclusive territories to its independent bottlers as well as its own bottling facilities, it allegedly became involved in a "horizontal" market division scheme for the sale of the allied products. (App. Br. 55-56 Amici Br. 13). Amici and complaint counsel contend that geographic market restraints imposed under these circumstances serve no purpose except to stifle competition. Both situations are said to constitute per se illegal horizontal market divisions under the Supreme Court' decision in Topco. (8) a. Acquisition of" Bottling Subsidiaries by The Co~a-Cola Company Although The Coca-Cola Company is both a supplier of syrup and a soft drink bottler, the record as a whole demonstrates that the restraints involved here are not primarily "horizontal" within the meaning of the court' Topco decision. Admittedly, the line which separates the "vertical" from the "horizontal" forms of a geographic market allocation arrangement is not always as easy to distinguish as the market plane to which they refer might tend to indicate. Both types of restraints at times may, at a given level of production or distribution, exhibit similar competitive characteristics which, on the true source. (U.S.surface, obscure the firm or firms which are their Sealy, Inc. 388 U.S. 350, 352 (1967)). Consequently, only by ignoring the essential relationships which exist between the respondents and the independent bottlers might it be concluded that the restraints (ire Topco-type horizontal" market allocations based solely on the fact that respondents operate bottling facilities and are thus potential competitors of the independents, and vice versa. l1 (9) II Dual-distributing manufacturers and their independent wholesalers obviously can be "in competition with each other" and have so been adjudged in cases which have, for example, construed the scope of the now-repealed Fair Trade Law exemptions to the Sherman and Federal Trade Commission Acts.See U.S, v. McKesBOn Robbi1UJ, Inc. 351 S. 305 (1956), and Rubbermaid, Inc., F, C. Dkt, 8939 (87 F. C. 676). (The Fair Trade Laws were repealed by the (Continual) , .... Opinion 91 F.

The Coca-Cola Company s forward integration by acquisition into the bottling industry did not alter in a substantive way either the nature of the restraints or the implementation (10) policies employed by The Coca-Cola Company with respect to established bottling territorial relationships. These restraints were in place nationwide for several years prior to Coca-Cola s entry into bottling.12 When it Consumer Goods Pricing Act, Pub. Law No, 94-145 (Dec. 12, 1975),) Both cases involved resale price maintenance agreements coupled with supplier-imposed customer restrictions. Notably, the interpretations applied in the fair trade cases cited actually narrowed the fair trade law resale price maintenance immunities, As we noted inRubbe-rmc.id . . . we will construe strictly any provision which deviates from fundamental antitrust policy, for exemptions from the antitrust law are to be strictly construed. . . ," (Slip Opinion, p. 24, fn. 45), Two cases traceable to McKesson have condemned, as "horizontal " agreements between dual-distributing suppliers and their independent distributors,See Interphoto Ccnp,v, Mi1Wlta Ccnp" 295 F. Supp. 711 (S. Y. 1969), affd per curiam 417 F,2d 621 (2nd Cir. 1969) (resale price maintenance and territorial restrictions), cited in Hobart Brothers Co,v, Malcolm T. Gilliland, Inc"471 F,2d 894 (5th Cir. 1973), In Hobart the supplier of welding equipment competed with its distributor in the sale of the equipment to other customers. The supplier and distributor were al~ competitors in the manufacture and sale of continuous wire feeder mechanisms. Efforts by the supplier to police its territorial restrictions, through disparagement and finally tennination of the welding equipment distributorship, damaged the distributor s business in the sale of both welding equipment and wire feeder mechanisms. (Id, at 898, 903), The Fifth Circuit found that the territorial restriction in these circumstances operated horizontally, The court also noted in dicta, however, that agreements limiting the area in which other independent distributors could sell Hobart products in competition with Hobart constituted horizontal territorial allocations,(Id, at 899). In non-fair trade cases, the Supreme Court has not applied the fair trade "in competition" standard in detennining horizontality in dual-distribution, territorial restriction situations. Had the standard been applied, for example, in White Motor the restraints before the court conceivably could have been treated as horizontal arrangements; Justice Clark, citing McKesson in his dissenting opinion, argued as much with respect to White' s customer restrictions. In fact White Motor had reserved to itself the business of selling its trucks to certain types of customers located within the exclusive" territories it granted to its independent distributors White Motnr Co,v. S., 372 U.S. 253 (1963). While the per se rule inSchwinn has been overruled, the opinion contains useful -guidance for purposes of classifying restraints. Notably, Schwinn shipped bicycles directly to retailers, while paying the order-taking distributor a commission on the sales(Schwinn, mpra at 370), and consequently the situation involved substantial participation by the manufacturer in the bicycle distribution chain. The court stated: we are here confronted with challenged vertical restrictions as to territory and dealers. The source of the restriction is the manufacturer. These are not horizontal restraints in which the actors are distributors with or without the manufacturer s participation. (at 372). Later inSchwinn the court again emphasized that it was: . .. dealing here witha vertical restraint embodying the unilateral program of a single manufacturer. We are not dealing with a combination of manufacturers. , , or of distributors. . . , We are not dealing with a division" of territory in the sense of an allocation by and among the distributors. . . or an agreement among distributors to restrict their competition, , . . We are here concerned with a truly vertical arrangement. . . (at 378, citations omitted).

12 Territorial monopolies, intrabrand, have been a dominant characteristic of respondents' distribution system since the beginning of the Coca-Cola bottling business, Looking back upon respondents' humble - origins, exclusive territories may have, as they contend, been necessary to attract local businessmen to invest in their bottling venture, We certainly ascertain nothing in the record which disputes respondents' characterization of the difficulties encountered by those who labored, nearly three-quarters of a century ago, to solicit investor interest in soft drink bottling.

Prior to 1900, bottled Coca-Cola was virtually unknown. At the time, Coca-Cola syrup was sold almost exclusively through fountain jobbers to retailers who perfonned the function of mixing the syrup with carbonated water, and the finished soft drinks were served, most often for immediate consumption by the consumer, at the retailer s place of business, The demand for Coca-Cola in containers capable of maintaining its effervescence which could be purchased at the store and taken home for later consumption was, in fact, an outgrowth of the fountain business. A brief survey of the economic landscape of 1900, as revealed in the record, leads us to conclude that businessmen of that era probably considered soft drink bottling little more than a newfangled invention with a questionable future, Having never before been done to any significant degree, it had virtually no financial track record to guide potential investors, Even the management of The Coca-Cola Company at the time had serious reservations about its feasibility, Coca-Cola bottling was not an innovation of The Coca-Cola Company; rather, it appears from stipulated record ­ evidence that its then-chief executive probably considered the scheme to bottle the product an undertaking more suited to the taste of adventurous speculators than serious investors, Thus viewed in its historical context, soft drink bottling was a fledgling industry when terri~rial exclusivity was (Continued) .,..., THE COCA-COLA CO., ET AL. 613 517 Opinion acquired a bottler, The Coca-Cola Company itself became subject the (11) same territorial limitations it had previously imposed upon the acquired bottler. (Tr. 512-13, 527). 13 With each (12) acquisition, then The Coca-Cola Company merely replaced an independent bottler within a preexisting distribution scheme.14 No evidence was introduced that the acquisitions actually changed either the competitive effects of the territorial restrictions or the basic relationships among the bottlers. While it is true that respondents may at times resolve border disputes involving territorial boundaries which occasionally erupt among the bottlers, unlike Topco it has not been established on this record that the independent bottlers exercise control over any respondent or the way in which a respondent implements the territorial aspects of its trademark licensing programs. See U.S. v. Sealy, Inc. , supra. Nor has it been established on this (13) record that the tapestry of Coca-Cola bottling territories is the product of horizontally contrived arrange­ ments among the bottlers actively blessed or passively accepted by any respondent. (See Fontana Aviation, Inc. v. Beech Aircraft Corp. 432 2d 1080, 1084 (7th Cir. 1970).

originally awarded to Mr. Thomas and others, and, by them, subsequently in smaller parcels to hundreds of local bottlers, In this way, they attracted the manufacturing and distribution capital to develop a new business and to expand the sale of a new product, finished Coca-Cola in bottles, into new markets. In these circumstances, the language in White Motm Co, v. United States,372 D,S. 253 (1963), quoting from Justice Brandeis inChicago Board of Trade, is appropriate:

The history of the restraint, the evil believed to exist, t~e reason for adopting the particular remedy, the purpose or end sought to be obtained, are all relevant facts. This is not because a good intention will save an otherwise objectionable regulation or the reverse; but because knowledge of intent may help the court to interpret facts and predict consequences, (at 261). Evidence concerning the history of respondents' territorial restrictions and the essential relationships which have . existed basically unchanged over the years among the respondents and between them and the independent bottlers confirms our conclusion that the restraints on the sale of Coca-Cola are not the offspring of a horizontal conspiracy or collusive horizontal agreements.

13 The record shows that there are several types of Coke, bottler licensees, (CPF 83). Those bottlers which originally acquired the rights to bottle Coke directly from The Coca-Cola Company or its predecessors are known as parent bottlers. This category now includes only the Thomas Company respondents, the other parent bottlers having been acquired by The Coca-Cola Company. The parent bottlers,in turn, parceled out pieces of their territory in which they granted exclusive rights to local investors known as first-line bottlers. (CPF 84). Territorial restrictions were imposed upon each of these first-line bottlers by the parent bottlers. In some instances, first-line bottlers have further carved up their territories and have licensed others, known as "sub-bottlers " to bottle Coca-Cola on an exclusive basis, 14 This conclusion is corroborated by the fact that The Coca-Cola Company entered Baltimore not as a parent bottler, but as a first-line bottler. It operates within the exclusive territory of the bottler which it acquired, and its parent bottler is a Thomas Company. (App, Br, 23, 55). Furthermore, the record shows that several bottling facilities were acquired by respondent to assure product availability in territories in which the independent bottlers were leaving the market and other independents with sufficie~t capital and know-how were unavailable to take their place. (Tr. 913, 922).

It should be noted that soft drink syrup producers and particularly small manufacturers may be able to enter new markets nationwide to compete with dominant firms like Coca-Cola and Pepsico by offering exclusive trademark licenses of limited duration to existing bottlers or by encouraging new bottlers into the market. If the search for independent capital is unsuccessful or if an independent bottlerdecides to withdraw from the market, a syrup company may then decide to integrate vertically in order to preserve its market position, Should it, in fact, integrate under these circumstances, it would, of course, be entering the "bottling level " but we do not read Topeo condemning this type of dual-distribution program as a horizontal market allocation arrangement, 15 An aggregation of geographic restraintsdesigned by a franchisor for the purpose of eliminating both intrabrand and interbrand competition between itself and its franchisees may, under certain circumstances, result in a horizontal" allocation of markets.See American Motor Inns, Iru;, v, Holiday Inns, 1m" 521 F,2c\ 1230 (3rd Cir, 1975), in which the court concluded that since the franchisor: (Continued) Opinion 91 F.

This is not to say that the type of territorial restrictions traditionally considered vertical are devoid of horizontal competitive implications; but on the facts before us, we cannot conclude that the horizontal aspects of these restraints are, for classification purposes, predominant in the Topco sense simply because they now prevent intrabrand competition among independents and Coca-Cola s subsidiaries, whereas previously they functioned as a barrier to intrabrand ~ompetition only among independents. In the latter situation and in makrets in which respondent Coca-Cola entered the distribution system below the level of a parent bottler, as it did in the Baltimore territory, complaint counsel concede the restraints are vertical (App. Br. 55), and for all that appears in the record, the essential nature of these restraints in instances respecting the distribution of bottled and canned Coca-Gola despite The Coca-Cola Company acquisition of parent bottlers remains vertical.16 (14) Introduction of New Product Lines by a Dual-Distributing Supplier/Trademark Licensor The allied products of The Coca-Cola Company, TAB, Sprite, Fresca Fanta, and Mr. PiBB were developed by respondent, at least in part, to satisfy the demands of its bottlers for additional soft drink flavor lines. These products were first introduced in the early 1960' , long after The Coca-Cola Company had entered the bottling level, and were offered to the integrate bottling operations and the independent Coca-Cola bottlers alike on an exclusive basis for distribution within their existing Coca-Cola bottling territories.17 Allied product licenses were granted by The Coca-Cola Company directly to the bottler. Unlike many of the Coca-Cola licenses, no parent bottlers are involved in these licenses. 18 Consequently, complaint counsel view the territorial aspects of these allied product licenses as market allocation agreements . .. in one of itscapacities, was dealing on the same market level as its franchisees, its contracts that, in effect, foreclose such franchisees from operating either Holiday Inns or non-Holiday Inns in cities where HI operated an inn, except with HI's permission, constituted market allocation agreements among competitors. (at 1254), Respondents' bottlers, in contrast, are not prevented from manufacturing or distributing soft drinks trademarked by competing syrup companies; nor do respondents have any control over the geographic area in which its bottler may distribute such products.

16 In Adolph Cows Company, 83 F, C. 174, the Commission considered, strictly in a vertical context, an aggregation of trade restraints, including price fixing and territorial restrictions, by a brewer which distributed its products through independents and "a wholly-owned subsidiary of respondent" (83 F. C, at 175), 17 See Tr, 540-41, CX 104A, CX 1ioa, CX lisa, 119A, 121A (allied product territories of the/Baltimore subsidiary); and for example, CX 199A, CX 202A, CX 2O6A, CX 2O7A (allied product territories of the Richmond bottler); CX 256 CX 259A, CX 264A (allied product territories of the Washington bottler); CX SMA, 565A, 566A (allied product territories of the Dover, Delaware bottler). In each instance, the boundaries within which the bottler may produce and distribute the allied product are identical to the territorial boundaries specified in its Coca-Cola license. IR The Coca-Cola bottlers were not required to handle the allied products, and many which were already producing soft drinks, made from syrups produced by other syrup companies such as Dr. Pepper or Sunrise flavors, declined the license for certain Fanta flavor lines or Mr. PiBB, Coca-Cola s "Pepper-type" drink. .... .

THE COCA-COLA CO., ET AL. 615 517 Opinion between potential competitors; specifically, respondent's own bottling subsidiaries and the independent bottlers.

While the allied product licenses are conferred by a manufacturer which also produces and sells finished soft drinks at wholesale to retailers within exclusive territories, absent evid~nce of collusive activity among the bottlers, we conclude that the introduction of new product lines by a vertically integrated soft drink syrup company using its existing channels of distribution would not, under White and its progeny, necessarily render the bottling agreements "horizontal-" (15) Since complaint counsel have the burden of proof, we shall assume that the "allied product" trademark licensing programs for each flavor line were conceived by The Coca-Cola Company, acting unilaterally as the syrup and concentrate supplier and trademark licensor. No evidence to the contrary was introduced. The record as a whole does not evidence any collusion among bottlers concerning the allied product territories or that bottlers jointly participated in or exerted any control over the territorial aspects of respondent' s allied products distribution scheme. (See GTE, supra at 71 901, fn. 28). Rather, the evidence indicates that respondent, alone, elected to distribute the allied products through the existing network of Coca-Cola bottlers . using the Coca-Cola-type licensing system. While not dispositive of its liability in this proceeding, it is also relevant, for purposes of classifying the restraint as horizontal, or vertical, that complaint counsel failed to demonstrate, in any respect that The Coca-Cola Company s presence at the bottling level substan­ tially altered either the competitive effects of the aliied product restrictions or the essentially vertical relationships respondent had with its bottlers before the allied products were introduced. We conclude that Topco is not applicable in this context. Vertical Per Se Th€O'f"ies As we mentioned previously, the Supreme Court, in overruling Schwinn has not entirely rejected the possibility that vertical restrictions may, in individual cases, be declared per se unlawful, but it has toughened the standard considerably. Only those restraints found to be 1'pernicious" and without "any redeeming virtue" now justify per se treatment. The types of competitive situations, other than price fixing, which may meet this standard are unclear, but beyond that, the trier of fact and appellate tribunals must be receptive to the fact that situations may exist in which the imposition of a vertical restraint may, under GTE still be per se unlawful.19 (16) 19 During the pretrial period following issuance of the complaint, complaint counsel' s predecessors, citing the Supreme Court's decision in Schwinn filed with the administrative law judge then assigned to the case a motion for (Continued) Opinion 91 F. C. .. On the facts before us, we believe the application of a per se rule would be inappropriate. Taking into consideration the competitive dynamics in this industry, there are important unresolved issues in this proceeding concerning whether open intrabrand competition among the bottlers of Coca-Cola and the allied products would adversely affect interbrand competition in the sale of soft drink beverages. The resolution of these issues in this case, we believe, requires a rule of reason analysis. The burden of proof justifying application of a per se standard has not been met on this record. The ~erritorial aspects these trademark licensing agreements, or those which may be imposed by other firms in this industry, have not been shown to be typically pernicious and without redeeming virtue under the Northern Pacif-k2O standard, as adopted in Continental T. V. , Inc. v. GTE Sylvania, Inc. supra. We now turn our attention to the record. III. EFFECTS OF THE TERRITORIAL RESTRICTIONS A. THE "CORRIDOR AREA Although respondents admittedly impose territorial restrictions on virtually all of their bottlers nationwide, at the trial, complaint counsel limited their proof of competitive impact to an area of the country extending from southern Virginia to upstate New York, an area which has been referred to in this proceeding as the "corridor area. Complaint counsel believe the "corridor area" is a microcosm of the soft drink bottling industry as a whole; thus if the restrictions are found to be anticompetitive in this geographic area, the findings can according to complaint counsel, be applied to the competitive situation nationwide. We believe complaint counsel have met their burden establishing the validity of the "corridor area" analysis. Respondents' objections to it notwithstanding, the business of bottling soft drinks in the "corridor area" is, in fact, essentially no different from the bottling business in other areas of the country.

The record shows that within the "corridor" there are urban suburban, and rural bottlers with single-plant and multi-plant operations, large and small bottlers, first-line bottlers, sub-bottlers and marketing bottlers with no production facilities. Several "corridor area" bottlers distribute within a single territory. Others, through consolidations or acquisitions, have obtained the rights to distribute partial summary decision declaring respondents' territorial restrictions per se illegal verticalon therestraintssale finished soft drink products. The judge denied this motion (See Order Denying Motion by Complaint Counsel for Partial Summary Decision, April 5, 1973); interlocutory review of his ruling was not sought; and the case subsequently proceeded to trial, the vertical per se theory having been abandoned. (App, Br, 3, 5 fn. 1). Amici have revived the theory for considerationon appeal.

20 See Narthern Pacific Railroad Co, v. U.S., 356 V.S, 1 (1958), ....

THE COCA-COLA CO., ET AL. 617 517 Opinion Coca-Cola in two or more territories. The "corridor area" also includes both private and publicly owned bottlers, a bottler-owned canning cooperative, a major bottling and canning subsidiary of The Coca-Cola Company, contract canners, and interbrand competitors. In addition Coca-Cola bottlers throughout the (17) country may manufacture and distribute, or "piggyback " soft drinks trademarked by competing syrup companies; and in virtually all instances, they use a route or store-door" delivery system to distribute at wholesale the soft drink products in various package sizes and types which they either bottle themselves or which are produced for them under agency agreements by neighboring bottlers or canners.

While respondents correctly note several perceptible but minor distinctions in the "corridor area" bottling business, those differences are really inconsequential for the purpose of this proceeding. Respon­ dents, for instance, alert us to the fact that the demand for returnable refillable bottles tends to be higher in other parts of the country than in the "corridor area" where convenience packaging seems to be more popular. (Tr. 1345-46, 2871- , 2064, 3781, 2368-69). As a packaging alternative, however, refillable bottles are offered in many markets and are an important factor in several bottling territories including within the "corridor area." The record shows that refill abIes represent 50 percent of the sales of Coca-Cola in bottles and cans in the Richmond territory; 60 percent in Charlottesville; 65 percent in the territory of the Washington, Pa., bottler; 47.9 percent in Westminster Md. ; 41 percent in Dover, De. ; and 74 percent in Montross, Va. (RPF 348). Recognizing, then, that the proportion of soft drinks sold in refillables may be greater in other parts of the country, there is ample use of this form of packaging and sufficient investment by the bottlers in refillable bottle inventories or "float" within the "corridor" to safeguard against any significant distortions in our analysis. Nor are we persuaded by the argument that the "corridor area" is atypical of the nation as a whole merely because territories may tend to be larger and the population ratio of large and small bottlers may vary in other areas of the country. (Tr. 133~7, 1345, 3266-67). We believe the record provides ample support for complaint counsel's contention that the ~'corridor area" represents a reasonable crosssection of the bottling firms which operate throughout the country. Setting aside respondents' protestations and references to insignificant distinctions in "corridor area bottling, we feel that an accurate assessment of the competitive dynamics in the territories of both large and small bottlers and the interrelationships between bottlers which would, absent the territorial restrictions, be likely to result can made on this record. Respondents called, as defense witnesses ....

Opinion 91 F.

numerous bottlers from Georgia, Iowa, Texas, California, and other locations beyond the "corridor area." Their testimony is remarkably similar to the testimony of the bottlers situated within the (18) corridor " including their assessments of the competitive effects of the restrictions under present market conditions and their estimation of the likely consequences of a Commission order eliminating the restraints. Under these circumstances, we find no basis for dismissing the "corridor area" as too dissimilar to the rest of the country to support an analysis of the nationwide competitive impact of respon­dents' trade restraints. B. SUPPRESSION OF INTRABRAND COMPETITION AMONG RESPONDENTS BOTTLERS Respondents acknowledge that territorial restrictions prevent intrabrand competition among their bottlers, but claim this effect is actually procompetitive and necessary in the interest of promoting the overall efficiency and productivity of its bottler network. (Ans. Br. 12­ , 54). Respondents contend, moreover, that the admitted restraint of intrabrand competition is of no concern unless "the restraint is imposed by parties with excessive market power " the "principal indication" of which "is the ability to set the price for a product free from the influence of interbrand competition. " (Ans. Br. 45, 47). On this premise they further contend that the evidence does not show that respondents have "unrestricted market power" with respect to price, packaging, or service (Ans. Br. 47), and that evidence concerning market share and profits does not demonstrate that Coca-Cola has "dominant or monopoly power." (Ans. Br. 49). Implicit in this contention is the idea that absent such market power the asserted efficiency and productivi­ ty benefits of restrained intrabrand competition will be passed on to the consumer as a result of interbrand competition. We do not agree that a showing of "dominant or monopoly power" or unrestricted market power" is necessary before it may be concluded that suppression of intrabrand competition is unreasonable and in violation of Section 5. Respondents and the ALJ cite the decision in United States v. Columbia Pictures Corp. 189 F. Supp. 153 (S. 1960), where the court made the following summary of the doctrine of ancillary restraints (id. at 178):It permits, as reasonable, a restraint which (1) is reasonably necessary to the legitimate primary purpose of the arrangement, and of no broader scope than reasonably necessary; (2) does not unreasonably affect competition in the marketplace; and (3) is not imposed by a party or parties with monopoly power.

Thus, the court did not hold that market power must be demonstrated THE COCA-COLA CO., ET AL. 619 517 Opinion Or- . before a restraint could be held unreasonable under the Sherman Act but rather held only that the absence of monopoly power was (19) one of several prerequisites before a restraint might be held reasonable. Indeed, in GTE the Court indicated that even a less sweeping restraint on intrabrand competition than we have before us here could be found unreasonable without a showing of market power, even though the coippany imposing the restraint had a small market share and was Jar removed from the dominant firm in the industry. GTE, supra 893.

While the territories in which Coca-Cola and the allied products are sold are not devoid of interbrand competition, nevertheless Coca-Cola and allied product prices have great competitive significance in the marketplace.21 Moreover, the record amply demonstrates that respon­ dents' territorial restrictions constitute a serious impediment to free market forces and diminish competition in the manufacture, distribution, and sale of several important soft drink product lines. The record also shows that intrabrand competition would invigorate price competition which would be likely to produce lower wholesale prices for Coca-Cola and the allied products. (Tr. 739, 887-889, 992-93, 1568 2459 2885). By suppressing the development of intrabrand competition in the sale of these products packaged in bottles and cans, the restrictions have, over the years, distorted the competitive dynamics of the industry, and have disrupted the natural economic forces which would have, in the absence of restraints, caused an evolution in the geographic market boundaries of respondents' bottlers. (20) Before we consider whether these restraints promote interbrand competition and efficiencies in distribution, as respondents contend, we must take a closer look at the intrabrand effects of the restraint. Intrabrand Syrup Jobbers Respondents argued below and again on appeal that Coca-Cola sold by licensed bottlers in bottles, cans, and pre-mix containers is subject to "vigorous" intrabrand competition from post-mix Coca-Cola syrup 2\ The complaint in this matter defines soft drink products as including non-alcoholic beverages and colas carbonated and uncarbonated, flavored and non-flavored, sold in bottles and cans, or through pre-mix and post-mix systems, or the like. (Complaint para, l(h)), Within this broad product market definition, however, there may be a number of relevant submarkets. For example, in Sulmeyer v, The Coca-Cola Company, 515 F.2d. 835, 848-49 (5their. 1975), the court found that a lemon-lime flavor segment of the soft drink market was a relevant submarket and, further, that all independent bottlers, as urged by The Coca-Cola Company in that case, constituted a relevant market. We note, however, that the trialbelow explored the implications of these restraints in an exceedingly broad framework which encompassed interbrand competition within the total context of the soft drink industry. The trial did not focus on structural characteristics in various arguably valid submarket categories; nor did it isolate the competitive effects of these restraints within strict submarket contexts. In a light most favorable to respondents, a record of competitive impact was developed in the context of virtually every liquid, except alcoholic beverages, a person may consume. Particular emphasis however, is placed on flavored carbonated soft drink beverages since virtually all of the bottlers tended to place their emphasis on these beverages in describing competitive products which influence their business decisions.

Opinion 91 F. C:-­ sold by independent wholesalers for use primarily at soda fountains and in cup vending machines. (RPF 171- , IDF 133-34). While the bottlers distribute the packaged finished soft drinks whhin exclusive territories, a syrup jobber is free to sell post-mix syrup in any geographic market in which a demand for the syrup exists to any customer who has a proper use for it. Several independent wholesalers may compete in the sale of post-mix syrup in any given area, including a few bottlers of Coca-Cola who also wholesale the post-mix syrup primarily to the cold drink trade and, like the jobber, may independent­ ly decide where and to whom they will distribute it. (RPF 171 , Tr. 1941).

In his initial decision, the judge, without qualification, found that intrabrand competition between jobbers of post-mix syrup and the bottlers of packaged finished soft drinks is indeed "vigorous." Only by ignoring relevant supply and demand factors, including the fact that the bottler sells a packaged product which is frequently purchased by the consumer in quantity and stored at home for later consumption would this conclusion be sustainable. (See Tr. 2384-85, 1684). The soft drink bottling industry grew out of the business of selling syrup to soda fountain retailers, but it has always been viewed by respondents and the bottlers as a different business. (Res. Ans. Br. 3-5 10-11; Tr. 1572-73 3262). See The Coca- Cola Bottling Co. v. The Coca- Cola Co. 269 F. 796 (D. Del. 1920)). This is evidenced by a relationship between The Coca-Cola Company and its bottlers predicated on the distinction between syrup sales to retailers who serve soft drinks to consumers for on-premise consumption and the sale of packaged finished soft drinks to retailers who resell it to consumers for home consumption. This distinction is as valid today as it was when respondent Coca-Cola sold its rights to manufacture and distribute bottled Coca-Cola to Messrs. Thomas and Whitehead. Admittedly, for certain types of soft drink retailers, there is a viable option to purchase either finished packaged soft drinks from a bottler or post-mix syrup which they can mix with carbonated water just as a bottler would, but the choice is really available only to retailers, such as restaurants, fast-food retailers, cafeterias, sports stadiums, and (21) other types of outlets which serve Coca-Cola in cups, bottles, or cans for immediate consumption. (Stip. No. , CX 1244--1). Competing for these accounts against the Coca-Cola post-mix wholesaler, however, a bottler is at a serious disadvantage precisely because he is selling a finished packaged product.

Unlike the bottling and canning of Coca-Cola and other soft drinks post-mix wholesalers are not required to perform any of the manufacturing functions a bottler performs. Nor is the wholesaler .,..

THE COCA-COLA CO., ET AL. 621 517 Opinion required to provide any dispensing equipment or service and often he does not perform any delivery functions since the post-mix syrup is frequently drop-shipped by The Coca-Cola Company directly to the retail customer.

The record further shows that fountain syrup is often incidental to the b9ttlers' overall business to the point that they make no effort to sell it. Mr. Navarre, Chairman of the Boards of the Coca-Cola Bottling Co. of Miami, the Delaware Coca-Cola Bottling Co., and the Coca-Cola Bottling Works of Havre de Grace, Maryland, testified about the fountain syrup business:

believe you stated that you don t sell fountain syrup - why have you elected Q. - I not to? Is there a contractual part of your doing so? A. No, sir, it is a competitive situation and ability to be able to furnish to these dealers at this price and the profit contribution under our form of doing business is not sufficient to interest me. (Tr. 1554--55).

Conversely, in selling to other types of outlets, such as retail food stores which cater to a substantial market for Coca-Cola and the allied products in take-home packages, the bottler need fear no intrabrand competition from any of the post-mix wholesalers. This comports with the basic rationale of the soft drink bottling industry. (See Tr. 4080--81). In fact, the entire bottling industry exists because of its ability to . service the demand for soft drinks in take-home packages which the fountain syrup wholesalers have never been able to reach. (See Tr. 1457). Consequently, in the sale of soft drinks in bottles and cans for home consumption, which the bottler alone is uniquely equipped to serve, intrabrand competition from post-mix wholesalers is virtually nonexistent. Mr. Navarre s testimony amply demonstrates that the intrabrand competition which may exist between syrup jobbers and bottlers is confined to a limited, rather well-defined class of customers who cater to the cold drink market, and even as limited, there will be competition between bottlers and jobbers only if the bottler elects to expand into the cold drink trade. Thus a bottler may,. in some instances actively solicit cold drink accounts, but jobbers are, by the nature of their product, foreclosed from competing for bottlers' take-home business. (22) Contrary to the judge s finding, then, it is evident there is virtually no direct competition between syrup jobbers and bottlers for the bulk of the bottlers' business to their traditional food store and other accounts which serve the consumer demand for Coca-Cola, TAB Sprite, Fresca, and other allied products in take-home packages. ..

Opinion 91 F.

2. Territorial Restmtions Prevent Procompetitive Geographic Market Expansion and Eliminate Potential Competition Complaint counsel contend that respondents' territorial restrictions rather than fostering greater efficiency, actually deter progress and the efficiency of the bottlers because they prevent the type of production and sales expansion which would enable bottlers to achieve maximum scale economies and further prohibit or discourage the bottlers from taking maximum advantage . of i~proved production distribution, transportation, and communications systems developed in the last five decades or so. (App. Br. 57). Respondents vigorously dispute each of these contentions. In their view, bottlers large and small have been able to adapt to changing economic conditions, to expand their sales within their territories, and to employ innovative techniques of marketing and packaging. (Ans. Br. 81). Respondents are correct in their assertion that many of the adaptable technological breakthroughs of the 20th century have not bypassed the bottling industry. Bottling territories were originally parceled out at a time when bottling facilities used manual equipment and finished soft drink products were delivered in horse-drawn wagons over dirt roads. (Tr. 681, 1656-59). Today, in contrast, even the small bottler uses modern delivery trucks (RPF 292), and unlike his predecessor, he operates on a much more efficient production-line basis, using automated equipment which cleanses containers and purifies and carbonates water. He has mechanized systems which mix the syrup and water, fill and cap the bottles, and package the filled containers at varying speeds depending upon the bottle size and the type of bottling equipment used.

These modern automated production lines have, in addition increased the potential production capacities of both large and small bottlers. At present, soft drink bottlers often produce and distribute, or piggyback " the soft drinks trademarked by several syrup companies and may, at times, distribute these brands in exclusive territories of various sizes assigned to them by different syrup companies. (See Tr. 3078, 3063- , 3067, 3236). Some Coca-Cola bottlers are also capable of supplying, in addition to the soft drink requirements within their own territories, the requirements of other Coca-Cola bottlers who have retained a territorial monopoly for the distribution of Coca-Cola but have temporarily discontinued producing it themselves (Tr. 529-30 555 , 788-89); other bottlers have entered (23) into agency arrange­ ments to supply neighboring bottlers with their requirements for certain package sizes 22 or have, by consolidations and mergers 22 The small Pepsi bottler in Dyersburg, Tennessee, also piggybacks Dr. Pepper and Bubble-Up and, working (Continued) THE COCA-COLA CO., ET AL. 623 517Opinion 0,-combined their territories, efficiently serving from one production center an area previously serviced by two separate bottling facilities. Originally, the bottlers' territories probably represented a rather close approximation of the geographic boundaries which would have existed in the industry if natural economic forces were left unre­ strained. While territories were granted in various sizes and shapes they probably encompassed an area roughly measured by the distance a turn-of-the-century vehicle could travel in one day. (Tr. 681). Given the technological and transportation limitations of the late 19th and early 20th centuries, under which the original bottlers operated, it seems reasonable to conclude that most territories probably covered an area not significantly smaller than the Coke bottler was capable of servicing efficiently and effectively. As time passed, however, the potential for direct competition among respondents' bottlers grew as automated production of soft drinks replaced manual bottling lines, as new types of packaging were introduced, and as truck transport and road surfaces improved. Despite these advancements, however respondents' territorial system stands impervious to natural geograph­ ic market evolution and procompetitive market extension by indepen­ dent bottlers.

3. Territorial Restrictions Indirectly Lessen Competition in the Delivery Services Bottlers Offer to Their Customers The record also shows that the restrictions impede the bottlers ability to respond to the demand for competing delivery services. Since the beginning of the Coca-Cola bottling business, the bottlers have used, almost exclusively, a route-delivery system (or store-door delivery as the bottlers refer to it) which entails frequent, direct delivery by the bottler to each of the customer s retail outlets. In the early days of this business before the chain store, central warehouse era of the 1930s and (24 J the introduction of nonrefillable containers in the mid 1950s 23 there may have been few competitive alternatives to store-door delivery. Today, as a result of soft drink packaging innovations improvements in transportation, and the widespread use of central warehouse facilities by retailers and independent wholesalers there is a market for service options, such as central warehouse overtime, was still able to supply larger bottlers, including the large Memphis Dr. Pepper bottler, a subsidiary of RKO, with Dr. Pepper in 32-ounce returnable bottles for ten months. Similarly, the Coke bottler in Las Cruces, New Mexico supplied the Coke bottler in San Antonio with 64-ounce nonreturnable bottles of Coca-Cola for a period of four months (RPF 253; Tr. 2483-84 2511-12), and the small Northern Neck, Va., Coca-Cola Bottling Co. supplied Coca-Cola in 32­ ounce returnable bottles to other bottlers, including the large Crass organization in Richmond. (Tr.1~6). 23 Pressure from competitive packages forced The Coca-Cola Company in 1955 to abandon its single-package (6 1/2 oz. returnable, refillable bottle) philosophy and authorize the bottlers to use various size refillable bottles and nonrefillable bottles and cans. (Tr. 714, 1344), Opinion 91 F.T:C:

delivery and plant pick-up by central warehouse and other customers and respondents' bottlers have the capacity to exploit it.24 Yet notwithstanding the demand for these competitive delivery services, a bottler may not, consistent with respondents territorial policy, ship to central warehouses or allow plant pick-up in instances which will result in distribution of the product by the customer outside the bottler territory. (RPF 49).

While the bottlers who appeared at the trial testified that they prefer store-door delivery to central warehouse delivery because it promotes deep market penetration and allows them to maintain some measure of control over the way the product is merchandised by the retailer on the retail shelves (See also RPF 49), it also appears that store-door delivery is preferred today by many bottlers, at least in part, because it is completely compatible with the preservation of exclusive (25) territories. (Tr. 1901.25 In fact, respondents and the bottlers concede not only a strong market demand for central warehouse delivery and plant pick-up by central warehouse customers (RPF 88--90), but also that some bottlers would (26) provide a competitive response to this demand were they free to do so. (Ans. Br. 55). 26 Consequently, by hindering central warehouse and plant pick-up H Central warehousing involves the purchase of soft drinks by the warehouser directly from a bottler or canner for delivery into the purchaser s warehouse, Subsequently, redelivery of the soft drinks is made in the warehouser own trucks to the warehouser s retailoutlets, Warehousers may. themselves be retailers (such as large chain supermarkets) who buy for redelivery to their own outlets in their own trucks, or independents who buy for redelivery to non-affiliated outlets or retailer warehouses. Although the agreements between respondents and the bottlers do not directly prohibit warehouse delivery, respondents concede that the bottlers may not sell Coca-Cola and the allied products to central warehouse customers or allow plant pick-up where the result would be redistribution of these products outside the selling bottler s territory. (RPF 47-49). As a consequence, respondents' territorial policy has indirectly but effectively blocked the development of these alternative modes of delivery.

25 Respondents and complaint counsel have joined issue over the comparative efficiencies of warehouse delivery and route delivery, A study of both methods of distribution prepared by respondents' expert, Mr, Cowart, shows that the average costs of delivering soft drinks packaged in nonreturnable bottles and cans are approximately the same for route delivery or warehouse delivery, Mr. Cowart' s testimony indicates, for example, that the average cost of delivering 3~unce nonreturnable bottles through warehouses would be 9,6 cents more per case than the cost of current store-door delivery, (Tr, 343&-39). A case of cans is an ideal package for central warehousing because cans are a compact, low-cubage container. Here average costs vary from 3-5 cents in favor of the warehouse in different parts of the country; and if merchandising the product is included in the cost, the warehouse advantage would decline to an average of about .06 cents, (Tr, 33614i2, 3348; RPF 319). While we cannot conclude on the basis of a study of average costs that central warehousing for soft drink products is more efficient than store-door delivery in all cases, neither is such a study indicative of actual costs in individual competitive situations involving different warehouses, bottlers, and package types and sizes, Route delivery may at times be more efficient than central warehousing for the distribution of large-volume containers such as the 64-ounce bottle; at times it may be a less-efficient method of distribution for soft drinks packaged, for example, in cans, In some instances, then, territorial restrictions may tend to rigidify delivery inefficiencies which a bottler free of the restraint may avoid, Of course, the complaint in this matter does not challenge route delivery as a method of distribution under any circumstances, including those in which its efficiency is suspect. We are concerned only with the practice of restricting territories, a secondary, indirect effect of which is to inhibit the bottlers from freely competing with respect to the delivery services they may offer depending upon the competitive situation and their own assessment of how best tc respond to it.(See Tr. 2786--87 Compare Tr. 3497-98 with Tr, 3532), 26 It has been suggested that the store-door method of product delivery is inconvenient for some of the bottlers customers, See TQTTUlC,Inc. v. Coca-Cola Co"1976-2 Trade Cases, '60 988 at 69 3814!2, However, there is more at stak( here than the convenience of some customers, Ins. v. Aluminum Co, of America 148 F.2d416 (2nd Cir. 1945), Chiej (Continued . . , THE COCA-COLA CO., ET AL. 625 0,­ 517 Opinion delivery, territorial restrictions impede the development of an important aspect of competition in the types of delivery services bottlers would offer to their customers in advantageous competitive situations.

Territorial Restrictions Deprive Retailers and Consul'iers of the Bemfits of Open lntrabrand Competition Complaint counsel introduced into the record as part of their case-in­ chief evidence which shows that the bottlers are not always able to adapt to changing economic conditions and improved technology in marketing and production to achieve efficiencies, especially if their initiatives are inconsistent with respondents' territorial policy. At the same time, the evidence shows that the restriction prevents any intrabrand (27) competition, including price competition28, in the sale of (28) Coca-Cola and allied products in bottles and cans. As a consequence, respondents' restrictions are, as alleged in the complaint depriving retailers and consumers of the opportunity to purchase Coca- Cola and the allied products in bottles and cans in unrestricted markets at openly competitive prices. Moreover, these restrictions have repressed the freedom of independent bottlers to expand their businesses or to seize opportunities they may perceive to increase their Judge Hand, commenting on the purposes of the Sherman Act, noted: that the spur of constant (competitive) stress is necessary to counteract the inevitable disposition to let well enough alone. . . that competitors, versed in the craft as no consumer can be, will be quick to detect opportunities for savings and new shifts in production, and be eager to profit by them, As the record in this proceeding indicates, at times respondents' territorial restrictions may necessitate a more costly and less competitive method of delivery than those which may evolve in an open market. 27 AB we observed indoors in a competitive free enterprise system, the decision to exploit the advantages of route delivery or central warehouse delivery:

. ., should be leftto the free, unimpeded play of market forces and the respective, independently exercised judgments of the relevant units of distribution. (at 202). 28 That the restraint has severe adverseeffects on price competition is abundantly demonstrated in the testimony of James Wimberly, Vice President of Coca-Cola U. A. In response to Judge Dufresne s questions, he testified that: . .. the experiences that I recall, sir, would only result when maybe one bottler raised his price and an adjoining bottler did not at that point in time, and customers or dealers would try to bring Coca-Cola from one territory to the other.

Judge Dufresne: The fellow who raised his price reported to you? The Witness: Yes, sir, sometimes, that is right, they did, Judge Dufresne: And what did you do about it? The Witness: I generally said two things: One is if we do anything about it we have got to be sure that it occurred, and that we are dealing with facts; and, secondly, on some occasions I went to the bottler in whose territory it was reported the merchandise was coming from to try to get him to talk to their dealers or salesman to persuade them not to do that.

Judge Dufresne: Suppose he says, Mr, Wimberly, I don t care what you say. I am going to sell this Coca-Cola to anybody who comes to my door and says, I don t want to pay Charley s prices in the next territory? The Witness: Well, yes, sir, but you see, most of the time the bottler who allegedly was purportedly involved in that, that one whose territory that the Coca-Cola was coming from; in most instances he was eager not to continue that sort of practice either because he felt pretty sure if he did, the other Coca-Cola bottler was going to try to do the same thing in his territory, and it would just lead to Judge Dufresne: He could be persuaded to discontinue it. Is that what you are saying? The Witness: In most instances they realized that that would lead to great trouble and biokering and fighting between them and were pretty anxious to discontinue -. (Tr, 887-89), .

Opinion 91 F.

output of Coca-Cola and the allied products by selling these products where and to whom they choose in markets governed by natural economic forces.

IV. CONSIDERATION OF RESPONDENTS ARGUMENTS SUPPORTING THE EUMINA non OF INTRABRAND COMPETITION In concluding that the type of transaction sale or consignment, a manufacturer uses to distribute a product "is not sufficient to justify the application of a per se rule in one situation and a rule of reason in the other " the GTE court noted that post-sale vertical restrictions may not always be without redeeming virtues. For example, the Court pointed out that vertical restrictions may promote interbrand competi­ tion by inducing capital investment and promotional and service activities by the supplier s customers, by increasing marketing efficiency, and by improving quality control. (See GTE at 71 900--901). While the Supreme Court did not indicate that lower courts should afford such inducements and efficiency factors dispositive weight, its opinion clearly makes the consideration of these issues relevant in determining whether the restraints are reasonable. A. CAPITAL INVESTMENT Respondents contend that territorial restrictions promote the business purposes of The Coca-Cola Company because the soft drink industry is capital intensive and the restraint creates a climate conducive to capital investment. While it is true, as respondents contend, that exclusive territories provide bottlers with a measure of certainty with respect to their ability to recover their investments (RPF 73), we are unable to conclude, on this record, that a free market would otherwise render the bottlers incapable of operating at a profit. The fact that the risks which attend a bottler s efforts to recover his investment would increase without territorial intrabrand monopoly protection is simply a corollary to the conclusion that as competition intensifies, business risks of capital recovery increase to the entrepren­ eur. While capital investment considerations, as we have previously noted, may justify a territorial restriction imposed by a new entrant or a failing or faltering firm, we do not, in (29) applying Section 5 ordinarily distinguish between capital-intensive and less capital-inten­ sive businesses by applying different antitrust standards to them license to restrain trade because it promotesgranting the former capital investment while mandating, in the case of the latter, that competition should be preserved. (Compare To'YYWA: , supra at 69 381). In competitive markets, prices may be expected to reflect the capital , THE COCA-COLA CO., ET AL. 627 517 Opinion requirements of the firms in the industry in addition to providing entrepreneurs a fair return on their investments. Shielded by artificial trade barriers created by The Coca-Cola Co. bottlers may well feel secure in making investments which might seem unwise to them if their decisions were being fashioned by free market demands; but this is further evidence of the significant degree to which competition may be lessened by these restraints. Here territorial restrictions are not serving the interests of competition in aid of an aspiring new entrant or a failing or faltering firm which cannot otherwise find investors to put up the distribution capital necessary to market its product. In this instance, the restraint is reducing the entrepreneurial risk of investment by lessening competition among the firms which wholesale one of the most (30) popular consumer product lines in American industry. (Tr. 685).30 While intrabrand competition may reduce the profit in bottling Coca-Cola and allied products respondents' failed to establish that these prized trademarks and premium products would not still remain viable interbrand competitive factors in an open, unrestricted marketplace. As such we cannot sanction anti competitive conduct for the purpose of allowing respon­ dents' bottlers to continue, in perpetuity, to make capital-investment decisions in response to the distorted economic forces within their exclusive territories.

B. A V AILABILITY AND MARKET PENETRA non By using route delivery in exclusive territories, the bottlers have maximized their market penetration and the availability of Coca-Cola putting it in every conceivable location a soft drink may be sold and placing it within "arm s-reach of desire." (Tr. 696). Numerous bottlers testified that deep market penetration and product availability are crucial to selling soft drinks in bottles and cans successfully. (RPF 77). This marketing philosophy has led the bottlers to service large 29 Evidence of the profit bottlers realize on the sale of Coca-Cola or the allied products is in a state of disarray. To begin with, profit is variously described by different bottlers as dividends on book value or as a return (1) on sales, (2) on book value, (3) on investment, (4) on invested capital (5) after taxes on the replacement value of investment, (6) on the market value of investment, and (7) on equity, For all that appears in the record, each bottler may calculate profits on a different basis, Moreover, seven of the ten witnesses, relied upon by respondents in support of their contention that the profit levels of their bottlers are reasonable (RPF 266), piggyback brands other than Coca-Cola and the allied products. Profit on the sale of Coca-Cola products by these bottlers is not indicated, There is, as a consequence, little basis for a comparison of the profitability of Coca-Cola bottling with other industries; nor do we find support for the conclusion that the return obtained by bottlers on the sale of Coca-Cola and allied products is not "abnormal" when compared with other industries, (RPF 265). Furthennore, the profitability of respondents is not reflected in the record. If respondent were a new entrant or a failing or faltering finn, profitability might be a relevant consideration in assessing these restrictions, However, we find it difficult to justify the restraint, in this instance, as a means of improving respondents' profits or those of its bottlers. 30 Evidence in the record indicates that Coca-Cola bottlers are firmly entrenched in the fabric of the bottling industry and that their Coca-Cola brand alone is often a leading brand in their territories, This is reflected in a stipulated survey of 36 cities, from Maine to California, in which Coca-Cola, as a single brand, consistently ranked among the top four brands in each city. (RX-2Y -Z38, See Tr, 2691). . . . ....

Opinion 91 F.

numbers of, vending machine accounts, small outlets, and "special events" which they claim are unprofitable. (RPF 83). Many of the bottlers who engage in these types of unprofitable activities do so, they say, to obtain "paid sampling" of their products "t~ get the product awareness to make the larger accounts profitable. It is a matter of developing a market, training people to drink Coca-Cola. " (Tr. 1454). In the opinion of the President of The Coca-Cola Company, territorial restrictions encourage this type of market penetration because "(t)he fellow who has a limited field to till obviously has to till it better in order to get the most out of it." (Tr. 696, RPF 84). The record does not indicate whether The Coca-Cola Company consistently sells syrup unprofitably to some of its bottlers as its bottlers sell unprofitably to a large number of accounts presumably to create a demand for Coca-Cola; but it would not be second-guessing the bottlers' business judgments to observe that The Coca-Cola Company may be "free riding" on the volume generated by its independent bottlers' give-aways and unprofitable sales. The Coca- Cola Company, in selling the syrup and (31) concentrate soft drink ingredients to its bottlers, profits by the expanded sales universe of its bottlers, even if that universe includes accounts which are unprofitable to the bottlers.

At the same time, a bottler typically charges a uniform or level price to all of his customers irrespective of the fact that price differences between customers may be justified on the basis of different delivery costs the bottler incurs in serving each outlet. Consequently, some accounts which may cost the bottler less to service probably contribute 31 The Court inGTE noted that vertical restrictions may increase economic efficiency because the manufacturer desires to minimize his cost of distribution and to encourage dealers to sell at "the lowest retail price possible because a lower retail price means increased sales and higher manufacturer revenues," 1977-1 Trade Cas. Para, 61 488 at 71 901, n. 24, citing Note, 88 Harv. L. Rev.636, 641 (1975), The trademark license to bottle and sell Coca-Cola contains a fixed syrup price which can change only in accordance with a formula tied to the price of sugar. (CX 9A- , CX 11A- , CX 13A-B). Hence, The Coca-Cola Company can raise its syrup price vis-a-vis bottlers only when sugar prices rise, The fixed syrup price means that The Coca-Cola Company cannot profit from higher prices charged by bottlers for Coca-Cola, Only more syrup volume produces more profit, (Supplemental Br, of Intervenors Coca-Cola Bottling Co, of L.A" et at. at 6). It is not possible on this record to state definitely whether the bottlers' market penetration in exclusive territories generates greater syrup and concentrate volume and profit for The Coca-Cola Company than would intrabrand competition among bottlers but the latter probably would, in many instances, result in lower wholesale prices for the finished soft drinks. It should be noted that several witnesses testified that while wholesale prices in open markets would probably be lower for some ' customers, they might be higher for other customers. As we notedBaisein Cascade Curp. Dkt. 8958 issued January 11 , 1978 (91 F, C, 1 J:

By "lower" (prices) we do not mean simply lower for all customers. Elimination of restraints of trade may result inraising prices to some purchasers (perhaps those whom it is costlier to supply) while lowering them to others, In a freight intensive industry the reallocation might occur roughly along lines of relative actual freight costs. (at 6 fn. 4).

The testimony in this proceeding reflecting the likelihood that prices might rise for some customers and be lower for others absent the restraint is consistent with our observationBaise inabout the workings of a competitive market. In this instance, the free market would be likely to provide those retailers who are efficient not only the opportunity to buy Coca-Cola and allied products from competing bottlers at prices which more accurately reflect 'costs but also the option to pass on to consumers the benefits of their efficiency. .... .

THE COCA-COLA CO., ET AL. 629 517 Opinion a disproportionately higher share of the overall cost of the bottler market penetration. (Tr. 4043). And eventually, those retailers who may be paying or "subsidizing" part of the costs associated with deliveries to other retailers will pass on to consumers, in the form of higher prices, any added cost they may be absorbing. (Tr. 4042). (32) We acknowledge that the elimination of exclusive territories may force the bottlers to abandon their level pricing policies and begin to charge prices which reflect the actual cost differences in servicing various retailers. A bottler who elects to compete for accounts in neighboring territories or who is forced to defend against the forays of intraband rivals which seek the business of his previously captive outlets will no doubt lose the leverage of intrabrand monopoly to extract a price from some retailers which reflects the cost of market penetration to other retailers. Consequently, if the degree of market penetration respondents now enjoy fails to reflect actual costs of servicing each customer, it is likely that some adjustments will be necessary: level pricing may give way to pricing which more closely approximates costs, or bottlers may establish a minimum volume which they will deliver to customers, or they may encourage plant pick-up by customers who cannot be serviced efficiently. But the marketplace would benefit from the increased competition, and we cannot conclude that respondents' interests in maintaining the status quo supercedes this consideration.

C. ADVERTISING AT THE LOCAL LEVEL Respondents also contend that a bottler s interest in advertising and promoting Coca-Cola at the local level will subside if another bottler selling the same brand can take advantage of his efforts. Exclusive territories prevent this type of "free riding," and thus encourage Coca- Cola s promotion at the local level.

Recently, the court in GTE noted that the extent to which vertical restraints on intrabrand competition alleviate market imperfections such as the "free rider" effect and promote interbrand competition may be a relevant consideration in assessing the reasonableness of a vertical restriction. Further guidance on this issue was provided in Bates v. State of Arizona 433 U. S. 350, 1977-2 Trade Cases, ~ 61 573. In Bates the court observed that where consumers have the benefit price advertising, retail prices often are dramatically lower than they would be without advertising. " (at 72 330). The court further noted in Bates that advertising may facilitate entry by a newcomer seeking to penetrate the market. (at 72 331). Under certain circumstances, price advertising, brand enhancement or image advertising by a new entrant, for example, and advertising which informs consumers about .

Opinion 91 F. C-:'~ distinct product attributes may, to a greater or lesser (33) degree enhance the competitive vigor of a market.32 In this instance, however the burden of the restraint exceeds the benefits of the advertising it is said to encourage. After 75 years of advertising by respondents and the bottlers, the record clearly shows that it is intrabrand competition which is likely to produce the pressure necessary to reduce the wholesale price of Coca-Cola. 33 (See text at 51-54 infra). Unlike GTE and Bates which involved advertising by those who offered goods or services to ultimate consumers, respondents' bottlers usually sell their products to retailers. As wholesalers, the bottlers admittedly have no control over retail prices charged by their customers. In contrast, Sylvania s retail dealers, like the lawyers in Arizona in the Bates case, advertise prices to their immediate customers. (34) Between the bottlers and consumers, however, an additional independent retail level of distribution usually intervenes. (Compare RPF 218 with Tr. 2496). Consequently, bottlers may only suggest retail prices, and while this may indirectly influence the retailers' pricing decisions, we do not consider suggested price advertising a substitute for intrabrand competition at the wholesale level which results in lower wholesale prices and, in turn, competition among retailers which results in lower retail prices. (Compare RX 5 RX 56, 58- and RX 101 (advertising by bottlers, Tr. 1982-83 2493 2497 , 3031) with RX 57A-57Z, (advertising by retailers, Tr. 2496). As the court observed in Bates. advertising is the traditional mechanism in a free-market economy for supplier to inform a potential purchaser of the availability and terms of the exchange. " (at 72 330). The record, in this instance, leaves little doubt that the bottlers would have every incentive to price promote their products in competition with intrabrand bottlers and to convey information relating to the terms of sale or the competitive packaging or service alternatives they 32 It has been argued that territorial restrictions cure the "free rider" problem. and thereby promote advertising and merchandising efficiencies at the local level. It is not inconceivable,however, that the pressure of intrabrand competition might encourage bottlers to increase their overall efficiency by cutting costs associated with advertising and merchandising beyond that which the free market might demand.(See Bates, supra at 72.331, fn. 35). 33 In this instance, we recognize that intrabrand competition may well have an effect on the types of merchandising and advertising a bottler may elect to provide to his customers in response to the types of merchandising efforts customers and consumers demand from the bottler. Presumably some customers would elect to purchase from a bottler offering lower prices and fewer merchandising services if a choice between lower price or increased merchandising were available, Conversely, in exclusive territories a bottler arguably gains a "free ride" on consumers who may end up paying for any excessive advertising, merchandising, or local sales efforts which would be . discouraged in favor of price competition. Ai! the Supreme Court observed inNartlwrn Pacific Ry.Co, v. U.s., supra, the antitrust laws rest:

on the premise that the unrestrained interaction of competitive forces will yield the best allocation of our economic resources, the lowest prices, the highest quality and the greatest material progress, while at the same time providing an environment conducive to the preservation of our democratic political and social institutions. But even were the premise open to question, the policy unequivocally laid down by the Act is competition. Thus, the potential efficiency-creating aspects of a practice which substantially diminishes competition cannot, alleged efficiency grounds alone, always be justified under Section 5 of the Federal Trade Commission Act. . . . . THE COCA-COLA CO., ET AL. 631 517 Opinion may offer to their potential purchasers, the soft drink retailer. (See , Tr. 2500). In fact, the amount of such information received by the potential customers of competing intrabrand bottlers would probably increase. The free rider problem is not likely, for example, to prevent a Coca-Cola bottler from advertising to retailers that his price is lower than that of his intrabrand competitors or that he offers them various delivery options or credit terms. Nor can we conceive of any reason why retailers who purchase their soft drink supplies from competing intrabrand bottlers would lack the incentive to' convey similar (35)information to their customers, the ultimate consumer.34 Although bottlers may be reluctant, absent territorial exclusivity, to engage in brand enhancement or image advertising which may be especially susceptible to same-brand free riders, it is highly unlikely that consumer recognition of the Coca-Cola and allied produce brands would fade appreciably as a consequence. The Coca-Cola Company, an established giant in the industry, has not shown itself to be in need of financial assistance to promote these brands. Unlike the situation in Sandura Co. v. 339 F.2d 847, 854, 858 (6th Cir., 1964), this record does not show that The Coca-Cola Company must depend upon its bottlers for funds to sponsor national, regional, and local, level advertising. This, then, is not a case in which the restraint is promoting interbrand competition by aiding a new firm to enter the soft drink industry or by helping to forestall the exit of a failing or faltering firm as in Sandura. According to its President, Coca-Cola: " . is the most widely recognized name in American commerce and indeed in world commerce . it has huge value." (Tr. 685). We conclude, in this instance, that advertising-related considerations which may justify the restraint in the interest of fostering interbrand competition by new or faltering firms do not apply here.

D. QUALITY CONTROL Respondents contend that territorial restrictions promote product quality in essentially two ways. Because a bottler has a limited geographic area, respondents submit that he cannot afford to risk 34 Of course, advertising by bottlers may sometimes convey information useful to consumers. For example, an ad in the "San Antonio Light" sponsored by the San Antonio, Texas, Coca-Cola bottler on December 3, 1970, which discussed the merits of refillable bottles may have been useful to some consumers. (RX Compare Tr. 2915). It is in this regard relevant, however, that consumer organizations have filed a brief as amici urging the Commission to lift the territorial restriction despite the advertising "free rider" problem.(See Bates, :rupra, at 72 331 fn. 35). Moreover, advertising is discretionary even by a bottler protected by territorial restrictions. He may choose not to advertise at all or he may direct his advertising to product attributes with which consumers are generally familiar (Tr. 2915) or that consumers may learn about, without incurring significant search costs, through advertising by The Coca-Cola Co. or by retailers or from other sources in the marketplace. Thus the value to consumers of advertising by the bottlers is highly speculative. We are, therefore, unable to conclude that advertising by bottlers which may on occasion convey information potentially useful to comparative shoppers(see RX 60; Tr, 2499;see also RX 56, 61 and 101), out-weighs the sacr;fice of intrabrand competition, in perpetuity. .,..

Opinion 91 F.

losing customers who become dissatified with the quality of his product. The restrictions presumably induce bottlers to manufacture a high-quality product and then to ensure that it is subsequently stored and merchandised in a way which prevents the buildup of stale inventory at retail outlets. (RPF 126, Tr. 762, 699). Respondents also contend that the restrictions enable them to monitor, at the retail level the quality of the product produced by each bottler. In Manufacturing To ensure that bottlers are properly preparing the finished soft drink products, The Coca-Cola Company has a Quality Control Department which inspects, on an average of three to four times a year, every bottling and canning facility which manufactures Coca- Cola and allied products. Its inspections are generally unscheduled and unannounced and include water analysis, bacterio-Iogical checks on water and processing equipment, bottle washing solution checks sanitation monitoring, and finished product syrup content and degree of carbonation. In addition, each production facility is required monthly to submit product samples for analysis by respondent Coca- Cola s quality control lab. (RPF 121). In this way, The Coca-Cola Company frequently and routinely monitors the bottlers' manufactur­ ing process to ensure that they are producing soft drinks in accordance with its standards of quality. (See Tr. 2669). Contrary to respondents' contentions, however, there is really no connection between these types of quality control inspections and the areas where independent bottlers sell the finished product; plant facilities can be inspected regardless of where the product is eventually sold. (36) Furthermore, even though the bottlers presently operate within exclusive territories, the possibility is ever present that a bottler despite his intrabrand monopoly, may be tempted for short-term profits or other reasons to cut corners by, for example, reducing the amount of syrup he mixes with carbonated water to produce the finished product, thereby reducing its quality.35 Recognizing this 35 Respondents' quality control inspectors also spot-check their bottlers by obtaining, for analysis, products which they purchase directly from retailers. (Tr, 921 974-75). Because the bottlers presently need not identify themselves on their packages, successful spot checking now depends, in large measure, on the assumption, vl1olidated by respondents' territorial policies, that the soft drinks found on the retail shelves within a given territory were sold to the retailer by the bottler in that territory. To this extent, territorial restrictions facilitate product-source identification. Yet the issue of whether territorial restrictions could ever be justified on the ground that they indirectly encourage quality control by assisting respondents' monitoring efforts need not be decided here. Rather, we find that respondent could as easily continue spot-check, quality control inspections at the retail level by requiring each bottler to place an identification mark on his product. Obviously, neither marking requirements nor territorial restrictions provide fool-proof safeguards against the production and distribution of defective products (RPF 125) or the midnight" batch of substandard products which a bottler could presumably intentionally produce. A cheating bottler could be difficult to trace under either monitoring mechanism, yet a simple product-source identification mark (Tr, 804) like dating codes which bottlers may now employ (Tr, 1116-1119) would allow respondents to determine product (Continued) ....

THE COCA-COLA CO., ET AL. 633 517 Opinion respondents have provided the bottlers with the added business incentive not to produce substandard soft drinks. They have condi­ tioned each bottler s right to continue to produce the trademarked product upon his faithful adherence to their quality standards. (Tr. 911-'-12). Thus a bottler s failure to meet respondents' standards of quality may result in the cancellation of his trademark license. (Tr. 778): At the manufacturing level, then, unscheduled plant inspections and frequent product sampling, coupled with the threat of termina­ tion, if not the act itself, should provide a strong deterrent to the bottler who might be inclined to cheat on quality, notwithstanding the markets in which he may ultimately distribute the finished products. (37) 2. In Distribution Respondents further point out that their trademarks appear on the finished products which reach consumers, so their interest in maintain­ ing product quality extends to the retail level. Respondents note that the bottlers assist in their overall quality control effort by offering stock rotation services to retailers and by removing the old product which may have deteriorated on the retail shelves. The Commission recognizes the interest of a supplier in maintaining the quality of a trademarked product in the channels of distribution through which it travels to the marketplace. In Coors we considered the needs of a brewer who sought to impose customer and territorial restrictions upon its distributors in order to ensure among other things, that its beer remained refrigerated in storage and distribution from the brewery to the consumer. We pointed out, however, that a supplier of a trademarked product may have available to it means less anticompetitive than territorial or customer restrictions to ensure a reasonable measure of quality control at each level in the chain of distribution. Coors beer was brewed by a unique process and required continuous refrigeration. Thus our order in that case permitted the brewer to establish refrigeration standards not only for its own distributors, but downstream for the distributors' customers. The brewer was permitted to hold distributors responsible for inventory rotation by central warehouse customers and at the retail delivery locations where the beer was received from the central warehollse. Having considered respondents' quality control objectives, we feel that the underlying rationale of our decision in Coors is clearly applicable. While the finished soft drinks need not be distributed origin as reliably as territorial restrictions. Further, if respondents employed a container dating which would be read without resort to a code, retailers and consumers would be able to monitor and detect the product' s age. .

Opinion 91 F.

through refrigerated channels, the shelf-life of these products is not indefinite; over time, the process of oxidation can sour the taste of these beverages. (Tr. 698, 978-79).36 Respondents may, however establish reasonable (38) quality control standards for distribution and storage, including inventory rotation policies, and may further require that each bottler identify itself on the bottle, bottle cap, or on the can so that respondents may reasonably monitor compliance with its quality standards. Clearly, quality control and intrabrand competition are not incompatible.

Under these circumstances, we are unable to conclude that territorial restrictions are reasonably necessary to ensure the taste uniformity or the purity of these products; quality control, trademark protection considerations do not, in this instance, justify the restraint imposed on the sale of the bottlers' finished soft drink products. V. INTERBRAND COMPETITION Buttressed by the judge s finding that the "corridor area" exhibits intense" interbrand soft drink price competition, respondents argue that their restraints on intrabrand competition are ' reasonable. The judge concluded that the prices which bottlers charge for Coca-Cola and allied products are determined by their costs and interbrand competition (IDF 106) and that bottlers cannot price Coca-Cola and allied products above the prices of other brands, such as Pepsi-Cola and Up, without losing sales. (IDF 108-09). He also found that the bottlers of Coca-Cola frequently offer price promotions (IDF 127-30) and that a restriction on intrabrand competition is procompetitive because it allows the bottlers to focus on interbrand rivals, therebyincreasing interbrand competition. The record shows that Coca-Cola and the allied products compete with a wide variety of beverages. Evidence was adduced at the trial from which a list was compiled of the brand or trade names of products which, to one degree or another, compete with Coca-Cola; the list of brands is lengthy and will not be repeated here. (See RFP 157-80). In summary, it includes the names of hundreds of national, regional, and local flavored carbonated soft drink brands; private label soft drinks the bulk of which are produced by contract canners for food chains and other types of chain stores; powdered mixes such as Kool Aid, Funny Face, and Wylers; and noncarbonated drinks, including such brands as Hawaiian Punch, Gatorade, and fruit juices and drinks. The Coca-Cola 36 Estimates of this time span are variously given for bottled products as two to four weeks (Tr. 111&-17), three to eight weeks (Tr, 1240-41), 60 days (Tr, 1632), a few weeks (Tr. 979), a month or so (Tr. 2087), and several weeks for cans, if stored in a cool, dark place. (Tr. 1381), In addition, the shelf-life of canned products depends on whether the cans are made of steel or aluminum, Aluminum cans apparently retain taste quality a little longer than steel cans. (Tr. 1ll&-17, 1239, 1343-44 2300), THE COCA-COLA CO., ET AL. 635 517 Opinion bottlers who testified in this proceeding agreed that all such products 4p.- ,­ compete, at least to some degree, with Coca-Cola in bottles and cans. However, their testimony clearly demonstrates that flavored carbonat­ ed soft drinks generally and the brands, such as Pepsi-Cola, distributed by (39) other bottlers are the Coca-Cola bottlers' primary competitive rivals. As the record in its entirety amply demonstrates, the suppliers of these products exert the greatest influence on their competitive decisions. (See Tr. 1324-25; 1533; 3243). Consequently, we will focus mainly on the products which the bottlers have identified as their most important interbrand competition. Presumably, this is where the intensity" of interbrand competition would be most evident. A. FLAVORED CARBONATED SOFT DRINK BRAND COMPETITION The judge found that there is intense competition in the sale of flavored carbonated soft drinks "which stems from the fact that there is a large number of brancls available to the consumer in local markets." (ID 36). As impressive as the number of brands on respondents' list may be, however, it is, in itself, no measure of the intensity of the competitive interaction among the brands or the bottlers or canners which supply them. Indeed, the judge s consider­ ation of interbrand competition at the finished soft drink production and distribution level glosses over the customary practice of major brand bottlers to carry the brands of several different syrup companies, a practice which they refer to as "piggybacking." Nor does the initial decision reflect any analysis of the anticompetitive interbrand effects of geographic market restraints which admittedly permeate the entire industry.37 We believe that an accurate assess­ 37 The judge found that intense interbrand competition was evidenced by data showing a decline in Coca-Cola food store market share during the period 1950 through 1971. In reaching this conclusion, the judge relied on two series of data stipulated by counsel and offered into evidence by respondents. In IDF 163, the judge found that Coca-Cola brand unit sales declined from 41.2 percent of total domestic flavored carbonated soft drink food store sales in 1950 to 24.4 percent in 1965. Unit sales, however, do not take into account the fact that soft drinks are packaged in containers of different sizes; it reflects only the number of bottles and cans sold, not liquid volume. (RX 2B). The record shows that prior to 1955, bottled Coca-Cola was available in only one size, the 6 1/2-0z. bottle. (RPF 253). In subsequent years new sizes were introduced ranging from 6 1/2 ounces to 64 ounces, Yet on a unit basis, one 32-0z. bottle is the equivalent of one 6 1/2-0z. bottle, although it contains nearly five times as much beverage. Under these circumstances the comparison of unit sales data before and after 1955 in IDF 163 is meaningless, The judge also compared statistical case sales of Coca-Cola between 1960 and 1971 and noted that it declined from 22,3 percent of food store sales to 20,8 percent. (IDF 164), The statistical case represents a conversion of the sale of soft drink cases containing all package sizes to the equivalent of 24 8-oz, containers or 192 fluid ounces. As reflected in the stipulation, the "Coca-Cola" sales trend, on a statistical case basis, is as follows: 1955 20, 1960 22, 1965 19, 1970 21. 1971 20. RX 2-Z44 However, the stipulation also shows that both total flavored carbonated soft drink food store sales and Coca-Cola brand food store sales increased rapidly during this period. Food store sales in 1955 exceeded 495 million cases. In the (Continued) Opinion 91 F.

40-' ment of the condition of interbrand competition in this industry, that , its "intensity" or "degree" as reflected in the record, must take these factors into consideration. (40) Effect of Piggybacking on Interbrand Co'mpetitWn at th€ Bottling Level Piggybacking involves the production and sale by a bottler of soft drink brands trademarked by two or more syrup companies. Each syrup company generally grants the bottler an exclusive territory for its brands. In piggybacking situations involving Coca-Cola bottlers, the territories are not always coextensive (41) in size or dimension with their Coca-Cola territories, but they usually overlap to a substantial degree.

The Coca-Cola Company argues that the brands piggybacked by its bottlers evidence interbrand competition. Some insight into respon­ dents' rationale for concluding that competition is intense among a bottler s piggybacked brands was provided by the President of The Coca-Cola Company. According to Mr. Smith: (42) (W)hen a product is put on the (retail) shelf the consumer is often unaware of its source. . .so what I am saying is when the consumer is shopping on a shelf or looking same year, Coca-Cola brand sales exceeded 98 million cases. By 1971 food store sales topped 1,573 million cases while Coca-Cola brand sales reached nearly 328 million cases, (RX 2-Z44). Evidence of the meteoric rise in the volume of soft drinks sold during this period and Coca-Cola s relatively stable portion of this sizable volume, particularly since the early 19608 when diet soft drinks emerged as a strong factor in the market (Stipulation No. , ex 1244H-I), contradicts the contention that interbrand competition has significantly eroded Coca-Cola s position in the market.

In addition, we note that food store sales data relied upon by the judge fails to reflect Coca-Cola sales in a large number of non-food store outlets (RPF 221- , RX 2Z41-42), and thus probably understates the brand's true strength. (See Tr, 2324), For this reason, the data cannot provide an accurate indication of either the Coca-Cola Company s soft drink syrup and concentrate sales to bottlers or the bottler s sales of finished, flavored carbonated soft drink sales in any local market. Moreover, even within the limited universe relied upon in the initial decision, the fact is ignored that the Coca-Cola Company s allied product lines, including Tab, Sprite, Fresca and Mr. PiBB, which were introduced in the 19608, had by 1971 captured about 4 percent more of total food store sales. (RPF 273). Had these brands been included, it is evidence that the share of this universe attributable to the Coca-Cola Company s brands did not decline; rather it increased from about 20 percent in 1955 to about 24.8 percent of food store sales in 1971. Similar distortions are noted in the judge s analysis of the Pepsi-Cola brand' s 1971 estimated market share of approximately 19.3 percent, This analysis also ignores "Pepsi" sales in non-food store outlets and Pepsico s sale of such allied products as Diet Pepsi, Patio flavors, and Mountain Dew.

38 Piggybacking is used extensively in the soft drink bottling industry, The record shows that in 1971 , 438 of the 726 domestic Coca-Cola bottlers also distributed at least one soft drink brand not licensed by respondents. (Tr. 689). As a consequence, important national brand soft drinks, such as Dr. Pepper or 7-Up, are in some territories produced and sold exclusively by the local Coca-Cola bottler. Similarly, Nestea, canned ice tea, is sold under a territorial licensing system by 135 national brand bottlers, including 55-60 bottlers of Pepsi-Cola, 45-50 bottlers of 7-Up, and 30 bottlers of Coca-Cola. (RPF 262-63). In New York City, for example, where Coca-Cola is the leading flavored carbonated soft drink brand with a 14 percent market share in 1973, the Coca-Cola bottler sells several allied products and piggybacks both Welch's Sparkling Grape Soda and Dr. Pepper, Other examples include the Reading Coca-Cola bottler who piggybacks Pennsylvania Dutch Birch Beer and Bottoms Up Chocolate (Tr. 1888-89); the Jamestown, N. , Coca-Cola bottler who piggybacks 7-Up, Nesbitts Orange, Dads Root Beer, Squirt, and Sunrise Flavors (Tr. 1957); the Coatesville, Pa" Coca-Cola bottler who piggybacks Dr, Pepper and Pennsylvania Dutch Birch Beer (Tr, 2173); the Herminie, Pa" 7-Up bottler who piggybacks RC Colaj and the Dyersburg, Tn., Pepsi bottler who piggybacks Bubble-Up and Dr. Pepper, (See Tr, 961, 1443-45, 1600-01, 2809-10 2863, 3005-m, 3063), ....

THE COCA-COLA CO., ET AL. 637 517 Opinion through a vending machine to pick a product, any product, (it) is in competition with any other product there in my opinion. (Tr. 781-82). Were we to concern ourselves only with the image of competition which a lengthy recitation of brand names may project, the consumer imperfect knowledge about who it is that actually supplies the brands might be superficially persuasive, and the effect of piggybacking might be safely ignored.

To the extent these brands represent the sale of syrups and concentrates by. competing syrup companies we acknowledge that they are a factor in the sale of soft drink ingredients to bottlers: (Tr. Oral Argument July 28, 1976 at 71-73). By contrast, however, in the sale of finished soft drink products to retailers, piggybacking allows a Coca-Cola bottler to control the pricing and marketing strategies for each piggybacked brand. (Tr. 1820-23). Thus he may determine unilaterally the extent to which pricing policies respecting one of these brands will be permitted to "cannibalize" sales of his other brands. (Tr. 2007-08). Consequently, if a competing bottler undercuts Coca-Cola and thereby cuts into Coca-Cola sales, the Coca-Cola bottler s only defense may be a responsive price cut. In contrast, if a Coca-Cola bottler who piggybacks Dr. Pepper finds that his price on a Dr. Pepper promotion is cutting too deeply into his Coca-Cola sales, he .may find it in his interest to raise the price of Dr. Pepper rather than lower the price of Coca- Cola. (See Tr. 3037-38). Thus, the Coca-Cola bottler in New York City, having assessed the potential strength of Dr. Pepper in New York and having determined that its entry was imminent, became a Dr. Pepper "piggybacker"39 because: "we would rather compete with ourselves than have somebody else compete with us. " (Tr. 2302). (43) N or is it surprising that a bottler would prefer to shadowbox with in-house" brands rather than meet the more rigorous competitive challenge of another bottler. In becoming "self-competitive " the bottlers' objective understandably is to " get more new sales volume from a competitor than from themselves." (Tr. 782 2008). As a result, a willCoca-Cola bottler who piggybacks Dr. Pepper, for example, employ marketing strategies which are designed to take sales away ~rom the brands of other bottlers without losing Coca-Cola volume in 39 It is a policy of The Coca-Cola Company not to license its allied products to bottlers other than Coca-Cola bottlers, (Tr, 675). Consequently, bottlers have at times elected to piggyback certain brands or flavors of another syrup company, knowing that Coca-Cola allied products would not be introduced as competitive brands in their territory, Thus Mr. PiEB, respondent's Pepper-type drink, was not introduced in New York City because the Coca-Cola bottler there elected instead to distribute Dr, Pepper. (Tr. 2301), Had Dr, Pepper entered New York via another bottler, such as the Pepsi bottler, the New York Coca-Cola bottler could have responded by introducing Mr. PiBE. Pursuant to The Coca-Cola Company s policy, however, the New York Coca-Cola bottler not only acquired control of the Dr, Pepper brand, it knew that no other bottler would have access to the competing Pepper-type drink, Mr. PiBE. Similarly, Coca- Cola bottlers who handle their own flavor lines understand that competing Fanta flavors will not be introduced in their territories by any other bottler,(See Tr, 1600-01, 1094-95, 1226, 1666). ....,.

Opinion 91 F.

the process. (Tr. 1558, 2691). He may, for example, prevent price interaction among his piggybacked brands by selling each of his brands at the same price (Tr. 1822-23); by packaging one brand in returnable bottles and another brand in cans or. nonreturnable bottles, thus minimizing head-on package competition between them; or by adopting other strategies depending upon the particular situation. (Id. Tr. 2392-93; See also Tr. 2553). While the record shows that bottlers are not always able, in the short run, to prevent one brand from cannibalizing the volume of another (See Tr. 2354-55), it also shows that the basic marketing strategy of brand proliferation is to increase the bottler s total sales in the long run or protect his other brands from erosion. (Tr. 2385-86, 3037-38, 2008, 2302). Notwithstanding respon­ dents' vigorous protestations about the "intense" interbrand competi­ tion among a bottler s piggybacked brands, their bottlers understand that being self-competitive is not "the real thing. Furthermore, evidence of the potential effect of piggybacking on the structure of the flavored carbonated soft drink bottling industry indicates that the practice tends to increase con~entration. For example, in the territory of the San Antonio, Texas, Coca-Cola bottler a large number of brands are available to the public. The bottler, when asked about interbrand competition, identified Pepsi-Cola, Diet Pepsi and the allied products of Pepsico. ; Royal Crown and its allied products; Dr. Pepper, Diet Dr. Pepper Canada Dry and its allied products; 7-Up; Shasta, Barqs; Nestea; Big Red; Orange Crush and its line of flavors called Matthews Dot; numerous flavor lines offered by other bottlers; and private label house brands of the major chains, such as Handy-Andy, among others. (44) The record also shows, however, that in San Antonio, Pepsi-Cola and its allied products and 7-Up and its allied products are manufactured and distributed by the same bottler who, in addition, offers his own line of flavors. (Tr. 2501). Another bottler manufactures and distri­ butes RC Cola, Diet Rite Cola, and the Nehi Flavor line. (Tr. 2501-02). The Canada Dry bottler, in addition to Canada Dry and its line of ginger ale, sodas, and tonics, also manufactured and distributes Frosty Root Beer, Orange Crush, and the Matthews Dot flavors. (Tr. 2502). Big Red, which respondents cite as a strong regional competitor, is manufactured and distributed by the same bottler who manufactures and distributes Barqs flavors. (Tr. 2503-04).40 (45) Evidence of this 40 Of all the flavored carbonated soft drink brands available in the food stores in San Antonio, Coca-Cola is the market leader. (RX2-Z37), The Coca-Cola bottler testified that his share of the flavored carbonated soft drink market sold through food stores in San Antonio varied anywhere from a low 34 percent to a high of about 40 percent (Tr. 2485--$6, 2532-83) over a period of several years, (Tr. 2533--34). He further estimated that the Pepsi-Cola share varied from 17 percent to about 21 percent; Dr. Pepper from 8-11 percent; RC Cola from 6-8 percent; and Big Red, the strong regional brand, from 9-10 percent of the market. Consequently, if the combined market of the San Antonio Coca-Cola bottler and the Pepsi-Cola bottler had fallen (Continued) THE COCA-COLA CO., ET AL. 689 517 Opinion high level of concentration among the suppliers of the various piggybacked products strongly suggests that the "intensity" of interbrand competition cannot be realistically assessed simply by naming and counting brands available in a market. Indeed, we find much more significant the fact that piggybacking tends to increase the concentration of brands controlled by the strongest bottlers in a territory, while territorial restrictions shield them from the com~ti­ tion of extra-territorial interbrand bottlers. B. TERRITORIAL RESTRICTIONS LESSEN INTERBRAND COMPETITION AMONG SOFT DRINK SUPPLIERS The judge also ignored evidence showing that territorial restrictions which prevent intrabrand competition also tend to lessen interbrand competition. (Tr. 960- , 1879, 1900-1900A). Because the universe potential customers available to a bottler is strictly limited by the boundaries of his territory41 (46) he is prevented from competing against Pepsi-Cola bottlers and other soft drink suppliers interbrand for the business of retailers located within another Coca-Cola bottler territory. (Tr. 887-88). Consequently, the restriction eliminates impor­ tant potential interbrand price competition between a Coca-Cola to their low points at the same time, these two bottlers still controlled 51 percent of the flavored carbonated soft drinks sold through food stores in San Antonio; and this would not reflect their sales through non-food store outlets. In fact, the Coca-Cola bottler alone placed about 8,000 vending machines throughout his territory. (Tr. 2481). Furthennore, according to the testimony of this defense witness, the four leading soft drink bottlers in his territory controlled about 68 percent of the total food store flavored carbonated soft drink sales, including the private label products sold by the chain stores. (Tr. 2533). Despite brand availability at the retail level, the evidence indicates that the San Antonio local bottling industry may be advancing toward fairly tight oligopoly. Similarly, the record shows that in the Albany, N. , territory, Pepsi-Cola, according to the bottler called as witness by respondents, is the leading brand with a flavored carbonated soft drink brand market share of about 21­ percent: Coca-Cola has about 16-17 percent. (Tr. 2935). Thus, these two brands alone account for about 87 percent food store sales in Albany. In addition, however, the Pepsi-Cola bottler also controls Hires Root Beer: Orange Crush; Schwepp s carbonated soft drink line; canned Lipton Tea; and PepsiCo's allied products, including Mountain Dew. (Tr. 2868).

4\ In overrulingSchwinn, the court inGTE made it clear that the degree of intrabrand competition foreclosed by a vertical restriction provided no basis for distinguishing situations in which the Schwinn per Be rule would or would not be applied.GTE, BUpra, at 71,896. Under GTE, however, territorial restrictions must be evaluated by the traditional rule of reason framework of analysis to detennine if they produce a demonstrable effect on competition. Thus the degree of foreclosure is a factor in assessing the overall competitive effect of the restraint. genero.Uy(See Ellman Motors, Inc.v. Chrysler Corp. 1977-2 Trade Cases, '61, 650 at 72,683 567 F.2d 1252). The Coca-Cola Company s territorial restriction is a demonstrably more severe restraint on intrabrand competition than the dealer location clause imposed by GTE. GTE designates the location of its retailer dealer s outlet, but apparently does not limit the area from which a retailer may draw its customers, The territorial restrictions involved here not only limit the area from which bottlers may solicit customers, they eliminate the retailer s option. to do business with the Coca-Cola supplier offering the most competitive deals. Furthennore, as the restrictions are applied by respondents, they limit retailers in reselling the Coca-Cola and allied products: usually the product purchased from a bottler may be resold by a retailer only at outlets located within the territory of the bottler from which it was purchased. Thus, the vertical restraint in GTE' s franchise does not constitute an exclusive territoryat(GTE 71 898); nor does it ensure GTE's retailers freedom from intrabrand competition. Unlike the situationGTE inand Snap-On­ Tools Corp,v. C" 321 F.2d 825 (7th Cir. 1963), in which cutomers were free to buy- in any territory from any dealer, thus leaving open the potential for intrabrand competition among the dealers, respondents' practice mandates exclusive territories and completely eliminates intrabrand competition among the bottlers of Coca-Cola and allied products.

, ....,. Opinion 91 F.

bottler confined to a territof.y and virtually all interbrand suppliers serving customers in areas adjacent to his territories.42 (47) Territorial Restrictions Industrywide Lessen lnwrbrand Competition When this effect is considered in light of the fact th~t respondents' territorial restrictions are nationwide in scope, and in light of the further fact that territorial restrictions are an industrywide practice restricting "Pepsi" bottlers and the bottlers of numerous other major and secondary brands throughout the country (RPF 17), it is difficult to avoid concluding that territorial restrictions, vertically imposed have seriously impaired interbrand competition. Not only are Pepsi- Cola bottlers and other soft drink suppliers shielded by respondents' restriction from the competition of all but one Coca-Cola bottler for the business of virtually any given retail outlet, the industrywide nature of the restraint insulates Coca-Cola bottlers from unimpeded competition of potential inter brand bottlers. Evidence of this insulating effect is reflected in the pricing behavior of respondents' bottlers in territorial overlap situations. Overlaps occur when, for example, the territory of a Coca-Cola bottler encompasses all or a part of the territories of two or more bottlers of a competing brand, such as "Pepsi." Like the Coca-Cola bottlers Pepsi" bottlers are also confined by territorial restrictions which prevent them from competing with each other. In these situations, the Coca-Cola bottler will, in any given segment of this territory, compete with only one of the Pepsi bottlers. For example, the record shows that Warrenton Virginia, is outside the territory of the Washington, D. , Pepsi bottler but within the territory of the Washington Coke bottler. Because the Warrenton Pepsi-Cola bottler has at times charged lower prices than the large Pepsi bottler in the metropolitan Washington area, the Coca- Cola bottler has been forced to respond with lower prices in that part of its territory. As Mr. Wilbert N. Sales, Vice President and General Manager of Washington Coca- Cola Bottling Cd. Inc., testified: A. ... Warrenton is priced well below Alexandria, and the reason for that is competition.

Q. Could you explain what you mean by that? A. Well, basically ip that market it is Pepsi-Cola. Pepsi-Cola out of Charlottesville 42 This is illustrated by pricing data relied upon by respondents. The record shows, for example, that cases of 24 12-ounce cans of both Coca.Cola and Pepsi-Cola have been offered at wholesale for $3 in Baltimore, At the same time a case of Coca-Cola in 12-ounce cans was offered to retailers for $2.90 by a different bottler serving Havre de Grace Maryland. Yet the Coca-Cola bottler who served Havre de Grace through a distribution center was prevented by the territorial restriction from offering or selling canned or bottled Coca-Cola to retailers in the Baltimore territory, thirty miles away (Tr. 2960), in competition with the Baltimore Coca-Cola bottler, intrabrand, and the Baltimore Pepsi bottler, interbrand. (See RPF 192, Tr. 1564), THE COCA-COLA CO., ET AL. 641 517 Opinion Va. They are priced way down. . . . This is the way he operates, and he couldn t care less Orwhether he makes money or not, so there is the problem. (Tr. 1259). (48) In respondents' view, the fact that the Washington Coca-Cola bottler charges lower prices in overlap areas to meet competition from Pepsi bottlers, other than its major Pepsi bottler competitor in the Washington, D. , metropolitan area, is evidence of interbrand price competition rather than the lack of it. (Ans. Br. 88). This is correct to the extent that interbrand competition exists in both instances, out it can also be reasonably inferred from this evidence that interbrand competition between Coca-Cola and Pepsi-Cola may be significantly less "intense" in the W ashingtonmetropolitan area than it is in the Warren ton area. Moreover, the Pepsi bottler serving Warren ton is, as a result of the territorial restrictions imposed by Pepsico, precluded from expanding into Northern Virginia and perhaps Washington whenever higher "Co~e" or "Pepsi" prices prevail in these areas. Consequently, retailers and consumers in a major metropolitan market are deprived of the benefits of an open market in which the Washington Coke bottler probably would have had to meet the interbrand competition of the Warrenton Pepsi supplier in a wider geographic area and, at the very least, would have had to consider this bottler a serious potential inter brand price competitor outside of the Warren ton area, a consideration which Washington metropolitan bottlers may completely disregard. (Tr. ~314). What has occurred between the Warren ton Pepsi bottler and the Washington Coca-Cola bottler is not simply an isolated episode without broader competitive significance. The situation in Warrenton illus­ trates a fundamental limitation on inter brand price competition in the soft drink bottling industry not only in overlap situations, but as a direct result of territorial restrictions nationwide. Recognizing this respondents contend that disparities in the wholesale list prices of bottlers in different territories have no probative value because bottlers use alternative pricing strategies; some bottlers offer lower list prices, other bottlers adopt higher list prices but engage in more (49)frequent promotions. (Ans. Br. 75, 88).44 The fact that different bottlers use different pricing strategies and price levels at wholesale and during promotions in different territories , itself, a strong argument for lifting the restriction in order to allow the various prices and pricing strategies to clash head-on in the 43 Conversely, the restriction would allow the Charlottesville Pepsi Bottler to raise his prices in the portions of his territory located outside of the Warren ton area without regard for a lower price which the Washington Coca-Cola bottler may be charging at the time. (Tr. 1259-60). 44 Obviously, a wholesale price considered "low" by one bottler may be considered high by another bottler, Similarly, what one bottler considers "frequent" promotions may be considered occasional by another, just as a "deep price promotion in one territory may be considered miserly in another. ...,.

Opinion 91 F.

marketplace. We regard the uncertainties created by the confrontation of pricing strategies as the very essence of competition which the present system of territorial restrictions, to a large extent, eliminates. Whether a Coca-Cola bottler s pricing strategy is to compete on the basis of wholesale list prices or price promotions, or both, the fact remains that territorial restrictions rule him out as an actual or potential competitive rival of all soft drink suppliers, intrabrand and interbrand, in every locale beyond the territory assigned to him by respondents. In view of the fact that respondents' bottlers and virtually all other major brand bottlers are similarly restricted, we conclude that the practice is, to a substantial degr~e, adversely affecting interbrand competition at the bottling level of this indus­ (50)try.45 Thus we find reflected in the testimony of virtually every bottler who testified at the hearing and the top management of The Coca-Cola Company the fear that intrabrand competition would, in fact, cause prices to fall. The President of The Coca-Cola Company testified that absent territorial restrictions, there would be price competition at the wholesale level which does not exist under the territorial system. (Tr. 739).46 In the opinion of another experienced executive of The Coca- Cola Company:

I think under this "walls down" thing. . . . No territorial exclusivity, no territorial restrictions, that Coca-Cola and our pther products, or products from other bottlers would find its way into chain stores warehouses. I think. . . that pricing would be more active than it ever had been. (I) think that it 45 Respondents contend that territorial restrictions promote competition at the syrup-producing level because they make possible a means for the lesser-known brands of their syrup company competitors to enter easily into new local markets. (Ans. Br. at 91, IDF 159-162, RPF 262-64). As we have previously noted, Coca-Cola bottlers and other major brand bottlers piggyback the "lesser" brands of other syrup producers within exclusive territories granted to them by those producers. To this extent both piggybacking and exclusive territories may assist the company s entry into the soft drink syrup production industry, However, we are not here dealing with the reasonableness of territorial exclusivity conferred by a small syrup producer of one of the "lesser" brands. Respondents argue further that removal of territorial restrictions on the sale of Coca-Cola and the allied products will generate competitive forces which will result in the demise of many Coca-Cola bottlers and the secondary brands which they piggyback into local markets, (Respondents' " small bottler" arguments are considered, in detail infra). While the record shows that the number of independent Coca-Cola bottlers has, notwithstanding respondents' territorial restrictions, declined significantly since 1968, the record does not indicate the fate, in local markets, of the secondary brands formerly piggybacked by bottlers who have sold their businesses to neighboring bottlers or have consolidated their territories or bottling plants,(See Text at 66, fn, 63 infra), Presumably the secondary brand syrup producers were free to franchise either the bottler which took up the distribution of Coca-Cola within the territory from which its predecessor withdrew or any other independent bottler distributing other brands in the local territory or neighboring territories, (See Tr, 1672, 1684 , 1668). Thus The Coca-Cola Company argued inSulmeyer v. Coca-Cola, supra in defense of a complaint alleging that it was monopolizing the business of the independent Coca-Cola bottlers which it secured for the distribution of its lemon-lime flavored product, Sprite, that the universe of independent bottlers capable of effectively bottling and distributing the lesser-known brands was not limited to those bottlers who market Coca-Cola, After reviewing the evidence adduced at the trial, the Fifth Circuit Court of Appeals agreed with respondents, noting that a secondary brand "could reach the consumers in a given area through a franchise agreement with any independent bottler," (at 850). We are, therefore, unable to accept respondents' contention that the territorial restrictions which they impose on the sale of Coca-Cola and allied products are necessary to ensure the competitive viability of syrup companies which compete with respondents, 46 See IDF 171.

...., THE COCA-COLA CO., ET AL. 643 517 Opinion would mean that. . . to be competitive, and to get the business, we would have to make up our minds either we want the business or don t want it. We would be forced to reduce our prices to the principal customers. (Tr. 992-93). This assessment, by key management personnel of The Coca-Cola Company, was echoed by bottlers who predicted wholesale price reductions if the restrictions were lifted. (Tr. 1568, 2459, 2855). Yet would not, as the record shows, just be the price of Coca-Cola which would be more active as a result of intrabrand competition; the suppliers of hundreds of other interbrand soft drink products must be responsive to the prices of Coca-Cola and the allied products all(~ they could not afford to ignore for too long any reductions in the wholesale price of these products. (51) 2. Intrabrand Competition in the Sale of Coca-Cola and the Allied Products Would be Likely to Result in Increase4 Competition in the Sale of Soft Drink Beverages Respondents argue that Coca-Cola and the allied products are sensitive to the prices of competing brands, and as a result, the bottlers' pricing decisions must be influenced by the interbrand competition in their respective territories. As respondents submit there is evidence of price sensitivity in the record; however, as we have determined, the "intensity" of interbrand competition in the soft drink bottling industry is affected by piggybacking and substantially diminished by the territorial restrictions imposed by respondents and similar restraints imposed industrywide by other syrup companies. a result, interbrand competition may not be fully exploiting the price sensitivity of respondents' soft drinks, and equally important, as a consequence of the restraints on intrabrand competition here chal­ lenged, Coca,-Cola and the allied products are not fully challenging the sensitivity of other soft drink products to their prices. Evidence adduced at the trial by respondents shows that Coca-Cola and the allied products compete with such products as local, regional and national brand flavored carbonated beverages; private label soft drinks; and to some extent, powdered mixes and noncarbonated drinks. (RPF 157-80). To the extent Coca-Cola competes with and is price sensitive to these types of products, it may be concluded, particularly in view of the fact that Coca-Cola is the nation s leading flavored carbonated soft drink premium brand and a dominant brand in many local markets across the country, that other soft drink products are equally, if not more, sensitive to Coke prices. According to the bottlers of other brands, price competition of Coca-Cola takes sales away from Pepsi-Cola (Tr. 2886-87, 2889-90), 7-Up (Tr. 2682), Nestea (Tr. 3456), and Lipton canned ice tea (Tr. 3562-63). (RPF 219). Mr. Hurst, the ...

Opinion 91 F.

marketing manager for Nestea Co., testified that his canned ice tea product loses sales if it is one cent higher per six-pack than the premium priced carbonated soft drinks such as Coca-Cola. (Tr. 3456­ 57). Similarly, in response to questions propounded by respondents' counsel, the defense witness, who bottles Pepsi-Cola, Dr. Pepper, and Bubble-Up in Dyersburg, Tennessee, testified: Q. Can you afford to sell Dr. Pepper at a higher price than ~~Cola is being sold in your territory? A. Oh, definitely not.

Q. Now, are we talking about a dollar more, or a few cents per case? (52) A. I don t think-well, we would not let ourselves be caught in a situation whereby, over an extended period of time, any major product was being sold at a cheaper price than our products.

Q. You mean when you say cheaper. . .

A. Not even one or two or three cents a bottle or carton. (Tr. 3046). While respondents, in their answer brief, attempt to minimize the importance of intrabrand competition among the bottlers of Coca-Cola the acknowledged sensitivity of interbrand soft drink products to the price of Coca-Cola and the allied products refutes respondents' contentions. Because Coca-Cola is, as respondents' evidence solidly confirms, an important interbrand competitive force in the market, a practice which eliminates intrabrand price competition has adverse repercussions throughout the entire soft drink industry. As the evidence clearly demonstrates, lower prices for Coca-Cola would, in turn, exert enormous downward pressure on the price of interbrand flavored carbonated beverages and, to a lesser degree, on Kool Aid Funny Face, fruit juices, and all other soft drink products which according to the bottlers, compete with Coca-Cola. For this reason, the judge s conclusion that competition among the independent bottlers of a premium brand soft drink such as Coca-Cola would "dilute" their competitive efforts against interbrand bottlers could not, consistent with the pricing dynamics of this industry, apply to pricing behavior. Rather than dilute the Coca-Cola bottlers competitive impact interbrand, the record shows that intrabrand price competition would, perforce, strengthen their impact considerably. Thus it does not appear that price competition in this industry enhanced by respondents' territorial monopolies. In fact, evidence in the record demonstrates that exactly the opposite is true. We conclude that respondents' territorial restrictions substantially lessen competi­ tion among soft drink suppliers in the "corridor area" and the rest of the country, in violation of Section 5 of the Federal Trade CommissionAct.47 (53) 47 Respondents and intervenors contend that the legality of their territorial licenses was judicially upheld in 1920 (Ccmtinued) ...

THE COCA-COLA CO., ET AL. 645 517 Opinion VI. RELIEF A. RETURNABLE REFILLABLE PACKAGING Respondents' bottlers package their products in two basic types of containers: those which the consumers usually discard after use and those which may be returned to the bottler, purified, and reused. Both types of containers offer consumers distinct advantages. Nonrefillable cans and bottles, or nonreturnables as they have been referred to this proceeding appeal to consumers who prefer convenience, throw­ away packaging and are willing to pay for it. Refillable bottles, in contrast, appeal to consumers who are concerned more with economy than (54) convenience.48 For reasons stated below, we find it unnecessary to disturb established bottling territorial relationships which now exist with respect to the sale of Coca-Cola and allied products in returnable, refillable bottles. As in litigation involving mergers which violate Section 7 of the Clayton Act in which partial rather than full divestiture provides satisfactory relief Federal Trade Commission v. Pepsico, Inc. 477 F.2d 24, 29-30 (2d Cir. 1973); U.S. v. Reed Roller Bit Co. 274 F. Supp. 573 (W.D. Okla. 1967); Warnerin The Coca-Cola Bottling Co.v. The Coca-Cola Co., supa, In ruling on the legalityof the territorial licenses in response to the actions The Coca-Cola Company had taken to put its bottlers out of business, however, the district court was not required to decide, nor did it have before it, the legality of the territorial restrictions which are now before us. The basic question before the court there was whether The Coca-Cola Company had the right arbitrarily to terminate the bottlers over disputes concerning the priceof syrup, The consent decree eventually entered in that case reflected, in part, the court's ruling that the bottlers have a right to purchase Coca-Cola syrup and to use the trademark, in perpetuity, and to be free, not from the threatof competition among themselves, but from arbitrary termination by The Coca.Cola Company. The regulation of competition among the independent bottlers was not an issue joined before the court; the provisionsof the consent decree respondents rely upon in supporttheof restrictions here challenged apparently were inserted by the parties to accommodate their private interests. More directlyon point is the decisionof the district court in Tomac I'M. v. Coca-Cola Co" 1976-2 Trade Cases, ~60 988, in which the court, in a private treble damage action, overturned a jury verdict finding respondents' territorial restrictions illegal. The judge in Tomac concluded that the restraint was reasonable because it promoted a legitimate business purpose by providing incentives for capital investment and enhancing competition, (at 69,381). For reasons discussed at length in this opinion, we respectfully reach a contrary conclusion. We feel it is, in these circumstances, appropriate also to note that decisions of federal and state courts approving a practice challenged by the Commission would not foreclose a contrary FTC Section 5 decision, nor would consent decrees entered by agreement of the parties in settlementof a private suit.FTC v. Sperry Hutchinson 405 S. 233, Z39 fn. 4 (1972), 48 As we mentioned earlier, prior to 1955, the only unit in which bottled Coca-Cola was offered was the 6 lI2-ounce returnable container. (Unless otherwise indicated, the term "returnable bottle" or "returnables" refers to the type bottle which can be refilled and reused by the bottler.) The popularity of returnables declined after disposable containers were introduced, then increased and stabilized. This recent stabilization is attributed principally to adjustments in the deposit structure and the fact that economy-minded purchasers are buying refillable bottles and returning them while convenience buyers are purchasing nonrefillable packages. (RPF 349), Today the refillables are an important competitive factor in the market, accounting for about 55 percentof the sales of Coca-Cola in bottles and cans on a volume basis. (RPF From territory to territory, the percentageof soft drinks sold348,in refillableTr.bottles 3633).varies. (Tr, 3758--59, 3777-78; RX 7). For example, 30 percentof the sales of Coca-Cola in bottles and cans in Washington, D.C., are packaged in refillable bottles (Tr, 1167); 65 percent in Hartwell, Georgia (Tr. 1384); 70 percent in Spirit Lake, Iowa (Tr. 1462) and the State of Iowa generally (Tr, 1463); 30 percent in Wilmington, Delaware (Tr, 1541-42);25 percent in Havre de Grace Maryland (Tr. 1542); 75 percent in Charleston, West Virginia (Tr, 1542); 54 percent in Miami (Tr. 1542); 74 percent in Montross, Virginia (Tr. 1633); 40 percent in Reading, Pennsylvania (Tr. 1916); 20 percent in Coatesville, Pennsylvania (Tr. 2172); 51 percent in San Antonio (Tr. 2487); 45 percent in Stockton, California (Tr, 2567); 55--57 percent in Palo Alto, Burlingame, and San Mateo, California (Tr, 2610); 60 percent in Jamestown, North Dakota (Tr. 1982); and 70 percent in Ada, Oklahoma (Tr. 2670).(See also, Text at 17 supra), Opinion 91 F.

4p-..

Lambert Co. 87 F. C. 889- , 88 F. C. 503 (1976); RSR Corp. C. 873, 892-97, we have here determined that partial relief, which is limited to lifting the restrictions as they apply to nonrefillable containers, is fully adequate in the interest in maximizing both intrabrand and interbrand competition. (55) We have carefully considered complaint counsel's suggested option of placing an identification mark on each bottler 8$ a method of preserving the competitive viability of the refillable bottle, but we are unable, on the basis of the record evidence, to agree with their contention that less restrictive measures are viable -alternatives in the context of a system in which refillable bottles are purchased and used by numerous independent producers of a nationally trademarked finished soft drink which is also offered in nonrefillable bottles and cans.

Economy of Returnable Bottles I t is uncontroverted in the record that in virtually every territory in which refillable and nonreusable packages are offered, Coca-Cola is, on the per-ounce basis, significantly cheaper in the refillables 49 and the advantage is evident notwithstanding the fact that these bottles initially cost the bottler more than cans and disposable bottles, and the further fact that retailers generally take a larger markup on returnable bottles to compensate for the additional cost of handling the empties returned by consumers. (RPF 132). (56) The price disparities reflected in this record are, to a large extent explained by the fact that when a consumer buys soft drinks in nonreturnable bottles and cans, the bottler, at wholesale, and the retailer must recover the full cost of each beverage container with each sale. 50 In contrast, the full cost of a refillable bottle ordinarily need not 49 Evidence of the economy of the returnable bottle is reflected in the per-ounce price differentials between Coca. Cola in returnable bottles and nonreturnable containers. For example, in July, 1971, it cost the consumer in Baltimore approximately 33 percent more per ounce to buy Coca-Cola in 16-ounce nonreturnable bottles than in l(h)unce returnable bottles, and 66 percent more per ounce in l2-ounce cans than in 16-ounce returnable bottles. (Tr. 982). In Wilmington, Delaware, the retail price of 32-ounce returnable bottles of Coca-Cola is four for $1.69 or 1.32 cents per ounce; the prevailing retail price for cans is six for $1.49 or 2.06 cents per ounce, 36 percent more expensive to the consumer on a per-ounce basis. (Tr. 1541), In Montross, Virginia, the 16-ounce returnable bottle retails in supermarkets at 1,08 cents per ounce (Tr. 1680); cans retail at 2 cents per ounce or approximately twice as much. (Tr. 1680, 1692). In Jamestown, North Dakota, the current retail price per ounce of Coca-Cola in 32-ounce returnable bottles is 1.2 cents; the price per ounce in 32-ounce nonreturnables is 1.5 cents and in cans, 2.2 cents. (Tr. 1981). Coca-Cola in 16-ounce returnable bottles is, on a per-ounce basis, 29 percent cheaper than Coca-Cola in 16-ounce nonreturnables;Z1 percent cheaper than Coca-Cola in 32-ounce nonreturnables; 16 percent cheaper than Coca-Cola in 64-ounce nonreturnables; and 61 percent cheaper than Coca-Cola in l2-ounce cans in Reading, Pennsylvania. (Tr. 1925). In San Antonio, Texas the prevailing retail price per ounce for Coca-Cola in 16-ounce and 32-ounce returnable bottles is about a penny. Coca- Cola in 48-and 64-ounce nonreturnable bottles retails at about 1.5 cents per ounce, or 50 percent more expensive, and Coca-Cola in cans retails at 1.9 cents per ounce, or 90 percent more expensive. (Tr. 2488, 2551; RPF 208), 50 The record shows that a case of 24 l2-ounce aluminum or steel cans costs about $1,44 or 6 cents per can. The Coca-Cola bottler in Spirit Lake, Iowa, for example, testified that Coca-Cola in his territory in 16- and 32-ounce returnable bottles is about 50 percent cheaper than Coca-Cola in cans, even though a case of 24 empty l2-ounce cans (Qmtin~) THE COCA-COLA CO., ET AL. 647 517Opinion 4p., be recouped all at once, but can be spread over the number of trips the bottler can expect the bottle to make before it is lost, destroyed, or no longer usable. (Tr. 997).51 Consequently, if a 16-ounce returnable bottle which costs 12 cents survives 18-20 trips, it generates a container cost of only a fraction of a cent per trip. (Tr. 1461-62, 2488, 3996). As one bottler testified:

when I price my packages I add right on top" the cost of the package. A l0-ounce package, for instance, of a returnable Coke is $2.50 a case . . . 240 ounces, so we are talking about 1.1 something (cents per ounce). When we talk about a l0-ounce NR (nonreturnable) package, we are talking about $3.60 a case for 240 ounces, probably 1.5 (cents) per ounce, so that is the price of convenience. (Tr. 2149, RPF 209). (57) Bottle Trippage The record shows that in pricing his packages, a bottler must be able to anticipate, with a reasonable degree of accuracy, his returnable bottle requirements and glass "float" inventories. 52 (Tr. 700, 735, 2486; RPF 131). While bottlers continuously invest in returnable bottles to replace those which are lost or no longer usable (RPF 69), the territorial restrictions permit the bottler to anticipate that most of the reusable bottles he puts into the market will be returned by the consumers to the stores within his territory and will be returned by those stores to him. As a result, a trippage rate, which represents the average number of cycles or reuses a bottler can expect from a bottle can be determined in each territory (Tr. 3635)53, and used by the bottler in allocating container costs in accordance with his anticipated trippage experience. Generally, the lower the trippage rate in a costs him $1.44 while a case of 12 empty 32-ounce returnable bottles costs him about $3.11. Yet because his trippage rate is about 25 per bottle, his container cost per case for 32-ounce returnables(i.e., 384 ounces) is about 11 cents per trip, in contrast with the full $1.44 per case (i, 288 ounces) on one-way bottles or cans. (Tr. 1462). 51 The nonrefillable bottle is not designed to withstand the punishment of reuse. Made of thinner glass than the refillables, products liability considerations dictate that it be used only as a one-way, one-fill container. (Tr. 3765-68). While some jurisdictions have enacted litter laws which require the consumer to pay a deposit, which is refundable upon the return of nonrefillable bottles and cans, the containers reclaimed are not returned to the bottler for reuse. Instead, the nonrefillable bottles recovered from post.-consumer waste streams are processed or recycled into crushed glass or cullet for glassmaking processes. Unlike the refillables, then, the bottler cannot spread the cost of a , returnable, nonrefillable bottle or can over more than one sale. 52 The term "float" refers to the total number of refillable bottles of a given type in the bottler s system; it includes those in the inventories of both the bottler and the retailers in his territory as well as those in the homes of consumers. (Tr, 3769). A bottler s minimum "float size" equals his sales multiplied by his anticipated turn-around period. Thus, if the bottler s turn-around time is six weeks and on the average he sells 100 24-bottle cases of returnables per day, five days a week, his float size would be approximately 3000 cases or 72 000 bottles (100 (cases) X 24 (bottles per case) X 5 (days) X 6 (weeks turn-around time).). (Tr. 3770). 53 The trippage rate, in turn, depends upon a bottler s "float" size and estimated turn-around time (or the anticipated time it takes each bottler on an average to recover an empty from the consumers in his territory) and the bottler s loss rate. Thus, in a territory in which sales remain constant, "float" size is constant, and turn-around time is constant, the replacement rate would be equivalent to the loss rate, and the bottler would purchase just enough bottles to replace those which have been lost or destroyed, In contrast, in territories in which sales are increasing, the bottler must not only replenish the bottles he has lost, but also invest in a bigger "float;" if sales are decreasing, on the other hand, the bottler s "float" itself may supply his bottle needs so no new investments in glass may be required, (Tr. 3636), Or- ­ Opinion 91 F.

territory, the more rapidly the bottler must recoup the bottle s full cost, thus increasing the per-ounce price of the soft drink.54 (58) 3. Bottle Recapture In assessing the impact of the order proposed by complaint counsel it is pertinent in the context of respondents' bottler' network that the use of refillable bottles makes economic and competitive sense only if each bottler is able steadily to recapture from the -market an adequate predictable supply of used bottles to service his production require­ ments. Consequently, there are two major impediments to intrabrand competition in the use of refillables: First, retailers will, from time to time, switch their Coca-Cola bottler supplier; and second, consumers will buy and return bottles to different retailers. Over an indefinite time, then, the refillable bottles provided by a number of bottlers will periodically be returned by consumers either to the store from which they were originally purchased or to a different store supplied by the same or different bottlers. As a result, bottle recapture, under these circumstances, would be unpredictable and economically burdensome. (Tr. 2996-98, 2027). Even if the bottler were to place an identification mark on his bottle, it would be impractical and costly to expect the retailer to notify each bottler whose bottles he may have collected or to require the bottler to divert his trucks to pick up a few empty bottles from retailers who, at the time, may be purchasing "Coca-Cola" from a competing bottler. (Tr. 2544-49).

Nor would the burden of recapture be substantially reduced if a . bottler picked up all of the empties, regardless of their source, from each of his customers. Each bottler individually purchases his returnable bottle float and must be able to anticipate his bottle needs based on trippage experience in his territory. Retail outlets which collect large numbers of returnable bottles would provide an abun­ dance of bottles to their suppliers, while other bottlers serving retailers which collect relatively few bottles may experience shortages. Because a bottler would be unable to predict the retail customers he may acquire or lose over time, or their locations, and because bottlers maintain glass inventories of varying sizes, there can be no assurance that the number of bottles a bottler puts on the market will, on a random basis, equal or even closely approximate the number he may pick up in return. It would be virtually impossible for a bottler determine, with a reasonable degree of accuracy, what portion of his float outstanding in the market will be returned to him for reuse. (59) Alternatively, if each bottle carried an identification mark and all 54 In territories in which returnable bottles are offered, the record shows that trippage rates vary from as low as five in some territories to as high as 30 in others. (Tr. 2579, 2995, 1859; RPF 344). ....

THE COCA-COLA CO. ET AL. 649 517 Opinion bottlers picked up all empties from their respective customers, each bottler would be picking up other bottlers' bottles, backhauling the empties, storing them, and notifying the other bottlers who would have to pick them up from widely dispersed collecting bottlers, thus substantially increasing the handling costs associated with the use of returnables while diminishing their economy advantage. It has been stipulated on this record that the use of returnable bottles incompatible with central warehouse distribution by retailers l~rgely due to the impracticality and costs of having the retailer collect backhaul, sort, and store empty bottles for the bottlers. Nothing in this record suggests that it would be any more practical or much less costly for a bottler to perform a central warehouse function for the return of other bottlers' empty bottles.

Similarly, a credit system which would permit a bottler to use bottles purchased by competitors would probably not result in a competitively viable distribution of empties in accordance with the bottlers' bottle needs or investments. Bottlers may offer a wide range of refillable options, including 6 1/2-ounce, 10-ounce, 16-ounce, 26-ounce, and 32­ ounce sizes with different investments in each size; and while some bottlers offer most sizes, other bottlers offer only one or two. Consequently, a bottler who maintains a sizeable float which presently services his production runs may end up, from week to week, with too few bottles actually on hand against which credits could be claimed to compete effectively for returnable bottle sales. This could occur, for example, either because a bottler may, as we mentioned, lose retail accounts which collect large numbers of empties or because he may be collecting an assortment of bottle sizes, some of which may not be compatible with his bottling line equipment, or because he has collected too few bottles in each size to offer any size on a competitive basis. 55 While an increase in the amount of the deposit a bottler may require might protect his investment in glass bottles (Tr. 2097, 3098-3100), the competitive potential of the returnable bottle system would likely be lessened since higher deposits would probably meet with appreciable consumer resistance and encourage a shift to disposables. (Tr. 3051 2522, 996, 2871, 1994). Nor would intrabrand competition be (60) fostered if bottlers had to invest continuously in new bottles or even used bottles, assuming a secondary used bottle market were to spring into existence, to compensate for wild, frequent fluctuations in float. (Tr. 998).

55 Even if several bottlers were to form a cooperative for the productionof soft drinks in refillable bottles, the recapture problem would still exist vis-a-vis the members and nonmembers of the co-op. (See Tr, 2139--41), .,.., Opinion 91 F.

Under these circumstances, we find it unnecessary to lift restrictions on the sale of Coca-Cola and allied products in refillable bottles. Rather, as the record ~hows, fully adequate relief in this matter necessitates only the lifting of the restriction as it affects the sale of these products in the nonrefillable containers. Because the relative market strength of convenience and returnable refilla,ble packaging is largely dictated by a price spread sufficient to maintain the consumers participation in the return system, any downward price movement resulting from intrabrand competition in the sale of nonrefillables would directly influence the price of refillables. Conversely, a viable refillable bottle system operating in the context of an exclusive territory will provide each bottler with a potent price-competitive package. The relief entered in this proceeding will, therefore differentiate between reusable and nonreuseable bottles and cans based upon demonstrated economic effect. (GTE, supra). 4. Split Delivery Weare mindful of respondents' defensive arguments that the use of refillable bottles is inexorably linked to territorial exclusivity and store-door delivery of each bottler s entire package mix, and the belief expressed by several bottlers that chain store outlets would substan­ tially reduce, if not eliminate, their refillable bottle purchases if warehouse delivery of other types of packages were offered to them. Respondents' scenario projects a decline in the volume of soft drinks packaged in refillable bottles and distributed via store-door delivery and, as a result, price increases to cover fixed costs at the reduced volume. (Ans. Br. at p. 57).

While the record shows that a few high-volume chain stores have refused to retail returnable bottles (Tr. 2170--72, RPF 13~6)56, we find no basis in the record for concluding that a (61) substantial segment of the nation s chain store population will follow this lead. According to the bottlers, some chain store customers complain about the handling costs associated with storing and sorting empty bottles (RPF 136), but respondents submit that the retailers take a markup sufficient to compensate them for their trouble, and there is no indication that the profit on returnables is not comparable to that which is made on nonreturnable bottles and cans. (RPF 132, 137). 56 The record shows, for example, that in CoatsvilIe, Pennsylvania, several food chains have declined to handle returnable bottles (RPF 143), but the bottler in that territory not only offers returnable bottles, he offers his l0-ounce returnable bottles of Coca-Cola for $1.10 less per case than his l0-ounce nonreturnables (RPF 209, Tr. 2149); and despite the refusal of the chain stores to retail this type of package, he is planning to introduce the 32-ounce returnable, resealable bottle. (Tr, 2171), Similarly, in Reading, Pennsylvania, four large chain stores do not carry returnables, yet 16-ounce returnable bottles, priced on a per-ounce basis, were 61 percent cheaper in that territory than Coca-Cola in 12-ounce cans, (RPF 208). In Albany, New York, no chains carry Coca-Cola in returnable bottles, but the package is available in that territory, (RPF 143). .,..., THE COCA-COLA CO., ET AL. 651 517 Opinion Apparently, the refusal to retail this package is a competitive decision and in view of the uncontroverted evidence that the wholesale price for returnable, refillable bottles is usually cheaper than nonrefillables, the retailer who rejects the former may be less price competitive as a consequence. (Tr. 1771-73). Although some retailers may justify the disadvantage, many more likely will not.

As the record shows demand for returnables has increased and stabilized in recent years at about 55 percent of "Coca~Cola nationwide can and bottle statistical sales volume 57, and while the percentage varies from territory to territory, returnables are significant factor in virtually every territory surveyed in this record. (RPF 348). We therefore find it difficult to conclude that most high-or low-volume soft drink retailers who now handle the package would ignore this demand by declining to offer consumers the choice between convenience and economy packaging. (62) Furthermore, while numerous bottlers subscribe to the contention and therefore conclude that returnable bottles could not be offered competitively in a split-delivery environment of store-door distribution and central warehousing, their testimony is largely based on speculation. (Compare. Tr. 3575-76). Only two of the Coca-Cola bottlers who testified in this proceeding ever experimented with split-delivery of Coca-Cola, and their testimony shows that chain stores which have obtained central warehouse delivery of Coca-Cola in cans have continued to purchase it in bottles delivered directly to their retail outlets. Thus we find unwarranted the assumption that high-volume (See also Textaccounts will disappear from store-door delivery routes. at 74-76 infra).

Moreover, the efficiency of a store-door route depends upon such factors as the number of customers on the truck route, the volume of soft drinks delivered to each customer, the distances between customers, and the time required to make each delivery. These factors , in different territories, inmay vary greatly on different routes various competitive situations.58 Consequently, a bottler can achieve delivery efficiencies by adjusting the type of accounts serviced on each 51 According to stipulated data, total food store sales of the Coca-Cola brand alone in 1960 represented the movement of 143 million statistical cases, including returnable bottles and nonreturnable bottles and cans,2Z-44).(RX By 1971 these statistical case sales of Coca-Cola had grown to 327.9 million cases. If we assume that only half of the Coca-Cola food store sales volume in 1971 were sales in returnable bottles (Tr. 661, 777­ 3633, 3653, 3755), Coca-Cola brand volume in returnable bottles alone was a little over 163.9 million statistical cases, exceeding by approximately 20 million statistical cases the returnable and nonreturnable food store package volume in 1960, (Tr. 3653). 58 Under the present system, for example, virtually all products are distributed on a store-door delivery basis and costs vary from one territory to another, In Hartwell, Ga" the bottler s break-even point per delivery is four cases (Tr. 1370-71); in Coatesville, Pennsylvar;tia, on his scheduled routes, the bottler s break-even point is five cases (Tr. 2191); in San Antonio, Texas, the bottler estimated that he broke even on deliveries involving about six cases. (Tr. 2554--55), ), ... Opinion 91 F.

route by each of his trucks. Such route adjustments are not unknown Bottlers are, for example, flexible inin the industry. (Tr. 4044).59 adjusting their routes in response to fluctuation in demand for soft drinks caused, for example, by seasonal variations (Tr. 2567, IDF 38 Stip. No. , ex 1244G, Tr. 476) or by the addition of piggybacked brands which they (63) may distribute to customers who are located beyond the limits of their primary territory. (See Tr. 2848-56, 3064-69). In addition, delivery costs may be reduced by route adjustments which eliminate deliveries to unprofitable accounts or by establishing a minimum volume which the bottler will deliver to a customer s place of business. (See Tr. 1932, 2554). We recognize, of course, that some territories may be too small and the returnable bottle volume too insubstantial to allow a bottler to operate efficiently. A similar problem exists under respondents' territorial system. Yet in such circumstances in which a bottler is unable to compete in returnable bottle sales, he may merge or consolidate his territory and plant with that of another bottler, as respondents now recommend to their small bottlers as a means of increasing their volume and efficiency. (Tr. 615). For these reasons, we conclude that territorial (64) restrictions which cover a bottler s entire package mix are not justified because part of the mix includes the use of refillable bottles. ~9 It has been suggested that central warehouse delivery would siphon away 50 percent of the store-door delivery volume in " most territories. This assumption, however, is speculative. The record shows that chain stores, large independent supermarkets, and convenience stores which are serviced by warehouses for other food items, as a class of customers, account for about ~ percent of the total sales of the bottlers of Coca-Cola nationwide. Within various territories, the percentage varies. In Washington, D. , this customer class accounts for about 27 percent of the bottlers' sales; in Herminie, Fa. about 20 percent; in Wilmington, about 33 percent; in Belmont, Calif. about 60-65 percent; in Westminster, Md. , about 18-20 percent. (RPF 325), We note, in addition, that the fact these customers are serviced by warehouses for other food items does not mean that all or any specific portion of their requirements for Coca-Cola would be centrally warehoused.(See Text at 75-77 with accompanying notes infra, For example, the percentage of the bottlers' sales volume which is packaged in refillable bottles and the large sizes of nonrefillable bottles may continue to be delivered store-door to the retail outlets of these customers. Consequently, that portion of the bottlers' total sales volume which may actually be centrally warehoused will probably be, in many instances, significantly less than the bottlers' total sales volume to the class of customers who are serviced by warehouses for other food items. 60 As we previously mentioned, prior to 1955, finished Coca-Cola was packaged solely in refillable bottles, Respondents' practice now applies to the bottlers'entire package mix, of refillables and nonrefillables, and we considered the effects of the restraint in . that context. Under our order, a market context will prevail in which intrabrand competition will be fostered in the sale of Coca-Cola and allied products packaged in nonreturnable containers, Under these circumstances, we believe the restraint, if limited to refillable bottles, is reasonable for the reasons discussed above, There is no occasion to determine whether the restraint,before the introduction of nonrefillables, was reasonable as applied solely to sale of Coca-Cola in refillable bottles prior to 1955, We should emphasize that our finding of reasonableness here is also limited to the use of refillable bottles, We note, for example, that no evidence was adduced that intrabrand competition would unduly burden the use of refillable containers which may be used in pre-mix or post-mix systems, These pre-mix and post-mix systems, unlike the refillable bottles, remain with the retailer who dispenses the beverage, Thus, it is significant that respondent Coca- Cola successfully packages and sells its fountain syrup in refillable five-gallon stainless steel tanks (Tr. 3773), even though the fountain wholesalers are not confined by territorial restrictions. In fact, the recapture and return system for post-mix containers seems to work well in view of the fact that The Coca-Cola Company has found it unnecessary to impose a deposit refundable upon the container s return, (Tr. 3773-74). As a prophylactic measure against the imposition of the restraint in the future, our order will cover post-mix syrup sales and distribution, ...

THE COCA-COLA CO., ET AL. 653 517 Opinion B. "SMALL BOTTLERS Respondents contend that an order eliminating the territorial restrictions which they impose on their bottlers would result in restructured, highly concentrated industry dominated by a few large bottlers. Reversing the thrust of their own argument that interbrand competition now places limits on the extent to which their bottlers both large and small, may increase their prices, respondents assert, " this concentrated economic environment (i. an environment free of respondents' territorial restrictions) in which hundreds of small bottlers61 had been forced out of business " wholesale prices would rise. (Ans. Br. 57). (65) Respondents' concern about a market structure in which their bottlers are competing intrabrand in the sale of Coca-Cola and allied products is indeed a curious defense of territorial restrictions which allow one bottler to be the sole source of supply of these products to the customers within his territory. Contrary to respondents' assertions, the removal of the restraints would probably result in substantial reduction in concentration as existing independent Coca-Cola bottlers expand geographically to encompass the previously captive retail outlets of other bottlers in areas they are now forbidden to penetrate. Rather than reducing competition and increasing concentration, the elimination of territorial restrictions will probably increase both actual and potential competition and decrease concentration. Territorial Restrictions as a Method of Protecting Small Busi ness Respondents' protestations about concentration and the future structure of the industry aside, the thrust of their argument is predicated on the notion that small independent Coca-Cola bottlers would be unfairly disadvantaged by intrabrand competition. N umer­ ous bottlers, particularly the smaller bottlers, testified that they were dependent upon the refuge of their territorial enclaves because intrabrand competition would force them out of business.62 This 61 At the time of his testimony at the trial, the President of The Coca-Cola Company could not define the term small bottler " and when asked by Judge Dufresne "what is a small bottler?" he testified: "Well, I don t know, sir. . . (w)e have never really tried to make such a definition. " (Tr. 590-91). The term "small bottler" is,of course, a relative term. Subsequent witnesses noted that the relative size of a bottler may be measured by the population in his territory, his annual case sales of soft drinks, and the number of people he employs. (RPF 279). The Small Business Administration classifies a manufacturer with less than 250 employees as a small business. In 1974 respondents conducted a census of Coca-Cola bottlers, and of the 567 bottlers who responded, representing about 75 percent domestic bottlers of Coca-Cola, 529 had fewer than 200 employees. (RPF 284). In this proceeding, however, the term small bottler" has been used primarily as a reference which encompasses a number of factors, such as a bottler s sales volume, production capacity, proximity to central warehouse customers, and his access to capital resources, among others, which are said to give one bottler a competitive advantage over neighboring bottlers. 62 It is a questionable hypothesis as to whether territorial restrictions promote theofviabilitysmall business in view of the fact that they nece.clSitate survival for many small bottlers by merger, rather than growth by internal (Qm.tin~) .. .... , . . . Opinion 91 F. C':

assessment was, in (66) turn, based on several assumptions which were adopted in a series of important findings in the initial decision.63 The judge concluded that without exclusive territories, large bottlers of Coca-Cola would drive smaller bottlers out of business. He further concluded that a Commission order lifting the territorial restrictions: would be in direct conflict with the purpose of the Congress in enacting and in agencies administering the antitrust laws Of. . . to perpetuate and preserve, for its own sake in spite of possible cost, an organization of industry i~ small units which can effectively compete with each other. S. v. Aluminum Co. of America, 148 F.2d 416 429 (2d Cir. 1945).

We acknowledge this admonition that one of the underlying purposes of the antitrust laws is to protect and preserve small business; indeed, in American Cyanamid64, the Commission noted that "This agency also has its very roots planted in that philosophy. " Our previous decisions implementing this philosophy clearly indicate however, that we have never condoned anticompetitive practices solely for the purpose of eliminating competition between large and small firms. We stated in Procwr Gamble. it may be appropriate. . . to note Congress' concern with the preservation (of small firms), to the extent compatible with social and economic progress, of the fundamental benefits of a small-business, decentralized economy. The interest fostering equality of opportunity for small business and in promoting the diffusion of economic power. . . was unquestionably intended by Congress to be relevant in any scheme for the enforcement of Section 7. (63 F. C. 1465, 1555-56 (1963)). (67) But in effectuating this policy, the Commission made clear that it does not subordinate "the protection of competition to the protection of small business competitors. (Id. citations omitted; Compare Ans. Br. at 66-67). "Otherwise " as the Third Circuit has observed in another context what is intended as a shield for small competitors becomes a sword against the consumer. NBD Industries Treadway Cas. , Inc. Brunswick Corp. 523 F.2d 262, 279 (3rd Cir. 1975), vacawd on other expansion. Between 1968 and 1971 there were 107 bottling plant mergers among Coca-Cola bottlers. (Tr. 650-51). In addition, respondent Coca- Cola has issued 14 temporary marketing bottler agreements, pursuant to which bottlers who have discontinued production continue to distribute, within their territories, Coca-Cola produced for them by neighboring bottlers. (CX 1245 A-M, CX 1246 A-J). When these agreements expire, The Coca-Cola Company does not intend to renew them. (Tr. 900). The marketing bottlers will then have the option to resume bottling or merge their territories with some other bottler. (Tr. 901). According to the President of The Coca-Cola Company, the demise of small independent units of production under its system is a function of improvements in transportation, economies of scale, shifting population, changing tastes, and income patterns which "have tended to reduce the number of bottling plants and .increase the size of some territories." (Tr.614). In circumstances in which bottlers are too small to operate efficiently and foreclosed by territorial restrictions from significant internal expansion, respondents recommend that they merge or consolidate their production with another bottler (Tr. 615; see also Tr. 9O().4)1), thus reducing the population of small bottlers. 83 See IDF 185-193, 84 63 F. C. 1747 1857-58 (1968).

. . .

THE COCA-COLA CO., ET AL. 655 517 Opinion grounds and remanded, Brunswk Carp. v. Pueblo Bowl- Mat, Irw. 429 U.S. 477, 1977-1 Trade Cases, ~61 255.

Consistent with our prior application of these principles, we conclude that territorial restrictions are not justified as a means of protecting small independent Coca-Cola bottlers from large independent intrab­ rand rivals, but that ancillary relief is necessary, in the public interest to prevent The Coca-Cola Company s integrated bottling operations from exploiting certain advantages which may accrue to it as a dualdistributing trademark licensor.

Large Independent Bottlers v. Small Independent Bottlers Recognizing that all of the bottlers who testified in this proceeding were concerned about intrabrand price competition within their respective territories, respondents and the bottler intervenors also adduced evidence from which it may be concluded that respondents' territorial policy is today the chief reason why many bottlers remain small. With access barred to retail accounts in densely populated areas now under the lucrative intrabrand domain of the metropolitan . bottlers, expansion by a small bottler largely depends, absent a territorial merger, on population growth in his territory and per capita consumption of his product. Under these circumstances, large bottlers in the nation s major cities may be the principal benefactors of the special protection " these restraints afford. (Tr. 872). The record shows that small bottlers, in some instances, may have overall cost advantages because, among other reasons, they have lower labor and land costs and lower taxes than large bottlers located in major metropolitan areas. (RPF 191 , 290; see Tr. 2248, 2363). In fact wholesale prices charged by large bottlers with high-speed, highvolume production facilities are often higher than the prices charged by small bottlers in adjacent territories. (RPF 295; Tr. 2179-80, 2832). Respondents submit this pricing behavior as evidence that many small bottlers may actually be more efficient overall than large (68) bottlers. (Ans. Br. 82; RPF 295, 188).66 This evidence adduced by respondents 65 In one instance noted in the record, territorial restrictions prevented a small Coca-Cola bottler from doubling his annual volume by selling canned Coca-Cola to a customer who intended to transship out of the small bottler territory. (Tr. 665-$), 66 Although the record is silent with respect to the bottling or canning plant volume necessary to achieve. full economies of scale, bottlers who have merged their geographic markets and consolidated their operations testified that they have improved their bottling line efficiency. According to respondents' expert, however, the relative economies of production between large and small bottling plants are not a "big factor" in the soft drink bottling business. (Tr, 1044), With respect to canning operations, the production efficiencies of large canning lines average 3--5 cents per case (Meyers 1737-38), and this obviously may be a significant competitive(Seefactor.Tr, 3179), However, nothing in the record suggests that economies of scale are any different for Coca-Cola bottlers than they are for Pepsi-Cola bottlers. Thus it is relevant, in assessing the relative advantage scale economies afford finns different size, that "small" Coca-Cola bottlers effectively compete with "large" interbrand bottlers. The bottler in Hartwell, Ga" for example, has an annual sales volume of about 340 000 cases in a territory serving 35,000 people (RPF 280 283); yet he apparently suffers no overall cost disadvantage despite the fact that Pepsi-Cola is sold in his territory (Continuro) Opinion 91 F.

from bottler witnesses tends to contradict their argument that small bottlers as a class would be unable either to defend their existing sales volumes or expand out of territories in which low costs and prices prevail into the territories of large bottlers who may be incurring higher costs and charging higher prices as a consequence. (See RPF 188, Tr. 881-86). Certainly, no prudent retailer of Coca-Cola and the allied products would continue to patronize a large bottler exclusively if he were able to purchase all or a portion of his products at a lower price from a competing (69) supplier. (See Tr. 3179).67 In effect, then territorial restrictions may, in some instances, be preventing small bottlers from fairly exploiting the competitive advantages which, in open markets, would ordinarily accrue to those who offer lower prices.68 (70) Competition for the Business of th€ High- Volunw Clwin Store Accounts Although the judge found that small bottlers are often located near large bottlers (IDF 186), he also found that large bottlers in metropolitan markets would have a competitive edge over small bottlers for important central warehouse accounts "because chain store warehouses are located mainly in territories of large bottlers." (IDF 185; RPF 329, 333). At the outset, reject the notion that tradeby a bottling operation of General Cinema Corporation, which "owns North Georgia - and most of Florida." (Tr, 1390--93; See also Tr, 1671A-73). In a reverse situation, the viability of small Pepsi bottlers apparently was not threatened despite the fact that their territories were encompassed by the territory of the huge Coca-Cola Bottling Co. of New York. (Tr, 2276-78).

67 The judge found that small bottlers do not have the production capacity to compete effectively for the business of thelarge chain store accounts.

Yet not only is the record unclear concerning the output capacity which would be required to serve all or part of the demand of large retailers from time to time, the finding ignores the fact that a bottler s ability to supply largevolume accounts poes not necessarily depend on his in-house production capacity alone, Bottlers have, in the past supplemented their production capacity by entering into agency canning agreements with contract canners (Tr. 837­ 38, see Tr. 3153--54), and as the record shows, the canned product is ideally suited to central warehousing.(supra fn, 25). Nothing in the record suggests that these canners could not produce canned Coca-Cola for small bottlers at prices which are competitive with the in-house canning lines of large bottlers,(Compare Ans. Br. 63, Fn. 70; Tr. 1325--26), Nor is there any evidence in this record which would suggest that those retailers which presently back haul private label soft drinks produced for them by contract canners would not, except where local union contracts prevent it, back haul Coca-Cola directly from a contract canning plant to the chain store warehouse, In addition to contract canning as a means of boosting the capacity to supply a product, the record also shows that small bottlers can overcome capital barriers by joining together in cooperative soft drink canning ventures, such as the Mid-Atlantic Canning Association owned by 16 bottlers, including many small bottlers, (Tr. 2138, 2923--25, 1500, 1561 1771- 2042), In these ways, small bottlers have arranged for additional production capacity to meet the demands in the markets they serve, Should the demand for the small bottler s product increase, the bamers would not appear to be insurmountable for those who attempt to accommodate it. 68 The administrative law judge concluded that small Coca-Cola bottlers would not have the financial resources to meet the price reductions intrabrand competition may stimulate. While numerous bottlers did, in fact, express concerns about the financial resources of their neighbors, the record also shows that "small" bottlers have been able to price compete with larger, so-alled deep-pocket interbrand bottlers serving customers within the small bottlers territorial boundaries. As in many sectors of the economy in which large and small businesses compete, it is the large firms which usually possess the greatest financial resources, if not superior efficiencies. The soft drink bottling industry is no exception, But the disparity in the financial strength among various firms in a market is not, by itself an accurate indicator of the ability of any particular firm, large or small, to compete effectively in the market. ...

THE COCA-COLA CO., ET AL. 657 517 Opinion restrictive territorial practices can be sanctioned as a means of eliminating fair advantages which may accrue to a bottler by virtue of his proximity to customers.

Beyond that, we find little in the record to support the judge sweeping conclusions in IDF 185. While several bottlers testified that many chain store warehouses are located within the present territodal boundaries of large bottlers, there is scant evidence reflecting shipping distances or the relative "proximities" of large and small bottling facilities to the various chain store warehouse facilities. In fact, food store warehouses which may be located on the outer fringe of the territory of a large urban bottler could actually be closer to the production plant of a small suburban or rural bottler than the plant or distribution facility of the bottler in whose territory the warehouse is actually located. (IDF 186). The judge cited evidence indicating that the Baltimore Coca-Cola bottling facility of The Coca-Cola Company may be closer to an A & P warehouse than the bottling facility of the Westminister, Maryland, Coca-Cola bottler. (IDF 188).69 But neither this example nor the fact that some central warehouses may be located within the territories of some large bottlers, but at undisclosed distances from bottling plants or distribution centers, supports a general conclusion that because of transportation disadvantages (71) small bottlers would be unable to compete effectively for the business of high-volume retailers. The bottlers' prices are influenced by many factors, including their overall costs. As a result, a bottler who enjoys by virtue of his location, a delivery-cost advantage with respect to one customer may be disadvantaged by his location vis-a-vis another bottler and other customers or by cost disadvantages he may incur in other aspects of his operation.

Central Warehouse Delivery and Backhauling by Central Warehouse Customers Furthermore, the judge ignored the fact that respondents' territorial boundaries are no measure of the distances finished soft drinks may be shipped economically. Small bottlers presently haul and backhaul Coca- Cola efficiently from their bottling or coopeatively owned canning plants to their distribution facilities 7O and they often transport the 69 In contrast with the testimony of the fonner president of respondent' s bottling operation in Baltimore, cited in IDF 188, the small 7-Up/RC bottler in Herminie, Pennsylvania, testified that he has two potential warehouse customers in his territory which are located 20-30 minutes from his plant, but one hour from the plant of the large bottler of 7-Up and Royal Crown Cola in Pittsburgh. (Tr, 2823,seeBut Tr. 1783-84). 70 For example, the 7-Up/Royal Crown bottler in Henninie, Pa" also owns the 7-Up franchise in Wheeling, W, Va. He testified that he ships soft drinks packaged in 28-ounce nonreturnable bottles from his Wheeling facility to his Herminie facility on a route which passes through the territory of the large Pittsburgh bottlers; however, because of the restrictions imposed upon his territories, he may not sell soft drinks at wholesale in the Pittsburgh territory. While this bottler would not consider it feasible to sell returnable bottles in Pittsburgh because of the problem of recapturing (Continued) Opinion 91 F. Or- ­ canned and bottled products produced for them by contract canners or other bottlers under agency arrangements over routes which some­ times traverse the territories of large neighboring bottlers. It therefore likely that sizable portions of a large bottler s territory, and the customers within it, may lie within an area which small neighbors might effectively service.

Moreover, while backhauling by high-volume soff drink retailers from the canning plants of contract canners to their central warehous­ es is a customary mode of private label soft drink distribution (Stip. No. , Tr. 2998),71 respondents (72) discount its importance as a means of distributing Coca-Cola and allied products. They claim the chain store trucks servicing the retail stores in the territories of small bottlers could not feasibly backhaul Coca-Cola from the bottlers plants. The judge below agreed with this contention. Relying upon respondents' proposed finding of fact, he cited three witnesses in support of the conclusion that a small bottler could not supply chain store warehouses by allowing backhauling "because the chain store truck servicing the few stores in that territory would not have enough room to pick up a significant supply on a backhaul." (IDF 188).72 We find the reference to "significant supply" vague in this context 73 but assuming it relates to the bottlers' sales volumes, the testimony upon which it is presumably predicated is hardly a compelling basis for the finding.

Mr. Rooks, the Coca-Cola bottler in Hartwell, Georgia, testified that the chains may have "space problems" on the trucks which deliver to the retail chain outlets in his territory (Tr. 1417); and Mr. Christian the President of the Charlottesville, Virginia, Coca-Cola Bottling Works, testified that he thought the chain stores could backhaul on the trucks they use to service their outlets in his territory, but he assumed the chain store trucks were, in fact, already backhauling other items. (Tr. 1843). Despite these assumptions, neither of these witnesses testified concerning the number of trucks servicing the chain outlets in the empties (Tr. 2849--50), he testified that it would be feasible for him to sell nonreturnable bottles in Pittsburgh (Tr, 2853-54), although if he did so, he would expect the bottlers there to respond by competing for customers in his territory, (Tr. 2855), 7J In one instance noted in the record, canned Coca-Cola is being backhauled by a bottl~r s customer. The Alpha- Beta chain in Los Angeles is presently backhauling canned Coca-Cola from the Los Angeles Coca-Cola bottler to its central warehouse and subsequently transshipping it in its own trucks to Alpha-Beta retail outlets located in the territories of neighboring Coca-Cola bottlers, (Tr, 2584-85, 2588, 2634, 2650-51), 72 Compare IDF 188 with RPF 333.

73 It is unclear whether this finding refers to a supply of soft drinks which chain store customers might consider significant" or a sales volume which the bottler would consider significant, However, no chain store customers were called to testify at the trial, and the record does not show what quantity of soft drinks any retailer would consider significant on a backhaul, although presumably a customer interested in backhauling from a particular bottler might consider, among other factors, how far removed the pick-up point is from the delivery truck' s normal route, the quantity of soft drinks it requires and the bottler has available, and the price at which the soft drinks are beingoffered. , THE COCA-COLA CO., ET AL. 659 517 Opinion their respective territories; the frequency of the chain store deliveries to these (73) outlets; the amount of space, if any, which might actually be available on these trucks from time to time; or the amount of space the bottlers thought they would require to permit the backhauling of significant" volume of soft drinks.74 The third witness cited by the judge, Mr. Roadcap, President of the Westminster, Md., Coca-Cola Bottling Company, doubted that backhauling would be feasible'- for reasons entirely unrelated to speculations about truck capacities in back haul situations. (74) Mr. Roadcap stated that, backhauling would not be feasible because other Coca-Cola bottlers would find out that he had allowed Coca-Cola to be shipped into their territories, and in his judgment they would keep cutting the cost and it would go down down, down to the point no one would make money. . . . " (Tr. 2459). Whether backhauling would always be feasible for all of respon­ dents' bottlers and their customers cannot be gleaned from this record but neither was it complaint counsel's burden to disprove respondents' contentions that individual backhaul situations, in some cases, might not be feasible. Absolute competitive equality among bottlers was not a prerequisite of their case. The fact that a particular delivery mode may not be feasible for some does not justify a restriction which virtually precludes all bottlers from freely using it. Store-Door Delivery to Central Warehouse Customers While instances in which respondents' bottlers have offered delivery services other than store-door delivery are, as a consequence of respondents' efforts to preserve their territorial arrangements 74 In finding 188, the judge, relying on the testimony of Mr. Hornsby, Executive Vice President and Treasurer of the K-S Canning Co., noted that an empty tractor trailer truck can accommodate 1800 to 1900 cases of 12-ounce cans. (Tr. 3175). (Both respondents' proposed finding of fact No. 333 and the judge s finding of fact IDF 188 erroneously cite Mr. Meyers, former President of Shasta Beverages, as the source of this statement.) While accurate in substance, the context in which this fact is used in IDF 188 seems to suggest that a tractor trailer truck, if used by a chain store to deliver other food items to retail stores in a bottler s territory, could not, even if empty, backhaul "significant" " supplies of Coca-Cola. Considered in light of other facts presented at the trial, the first two findings in IDF 188 lack the scope necessary to give them any realistic perspective. Recognizing that we cannot, on this record, state definitively the chain store backhaul capacity, if any, which may be available to individual bottlers, we note that if one empty tractor trailer truck, or its equivalent from partial truckloads backhauled by sever:al customers, were available to a bottler once a week, for example, it would provide a back haul capacity of approximately 100 000 cases of cans annually, or the equivalent volume of about 150 000 statistical cases of 24 S-Qunce bottles. We note further that the record shows this would be more than sufficient to haul the total annual soft drink volume many small bottlers now sell to chain store customers, (But see fn, 59 supra), Mr. Rooks of Hartwell, Georgia, for example, had total sales of 340 000 statistical cases (Tr. 1422, RPF 283), but only 25 percent of his sales went to customers with" warehouse facilities, (Tr, 1371, 1438; RPF 325). Consequently, a back haul capacity of about 85 000 statistical cases would maintain his sales volume to chain store customers, Similarly, while Mr. Roadcap s testimony about the feasibility of backhauling was concerned with other matters the record shows that he has total sales of about 500 000 statistical cases (Tr. 2434-35, RPF 283), but only 20 percent of his total represented sales to customers which are served by warehouses for other food items, (Tr, 2436, 2438; RPF 325). An annual backhaul capacity of 100000 statistical cases would maintain his sales volume to chain store customers.

Opinion 91 F. C. 0,­ admittedly rare, actual warehouse delivery situations are not unprece­ dented even within respondents' bottling network. 75 And limited though this experience may be, it shows that bottlers can provide, and their customers have accepted, both warehouse delivery of Coca-Cola in certain types of packages, such as cans, and store-door delivery of (75 J Coca-Cola in bottles.76 Indeed there appears to be a significant market among high-volume retailers for various delivery options. As a consequence, the competitive opportunities for small bottlers in open markets include not only the business which might evolve from central warehousing, but also the store-door trade to chain store outlets both within and outside their present territorial borders. Store-door Delivery W CuswmJ3rs Without Central Warehousing The record further shows that many small bottlers would, absent territorial restrictions, have access to huge metropolitan markets in which thousands of soft drink retailers not serviced by central warehouses for other food items presently obtain Coca-Cola and allied products on a store-door delivered basis. While chain stores, large independent supermarkets, and convenience stores serviced by ware­ houses are important to (76) the bottlers (RPF 325), in the largest metropolitan areas, as much as 73 percent of the bottlers' volume is 77 (Tr.delivered on a store-door basis. 2309-09). Although a part of this 75 The record shows that in the early 60s' a group of Coca-Cola bottlers on the west coast entered into a cooperative agreement for the purpose of experimenting with warehouse delivery of Coca-Cola in cans through the Safeway, Lucky, and Purity food chains. According to respondents and the bottlers, these experiments failed, While the bottlers were apparently dissatisfied with the way some store managers at the retail outlets were merchandising the product after the chains had purchased it, it is arguable that merchandising decisions in individual retail outlets such as the number of shelf-facings a product will receive in a store, are not misplaced if left to the discretion of the retailer who buys the product for resale.

Nor does the record show that these experiments demonstrate the failure of split delivery. The tests lasted several years during which time, participating customers who picked up Coca-Cola in cans from. the canning plant and backhauled it in their own trucks to their respective warehouses (Tr, 2623) still purchased bottled Coca-Cola from individual ~ottlers for store-door delivery to the chain store retail outlets, (RPF 110). Although respondents claim such delivery is infeasible, these early tests with central warehousing involved split delivery to a significant degree, Nor are they of purely historical significance, As we noted previously, the record shows that the Alpha-Beta chain receives at its warehouse canned Coca-Cola which it obtains from the Los Angeles Coca-Cola bottler and transships into the territories of neighboring bottlers; Alpha-Beta, however, still purchases Coca-Cola in returnable bottles delivered store-door by the bottlers in the territories in which its retail stores are located, (Tr. 2584-85, 2588, 2634, 2650-51; See also Tr. 357~76). 76 On the large size bottles, for example, store-door delivery may be more efficient than central-warehouse delivery, (Tr, 3438-39), 77 For example, in New York about 70 000 accounts purchase Coca-Cola; in Washington, D. , 15 000 accounts purchase it; 6 400 accounts purchase it in Richmond; 12 000 accounts purchase it in San Antonio, Texas; and 3 000 accounts purchase it in Wilmington, Delaware, In contrast, the small bottlers in Annapolis, Maryland, and Charlottesville, Virginia, service 1 335 and 1 375 accounts respectively. The bottler in Westminster, Maryland, services a total of about 1 000 accounts and the Coatesville, Pennsylvania, bottler services about 1,200 accounts, The Dover Delaware, bottler services a total of about 650 accounts while the neighboring bottler in Wilmington services about 300 Mom and Pop stores alone, (RPF 225), Respondents correctly assert that in the absence of exclusive territories, a big bottl!!r may compete intrabrand for the relatively few accounts the small bottlers presently serve, but the potential for a small bottler to expand might include thousands of accounts now foreclosed whim, THE COCA-COLA CO. ET AL. 661 517 Opinion volume may represent sales in refillable bottles, the exclusivity which will remain undisturbed, store-door delivery of nonrefillable containers in these metropolitan areas still holds substantial opportuni­ ties for growth and market expansion by small bottlers. 4. Terriwrial Restrictwns Foreclose Fair lntrabrand Competitwn in the Sale of Coca-Cola and Allied Prodw:ts Packaged in Nonrefillable Bottles and Cans While larger and potentially more fertile markets would, absent the restraint, open to small bottlers, we acknowledge that the free market provides no assurance that all of respondents' bottlers will compete effectively or thrive in an unsheltered environment.78 Nevertheless we reject respondents contentions that the antitrust laws embody a pledge to protect (77) small bottlers from competitive risk and that The Coca-Cola Company may redeem the pledge by keeping captive the demand side of a market which includes soft drink retailers from coast to coast and indirectly the consuming public served by those retailers.

Respondents simply misapply the thrust of our decisions and those appellate tribunals directed toward the preservation of small business. The precedents respondents invoke for example, involve situations in which anticompetitive behavior, such as monopolization 79, merger activity 80, exclusive dealing-type franchise arrangements which impede independent franchisees from purchasing supplies from their franchisor s competitors 8!, boycotting82, and discriminatory pricing or promotional practices in violation of the Robinson-Patman Act83, were condemned by this Commission or the courts. These cases, to the extent they implement the concern of Congress for the preservation of small business, demonstrate a strong public policy to protect small business not from open and fair competition, but from unfair anticompetitive acts and practices of larger rivals. In essence, the decisions concerned with small business problems issued by appellate tribunals share in common the singular proposition that small business may be shielded from the unfair, anticompetitive practices large firms sometimes 78 We noted previously that many small bottlers have been locked into territories that are so small they cannot generate enough volume to support an independent bottling operation, Thus the number of small bottlers forced to merge with or sell out to neighboring bottlers is substantial. The record shows that the survival of the independent small business unit of production and distribution of Coca-Cola and allied products is, under respondents territorial system, threatened in numerous instances by inescapable inefficiency due to their confinement in small territories. (Tr, 615, 895-901), 79 S. v, Aluminum Co, of America, 148 F.2d 416 (2d Cir. 1945), 80 C, v, Procter Gamble Co. 386 V,S, 568 (1967); Brown Shoe Co. v. S" 370 V,S, 294 (1962); Naticna1 Tea Co" 69 F, C. 226 (1965), 81 Brown Shoe Co. 62 F. C, 679 (1963), 82 Fashion Originatms ' Guild v. C" 312 V,S. 457 (1941), 83 C. v. Fred Meyer, Inc. 390 V,S. 341 (1968), Opinion 91 F. G; ­ employ against them, but the case law is equally clear that the antitrust laws afford small business, as a class, no license to engage in anticompetitive market segmentation activity for its own protection (See Topco, supra); nor do we find in them sanction for the patronage of respondents' anticompetitive activities, presumably on behalf of the small bottlers.

The threat of competitive confrontation between- large and small independent bottlers is alone not enough to justify the imposition of a restraint preventing consummation of the threat. To conclude otherwise would, in our judgment, clearly represent a novel departure from free market principles; neither the precedents cited by respon­ dents and the judge nor the circumstances revealed in this record lend any support for it. (78) 5. Independent Bottlers v. Tlw DorYl€stw Bottlers' Subsidiari€s of Tlw Coca-Cola Company (DBS) Intervenors' most vigorous objections to an order lifting territorial restrictions concern the competitive imbalance which they assert might exist between the independent bottlers and The Coca-Cola Company DBS operations. Intervenors contend that The Coca-Cola Company, as a dual-distributing trademark licensor, may have critical advantages over its bottlers, unrelated to the efficiency of its syrup-producing and bottling integration. We have carefully considered the evidence relating to the competitive imbalance which intervenors perceive. Respondents' Access to Confidential Trade Information The record shows that the The Coca-Cola Company, in the course of its business as a trademark licensor and syrup supplier, acquires detailed and sensitive, competitive information about each of its bottler s business operations. For example, during routine quality control inspections of bottling plants, respondents can obtain access to the type of information which may reflect a bottler s production capacities and competitive capabilities, including the innovations and methods a bottler may employ to reduce his production-line or plant costs, or increase his capacity and competitive potential.84 In addition 84 While it may have, in the past, been beneficial to the overall efficiency of the Coca-Cola and allied product bottling network for each bottler within it to pass on useful commercial infonnation to other bottlers, such efficiencies may not be possible and may have to be sacrificed to some degree in the interest of preserving the free market. these circumstances, as Bork has observed, a manufacturer: . '. is much less likely to make known to others in the system any particularly successful selling or manufacturing techniques it devises if there is a substantial possibility that such techniques will be used to take business away from it, (Bork The Ruw of Rea.wn and tM Per Se Co7u;ept: Price Fi:&ing and Market Dit..;.Mn 75 Yale L, J, 373 (1966) at 439-40­ If a firm cannot' be protected by a market division agreement from the danger of the " free rider," it probably would, in its own interest, cut off the infonnation flow.(Id. at 445). Certainly an independent bottler would not (Continued) THE COCA-COLA. CO., ET AL. 663 517 Opinion 0,­ a bottler must obtain The Coca-Cola Company s approval before using new, previously unauthorized types of packaging, and as the record amply demonstrates, packaging decisions in this industry can be a vital aspect of a bottler s marketing strategy. (RPF 253). The Coca-Cola Company would have advance knowledge of, and the right to approve new packaging innovation and, unlike its independent bottlers, could begin to react to a bottler s innovation before it was actuaUy introduced into the market. (79) The sensitivity of this type of information is further evidenced by the fact that The Coca-Cola Company itself "argued persuasively for in ca'Yt'Wra treatment iri this proceeding of similar types of commercial data in order to prevent it from falling into the hands of syrup company competitors and their bottlers (Tr. 486-87; CX 1- in ca'Yt'Wra), and complaint counsel agreed the request was not "wholly without merit." Thus it appears that, as trademark licensors, respondents' relationship with their bottlers is more in the nature of a fiduciary than a competitor. (Tr. 487). Under these circumstances, we believe it would be inequitable and unfair to ignore intervenors' concern that the bottling operations of the trademark licensor may easily obtain access to competitively sensitive information and may easily exploit the" advantages this would give them.

Our order will, therefore, require respondents to safeguard the information they acquire from their independent bottler licensees in the course of respondents' business as trademark licensors and syrup suppliers. Disclosure of this type of information to those of respon­ dents' employees involved in or responsible for the production and sale of finished soft drinks will be prohibited. Respondents will also be enjoined, pursuant to paragraph II F. of our order, from enforcing or aiding in the enforcement of plant inspection provisions incorporated into licensing agreements, which respondents have approved or consented to, between any bottler and the bottler s sub-bottlers, term sub-bottlers, or temporary bottlers. (See, e. CX 20B Para. (f), CX 35C Para. (f), CX 36D Para. (f)). This provision is necessary to prevent exploitation and competitive abuse of information which may be acquired by bottler/licensors, and should impose no undue (801 burden upon respondents' quality control program in view of the fact that respondents retain the right to inspect the sub-bottlers' facilities, and the further fact that respondents customarily conduct inspections of voluntarily, yield his production and marketing ideas and strategies to any direct competitor. It is true that much of the data, such as monthly sales to chain stores and planned promotions, is apparently supplied voluntarily by independent bottlers to The Coca-Cola Company. This type of data presumably could be withheld if it were in the bottlers' interest to do so. Other types of information, however, concerning the independent bottlers' plant facilities and production capacities, for example, would as a consequence of the trademark licensing relationship, be extremely difficult for the bottler to withhold or safeguard. ...

Opinion 91 F.

every plant at which soft drinks bearing their trademarks are produced. The injunction will, however, include an exception which will allow respondents to continue to fill orders for finished packaged soft drinks from licensed Coca-Cola and allied product bottlers pursuant to agency bottling or canning agreements.

Divestiture Stipulation The bottlers also" contend, however, that divestiture of integrated bottling operations by respondent Coca-Cola and other integrated syrup companies would be the only effective way of dealing with unrestrained dual distribution in this industry. (See Ans. Br. by Coca- Cola Bottling Company of Los Angeles et al. Para. at 33-34). The Coca-Cola Company and its bottlers have negotiated a stipulation pursuant to which respondent Coca-Cola has agreed not to object to a divestiture order, provided the Commission enters equivalent relief against seven other syrup suppliers.85 Yet we cannot, in the abstract endorse a proposal premised on remedies in cases not yet adjudicated; nor are we, on the record before us, prepared to decide a general rule of vertical divestiture, including situations possibly involving de novo entry or toehold entry by acquisition, which could rule out the potential efficiencies of integration as well as the potential procompetitive effects it may have in this industry. Certainly, nothing in this record demonstrates that such measures would be appropriate. To the contrary, although we reserve judgment on cases involving other syrup companies now pending before the administrative law judge, it is not inconceivable that vertical integration by acquisition or de novo entry into bottling might be justified by a smaller syrup company attempt­ ing to piece together a nationwide bottler distribution network to compete with (81) the industry giants such as The Coca-Cola Company and Pepsico. (See Text at 12, fn. 14 supra). Nor is there sufficient independent record basis for extraordinary divestiture relief against respondents in this proceeding.

85 The divestiture stipulation is limited by the following caveat: , .. the other seven manufacturers of nationally branded soft drink syrups against whom the Commission now has complaints pending and their subsidiaries and affiliates are required by the Federal Trade Commission, in Docket Nos, 8853 (Crush International, Limited), 8854 (Dr. Pepper Co.), 8856 (Pepsico, Inc.). 8857 (The Seven-Up Co.), 8858 (The Royal Crown Co,), 8859 (Nationallndustries, lnc.) and 8877 (Norton Simon Inc.), to divest and do divest all other bottling, canning, and distributing operations. , . . (Tr. 4104-05; Stipulation No. 10, Docket Binder 1~, filed June 2:7 1975), 86 Intervenors note that The Coca-Cola Company has the capacity to exploit its resources as a dual-distributing syrup producer for the purpose of increasing the market share of its bottling subsidiaries, Citing the testimony of John H. Ogden, Executive Vice-President of Coca-Cola U, A" intervenors point out that respondent' s Chicago DBS has since 1975, incurred losses because its management viewed that territory as "an area for investment spending, believing that leadership in a market ultimately moves to a profitable position," (Tr.~1), While intervenors emphasize that the profits of the DBS operations constitute approximately 1 percent of The Coca-Cola Company s pre­ tax profits and that it might be economically feasible for respondent Coca-Cola to operate its DBS on a break-even basis for an extended period of time, no evidence was adduced at the trial that respondent provides deep-pocket (Ccmtinued) . .

THE COCA-COLA CO., ET. AL. 665 517 Opinion Or- .

The burden of establishing the necessity of ancillary relief, such as divestiture or supply limitations in the nature of convenants not to compete, rests with the party asserting the need for such protection. Frequently, this burden is assumed by government counsel in cases in which it appears that ancillary relief is necessary in the public interest to preserve the competition fostered by the primary remedies of antitrust litigation Ford Motor Co. v. 405 U. S. 562 (1972); Balfour Co. v. F. T. 9 S&D 26, 56 (7th Cir. 1971); lIuria Bros. &',Co. Inc. v. 8 S&D 615 (3rd Cir. 1968), by intervenors who seek to protect interests they believe will not be adequately represented by the parties Ford Motor Co. v. , supra or by plantiffs in private treble damage actions. Brunswick Corp. v. Pueblo Bowl- Mat, Inc. , supra 775. But as the court observed in Papercraft Corp. v. 9 S&D 530, 536 (7th Cir. 1973), " . divestiture orders have included special provisions designed to insure the survival of the divested business, but in each instance the supporting findings demonstrated the need for a special protective provision." No evidence of need for ancillary divestiture relief has been adduced in this case. (82) c. Request for Further Hearings on Relief The bottlers further argue that a remand on issues of relief is necessary if their interests are to be adequately protected. Intervenors had fair notice that issues of relief were before the judge. Intervenors were afforded every opportunity to participate in the development of the trial record; they were authorized to offer documents into evidence, to call witnesses to testify in their behalf, and to examine or cross-examine witnesses called by respondents and complaint counsel. Along with the 26 bottlers, representing a cross section of the industry, seven executives of The Coca-Cola Co., the former President of Shasta Beverages, the Executive Vice-President of K-S Canning Corp. , a contract canner, and two representatives of canned ice tea producers appeared at the trial. As the record shows, these witnesses addressed issues of relief as well as issues of liability; and intervenors' counsel present at each hearing session, were free to pursue with these witnesses lines of inquiry relevant to questions of relief at intervenors discretion. The fact that the record now fails to support intervenors theories concerning the need for ancillary protection, in all respects, is subsidies to its Chicago DBS or supports below-cost sales. A1J intervenors must fully appreciate, even independent bottlers sometimes operate unprofitably, (Tr. 1475), Evidence such as this hardly establishes the necessity for drastic ancillary divestiture relief. Nor are intervenors' other theories, analogizing this situation to vertical merger cases, supported by this record.See Bru'n8'UJ'U;k Cclrp, v. Pueblo Bowl-O-Mart S. --- , 1977-1 Trade Cases, Para. 255; Elfman Mows, lru;, v. Chrysler Cclrp,(567 F.2d 1252) 1977-2 Trade Cases, ~ 61,650. Opinion 91 F.

Of.. ­ no basis for concluding that a remand of this proceeding is either justified or necessary.

Several intervenors also request consolidated, industrywide hearings on relief. While this seems to assume the liability of other respondents in proceedings involving rule of reason inquiries, stiU pending before the administrative law judge, the contention that such hearings are necessary is otherwise lacking in merit. For even if we momentarily assume, for the sake of argument, the liability of respondents in proceedings before the judge, it would not necessarily follow that uniform, industrywide remedies or uniform anciIIa:ry relief would necessary or appropriate. To the contrary, fact records different from the record here before us may weII justify different remedial provisions.87 Under the circumstances, (83) we believe that a remand on issues of ancillary relief in consolidated, industrywide relief hearings is unwarranted and would unnecessarily delay final disposi­ tion of these cases.

C. NATIONAL ENVIRONMENTAL POLICY ACT (NEPA) CONSIDERATIONS Finally, respondents contend that an Environmental Impact State­ ment (EIS) must be prepared by the Commission, pursuant to the National Environmental Policy Act (NEPA) (42 U. C. 4321 et seq. before a final order is entered in this matter. Our rules provide that a formal EIS need not be filed in our adjudicatory proceedings. (16 CFR 82(d)). The issue has never been squarely before a court. A t the trial, respondents caIIed two experts on the ecological impact of beverage containers. These witnesses concluded that the returnable refillable bottle may be an ecologicaIIy sound form of packaging. At two trips, for example, the refillable bottle has roughly the same impact on the environment (including water use, solid waste genera­ tion, air pollutants, water-borne wastes, and energy effluents) as the nonrefiIlable, nonreturnable bottle (Tr. 3801, RX 126Z20-23); at four trips it has about the same impact as the conventional steel can (Tr. 3801); at five trips its impact is about that of an aluminum can which recycled at an 80 percent rate. (Tr. 3802). Evidence also suggests that 87 We note that respondents have vigorously opposed consolidated industrywide hearings. (Opposition of Respondents to Motion For Consolidation, Dkt. Binder 8855, 1-.'3-1 filed August 12, 1971; See Order Denying Motion To Consolidate Proceedings, filed September 29, 1971). 88 See Gifford-Hill Co, v. 389 F. Supp, 167 (D. C, 1974) affd 5Z3 F.2d 730 (D.C. Giro 1975); Mobil Oil Corp, v. 1977-2 Trade Cases, ~61 632, The Council on Environmental Quality, whose interpretation of statutory requirements under NEPA is entitled to great deference(Warm Springs Dam Task Force v. Gribble, 417 V,S. 1301 1310 (1974); JicarW.a Apcrehe Tribe v. Murtcn 471 F,2d 1275 (9th Cir. 1973); Environmental Defense Fund v, Ten1!e88ee Valley Authority,468 F,2d 1164, 1177- , (6th Cir, 1972)), has concluded that the Commission s Rule 1.82(d), exempting adjudicatory proceedings from the EIS requirement, was consistent with NEPA. (Brief for Defendents-Appellees, Addendum, Gifford-Hill mpra, .,..

THE COCA-COLA CO., ET AL. 667 517 Opinion the refillable bottle with a trippage of 10 is more energy efficient than steel or aluminum cans or glass, nonrefillable bottles.89 Under certain circumstances, then, the returnable, refillable bottles may be ecologi­ cally superior to other package forms used by the bottlers. (84) We noted previously, however, that territorial restrictions on the use of returnable, refillable bottles will not, for reasons heretofore stated be lifted by our order. As a result, a bottler will have even greater incentives than exist now to promote reusable bottle sales, since ' increase in the intrabrand market share of this container will increase the bottler s soft drink volume protected by exclusivity. Nor would the free rider" problem (See text at 32-35 supra) deter a bottler from actively promoting, as some have in the past, any economic and ecological benefits of the package in his ~rritory. (RX 60, Tr. 2499). Use of refillable bottles is unlikely to change significantly as a result of the relief entered in this proceeding.

Beyond these observations based on the record compiled at the trial below, we note that NEPA was not designed to stymie the Commis­ sion s enforcement activities which seek to redress violation of the antitrust laws. Nor does NEPA its legislative history, or its precedential case law require the preparation of a formal EIS in this proceeding. We find no basis for respondents' claim that these requirements apply to the adjudicatory activities of law enforcement agencies.9O (85) VII. CONSIDERATION OF THE DISSENTING STATEMENT OF COMMISSIONER CLANTON In his dissenting statement, Commissioner Clanton recommends that this matter be remanded to the administrative law judge for hearings which would further explore the competitive effects of the challenged practice. The rationale which leads to this recommendation is believe, erroneous in two basic respects. First, it misapprehends complaint counsel' s burden of proof. Second, it concludes that there is not enough evidence in the record to decide this case. The dissent contends this case cannot be resolved without a full 89 Little evidence was adduced concerning the environmental impact associated with litter attributable to the use of one-way, throw-away containers, 90 We note further that NEPA requires preparation of an EIS only in connection with "major Federal actions significantly affecting the quality of the human environment. , , ," 42 U. C, 4332(2Xc), Based on our review of the record in light of that standard, we conclude that our order would not in any event require preparation of an EIS. Moreover, our decision, permitting respondents to continue their territorial restraints with respect to refillable bottles, , we believe, less likely to have any effect on the use of this container than any resolution of this case other than allowing respondents to continue to restrain competition in violation of Section 5. NEP A, however, does not immunize respomlents' unlawful activities, for environmental reasons, from the Commission s law enforcement processes, Opinion 91 F.

0,-" ­ structural analysis of the soft drink syrup producing and bottling industries.91 (Dissent p. 2). At the bottling level where the restraint precludes intrabrand competition, the undertaking recommended would include surveys of each territory to determine (1) Coke s market share, (2) concentration trends over time, (3) barriers to new entry and barriers to effective competition, (4) the degree of product differentia­ tion, and (5) market performance and profitability of fountain syrups. In addition, after further discovery, a "rigorous analysis" of profitabil­ ity at the manufacturer and bottler levels would also be required. (Dissent p. 21).92 If, after examining the (86) structural characteristics of numerous territories, it can be inferred that bottlers possess substantial market power, this might justify "striking down the restrictions irrespective of any countervailing benefits." (Dissent 10).

The critical question raised by the dissenting opinion is whether complaint counsel, having demonstrated that respondents' vertical restraint adversely affects competition in the soft drink industry, were also required to adduce evidence showing the effect of the restraint on market shares and concentration, entry barriers, product differentia­ tion, or the profits of the manufacturer and bottlers. We think not. We do not dispute, as the dissent suggests, that statistical data and market structure evidence might be. relevant, and in some instances necessary, to determine the competitive effects of vertical restraints. 91 It should be noted the territories imposed by Pepsico, Inc" challenged in a companion matter, are not necessarily co-extensive wtih the territories of the Coca-Cola bottlers in the "corridor area" and arguably would have to be separately surveyed, 92 Responding to a note in theHarvard Law Review, the dissent suggests that "the indicia for measuring market power are familiar concepts which do not present unmanageable problems of proof in a rule of reason case." (Dissent The burden of the inquiry proposed by the dissent should not be underestimated. We know from experience in merger cases involving one or two geographic markets and similar structural inquiries that such litigation is complex, extremely time-consuming, and burdensome to all parties. In this instance, the trial would begin again from scratch, extensive pre-trial discovery would be required, and the structural characteristics both of the syrup industry and the bottling level in numerous territories, each the equivalent of a separate geographic market, would have to be surveyed and litigated. In all likelihood, years of costly trial would ensue; this, we believe, is unnecessary.

93 It should be noted that if it were established that a bottler had "substantial market power," the dissent would apply what is virtually a per se standard of iIIegaUty. It is unclear, however, whether this per se rule would prevent a new entrant, for example, at the syrup producing level, from offering exclusive territories in piggybacking situations standardsto bottlers with marketforpower.applyingWhether marketa perpowerseevidencerulealoneneedwould benotsufficientbeto decided.meet the rigorous 9~ Economists sometimes use the terms "market power" and "monopoly power" interchangeably.See Scherer Industrial Market Structure andEcO'lWmic PerfOT'l1lance (1970) at 10. The dissent notes "that "The Commission determined correctly that proof of monopoly power or unrestricted market power, as argued by respondents, is an unnecessary prerequisite to a finding that a particular restraint is unreasonable." (Dissent p. 14). In recommending a remand to adduce market power evidence, it is unclear whether the focus would be to determine the bottlers' market power in light of all the brands they may piggyback or just the Coca. Cola brand. (The dissent's analysis of the effect of piggybacking on entry barriers at the syru~producing level, which we previously noted supra at 49, fn, 45, is entirely consistent with our conclusion that piggybacking also tends to concentrate brands at the bottling level.) This is important because the focus of the remand sought by the dissent seems to be Umited to a determination of the market power of Coca-Cola, Yet this would ignore the fact that piggybacking tends to concentrate brands and the power to price piggybacked brands in the hands of the strongest bottlers in a territory, , THE COCA-COLA CO., ET AL. 669 517 Opinion 0,- Yet, the adverse effects of the restraints in this case have been established without such evidence. The (87) record demonstrates that respondents' territorial policy (1) impedes competition in the types delivery services bottlers offer to their customers 95, (2) prevents efficient bottlers from fully exploiting their competitive advantages and (3) prevents retailers located within the territories of less efficient bottlers from purchasing Coca-Cola and allied products from efficient sources of supply. Moreover, the record leaves little doubt that the practice substantially lessens both intrabrand and interbrand price competition. The testimony of the President of The Coca-Cola Company, other officials of the company, and bottlers, which virtually constitutes admissions of substantial adverse competitive effects clearly supports these findings.

Such anticompetitive effects have indeed been inferred in cases where the evidence was much less direct than it is here. Relying on S. v. Continental Can Company, 378 U.S. 441 (1964), the Commission recently noted that concentration and market share data alone suffice to establish illegality of a merger in the absence of convincing proof to the contrary. Jim Wallet COTp. Dkt. 8986, issued December 20, 1977 (90 F. C. 671). The evidence in this record that respondents' practice substantially lessens price competition is, we believe, more compelling than would be the case if such effect were inferred from concentration and market share data alone. The dissent, moreover, would require a more detailed evaluation of pricing patterns in the industry. As we recently noted, however The absence of discernible effect on pricing or the lack of small company failures attributable to a merger can be given little weight. . . . At best, such effects are difficult to measure particularly if prices are already at non-competitive levels."97 Adverse effect on price is, however, (88) clearly discernible in this record. Similarly, evidence, which the dissent would require, showing whether new entrants have made inroads into the various territories would be of limited utility in rebutting the evidence of anti competitive effect reflected in the record.

As we mentioned, a prima facie case was established through the 95 The reservations expressed in the dissent about the demand for central warehousing would certainly surprise the witnesses who testified in this proceeding, While the pros and cons of this method of distribution were hotlycontested, the demand for the service was never seriously disputed, (RPF 88-91, Ans, Br. 55), 96 The dissent without elaboration would dismiss, as "anecdotal " testimony reflecting the adverse effects of the practice provided by these witnesses in response to questioning by the judge and by counsel. We are unable to depreciate such testimony in this manner. (See Text at 27 en. 28; 47-52 supra. 97 Jim Walter Corp., supra, 98 Id, at 45-46. As we stated RSRin Corp" 88 F. C. 797 (1976): even proof of low entry barriers, ,. can be at most of slight exculpatory value in the face of probable anticompetitive effects, since all it suggests is that such effects may be smaller or short lived, not that they are unlikely to occur. (at 289).

Furthermore Jim Walter clearly indicates that the burden of proof rests with respondent to show whether "new finns have eroded the market position of the industry leaders, (Supra- at 46), . . ). ... . Opinion 91 F.

testimony of the President of The Coca-Cola Company and other industry witnesses. Thus as Commissioner Clanton, writing for a unanimous Commission in the Jim Walter case, correctly observed, it is respondent's burden, once a prima facie case has been established based upon other evidence of anticompetitive effect, to provide exculpatory evidence "pertaining to the structure, history, and probable future of the asphalt and tar roofing industry sufficient to overcome the presumption (arising from concentration and market share data alone) that the merger threatens a subst~ntiallessening of competition. (Jim Walter Corp. , supra at 42 et seq. In this instance we believe it was unnecessary for complaint counsel to resort to further statistical data to confirm the testimony upon which a prima facie violation of Section 5 had been established, and to the extent such data may have been relevant to the defense, it was respondents' burden to adduce it.

We agree with the dissent that " . one s preference for one kind of competition over another (price competition v. nonprice competition) should not automatically condemn" respondents' practice, although we believe that emphasis on the tendency of respondents' practice to impede price competition (89) is not misplaced. A practice which lessens price competition touches the core of the free enterprise system. The Supreme Court has described the price mechanism as critical" and "sensitive. S. v. Container Corp. of America 1969 Trade Cases at 86 413. In S. v. Socony- Vacuum Oil Co., Inc. the court, citing Handler Federal Antitrust Laws-A Symposium (1931), noted that this aspect of competition is "the central nervous system of the economy." Thus the alleged justifications for a practice which substantially lessens price competition requires, and in this instance (90)has received, the closest scrutiny.99 The dissent reexamines these justifications and raises a number of 99 It is likely that the recommended surveys of various territories might disclose that some bottlers have substantial market power" while others may not, and it is unclear what outs:ome the dissentwould regard as appropriate in these circumstances. If a certain percentage of the bottlers surveyedposssessed substantial market power " would this justify striking down the restraint as it applies to the others "irrespective of any countervailing benefits?" If not, would the restraint be illegal only when it applies to bottlers with "substantial market power?" The surveys called for by the dissent might reveal, for example, that Coca-Cola bottler A has "substantial market power but not Coca.Cola bottler B. Would the restriction then be lawful as applied to bottler B and unlawful as to bottler A? This would leave bottler A with "substantial market power" free to compete while bottler B would remain restrained. Yet in order to dissipate the power of bottler A presumably bottler B should be free to compete in bottler A's territory. If the dissent is concerned about the restriction only when it serves to "protect" bottlers with "substantial market power," then it would seem to follow that a bottler without such power might remain protected from intrabrand competition in his territory, since the dissent' s per se rule based on market power might not apply to him. He would, however, apparently be free to compete in the territory of a bottler with market power, at least until intrabrand competition dissipates that power, Once the power has been dissipated, the market power per se rule would no longer apply, and the restraint might again be lawful as it would presumably be for similarly situated bottlers who were found not to possess market power, It might then be necessary to monitor each bottlers power periodically to determine when, where, and how long intrabrand competition might be needed to prevent the build-up of "excessive market power, ...

THE COCA-COLA CO., ET AL. 671 517 Opinion questions concerning whether investments by bottlers operating in exclusive territories enhance or impair competition, whether exclusive territories facilitate interbrand competition by enhancing availabili­ ty 100 and by inducing greater demand for soft drink products, whether ter~itorial restrictions facilitate advertising by the bottlers which promotes interbrand competition, and whether obstructions to intrab­ rand competition are necessary to maintain product quality. The issu-es now raised in the dissent, concerning which it finds the record inadequate were previously raised by respondents in the form of affirmative arguments in justification of these restraints. In each instance, the evidence respondents relied upon in support of their contentions that the restrictions were reasonably necessary to maintain at current levels the interbrand viability of Coca-Cola and allied products were carefully examined by the Commission ami found wanting. 101 Thus the dissent reviews the alleged relationship between the restraint, capital formation, and interbrand competition, and apparently unable to conclude from the record that investments in exclusive territories enhance interbrand competition, are necessary to the continued competitive viability of Coca-Cola and the allied product or that respondents' capital formation arguments, and the evidence relating to them, justify the restraint. (Dissent at 7). If the burden rests with respondents to establish this defense, as we believe it does 102 the dissent seems to confirm our finding that respondents have not, in this respect, adequately justified their restraint. The dissent also examines respondents' arguments to the effect that exclusive territories facilitate level pricing by the bottlers and thus intrabrand competition by enhancing (91) availability." While it is apparently not disputed that market penetration based on level pricing results in price discrimination which "means. . . that some Coca-Cola is provided at less than its actual cost and some is priced above" it is suggested that the cost differentials may not be substantial enough to warrant price differences (Dissent at 9) and that accounting ancl billing costs may exceed cost differentials or may not justify an expanded price list.103 Such assumptions, while perhaps a plausible rationale for level pricing in some instances, are largely contrary to 100 As we noted previously, it is not possible on this record to state definitively that exclusive territories enhance output to a greater degree than would lower prices resulting from intrabrand competition. (Text at 81, fn. 81, supra). 101 The dissent, while coming close to accepting respondents' arguments that restraintpromotes interbrand competition, does not actually do so, (Dissent at 12). 102 Sandura, supra; Srmp-On Tools, supra; Jim Walter, supra, 103 The same analysis might also hold true even if the bottlers have "substantial market power. ...

Opinion 91 F.

evidence cited in this opinion and elsewh~re in the dissent. (See Dissent at 14, .In. 27). 104 If prices more accurately reflected actual costs as a result of intrabrand competition, efficient retailers would be in a position to pass any cost savings on to consumers. Under the present system, however, level pricing deprives efficient retailers and their customers of the benefits of such competitive options. 1O5 Thus the dissent does not seem to resolve the issue of whether r~spondents have adequately justified the restraint because it aids market penetration by permitting level pricing.

The same is true of respondents' advertising and " free-rider arguments. Judgments concerning the nature of the advertising for Coca-Cola were based on a thorough review of the advertising respondents or bottlers elected to introduce into the record. We certainly do not believe complaint counsel were obligated to provide the evidence upon which a more "systematic and thorough review of Coke advertising" might have been made. (Dissent at 10). Nor do we believe complaint counsel can reasonably be expected to offer evidence showing both the efficiency of the promotional methods respondents now employ and "the relative efficiency of manufacturer (and presumably retailer) advertising versus bottler advertising. " (Dissent at 10). (92 J The court in GTE was concerned that the Schwinn rule declaring exclusive territories per se illegal might result in "a shift to less efficient methods of obtaining the same promotional effects. (GTE supra at 71 901, fn. 25). In applying the rule of reason to these restraints, the court thus opened for further inquiry, on a case-by-case basis, the possibility that promotional methods employed in exclusive territories may be more efficient than alternative promotional methods absent the restraint. The court did not hold, however, that the mere assertion of such efficiency by a respondent without supporting facts was enough to require what the dissent acknowledges to be the very difficult" process of exploring the "relative efficiency" of alternative methods available in unrestricted markets. The dissent renders no judgment either about the efficiency of the promotional methods respondents now employ or about the promotion­ al effects they obtain. 1O6 This is not surprising since respondents did 104 Yet even if bookkeeping costs justified the continuation of level pricing, the evidence shows that some bottlers are more efficient than others. Thus the level price of some bottlers is likely to be lower or more competitive than the level price charged by others.

105 With respect to brand availability, the dissent does not contend that if a demand exists for these products at prices which reflect actual costs, the market is unlikely to supply them at competitive prices. 106 The dissent notes that one such effect might be that the promotion of the Coke brand has conferred substantial market power upon respondents and their bottlers by successfully differentiating their product, but neither this nor any other brand-enhancement effect can be measured based on the evidence in this record. The dissent does not otherwise dispute our analysis which shows that the "free-rider" problem is unlikely to reduce the bottlers' incentives (Continmd) . . . , THE COCA-COLA CO., ET AL. 673 517 Opinion Of.

not, in asserting this defense, adduce evidence which would allow such judgments to be made. Consequently, even if complaint counsel had produced evidence of the efficiency of alternative methods of promotion, respondents' failure to establish the efficiency of their own methods would have made, as noted in the dissent fine-tuned assessments" of relative efficiency very difficult. The dissent's consideration of respondents ' quality-control justifica­ tions focuses only upon the alleged relationship between territorial restrictions and quality control in distribution.lo7 The issue here seems to be whether the (93) Commission may independently evaluate the alleged quality-control justification to determine " whether, assum­ ing some justification for the limitation can be shown, their operation is reasonably related to the needs which brought them into being. White Motor Co., supra 372 U.S. 253, 271 (1973) (Brennan, J. concurring).

According to the dissent, any effort to determine whether the restriction is excessively restrictive "implicitly second-guesses Coke belief that obstructions to intrabrand competition are needed to maintain the high quality of its product." (Dissent at 12). The situation is the same, though the reverse of the problem considered by the Third Circuit in American Motor Inns, supra. In Amel1,ean Motor Inns the court was concerned that plaintiff's lawyers, in a private treble damage action under the Sherman Act, might "conjure up some method of achieving the business purpose in question which would result in a somewhat lesser restriction of trade." Our concern here is in protecting the public interest against the imaginations of entrepren­ eurs and lawyers who are students of antitrust practice and skillful advocates in defending trade-restrictive conduct. This requires us to assess the competitive effects of respondents' action. lo8 The Commis­ sion is not bound to accept Coke s belief that obstructions to intrabrand competition are needed when the consequences of its action are excessively trade-restrictive.lo9 Further, respondents did not substan­ tiate, and there is really no basis on this record for measuring, the efficiency of territorial restrictions, including, for example, the costs associated with policing and enforcing them, as a quality-control to advertise desirable information about price, quality, and services to their customers, (GTE supra at fn. 25; text at 33-34 supra), 107 The relationship between territorial restrictions and quality control in manufacturing is not considered in the dissent, 108 Certainly no firm is omniscient. The Coca-Cola Co" for example, (1) doubted that carbonated soft drinks could be bottled successfully and sold for home consumption (See Text 10 fn.supra;12 RPF 28) and (2) agreed to sell its syrup at a setprice, in perpetuity, without provision for market conditions which might increase the cost of the ingredients used to make the syrup (SeeThe Coca-Cola Bottling Co,v, The Coca-Cola Co. supra), 109 See Coors, $upra. While the combination of price fixing, territorial restrictions, and customer restrictions were found to be per se iJIega! in Coors the Commission nevertheless fully considered and found merit in some of the quality-control arguments advanced by Coors, ...... ,.

Final Order 91 F.

monitoring mechanism. Thus it is unclear whether alternatives, such as an open dating system which might allow the market to monitor product age, would be "less efficient." 110 (94) Under these circumstances, the language of the Supreme Court in Northern Pacific Ry.Co. noted earlier in this opinion, is appropriate here. The court in that case emphasized that the antitrust laws rest: on the premise that the unrestrained interaction of competitive forces will yield the best allocation of our economic resources, the lowest prices, the highest quality and the greatest material progress, while at the same time providing an environment conducive to the preservation of our democratic institutions. But even were that premise open to question, the policy unequivocally laid down by the Act is competition. Nc;rthern Pacifw Ry. Co., supra.

CONCLUSION Other arguments of the parties, intervenors and amici not specifical­ ly addressed in this opinion have been considered and found to without merit. Having reviewed the record in its entirety, and all of the arguments advanced by respondents in support of these restraints and having found no adequate justification for the substantial adverse affects these restraints are having on competition in this industry, we conclude that territorial restrictions on the sale of finished Coca-Cola and allied soft drink products are unreasonable restraints on trade, and constitute unfair methods of competition in violation of Section 5 of the Federal Trade Commission Act.

An appropriate order is attached.

FINAL ORDER This matter having been heard by the Commission upon the appeal of complaint counsel from the initial decision, and upon briefs and oral argument in support thereof and in opposition thereto, and the Commission, for reasons stated in the accompanying opinion, having granted the appeal:

I T IS ORDERED, That the initial decision and order of the administra­ tive law judge be, and they hereby are, vacated, and the findings of fact and conclusions of law contained in the accompanying opinion of the Commission be, and they hereby are, adopted as the findings and conclusions of the Commission in this matter. Accordingly, the following cease and desist order is hereby entered: (2) 110 The dissent invokes what seems to be a "rule-of-plausibility" which would virtually end the evaluation of an alleged justification upon the lISSertion by a respondent of a plausible link between the restraint and some legitimate business purpose, , THE COCA-COLA CO., ET AL. 675 517 Final Order Or- .

ORDER IT IS ORDERED That the following definitions shall apply in this order:

A. Allied products - the soft drink products of . The Coca-Cola Company, other than "Coca-Cola " including Sprite, Fresca, Fanta, Tab and Mr. PiBB, among others;

B. Bottler - any individual, partnership, corporation, association, or other business or legal entity which purchases respondents' syrups or concentrates for use in the manufacture and sale, primarily at wholesale, of finished soft drink beverages; C. Central warehousing - a method of distribution in which soft drink products are received at a storage facility and either resold or delivered to retail outlets or wholesalers; D. Concentrate - the basic soft drink ingredients, either dry or liquid, to which sugar is added to prepare a syrup; - facts, data, statistics, or E. Confidential commercial information other material which concern the business of licensed Coca-Cola or allied product bottlers including, but not limited to, trade secrets customer lists, plant equipment or production capacities, or syrup and concentrate purchases obtained by or available to, respondents pursuant to, or as a result of, any agreement, understanding, or provision of a trademark license, and which could, if disclosed to a competitor, cause substantial harm to the competitive position of the bottler from whom the material was obtained; once F. Nonrefillable - a special container designed to be filled only with finished Coca-Cola or allied soft drink beverages; G. Post-mix syrup - a soft drink ingredient which is used in fountain-dispensing or vending equipment and which is usually sold by bottlers and other wholesalers in steel tanks. A typical post-mix system draws one ounce of syrup from a tank, usually having about a fivegallon capacity, and mixes it at the point of sale with five ounces of carbonated water to produce finished soft drink beverages; (3) H. Pre-mix system - a system which draws from a tank, usually having about a five-gallon capacity, a finished serving of a soft drink product containing both syrup'and carbonated water pre-mixed " to produce finished soft drink beverages;

1. Soft drink products - nonalcoholic beverages and colas, carbonat­ ed and uncarbonated, flavored and nonflavored, sold in bottles or cans or through pre-mix or post-mix systems or the like; J. Syrup - a mixture of ingredients in liquid form which, when mixed with carbonated water, becomes a finished soft drink product. ...,.

Final Order 91 F.

IT IS FURTHER ORDERED, That The Coca-Cola Company; Coca-Cola Bottling Co. (Thomas), Inc. ; Coca-Cola Bottling Works (Thomas), Inc. and Coca-Cola Bottling Works 3rd, Inc., and the officers, agents representatives, employees, successors, and assigns of each respondent directly or through any corporate or other device, in connection with the advertising, merchandising, offering for sale, and sale or distribu­ tion of soft drink products, including syrups and concentrates, in or affecting co~merce, as "commerce" is defined in. the Federal Trade Commission Act, do forthwith cease and desist from directly or indirectly:

A. Attempting to enter into, entering into, continuing, maintain­ ing, enforcing, or renewing any contract provision, combination understanding, or agreement to limit, allocate, or restrict the territory in which, or the persons or class of persons to whom, licensed Coca-Cola or allied product bottlers may sell or distribute post-mix syrup or finished soft drink beverages packaged in pre-mix containers or in nonrefillableB. Imposingbottlesor attemptingor cans. to impose any limitations or restrictions respecting the territories in which, or the persons or class of persons to whom, bottlers may sell or distribute post-mix syrup or finished soft drink beverages packaged in pre-mix containers or in nonrefillable bottles or cans. (4 C. Refusing to sell, threatening to refuse to sell, or impairing sales to any bottlers, operating pursuant to a license consented to, granted by, approved by, or ratified by The Coca-Cola Company; Coca-Cola Bottling Co. , Inc. ; Coca-Cola Bottling Works, Inc. ; or Coca-Cola Bottling Works 3rd, Inc., for the duration of the license, anything used in the manufacture and sale of soft drink products, including, but not limited to, syrups and concentrates or the container in which they are sold, or otherwise in any way penalizing any such bottler because of the territory in which, or the persons or class of persons to whom, the bottler sells or distributes post-mix syrup or finished soft drink beverages packaged in pre-mix containers or nonrefillable bottles or cans.

D. Refusing to deliver all of a licensed Coca-Cola or allied product bottler s order for syrups, flavoring, or concentrates because the bottler has made, or intends to make, sales of post-mix syrup or soft drinks packaged in pre-mix containers or nonrefillable bottles or cans to customers outside of the territory granted to the bottler, or because the bottler has made, or intends to make, such sales to customers within the territory granted to the bottler, with knowledge that the ... .

THE COCA-COLA CO., ET AL. 677 517 Final Order customer has transshipped or will transship such soft drinks outside of the territory.

E. Impeding, hindering, or preventing, either directly or indirectly, the methods, including, but not limited to, central warehouse delivery, by which licensed bottlers may distribute Coca-Cola or allied products Provided, however that respondents may (1) establish quality stan­ dards, including standards for the rotation of Coca-Cola and allied products inventories in the central warehouse and at retail delivery locations, irrespective of whether the soft drinks are redelivered from a warehouse or delivered directly to the retail outlet by a bottler; (2) require the bottlers to use a uniform container dating system so that bottlers and retailers will recognize the date without reference to a code; (3) require the bottlers to be responsible, directly or indirectly, for the maintenance of such standards of quality; and (4) require each bottler to place an identification mark of origin on each bottle, bottle cap, or can for the purpose of monitoring compliance with such quality control standards.

F. Enforcing or aiding in the enforcement of any contract provision, agreement, or understanding providing for entry into or examination of the plant and facilities of any independent bottler by another independent bottler. (5 III I T IS FURTHER ORDERED, That respondents shall provide for the protection of confidential commercial information acquired from bottler licensees of Coca-Cola or allied product brands as follows: A. Access to or use of confidential commercial information obtained by respondents, their officers, employees, or agents concern­ ing the production, packaging, distribution, promotion, or sale of Coca- Cola or allied product brands by any licensed bottler shall be restricted to those of respondents' officers, employees, or agents who are neither involved in nor responsible for the production, marketing, promotion or sale of finished soft drink products by respondents' bottling or canning operations, divisions, subsidiaries, or affiliates. B. Such officers, employees, or agents who receive, process, or evaluate package-approval requests; process or fill syrup or concen­ trate purchase orders; conduct on-site inspections of independent bottling plants and facilities; or receive or review confidential commercial information obtained from any independent Coca-Cola or allied product bottler in the course of carrying out the provisions of any soft drink trademark licensing agreement, shall refrain from making any such confidential information available to, or communicat­ ing or discussing any such information with, any person involved in or 678 , FEDERAL TRADE COMMISSION DECISIONS Final Order 91 F. C...- . responsible for the production, marketing, promotion, or sale of finished soft drinks by respondents' bottling or canning operations, divisions, subsidiaries, or affiliates.

process, or have C. Such officers, employees, or agents who receive, access to confidential information concerning the business of individual independent Coca-Cola or allied product bottler licensees, shall refrain from suggesting, influencing, or making recommendations to any person concerning the production, distribution, marketing, promotion or sale of finished soft drinks by respondents' bottling or canning operations, divisions, subsidiaries, or affiliates Provided, however that this provision shall not apply to respondents' officers, employees, or agents who receive, review, or evaluate data, information, or statistics only in aggregate form or quality control inspection reports which include such information as bacteriological tests, water analyses, water inspection checks, orcarbonation and syrup content tests, sanitation bottle washing solution analyses, so long as such reports do not also contain information concerning the bottler s plant equipment, produc­ tion capacity, or similar types of confidential commercial information. employee, or agent who(6)D. Respondents shall provide each officer, receives, reviews, or has access to confidential information as set forth in subparagraphs A. through C. above with a copy of this order and explanation, in writing, of the restrictions this order imposes on access to and the use of such information.

above shall not apply (1) to data or E. Subparagraphs A. through C. information which is in the public domain or which has entered the public domain from a source other than respondents or their officers from employees, or agents; or (2) to transactions involving orders licensed Coca-Cola or allied product bottlers for finished canned or bottled soft drink products prepared by any respondent for a bottler pursuant to an agency canning or bottling agreement. IT IS FURTHER ORDERED, That within sixty (60) days from the date The Coca-Cola Company receives service of this order, it shall service a copy of this order upon all bottlers of its soft drink products. I T IS FURTHER ORDERED, That respondents The Coca-Cola Company; ; Coca-Cola Bottling Works Coca-Cola Bottling Co. (Thomas), Inc. (Thomas), Inc. ; and Coca-Cola Bottling Works 3rd, Inc., shall forthwith THE COCA-COLA CO., ET AL. 679 517 Final Order distribute a copy of this order to each of their subsidiaries and operating divisions.

I T IS FURTHER ORDERED, That respondents The Coca-Cola Company; Coca-Cola Bottling Co. (Thomas), Inc. ; Coca-Cola Bottling Works (Thomas), Inc. ; and Coca-Cola Bottling Works 3rd, Inc., shall notify the Federal Trade Commission at least thirty (30) days prior to any proposed change in the corporate respondents, such as dissolution assignment, or sale resulting in the emergence of a successor corporation, the creation or dissolution of subsidiaries, or any other change which may affect compliance obligations arising out of the order.

IT IS FURTHER ORDERED, That each respondent shall, within sixty (60) days after service upon it of this order, file with the Commission a report in writing setting forth in detail the manner and form in which it has complied with this order.

Chairman Pertschuk did not participate in the consideration of this matter. Commissioner Clanton dissents.

Complaint 91 F.T.C.