International Telephone & Telegraph Corporation

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

IN THE MATTER OF INTERNATIONAL TELEPHONE & TELEGRAPH CORPORATION, ET AL.

FINAL ORDER, OPINION, ETC. IN REGARD TO VIOLATION OF SEC. 5 OF THE FEDERAL TRADE COMMISSION ACT AND SEC. 2(a) OF THE CLAYTON ACT AS. AMENDED Docket 9000. Complaint November 1974-Final Order July, 1984 For reasons set forth in the Commission s Opinion, this final order reverses the ALJ' initial decision, denies complaint counsel's appeal, grants appeal of respondents and dismisses the complaint charging a New Yark City conglomerate and its wholly-owned baking company subsidiary with alleged violations of ferlerallaw. The complaint had alleged that the baking company had attempted to monopoliz the white bread productsales market in five geographic areas and caused competi. tive injury in those markets by, among other things, engaging in predatory or discriminatory pricing practices for significant periods of time. Appearances For the Commission: Jerry A. Philpott, Renee S. Henning, Stephen E. Nagin, Richard Malatt, Gilda E. Rodriguez, Marimichael O. Skubel and Allen M Hickey For the respondents: John H Schafer S. William Livington, Jr. David J. Cynamon, Bruce D. Soklerand Jane H Chalmers, Covington & Burling, Washington, D.C. and Craig D. Walley and Gordon Thomas New York City.

COMPLAINT The Federal Trade Commission, having reason to believe that the above-named respondents have violated and are now violating Section 5 of the Federal Trade Commission Act, as amended (15 D. 45), and subsection (a) of Section 2 of the Clayton Act, as amended by the Robinson-Patman Act (15 D. C. 13), and believing that a proceeding by it in respect thereof is in the public interest, hereby issues this complaint charging as follows:

Definitions 1. For the purpose of this complaint, the following definitions shall apply:

280 Complaint (a) Bread shall refer to white pan bread and bread type rolls and related products but not to specialty breads. (b) Wholesale bakers are bakers of bread which sell at wholesale to other establishments including groceries, restaurants, hotels and institutions. Bakeries owned and operated by grocery chains which distribute the bakeries' products through the chains' own retail grocery stores are not wholesale bakers. (2) Respondents 2. Respondent International Telephone and Telegraph Corporation ("ITT") is a corporation organized and existing under the laws of the State of Delaware with its principal place of business located at 320 Park Avenue, New York, New York. ITT is a conglomerate engaged directly and/or through subsidiaries in numerous and diverse businesses including, among others: the development, manufacture, distribution servicing and operation of electronic and telecommunication equipment and other industrial and consumer products; life, fire and casualty insurance; processing wood pulp; mining; business and consumer services; and the manufacture and distribution of food products and automotive parts. In 1973, ITT had sales of approximately $10.2 bilion, making it the ninth largest domestic corporation in terms of sales, and assets of approximately $10.1 billion, making it the eighth largest in terms of assets. 3. Respondent ITT Continental Baking Company, Inc. ("ITT Continental"), is a corporation organized, existing and doing business under the laws of the State of Delaware, with its offce and principal place of business located at Halstead Avenue, Rye, New York. ITT Continental is a wholly-owned subsidiary ofITT, which acquired it in 1968 from its predecessor, Continental Baking Company. ITT Continental is engaged in the manufacture, sale and distribution of bread, cakes, snacks such as potato chips, frozen prepared foods, candy and ingredients for the bakery industry. In 1972, ITT Continental's net sales were approximately $865 milion. In 1972, ITT Continental would have been ranked as the 165th largest domestic corporation in terms of sales if it were not owned by ITT. 4. ITT controls, approves and benefits from the practices of ITT Continental.

Jurisdiction 5. In the course and conduct of their businesses, respondents are and for a substantial period of time have been engaged in selling bread throughout various States of the United States, and have caused bread baked in various states to be shipped to purchasers in various other states. Thus, respondents are and at all times relevant Complaint 104 F.

herein have been engaged in a continuous and substantial course of trade in commerce, as ucommerce" is defined in the Federal Trade Commission Act and the Clayton Act, as amended by the Robinson- Patman Act. (3) COUNT I Alleging violations of Section 5 of the Federal Trade Commission Act.

Trade and Commerce 6. The relevant product market for purposes ofthis complaint is the baking, sale and distribution of bread by wholesale bakers. The wholesale baking of bread is the largest and most significant submarket of the bread baking industry. Relevant geographic markets consist of regional and local markets in the United States where bread is sold by ITT Continental, and aggregates thereof. 7. The baking, sale and distribution of bread is a substantial industry in the United States. In I972, approximately I5.6 billon pounds of bread including specialty breads were sold in the United States with a retail value of approximately $3.3 bilion. Of this, approximately 87.5% was baked and sold by wholesale bakers. 8. Concentration in the nation s bread industry is increasing. In 1963 the four largest bakers of bread and specialty breads accounted for 24% of the nation s bread sales; the eight largest accounted for 36%. By 1971 the four largest had increased their share to 30% and the eight largest to 40%. Concentration among the nation s wholesale bakers is higher. In 1963 the top four wholesale bakers of bread and specialty breads accounted for 31 % of the total sales of the nation wholesale bakers. By 1970 this had risen to over 37%. Between 1958 and 1972 the number of bakery plants declined by 45%. From January 1972 to mid-I973, 43 wholesale baking firms and 80 wholesale bakery plants closed.

9. Shares of total sales at the national level understate concentration in the bread industry. Due to its short shelfli!" and high transportation cost, most bread is sold within 150 miles of the bakery. Concentration is significantly higher in local and regional markets than in the nation as a whole, with 4-firm concentration ratios substantially exceeding 50% in many such markets. 10. Barriers to entry into wholesale bread baking are high. Significant start-up costs are associated with a new bakery, particularly with respect to establishing routes and obtaining suffcient shelf space. Because bread shelf space of grocers is limited, the introduction of a new brand means displacement 0((4) established brands. Established multiplant wholesale bakers with interstate treasuries histori- 280 Complaint cally have met or bettered price concessions and promotions offered by bakers trying to expand or enter into their markets. Moreover since bread is a homogenous product which is differentiated on the basis of sales and advertising expenditures, a new entrant who seeks to alter consumer preferences for established brands is placed at a substantial cost disadvantage.

11. ITT Continental is the world' s largest bread baker. Its net sales of bread in 1973 were approximately $475 milion. This represents a 47% increase over 1968, the year in which ITT acquired Continental Baking Company. In 1970 ITT Continental' s net sales of bread were almost half again as large as the next largest wholesale baker, Campbell-Taggart Baking Company. ITT Continental bakes bread in 48 bakeries located in 30 states and the District of Columbia and distributes it through in excess of 300 depots to areas in 46 states occupied by 70% of the population of the United States. ITT Continental's share of the nation s bread industry has grown steadily. 12. ITT Continental' s market shares in regional and local markets are substantially higher than its share of national sales. ITT Continental has over 75% ofthe wholesale baker market in 4 regional or local markets; over 60% in 8; over 50% in 17; over 40% in 37. 13. ITT Continental is also the nation s largest producer of snack cakes, which it markets under the "Hostess" label. In 1973, ITT Continental's snack cake sales were $315 millon. In some areas snack cakes and bread are distributed by the same ITT Continental routes. 14. ITT Continental is one of few multistate wholesale bakers. In competition with other multistate wholesale bakers such as Camp. bell-Taggert Baking Company, American Bakeries Company, and Interstate Brands Corporation, as well as those local wholesale bakers who are stil in existence, ITT Continental sells and distributes bread in most of the populous portions of the United States. (5) 15. ITT Continental's wholesale baking business was built and conducted and is now conducted on a national basis. From its headquarters in Rye, New York, ITT Continental centrally purchases raw materials for the production of bread, as well as supplies, equipment and other needs. Advertising, both national and local, is prepared and placed in media by ITT Continental' s headquarters. ITT Continental at all times maintains control, directly from its headquarters or through various regional offces, over the activities of its bakeries such control being exercised with respect to, among other matters planning and sales objectives, national accounts, the area in which and the price at which each bakery is permitted to sell, standards of products maintained by said bakeries, all but minor repairs to plants and equipment, personnel policies, and funds collected and disbursed by said bakeries.

Complaint 104 F.

Acts, Practices and Methods of Competition 16. ITT Continental and ITT jointly set for ITT Continental its sales and profit budgets and policies, including sales growth goals, which force ITT Continental managers to behave in a predatory manner. Through a requirement for detailed and systematic reports from ITT Continental with respect to all of its significant business decisions, ITT agrees upon, and/or concurs and acquiesces in most of the acts practices and methods of competition engaged in by ITT Continental including most or all of those hereinafter set forth. 17. Respondents plan to achieve dominance in wholesale baking in all relevant geographic markets, seeking to attain their objective by engaging in the acts, practices and methods of competition hereinafter set forth, most of which have been and are directed to specific geographic markets.

18. With the intent to lessen, hinder or restrain competition and/or to attain monopolies in wholesale baking in one or more relevant geographic markets, beginning at least as early as 1952, and continuing thereafter up to and including the date ofthis complaint, respondents and the predecessor of ITT Continental, Continental Baking Company, have engaged in various acts, practices and methods of competition including, but not limited to, the following: (6) (a) Acquisitions of a number of wholesale bakers; (b) Since 1962, acquisitions of at least three independent bakers in violation of a Federal Trade Commission order forbidding such acquisitions;

(c) Sales of bread below their cost or at predatory prices for substantial periods of time in various geographic markets; (d) Subsidization of sales below cost or at predatory prices in various geographic markets by sales at higher pripes in less competitive geographic markets;

(e) Discriminations in price, directly or indirectly, between purchasers of bread of like grade and quality;

(f) Paying for services or facilities furnished by selected customers and/ or furnishing services to selected customers to induce such customers to maintain and/or increase purchases of respondents' bread; (g) Furnishing allowances, discounts and other things of value to customers or customers' employees for shelf and/ or other merchandising space in grocery stores;

(h) Systematic concentration of advertising and consumer and trade promotions in connection with the sale of bread at predatory prices in various geographic markets; and (i) Use of its dominant position in the sale of snack cakes and profits llu!'ItNATIONAL TELEPHONE & TELEGRAH CORP. , ET AL. 285 280 Complaint resulting therefrom to increase bread sales and to subsidize losses thereon. (7) Effects 19. The effects of the acts, practices and methods of competition described in Paragraphs 16, 17 and 18 are, among others, to: (a) Impair the ability of wholesale bakers to compete with ITT Continental;

(b) Induce small wholesale bakers to discontinue the production and sale of bread;

(c) Aggravate the trend towards concentration in wholesale baking; (d) Deter new entrants and raise barriers to entry into the wholesale baking industry;

(e) Deter existing wholesale bakers from undertaking competitive initiatives;

(f) Inhibit growth of existing wholesale bakers; (g) Substantially hinder, lessen, eliminate, injure, destroy and/or foreclose actual and potential competition in wholesale baking; and (h) Increase the probabilty that respondents will attain a monopoly in the wholesale baking industry in each and all relevant geographic markets. (8) Violations 20. The acts, practices and methods of competition alleged in this complaint constitute attempts by respondents to monopolize and injure competition in the wholesale baking industry in relevant geo- Section 5 of the Federal Tradegraphic markets in violation of Commission Act.

21. The acts, practices and methods of competition alleged in this complaint constitute unfair methods of competition or unfair acts or practices by respondents in violation of Section 5 ofthe Federal Trade Commission Act.

COUNT II Alleging violations of subsection (a) of Section 2 of the Clayton Act as amended by the Robinson-Patman Act.

22. Each of the allegations in Paragraphs 1 through 5 herein are incorporated in this Count II as though set forth in full. 23. In the course and conduct of their business in commerce, ITT and ITT Continental, now, and for many years past have been in substantial competition with other corporations, partnerships . individuals and firms engaged in the manufacture, distribution and sale of bread.

286 FEDERAL TRADE COMMI lU" u v'v,-- Initial Decision 104 F. 24. Respondents, in the course and conduct of their business in commerce, as above described, have for a number of years discriminated and now are discriminating in price, directly or indirectly, between different purchasers of bread, by selling bread oflike grade and quality to some of such purchasers at substantially higher prices than to other of such purchasers.

Among the methods by which respondents discriminate between said purchasers is the granting of discounts of 5 to 12% and more off the list or regular prices to some customers in a trading area and denying such discounts to competing customers in the same trading area, and to customers in other trading areas. (9) 25. The effect of such discriminations in price as alleged herein may be substantially to lesse!\competition or tend to create a monopoly in the lines of commerce in which respondents and respondents' customers are respectively engaged; or to injure, destroy or prevent competition with respondents or with purchasers therefrom who receive the benefit of such discriminations.

26. The aforesaid acts and practices of the respondents constitute violations of subsection (a) of Section 2 ofthe Clayton Act, as amended by the Robinson-Patman Act.

INITIAL DECISION BY MILES J. BROWN, ADMINISTRATIVE LAW JUDGE MAY 1, 1981 INTRODUCTION The Federal Trade Commission issued its complaint in this matter on November 26, 1974 (mailed December 16, 1974), charging respondents, International Telephone & Telegraph Corporation ("ITT") and ITT Continental Baking Company, Inc. ("Continental") with unfair methods of competition or unfair acts or practices in violation of Section 5 ofthe Federal Trade Commission Act (15 C. 45) and with discriminations in price in violation of Section 2(a) ofthe Clayton Act as amended by the Robinson-Patman Act (15 C. 13(a)). (2) Initial Pleadings More particularly, the Commission charged that Continental, until its acquisition by ITT in 1968, and Continental and ITT since that time, have engaged in certain practices with the intent to lessen hinder or restrain competition or to attain monopolies in wholesale baking in one or more relevant geographic markets. The Commission 280 Initial Decision alleged that the practices challenged included, but were not limited , the following (See Complaint U 18): (a) Acquisitions of a number of wholesale bakers; (b) Since 1962, acquisitions of at least three independent bakers in violation of a Federal Trade Commission order forbidding such acquisitions;

(c) Sales of bread below their cost or at predatory prices for substantial periods of time in various geographic markets; (d) Subsidization of sales below cost or at predatory prices in various geographic markets by sales at higher prices in less competitive geographic markets;

(e) Discriminations in price, directly or indirectly, between purchasers of bread of like grade and quality;

(0 Paying for services or facilities furnished by selected customers or furnishing servces to selected customers to induce such customers to maintain or increase purchases of respondents' bread; (g) Furnishing allowances, discounts and other things of value to customers or customers' employees for shelf or other merchandising space in grocery stores;

(h) Systematic concentration of advertising and consumer and trade promotions in connection with the sale of bread at predatory prices in various geographic markets; and (i) Use of its dominant position in the sale of snack cakes and profis resulting therefrom to increase bread sales and to subsidize losses thereon.

The Commission alleged that the effects of the challenged practices are to (Complaint U 19):

(a) Impair the ability of wholesale bakers to compete with Continental;

(b) Induce small wholesale bakers to discontinue the production and sale of bread;

(c) Aggravate the trend towards concentration in wholesale baking; (d) Deter new entrants and raise barriers to entry into the wholesale baking industry;

(e) Deter existing wholesale bakers from undertaking competitive initiatives; (3) CO Inhibit growth of existing wholesale bakers; (g) Substantially hinder, lessen, eliminate, injure, destroy and/or foreoclose actual and potential competition in wholesale baking; and (h) Increase the probability that respondents wil attain a monopoly in the wholesale baking industry in each and all relevant geographic markets.

The Commission also alleged that respondents have discriminated Initial Decision 104 F. in price between different purchasers of bread by sellng bread oflike grade and quality to some of such purchasers at substantially higher prices than to other of such purchasers, the effect of which may be substantially to lessen competition or tend to create a monopoly in the lines of commerce in which respondents and respondents' customers are respectively engaged (Complaint n 24).

In the notice of contemplated relief that accompanied the complaint, the Commission advised that, if the charges alleged in the complaint were sustained, it may, among other provisions, order respondents to divest wholesale baking assets, including bakery plants and other facilities in one or more areas, and require the licensing of brands or trademarks for respondents' bread under such terms as are consistent with the restriction of competition in the relevant markets (Complaint at 11).

In their answer (fied March 21 1975) respondents admitted certain background facts, but denied all allegations as to any violation of Section 5 of the Federal Trade Commission Act or Section 2(a) of the Clayton Act, aB amended. Respondents also put forth as defenses certain prior actions of the Commission inconsistent with the theory of the complaint in Docket 9000 and the fact that certain challenges to their practices should be barred because of undue passage of time. Respondents further asserted that respondent ITT was not responsible for Continental's actions before September 1968, and that after that date it, being a corporation separate and distinct from Continental, has not engaged in the manufacture, sale, or distribution of bread. As to the charges of price discrimination, respondents raised the "cost justification" and "meeting competition" defenses. Finally, respondents asserted that the Commission lacked authority, on the basis of the allegations of the complaint, to secure the relief set forth in the notice of contemplated relief.

Sanctions Order In the course of extensive post complaint discovery, complaint counsel sought detailed information on which they could recompute cost figures for indivicjual bread varieties, cost figures which they considered essential to sustain the "sales below cost" allegations of the complaint. Respondents, stating that such an investigation was, in their opinion, beyond the scope of the Commission s complaint, on December 10, 1976, (4) refused to produce any further documents to support complaint counsel's efforts to develop a different cost accounting system from that which was employed by Continental during the years covered by the complaint.

Following complaint counsel's application for sanctions available under Section 3.38 of the Commission s Rules of Practice, and the INRNATIONAL TELEPHONE & TELEGRAPH CORP., ET AL. !j!J 280 Initial Decision Administrative Law Judge s order of February 16, 1977, granting the application for sanctions, an order was issued March 22, I977, imposing sanctions on respondents for failure to produce all documents deemed essential to recompute Continental's costs. The "sanctions order" provided that, for specific geographic markets during specific time periods, complaint counsel were deemed to have established that Continental sold bread below cost no matter how cost is measured. The sanctions order also provided that Continental could not introduce into evidence or otherwise rely upon the documents which it failed to produce and that it could not object to complaint counsel' use of other relevant material and reliable evidence to demonstrate that Continental made sales below cost in the designated geographic markets for the time periods specified.

Respondents were granted leave to fie a limited appeal to the Commission on the policy question of whether it would be more appropriate for the Commission to seek district court enforcement of the subpoena as it related to the "withheld cost documents" rather than impose the sanctions for withholding the documents. On June 29 1977, the Commission issued its order denying respondents' application for review.

Following the Commission s denial of the application for review respondents made a continuous effort to require complaint counsel to accept their tender of the withheld documents and to obtain an order vacating the sanctions order. Complaint counsel steadfastly refused to accept those documents and respondents ' motions were denied, (See orders dated November 7, I977; September 28, 1979, April 15, I980). Respondents ' Discovery As complaint counsel's pretrial discovery neared completion at the end of 1977, it was agreed that respondents' discovery directed to Continental's wholesale baker competitors would be held in abeyance until the conclusion of the case-in-chief. Although broad subpoenas were issued to these nonparties in order to ensure preservation of documents, return dates thereon were suspended. After the case-inchief was concluded, limited returns were negotiated with four of Continental's wholesale baker competitors.

Hearings Adjudicative hearings commenced June 19, 1978 and 81 days of trial were held intermittently until October 15, 1980. (5) Complaint counsel' s case-in-chiefused 53 trial days, respondents' answering case used 16 trial days, complaint counsel' s rebuttal case used 11-1/2 days and respondents surrebuttal case used 1/2 day. Ten of these hearing dates were devoted entirely to the handling of documentary material. Initial Decision 104 F. After the close of complaint counsel's case- in-chiefon June 7, 1978 respondents fied numerous motions to dismiss the complaint in certain particulars, all of which were denied (Order Denying Respondents' Motion to Dismiss the Complaint as to International Telephone and Telegraph Company ("ITT"), dated September 21, 1979; Order Denying Respondents' Motion to Dismiss the Seattle phase of the Case, dated September 24, 1979; Order Denying Respondents' Motion to Dismiss the Complaint as to the Price Discrimination Allegations (Count rn, dated October 5, 1979)). Earlier motions to dismiss the Denver and Minneapolis phases of the case on the grounds that the evidence would be stale were also denied (9182; order dated January , 1979).

Time to File Initial Decision On October I7, 1980, the Administrative Law Judge closed the record for the receipt of evidence, although the record was reopened on November 4, 1980, to permit the receipt into evidence of certain documents that complaint counsel acquired from respondents after October 17, 1980. On October 27, 1980, the Commission granted ajoint motion of counsel for extensions of time in which to fie proposed findings and reply briefs. The Commission also extended the time for fiing the initial decision until February 27, 1981. On December 23 1980, the Commission issued an order, on complaint counsel's motion granting the parties a further extension of time in which to fie proposed findings and reply briefs and extended the time for fiing the initial decision until March 14, 1981. Proposed findings were fied on January 9, 1981, and reply briefs were fied on February 6, 1981. On March 4, 1981 , the Commission granted the Administrative Law Judge s request for an extension oftime until April 10, 1981, in which to file the initial decision, and on April 7, 1981, further extended that time to May 1, 1981.

Motions and Abbreviations Any motions appearing on the record not heretofore specifically ruled upon either directly or by the necessary effect ofthe conclusions in this initial decision are hereby denied. The proposed findings and conclusions submitted by counsel supporting the complaint C'CCPF" and counsel for respondents C'RPF") have been given careful consideration and to the extent not adopted by this decision, in the form proposed or in substance, are rejected as not supported by the evidence, as argumentative, as immaterial or based on material not in evidence. Other abbreviations used in this initial decision are: CX - Commission Exhibit; RX - Respondents' Exhibit; CC Mem - Complaint counsel' s memorandum oflaw in support ofCCPF; RBr - (6) Respond- 280 Initial Decision ents' brief in support ofRPF; CC Reply Br. - Complaint counsel' s reply brief. and R Reply Br. - Respondents' reply brief. Collateral Proceedings Most of the factual and legal issues presented in this case arise from the facts and circumstances relating to Continental's pricing practices in five general market areas; i.e. Denver, Minneapolis-St. Paul Los Angeles, Northern California and Cleveland-Akron. Continental's practices have been, and in some cases, continue to be, the subject of private litigation. In Denver, a District Court' s determination that Continental had violated Section 2(a) ofthe Clayton Act, as amended was upheld by the Court of Appeals Continental Baking Co.v. Old Homestead Bread Co. 476 F. 2d 97 (10th Cir. 1973), cert. denied 414 U.s. 975. In Minneapolis-St. Paul, private damage suits fied by Zinsmaster Baking Co., Pan 0 Gold Co., and Tender Krust/Creamy Crust were settled before trial (CX 299F , 300AJ, Zinsmaster Baking Co., et al. v. ITT Continental Baking Company, No. 4-8 Civ. 96 (D. Minn). In Los Angeles, a suit brought by several of Continental's wholesale baker competitors was being prepared for trial during the final hearings in this matter. See Wm. Inglis Sons, Inc. v. ITT Continental Baking Co. (S.D. Cal. No. 78-3604). In Northern California a District Court's jury verdict for plaintiff was reversed by the district judge on a motion by Continental for judgment notwithstanding the verdict Wm. Inglis Sons, Inc. v. ITT Continental Baking Co. 461 F.Supp. 410 (N.D. Cal. 1978), appeal pending in the Court of Appeals for the Ninth Circuit. In Cleveland, the representatives of the creditors of Laub Baking Co. have fied a treble damage suit against Continental a matter that was being prepared for trial during the final hearings in this matter Melamed v. ITT Continental Baking Co. No. 74-960 (N.D. Ohio). With respect to a sixth market, Seattle, complaint counsel have alleged that Continental engaged in certain anticompetitive practices, although no price discrimination conduct or sales below cost are alleged. Continental' s practices in Seattle were also the sub, ject of a private damage suit. The Court of Appeals applied the labor exemption ofthe antitrust laws and remanded it to the District Court for dismissal Granddad Bread Inc. v. Continental Baking Co., 612 2d 1105 (9th Cir. 1979), cert. denied 49 USLW 3493 (January I2 1981).

Although the record in this matter contains certain materials collected or used in evidence in these other cases, this case is a de novo consideration of all issues, factual and legal, raised by the allegations of the complaint in Docket 9000, as modified or explained during the administrative proceedings.

In addition, the Commission investigated Continental' s practices in Initial Decision 104 F. Denver and Minneapolis, and closed both fies (7) without prejudice to the right of the Commission to reopen those investigations. (See Attachment A to Respondent' s Motion to Dismiss the Denver Phase of the Case; RX 402C). A compliance investigation in Seattle during the early I970' , relating to a 1964 Commission order prohibiting price-fixing, was closed without action (See Order Denying Respondents' Motion to Dismiss the Seattle Phase of the Case, dated September 24, 1979.

In Camera Treatment In camera treatment has been afforded many of the documentary exhibits in this case (including some in the rejected fie). Although it was generally understood that all in camera status would expire on the date of the initial decisio in this matter the record contains some exceptions and ambiguities on this subject (3999, 9166, 11466, 11429- , 11829- , 11319).

During the adjudicative hearings, counsel for Interstate moved to have Rxs 105IC-D; 1145A-J; 1146A-M; 1147A-B and 1148A-B (rejected) placed in camera with certain restrictions on their availability to employees of respondents (10382). These documents contain information about Interstate s marketing strategies and techniques, as well as analyses, and, if made public, may be detrimental to Interstate s business (See 10382). Interstate s motion was granted and camera treatment was directed until further order ofthe Commission (I0383, 10688).

First National Supermarkets ("National", the parent corporation of Pick N Pay, a Cleveland, Ohio, chain supermarket) has fied a motion for post-decision in camera protection for certain documents that it supplied to the parties for use in Docket 9000. These documents contain the formula by which Continental's wholesale prices of Pick N Pay private label bread products are determined, a formula that is used currently. National contends that this formula, as well as current prices charged by Continental, is highly sensitive business information, the disclosure of which to competitors would be detrimental to National' s business. All information which is essential to decide the issues in this matter is contained in the portion ofthe record that wil be open to public inspection. Accordingly, to protect the legitimate business interest of Pick N Pay and National in camera treatment for CXs 973 (portion), 974, 975, 976, 978, 979, 980, 981, 982A- , 984 A- 986 A- , 2602 A-B and 2601 (A- , rejected) is extended until the date ofthe Commission s decision in this matter or until otherwise ordered by the Commission.

There does not appear to be any reason to extend in camera treatment for any other document of record. Accordingly, the in camera --...... vv, s-. AL.

280 Initial Decision status for all other in camera documents expires as of the date ofthis initial decision. (8) Preliminary Summary of Facts and Contentions Although a long and detailed recitation ofthe facts annotated to the record is presented in this initial decision it seems appropriate to set forth a brief summary of the salient facts and the parties' basic contentions as an introduction to those findings. The principal producers of bread products are either wholesale bakers or chain grocery stores that operate their own bakeries. Some wholesale bakers are multiplant companies operating a large number of bakeries throughout the United States. There are also wholesale bakers that operate in a few or only in one market area. Bread bakers usually produce a great number of bread products. White bread constitutes, by far, the largest single bread product, and it is considered to be a homogeneous product. The same dough mix may be used to produce loaves of white bread, hot dog rolls and hamburger buns. These white bread products are generally referred to as "white pan bread." Some other bread products such as whole wheat, rye and pumpernickel breads are usually referred to as "variety" breads.

Wholesale bakers sell bread products to retail grocery stores, restaurants and institutions. They usually deliver bread products to the purchasers' stores or places of business. Chain store bakers, called captive bakers " distribute their bread products exclusively through their own stores. Although production costs are approximately the same for all bread producers, the chain store bakers' costs of distribution are lower than the delivery costs of wholesale bakers. The retail price of captive label bread is usually lower than the retail price bread baked by wholesalers.

Wholesale bakers sell bread under different labels and at different wholesale prices. The bread products sold under the wholesale bakers principal trade name are usually called "advertised label." Wholesale bakers sell bread to chain stores that do not have their own bakeries under the stores' label which is generally called "private label" bread. The wholesale price of private label bread is usually lower than the wholesale price of advertised label bread which permits the grocer to sell it at a retail price comparable to the retail price of captive label bread.

Retail grocers, including chains with captive bakeries usually purchase the advertised label of most ofthe wholesale bakers doing business in a market area. Bread products are displayed in the grocery store on special bread racks. This display is usually referred to as the bread table." The position and the amount of space on the bread Initial Decision 104 F. table which a grocer allots to a wholesale baker wil determine the volume of that baker s products that are sold at retail in that store. (9) Grocery stores operating captive bakeries give their captive label bread the largest space and the best position on the bread table. Grocers purchasing private label bread will give that bread product the largest space and best position on the bread table. The wholesale baker supplying a grocer with private label bread customarily will get the next best position on the bread table for its advertised label bread. The wholesale and retail prices of the various wholesale bakers advertised bread are usually identical in any particular marketing area. In fact, no wholesaler sells its advertised label white pan bread at a higher wholesale price than its competitors' wholesale price for the same product.

Notwithstanding the growth in population, the overall demand for bread has been static. Over the years the per capita demand for bread has decreased. In fact, the public s preference for white bread has decreased while the preference for variety bread has increased. In many marketing areas, competition among the chain stores has resulted in low retail bread prices. As the retail price of captive bread decreases, noncaptive chains demand lower wholesale prices for private label bread. As the retail price spread between private! captive label bread and wholesaler bakers' advertised label bread increases less advertised bread is purchased by the consumer. If one wholesale baker lowers its wholesale price of advertised label to protect its volume in advertised bread, all other wholesale bakers must match the lower price or they soon wil face a loss of volume of their sales. Another method of protecting the sales volume of advertised label as well as increasing sales volume, is by obtaining a grocer s private label business and the additional sales of advertised label products that result from obtaining favorable space and position on the bread table.

Economies of scale are important in the baking industry and wholesale bakers must operate their bakeries at approximately 80% of maximum possible production capacity (2 shifts and some overtime) in order to obtain maximum effciency.

As a result oflower wholesale prices. the profitability ofthe wholesale bakers is affected adversely. Besides lower wholesale prices, the extent of loss of profitability may also depend on a loss of volume of sales or on the percent of a wholesale baker s business that is in private label products.

Complaint counsel contend that in certain marketing areas Continental has lowered the wholesale price of private label bread or adver- .:"O hcl hrPRlJ below Continental' s fully allocated costs for long 280 Initial Decision periods of time and that as a result (IOJ other wholesale bakers usually those operating only in one marketing area, have not been able to continue in business. Complaint counsel contend that such practices have been engaged in with the intent of injuring competition and creating a monopoly in an industry that has high barriers to entry. They also contend that the differences in price that result from respondents' pricing practices constitute price discriminations which have the effect of injuring competition and creating a monopoly. For purposes of antitrust analysis, complaint counsel contend that white pan bread baked by wholesale bakers constitutes a relevant product market, and that price discriminations and below cost sellng can be measured on particular bread products within that product market (See generally CC Mem; CC Reply Br. Respondents contend that there is "chronic" excess capacity in the bread industry, usually caused by the backward integration of chain stores acquiring their own baking facilities, thus reducing the space available to wholesale bakers on the bread tables in a market. In this respect they argue that there are no significant barriers to entry into the bread baking industry. They argue that low retail prices of bread charged by the chain stores are the cause of low wholesale prices of private label and advertised label bread. They also contend that the wholesale bakers must compete for whatever space is available on the bread tables of the grocers who do not have captive bakeries and that the wholesale bakers must match the lowest wholesale price on the market. No wholesale baker can unilaterally raise prices. Respondents contend that they never beat other wholesale bakers' prices, but always responded to the prices offered to grocers by their wholesale baker competitors. They argue that it is not improper for a wholesale baker to sell below its fully allocated costs, in that sellng at prices above marginal costs contributes to the cash flow and profitability of a baker and that such pricing makes good business sense. Respondents assert that variety breads are interchangeable with white bread and that captive bread competes with wholesale baker bread. In this respect, they contend that the relevant market is all bread sold by both wholesale bakers and captive bakers. They also contend that wholesale bakers actually sell a line of bread products to grocers and that the product mix of private label and advertised label bread as well as white bread and variety breads, determines the profitability of any customer s account.

respondents contend that the withdrawal of undercapitalized wholesale bakers that do not advertise their products is inevitable under the market conditions that exist in the bread industry and that such withdrawal is not the result of their pricing activities. Finally, respondents contend that their market share of the bread industry is ).

Initial Decision 104 F. decreasing and that there is no way that they or any wholesale baker can achieve a monopoly position therein (See generally RBr.; R Reply Br. (llj Expert Testimony Both sides presented the testimony of economic experts in support of their respective theories of the case. Complaint counsel's expert witnesses were Dr. Richard G. Walsh, a professor of economics at Colorado State University (8015), Dr. Stanley E. Boyle, a visiting professor of economics and independent consultant (7892), Dr. Rodney D. Patterson, also a professor at Colorado State University (7707), and Dr. Lee E. Preston, a professor of business and public policy at the University of Maryland (1l574). Respondents' expert witness was Dr. Michael L. Wachter, professor of economics at the University of Pennsylvania (10386). Only Dr. Walsh has made specific studies of the baking industry, and he published a major work on the subject in 1963 (8017 et seq.

The expert witnesses testified on a number of issues that have been raised in this case including their opinions on market (monopoly) power, relevant product markets, excess capacity, barriers to entry and whether certain pricing practices should be considered predatory instead of being seen as predictable and reasonable business responses to industry conditions. For example, Dr. Wachter testified that in an industry where there was excess capacity and low barriers to entry pricing below fully allocated costs and above marginal costs (or average variable costs) in the short run were to be expected (10421 10453, 10650). He stated that under such circumstances, predation would not be an economically sound business choice, because it would be highly improbable that the firm could recapture the loss in profits through higher prices in the future (10483, 10625, 10655). In addition several expert witnesses expressed opinions on whether divestiture would be an adequate remedy ifthe allegations of the complaint were sustained.

For the most part, the expert testimony is general and theoretical in nature and is n!'t directed to the specific fact situations. that exist in the several local marketing areas which have become the subject of this proceeding. In addition, the various opinions seem to be inconsistent in many respects. Accordingly, except for providing some understanding of technical economics concepts, this testimony is not very helpful in assessing the legality of Continental's pricing practices under the several statutes involved. However, it would appear that such expert testimony may be of great value to the Commission when it makes its policy determination as to the future regulation, if any, of respondents' conduct in the baking industry. 280 Initial Decision Introduction to Findings of Fact Having reviewed the entire record in this proceeding, and having considered the demeanor of the. witnesses who testified at the adjudicative hearings together with the (12) pleadings, admissions, proposed findings, conclusions and briefs submitted by complaint counsel and counsel for respondents, I make the following findings of fact based on the record considered as a whole. Because the record contains much duplication offactual material, not all references possible may be included in the citations to the record, especially where the parties do not differ significantly in their understanding of such facts.

FINDINGS 0.' FACT Respondents 1. Respondent ITT is a Delaware corporation with its principal place of business at 320 Park Avenue, New York, New York. ITT is a conglomerate engaged, indirectly and through subsidiaries, in numerous and diverse businesses including, among others: the development, manufacture, distribution, servicing and operation of electronic and telecommunications equipment and other industrial and consumer products; life, fire and casualty insurance; processing wood pulp; mining; business and consumer services; and the manufacture and distribution of food products and automotive parts. In 1973 ITT had sales of approximately $10.2 bilion and it was ranked ninth among domestic corporations in the Fortune 500 in terms of sales (Complaint 2; Answer 2; CX 2051). 2. Respondent Continental is a Delaware corporation with its principal place of business at Halstead Avenue, Rye, New York. Continental is a wholly owned subsidiary ofITT which acquired it in I968 from its predecessor, Continental Baking Company. Continental is engaged in the manufacture, sale and distribution of bread, cake, snacks such as potato chips, frozen prepared foods, candy and ingredients for the baking industry. In 1972, Continental' s net sales were approximately $865 milion (Complaint, Answer 3). Ifit had been a separate entity, Continental would have been ranked 177th in the "Fortune 500" in terms of sales (CX 205I). In 1972, some of Continental' s subcompanies or divisions were Morton s Frozen Food, Pearson Candy Company, Gwaltney (meats), and Panaplus (ingredients) (9905 Dierker). 3. In 1973, Continental baked bread in 48 bakeries located in 30 states and the District of Columbia and distributed it through over 300 depots to areas of 46 states. In 1973, Continental' s sales of bread were approximately $445 milion and the I973 dollar sales of bread Initial Decision 104 F. were about 35% above its 1968 annual sales (See Complaint, Answer n 11).

4. Continental Baking Company was founded in 1924 by the consolidation of90 baking companies (463 Woodward). By 1978 Continental had 61 plants, 31 of which baked bread exclusively and 7 of which baked cake exclusively (468 Woodward). Since 1924, Continental Baking Company and its successor Continental have been the largest bread baker and snack cake baker in the (13) United States (464 472-73 Woodward). In 1974, Continental's geographic coverage for bread products was approximately 70% of the U.S. population (471 Woodward; CX 116C). It served approximately 80% of the U.S. with cake (CX 116C).

5. In the early 1970's Continental's bakeries were grouped into 7 regions which are sometimes identified by the location ofthe regional offce: New York, Atlanta, Detroit, Chicago (2), San Francisco, Los Angeles (473-79 Woodward). Each functional offcial at the bakery level bakery manager, bakery sales manager, etc. has a counterpart at the regional level to whom he reports (474, 478, 480-81 Woodward; see 3164-5 Frielink). The regional offce personnel have staff functions instead of line functions. The regional vice president has profit responsibility for his region (473 Woodward). Each regional vice president transmits a monthly management letter to the President of the Bakery Division in Rye, New York (479 Woodward). 6. Continental's Rye Headquarters purchases all major commodities (ingredients) used by its bakeries. It also prepares advertising copy and places advertising with various media. New product development and laboratory work is performed by headquarters personnel. Rye personnel also assist the various regional offces with labor negotiations, in making sales to national accounts, or multiple outlet accounts, and in making sales to certain t!institutional accounts" such as hotel chains and military installations (660-61 Meyn). 7. Continental has a very sophisticated money management system by which the cash generated at the various plants is withdrawn each day and moved to Rye, N.Y. and then to ITT in New York City (1324- 25 Willmont).

8. The day-to-day accounting is conducted at the individual plants and the results are combined or summarized at the regional offces and forwarded to Rye headquarters (490 Woodward). Since 1973, Continental has employed a Standard Cost system in its individual bakeries. Before implementation ofthis cost system, Continental developed estimated costs from a test cost program (528-29 Woodward). 9. Continental produces a full line of bread and "snack cake products (467 Woodward). Continental' s bread labels include Wonder Home Pride Butter Top, Fresh Horizons, Country Fair and Country 280 Initial Decision Style (467, 493-94 Woodward). Continental's cake labels include Hostess and Twinkies (467-68 Woodward).

Continental's Wholesale Competitors 10. After Continental, the largest wholesale baking companies in the bread industry are Campbell-Taggart, American Baking Company ("American ), Interstate Brands Corporation ("Interstate ), Flowers Bakery and Metz Baking Company (514 Woodward). Today, Continental, Campbell-Taggart and Interstate are the "primary national bakers" (CX 4 (116)). (14) 11. Campbell-Taggart was formed as an offshoot of Continental in the late 1920's. In 1974, it had approximately 70 bakeries (CXs 69R 2107). Many of these bakeries were located in the smaller cities of the South, Southwest, Midwest and Far West (CX 3 (105); see alsoCX 69R Z-20). Campbell-Taggart serves approximately 30% of the U.s. population (CX 69F). Continental and Campbell-Taggart generally do not sell in the same geographic markets, although major overlapping markets are San Francisco, Denver, Sacramento, Kansas City, Mem phis and Indianapolis (CX 3(105); see also CX 69R, Z-20). Although Campbell-Taggart performs in-depth services for the various functional departments of its bakeries, the local baking subsidiary companies operate as separate legal entities (CX 69Z-15; 1006- Z Mackaman). Campbell-Taggart' s labels include Colonial, east of the Mississippi River, and Rainbo, Manor and Kilpatrick, west of the Mississippi River (10063-5 Mackaman; CX 2107). Continental has rated Campbell-Taggart as "excellent" in product quality and management (See CX 2107).

12. American operates approximately 20 plants located generally in the Northeastern, Central and Southwestern parts of the United States (CX 2107; see also CX 9 (204)). In 1973, Continental considered that American was "retrenching" and in 1979 Continental regarded American as the weakest of its national competitors (CXs 2107; 8 (038)). American s labels include Tastee and, on the west coast, Langendorf (8855 Murray).

13. Interstate operates approximately 30 bakeries located in the Eastern, Central, Southwestern and Northwestern United States (CX 2107). In 1973, Continental considered that Interstate was "retrenching" (CX 9 (IM)). Continental generally regarded Interstate as a weaker competitor than Campbell-Taggart, but considered it to be strong in marketing approach and creativity (CX 2107; seeCXs 2 (087); 4 (133)). Interstate s labels include Milbrook, Butternut, Weber and Blue Seal (3942 Heuter; 4706 J. Walsh).

14. As late as 1972, Ward Baking Company had 15 bread producing plants but was not regarded by Continental as a strong competitor Initial Decision 104 F. (CXs 2 (087); 4 (133); 9 (204)). Apparently Ward is no longer considered to be in the bread baking business (9425- , 9376, 9532 Gase). 15. Flowers serves the Southeastern United States from approximately 2I plants. Continental regarded Flowers very highly (CX 2107). During the times relevant in this case, Flowers and Continental apparently did not compete in the sale of bread products, at least not in the six marketing areas studied in this case (See 514 Woodward; 10267 Jakaki).

I6. Metz has approximately I4 bakeries located in the North Central region of the UB. (10200 Metz). Its principal label is Old Home (10201 Metz). (15) 17. Major bakery cooperatives are Quality Bakers of America C'QBA" ), W. E. Long, and American Bakers Cooperative (8053-54 R. Walsh; 514-I5 Woodward). These cooperatives are management service associations organized on a nonprofit basis to provide the kind of services that a multiplant corporation might provide to its individual plants, including counseling services on production, engineering or accounting problems. Many independent bakers belong to cooperatives (8053 R. Walsh; seeCX 3 (108); 10266-9 Jakaki; 11199 Schaus). The cooperatives also have advertised labels. Members of QBA can use the label Sunbeam and members of W. E. Long can use the label Holsum in many marketing areas (490-91 Woodward; RX 1039). I8. It is estimated, based upon information supplied by wholesale bakers to the Commission in 1970, that the shares of the national wholesale bakers' bread and bread type rolls national market were as follows: Continental 14.2%; Campbell-Taggart 9.7%; American 7. and Interstate 5. 6% (See RX 1042 in camera). Captive Bakers 19. In addition to the wholesale bakers, many retail grocery chains operate their own bakeries and produce bread which they sell in their own retail outlets (10208-9 Metz). In some marketing areas the chain store or "captive bakers" may be the largest producers of white pan bread.

Bread Baking 20. The bakery products produced by large wholesale bakers include white bread, hamburger buns and hot dog rolls, variety and hearth breads such as whole wheat, rye, pumpernickel and raisin breads, brown and serve rolls, specialty rolls such as Parker House rolls and English muffns, French bread, bread stuffng, snack cakes doughnuts, pastries, sweet rolls and fresh pies (RX 1000C&E). White pan bread ("white bread") is white bread that is baked in a pan to the standards set by the government and it is considered to be a homo- 280 Initial Decision geneous product no matter who bakes it (499-500, 508 Woodward; see 7909 Boyle).

21. White pan bread is produced by.either a conventional process (which entails the mixing of each batch of dough separately) or a continuous mix (which permits a continuous mixing of the dough) (518-19 Woodward). Although the continuous mix process became popular in the 1960' , the present trend is back to conventional mix process because it is more flexible in the production of variety breads and because it produces white bread, the texture of which may have better consumer acceptance (519-20 Woodward; see 3358-59, 3396 Frielink).

22. In the conventional mixing process, a sponge is produced. It ferments. Other ingredients are added after (16) fermentation. The mixture is then passed through equipment to be scaled, divided and placed in a pan to be baked. In the continuous process the ingredients are fed into a mixer or processing unit. The dough continues in a steady stream without fermentation, then it is scaled, divided and placed in a pan and baked (724 Meyn). Hearth breads are baked on a hearth, that is, a solid surface. Regular pan bread is baked in a pan on a shelf oven over a flame which is applied to the pan (724 Meyn). Continental uses special pans to produce hearth type breads (725 Meyn). Continental also uses a "brew system" which enables it to produce bread with properties of the conventional mix process by the continuous mix method (3199-3200, Frielink). 23. Economies of scale are important in the baking industry and wholesale bakers must operate their bakeries at approximately 80% of maximum production capacity (2 production shifts and some overtime) in order to obtain maximum effciency (11204-5, 11223- Shaus; see 10295, 10300 Jakacki). Independent wholesale bakers are generally as effcient, if not more effcient, than the large multiplant wholesale bakers (11199-11200 Shaus).

Bread Distribution 24. Continental and other wholesale bakers sell bread principally to retail grocery stores. Wholesale bakers also sell to restaurants hotels, institutions, and other customers (467-68 Woodward). It is estimated that Continental's sales to restaurants and institutional accounts, which consisted predominantly of buns and rolls was approximately 13% of its total route sales (RX 1039; see also CX 2361 B).

25. Continental and most other wholesale bakers distribute bread though a system of route salesmen who deliver bread to retail grocery stores and restaurants. These route deliveries are made from the bakery or from depots or agencies to which bread is transported in Initial Decision 101 F. large trailer trucks from the bakery for transfer to route trucks (482- 525 Woodward; 3167 Frielink).

26. Bread is usually sold on a guaranteed sales basis, that is, the wholesaler takes back loaves that are not sold within a few days. re-These loaves are called "stales" and the service is called "stale turn." Sometimes bread is sold on a Hno return" basis at a lower wholesale price (See, e. CXs 207; 1714F; 1515 Z-95). Bread that is returned is usually placed in the bakery "thrift store" for resale at a reduced price (664 Meyn) or sold to farmers as feed (1120-21 Brout) Continental sets 48 hours as the shelflife of its Wonder label bread (726 Meyn; 1120 Brout) The driver-salesmen provide "rack service to grocery stores. They physically place the fresh bread on the grocery store shelves two or three times a week and take away the stale returns.

27. Driver salesmen are members of the Teamster Union. Most union contracts provide that their compensation be based on (I7) a certain salary plus commission. Chain store bakers usually distribute oftheir retail outlets; thebread products by semitrailer to the "door" bread is placed in the grocery shelves by store employees. Drivers of semitrailers usually belong to a different "local" of the Teamsters Union. They are paid on a straight hourly rate (726 Meyn). Generally, the costs of producing white bread are uniform among bakeries, but chain store bakers enjoy a distribution cost advantage because of the difference in delivery methods Ud. CX 22(ii); 902-03 Stolle). 28. Because ofthe relatively small area that a bakery can serve with fresh bread, bread is marketed and sold on a local basis. No major baker competes in every market. The competitive conditions, price levels, identity and number of competitors differ from marketing area to marketing area. The pricing in one market generally has no direct impact on prices in other markets. There is no dispute that the relevant geographic markets are generally local (Answer n 6). 29. Most wholesale bakers, including Continental, market an "advertised label" or "primary label" line of bread products sold under the brand name owned and usually promoted and advertised by the wholesaler. "Wonder" is Continental's principal advertised brand and s white bread sales (RX 1039 (inaccounts for the bulk of Continental' camera); 490-91 Woodward; see 649 Meyn). 30. Continental and many other wholesale bakers also sell bread packaged under the "private labels" of retail grocery chain stores. The brand name is owned by the retailer and the retailer promotes and advertises the bread. Wholesalers sell private lahel bread at lower wholesale prices than their advertised label, and private label bread is sold hy the retailer at a lower retail price than advertised lahel hread (490-98 Woodward; see 727 Meyn). A private lahel program g., 280 Initial Decision usually includes at least the popular size loaves of white bread, hot dog rolls and hamburger buns (3177-78 Frielink; 4537 A. Gordon; 4734 J. Walsh; see e. CX 1526B-).

31. In addition, wholesale bakers sometimes sell "secondary" label bread. Secondary label bread usually sells at a price comparable to private label and is often provided to independent grocers that are too small to have an economical private label program (491 , 494 Woodward; 650-51 Meyn). Some wholesale bakers will supply bread to a retail chain store packaged in a "controlled" label owned by the wholesale baker to which that chain store customer is granted exclusive use in a particular marketing area (650 Meyn; 4148 Heaps). Controlled label bread is usually sold at a private label wholesale price (Jd.

32. Bakeries owned by retail chain stores are called "captive" bakeries and the bread baked and sold in the chain s (18) retail outlets is called "captive label" bread. It is estimated that 50% ofthe white bread produced and sold in the U.s. is sold under either private label or captive label (604 Woodward).

33. Shelf position and the amount of shelf space in the retail grocery store are the most important factors in the marketing of bread, in that a better shelf position will sell more bread and larger shelf space wil sell more bread (504-5 Woodward; 690-9I Meyn; 854-55 Vail. In addition, the consumer franchise (consumer acceptance ofthe labeled product) is an important factor. The grocer determines the allotment for bread shelf space among his various suppliers (652-53 Meyn). A preferred position for bakery products is the position that is exposed to the major flow oftraffc past the bakery products display (656, 692 Meyn). It is customary for a grocer to allot the preferred position and the largest amount of shelf space to his private label bread and the next best position to the supplier ofthat private label bread which the supplier wil use to merchandise his advertised label (506-8 Woodward; 656 Meyn). Chain stores with captive label bread wil allot the preferred position and the largest amount of shelf space to the captive label bread (10137-38 Nissen; 10210-Il Metz). 34. Most grocers including chains who have their own bakery wil usually carry a limited quantity of the advertised labels of all the wholesale bakers selling in a particular marketing area in order to meet consumer demand. Restaurants and institutions, on the other hand, usually purchase bread products from only one wholesaler. 35. Although all white bread is virtually a homogeneous product with regard to quality, nutrition, palatability and physical features wholesale bakers, by advertising their primary label bread, are able to create a consumer franchise for their brands. Because ofthis consumer franchise, a retail grocer is able to sell advertised white bread Initial Decision 104 F. at a higher retail price than brands that do not have a consumer franchise (CX 104; see 8I84 R. Walsh). 36. Historically, the retail price of captive label bread is below the price of advertised label bread, and private, controlled, and secondary label breads are used by chain stores, which do not have their own bakeries, to compete at the retail level with the captives. Wholesale bakers sell bread products under private controlled and secondary labels at a lower wholesale price than the wholesale price of advertised label. A large retail price spread between captive and private label bread, on the one hand, and advertised label bread, on the other will result in a loss of sales of advertised label bread. If this price spread is reduced the captive and private label products wil lose sales and there will be an increase in sales of advertised label bread (See 10139 Nissen; 10239--0 Metz). (19) Bread Industry Trends 37. During the 1960's and 1970' , the bread industry has enjoyed little, if any, growth (522, 605 Woodward; 1022 Sweeney), and in recent years the per capita consumption of bread has decreased (522 605 Woodward; 10088 Mackaman; 11224 Schaus). The market for standard white bread has declined from 8 808 milion pounds in 1967 to 8 657 milion pounds in 1972, and further decreased to 7 327 million pounds in 1977 (CX 1355, (E-15); RX 1000E). White bread has accounted for an increasingly smaller portion of total bread sales. White bread represented 79% of all bread sold in 1947, 6I % in 1967, and 59% in 1971 (CX (178, 192)). On the other hand sales of variety breads have increased (10162 Nissen; I0231 Metz; 11234 Schaus). 38. Although total white bread sales have decreased, the total of private and captive label bread sales have grown. In this respect the private and captive label share of white bread sales nationally increased from 18% in 1960 to approximately 36% in 197I, and Continental expected that that captive and private label share ofthe white bread marketshare would increase to 50% by 1977 (604 Woodward; CXs 4(130), 21(9); 94A). In 1971, Continental estimated that one third of the white bread that was not sold under an advertised label of wholesale bakers was baked by captive bakers, while two-thirds was private label bread baked by wholesale bakers (CX 4(I30)). 39. The cost of ingredients and labor increased sharply during the early 1970's. Because Federal Price Controls often precluded commensurate wholesale price increases, numerous small bakers were forced out of the market by the resulting cost-price squeeze (8992- Brown; 102O-5 Metz; 10132-33 Nissen). As many as 140 independent bakers withdrew from the market in the 1972-1974 period (10205 Metz; see also 112I4 Schaus). Between 1969 and 1974, 83 bread baker- 280 Initial Decision ies that competed in the areas served by Continental went out of business (CX 37C).

40. Individual geographic markets in the bread industry are highly concentrated. According to Continental' s 1972 Business Plan, in 75 Markets surveyed, the average white bread market share of Wonder and the leading wholesale baker competitor was 59.7%, in a universe that included private label and captive label bread. Excluding private label and captive label bread, the average two-firm share in advertised label white bread would be over 75% (See CX 4(131)). 41. There have been few new entrants into the wholesale baking industry during the 1970' (See 10171 Nissen; 10282-83 Jakacki). The number of captive bakers, which grew dramatically in the 1960' , has decreased in the 1970's (10140-2 Nissen; 10202--3; 11230-31; 11236 11243 Schaus).

Continental's Sales and Profits 42. Although Continental' s share of total bread sales on a national basis has not increased, but remained between 11% and (20) 12% from 1967 to 1977, its advertised label's share of the overall white bread market increased from 15.4% in I975 to 20. 1 % in I978 (SeeCXs 1355; 2162B; (in camera); RXs 1000; 1142 (in camera)). Respondents estimated that as of February of1973, Continental had a 32.7% share of the advertised label segment of the white bread market (CX 9(203); CX 1371B).

43. Continental's dollar sales of all bread products increased from $309 milion in 1967 to $609 milion in 1977. Continental's pretax profit on bread as a percentage of sales for the period 1964 through I977 were as follows (RX 1142 in camera): 1964 4.

1965 3.

1966 4.

1967 5.

1968 4.

1969 4.

1970 5.

1971 4.

1972 (0.

1973 (1.

1974 0.

1975 2.

1976 0.

1977 3.

g., )). Initial Decision 104 F. Continental's Costs 44. Before I973-1975, when it installed a Standard Cost Accounting system, Continental did not maintain ongoing records of the cost incurred in producing particular varieties or labels, but instead maintained records on the total costs for the weekly operation of each bakery. Continental did attempt to estimate the costs of producing any new variety or label on its Form 452B (See9688-93 9887 Dierker). 45. The bread industry has a relatively low ratio of fixed to variable costs (10486-7 Wachter; 11633 Preston). Apparently, Continental's variable costs in 1971 were approximately 80% ofits revenues (SeeCX 262D (in camera); CX 747 (in camera); see also CX 2I04-Z-24). For purposes of this initial decision, it wil be assumed that Continental' variable costs were 80% of its total allocated costs. Distribution and selling costs for wholesale bakers were relatively high. For example in 1973, Continental's distribution costs were 35% or more ofthe total costs on all route sales of bread (CX 58B; CX I06; CX 747 (in camera)). Market Shares 46. Each year Bruskin and Associates, an independent research organization, conducts a telephone survey for (21) Continental in most of the markets where Continental competes in the sale of bread products (768 Hackett; 595- , 602 Woodward). The purpose of this study is to ascertain bread producers' share ofthe markets, including private label and captive label bread, and to generate market trend information (768 Hackett; 630 Woodward). Although these studies tend to understate private label shares, they are fairly accurate measures of market shares on a national and regional basis and are relied upon by Continental to determine competitors' shares of individual markets. However, the samples are too small to provide accurate market shares for a local market and should be used only to determine trends in such markets (See771-72 Hackett; 602, 6I4, 630 Woodward). The Bruskin studies (sometimes referred to as "Bread Metrics (CXs 135-143)) form the basis for most ofthe market share data that appear in Continental's business reports and planning documents (See e. CXs 11 (137); 2167B (in camera ITT's Involvement 47. After it became part of ITT in 1968, Continental changed its reporting, business planning and budgeting systems to conform to ITT' s methods (1328-29 Willmont). Continental became subject to ITT' s high growth philosophy (896-97 Stolle). ITT's influence made Continental' s management more responsive and alert to identifying . . ..... .... .. .j .j ,.( ,.. .. , INTERNATIONAL TELEPHONE & TELEGRAH CORP., ET AL. 307 280 Initial Decision opportunities and problems. Woodward, Continental' s president during the early 1970' , stated (636 Woodward); ..... ITf's strength is an extremely effective management system. We are working very closely with ITT headquarters and its staff. lit is a good risk taker, provided you effectively support your plan. They have made us more aggressive in our outlook toward opportunities. I'M expects its units to grow 15% a year in profits. ITT personnel with oversight responsibilities for Continental were aware that the growth in private label bread was preventing Continental from meeting its growth objectives (CX 25A-C; see CXs 3(091); 5(018)), ITT was directly involved in establishing strategies for gaining additional private label sales and in evaluating Continental's pricing strategies (CXs 90J; 128B; see CX 58B). ITT knew that Continental sold private label products below cost (CX 94B-D). 48, In May 1971, ITT initiated a study to investigate the impact of private label (including captive label) merchandising upon Continental's bread business (796, 802 Butler). The primary purpose of the study was to generate an independent confirmation of Continental's internal strategies and to gain information on captive label businesses including their inside manufacturing costs and their distribution costs (761 Hackett; 802 Butler; 916 Stolle). McKinsey & Co. was chosen to conduct this study because they had good business relationships with a (22) number of retail chain stores (908 Stolle). Although Continental was attempting to meet the problems of captive label merchandising, ITT did not think it was doing enough (913-14 Stolle). 49. McKinsey personnel held many conferences with Continental personnel (805 Butler), At one such meeting, according to notes taken by R. Ronald Daniel, the McKinsey partner responsible for the study, the question arose as to whether Continental should hold the line on prices in hopes of driving weaker competitors from the market (CX 28S; seeCX 28T). In the final report submitted to ITT in October I971 McKinsey identified a number of issues that warranted further intensive analysis, one of which was the question of whether "holding retail prices (would) hasten the exit of wholesale competitors?" (CX 21 at 3; see CX 20Z-51-Z-3). According to Continental offcials, Continental did not adopt any new strategies as a result of the McKinsey report (607 Woodward; see 1349-50 Wilmont). Denver Marketing Area 50. Complaint counsel challenge certain pricing actions engaged in by Continental in the Denver, Colorado, area during the period 1963 through 1969. More particularly, complaint counsel challenge Continental's private label agreement with Associated Grocers, a vol unto;ru "not\"""'o;f.;no f'.;.. .-WO"'''''-," .... .. 4-h.. ",I'",.. 1",,. -_ . .

Initial Decision 104 F. pursuant to that agreement upon competition among wholesale bakers in the Denver area (CCPF 8-1-8193; CC Mem. 71-95). 51. Continental's Denver bakery was part ofits Southern California (Santa Monica) region which also included bread bakeries located in San Diego, San Pedro (DiCarlo), Beverly Hils and Salt Lake City (4677- , 4693-94, 4751 J. Walsh).

52. The other major wholesale bakers sellng white bread in the Denver market were Interstate, Campbell-Taggert, Old Homestead and Star Bread Company (6986 Downing). Star was primarily a baker of variety bread and it was not a great factor in the white bread market (6863-66 Downing). All white bread sales in the Denver market were made from plants located in that market. Continental Baking Co. v. Old Homestead Bread Co. 476 F.2d 97 (10th Cir. 1973), cert. denied, 414 U. S. 975 Old Homestead"

53. Continental's Denver bakery sold products, including white bread, in most of Colorado, Southeastern Wyoming and Southwestern Nebraska (CXs 1401E; 1427G; 1718Z-37 Ward). 54. In the late 1950' , Continental operated bakeries in Denver and Pueblo, Colorado (CX 1723D Barsch). In 1962, Continental replaced these two plants with a new bakery located in Denver (6864 Downing; CX 1723H Barsch). This new bakery was twice as large in floor area and could produce twice the amount (23) of bakery products as its two old bakeries combined (RX 831Z-7, Z-16 Hildebrand). When it began operations, the new bakery produced at a rate of 33 milion pounds of bakery products per year, which was only at 50% of its capacity (Old Homestead, supra 104). In 1962, Continental's Denver plant sustained a loss of $600 000 (CXs 1401F; 1427K). 55. During the period 1960 to 1968, the Denver bakery produced bakery products under the following labels: Wonder, Tender Crust Oak Farm, County Fair, Home Pride, Daffodil, Profie, Frenchef, Mrs. Wright' , Jils and Hostess (CXs 1427E; 140lB). It sold private label bread to Associated Grocers, Red Owl and Southland Corporation (7-1I stores) (CXs 1423B- , B- , N-71-72 Vanwyk). 56. Interstate had entered the Denver market by acquisition in the late 1950's and built a new plant which was completed about 1960 (6863, 6984, 6986 Downing). Its main advertised label bread was Butternut (6863, 6887-88 Downing).

57. Campbell-Taggart built its Denver plant some time in the 1950' (6986 Downing). Its advertised label bread was Rainbo (6865, 6887-88 Downing).

58. Old Homestead, established in 1892, was a pioneer in the Denver bakery market (6852-53 Downing; CX 1706N Roberts). For many years it enjoyed excellent shelf position in grocery stores because it "Moron INTERNATIONALTELEPHUNE & TELELiHAPH CUlt., E'I AL. ;jU 280 Initial Decision Rheinfrank; 7075 Tipton). In 1962, Old Homestead changed its baking process from the conventional mix to continuous mix process and as a result it became very effcient in white bread production (6859--O Downing). Old Homestead also operated a bakery in Colorado Springs (Star Baking Co.) and a bakery in Cheyenne, Wyoming (7048-50 Downing, Jr.

59. Star, (not to be confused with Old Homestead' s subsidiary ofthe same name), which was primarily a variety bread baker, was the leading wholesale baker in sales to restaurants and institutions (7099 7101 Cauble).

60. In 1963, the total of independent wholesale bakers' sales in the Denver market was approximately $14 milion. Continental had the largest share with about 35%, Old Homestead was second with about 29%, Campbell-Taggart had about 18% and Interstate about 16% (Old Homestead, supra at I05).

61. Safeway, King Soopers, and Milers were major grocery store chains in the Denver area that operated their own captive bakeries (6873-78 Downing; 7082-3 Tipton). It was estimated that these three chains accounted for about 75% of the grocery retail business in the Denver market (6878 Downing; 7237-38 Johnson). In volume of white bread sales, the bakery operations of King Soopers and Safeway were larger than the combined volume of all the wholesale bakers (7663 Stevens). (24) 62. The largest sellng size of white bread in the Denver market was the 1 lb. expanded loaf (7326 Aldrich).

63. Before 1964, wholesale bakers did not compete on the basis of the wholesale or retail prices of their products (7629 Stevens; 7087 Tipton). There were meetings or communications between them prior to any price move, which all the wholesale bakers took at the same time (6971-73 Downing). Competition among wholesale bakers was usually limited to salesmen s personalities and their relations with various grocers (7614, 7628-33 Stevens). The wholesale price of 1 lb. white advertised bread fluctuated between I6 and 20 during the early 1960's and, in July 1964, the price was increased to 20. (CX 1477 A-C).

64. The relative stability of wholesale price from 1960 to 1964 combined with rising costs of production, caused Old Homestead to operate at a loss of approximately $83 000 in 1964 (CX 15IOB; 7573- Wiliamson).

65. Before July 1964, none ofthe wholesale bakers sold private label bread (Old Homestead, supra, at I02). In the early 1960' , when Red Owl entered the Denver retail grocery market, it sought a private label program, but no wholesale baker was wiling to bake private label for Red Owl (6963-64 Downing). Red Owl acquired a baking Initial Decision 104 F. facility and produced its own bread until early 1965 (6963-64 Downing).

66. Associated Grocers is a cooperative of independent retail grocery stores. It was formed to purchase in quantity for its members and it performs wholesaler and other functions for its members (74 7G-71 McCutcheon; 6878 Downing). In 1960, grocers paid an entry fee in the nature of a deposit to join Associated Grocers and application approval was made by its Board of Directors on the basis of financial condition, location and accessibilty for delivery. Some applications for membership were turned down (7472 McCutcheon; 7181 Johnson). 67. Although the majority ofthe Associated Grocers members were located in Colorado during the 1960' , it also served grocer-members in Wyoming, Kansas, Nebraska and New Mexico (7471 McCutcheon). Associated Grocers' members accounted for about 4% of the retail grocery business in the major cities in the Denver area, including Boulder and Colorado Springs (7293-94 Johnson). 68. In 1961 and 1962, Associated Grocers advised the wholesale bakers in the Denver area that it was interested in obtaining either a private label program or in acquiring and operating its own bakery (7185-7 Johnson). Although at first all the wholesale bakers refused to supply private label bread, in late 1963 or early 1964, Interstate began serious private label negotiations with Associated Grocers (7186 7246-7 Johnson). When Continental learned of these negotiations, Ralph Ward, Regional Vice President, contacted Lenhard Johnson of Associated Grocers and began negotiations for a private label program (CX 1718 R. Ward; 7248, 7257 Johnson). (25) 69. On July 1, 1964, Associated Grocers organized a company named Five States Supply Company ("Five States ) for the purpose of the private label bread program. In practice, only members of Associated Grocers could become members of Five States. They were charged a $100 refundable membership fee per store and were required to agree to comply with the Five States Bylaws (CX 1519; 7183- , 7300 Johnson). Five States had no separate employees, offices or facilities and operated through Associated Grocers which handled its bilings and payments (7183 Johnson; 7473 McCutcheon). 70. The private label bread supplied to the Five States members was labeled "Tender Crust", Five States having secured exclusive use of this trademark (CX 1522D). The Five States Bylaws provided that the member grocers would exert their best efforts to sell the Tender Crust labeled products, would give first position on the grocers' regular bread rack to Tender Crust label bakery products, would give second position to the advertised label of the wholesale bakery supplying Tender Crust oroducts. and would orice Tender Crust oroducts at the .

280 Initial Decision same retail price as any other comparable bread product (CX 1522D- E).

71. Under the date of August 10, 1964, Continental entered into a written agreement with Five States to supply its member grocery stores with private label Tender Crust and other bakery products (CX 1526). That document provided, in pertinent part, that Five States (CX 1526 B):

4. lagreesl to make available to (Continental) preferred space on (itsJ member bakery products racks for the display of(Continental'sJ regular Bread, Cake and Sweet Goods products so that the fun sales potential of these advertised brands can be realized. (Five States further agreesJ in requesting Lists) members to cooperate with (Continental) in arranging special weekly promotions on said advertised bread items. 5. Deliveries of all products to (its) member stores pursuant to this agreement shall be made by fContinental'sJ regular Wonder and Hostess route salesmen from their regular delivery trucks. ..n 72. Private label products were to be the lib. expanded white bread, the 1 1/4 lb. round top white bread, 1 I/4 lb. sandwich bread and hamburger buns and hot dog rolls. The agreement provided for three methods of delivery at different prices. For example, the wholesale price for I lb. expanded bread with stale returns was 17. 5v; with no stale returns was 16v and on a store drop shipment was 15v (CX 1526 , D. nn 7, 8, 9). At that time the wholesale price for Wonder 1 lb. expanded white bread was 18.5 (ld. ). (26) 73. Paragraph 12 of the agreement provided that the wholesale price set by Continental would rise if any general price increases of controlled label bread occurred in the market (CX 1526D). Paragraph I3 of the agreement provided (CX 1526D):

In the event prices on any of (Continental's) regular advertised products shall, in response to a general price decrease in any market, fall below the prices in effect as of June 1 , 1964, then any private label price set forth in paragraphs 7, 8 and 9 shall be reduced in that market to a price at lea."t one cent below the prices on (Continental's) comparable advertised item.

74. The Tender Crust program, including prices, were always conducted under a written agreement and each agreement was in effect until amended (7199 Johnson). After 1964, all deliveries included rack service with or without stale returns (7203 Johnson). 75. Continental obtained the Tender Crust business because it was willng to enter into a written agreement with Five States (7265-6 Johnson).

76. During the I960's Continental representatives were invited to and frequently attended meetings of Five States Board of Directors (7213-14 Johnson). In 1964 and I965, Five States newsletters urged Initial Decision 104 F. Associated Grocers members to join Five States and to participate in the Tender Crust program (CX 1518G). Continental personnel served as liaison with Five State members and helped them with bread rack arrangements and solicited new members (7211 Johnson). During the 1960' , Continental wrote, typed, made copies, and mailed the Five States newsletters. Five States would supply Continental with information for its newsletter. Paper for the newsletter was kept at Continental's Denver plant (7209, 7213 Johnson). 77. The newsletter advised Five States members of the retail prices on Tender Crust, of the weekend specials and also urged the Tender Crust grocers to maintain those retail prices (CX 1518A-Z-25). 78. At the beginning of the Tender Crust program all labels of comparable bread were identically priced at retail. The wholesale price of Tender Crust was by contract, lower than the price of the other labels (7221 Johnson; CX 1723-Z-5 Barsch). The automatic one cent price differential between Wonder and Tender Crust was created to discourage other wholesale bakers from attempting to compete with the Tender Crust program (7233 Johnson). 79. In September of 1966, Campbell-Taggart began sellng a white bread loaf under the label Rainbo Buttermilk at the same price as Tender Crust. Pursuant to its agreement with Five (27) States, Continental granted a 6% discount so that the wholesale price of Tender Crust would be below the wholesale price of Rainbo Buttermilk (CX 1727Z-66-67 Vanwyk).

80. At other times the difference in price between Wonder bread and Tender Crust bread was more Clan one cent (CXs 1537, 1539 1549). A 1967 agreement between Continental and Five States provided for a 3v differential and a 2.25% central biling discount on Tender Crust purchases (CX 1571).

81. In September 1967, Continental, in response to a 10% discount offered by Campbell-Taggart, instituted its "sliding scale discount" on sales of Tender Crust products. Discounts of 7%, 9% and 11% were granted on certain volumes of purchases. The wholesale list price of Wonder bread was Iv greater than the wholesale list price of Tender Crust and a 2% central biling discount was in effect on Tender Crust purchases (CX 1549B). On November 6, 1967, the wholesale price of 21v to 16v and the wholesale lib. Wonder bread was reduced from price of the comparable loaf of Tender Crust was reduced from 20v 15v (CX 1652 Z-7). 82. On November I6, I967, the wholesale price of lib. white bread to Continental's other private label customers was reduced from 20v to 15v (CX 1652Z-50). To grocers entitled to 11 % discount and the 05v, or an 18.4%2% central biling discount, the net price was 13. n1"rfp -iffprpnt.l::1 nmn HPrl to the 16d Wonder bread price; ........ H. VU.n., j:T AL. 3I3 280 Initial Decision 83. Complaint counsel's chart showing Continental' s wholesale prices and the percent of differential between Wonder bread private label bread is incorporated into this initial decision, as it appears on the following page. (27a) 13, 1652- ex eX1698DCX1578 CX1650 1649CX CX1S77 Source CX1526 1727-ex 1652-ex eX1852- eX1652- 1537ex 1539ex 1553Aex CX1549 eX1549CX1652-Z-47 13518.9 10. 12. 12.4 13. 15. 17. 14. 16.18A 14. 23. 12. %ofPrlceDiference 4.4Tender Price (cents) DlfeJ'nce 1961969.Wonder, leettlePeriod 17. the JanetUnderIn CBClabels bylee Bake 13.S"" ProduceJenet Magic and ofWPBBake, Gem loafMagIc Pound Owl Farm, One Red Oak ofttle Gem, 3% andcents) Owl, 0% OakFann(In Red 101 PrlcN Crust, discountsNet 1)2) 3) Crust (Plan(Plan (Plan TenderCrust subjecto Tander 17.516 15 16. 18. 18.4 18. 18.2(7%&2%)17.8(9%&2%)17.4(11%&2%) 13.65(7%&2%)13.3S(9%&2%)13.05(11%&2%\ 13.05"" 15. on Bake discount Magic 18. 17. scaie and Wonder sliding 19S7'" 196B 196B 1988 1969 1969 CBC's6%discountonTenderCrlJstCSC's2%discountonTenderCrustCSC's TenderCrust 1964 19S6 1986 1968' 19S5" 19S7 Date July Oc!.1965 Mar. July Sept. Oct. Aug. Sep!.1967'" Nay. Jan. Mar. Sept. April July . .. ... .... g., 280 Initial Decision (28) 84. The Tender Crust weekend specials were in effect from the original Tender Crust agreement of August 1964 through at least 1970 and were offered by Continental every week (7205-06 Johnson). At first they were offered under the Wonder label, and later, at the request of Five States, under the Tender Crust label (7207-09 Johnson). Some of these specials were bread items, such as sandwich or wheat loaf or hot dog rolls and hamburger buns (CX 1699E; 7023- Downing; 7206 Johnson). These specials were offered at reduced wholesale prices and were sold at retail at reduced prices (CX 1518; see e. 7207 Johnson).

85. In the third quarter of 1967, Continental's total cost for the 1 lb. expanded loaves of Wonder bread and Tender Crust white bread were 19.62\1 and 16.59\1, respectively (CX 1722Z-19-Z-20 Swan). The average variable cost for Tender Crust would be 13.27\1, applying the 80% ratio adopted for the purposes of this decision. Accordingly, the 13.05\1 wholesale price on Tender Crust 1 lb. white bread was below Continental's average variable cost. Many other wholesale prices were below Continental's fully allocated costs (Admission, Set 6 (amended) Item 9)).

86. At the outset of the Tender Crust program 50 stores had become members of Five States. There were more than 100 member stores in Five States by early 1965, and more than 150 had joined the program by February 1967. Five States had 200 member stores by January 1969. At that time Associated Grocers had 415 member stores (CXs 15I5Z-22 , Z-62; 7214 Johnson; see CX 1423B-6 Vanwyk). 87. In I966, Continental's sales of Tender Crust products to the Five States grocers amounted to $670,000. In 1966, such sales were between $670 000 and $1 000 000 and in 1968 were over $1 000 000 (7217- 7291-92 Johnson; CX 1515-Z-22, 62). In the first four weeks ofi965, Continental sold an average of 60 000 loaves per week of the 1 lb. expanded size. In December 1967, when the so-called "sliding scale" was in effect, sales of 1 lb. Tender Crust averaged 90 000 loaves per week (CX 1727G Vanwyk).

88. In 1967, the total of independent wholesale bakery sales in the Denver market was still approximately $14 000 000. Continental had the largest share with 51.2%, Campbell-Taggart had 22.1%, Interstate had 20.1 % and Old Homestead had 6. 3% (See Old Homestead supra at I05).

89. When the Tender Crust program came to a grocery store that Old Homestead serviced, Old Homestead immediately lost shelfposition, space and volume (CX 1706 Roberts; 7079 Tipton; 7327-29 Aldrich). In the Five States grocery stores, Tender Crust and Wonder bread occupied 75% of the bread rack (CXs 1706E-G Roberts; 1717C- D Chapman; 7328-29 Aldrich; 7398 Phillps). In some Five States 3I6 FEDERAL TRADE COMMISSION DECISIONS Initial Decision 104 F. member stores Old Homestead lost the account completely (7616 7618, 7620 Stevens; 7356--57 (29) Rheinfrank). The erosion of Old Homestead' s market position was a continuing process that started in August 1964 (CX 1706H Roberts; 7669 Stevens). 90. Continental's weekend specials also had a very detrimental effect on Old Homestead. These specials greatly increased Old Homestead' s stale percentage, which became one of the worst elements of Old Homestead' s costs (6924--25 Downing).

91. Continental' s plant manager thought that there was a possibility that Old Homestead would go out of business (See CX 1423B-71 Vanwyk; see also 6939--I Downing). On December 31 , 1967, Old Homestead withdrew from the baking industry (6851 Downing). 92. On December 18, 1967, Continental announced price increases to be effective January 2, I968. One pound advertised bread was increased to 21 , secondary label was increased to 19. and private label increased to I8 on a no stale return basis. All discounts were eliminated (CX 1569; RX I150). In March I968, prices were reduced lib. Wonderto 17 and lib. Tender Crust to 13. (CXs I696D; 1578). 93. Continental, Interstate and Campbell-Taggert each made a profit on their bread operations in Denver in 1967 (CX 1401F). Only Old Homestead showed a loss in that year (CX 1510C). 1967 was the first year since 1960 that Continental made a profit $14 000 (CX 140IF).

94. Continental retained the Tender Crust private label business until the middle of 1978. At that time Interstate offered a lower price and became Associated Grocer s supplier of private label bread. However, in November 1979, Tender Crust, dissatisfied with Interstate solicited bids from the wholesale bakers and Continental again became the supplier of Tender Crust (9562-64 Nuziard). 95. From the various reports on production and market surveys it appears that in the I970's Campbell-Taggert has become the largest wholesale bread baker in the Denver market with Continental and Interstate about equal (9543, 9554 Nuziard; see CXs 1385-89; RXs 1024 (in camera); 1031 (in camera); 1032A- (in camera); 1044 (in camera); 105IA- (in camera); see also RX 843A.) Oroweat purchased Star of Denver in 1977 (7103 Cauble). Continental's Regional Vice President testified that during the last ten years Safeway and King Soopers have a share of approximately 85% ofthe retail grocery business in the Denver market (9543--4, 9546 Nuziard). Recently, both of these large chains have constructed large new bakeries in Denver (9546-5I Nuziard).

96. It is found that during the period 1964 to I968 Continental discriminated in price between purchasers of its advertised bread and purchasers of its private label bread and that during most of this 280 Initial Decision period those lower discriminatory (30) prices were below Continental' s fully allocated costs. It is further found that in the last quarter of 1967, Continental's lower discriminatory prices were below its average variable costs. In most instances, these lower prices were not granted to meet the equally low price of one of Continental's competitors. It is found that Continental's pricing practices contributed substantially to the failure of Old Homestead and that as a result thereof the wholesale white bread market has become more concentrated in the Denver, Colorado, market area.

97. Complaint counsel also contend that the discriminations in price between competing Tender Crust customers which resulted from Continental's sliding scale discount program in effect in September, November and December of 1967 caused the type of injury to competition between such customers proscribed by Section 2(a) (See CX 2101 Stip; CC Memo. 93). The maximum price difference was 8/loth of one cent and 4% of $400 worth of purchases, $16 (See CXs 1549; 1652Z-7 Z-50). This maximum price difference was in effect between September 16, 1967 and November 6, 1967. These discriminations were eliminated on January 2, 1968. 98. During this period Continental had many Tender Crust customers in Denver, in other cities in Colorado and in Wyoming (See 1652Z-8Z49; 1523). Although it is highly probable that some of the Five State grocers (the purchasers of Tender Crust) competed with each other, there is nothing in the record to demonstrate that Tender Crust purchasers at the different discount rates of the sliding scale actually competed with each other.

99. Given the small amount of the price difference challenged as well as the relatively short period of time that it was in effect, it cannot be found on this record that the effect of such price differences may be to substantially injure competition between retail grocers. Minneapolis (Twin Cities Trading Area) 100. Complaint counsel challenge certain pricing practices engaged in by Continental in the so-called Twin Cities Trading Area ("TCT A" in Minnesota during the period 1966 through 1969 (SeeCSCPF 9- I- 177; CC Mem. 96-1I5). This market area includes the Twin Cities of Minneapolis and St. Paul, and seven counties surrounding them (CX 306; 6075 Alton).

101. Continental's Minneapolis bakery was part of one of its "Chica- " regions which also included bakeries located in Rochester, Minnesota, Davenport and Sioux City, Iowa, Omaha, Nebraska Milwaukee, Wisconsin, and Chicago ("Shiler Park") (CXs 220A; 307B).

102. The other major wholesale bakers selling white pan bread in 3I8 FEDERAL TRADE COMMISSION DECISIONS Initial Decision 104 F. the TCTA from 1965 to I970 were American, Zinsmaster Baking Company ("Zinsmaster ), Pan 0 Gold Baking Company ("Pan 0 (31) Gold"), Creamy Crust Baking Company CCreamy Crust " a subsidiary of Tender Crust, Eau Claire, Wisconsin) and Emrich Baking Company CEmrich"). Emrich sold primarily to restaurant and institutional accounts (6077 Alton; 6189 Peterson; 6338 Monahan; 6556 Coolidge; 8714 Wiliams).

103. In the late 1960' , Continental's Minneapolis bakery used a continuous mix process as well as a special conventional mix process to produce mostly white bread and bun varieties (CXs 208A; 303A- 8761 Wiliams; see 8975-77 Brown). It ran approximately 63 bread routes from its plant or from agencies located in the Minnesota communities ofSt. Paul, New Hope, Wilmar and Sauk Center as well as Eau Claire, Wisconsin (CX 200F-F; seeCX 250A-p; 6319 Hueller). It also supplied Continental's Rochester bakery with some bakery products (6298 Glass; see CX 200B-DJ. 104. Continental' s Rochester bakery used a conventional mix process and produced white bread, dark bread varieties and buns, rolls and doughnuts (6282 Glass; CXs 208A, 304A-B). It had agencies located in the Minnesota communities ofOwatonna and lViankataz, and in La Crosse, Wisconsin (CX 200G-H). It supplied bread products to Continental's Minneapolis bakery as well as many other Continental bakeries (6298 Glass; see CX 200B-D). I05. The Rochester market area was considered separate from the TCTA. The areas served by Continental' s Minneapolis and Rochester bakeries did not overlap and bread prices in Rochester and Minneapolis-St. Paul were unrelated to each other and fluctuated independently (6289- 6299-6300 Glass; 8697-98 Williams; 8820 Brown). 106. Continental' s advertised bread was sold under the Wonder label while its largest selling secondary label was Wonder Country Style (CX 208B-C). The Minneapolis bakery sold private label to three large chain stores: Applebaum, Penny s and Kroger (8783 Wiliams). 107. American had bakeries in Minneapolis and St. Paul (6290 Glass). It was the largest wholesale baker in the TCT A and sold its advertised bread under the Taystee label (6196 Peterson; 6321 Hueller; 8684 Wiliams; 8813 Brown). It used the continuous mix process during the 1966-1969 period (6210 Peterson). It ran approximately 80 routes out of its plants and agencies (6138 Alton). American also sold bread under private label (RX 407B).

108. In the late 1960' , Zinsmaster had two bakeries in the TCTA one in Minneapolis and one in St. Paul. The Minneapolis plant was operated as a Kosher plant and baked primarily white bread and Zweiback toast (6181-82 Peterson). The St. Paul plant baked buns rolls and sweet goods (6182 Peterson). Zinsmaster was considered to 280 Initial Decision be the largest seller of advertised bread in the TCTA, which it sold under the Master label (6196 Peterson). It used a conventional mix process which apparently gave it an (32) advantage over American and Continental, both of which used the continuous mix process (6210 Peterson). Zinsmaster sold secondary bread under the Thrifty, Oven- Glo and Sky Line labels. It did not sell private label bread in the TCTA (CX 268I; 6I83 Peterson).

109. In 1966, Pan 0 Gold, which had a bakery in St. Cloud, Minnesota, bought Regan Bakeries Inc., including a bakery in Minneapolis (6068-71 Alton). It also purchased the Twin Cities sales operation of Eddys Bakery Company (6070 Alton; see RX 1072Q). Thereafter it sold primary bread under the Holsum label, and secondary bread under the American Beauty, Holsum, Better Buy and Eddy s Cabin Style labels (6080-82 Alton). It also sold private label bread to chain stores including Country Club (20 stores; Country Club label) and Super Value (30 stores; Favorite label) (CX 283C-F; 6113- , 6081- Alton). Pan 0 Gold used a conventional mix process (6210 Peterson). Most of its white bread sales were made either under private labels or secondary labels (RX 411E-G; 6150-51, 6163 Alton; 6517 Reinartz). 110. After Creamy Crust merged with Tender Krust in 1960 , it discontinued baking operations, and was supplied with bakery products by Tender Krust (6381, 6389-90 Lindebck). It sold advertised bread under the Bunny Bread label and secondary bread under the Creamy Crust Jabel (6385 Lindebak). It also supplied private label bread to the following major chain stores: Shopper s City, Johnson Brothers, Piggly Wiggly (Sunset GoJd label) and Walsh' s (6334- Monahan; 8729 Wiliams). Approximately 90% of Creamy Crust' sales were made under private label (6336, 6347--8 Monahan). It was the first wholesale baker to ofter private label bread to the chain stores in the TCTA (6357 Monahan).

111. In 1966, the approximate market shares of the wholesaler baked white bread market in the TCTA were American, slightly over 30%; Continental, slightly under 30%; Zinsmasters, 20%; Pan 0 GoJd 15%; and Creamy Crust 5% (6339, 6550-52 Monahan; 6195-98 Peterson; 6117- , 6138 Alton; 6424-25 CooJidge; seeCX 270K). It appears that Zinsmaster and American were Jeaders in sales of advertised white bread with Continental in third position (6196-97, 6248-9 Peterson). Continental appears to have been the largest seller of secondary and private label white bread (See CX 204 J-K). 112. Two major chain stores had "captive" bakeries in the TCTA in 1966: Red Owl and National Tea (6190-91 Peterson). Holiday Discount Stores, a two store chain sellng general merchandise, operated a bakery in one of them in 1966. It produced a "poor" loaf of bread by a fast and cheap process (6192 6245 Peterson; 6439 Coolidge; CX Initial Decision 104 F. 2355B, C Coolidge). In I966, the captive bakeries sold about 30% ofthe white bread in the TCTA (6120-21 Alton). In addition to these captive bakers there were many "in-store" bakeries. In 1968, Penny s opened its own bakery, in which it baked buns, rolls and sweet goods. (33) It is estimated that the captives and in store bakeries accounted for 50% of the white bread sold in the TCT A (6120-21 Alton). 113. The highest volume white bread loaf sizes sold in the TCTA were the 1 I/2 lb. round top and sandwich loaves (6115 Alton; 6190 Peterson; see CX 204C-L).

114. During the period 1966 to 1970, Continental was considered to be the price leader among the wholesale bakers in the TCTA, because historically it was the first wholesale baker to increase or decrease the wholesale price of bread (6339, 6362 Monahan; 6387, 6408 Lindebak; 6229 Peterson; see8965 Brown). Continental was also considered to be the price leader in the Rochester market (6387 Lindebak). 115. No wholesale baker could sell bread products in the TCTA at a wholesale price higher than its wholesale baker competitor s price for a comparable product (6352-53 Monahan; 6408 Lindebak; see6211 Peterson).

116. Although there was always keen competition among retail grocers in the pricing of white bread products, by the middle of 1966 the retail price spread between private label bread and advertised label bread had been reduced to about 5if (RX 1072D). The wholesale price for a 1 1/2 lb. loaf of advertised label bread was 28if and the retail price of that loaf was 35 . The wholesale price for secondary bread was 24. 5if and the retail price for secondary bread was 31if. The wholesale price for private label bread with rack service was 23.5if and the prevailing retail price for private label bread was 29if (RX 1073J).

117. In November 1966, Holiday Discount Stores opened a new store in Fridley, Minnesota, a community in the Northern suburbs of the Twin Cities. Its retail price for bread baked in its own bakery was 21if. The chain stores reduced their retail price on private label bread from 29 to 21 . Penny, to which Continental supplied private label bread, advertised its 21 retail price. Although Holiday raised its retail price to 23 on November 12 , it reduced it to 21if shortly thereafter (CX 315; 8938-40 Brown; see RX 1072W, 1073J; 6246 Peterson). 118. On December. 15, 1966, Continental reduced the wholesale price of its secondary label Country Style white bread from 24. 20if in TCTA (CX 207C; RXs I072V, W; 1073 J). At that time, the fully allocated costs of manufacturing and distributing Wonder Country Style from the Minneapolis plant was 23.45 (CXs 204Q, 249). At the wholesale price for this secondary bread, each loaf was sold at 3.45if below Continental's fully allocated costs. For over two months r... l1until Jiphrll r" ? qR7 rnntiopnhll' s wholesale Drice for Wonder INTERNATIONAL TELEPHONE & TELEGRAPH CORP., ET AL. J41 280 Initial Decision Country Style white bread remained below its fully allocated costs (CXs 207C; 204Q; 249). (34) 119. Continental' s Minneapolis plant stated that this price reduction was to meet the competition from Red Owl and Holiday (RX 413A; see also CXs 300B 295C).

120. At the time the Wonder Country Style was being sold at wholesale at 20 in the TCTA, it was sold in the Rochester trading area at 24. , and Continental' s advertised Wonder bread was sold at 28 both the TCTA and the Rochester trading area (CX 207C). 121. After Continental reduced its wholesale price for Wonder Country Style it reduced the wholesale price of private label bread to its three large chain store accounts (CX 202F; see CXs 207C; 209P). The record shows that the losses incurred by Continental's Minneapolis plant increased substantially after these price reductions were made (CX 308D,I, S).

122. On March 7 1967, Zinsmaster introduced a 11/21b. "split top loaf of white bread, for a "premium" wholesale price of30 . Continental had been successful with this product in Chicago selling it at the premium price of31 (6116-17 Alton; CX 202Q). On March 24 1967 Continental introduced its own split top loaf with a wholesale price of23 (CXs 207C, 316, 318). Pan 0 Gold also introduced a similar split top loaf under the "Bonanza" label in March 1967 (6I16 Alton; CX 283G-E). Zinsmaster stopped selling its split top loaf on May 20 1967 because ofa sales decline after Continental's split top loaf came on the market at a lower price (CX 300 D; RX 409B; 6221-62 Peterson). Continental used a mechanical device to split the bread (8705- Wiliams; 8815 Brown). The Minneapolis plant ordered production of two "loaf splitters" on December 16, I967 (CX 310). 123. The retail price was continued into mid-1967, with Red Owl Penny, Johnsons and Shopper s City advertising special low prices on white bread (CX 319; RX 1073J-K). By July I967, an 8 to I6 differential had developed between the retail price of Wonder advertised bread (as well as other advertised bread) and the retail price of private label (including captive label) bread in the TCTA (RX 406; 8687-88 Wiliams; 8816 Brown). This resulted in a loss in volume of advertised label sales for all wholesale bakers (8816 Brown; 8687-88 Wiliams; see e. RX 405A-C).

124. On June 28, 1967, Pan 0 Gold's Minneapolis plant was damaged by a severe wind storm and it did not bake bread for six weeks (6097-98 Alton). During that period it supplied about 15% of the white bread sold by it in the TCTA from its Pipestone and St. Cloud plants (6099 Alton). It purchased bun products from American and Zinsmaster (6101--2 Alton), sandwich bread from Emrich and most of its white bread from Red Owl's captive bakery (6102-03 Alton). Initial Decision 104 F. Although Alton testified that Continental refused to supply Pan 0 Gold with bread products (6103), Wiliams, who became Continental' Minneapolis plant manager the weekend of the storm, testified that Continental never received (35) a request from Pan 0 Gold (8696). In any event, Continental did not supply bread products to Pan 0 Gold during the period its Minneapolis plant was out of production. 125. After a strategy meeting between Continental personnel from the Minneapolis plant, Chicago Regional Offce and Rye Headquarters, Continental on July 24, I967, reduced the wholesale price of its 11/2 lb. Wonder advertised bread (round top and sandwich) by 8!" a loaf, from 28!" to 20!" in the Twin Cities metropolitan area, reduced the wholesale price of Wonder Country style from 21.5 to 18!", and reduced the Wonder split top white bread from 23 to 20!" (8733- Wiliams; CX 207C; see RX 1073P). These price reductions were met by other wholesale bakers on their advertised and secondary bread. Zinsmaster lost all sales to Penny s on July 24 1967, because it didn reduce its prices immediately (6222-23 Peterson; 6431-32 Coolidge). 126. Continental offcials testified that they hoped that a short term price reduction would preserve its advertised bread franchise, that they expected that all wbolesalers would benefit from increased sales of advertised bread, and that bread prices would eventually return to higher levels (RX 406; 8690-91, 8698 Willams; 8817-8820 Brown). 127. In July and August, I967, in conjunction with the price reduction on Wonder bread, Continental engaged in a newspaper advertising campaign in the Twin Cities area announcing a 1O!" reduction in the retail price of advertised bread and a 4!" reduction in the retail price of Wonder Country Style, its secondary bread (SeeCXs 222-227 3I4A-D; 6106-07 Alton; 8695, 8764-67 Willams). Similar advertisements were published in the outlying Minnesota and Wisconsin communities (CX 221F-J; CX 222F-N).

128. At the time of the July 24 1967, price drop on Wonder round top and sandwich breads, the fully allocated costs of manufacturing and distributing these products from Continental' s Minneapolis bakery were 25.12!" and 25. , respectively. The price reduction to 20!" was therefore, below Continental' s fully allocated costs by more than 5!" and, pursuant to the formula adopted for this case, was apparently below average variable costs (80% of 25.12!" equals 20. (See 204Q; 249; Finding 45 supra). The so-called July 24 1967, prices were in effect for approximately a month and a half, or until September 11 1967, when the wholesale price of advertised label bread was increased to 23!". Thereafter, for a period often months the prevailing wholesale price in the TCTA for Wonder round top and sandwich white bread was over 2!" per loaf below Continental's fully allocated costs (CXs 207C, 204Q, 249; see 8798 Willams). ..

INTERNATlUNAL" .rLI!rnv.l'J.

280 Initial Decision 129. At all times from July 24, I967, until July 21, I969, the Wonder round top and sandwich white bread were sold at wholesale in the Rochester trading area for 28 and Country Style was sold at wholesale for 24. in the Rochester area (CX 207C). (36) I30. The retail price war continued and on October 5, 1967, Continental reduced its wholesale price on I 1/2 lb. private label round top and sandwich bread to its three chain store accounts by 4 (CXs 202F 209P 245C , 309A-D). The prevailing wholesale price to Applebaums on rack service and stale return was reduced from 19. 15. (CX 309A). The actual wholesale price, including a 2% central billing discount, was 15. (See CX 318). 131. At the time of the October 5, 1967 price reduction the fully allocated cost of manufacturing and distributing Pantry Pride round top and sandwich bread to Applebaum from Continental's Minneapolis bakery was 19. and 19. , respectively (CXs 204R, 249). The wholesale price of15. was below Continental's fully allocated costs and, pursuant to the formula adopted for this case, was apparently below Continental's average variable costs (80% of 19. equals I5. ). The October 5, I967 prices remained in effect for approximately two months, until November 9, 1967 (CXs 207C, 204R, 249). 132. Between the weeks of October 7 1967 and November 11, 1967 Continental' s Minneapolis bakery sustained a net loss in profits before taxes of $45 336.85 (CX 250 K- P).

133. On March 14, 1968, Pan 0 Gold announced a 3 per loaf increase on advertised and secondary bread and an increase of 2. private label to become effective March 25 (6506, 65I8 Reinartz; 6115 Alton; CX 283C-E). On March 20, 1968, Continental announced an increase of 1. on secondary and private label to become effective on April 1, (6506-7, 6518 Reinartz; 8701 Wiliams; CXs 229A, 230B 232B, 233B, 234B, 235I, 245H; RX 416; see RX I074C). The other wholesale bakers followed Continental's price increase, and Pan 0 Gold met Continental's price. Zinsmaster did not consider Pan 0 Gold to be a significant factor in the advertised bread market (6230 Peterson).

134. On April 2, I968, Zinsmaster instituted a treble damage suit against Continental for its pricing practices in the TCTA (CX 299F; 6436-37 Coolidge).

135. On July 3, I968, Continental announced an 1.5 price increase on primary, secondary, and private label bread, effective July 15, 1968 (CX 207C).

136. On December 2, 1968, Pan 0 Gold announced a wholesale price increase of 3. on advertised and 3 on secondary to become effective on December 23, I968 (6507-09 Reinartz). Continental did not follow this price move, but announced, before December 23, that it would Initial Decision 104 F. increase its wholesale prices by 1.5(/ and 2(/, on advertised and secondary labels respectively (6509 Reinartz). The other wholesalers, including Pan 0 Gold, increased their prices to match Continental' s prices (6509, 6514 Reinartz). (37) 137. On July 21, 1969, Continental raised the wholesale prices of Wonder advertised bread again and the wholesale prices of the 11/2 lb. loaf of Wonder round top and sandwich and the split top loaf were 28(/ a loaf (CX 207C). These were the prices prevailing in the Rochester area (CXs 202H, 207C).

138. During the period of "low" prices on advertised bread, the profits of all the major wholesale bakers including Continental were substantially decreased (CXs 248, 250; RXs 1073P, lO74D, I, 6112 6142-45 Alton; 6509 Reinartz; 8775-76 Willams). Zinsmaster lost profits during this period and was forced to curtail its advertising (6226 Peterson; RX 1074D). Pan 0 Gold lost sales and estimated a loss of profits of $1 600 000 (6112, 6142 Alton; 6509 Reinartz). It discontinued all advertising and capital expenditures except those for emer- 11"gency repairs (6113 Alton). In 1970, it went through a "Chapter bankruptcy proceeding (6145 Alton). Creamy Crust suffered a net loss for bread sales in the TCT A each year during the period; its loss rising from $13 748 in fiscal year 1967 to $70 391 in fiscal year 1968. In 1969 Creamy Crust lost $11,767 (6386 Lindebak; CX 294F; RX 4IOD). 139. The treble damage suit instituted by Zinsmaster and joined in by Tender Krust/Creamy Crust in July 1968 and Pan 0 Gold in 1969 was settled and certain agreed-upon damages were paid to the plaintiffs by Continental Pan 0 Gold received $200 000 (6145 Alton). In addition, the Court entered an order on January 11, 1971, prohibiting the parties from sellng bread products at different prices in an extensive area of southern Minnesota, an injunction that was to be in effect for four years (CX 299).

140. Although Continental reported in 1971 that it had surpassed American as the number one seller of white bread, it appears that American is the Jargest producer of white bread in the TCTA today (SeeCX 4(131); 10227- , 10242 Metz). In 1978, Metz Baking Company purchased Zinsmaster and in early 1980 was in the process of constructing a new bakery in Minneapolis (10202, 10242 Metz). Metz and Continental are about equal sellers of bread products, following American. Heileman Brewery has purchased Pan 0 Gold and Emrich (10227 Metz). Pan 0 Gold's old Minneapolis bakery has been closed and sales to the TCT A are made from the St. Cloud plant (8794 Williams; 10262 Metz). Creamy Crust is no longer in Minneapolis (8793 Wiliams). The captive bakeries continue to maintain very strong positions in the market (RX 1078D; 8716 Wiliams; 10209- , 10222- 23 10227 Metz).

_ _u.. "'.1 jJ.JJCrnvl'\jJ'&'l LgGRAHCORP. ET AL. 325 280 Initial Decision 141. Minneapolis is stil a private label bread market. Variety breads have taken over much ofthe bread market and sales of white bread have declined. Continental now concentrates its efforts on its sales of advertised bread (8707-13 Wiliams). (38) I42. On December 10, 1968, the Commission s Bureau of Restraint of Trade (now Bureau of Competition) advised Continental that the Commission had received a complaint concerning the pricing of Continental's products in the Twin Cities trading area alleging violation of Section 2(a) of the Clayton Act (RX 404A). After Continental had submitted certain information, the matter was closed by the Commission on October 27 1970, without prejudice to the right ofthe Commission to reopen the matter if and when warranted by the facts (RX 403G; see RXs 402-404).

143. As to the TCTA, it is found that Continental discriminated in price among its customers and that the effect of such discrimination may have substantially injured competition between Continental and its wholesale competitors. It is also found that the lower prices to certain customers were below Continental's fully allocated costs and in several instances, were below its average variable costs. It is also found that Continental's lower discriminatory prices were not made to meet the equally low price of a competitor, but were made with the express intention of restructuring the wholesale bread market in the TCTA. In this respect, it is found that Continental maintained its lower discriminatory prices over long periods of time when it knew or had reason to believe that its wholesale competitors were suffering large economic losses and, perhaps, were being irreparably injured in their ability to compete in the sale of advertised label bread, secondary bread, and private label bread in the TCTA. Finally, it is found that the wholesale baking segment ofthe white bread market in the TCTA has become more highly concentrated over the years. Seattle Marketing Area 144. Complaint counsel challenge certain actions taken by Continental in the Seattle marketing area during the 1960's and early 1970' s. Unlike the other five marketing areas studied in Docket 9000 the Seattle phase of the case does not contain allegations of ilegal price discriminations. (CCPF 10-1- 10-4; CC Mem. 116-119; CC Reply Br. 76-78).

145. Continental's Seattle bakery is part ofthe Northern California Region (Redwood City) which also included bakeries located in Portland, Oregon, Spokane, Honolulu, San Francisco, Sacramento and Oakland (3163 Frielink).

146. The Seattle marketing area includes generally the metropolitan areas of Seatte and Tacoma (See CX 119I). , Initial Decision 104 F.T. 147. Other wholesalers in the Seattle area during the last halfofthe I960' s included American, Gai's Baking Co. ("Gai' ), Hansen, and Oroweat (6621 Vail. Another baker, Ruth Ashbrook ("Ashbrook" produced bread for Associated Grocers, a voluntary cooperative association, and distributed bread products to members ofthe cooperative. Gai's was primarily a (39) variety bread baker until it built a new plant in the late 1970's and began producing large volumes of white bread (9327 Frielink).

148. During the mid 1960' , Safeway was the only captive baker in the Seattle area. Later Fred Meyer sold captive label bread in Seattle bread that was baked in its Portland, Oregon bakery (6683 Vail; 9303-04, 9328 Frielink).

149. At all times relevant herein, Continental was shipping bread baked in the Seattle bakery to Alaska (6581-83 Stip. I50. It appears that Continental's share of the wholesale white bread market in the Seattle area was approximately 25% in the early 1970' (See CCPF I0-6; CX I35).

151. The most popular sized loaf of white pan bread sold in the Seattle area during the late 60's and early 1970's was a 22.5 oz. loaf (4368-9, 4377 Salle; 6622-23 Vail.

152. In 1963, George T. Vail and Boyd Hall, who had been sellng bread house-to-house, began sellng bread at wholesale in the Seattle area (6612 Vail). At first they purchased bread fieom Ruth Ashbrook but this source of supply was terminated after a short while, due to certain pressures by Local 227 of the Teamsters Union (See 6615 6672-73 Vail). Thereafter, and until 1972, Vail purchased bread from the Mother Hubbard Bakery in Abbotsford, Canada (6616-17 Vail; see 4379 Salle). Vail and Sandland, who replaced Hall as Vail's partner formed the Granddad Bread Company, and the so-called Canadian bread was sold by them in the Seattle market under the Granddad label (66I2, 6616-19, 6626, 6628-29, 6643 Vail; 6780 Sandland). 153. Vail and Sandland drove to Canada, picked up the Granddad bread products (white bread and whole wheat bread), trucked such products to the Seattle area and sold and delivered them at wholesale prices to grocers on a store drop delivery basis (See6673-74 Vail; 6815 Sandland). Granddad packaged its bread in fold-up, reusable cartons that held 16 loaves. Granddad bread could be merchandised directly from these cartons (6625 Vail; CX 1188 (52) Eastey). 154. Granddad' s method of transport and delivery, as well as a favorable exchange rate on the Canadian dollar during the 1960' permitted Granddad to sell its bread at significantly lower prices than the wholesale prices charged by the wholesale bakers in Seattle (6673 6696-97 Vail; 6815 Sandland). Granddad bought the bread from . l "f (1'1'1 R Vail. It cost Granddad about 280 Initial Decision 21 a loaf to transport the bread from Canada (6673 Vail). In I964- 1965, Granddad charged its grocery customers 181 per loaf(6621 , 6632 Vail; see, e. CX 1194). Those customers usually sold Granddad bread at retail for 5 loaves for $1.00 or less (6631-32 Vail; 6754 Hinshaw). (40) 155. In 1964-1965, the wholesale price of the wholesale bakers advertised labels was 281, based on a conventional 20% discount from the prevailing retail price of 351 (6632 Vail; CX 1197 (5) Thompson). 156. The fact that Granddad was nonunion caused it considerable diffculty throughout its existence with Local 227 of the Teamsters Union in Seattle which represented the driver-salesmen of all the wholesale bakers (6676, 6684-5 Vail; 6792-94 Sandland). I57. Early in Granddad' s existence, Vail and Sandland made arrangements whereby they would purchase Hostess Cake products from Continental for Mother Hubbard and "back haul" those products to Abbotsford (6648 Vail; 6819 Sandland). Granddad was the only firm with an ICC permit to haul goods between Seattle and Abbotsford (6649 Vail; 6818 Sandland). Alternative modes of shipping involved a t.transfer at Vancouver. The delay incurred by such transfer adversely affected t.he freshness of the cake products as finally delivered to the Mother Hubbard bakery (CX 1196(165) Taylor; 1197 (204- 05) Thompson; 6738 Vail. After Thompson, the plant manager of the Seattle bakery, questioned the financial arrangement. whereby Vail and Sandland paid for the cake products, Mother Hubbard began paying Continental directly for those products (CX 1197 (197- , 205) Thompson).

158. Granddad bread was equal in quality to that of other wholesale bakers (6629 Vail; see6754 Hinshaw; CX 1188 (57-58) Eastey). Granddad' s business "started booming right away" after it started getting bread from Canada (6626 Vail. In 1964 and the early part of 1965 Granddad sold about 40 000 loaves a week in the Seattle area (6626- 30 Vail).

159. In early 1965, Continental introduced a 22.5 oz. secondary loaf of white bread called Wonder Country Style "to meet the competition of Canadian Bread, Pink Elephant and other baking companies" (CX 1022B; 6627 Vail. Pink Elephant was a home delivery baker (6612 Vail). This secondary loaf was priced to sell at retail for 4 loaves for $1 (6635 Vail). At that time neither American nor Hansen were selling secondary label bread (6713-14 Vail.

160. As a result, Granddad lost substantial sales volume in the stores that retailed both Wonder Country Style and Granddad (See 6633-35 Vai! During this period, Continental employees offered one grocer, Hinshaw a better price on Wonder secondary label ifhe would give Continental an "exclusive" and "exclude" Granddad' s from the , Initial Dccision 104 F. store (6756-57 Hinshaw). Hinshaw rejected both offers (6758 Hinshaw).

161. Continental also ceased doing business with several grocers who refused to stop carrying Granddad bread (CX 1195 (8-9, I4, 15) Lundstrom; 4370-81 Salle). (41) 162. Despite the competition from Continental's secondary label and private label bread which appeared in Seattle in 1966, Granddad increased its sales to single-owned non-chain grocery stores and by the middle of1968 was sellng to 70% of those stores in the greater Seattle area (6781-83 Sandland) at which time it reached one of its two volume peaks (6782 Sandland).

163. In early 1968, Granddad' s wholesale price was 22 a loaf and Granddad' s bread was being retailed at 4 loaves for $1.00, a price that was below the retail prices of other wholesale bakers' products (6641 Vail. A general price drop on private label and secondary label bread in late I968, which lasted for five months almost wiped (Granddad) out of business" (6642-44 Vail). The price of advertised bread was not reduced during this period (6644 Vail).

164. Sometime between February 27, 1969, and March 4, 1969, the wholesale price of Continental' , Hansen s and American private lahel and secondary label bread increased (CXs 1003; 1005; 1006; 1192 Stip.; 1197 (175- I 76) Thompson). Granddad raised its wholesale price back to 22 and started to make money again and increased its volume of sales (6646 Vail 165. During the last quarter of 1968, Continental' s Seattle bakery incurred a monetary loss on its bread production and sales (RX 703B). I66. In the spring of 1969, Thompson, Continental's Seattle bakery plant manager from 1966 until I977, met with Taylor, the plant manager of Mother Hubbard (CX 1197 (125, 129) Thompson). Thompson told Granddad's supplier that "the Granddad price was not in keeping with the standard market price of what bread seemed to be (CX 1196 (166-173) Taylor). Shortly thereafter, Vail and Sandland met with Thompson in the latter s offce at the Continental bakery (6784-85 Sandland). Granddad's oficials testified that Thompson asked them what the backhaul business meant to Granddad, advised them that bread prices were going to be increased again shortly and that Granddad should go along with the price increase so that everyone could make money. They further testified that Thompson told them that if Granddad did not raise its prices, Continental would reduce prices to five cents a loaflike it had done in other parts of the country (6648-49 Vail; 6785-86 SandI and). They testified that Thompson also told them that Granddad should not seek any more business (6649 Vail; 6786 Sandland). Thompson denied making such state- 280 Initial Decision ments, but admitted that he had discussed the backhaul business with Vail and Sandland (CX 1197 (200-02) Thompson). 167. Granddad did not raise its prices and did not stop seeking new business. Shortly after the meeting in Thompson s offce, Granddad got the K-Mart account which added 13 000 loaves per week to Granddad' s volume. About a week later, Granddad got the Food King account which added 8000 loaves a week (6650-51 Vail; see 6786 Sandland), at which time it reached its second volume peak (6782 Sandland). (42) 168. Also sometime after the meeting in Thompson s offce, Continental discontinued the backhaul business with Granddad but reinstated it within six weeks because Mother Hubbard complained about the poor service it was getting from the transport through Vancouver (CXs 1196 (165) Taylor; 1197 (204--5) Thompson; 6652, 6719 Vail; 1197).

I69. Also about that time, Thompson had a survey prepared to determine how much Granddad bread was being sold and where it was being sold. Thompson recognized that Granddad was a "factor in the market" and "could cause some problems" (CX 1197 (180-84) Thompson).

170. Granddad had displaced a secondary label of Hansen when it acquired the K-Mart account. Joe Collins, secretary of Local 227 ofthe Teamsters Union, pressured K-Mart to discontinue carrying Granddad bread (67947 Sandland; 6706-7 Vail). Granddad also lost Johns IGA when the Union threatened to picket that store for carrying Granddad bread (6705--7 Vail; see RX 700). 171. Granddad ceased buying bread products from Mother Hubbard in 1972 , when it purchased the Pink Elephant bakery (See 4379 Stolle). The favorable exchange rate between the United States and Canadian dollar had disappeared and Vail and Sandland thought they could produce bread more cheaply themselves. They withdrew from the market shortly thereafter. In I972, Hansen closed its Seattle bakery and consolidated its baking operations in its Portland plant (RX 707D).

172. Although Continental' s sales of white bread in units increased during the period 1970 to I973, since then the volume has returned to the 1970 level (RXs 1025 in camera 1031 in camera). Ashbrook is the leading seller of white bread products and Gai's is the leading bread baker in the Seattle area. Continental is second to Ashbrook in white bread sales (9302--5 Frielink). Complaint counsel estimate that Continental has over 30% share ofthe wholesale white bread market in Seattle (See CCPF I0-4).

I73. It is found that none of Continental' s actions in Seattle, as demonstrated on this record, are suffcient to support a conclusion Initial Decision 104 F. that Continental, as a single firm, engaged in conduct that could be construed as an attempt to monopolize or as an unfair method of competition in violation of Section 5 ofthe Federal Trade Commission Act. Although it could be inferred that Continental's secondary label prices during the 1968 price war were below its fully allocated costs there is nothing in the record to show that it initiated these low prices. Complaint counsel do not allege any ilegal price discriminations. In my opinion, the introduction of secondary label bread into a market is not ilegal. Merely competing with Granddad is not an ilegal attempt to monopolize. All of the other events relied on by complaint counsel involve efforts of salesmen to sell their products. The withdrawal of( 43) Continental's business from certain stores did not injure either the grocers or Granddad. It would appear that the volume of business in Continental products involved in those stores was minimal and Continental's claim that it was not economical to continue to service these accounts was probably reasonable under the circumstances. Moreover, it appears that Granddad's principal adversaries in Seattle were Local 227 of the Teamster Union, Hansen, and the driver salesmen of all the wholesale bakers. Los Angeles- Ventura Marketing Area 174. Complaint counsel challenge certain pricing practices engaged in by respondents in Southern California generally, and in the Los Angeles and Ventura areas in particular, during the period 1966 through 1974 (CCPF 11-36-11-I38; CC Mem. I20-128; CC Reply Br. 79-92).

175. Continental's Beverly Hils and DiCarlo (San Pedro, California) bakeries were part of its Southern California or Los Angeles Region (Santa Monica) which also included bakery plants located San Diego, Denver, Colorado, Ogden and Salt Lake City, Utah, as well as the Hostess Cake Kitchen, a cake bakery, also located in Los Angeles (CXs 5I8; 5I9; 4550 A. Gordon; 4677- , 4693-94, 475I J. Walsh). Continental considered Los Angeles and Ventura as part ofthe same market, served by the same major competitors (4747, 4805-06 J. Walsh).

176. The Beverly lElls plant had both conventional mix and continuous mix process lines whereas DiCarlo Bakery was a conventional mix plant. The San Diego Bakery, originally a continuous mix plant was converted to conventional mix in the 1970's (4691- 4723-24 J. Walsh). These bakeries supplied each other with bakery products (4688-92 J. Walsh).

177. Continental sold bread products produced by its Beverly Hils and DiCarlo bakeries to customers located in Arizona (9617 Nuziard). 280 Initial Decision These products were sold and delivered through its San Diego bakery (3182-83 Frielink; 4697 J. Walsh; 4759 Stip. 178. The Beverly Hils bakery had over 200 bread routes originating from its plant or from depots in its marketing territory. The DiCarlo Baking Company served its smaller marketing territory with approximately 80 routes (See CX 518; see also RX 290; 4828-29 J. Walsh). Although these two bakeries called on some of the same grocery customers, they did not generally sell products wrapped in the same labels to the same customer (4697, 4749, 4844 J. Walsh). The two plants did not generally serve the same institution, school or restaurant accounts (4697-98 J. Walsh).

179. During the period 1966 through 1974, Continental's labels at its Beverly Hils Plant were Wonder Bread, Wonder Country Style Holsum, Profie, Home Pride White and Wheat and Wonder Wheat. The labels used by the DiCarlo plant were (44) DiCarlos Bread, 0 Settlers, DiCarlo Parisian French, DiCarlo Sour Dough French DiCarlo Taix bread, and until 1974, Big D and Mr. Big (4683-84 J. Walsh).

180. The other major wholesale bakers of white pan bread and rolls distributing products to grocery stores and restaurants in the market area serviced by Continental's Beverly Hils and DiCarlo bakeries in the late 1960's and early 1970's were Interstate, American, Prosser Baking Company ("Prosser ) and Gordon Baking Company ("Gordon ) (4522-23 A. Gordon; see 9598-99 Nuziard). 181. By the 1970's Interstate had five or six bakeries in Southern California, two of which produced white pan bread (4527 A. Gordon; 10959 Vander Geissen). Interstate closed its Long Beach, California bakery in 1974 and its Santa Barbara bakery in 1964 or 1965 (RX 1051A- in camera). Interstate sold its bread under the Weber, Blue Seal, Milbrook and Log Cabin labels (4706, 4782 J. Walsh; 10959 Vander Giessen). Roman Meal, Hollywood and Jack LaLanne breads were also baked by Interstate (4769 J. Walsh). 182. American Baking Company, which had acquired Langendorf Bakery around 1964, had four bakeries in the Los Angeles area, two of which were located in Los Angeles, with the others located in San Bernadino and Pasadena, California (8852-55 Murray). American served both the Los Angeles and Ventura markets as well as San Diego and Bakersfield (4527 A. Gordon; 8855 Murray). By 1974 American had closed three of its four plants and, in December 1974 American terminated its grocery routes and withdrew from the grocery white pan bread business, redirecting the sales efforts of its Los Angeles Bell Gardens plant to the restaurant trade, concentrating on fast food establishments (RX 377; 8867-68 Murray; 4996 G. Gordon; 4897-98 Mitchell; 4713-14 J. Walsh; see 9608 Nuziard). Initial Decision 104 F. 183. Gordon was founded in 1906 as a family business. Although the business was sold to Interstate in I930, the Gordon family continued to operate the bakery for Interstate until 1936 when Gordon repurchased it. The bakery, modernized in 1963, used the conventional mix process (4510- I5 A. Gordon; 4924 G. Gordon). In I966, the Gordon Bakery was running two shifts with some overtime (4519 A. Gordon). 184. The Gordon plant, located in the center of Los Angeles, distributed its products in the greater Los Angeles area including Santa Monica, South Laguna, San Bernadino and San Fernando, California (CX 516; CX 519; 4511- , 4583, 4661 A. Gordon; 4919, 4950-51 G. Gordon; see 4661 A. Gordon).

185. Gordon s primary product, the 1 lb. expanded loaf, was distributed under the Gordon label as well as several private labels (4520, 4522-23 A. Gordon; see 4517 A. Gordon). It also sold bread to grocery stores under the Springfeld label through Certified Grocers, a grocery-owned cooperative (RX 207 A-B; 4607-(45)09, 452I A. Gordon). In the mid 1960's, Gordon s had a market share of3%-5% (4917 G. Gordon).

186. Prosser was established in early 1939 (4983 Prosser). Its plant was located in Ventura, California, a coastal community located approximately 65 miles northwest of Los Angeles (CX 517; 4984 Prosser). Prosser distributed its products in the coastal area from Paso Robles south to Burbank and eastward to the San Fernando Valley (CX 517; 4987-89 Prosser). It sold bread products under the Banquet Bread, and Family Bread labels (4993- , 5006-7, 5I11-I2 Prosser). During the period in which Gordon made Springfeld bread for Certified Grocers, Prosser made and distributed that label in the Ventura area (RX 207A-B; 46I9 A. Gordon; 5044-45 Prosser). Prosser s plant which used the continuous mix process had been expanded and modernized in 1964, and, in 1966, was operating two full shifts plus overtime (4985, 5002 Prosser).

187. Some other wholesale bakers included Eagle Baking Company which sold exclusively to restaurants and went out of business on February 27, I971 (CXs 481, 483; 4523-24 A. Gordon; 4707, 4715, 4771-72 J. Walsh), Modern Baking Company, which produced variety breads and sold some white sand which bread to grocery stores (4524 A. Gordon; 4711 J. Walsh), and Oroweat Bakeries, primarily a variety bread baker which sold some white pan bread to grocery stores (CX 460N; 4533 A. Gordon; 4783 J. Walsh). Several San Diego bakers sold white pan bread in sections ofthe Los Angeles market (4710 J. Walsh). Golden Creme, a cooperative bakery owned by several grocers, sold bread to members of the cooperative and other grocers. It was purchased by Food Fair in 1971 and became Food Fair s captive bakery (4523 A. Gordon; 4747, 4772 J. Walsh).

g., 280 Initial Decision 188. Interstate was the largest producer of white pan bread in the Los Angeles area during the period 1966 through 1974, with the Weber label as the largest sellng brand (RX 228A-B; 4646 A. Gordon; 4782 J. Walsh; see8887 Murray; 9567, 9607-08 Nuziard). Continental and Interstate s shares of wholesale bread business with grocery stores were approximately the same by 1974 (9607-08 Nuziard). 189. In 1966, the captive bakeries in the Los Angeles area were Safeway, Ralph' s and V ons. Lucky opened its captive bakery in 1967 (4585 A. Gordon; 4772-74 J. Walsh). By 1974 other major grocery supermarket chains were operating their own captive bakeries. Food Fair had purchased Golden Creme. Alpha Beta, which had discontinued its captive bakery in 1961 in favor of private label bread resumed baking its own bread in early 1972. Albertson s started baking its own bread in 1973 or 1974 (RXs 239A; 246B; 367 A; 4523, 4526 4532 4539 A. Gordon; 4747, 4772-73 J. Walsh; 9568 Nuziard). 190. Particularly after 1966, most supermarket chains which did not have captive bakeries purchased private label bread from wholesale bakers (4523, 4534, 4539 A. Gordon; 4772-73 (46) J. Walsh; CXs 409; 412; 416-21; 429-430; 442-445; 447-452). 191. The most important loaf size in the Southern California area was the 1 lb. expanded loaf of white bread (8854 Murray; 4734 J. Walsh; 4995 Prosser).

192. Continental considered that it was generally the leader of price increases in Southern California on advertised label and private label bread. (CX 490B).

193. Before 1966, the major grocery chains did not have many outlets in the Ventura area (4997-99 Prosser; see also 5002 5069- Prosser). Prosser had a 40% share of the Ventura market in 1960 while its competitors, operating from depots in the Ventura area, had lesser shares: Interstate 30%; Continental 20%; and American 10% (4989, 4993, 5004 Prosser).

194. Jordano, Williams Brothers, Scolari's, Santa Cruz, B&D McDaniel's and Bayles were Prosser s most important grocery store customers in 1966 (5007 Prosser). Prosser did not sell private label bread to grocery stores and 70% of its sales were of its primary label bread, Prosser Banquet (5007 Prosser). Grocery store customers accounted for approximately 50%-60% of Prosser s sales (5099-5100 Prosser; see also CX 505A-I; 5055-58 Prosser). 195. As the major grocery chains expanded into the coastal area the major wholesale bakers, including Continental, began granting discounts to the other grocery stores, discounts which Prosser met (RX 363F-H; 5009 Prosser; see 5I3I-32 Thompson; 5160 Hauser; see also 5040 Prosser).

196. In 1966, Prosser still had 40% of the market and Continental Initial Decision 104 F. had increased its share to 30%, whereas Interstate share had declined to 20% and American share remained at 10% (5008 Prosser). 197. During 1966, Prosser developed, proposed, and successfully negotiated a private label program for the approximately fifty Food Fair Grocery Stores in the Los Angeles area, some of which he had been supplying with Prosser Banquet label bread. Prosser began distributing private label bread to Food Fair s eleven stores in the Ventura, Santa Barbara and San Fernando area. He leased a warehouse in central Los Angeles from which he planned to distribute private label bread to the rest of the Food Fair stores in the Los Angeles area (CX 519; 5010-12 Prosser; 5136-38, 5I41 Ashton). 198. Plans to establish a Los Angeles depot, as were the plans to serve Food Fair stores in Los Angeles, were abandoned by Prosser after discussions with representatives of Interstate and Continental (5014 Prosser; 5137--0 Ashton). Prosser did continue to serve some Food Fair stores in the San Fernando Valley while Continental, and later, Interstate, supplied (47) private label bread to Food Fair s Los Angeles stores until it purchased the Golden Creme Bakery (CX 444A- J; RX 354M; 4747 J. Walsh; 5014 Prosser; 5139--0 Ashton). 199. In late 1966 Continental, Interstate and American reduced their wholesale prices of lib. expanded advertised label bread by 6 (5019-24 Prosser). At the same time, these wholesale bakers reduced their prices on eight-pack hot dog and hamburger buns (5026 Prosser). Prices did not increase until March 19, 1971 (5025 Prosser). 200. Beginning in 1966 and continuing at least through 1968, Continental engaged in widespread sampling of Prosser s restaurant accounts (5026 5028-29 Prosser; 5122-23 Thompson; seeCX 424A- , D , 2-, Z-9, Z-16-I8; see also CX 2361A-I; RX 378B, D, F; 5028-29 5077-78 Prosser).

201. In 1967 Continental introduced its secondary bread under the Holsum label in the Ventura area (503I Prosser but see 5IOI- Prosser (1964 or 1965)). The secondary label 1 lb. expanded white bread, which later was replaced by private label bread, was sold at a retail price of51oaves for $1.00. In 1966, Prosser s fully allocated cost of baking and distributing 1 lb. expanded bread was 20 (5006 Prosser). Prosser lost much shelf space in its principal accounts to Continental's Holsum label (5035- 42 Prosser). 202. As a result of these pricing moves by Continental and the other wholesaler bakers Prosser lost many restaurant and grocery store customers. Prosser s sales were reduced to some of its principal grocery accounts, Prosser was required to grant discounts to hold business, and Prosser sold bread products at net prices below its costs (5027--4, 5050-51, 5079, 5099-5100 Prosser; 5I23-24 K Thompson; 5149-50 Hauser).

g., 280 Initial Decision 203. In November 1969, after Gordon replaced Continental as the supplier of Certified Grocers' Springfeld private label bread, Gordon entered into an arrangement whereby Prosser would bake and distribute Springfeld Bread to grocery stores in Prosser s marketing area (RX 207; 5044 Prosser). Prosser, however, did not get the Springfield private label business in many of the grocery stores Continental had previously supplied with Springfeld bread. The stores substituted Continental' s secondary label, Holsum for Springfeld, apparently at a better price (5045-46 Prosser).

204. In June of 1970, Continental Beverly Hill bakery advised its Regional Offce that future sales opportunities might open up because (CJontinued sellng pressure may force withdrawal of some wholesale competition from the market. Weak bakers in the market are American, Prosser and Eagle" (CX 410B). "(EJliminating competition" was included in the report "(A)among the methods for obtaining sales" (CX 410J). (48) 205. In 1970 and 1971 Prosser lost more grocery and restaurant customers when he could not meet or hold prices at Continental's levels since they were below Prosser s cost. In 1971 Prosser s fully allocated cost for lib. expanded bread was 22 compared to 20-21 in 1966 (5006, 5051-53 Prosser; RX 280A-U; CX 441N, Z-9, Z-22 Z-26, Z-188, Z-189, Z-228-339; 4799-4803 J. Walsh). In 1970 Continental' s price of around 18 for private label lib. expanded bread was below Prosser s fully allocated costs (5042 Prosser). In 1971 Prosser tried to meet Continental and Interstate prices by selling private label bread to Willams Brothers and Jordano s. Prosser lost those accounts when he tried to raise the 18.5 price to cover his fully allocated costs of 22 (5051-53 Prosser).

206. In 1970 Continental was ranked first in the sale of the dominant I lb. expanded loaf in the Ventura area (5146 Ashton). In 1971 Continental' s share of the Ventura market was 40%, Interstate ranked second at 30%, Prosser had 20%, and American 10% (5062 Prosser; 5146-7 Ashton).

207. By 1971 Prosser s weekly route average had fallen to approximately $1000 from $1400 in 1966 and the number of routes had declined to 23 from 39 in 1966 (5002, 5038 Prosser). Prosser s bakery was profitable until 1967 and ran at a loss until it closed, with the exception ofi969 in which the company showed a marginal profit (CX 505A-I; 5057-58 Prosser).

208. Prosser discontinued its wholesale routes on November 20 1971, and after baking Pepperidge Farm Bread for a time, Prosser ceased all operations on February 20, 1973 (4983, 5054, 5059--0, 5063 Prosser; see CX 505A).

209. Continental made plans to fill the vacuum left by Prosser Initial Decision 104 F. imminent exit by expanding its routes and coverage in the Ventura area. One driver testified that he was offered positions with Continental even before it became evident that Prosser would soon close (CX 480C; seeCX 480F; ex 470D; 5125-27 Thompson; seegenerallyCX 483; RX 365A-B).

210. Sometime after Prosser went out of business, Continental discontinued sales of Hal sum in the Ventura area (9614 Nuziard; see also 9676 Nuziard).

2II. Continental has established itself very firmly in the Ventura market since Prosser withdrew. Continental now has the greatest share of that market (10945-46 Vander Giessen). 212. The introduction of secondary and private label bread in the Los Angeles area by the major wholesale bread bakers also had an adverse effect on Gordon s sales (4549 A. Gordon). In 1966 or 1967 Gordon s lost the Market Basket account, which comprised approximately 20% of the bakery s volume, to American which offered private label at a price below Gordon s cost. Gordon then sought a private label program at Boy s Markets and redirected his sales efforts to fast food restaurants (4556-57 A. Gordon). (49) 213. Until the loss ofthe Market Basket account, Gordon had not sought other private label business, since it felt such business was not profitable. The Boys account, marginally profitable to Gordon s when first obtained, did not continue to be profitable because Gordon was forced to lower net wholesale prices below its costs to meet periodic offers from Continental (4563--5, 4570-71 A. Gordon; 4894-96 Mitchell). Gordon started the private label account with Boys because it needed the volume (4563--5 A. Gordon). Gordon s sales to Boys amounted to approximately 35 000 to 40 000 units per week and totalled yearly sales of approximately $900 000 (4562 A. Gordon; 4891 Mitchell 214. Gordon s obtained the account at Jack-In-The-Box, a rapidly expanding fast food chain (4558-59 A. Gordon). Continental eventually acquired the account which it served at prices below Continental' costs (CXs 465; 466; CX 49IA). In January 1974, Continental lost over $82 000 from its sales to Jack-In-The-Box and over $4000 from sales to Denny s (a similar fast food operation)(ld. ). In March and October 1974, Continental stil reported losses on the Jack-In-The-Box account (CXs 491D, E; 465; 466).

215. When Gordon took over the Certified Grocers Springfeld Bread account from Continental in 1969, Continental ofiered the grocers it had been selling Springfeld bread, reduced prices on secondary label bread. As a result, only 50% of the grocers continued with the Springfeld program (4555 A. Gordon).

216. After Continental began to lose the Alpha Beta private label g.;

INTERNATIONAL TELEPHONE & TELEGRAH CORP., ET AL. 337 280 Initial Decision business in 1971 , it actively pursued private label business (See 4547 A. Gordon; CXs 53A; CXs 409; 412; 416-21; 429-430; 442-445; 447-452).

217. Continental discriminated in price between Wonder products and private label bread (CX 400). For example, Continental's wholesale price on the 1 lb. expanded Wonder advertised label loaf was 23.4 during the period from June 1970 to June 1972 (CX 400) whereas the wholesale prices on the same sized private label products were 20 to Alpha Beta and 21. to Albertsons. In May 1973, the wholesale price on 1 lb. expanded Wonder was 26.4 , and the private label wholesale prices were 23 to Stop N Go and 19. to both Market Basket and Heritage House (CX 4002-30; 2-7). 2I8. As a result of Continental's efforts to obtain a larger share of the secondary and private label market, Gordon s lost much shelf space for its primary line bread, since Gordon s was no longer able to control the bread table (4515- , 4537- , 4560-61, 4597 A. Gordon; see 4719-20 J. Walsh).

219. Respondents Profit by Variety" study for its Beverly Hills bakery shows that almost all of Continental's private label white bread, including the lib. expanded loaves, were (50) sold at wholesale prices below Continental' s fully allocated costs (See CX 441). 220. In August I973, Continental noted a "strong probability" that American and Gordon would be forced to withdraw from the marketplace resulting in an increase in Continental's market share of 5% to 6% (CX 7 (L-2)).

221. During the 1972-1973 period Gordon supplied branded bread to Hughes Markets' twenty- five stores (4936-37 A. Gordon; 5068 Prosser (including one in Oxnard)). In 1973 Continental offered Hughes a private label program and obtained the account. Continental was aware that the loss of bread table space at Hughes would put Gordon out of business. Hughes accepted Continental's offer of 17. for lib. expanded bread (RX 278 A-E; 4796-99, 4840 J. Walsh). This account was projected to yield $10 000 weekly in sales (RX 278A). 222. An examination of Continental' s Profit by Variety study relating to the Hughes private label account reveals that Continental was sellng all of the products which Continental supplied under Hughes private label below fully allocated cost. Borne of the items were close to being below variable cost as well (CX 441 K, T, 2-20, 2-28, 2-37 2-2 2-52, Z-137, 2-139, 2-220, 2-221).

223. In March 1974, Beverly Hils sales to Hughes totalled $65 303.18 of which $19 639.86 (29.66%) was for Wonder bread products. The sales of Wonder Bread (assuming a profit) would not have been enough to offset the unprofitable private label program sales (CX 461B; CX 493A).

.. .. (, Initial Decision 104 F. 224. On June 22, 1974 Gordon leased its baking plant to Certified Grocers in order to avoid going bankrupt (4510, 4520-2I, 4575 A. Gordon). Thereafter Certified produced Springfeld private label bread for the member grocers. The bread products produced by Certified included only white bread and bread type rolls and the Springfield label products were delivered on a store drop basis without returns by salaried employees, similar to the distribution methods of captive bakers, except that Certified did not control the shelf space of its members. (See 1I334-46 Andikian; see also 10297 Jakacki). In addition, Certified sold some private label bread (other than Springfield), and solicited this portion of its business in competition with the other wholesale bakers (4608 A. Gordon; 1I340-3 Andikian). Certified did not sell advertised label bread and did not sell to restaurants or institutional accounts (1I337-38 Andikian). 225. Thus, in 1974, with the withdrawal of American from the grocery segment of wholesale baking and the withdrawal of Gordon from the advertised label business, Interstate and Continental became the dominent wholesale bakers in the Los Angeles area (4529- 4574 A. Gordon). Each had approximately 40% of the white bread market (4531-32 A. Gordon; see CX 493A). (51J 226. It is found that, during the period 1966 through 1974, Continental discriminated in price between purchasers of its advertised label bread and purchasers of its secondary label and its private label bread and that during most of this period the lower discriminatory prices were below Continental's fully allocated costs. In most instances, these lower prices were not granted to meet the equally low prices of any of Continental's competitors. It is found that Continental' s pricing practices contributed substantially to the failure of Pross- , to the withdrawal of American from the wholesale bakery business, and to the transfer of Gordon s baking operation to Associated Grocers. As a result, the wholesale white bread market has become more concentrated, with two multiplant bakers sharing most of that market.

Northern California Marketing Area 227. Complaint counsel challenge certain pricing practices engaged in by Continental in the Northern California marketing area during the period 1972 through 1974. More particularly, complaint counsel challenge the low wholesale prices at which Continental sold private label products to chain store grocers throughout that marketing area from July 1972 until 1974 (CCPF 12- 12-222; CC Mem. 129-137). 228. The Northern California marketing area included the San -11." t--.." l H" llou l,f'n"''' f'''''TY V..o:Qnf) tn , .. .

.1.."1 .LCI.L\..."In..L.L,,.."I.nu .LClu...a .L.L,,.."ICI U( .LCluClU.Lld.L.L.I V"-'' .OU. 280 Initial Decision seeReading as well as Reno and the Lake Tahoe area (7914 Stip. 781; 3251-52 Frielink).

229. Continental's San Francisco, Sacramento and Oakland bakeries sold bread products in this Northern California marketing area. These bakeries were part of Continental' s Northern California Region (Redwood City) which also included bakeries located in Honolulu Portland, Seattle and Spokane (3163 Frielink). 230. Other wholesale bakers in this marketing area in the early 1970' s were Campbell-Taggart which had six plants, American, William Inglis and Sons Baking Co. ("Inglis ) and Welsh Baking Company ("Welsh") located in Reno (3852 Albrecht; 3183-84 Frielink). 231. Continental' s Oakland bakery produced variety and pan white breads. Its San Francisco and Sacramento bakeries produced cake products as well as variety and white pan breads (3162-63, 3175 Frielink; 4139-40 Heaps). In 1974, the San Francisco bakery was converted into a cake producing plant and Sacramento s cake production was transferred to the San Francisco bakery (3163 Frielink). These bakeries supplied bread products to each other (3161- , 3175 Frielink). 232. Continental's advertised label was Wonder while its secondary labels were Country Style and Home Pride (3175, (52) 3I80 Frielink). In 1970, Continental' s private label customers were United Grocers a voluntary cooperative of retail grocers ("Bonnie Hubbard" label) (3214 Frielink; CX 771A-B), Albertson s in California and Nevada (3185- 3218 Frielink), Wentz Markets, Cola Markets and Mayfair (CX 61OA).

233. At all times relevant hereto, Continental sold bread products produced by its Bay Area bakeries to grocery customers located in Nevada (CX 61OA).

234. In the relevant time period, Campbell-Taggart had two subsidiaries in the Northern California marketing area, Rainbo and Kilpatricks. Rainbo operated four plants located in Chico, Sacramento Modesto and Fresno (3184 Frielink). Its advertised label was Raini,o. In late 1970, Campbell-Taggart was selling private label products to United Grocers and Farmers Market located in the northern part of the marketing area (CXs 771B; 610A).

235. Kilpatricks operated two plants located in San Francisco and see CX 785). Its adver-Oakland (3184 Frielink; 4956, 4965 Roberts; tised label was Kilpatricks (3356 Frielink). From 1970 to I974 Kilpatricks sold private label to Alpha Beta (4961 Roberts). 236. In the early 1970's American had one bakery in the Northern California area located in San Jose. It distributed bread products throughout the Northern California Marketing area (CX 784; 3852 Albrecht). American s advertised label was Langendorfwhile its larg- .. .. .... _.. Initial Decision 104 F. est private label customers were Mayfair and K Mart (CX 610A; 3336 Frielink).

237. Inglis had one bakery located in Stockton (CX 782; 3416-17 Sitter). This plant was rebuilt and modernized in 1957 and it used a continuous mix process (3717 19 W. Inglis). In 1966, Inglis acquired the Sunlite bakery located in San Jose, but transferred the production of that plant to San Jose in 1969 (3417 Sitter; 3728-29, 3753, 3784 W. Inglis). Inglis' advertised label was Sunbeam and it sold some bread products under the Sunlite label (3340 Frielink; 3416-17 Sitter; 3611 Jones). In 1970, its private label customers were Farmers market and Save Mart (CX 61OA).

238. Welsh, located in Reno, sold bread products under the Welsh and Sheepherder labels (3340 Frielink; 3515-16 Biechner; 3956 McGinley. It was purchased by American in 1971 (3466-9 Biechner; 3957 McGinley). It had tried to enter the Sacramento area in the late 1960' , but was not successful (3466-9 Biechner). 239. Interstate, which had entered the Northern California market in 1954 by purchasing bakeries in Sacramento and Oakland, sold these bakeries in 1967 and 1969, respectively, and withdrew from the market (3941 Hunter). (53) 240. In October 1972, the approximate market shares of the wholesale bread market in the Northern California area were Campbell- Taggart 40%; Continental 30%; American 25%; and Inglis 5% (RX 103C; 4046 McCarthy).

241. Continental was considered to be a price leader in the Northern California area in that its wholesale price increases were usually followed by other wholesalers (3891-92 Johnson). 242. The principal captive bakers in Northern California in the early 1970's were Safeway, Lucky Stores and Alpha Beta (3228, 3353 Frielink; 3429 Sitter; 3871 Albrecht). By 1974, Safeway s market share of all white bread was 2.7% in the Sacramento area and 11.0% in the San Francisco/Oakland area. In 1974, Lucky Stores' share of the total white bread market was 9.0% in the Sacramento area and 6.8% in the San Francisco/Oakland area. Lucky Stores used the Lucky and Harvest labels. The combined shares of Safeway and Lucky Stores were 17.8% in San Francisco/Oakland area and 11. in the Sacramento area (CX 141).

243. From 1970 to 1974, the lib. expanded loaf of white bread was the largest volume private label product in the Northern California area and was the focus of price competition among the wholesalers as well as the retail grocers (3514 Biechner; 3676 Jones). 244. Private label and captive label bread became an important factor in the Northern California market area in the latter part ofthe 10CfV.. (')0"7'7 '7Q i\ 11.-.... 1..., rr1... !'_.. r....._ 1... 280 Initial Decision was the "Bonnie Hubbard" program which Continental negotiated with United Grocers Cooperative, the largest grocery cooperative in the Northern California market (3214 Frielink; CX 771A). Although the wholesale price for the 1 lb. expanded loaf of Bonnie Hubbard private label bread was originally announced as 20v with stale returns and 19v without stale returns, before the program started on July 1, 1968, Kilpatricks offered to sell its advertised label bread to Albertsons at 18 . Continental lowered its price on 1 lb. expanded bread to 18 at the start of the Bonnie Hubbard program (4135- 4196 Heaps; 3214 Frielink; RXs 61A; 62). The other major wholesalers also lowered their wholesale price to 18 (RX 61A). 245. Continental supplied Bonnie Hubbard bread to United Grocer s members until mid-1970, when Campbell-Taggart replaced Continental (3214, 3333 Frielink; 4186, 4189-90 Heaps; CX 771A-B). Continental gave up the Bonnie Hubbard program because it was not profitable (3214, 3217 Frielink).

246. Continental actively sought private label business of the member grocers before and after its participation in the Bonnie Hubbard program (CXs 623A-B; 630A-B; 689A-B). From 1960 to mid 1970, Continental sold private label bread to various retailers at wholesale prices either identical to, or lower than, the price at which it sold Bonnie Hubbard label bread (54) (3216-19 Frielink; CXs 604A- 605A-B; 606A-C; 607 A-B; 613A-K).

247. As a result of the vigorous price competition among wholesale bakers the discounting of wholesale prices to retailers became prevalent (CXs 614A-C; 619A-C; 621A-B). Discounts were offered on private label as well as secondary and advertised label bread (3215-18 Frielink; CXs 610B; 759Z-24-Z-28; see 4184-88 Heaps). 248. From mid-1969 unti September 16, 1970, the prevailing wholesale price for I lb. expanded private label white bread was 17. 2v per loaf. Inglis considered this price to be "below the cost of doing business" (RX 61A). From September 16, 1970, until July 10, 1972, the prevailing wholesale price for the lib. expanded loaf of private label white bread was 18 (3376 Frielink; CXs 722; RX 2B). 249. On July 10, 1972, Continental reduced its wholesale price for the I lb. expanded loaf of private label white bread to 17. and this price was the prevailing wholesale price for over one year until August 20 1973 (3221-25 Frielink; CXs 676; 722). This price was offered to all of Continental's private label customers in the Northern California market area as the company had maintained a single price policy for private label accounts since 1970 (3221 Frielink; 4149 Heaps). Such a uniform price for private label bread was unique to the Northern California market area. The other major wholesale bakers gave Initial Decision 104 F. private label discounts in meeting particular competitive offers (3618 19 Jones; 4007, 4013 McCarthy; 4962 Roberts). 250. Although Inglis, American and Camphell-Taggart (Rainho Chico hakery) had sold 1 lb. expanded bread at prices below 18 before July 10, 1971, to several of their private label customers, these customers were located in a relatively small portion of the Northern California area (RX 31) None ofthe other wholesalers had offered the 1 lb. expanded hread below 18 to the entire Northern California market before Continental's offer of July 10, 1972 (CXs 647 A; 676; RXs 90A-C; 31; 33-40; 63A; 3380-2 Frielink; 3460-1 Sitter; 3979- McGinley).

251. Although all of the wholesale bakers in Northern California had similar costs, it appears that Continental' s costs during the period 1966 through 1972 were consistently higher than the industry average (CXs 747; 753 in camera; 3192 Frielink; 3607 Jones). Wholesale bakers' costs of producing and distributing white bread increased each year from 1966 to I974 CXs 648F; 655; 687C; 688B; 689A; 69IB; 747 in camera; 3265 3373-74 Frielink; 3607-08, 3618, 3628 Jones). 252. In 1966, Continental stated that its fully allocated costs of producing and sellng 1 lb. expanded white bread was 19. per loaf (CX 740A-B). Thus, from at least July 10, 1972, when it lowered the wholesale price of 1 lb. expanded bread to 17. , Continental was knowingly sellng that bread product (55) below its fully allocated costs in Northern California (See, 3214, 3217 Frielink). 253. Continental's losses (in thousands) on bread products during the years 1971 through 1974 were as follows (RX 3 (1971-1974); CX 1301C (1972 and 1974):

1971 1972 1973 1974 Oakland (66) (141) (435) (416) Sacramento (32) (199) (645) (639) San Francisco (192) (470) 0.1 (710 Total (290) (810) (2429) (1765) In four years Continental lost over $5 milion on bread sales in Northern California.

254. American s plant in San Jose was an extremely effcient, highly automated, modern plant (3854 Albrecht). On May 9, 1973, American determined that the total cost of producing and selling 1 lb. expanded white bread Was 25. (CX 751B). Accordingly, Continental was selling its lib. expanded bread considerably below American fully allocated costs (CompareCX 722 with 751B). In fact, assuming Continental's costs were comparable to American, the 17. price was below Continental' s average variable costs pursuant to the for- 280 Initial Decision mula adopted for this case in Finding 45 supra (80% of25. equals 20.42 255. Inglis' production costs were lower than Continental's because it used the "continuous mix" process whereas Continental was using the "Continental brew process" (3199-3200, Frielink; 3724-25 W. Inglis). During 1973, Inglis' fully allocated costs of baking and distributing I lb. expanded white bread was between 25.3343 (January 1973) and 29.939 (September 10, 1973). Thus, Continental was selling its private label I lb. expanded bread at a wholesale price considerably below Inglis' fully allocated costs (CX 722; 3617- 18 Jones). 256. During this time period, Continental guaranteed its wholesale prices to existing and prospective customers and predicted that the weaker companies would have diflculty surviving in view ofthe state of the market (CX 606B; 607 A; 3267 Frielink). Also during this time period and through 1974, Continental specifically recognized that American and Inglis were continually growing weaker, that their financial condition was not stable and that it intended to eliminate certain of its competitors from the Northern California market or portions thereof (CXs 616D; 619A; 666B-C; 686; 688B; 4165-7 Heaps). 257. For example, in a 1974 business plan, drafted in 1973, Continental' s San Francisco plant manager predicted that 1974 would present great opportunities. He stated that "(wJe are gaining more market domination for our products and as we do so, (56) our competitors fall back, and grow weaker" (CX 666B). Continental's regional business plan summary for Northern California predicted that "(courrent conditions in the industry may cause some bakers to close their plants." Continental's strategy was to " aggressively seek to increase sales tonnage through new varieties and through the assignment of specific target accounts to all sales management personnel" (CX 688BJ. In a 1974 business plan, dated June 23, 1977, Lund, manager of the Oakland bakery, targeted Louis Stores, an Inglis private label account, to advance Continental's private label program (CX 689B). In the same business plan Heaps targeted Save Mart, also an Inglis account, for Continental's private label business (CX 665G). 258. By 1972, Continental had acquired a number of chain grocers as private label customers. They included Brentwood Markets, Eagle Thrifty, Albertsons, Stop N Shop, Shop N Go, Ralphs, Bazar, Waremont, Wertz and Mayfair (CX 720; 3225 Frielink). 259. In response to private label offers to its customers by Continental before July 17 1973, Inglis lowered its price to meet Continental's offers (See CXs 722; 669; 3495-3509 Biechner; 3609-10 Jones). 260. In November 1972, the Campbell-Taggart and Continental Bakeries were closed by a labor strike. American and Inglis supplied the Northern California market with bread products, although, durg., Initial Decision 104 F. ing most ofthe strike period, they only produced the 1 1/2 lb. advertised label loaf on which they made more profit than if they had produced private label or the lib. expanded loaf of advertised label bread (CXs 2205C-D Jones; 2207F-G Albrecht). 261. On July 17, I973, Inglis sent out a "discount elimination letter to its customers in an attempt to increase its profits (3511- , 3547-49 Biechner; CX 764). American was the only other major wholesaler to follow Inglis' price move. Inglis and American thereafter lowered their prices to the prevailing 17. wholesale price (3500-- , 3511- 3547-49, 3553 Biechner; 3627-34 Jones; see CX 764). At the time of Inglis' letter, Federal price controls were in effect (See RX 75). 262. In a period following the Inglis "discount elimination letter Continental made a concerted effort to acquire private label business (RX 500G-I Letson; 4250-51 Konkel; 4389-92 Posesto; 4270-71 Teel; 4080-82 Miller; 3510 Biechner; CXs 769Z-6; 769Z-50). 263. During all relevant times. Continental's wholesale prices for its advertised label bread was substantially higher than the prices for private label. For example, from July 10, 1972 until August 6, I973 Continental's wholesale price for its lib. expanded Wonder loaf was , and from August 6 until (57) August 20 1973, its wholesale price was 26. (CX 721). During this period, its price of the comparable private label product was I 7. (CX 722). During this period the difference in price between the 1 1/2 lb. loaves of white private label and advertised Wonder label was at least 9 (CXs 721; 722). 264. Inglis first began to experience serious financial problems in the late 1960's and in 1969 fied for bankruptcy under Chapter 11 proceedings. The Inglis family personally loaned the Inglis Company approximately $4.5 milion in an effort to "save" the business (3734 3739-40, 3770 W. Inglis). It was well known in the baking industry in Northern California that Inglis had failed to make a profit from 1965 to 1976, except for the year 1972 when it supplied bread products to retail grocers during the labor strike against Campbell-Taggart and Continental (See also 376I-62 W. Inglis). 265. During the early 1970's, American withdrew from seven major areas of the Northern California market, including Sacramento, Bakersfield, Reno, Ukiah and Ft. Bragg. Thereafter American concentrated its operations in the San Jose area (3957- , 4005 McGinley; 4160 Heaps). In addition to the high costs of distribution, American attributed its territorial withdrawal to the long period of price erosion in Northern California (4006 McCarthy), lack of consumer acceptance, and the expansion of captive label bakeries (See 4006-7 McCarthy; 396I McGinley).

266. Between 1970 and 1975 American s net bread sales decreased from approximately $21 000,000 to approximately $17 700 000. Conti- 280 Initial Decision nental's bread sales increased from approximately $16 600 000 to $21 600 000 (RX 115A). Campbell-Taggart's sales increased from approximately $26,400 000 to $60 700 000 and Inglis had a slight increase from $9 800 000 to $10 800 000 (RX 115A). 267. By 1974, the only two major wholesale bakers of white pan bread sellng in the entire Northern California marketing area were Continental and Campbell-Taggart (4159-60 Heaps). 268. In 1975, Continental made a net profit for the first year since 1970 (RX 3). In 1976, Inglis withdrew from the market, Campbell- Taggart purchasing the plant and assets (9291 Frielink). 269. It is found that during the period 1972 to 1974, Continental discriminated in price between purchasers ofits advertised bread and purchasers of its private label bread and that during most of this period the lower discriminatory prices were substantially below Continental's fully allocated costs. The lower prices were not made iil good faith to meet the equally low prices of any of Continental' competitors. It is found that Continental's pricing practices contributed substantially to the failure ofInglis and that Continental knew or should have known that Inglis would be injured by its discriminatory sales below cost. As a result, the wholesale white bread market has become more concentrated. (58) 270. Complaint counsel also challenge certain wholesale price differences charged by Continental to certain customers in Nevada between I972 and 1974 on lib. expanded and 11/2 lb. loaves of white bread. More particularly, complaint counsel allege that price differences of as much as 12.4if (or 38%) between certain independent grocery stores on their purchases of lib. expanded Wonder and lib. private label bread sold to their chain store competitors had the adverse effects on competition between retailers proscribed by Section 2(a) of the amended Clayton Act (CCPF 12-187; CX 210lB Stip. 271. Continental's private label customers in the Nevada portion of the Northern California market were Albertson, Mayfair and Eagle Thrifty (3185-86 Frielink; CX 610). Continental's Wonder customers in Nevada included Clarkson s Market and Miler s Superette and Park Grocery stores (4498-500 Clarkson; 4313, 4315 Miler). 272. Clarkson and Miler paid substantially more for their Continental-produced 1 lb. expanded Wonder label than did Albertson Eagle Thrifty and Mayfair for private label bread of like grade and quality. During 1972 through 1974, the difference in price ranged from 7if in January I972 to I2.4if in February 1974 (CXs 674; 721; 722; 733; RXs I; 2; see also CXs 631; 625; 636). 273. Bread was a high volume sales item for Clarkson s and 5% of its business was in sales of bread (4503-05 Clarkson). During 1972 to 1974, Clarkson s purchased only Wonder products from Continental. Initial Decision 104 F. Clarkson never received any discounts, rebates, free goods, or cash in connection with the purchase of Wonder bread. Continental never offered private label or secondary bread to Clarkson s (4500-4 Clarkson).

274. Clarkson s usually purchased between $400 to $600 worth of white bread each week, 60% of which was purchased from Continental (Clarkson 4500). Clarkson had an overall net profit objective of 3% and a few percentage points difference in the wholesale price of bread was important (Clarkson 4503--5).

275. Mayfair was the closest grocery store to Clarkson, located approximately 1. 5 miles away. Albertson s and Eagle Thrifty were approximately 2 miles from Clarkson s. Continental supplied Albertson s and Eagle Thrifty with private label bread from 1972 to 1974 and Mayfair from January I972 until December of I973 (4502 Clarkson; CX 720).

276. The Savetime Park Grocery (Park Grocery) and Superette are grocery stores in Sparks, Nevada, operated by Ben Miller (4312- Miller). From 1972 to 1974, Miller s two stores competed with the larger chains, including Albertson s and Eagle Thrifty (4319-25 Miler). There was a Safeway store across the street from one of Miller stores (4318 Miler). (59) 277. During 1972 to 1974 all of Park Grocery s and Superette purchases from Continental were packaged under the Wonder label (4313-16 Miller). Bread was an important item for these stores (4320- 22 Miler). Miler s stores purchased approximately $1500 in white bread products each week. One third of that amount was Continental' s Wonder bread (4315 Miler). Miler always paid the wholesale list price for Wonder white pan bread and never received any free goods or cash payments from Continental on his purchase of bread (4316-18 Miler).

278. From 1972 through 1974, there were times when Miller s stores had to sell bread at cost or below cost to meet the retail price of their chain store competitors, Eagle Thrifty, Albertson s and Safeway (4321 22 Miller).

279. It is found that Continental discriminated in price between competing grocery customers in the Nevada area, and that such discriminations were, at times, substantial. The lower discriminatory prices were granted on sales of private label bread to chain stores. It is found that the effects of such discriminations, when in the magnitude of 10 to 12 a loaf(or 30% to 35%), may be to substantially injure competition between grocers paying the higher price for advertised label bread and chain stores paying the lower price for private label hread.

280 Initial Decision Cleveland 280. Complaint counsel challenge certain pricing practices engaged in by Continental in Northeastern Ohio during the period 1970 through 1974. More particularly, complaint counsel challenge Continental's private label agreement with Pick N Pay and the effects of that agreement upon competition in the greater Cleveland marketing area (See CCPF 13- 13-466; CC Mem. 13&-166). 281. Continental' s Akron and Youngstown bakeries were part of its Detroit region which also included bakeries known as Detroit Wonder, Detroit Hostess, Toledo Wonder, Toledo Hostess, Cleveland Hall Columbus Wonder and Dayton Certified. Continental's Cleveland Hall bakery produced cake and sweet goods only (5173 Miler). 282. The other wholesale bakers sellng white pan bread in the Northeastern Ohio area in the early 1970's included Interstate American, Ward, Laub, Schwebel, Nickles, Orlando, Tonys, Genest Firth and Keystone (5194-95, 5216-18 Miler; 5278 Schreck; 5537 Stonbraker). The major wholesaler bakers in the Cleveland area were Continental, Interstate, Ward, Laub and Nickels (9352 Gase; see5I94 Miler; 5318 Meehan; 5387 Kravitz). American was not considered to be a factor in the Cleveland market (5538 Stonbraker). Nickels and Orlando, a small baker of specialty items, sold primarily to the restaurant trade in the early 1970's (5705-06 Bronczek; 9353 Gase; 5537 Stonbraker). (60) 283. The Akron and Youngstown market areas were considered separate from the Cleveland Market. Competitors, competitive conditions and prices sometimes varied from area to area during 1970 through 1974 (5194- , 5216-18 Miller; 524&-50 Shreck; 9361 Gase). In Akron, the major wholesale bakers were Continental, Nickels Schwebel, Millbrook, American and Ward (935&-59 Gase; see 5195 Miler). In Youngstown, the major wholesale bakers were Schwebel Continental, Nickels and Keystone (9359-60 Gase). 284. In the early 1970' , the Akron bakery used a continuous mix process and baked only white bread (9361 Gase; CX 993"S" Davis). It ran approximately 100 bread routes from its plant or from agencies located in the Ohio communities of Cleveland, Canton, Mansfield and New Philadelphia (CXs 992L, 993D Davis; 5I60-70 Miler). In 197I the Akron bakery s bread sales were approximately $110 000 per week (5207 Miller).

285. The Youngstown bakery used a conventional mix process and baked hot dog and hamburger rolls and a wide assortment of variety breads (9361 Gase). It ran bread routes from the plant and from agencies located in Van port and Franklin, Pennsylvania and Andov- , Ohio (CX 922J; 5233 Shreck;). Approximately 70% of Young- Initial Decision 104 F. stown s sales were to grocery customers (5232 Shreck). In 1973 Youngstown s weekly route sales of white bread were approximately $35 000 (5236 Shreck).

286. Continental's Akron and Youngstown bakeries supplied each other with products and they were also supplied with additional varieties by other Continental bakeries (5259-62 Shreck; 5171- , 5183 5198-5200 Miler).

287. The Akron bakery sold bread products under the Wonder Wonder Country Style, Home Country Fair, Nancy Martin and Edwards labels (CX 2695; 5188-90, 5192-93 Miler). It sold private label products to the following chain stores: Pick N Pay, Stop N Shop, Open Pantry, T&A Saveway; Acme and Stop- Go (See CCPF 13-85 - 13- 98).

288. Interstate purchased Nabisco s Bread Division in 1968 (5301- 01 Meehan). Its Cleveland plant is usually referred to as the "Millbrook" plant (5302 Meehan). This plant used a continuous mix process (5303 Meehan; see 9462 Gase). In 1973, it had approximately 68 routes none of which were exclusively restaurant routes. The greater portion of its sales were in Cuyahoga County, Cleveland (5309 Meehan). It sold bread under the Milbrook and Good Cheer labels as well as private label bread under the Convenient, Stop N Shop and BiRite labels (5311, 5468-69 Meehan). Milbrook' s largest sellng loaf size was the 22 oz. loaf, which was 2 oz. lighter than other wholesale bakers' popular 1 1/2 lb. loaf (5480 Meehan). (61) 289. Laub, which was owned by the Georgetown Investment Company in 1970, had three bakeries (5528 Stonbraker; see RX 305). The Toledo bakery was fully automated and baked only white bread (5529 Stonbraker). Its Sandusky bakery produced only hot dog and hamburger rolls (5528-29 Stonbraker). The Cleveland plant was a conventional mix plant and produced mainly variety breads and rolls (5528 Stonbraker). In 1971, Laub had approximately 40 restaurant routes and 70 grocery routes in the Cleveland area. Approximately 58% of its sales were to restaurants, and included such products as hard rolls buns, club bread, special sandwich bread, Jewish sour rye and pumpernickel (5533-34 Stonbraker; see 5701 Bronczek). Laub's largest volume product sold to grocery stores was a 20 oz. open top loaf of white bread. It also sold a 1 1/2 lb. open top loaf and a 1 1/2 lb. sandwich loaf, as well as many ethnic breads (5535 Stonbraker; see 9380 Gase). Laub had the most extensive commercial line of bread and roll products in the Cleveland area (5779 Bateman). Laub had sold private label bread to Savemore, but lost that account to Ward in 1972 (5608 Stonbraker).

290. Ward had a bread bakery in Cleveland. That plant operated at a loss prior to June 1973 (5756 Bateman). It was operating two full . .

280 Initial Decision shifts at that time (5774 Bateman). During the period June 1972 through June 1973, Ward had gained some grocery store business and reduced its losses "quite a bit" (5783 Bateman). Ward closed its Youngstown bakery in I97I and its Cleveland bakery in 1978 (See 5368 Parks). It continued to supply its customers in the Cleveland area from its Toledo and Detroit bakeries (9425-26 Gase). 291. American had a cake plant in Cleveland and bread plants in Akron and Toledo. In 1973, it closed the Akron plant (5538 Stonbraker).

292. Schwebel was located in Youngstown and sold bread and rolls, hard rolls and hearth products in the Ohio communities of Y oungstown, Niles, Warren, Canton and Cleveland. When Laub closed in January 1974, Schwebel entered the Cleveland restaurant market at the request of some customers who did business with Schwebel in Youngstown. Schwebel operated four restaurant routes from its Cleveland depot (See 5808- I6 Schwebel). 293. Nickels, headquartered in Warren, Ohio, had five plants by the end of 1974. They were located in Navarro, Lima and Martins Ferry, Ohio, and Elkhart, Indiana, and in 1974 it purchased Tony s bakery, located in Fremont, Ohio, near Sandusky. Although Nickels sold a small quantity of bread products to grocery and restaurant customers in Cleveland prior to I974, it increased its Cleveland restaurant business after Laub closed its bakery (5893- , 5901 Gerber). Tony s bakery may have made some sales in Cleveland (5895, 5900 Gerber). 294. Genest Midwest took over the Bond Baking Company of Canton, Ohio, in the early I970' s (5898-99 Gerber). Genest (62) withdrew from the Cleveland area in 1970 or 197I and went out of business in Canton in 1972 (5899 Gerber; see 5217 Miler). 295. The record contains little information on Keystone or Firth except that these bakeries were located in Pennsylvania and that they sold bread products in the Youngstown market area (5275, 5279- Shreck; CX 994A-F Fischer).

296. In 197I Milbrook was the leading wholesale baker seller of white bread to grocery stores in the Cleveland area. Ward was second and Continental and American were tied for third and Laub was last (9352 Gase; see 5194 Miler; 5318 Meehan; 5387 Kravitz). Laub was the leading seller of white bread to restaurants with over 50% share ofthat market. Ward was second with 20% and other wholesalers had the remaining business (5705-006 Bronczek; 9353 Gase). Continental was the leading seller in the Akron area, having a slight edge over Nickels (9358-59 Gase; see 5195 Miler). In 197I Schwebel led Continental in sales in the Youngstown area (9359-60 Gase). 297. In 1970, the following chain stores operating their own captive bakeries did business in Northeastern Ohio: Pick N Pay, Fisher-Fazio Initial Decision 104 F. A&P, Kroger and Lawsons (9352, 9431, 9435 Gase; 5315 Meehan; 5219- 5225 Miller). Pick N Pay and Fisher Fazio were the largest chain stores. Combined they accounted for over 50% of the grocery store sales in the Cleveland market area (CX 968C F; see 9429- Gase; 5380 Kravitz). Other major chain stores were Stop N Shop, L' Shopper, Savemore, BiRite, Heinen, Open Pantry, Eagle, Super Val- , Convenient East, Convenient West, and K Mart (5309-10 Meehan; 5535-36 Stonbraker).

298. The most popular loaf sizes of advertised bread in the northeastern Ohio area were the 20 oz. loaf of white bread and the I 1/2 lb. loaf of white sandwich bread (9490 Gase; seeStonbraker). The most popular private label white bread were the lib., the 20 oz. and the 1 1/2 lb. loaves (CX 2695; CX 261B, E; 5419-20 Kravitz). 299. During the period 1971 to January 1974, Continental was never the price leader upward in the Cleveland market in either the grocery trade or the restaurant trade (5544 Stonbraker). Generally, a major wholesale baker could not charge a higher wholesale price that other wholesale bakeries without losing substantial sales (5216 5225 Miler). However, for a period of approximately a year, Miibrook did sell bread for a 1 higher wholesale price than Continental (5370 Meehan).

300. In 1967, five of the major bakeries in Northern Ohio were indicted for conspiracy to fix prices (RX 30lA-E). A civil action was fied simultaneously by the Department of Justice, also alleging a price-fixing conspiracy (RX 302A-G). In 1969, both actions were settled (RX 302H-Q). Cleveland was a static bread market in the early 1970' s and the volume of bread sales by the two principal captive bakers (Pick N Pay and Fisher (63) Fazio) was increasing (5613- 5619 Stonbraker). The spread between the retail price of advertised bread and captive label bread was increasing (5614-15 Stonbraker). There was excess bread bakery capacity that could service the Cleveland market, especially Continental' s Akron bakery which was operating at about 50% of capacity, one and 1/4 shift (5619 Stonbraker; 10047 Signore; CX 2634A-B; CX 2633A-F) The restaurant business, especially the fast food segment, had been growing (5658 Stonbraker).

301. In mid-1971, Continental began an aggressive program to increase its restaurant and institutional business in the Cleveland market by offering products at prices reflecting substantial discounts from its list prices (9383-84 Gase; 5545 Stonbraker; 5675 Bronczek). In this way Continental acquired the business of several large accounts previously serviced by Laub or Ward (5677 Bronczek). For example, Continental acquired the business of Interstate United at Cedar Point, an amusement and recreation facilitv located near San- ..

INRNATIONAL TELEPHONE & TELEGRAPH CORP., ET AL. 280 Initial Decision dusky, Ohio, for the summer season of 1972, by offering a I2% discount from 1970 wholesale prices ostensibly meeting Laub' s price offer (5551 Stonbraker; see CX 956A-B). Bronczek testified that Laub never offered such a discount (5680), Continental supplied bread and hamburger and hot dog rolls from its Toledo bakery, Laub had supplied Cedar Point with bread from its Toeldo bakery and buns from its Sandusky bakery (5553 Stonbraker), Continental retained the Cedar Point business in 1973 , offering a 15% discount (SeeCX 210lA- B). In the 1970' , that account was a high volume account for the five month resort season worth approximately $50 000 annually to the bread supplier (Il455-56 Bell; 5556 Stonbraker). 302. Laub also lost the Sky Chef account to Continental which offered a 12% discount (5557 Stonbraker). Laub had not offered Sky Chef a I2% discount (5573 Stonbraker; seeCX 909 (136)) and Bateman testified that Ward had not offered Sky Chef a 12% discount (5757). That account was worth about $26 000 a year in sales. 303. Laub also lost the following accounts to Continental and the annual volume of sales listed: Hospitality Inns - $18 000 (5574 Stonbroker); Ponderosa Chain - $104 000 (5575-78 Stonbraker); and Red Barn - $30 000 (5683-84, 5677 Bronczek), Laub also lost other restaurant and institutional accounts on the basis of price (5685, 5696 Bronczek; 5579, 5582, 5584-7 Stonbraker), During this period Laub discontinued 8 restaurant routes, reducing its total routes to 54 grocery-restaurant combination routes (5588 Stonbraker), whereupon Continental increased the number of its routes from 1 to I3 (5589 Bronczek).

304. After I970, the wholesale bakers granted discounts to their grocery and restaurant customers on advertised label products. Milbrook' s usual discount was 5% to meet competition (5322, 5326-29 Meehan; seeCX 931A-Z-21). Interstate (64) granted Pick N Pay a 12% discount directly from its Kansas City offce (5328-29 Meehan). Laub also granted a 5% discount to selected customers, although it would grant up to 10% to meet competition (5543 Stonbraker; 5683 Bronczek; seeCX 962A-N). Gase testified that Laub had the "cheapest prices in town" (9381), Ward reduced its discounts in November 1972 (5785 Bateman), Genest was considered to be a price cutter in the Cleveland market before it withdrew in 1970 or 1971 (5541 Stonbraker). 305. Continental learned, in early 1972, that Pick N Pay was considering closing its captive bakery and buying private label bread products from a wholesale baker (See CX 968C). At that time Continental was seriously considering closing its Akron and Youngstown bakeries. The Akron bakery, which was operating at about 50% capacity, was operating at a pre-tax loss of over $3000 per week (CX 2634A-B; CX 2633 A-F). Signore, Regional Vice President of Cantin en- ;t rp"ion, opposed these closings (CX 2635A-B; see JDL: FEDERAL TRADE COMMISSION DECISIONS Initial Decision 104 F. 2633A-D). In his view a Continental private label program for Pick N Pay would create a substantial profit for the Akron bakery and would completely open up the Cleveland marketing area and would in fact, make (Continental) the (dominant) factor on the market" (CX 2683B).

306. National Supermarket' s purchase of Pick N Pay delayed Continental's negotiations for the private label business (See CX 969H; see also RX 304). In fact National Supermarkets contacted Continental about the private label subject before the purchase (9966-7 Vail. After Kravitz became president of Pick N Pay, Vail and other Continental personnel from Rye Headquarters and Signore had many discussions with Bogomolny and Kravitz (9986 Signore; see 9974 Vail; 5397-98 Kravitz; see also CX 835). One specific condition for any private label arrangement, as conveyed to Vail, was that Continental (or any other wholesale bakery that obtained Pick N Pay s private label business) would have to compensate Pick N Pay for the value of its captive bakery (9968, 9971 Vail).

307. Thereafter Signore became Continental's principal negotiator (9986 , 10032 Signore; see CXs 886, 85, 807, 836). Although the early discussions contemplated delivery of private label products to Pick N Pay s Cleveland warehouse for store delivery by Pick N Pay, later negotiations centered on a store door delivery of private label products by Continental. Store door delivery by a wholesale baker became feasible when Continental secured a modification of the wholesale bakers contract with the local Teamsters Union that would permit any wholesale baker in circumstances where a captive bakery ceased operations, to make store door delivery by salaried employees instead of employees working on commission (9986-88 Signore; 9399 Gase; 5871-73 Bogomolny; seeCXs 805A, 809). 308. Signore was of the opinion that Laub, Omar (Fisher Fazio captive bakery located in Columbus, Ohio), Ward, Nickels and Schwebel were capable of supplying Pick N Pay with private (65) label bread (9990). MiJbrook was not considered to be in a position to supply private label to Pick N Pay because it was operating at full capacity (9989- , 10051 Signore). Continental' s Akron bakery was 29 miles from the Pick N Pay warehouse. Continental believed that it had the best capability in the area to produce the volume of bread products 1necessary to supply Pick N Pay with private label (CX 968D, G). 309. During the negotiations between Continental and Pick N Pay, ontinental personnel were advised by Kravitz that Pick N Pay had eceived offers for the private label business from other bakers rices lower than those offered by Continental (9992, 9994, 10051 gnore; see 9982 Vail). Most of these offers had been made in 1972 hen warehouse delivery was being discussed (See 9993, 10031 Si- 1.J.,I.J.I...,.nJ.I.OIJ.,..J.J..LJ.I.J.OIJ.'ClUO J.J..LU'J.""J.J.V""I. .&I."". uuu 280 Initial Decision gnore; 5394, 5446 Kravitz; 5817 Schwebel; 5868 Bogomolny; CXs 809 884). The record shows that Omar, Ward, Laub, Millbrook, Schwebel and Nickels, as well as Continental, had responded to Pick N Pay request for bids in 1972 (See 5393-95; 5408 Kravitz; 5817 Schwebel; 5869 Bogomolny; CX 884). Signore did not know the details of the offers made to Pick N Pay by any of the other wholesale bakers (9994 10031 Signore).

310. Although there was some delay in the negotiations in early 1973, Vail was confident that Continental would be Pick N Pay supplier of private label bread (CX 884). By J ne 1973, Continental was the only serious contender for this business. Bogomolny testified that Continental's proposal "overall was a lower price proposal than the others" (5870). Laub had not made an offer on the store door delivery arrangement in June 1973 (5594-95 Stonbraker; 5451 Kravitz; 5883 Bogomolny). Omar had not made a firm proposal with prices for store door delivery (5881 Bogomolny). In any event, Pick N Pay had substantial misgivings about placing itself in a position of depend ence for bakery products on its principal competitor, Fisher-Fazio, for which Kravitz and Bogomolny had once worked (5869, 5880 Bogomolny). Neither Ward nor Milbrook had pursued the matter after their initial response to Kravitz (5777 Bateman). 311. On July 13, 1973, Pick N Pay and Continental finalized their private label agreement. The written agreement (CX 803) detailed the varieties to be provided by Continental under the Edwards label drop shipped to the store door and established the following prices for the white bread products:

Variety Price 1 lb. White Bread 145 1/4 lb. White Bread 1575 1/2 lb. White Bread 8 Pack Hamburger Buns 8 Pack Hot Dog Buns 312. The agreement provided that the private label products would be of a quality comparable to Continental's Wonder (66) products. It also provided that Continental would supply a full-time merchandiser to promote and develop the full potential of Pick N Pay s Bakery Department. It provided for an increase (or decrease) in the price of the private label products if flour prices rose (or fell, the amount of price increase (or decrease) to be determined each quarter by a specific formula. This clause was subsequently revised by oral agreement to provide for review of flour costs on a monthly basis (CX 2608F). The agreement also provided that it would remain in effect for three years .... , ,,, Initial Decision 104 F. and was renewable, although it could be terminated by either party with 90-days advance notice.

313. Not mentioned in the agreement was Continental' s commitment to reimburse Pick N Pay for promotions of private label products commensurate with the estimated book value of the Pick N in camera). Conti-Pay bakery (See 9971- 9982 Vail; CX 984Anental actually paid Pick N Pay approximately $210 000 as promotional allowances, sometimes referred to as ttproof of performance from 1973 through the first three months of 1976 (CX 2682; 984Ain camera;2602A-B in camera). Continental also provided Pick N Pay specific varieties (10035with unit billing, which was a biling by Signore). Continental also hired Bil Owens who had handled Pick N Pay s bakery order desk. Thereafter he performed the same service for Continental in handling the Pick N Pay account (10017 Signore; 9406 Gase).

314. Other verbal arrangements not mentioned in the agreement included a 2 allowance to Continental for transportation and delivery of Pick N Pay s outside purchases of certain private label products, the agreement that Continental would lease Pick N Pay transport truck and bread racks and dolles for three years and the provision that Continental would pay rent for the Pick N Pay warehouse for one year (CX 2603G; see 10021- , 10033-34 Signore; 5412 Kravitz; 9407-D9 Gase). Continental also agreed to provide the Pick N Pay stores with cake display racks at a total cost of $19 000 (CX 800B). Continental was also obligated to deliver to Pick N Pay stores certain private label sweet good products that Pick N Pay was purchasing from other suppliers, the largest, Big Bear, located in Columbus, Ohio (10020-21 Signore; 5418 Kravitz). 315. According to Continental offcials, studies conducted by the Detroit Regional offce and Rye Headquarters during Continental' negotiations with Pick N Pay (studies which could not be located during this litigation) showed that the Pick N Pay private label business would be profitable to Continental on the basis of fully allocated costs (9996-10002 Signore; 9402 Gase). Continental had expected to make a "profit" of $6000 to $7000 per week on sales of private label to Pick N Pay (10002 Signore; CX 883C; see CX 805A). His estimate was for Akron and the other three plants producing Pick N Pay private label products, but did not include full fixed overhead costs (CX 2639R). In addition, it did not include the "proof of performance payment of approximately $70 000 per year, and did (67) not include certain costs that would be incurred using the Pick N Pay warehouse equipment and trucks.

316. Meehan testified that he doubted whether Millbrook could l._T' t- t- l.- 1 lh h..a"' t 1.1 h,j nt'.. ln!: (F\ .1. oorp 280 Initial Decision 930A- , E). Stonbraker testified Laub could not have made a profit on store delivery at that price (5598; CX 849A-N). Bateman testified that Ward could not have made a profit at the prices quoted in CX 803, if it had selling and delivery costs (5768-9). 317. Continental, as Pick N Pay s private label supplier, acquired the first position on the bread table (after the private label bread) for its advertised bread, which amounted to one half of the space allotted for all advertised bread (5457-58 Kravitz; 9412-13 Gase; see 5890-91 Bogomolny). In addition, Continental received authorization for a number of Wonder varieties that it had not theretofore been selling to Pick N Pay (9930-3I Dierker; 10026-27, 10047 Signore; 9412- Gase). The volume of sales in Wonder bread products increased approximately $20 000 per week (10027 Signore; 5891 Bogomolny; CX 845A-C).

318. The resetting of the bread table in the Pick N Pay stores after Continental became the private label supplier was stil controlled by Pick N Pay (5888 Bogomolny; 9411 Gase; 5410, 5448 Kravitz; I00l6 I0023 Signore). Less space was allocated to the Edwards label than had been allotted for Pick N Pay s captive label (10027 Signore). Milbrook lost its number one position for advertised bread to Continental and lost a considerable amount of shelf space (5338 Meehan; 5415 5450 Kravitz). Although Laub gained both space and authorization for varieties, it did not sell popular loaf sizes of white bread to Pick N Pay (5590-91 , 5602-03 Stonbraker; see5888 Bogomolny; 10024 Signore; 9413 Gase; RX 3009B-Z-122). Laub had complained to Pick N Pay about its shelf space assignment and because of a long time friendship between Kravitz and Beyer, Laub's president, Laub was given more rack space (5416, 5450-51 Kravitz; 10024 Signore; but see 5633 Stonbraker). Laub' s sales to Pick N Pay in all products dropped from over $IO OOO per week to $7 000 per week (5603 Stonbraker; see RX 3009B-Z-122). Laub's sales of its large white bread milk loaf dropped from over $1200 a week to under $900 per week (See 3009B-Z-122). Stonbraker testified that Laub also lost important exposure to the consumer for its advertised products in Pick N Pay stores (5605).

319. The distribution of private label products to Pick N Pay turned out to be much more expensive than Continental had anticipated (9409 Gase). Because of limited loading facilities at the Akron plant Continental had to use the Pick N Pay warehouse as a depot from which to deliver the private label products to the Pick N Pay stores (See 9404-06 Gase). There was an "explosion of orders" both on sweet goods and in private label bread (10020 Signore). The volume of Big Bear sweet goods purchased by Pick N Pay was three times the amount (68) forecast and Continental was required to use more trucks Initial Decision 104 F. (10021 Signore; 9404-band more employees to handle this product Gase; CX 976B in camera). Because of Pick N Pay s special promotions on its private label products, the volume of white bread was also greater than forecast and Continental was required to run two shifts at its Akron plant on overtime on numerous occasions (10039-42 Signore; see5419 Kravitz; CX 976B in camera). In addition, the price of flour, as well as other ingredients, and the price of gasoline and production energy increased dramatically (5336 Meehan; 5424 Kravitz; CX 976A- in camera). Continental experienced some service problems in its delivery of private label products to the Pick N Pay stores (10044 Signore).

320. After a couple of months it became apparent that the Pick N Pay program had not realized the profit that was expected to show (CX 882). In the fall of 1973, Continental began a series of studies of its Pick N Pay business (See CXs 2628; 882). One of the first studies revealed that during the four week period ending August 25 1973, the Akron Bakery sustained a $16 000 loss on its Pick N Pay business (CX 800B).

321. Only three Pick N Pay private label varieties showed a positive profit after variable costs, although these three varieties, all produced at Akron, accounted for 42% of the total Pick N Pay private label sales (CX 800C). Ingredient prices had a dramatic effect on Akron profits. Before production for Pick N Pay began the average price of ingredients was 9. per lb., but in August the average price of in- , T; see a/soCX 883A-C).gredients was I1.16 per lb. (CXs 800C; 2639R 1973, 322. Three Continental Form 452Bs dated December 28, showed that Continental' s prices to Pick N Pay on the 20 oz. loaf of white bread and the 24 oz. of white sandwich bread were below incremental costs (CX 2632C-F), whereas the price to Pick N Pay on the 1 lb. loaf of white bread was below fully allocated costs but above incremental cost (CX 2632A-B). A similar analysis on the Pick N Pay private label bun products produced at the Youngstown bakery shows that the 8 pack hamburger buns and 8 pack hot dog buns were sold below full cost, but above incremental costs (CX 2639H-K). 323. On March 1, 1974, Continental moved from the Pick N Pay warehouse to another facility in Cleveland, ceased paying for temporary employees and began using its own vehicles and drivers for redistributing private label products to the Pick N Pay stores (CX 2639S).

324. Continental's in-depth analysis of the Pick N Pay account covering a four week period during May 1974 showed sales below cost on almost all private label varieties (CXs 2663C-N; 2664A-D; 2665A- 2673-75). This analysis showed a total unfavorable figure of $38 550 for the orivate label see:ment ofthe Pick N Pav account (CX 2663D). INTRNATIONAL TELEPHONE & TELEGRAPH CORP., ET AL. 357 280 Initial Decision At that time the wholesale prices ofthe three Pick N Pay private label varieties produced (69) at the Akron bakery were as follows (CX 2673 A):

16 oz. White 1825 20 oz. White 1950 24 oz. Sandw 2275 The sales to Pick N Pay of these three items shows a total unfavorable figure of $33 528 (CX 2673A). According to Frederick Breines, on an incremental basis, Continental "basically broke even on the total (Pick N Pay) business" showing a $16 000 loss on Pick N Pay private label and a $15 100 profit on sales of branded products (SeeCX 2608B). In his opinion, the loss on private label products, both produced and purchased, resulted from not charging high enough prices to recover transportation, warehouse, delivery and administrative expenses (CX 2608C). Since the contract was signed in July 1973, there had been three price increases totallng 3. per unit (CX 2608E). In making this analysis, Breines was not aware of the "proof of performance payments that were to be made to Pick N Pay at a rate of about $70 000 per year (See CX 2693A-B). 325. According to Breines, Continental would have to obtain a net price increase of 4. per unit to break even on the Pick N Pay private label products produced by Continental (CX 2614A; see also CX 2636T).

326. Following a meeting in mid June 1974 between Vail and Kravitz, Gordon Thomas, by letter dated June 27, 1974, made Continental's proposal for a renegotiated agreement. Although containing most ofthe provisions of the July 1973 agreement, this proposal incorporated a more sophisticated formula for price changes, relating to all costs on products produced by Continental including ingredient cost changes in sugar and shortening, as well as flour. Ingredient cost were to be calculated every month and other costs calculated quarterly. The proposal also provided for actual costs incurred in transporting private label products that Pick N Pay purchased from other suppliers. The proposed agreement also contained a provision that Pick N Pay would notify Continental three days in advance of any special promotion of private label products (CX 975A- in camera). The base prices for June 1974 set forth in the proposed agreement were as follows (CX 975B in camera):

""IS FEDERAL TRADE COMMISSION DECISIONS Initial Decision 104 F. Variety Base Price 1 lb. White Bread 2244 1- 1 /4 tb. White Bread 2407 1/2 lb. White Bread 2829 327. In Continental' s analysis of its June 1974 business with Pick N Pay, the private label products produced by Continental showed an unfavorable result of $40 884 or a loss of 4. per unit (CXs 2679A- 2680A-D). (70) 328. After extensive negotiations and exchanges of views and information (See CX 2607 A-C; CXs 979A-D - 983 in camera; 2616, 2617), Continental and Pick N Pay entered into an amended agreement on September 25 1974, to become effective on October 7 1974 (CX 829A- H). This amendment generally adopted the features contained in seeCX 2616B). The baseThomas' proposal of June 27 1974 (CX 829D; prices on Pick N Pay label products were established as follows (CX 829B):

Variety Base Price 1 lb. White Bread 1792 1 1/4 lb. White Bread 2117 1 1/2 lb. White Bread 2242 329. In his memorandum dated August 26, 1973, explaining the possible impact of this agreement on annual profits before taxes Breines had presented the following projection to Continental's top offcials (CX 2616A):

Annual PST ($000' Direct Plus Full Load Direct only Plant Overhead Private Label Produced $(139) $204 $(75) Branded 135 200 135 Total ( 4) $404 $60 Memo: Proof of Performance $(70) $(70) $(70) It should be pointed out that this is the first analysis of Continenais Pick N Pay business in the record in Docket 9000 in which the proof of performance promotional payments, is included. In a 3cently discovered document it appears that Breines first learned of Ie arrangement about June 11 , 1974 (See CX 2693A-B). 330. Breines also reported (CX 2616B);

prices negotiated represent an average increase of .67 cents on all products except )z. white, 20 oz. butterball and the 8 d. hamburger buns. On these items we negotiat- 12.67 cent increase on the 20 oz. white (48.8% of tot a! volume) and the butterball 280 Initial Decision (6. 1% of total volume) and a 1.33 cent increase on the 8 cl. hamburger bun (11.9% of total volume).

331. In his suggestions as to additional areas for profit improvement under the topic "Better Mix Management", Breines stated (CX 2616C):

On a full load basis we have approximately five products that are either breaking even or making 8. 0 to 13.4% profit based on the new pricing. Three of these items are variety products family rye, hearty rye and wheat bread. The other two products are 16 oz. (71) white and 8 cl. egg hot dog buns. These items now only account for 21.7% of the total mix.

332. After the modified agreement was executed, the prices to Pick N Pay were adjusted monthly pursuant to the formula set forth therein. The price changes for Continental-produced product may be summarized as follows (CX 974A- in camera):

10/21/74 12/16/74 1/20/75 1 lb. white 1845 1863 1920 1/4 lb. who 2180 2201 2268 1/2 lb. who 2308 2331 2402 8 Pk Ham 2239 2261 2330 8 Pk. Hot 2102 2123 .2188 2/17/75 3/17/75 4/9/75 1 lb. white 1960 1937 1904 1/4 lb. who 2316 .2289 2250 1/2 lb. who 2452 2424 2383 8 Pk Ham .2379 2351 .2312 8 Pk Hot 2234 2208 2170 333. Continental's analysis ofthe Pick N Pay private label business for five weeks in December 1974 covering Continental-produced product showed a loss of $34 588. The wholesale value of these products was $221 444, the "cost to doors" was $212 606. Distribution costs totaled $38 992 (CX 2678A-B).

334. Analysis of Continental' s Pick N Pay private label business for January 1976 showed a full load loss before taxes of $25 500 and an incremental profit before taxes of$14 300. Continental's overall business in private label and Wonder/Hostess label showed a full load loss before taxes of $17 100 and an incremental profit before taxes of $5I 700 (CX 2690 A-B; see CX 2669 A- , CX 2670 A-G). The work sheets also showed that Continental' s prices on the 24 oz. Giant Sandwhich loaf and the 8 Pack hamburger buns were below Continental' incremental costs (CX 2669A).

335. Work sheets covering the first three months of 1976 showed rnitial Decision 104 F. T. that Continental's prices were also below incremental costs for three other bun varieties baked by the Youngstown bakery (CX 2668B), 336. Statistics for July 1976, showed a loss on the private label Pick N Pay business in Continental produced product on both a variable and full load basis (CX 2661C). Statistics for August 1976, showed a loss on a full load basis and a profit on a variable cost basis (CX 2686A-B). (72) 337. During the period 1971 - I979 quarterly sales by Continental to Pick N Pay may be summarized as follows (CX 2686 A-B); Sales Period Bread Sales to Pick N Pay Ending Main line Private Label Quarterly Quarterly 12/11/71 S 48 373 6/1 0/72 30,433 12/3/72 566 6/23/73 28,925 12/15/73 140 075 $654 316 6/22/74 125 788 723,09 12/14/74 143, 117 541 229 3/26/75 107 900 520,700 6/30/75 115,400 687 500 9/24/75 125 300 591,000 12/25/75 133,400 595 800 3/26/76 141 700 541,400 6/25/76 150 000 581 ,800 9/24/76 243 600 606 200 12/31/76 259 000 610 200 3/26/77 239 500 603 700 6/25/77 255 800 602 100 9/24/77 266 800 791 400 12/31/77 299 800 672,500 3/31/78 257 700 575 200 7/1/78 271 300 661 900 9/29/78 259 600 605,600 12/30/78 273 500 568 000 3/31/79 284 600 500 200 6/30/79 315 100 562 600 9/29/79 319 200 553,400 12/29/79 325 800 566 600 338. During 1973 and 1974, Continental sold its advertised label bread (Wonder) to a number of chain stores located in Northeastern Ohio, most of which were in direct competition with Pick N Pay. The I4. price on a I lb. white bread afforded Pick N Pay was not offered to any other customer of the Akron plant in July 1973, nor was the opportunity to accept store door delivery optional for all customers (CX 993M-N Davis).

280 Initial Decision 339. The wholesale list prices on branded products may be summarized as follows (See CX 912):

12/27/72 5/23/73 7/25/73 1 lb. Country Style 265 20 oz. White 355 24 oz. White 385 8 pk. ham 355 8 pk. d09 355 8/17/73 11/12/73 1/14/74 (73) 1 lb. Country Style 295 20 oz. White 395 .405 24 oz. White .425 .435 8 pk. ham .405 8 pk. dog .405 2/18/74 1 lb. Country Style 20 oz. White 24 oz. White 8 pk. ham .435 8 pk. dog .435 340. Pick N Pay promoted Edwards label bakery products weekly (10025 Signore). Meehan testified that the retail price of Pick N Pay private label bread was "more competitive" after Continental began supplying it and Pick N Pay ran more promotions on its private label bread (5341-42). Stonbraker testified that retail price competition between Pick N Pay and Fisher-Fazio immediately after Continental began supplying Pick N Pay had an adverse effect on all branded bread sales (5607-08). However, Signore testified that Continental' private label agreement with Pick N Pay did not have any effect on the retail pricing of bread in the Cleveland market (10025). On August 1974, Vance of Continental reported that the bakery manager of Fisher-Fazio blamed Continental for the low bread prices in Cleveland because it had not "forced the price up to Pick N Pay." As punishment Fisher-Fazio cut Wonder s shelf space in its stores in half (CX 828). 341. On January 8, 1974, Laub ceased its bakery operations (5610- 11 Stonbraker). Upon Laub's exit from the restaurant market, Nickels began an aggressive enort to obtain a substantial portion of the restaurant business in the Cleveland market (9355-56 Gase). Schwebel also entered the Cleveland restaurant market at the request of some of its customers (5808-16 Schwebel: 9356 Gase). Nickels gained some grocery business in the Cleveland market (9356 Gase). Both Nickels and Schwebel have increased their plant capacity in the last decade (9354-57 Gase).

342. In 1980 the ranking of the wholesale bakers in the Cleveland marketing area in sales to grocery stores was Milbrook, Continental Initial Decision 104 F. Ward, Nickels, Schwebel and American. Including captive bakers Fisher-Fazio had a third place ranking behind Millbrook and Continental (9354 Gase). A&P no longer has a captive bakery and has almost withdrawn from the Cleveland market, and Kroger sold captive bread to only six stores that do not do business under the Kroger name (9354, 9372 Gase). Milbrook and Continental are now considered to be the "price leaders" in the Cleveland market (9446-7 Gase). 343. In sales to the restaurant trade, Nickels is the first ranked wholesale baker followed by Milbrook, with Schwebel, (74) Continental and Ward sharing the balance about equally (9354 Gase). Continental's share of the restaurant market has not grown since 1971 (9355, 9383 Gase).

344. Continental' s total sale of white bread in Cleveland has declined slightly since 1974, with branded sales declining and private label sales increasing (9371 Gase). In the last two years, Continental has increased its promotion of its Wonder branded bread (9372). 345. Although the wholesale prices of private label white bread and branded bread have increased since 1974, the spread between them has also increased (9373 Gase).

346. Continental closed its Youngstown plant in 1976, and transferred the Youngstown production to its Columbus and Toledo bakeries (9377 Gase). The Akron bakery, which is now a conventional mix plant, has not been profitable since 1972 (9401 Gase). 347. It is found that Continental discriminated in price between Pick N Pay on private label white bread and all of its other customers located in Northeastern Ohio and Pennsylvania. The lower discriminatory prices to Pick N Pay were below Continental's fully allocated costs at all times between July 1973 and September 1974 and when the proof of performance payments are taken into consideration, it appears that these low prices to Pick N Pay were below Continental's incremental costs and its average variable . costs. 348. It is found that Continental knew or should have known that no other wholesale baker had offered Pick N Pay as low prices as did Continental on the private label products.

349. As a result of Continental's discriminatory pricing, as well as certain pricing practices designed to capture restaurant and institutional customers, competition between Continental and other wholesale bakers has been substantially and unfairly injured. In addition concentration in the wholesale baking of white bread has increased since 1970.

350. Complaint counsel also challenge Continental's price discriminations from July 1973 through 1974 on purchases by Pick N Pay of private label products and purchases by Valu King, of private label controlled label, advertised label and secondary label products (See 280 Initial Decision CX 2102 Stip.). Complaint counsel contend that such discriminations had the adverse effects upon competition between these unfavored purchasers and Pick N Pay prescribed by Section 2(a) ofthe Robinson- Patman Act, as amended.

351. There is no question but that Continental discriminated in price as alleged and that the discriminations were substantial (See Set 3 Adms. 36-60). (75) 352. Stop N Shop is a voluntary association of Cleveland supermarket owners who purchase and advertise collectively in order to compete effectively with the major chain stores (5493 Caster). During 1973 and 1974, the 20 Stop N Shop supermarkets were supplied in bread products by Milbrook, Continental, Laub and in a few stores Ward (5495 Caster). It purchased approximately $4 000 worth of Wonder products per week and paid the wholesale price less 5% (5496-98 Caster; see CX 846). It also purchased private label bread from Milbrook and private label buns, hot dog and hamburger, from Continental (5498 Caster; CX 959). Although its sales of private label bread were greater than either Milbrook or Wonder advertised labels, sales of these advertised labels combined was greater than sales of private label (5503-04 Caster).

353. The list price to Stop N Shop for Continental' s private label products were as follows (CX 912):

before 7/25/73 7/25/73 8/18/73 1/14/74 2/25/74 8 Pk Ham 203 228 265 272 8 Pk Dog 203 228 265 272 354. Sometime between 1970 and 1973, Continental offered Stop N Shop private label bread at 17 if to 17. 5if per lib. loaf. Although Caster would have recommended private label from Continental at the price which it charged Pick N Pay in 1973, he doubted whether the store owners would have accepted store door delivery service (5500, 5505 Caster; see CX 803).

355. In 1973 and 1974, Stop N Shop was able to "answer and compete" with any advertised price by Pick N Pay (5501, 5506 Caster). 356. BiRite Advertising Association was a group of approximately 30 independent supermarkets in greater Cleveland (5508 Rubin). Bi- Rite carried the advertised labels of Mill brook, Wonder and Laub and also carried Bond's while it was stil doing business in Cleveland (5509-10 Rubin). BiRite purchased all the white bread loaf izes and 8 pack buns from Continental (5509-IO Rubin) and paid list price less a 5% discount (5509-11 Rubin). Its purchases from Continental were between $3000 and $5000 per week (5510 Rubin). 357. In 1973 and 1974, BiRite purchased private label bread and Initial Decision 104 F. buns from Millbrook. It paid approximately 18\1 for the Ilb. loaf(5512 55I6 Rubin). Rubin did not believe he had ever received a better offer from Continental (5513). He testified that if he had been offered the same prices as afforded Pick N Pay in 1973, he would have tried to arrange for BiRite to "work with the program" (5514). BiRite s retail prices on bread were identical to the retail prices of Pick N Pay and Fisher-Fazio (5516 Rubin). Rubin thought BiRite was not able to buy on the same basis as some of his competitors (5519). (76) 358. In the 1970 to 1974 period, A&P's Cleveland division had 100 stores, approximately 30 of which were located in the greater Cleveland area (5845 Reed). Most of A&P's bread products were captive label; only 16% of the bread rack was open to the wholesale bakers (5851 Reed). Continental, Ward, Laub, Milbrook, Schwebel, Keystone and Nickel supplied various A&P stores in certain areas of the Cleveland division (5833 Reed). All of A&P' s bread suppliers, including Continental, offered a 5% discount (5835 Reed). The A&P stores in the Cleveland Division purchased a total of between $4000 and $5000 per week from Continental (5840 Reed).

359. Between 1968 and early 1970, A&P requested bids from wholesale bakers for private label (5842 Reed). Continental's offer was not as low as the prices which it charged Pick N Pay in 1973 (5843 Reed; seeCX 803A). A&P did not start a private later program at that time. 360. In 1974, Valu King Supermarkets was a voluntary group representing approximately 40 supermarkets located in a 60 mile radius of Youngstown (CX 992G; 5819 Tamarkin; 5256 Shreck). Continental sold a full line of Wonder bread and bun products to Valu King, the 18 oz. Wonder Loaf being one of the most popular white bread products (5820 Tamarkin). Continental provided full rack service to Valu King (5822 Tamarkin). Most Valu King stores paid list price less 7% discount of purchase of Wonder products (5823 Tamarkin). Other suppliers including Schwebel, Valu King s major supplier, granted 7%, except Keystone which offered 5% (5824 Tamarkin). Schwebel supplied Valu King with private label bread (5829-30 Tamarkin). Continental never offered Valu King private label bread at the prices afforded Pick N Pay in 1973 (5828 Tamarkin; CX 803). Mr. Tamarkin testified that bread was one of the most price sensitive items in the grocery industry (5829). Valu King did not compete with Pick N Pay (5828-29 Tamarkin).

361. In the early 1970's Kroger had 60 stores served by Continental' s Akron plant, some of which were located in the Cleveland market (See CX 914A).

362. Open Pantry had 42 stores served by . Continental' s Akron plant and was one of Continental's private label customers (See 900, 914B; Set 3 Adm. 56-58; 9366, 9503, 9507-08 Gase). 280 Initial Decision 363. It is found that Continental' s lower discriminatory prices such as those granted to Pick N Pay on its private label products may have had the effects on competition between competing grocers that is proscribed by the statute.

DISCUSSION Prior Proceedings Although this initial decision is based solely on the administrative record in Docket 9000, references are made by (77) complaint counsel and respondents to a series of administrative and court proceedings involving Continental and the Federal Trade Commission relating to antitrust matters. These proceedings may be summarized as follows: On May 11, 1962, the Commission issued an order pursuant to a consent agreement requiring Continental to divest itself of one bakery that it had acquired in 1958, and barring for 10 years Continental's further acquisition of an interest in any bread producer, unless permitted by the Commission. Continental Baking Co. 60 F. C. 1134 (1962).

On December 31 , 1963, the Commission dismissed. a complaint against Continental that was issued in October I959, charging violations of Sections 2(a) and 2(d) of the Clayton Act, as amended by the Robinson-Patman Act, on the grounds that Continental had proved that the challenged discriminatory prices had in each instance been granted in good faith to meet the equally low price of a competitor within the meaning of Section 2(b) of the statute. Continental Baking Co. 63 F. C. 2071, 2162 (1963).

On February 28, 1964, the Commission issued an order prohibiting Continental and others from engaging in price fixing activities, as found to have taken place in the Seattle marketing area. Bakers of Washington, Inc. 64 F. C. I079, I099 (1964), aff'd sub nom. , Safeway Stores, Inc. v. Federal Trade Commission, 366 F.2d 795 (9th Cir. 1966), cert. denied 86 U.S. 932 (1967).

On September 24 1973, the United States Court of Appeals for the Tenth Circuit affrmed a District Court ruling that Continental had violated the terms of the 1962 consent order by acquiring an interest in two bread producers. This suit was a civil penalty action fied by the Department of Justice on the Commission s behalf. United States v. ITT Continental Baking Co. 485 F. 2d 16 (IOth. Cir. 1973), reversed on the question of the amount of penalties, United Statesv. ITTContinental Baking Co., 420 U.S. 223 (1975). On December 12, 1973, the Commission issued an order to show cause why the I962 consent order barring Continental' s acquisition Initial Decision 104 F. of any interest in bread producers should not be extended for a five year period. On November 26 1974, the Commission rejected an Administrative Law Judge s recommendation that the I962 consent order be extended for a 5 year period and dismissed the order to show cause. ITT Continental Baking Co. 84 F. G 1349, 1400 (1974). On that same day, November 26 1974, the Commission issued the complaint in this proceeding, Docket 9000.

This Proceeding Interpreting the complaint in this case in light of certain statements contained in the Commission s decision (78) dismissing the order to show cause, respondents contend that complaint counsel have made no serious effort to support the broad charges in the complaint to the effect that exits of wholesale bakers from the baking industry have been the result ofITT forcing Continental to behave in a predatory manner, that respondents plan to achieve dominance in wholesale baking in all relevant geographic markets, and that respondents, since 1952, have operated with the intent to lessen, hinder or restrain competition and to attain monopolies in wholesale baking (RBr. 2).

In respondents' view, complaint counsel have abandoned the complaint' s broad charges of a nationwide program of predation and instead have "adopted the theories, contentions, and proof developed by lawyers in private litigation" involving six marketing areas. Respondents contend that "(t)rivial disputes over technical lawfulness of isolated transactions simply cannot sustain the basic charge made in the complaint that respondents have been predatorily attempting to take over the whole bread business throughout the nation" (RBr. 4-5). Respondents contend that, at this stage of the administrative proceeding in Docket 9000, it is apparent that the broad allegations ofthe complaint have no factual or legal support and that such allegations are based upon a fundamental misapprehension ofthe nature ofthe competition faced by wholesale bread companies such as Continental (RBr. at I).

I do not think that complaint counsel's approach to the issues raised by the complaint is as limited as respondents appear to argue. Although the primary emphasis is placed on respondents' practices and the effects of such practices upon competition in six marketing areas complaint counsel have presented some evidence about Continental's bakeries in other marketing areas. However, in its 1974 decision dismissing the order to show cause, the Commission stated that it would not infer a lessening of competition in local marketing areas from trends in concentration at a national level, especially where there is admittedly no national geographic market. The Commission 280 Initial Decision also pointed out that, although the exit of bakers from any market and the resulting increased concentration might be due to changes in technology and the industry members' adjustments thereto, it was the Commission s responsibility to "make very sure indeed that the larger members ofthe industry are not using any of the techniques prohibited by the antitrust laws to achieve a stil higher level of concentration . The Commission added that the "critical issue of competitive injury must be resolved on the basis of changes in the structure of actual rather than hypothetical markets. " (84 F. C. at I396, 1399; Commission s emphasis).

In my opinion, the instant proceeding is consistent with the Commission s concern about the changing structure of the bread industry in local marketing areas. (79) Price Discriminations Section 2(a) of the Clayton Act, as amended by the Robinson-Patman Act, provides in pertinent part:

It shall be unlawful for any person engaged in commerce, in the course of such commerce, either directly or indirectly, to discriminate in price between different purchasers of commodities of like grade and quality, where either or any of the purchases involved in such discrimination are in commerce, Ht and where the effect of such discrimination may be substantially to lessen competition or tend to create a monopoly in any line of commerce, or to injure, destroy or prevent competition with any person who either grants or knowingly receives the benefit of such discrimination or with customers of either of them.

Complaint counsel claim that, in the five separate marketing areas of Denver, Minneapolis-St. Paul, Southern California, Northern California, and Northwest Ohio, Continental has discriminated in price between purchasers of its white bread products and that such price discriminations have had the effects proscribed by the statue on competition between Continental and other wholesale bakers in the respective markets, and constitute primary line violations of the statute.

Respondents contend (RBr. 40-4) that complaint counsel have not demonstrated that the challenged pricing practices were predatory, have not demonstrated that the alleged injury was the result of price differences in commerce, have not demonstrated that the lower prices were subsidized by higher prices, or that the intensity of competition has actually been diminished in any of the markets involved (See Reply Br. 49-51). Respondents contend that the Robinson-Patman Act must be construed to protect competition, not competitors, and that the exit of a handful of marginal bakers, the diversion of business Initial Decision 104 F. from one seller to another or the lessening of profits is not suffcient to establish the injury element of a prima facie case (RBr. 43-4). Respondents also contend that the Commission s price discrimination charge is "completely out ofline" with more recent Commission enforcement of the Robinson-Patman Act (RBr. 40, n. Section 2(a) prohibits price discriminations when the effect "may be substantially to lessen competition *'* or to injure, destroy or prevent competition or tend to create to a monopoly in any line of commerce. It has long been considered that the statute seeks "to reach such discriminations 'in their incipiency' before the injury to competition is effected. Corn Products Refining Co. v. Federal Trade Commission 324 U.S. 726, 738 (1945). The prohibition of the statute goes to the price discrimination itself and the probable result of the price (80) discrimination if it is not stopped or prohibited. Each case, therefore must be adjudicated on its own merits, and, generalizations taken from previous cases are not necessarily controllng. Commerce In my opinion, complaint counsel have met the jurisdictional requirements ofthe statute. Admittedly, Continental is engaged in commerce. In each market area considered, Continental's local plant made sales of Wonder advertised label white bread at the higher price across state lines. Accordingly, the holding in Gulf Oil Corp. v. Copp Paving Co. 419 U.s. 186, 200 (1974), relied upon by respondents, is not applicable here. The Denver plant sold Wonder advertised label bread in Wyoming, the Minneapolis and Rochester, Minnesota, plants sold Wonder advertised label bread in Wisconsin, bread baked by the Beverly Hills and DiCarlo plants in Southern California was sold in Arizona, the San Francisco, Oakland and Sacramento, California, plants sold Wonder advertised label bread in Nevada, and the Akron and Youngstown, Ohio, plants sold Wonder advertised label bread in Pennsylvania.

Actually all of Continental's sales which are the subject of this proceeding are in commerce. Notwithstanding the local nature of each of Continental' s bakery operations, each grocery store purchaser is doing business with a corporation located in New York. Not only does Continental's headquarters engage in purchasing ingredients and supplies for the local bakeries, but its personnel engage in sales negotiations on behalf oflocal bakeries. It also finances the operations of each bakery, and creates and places the national advertising which makes possible the higher wholesale and retail prices for Wonder advertised label bread. See Holland Furnace Co. v. Federal Trade Commission 269 F. 2d 203 (7th Cir. 1959), cert. denied 361 U.S. 932 (1960).

280 Initial Decision Lower Price is Subsidized In this respect, the higher price for Wonder advertised label bread clearly subsidizes the lower prices for private label and secondary label whatever the pecularities of any individual market may be. In my opinion, the effects of a price discrimination on competition may be measured by the effects ofthe lower price on competition as well as the effects ofthe amount of the discrimination itself. This is especially true where the -lower prices are below the seller s fully allocated costs.

Like Grade and Quality It is undisputed that bread products are commodities oflike grade and quality, that Continental's advertised label white bread is the same bread product as its private label (81) or secondary label white bread. It is also undisputed that competing wholesale bakers' white bread is the same product as Continental's white bread, whether wrapped in advertised, private or secondary label. The only significant difference is the loaf size. In this respect, wholesale bakers almost invariably sell a line of products to their grocery customers which may include a mix of white bread sizes and shapes. Products Continental claims that wholesale bakers and grocers do not negotiate for private label bread products on the basis of one product, but rather on the basis of all private label items as well as advertised label products. In this respect, they challenge any Robinson-Patman analysis that concentrates on one item for purposes of determining the amount of the price discrimination or the effect of the price discrimination.

One of the peculiarities of the bread industry is that certain loaf sizes of white bread are the high volume products. Although the most popular loaf size varies from market to market, grocers concentrate in marketing the popular sized loaf. In these circumstances, wholesale bakers and grocers negotiate on the basis of the price of the largest sellng item. In addition, the wholesale baker considers the length of a production run that wil be involved in producing the most popular item. In my opinion, it is proper to determine the amount of discrimination and its effect on competition by looking to the prices of the one bread product that dominates the market. See Kroger Co. 2d 1372 (6th Cir. 197I), cert. v. Federal Trade Commission 438 F. denied 404 U.s. 87I.

Initial Decision 104 F. Predatory Pricing-Injury to Competition (Primary Line) I agree with the parties that the principal issue in determining whether Continental' s price discriminations are unlawful as charged is whether Continental's conduct was predatory. Predation is now considered to be a crucial element in primary line discrimination cases, although the Commission has indicated that market analysis may be suffcient in a particular case to raise the probability of injury to competition, proscribed by the statute. See Beatrice' Foods Co. 76 C. 719 , 799 (1969), aff'd sub nom. Kroger Co. v. Federal Trde Commission 438 F.2d 1372 (6th Cir. 1971), cert. denied, 871. In the circumstances of each of the five market areas considered for Robinson-Patman enforcement in this case, it is clear that Continental knew or should have known that the consequences of its price discriminations would be to eliminate one or more of its competitors or substantially to injure competition and would result in further concentration in the wholesale baking industry. In most instances Continental knew that the lower discriminatory prices were below its fully allocated costs and below its competitors' fully allocated (82) costs. Moreover, in certain situations Continental knew or should have known that the prices of certain products were below its average variable costs. See Utah Pie Co. v. Continental Baking Co. 386 U. 685, 698-700 (1967).

In Denver, through the Tender Crust program with Associated Grocers which granted a lower price on private label bread, Continental secured a substantial portion of the shelf space available to the wholesale bakers. The expected pricing reactions of competing wholesale bakers resulted in greater price discriminations and in Continental's sale of private label products well below its fully allocated costs. Continental knew or should have known that its lower prices were also below its wholesale baker competitors' fully allocated costs. At one time price discriminations were as great as 18% of the wholesale price of Continental's advertised Wonder loaf Continental knew or should have known that Old Homestead would not be able to successfully compete under these conditions. In fact, Old Homestead ceased doing business as a wholesale baker shortly after the 18% discriminatory price was established by Continental. In the circumstances, such price discriminations are predatory in nature and their effects may be to (and actually did) substantially lessen competition between wholesale bakers.

In the TCTA, Continental reduced its price of Wonder advertised label by 8 a loaf when it established a wholesale price of 20 while maintaining a 28 wholesale price in Rochester, Minnesota. The 20 hnlp lp nrl on Wonder advertised label was below Continental's 280 Initial Decision fully allocated costs. Continental's express purpose in initiating this price discrimination was to regain its volume of sales of advertised label bread with the expectation that retail prices of private label and captive label bread would thereafter be increased. Such a price increase would reduce the retail price spread between advertised label and private label bread and permit a price increase of advertised label bread without a loss in sales. However, the low retail prices of private label and secondary label bread, in turn, were due, in part, to Continental' s low discriminatory prices to its private label customers and purchasers of its secondary label bread.

Continental knew or should have known that its low discriminatory price on advertised label bread would seriously reduce the profitability of its wholesale competitors by lowering the wholesale prices of advertised bread below its wholesale competitors' fully allocated costs or by substantially reducing its competitors' volume of secondary and private label sales. Continental knew that most of Zinsmaster s sales were advertised label bread, whereas most of Pan 0 Gold and Creamy Crust' s sales were private label and secondary label bread. Continental also knew or should have known that the independent wholesale baker competitors would not be able to continue competing in the TCTA at such low wholesale price levels. Such price discriminations are predatory in nature and (83) their effects may be to substantially lessen competition between wholesale bakers. In the Los Angeles (Southern California) market, Continental discriminated in price between purchasers of its Wonder advertised label and purchasers of its private label and secondary label bread. During the latter part of the I960's and early 1970, the lower discriminatory prices in the Ventura area were below Continental's fully allocated costs and below Prosser s fully alloted costs. Continental knew that Prosser was a weak competitor and knew that Prosser could not survive ifit was required to sell at such low wholesale prices. In fact Prosser withdrew from the wholesale baking business in 1971. In late 197I and 1972, Continental sold private label bread in the Los Angeles area at low, discriminatory prices that were below Continental's fully allocated costs. With these low discriminatory prices Continental obtained many new private label customers and foreclosed much available shelf space from competing wholesale bakers including Gordon and American. It also offered the lower discriminatory prices to Gordon s only private label customer, requiring Gordon to lower its wholesale price below its fully allocated costs to retain that customer. Continental knew that Gordon and American were weak competitors and would not be able to compete at wholesale price levels below fully allocated costs. In fact, in 1974, Gordon withdrew from the wholesale baking industry, leasing its bakery to the As- Initial Decision 104 F. sociated Grocers Cooperative, which thereafter produced its own private label bread for its member grocers. Also in 1974, American discontinued sellng to grocery stores and concentrated its efforts on sales to restaurant customers.

Continental's price discriminations are predatory in nature and their effects may be to (and in fact did) substantially lessen competition between wholesale bakers.

In Northern California in 1972 Continental offered a 17. wholesale price on private label bread which was substantially lower than the wholesale price on its Wonder advertised label. This lower price was below Continental' s fully allocated costs. Continental knew that Inglis, which had fied a bankruptcy petition in 1969 and which it considered to be a weak competitor, would not be able to compete effectively at such low wholesale price levels. Such a price discrimination is predatory in nature and its effect may be to (and in fact did) substantially lessen competition between wholesale bakers. In the Cleveland area, Continental discriminated in price between purchasers of its Wonder advertised label and Pick N Pay on its purchases of private label bread. During the last half of 1973 and during most ofl974, the wholesale prices on Pick N Pay private label products were below Continental's fully allocated costs and close to if not below, Continental's average variable costs. Continental knew or should have known that its weaker wholesale baker competitors in (84) the Cleveland marketing area would not be able to compete in a market oflow retail bread prices which were a result of Continental's low discriminatory prices to Pick N Pay. Such price discrimination is predatory in nature and its effect may be to (and actually did) substantially injure competition between wholesale bakers. Good Faith Defense The Robinson-Patman Act permits certain price discriminations notwithstanding their proscribed effect. Section 2(b) provides: That nothing herein contained shall prevent a seller rebutting the prima facie case thus made by showing that his lower price H* to any purchaser or purr:hasers was made in good faith to meet an equally low price of a competitor. The "good faith" requirement of Section 2(b) has been interpreted as meaning that the grantor of the lower discriminatory price must show the existence of facts which would lead a reasonable and prudent person to believe that the granting of a lower price would in Hict meet the equally low price of a competitor. Federal Trade Commission v. A. E. Staley Manufacturing Co., 324 UB. 746, 759-60 (1945). Respondents argue that in every instance where they granted a INTEltATIUNAL TELE.t.tUNJ: & TJ:LJ:lil'rn \.UIU. .rl iU. 0j0 280 Initial Decision lower discriminatory price in the five marketing areas under consideration for Robinson-Patman enforcement, Continental believed that it was meeting the equally low price of a competitor. However, on the basis of the record in this proceeding, Continental has not made out its meeting competition defense.

In Denver, the agreement to sell Tender Crust at a wholesale price at least 1 below the wholesale price of Wonder advertised label bread was to be afforded without regard to a specific competitive offer. Subsequent pricing under modifications of the original Tender Crust agreement always established a wholesale price on Tender Crust below the wholesale price of competing wholesale bakers. Such conduct does not meet the requirements of Section 2(b). In the TCT A, Continental's lower discriminatory price on its Wonder advertised label bread was not intended to meet any ofter of a competing wholesaler to any of Continental's grocer customers. Moreover, the discriminatory prices given to its private label customers on private label bread were to meet Continental's customers ' competition with captive bakers. These situations do not meet the requirements of Section 2(b).

In Southern California, the record is not clear which wholesale baker initiated the lower prices on advertised label (85) bread in the Ventura area in the late 1960's or whether such prices were discriminatory. It appears that Continental, Interstate and American matched each other s lower prices. However, it appears that Continental granted lower discriminatory prices on its secondary label in the Ventura area without regard to its wholesale baker competitors prices. In addition, such prices were below Continental' s fully allocated costs as well as below Prosser s wholesale price and fully allocated costs. The record is clear, however, that Continental's lower discriminatory price to Hughes on the largest volume loaf size of private label bread was below Interstate s offer. The requirements of Section 2(b) were not met.

In Northern California, Continental offered a 17. wholesale price on private label bread to all chain stores throughout the marketing area. The only other 17. price offered on a comparable product were on sales of private label bread by other wholesaler bakers to grocers in a few local areas. The 17. price was below Continental's fully allocated cost. Accordingly, Continental was knowingly undercutting the price of its competitors to many individual grocery stores and did not meet the requirements of Section 2(b).

Finally, in Cleveland, Continental's contract with Pick N Pay in 1973, and as renegotiated in 1974, set wholesale prices of private lab products that were below Continental's fully allocated costs and which were below any competitive offer. Continental knew or should Initial Decision 104 F. have known that it was not meeting any competitors' price ofter and accordingly, the requirements of Section 2(b) were not met. There have been many other interpretations of the "good faith" element of Section 2(b), other than the question of whether the seller acted reasonably in determining the actual amount ofthe competitive offer he was allegedly meeting. For example, it has been held that 2(b) applies only to individual price concessions and not to systematic or generalized price discriminations. Federal Trade Commission v. Cement Institute 333 U. S. 683 91948); Federal Trade Commission v. E. Staley Manufacturing Co. 324 U. S. 746 (1945); Federal Trade Commission v. Standard Oil Co. 355 U.S. 396 (1958). There is a question as to whether 2(b) is applicable where the discriminations are used aggressively to obtain new business or more business, instead of protecting one s business with a customer. Standard Oil Co. v. Federal Trade Commission, 340 U.s. 231, 249-50 (1951); Great A&P Co. Federal Trade Commission 440 U.S. 69 (1979); Standard Motor Products, Inc. v. Federal Trade Commission, 265 F.2d 674 (2d Cir. 1959). And it should seriously be considered whether discriminatory sales below cost to obtain new business could ever be considered a good faith meeting of the equally low price of a competitor. See Sunshine Biscuits 59 F. C. 674, 678 at 679-81 (1961), rev 306 F.2d 48 (7th Cir. 1962). In my opinion, Continental also has failed to meet the requirements of Section (86) 2(b) for one or more of these other reasons in each of the five marketing areas in which it granted the challenged price discriminations.

Respondents contend that the good faith meeting of competition defense has already been adjudicated in their favor, citing the Commission dismissal of the Section 2(a) price discrimination matter in 1963. Continental Baking Co. 63 F. C. 2071 (RBr. 47). It appears that the evidence in that proceeding concerned specific discounts granted from Continental's list prices to 20 retail grocery customers. All but two of those customers were shown to have been, at the time the discounts were granted and thereafter throughout the period covered by that case, receiving at least as high discounts from one or more of Continental's competitors. The other two grocery customers had been offered, and had available to them, discounts from Continental's competitors at least as high as those granted by Continental. The circumstances in the instant case under which the lower discriminatory prices were granted are markedly different than the circumstances involved in the prior proceeding. Not only has Continental not met the basic requirements of the statute in the instant case, but Continental has also failed to demonstrate compliance with its own policies on meeting competition, because competitive offers were not docu- 280 Initial Decision mented with respect to the private label arrangements in most instances.

Cost Justification Defense Respondents contend that the W onder-Tender Crust price difference in the Denver marketing area was cost justified (RPF 168-169). They rely upon a study prepared by Lloyd Swan for use in the Old Homestead trial in 1971 (RPF 169; CXs 1722 Swan; 1728). The comparison made by respondents for cost justification purposes is between the profits Continental ostensibly made on Wonder advertised label bread and Tender Crust private label bread. There is no dispute that Mr. Swan calculated hypothetical costs applying percentages to the list prices in effect during relevant periods. It is doubtful whether such allocations of costs are adequate for purposes of cost comparisons. Continental did not keep records in those days from which it could calculate the actual costs of producing and sellng any single variety of bread product.

Moreover, the price differences shown on the Forms 452B were usually one cent. The actual price differences shown on this record during the last quarter of 1966 through December 1967 ranged between 2.2 and 3. . In many instances, the difference in profit between Wonder advertised label bread and Tender Crust private label bread were not as great as the price difference between them. In my opinion, respondents have not carried their burden to demonstrate a cost justification defense. They do not appear to assert such a defense for any other price discrimination challenged in this proceeding. (87) Competition One of respondents' defenses to the " primary line" Robinson-Patman allegations is that the competing wholesale bakers that withdrew from the baking business during or shortly after Continental engaged in the challenged price discriminations were marginal firms who were not advertising their products, who were mismanaged, inefficient, under capitalized, and doomed to failure irrespective of what Continental did (Seeger. 43). In effect, Continental is contending that its pricing actions, considered necessary for its own survival, are not ilegal because they happened to hasten the departure of competitors. Realistically, however, the survival in a local market of a company operating a large number of bakeries in many markets at the expense ofthe early withdrawal of a local bakery is exactly the type of situation the Robinson-Patman Act was intended to prevent. The resulting concentration of market power is the indicia ofa lessening ofcompetition and the type of competitive injury with which the statute is concerned.

Initial Decision 104 F. Respondents also contend that the degree of effective competition in each of the markets has not diminished over the years and that accordingly, enforcement of the Robinson-Patman Act is not in the public interest. But, as stated before, the Robinson-Patman Act was to prevent those price discriminations in their incipiency that may be to substantially lessen competition. It is found that respondents' price discriminations are such that, if permitted to continue, or if resumed or pursued in other markets, may have the effect oflessening competition between wholesale bakers. The record in this case is clear that effective competition between wholesalers in the five markets substantially was diminished over time.

Enforcement Policy I do not agree that the Commission s lack of Robinson-Patman enforcement against sellers since the Kroger-Beatrice decision in 1969 reflects a Commission judgment as to the lack of public interest in such enforcement. The complaint issued by the Commission in Docket 9000 contained a Robinson-Patman price discrimination charge. It was complaint counsel's duty to pursue that matter should the facts and circumstances warrant, and it is the duty of the Administrative Law Judge to adjudicate the issues presented. It is a grave mistake for any businessman to believe that the Commission s temporary inaction grants him a license to disregard the prohibitions of statutes such as the Robinson-Patman Act.

Price Discriminations-Injury to Competition (Secondary Line) It is well settled that substantial price discriminations on products sold to competing grocery store customers may be to substantially injure competition between them. See United Biscuit Co. of America v. Federal Trade (88) Commission 350 F. 2d 615 (7th Cir. I965), cert. denied 383 U. S. 926; Federal Trade Commission v. Morton Salt Co. 334 U. S. 37 (1948); Foremost Dairies Inc. v. Federal Trade Commission 348 F.2d 674 (5th Cir. 1965), cert. denied 382 U.S. 959. It has also been held that where a price differential between an advertised brand and private label brand reflects no more than the consumers' preference for the advertised brand, such price discrimination does not create the injury prescribed by the statute. See Bordon Company Federal Trade Commission 381 F. 2d 175 (5th Cir. I967). In the instant case, the price discriminations arising from the private label prices afforded Pick N Pay in Cleveland and the chain stores in the Reno area appear to be more than a difference which merely reflected consumer preference between advertised label and private label bread. It is found that Continental's pricing practices in Cleveland and Reno constituted secondary line injury violations of the statute. 280 Initial Decision For reasons already stated, respondents' lower discriminatory prices in the Cleveland market area and in the Reno portion ofthe Northern California area were not granted in good faith to meet the equally low prices of a competitor within the meaning of Section 2(b) of the Clayton Act, as amended.

Attempt to Monopolize Section 5 of the Federal Trade Commission Act provides in pertinent part:

Sec. 5(a)(). Unfair methods of competition in or affecting commerce, and unfair or deceptive acts or practices in or affecting commerce, are declared unlawful. Sec. 5(a)(2). The Commission is empowered and directed to prevent persons, partnerships, or corporations, from using unfair methods of competition in or affecting commerce and unfair or deceptive acts or practices in or affecting commerce. It is well settled that Section 5 covers conduct that either violates the prohibitions of the Clayton Act and the Sherman Act or conduct that could lead to unreasonable restraints on competition if not prohibited. See Federal Trade Commission v. Brown Shoe 384 U.S. 316, 32I (1966); Federal Trade Commission v. Cement Institute 333 U.S. 683 (1948).

An illegal attempt to monopolize, constituting a violation of Section 2 of the Sherman Act, involves a "specific intent" to control prices or destroy competition in a relevant market, predatory or anticompetitive conduct directed to accomplish those ends, and a dangerous probability of success. Purex Corp. v. Proctor Gamble Co. 596 F.2d 881 890 (9th Cir. 1979); Janich Bros. , Inc. v. American Distilling Co., 570 2d 848 853 (9th Cir. 1977); Golden Grain Macaroni Co. 78 F. C. 63 164 (1971), enforced in part 472 F. 2d 882 (9th Cir. 1972), cert. denied 412 U.S. 918 (1973). (89) These criteria are not mutually exclusive but rather are interrelated. Evidence of conduct may shed light on intent and the probability of success. Evidence of a company s purpose may reveal whether the conduct is reasonably related to legitimate business needs. See Janich Bros. 570 F. 2d at 853; Transamerican Computer Co. v. IBM Corp., 481 F.supp. 965, 989 (N.D. Cal. 1979).

With respect to the "dangerous probability" issue, there is conflict in the law as to what degree of market power, or proximity to monopoly status need be shown before a finding of liability can be made. Compare Greyhound Computer Corp., Inc. v. IBM Corp. 559 F.2d 488 496 504 (9th Cir. 1977) with United Statesv. Empire Gas Co. 537 F. 296, 305 (8th Cir. 1976), cert. denied 429 U.S. II12 (1977). It is generally accepted that monopoly power exists when an industry member has the power to raise prices or exclude competition when Initial Decision 104 F. it so desires, and that such monopoly power is unlawful ifit is willfully maintained or acquired as distinguished from arising from growth or development as a consequence of superior product, business acumen or historical accident. Purex Corp. 596 F.2d at 890; Golden Grain Macaroni Co. 78 F. C. at 157.

Contentions of the Parties Complaint counsel contend that Continental's predatory pricing practices, including price discriminations and sales below cost, along with other instances of predatory conduct, demonstrate that Continental intends to destroy competition between wholesale bakers and intends to gain monopoly positions in local wholesale white bread markets. Complaint counsel argue that, absent some remedial action Continental may be successful in gaining such monopoly positions (CC Br. 16-33).

Respondents contend that complaint counsel' s attempt to establish a relevant market limited to white bread produced by wholesale bakers is not realistic (RBr. 22-27). They point out that captive bakers which produce an identical white bread product are a dominant competitive factor in the bread industry, especially in the six markets which are the subject of this case (RBr. 23-24). Respondents contend that the retail prices established by the captive bakers have a controlling effect on the wholesale bakers' wholesale price of advertised label and private label bread in that the captive label bread and wholesaler baked bread compete directly for the consumer dollar (Ld.; See Reply Er. 60-1). Respondents also contend that white bread by itself is not a proper product market because other bread products are acceptable substitutes, as demonstrated by the fact that during the 1970' s there has been a substantial decrease in the consumption of white bread whereas the consumption of variety breads has increased during the same period (See REr. 26-27). (90) Respondents argue that Continental's pricing practices were not predatory, but merely responded to the competition of the captive bakers. They contend that there is "chronic" excess capacity in the bread industry created by the growth ofthe captive bakers, as well as the decreased consumption of bread, and that the wholesale bakers competition for the shrinking available shelf space has created chaotic conditions in certain markets in the bread industry (REr. 5-6; see R Reply Er. 6I-62). Respondents also contend that in such circumstances, to sell below fully allocated costs, but over "marginal" or variable" costs, is a reasonable business decision, in that such sales contribute to a positive cash flow and to the overall profitability ofthe business (See REr. 36-37).

F-in!' llv RI:"::nnnrJpnt, d:;im t.hat entry into the bread baking busi- 280 Initial Decision ness is easy and, accordingly, there is no way that Continental could achieve a monopoly position in any bread market (RBr. 31, 33-35; see R Reply Br. 66-7). They argue that the wholesalers must respond to the grocers' demand for low cost private label bread or face the possibility that those grocers wil become captive bakers (RBr. 10, 31). Relevant Product Market In my opinion, wholesale baked white bread has such distinctive product and market characteristics that it may be considered a relevant market or sub market upon which to determine the effects of Continental's challenged practices in the context of an alleged attempt to monopolize case. Quite clearly, concentration of the wholesale white bread business in the hands of one wholesale baker in a marketing area would constitute a monopoly. Any retail grocer which did not have its own bakery would have to buy white pan bread from one source. Moreover, retail grocers could not remain competitive by selling only bread products other than white pan bread. Captive bakers do not compete in selling to grocery stores; they limit distribution through their own retail outlets. Wholesale bakers on the other hand, compete with other wholesale bakers in the sale of bread products to grocers.

White pan bread is the most important bread product to the large wholesale baker. The quantities involved permit long production runs which maximize effciencies and increase profiability. The record shows that many bakeries concentrate their production efforts in white pan bread and that the volumes of white bread produced and sold do, in fact, reflect the profitability and competitive strength of wholesale bakers. For example, Continental's Akron and Minneapolis bakeries appear to have been exclusive white bread bakeries. These white bread products are also the most important bread products to the grocer. The largest and best shelf space (91) is allocated to these products. Most private label arrangements concentrate on white bread and negotiations are usually centered on determining the wholesale price of the most popular loaf size of white bread in a particular marketing area.

The fact that wholesale baked white bread is sold at retail by grocers in competition with captive label bread, either from the same bread table, or from the bread tables of competing grocers, does not place captive bakers in competition with wholesale bakers at the wholesale level of competition. Delineation of wholesale baked white bread as a relevant market conforms to the requirement that a relevant market be an area of effective competition and is consistent with the realities of competitive practice in the market place. See L. Balfour Co. v. Federal Trade Commission 442 F. 2d 1 (7th Cir. 1971). Initial Decision 104 F. No Chronic Excess Capacity The record does not show that there was "chronic" excess capacity in the bread baking industry, during the relevant periods, at least not in the six market areas that are examined in this record. The practical production capacity of a bakery consists of two shifts and some overtime or approximately 80% ofthe maximum production possible. Capacity is not, as Continental suggests, measured by the maximum output possible. Bakers need reserve capacity to handle unusual demands that occur from time to time and from season to season. The excess capacity that is indicated in this record existed where Continental built a new plant (Denver), was operating a plant at a little over one shift and sustaining a loss over a number of years (Akron), or was losing a large private label customer which built a captive bakery (Los Angeles). Accordingly, the competitive conditions in those markets seemed to have been generated by Continental' attempt to "fill" or "over fill" the optimum effcent capacity of its plants. The question here is whether Continental's aggressive price discriminations involving below cost pricing to take business from its competitors in order to fill its plants is really a response to excess capacity.

Barriers to En try The record does not demonstrate that it is easy to enter the wholesale bread baking industry. Although there are no technical barriers there are certain market barriers, the most important of which is the creation of a consumer franchise. For unless a wholesale baker has an acceptable advertised label, it wil not be able to obtain the necessary bread shelf position and space to fil a modern bakery. The record shows that practically no wholesale baker has entered a market in the 1970' s except by acquisition of an existing baker. The only new entrants have been several large chain stores that have built their own captive bakeries. However, the record also shows that a number of large chain stores closed (92j their captive bakeries during the 1970' and that the trend toward captive baking has abated. Another barrier to entry into the wholesale baking business may be the propensity of the large multiplant companies to engage in below cost pricing in particular markets. Such parallel behavior was practiced in the Denver, Minneapolis, Los Angeles and San Francisco markets. The record shows that a local bakery cannot survive in such a competitive atmosphere. The margin of profit and the return on investment are not great enough to attract capital expenditures in such an unstable atmosphere as exists in the wholesale baking business.

280 Initial Decision Pricing Below Cost In defense of its below cost pricing, Continental relies on a commentary by Areeda and Turner and certain court decisions that have adopted the view, that, if such pricing is above marginal or variable costs, pricing below fully allocated costs should not be presumed to be predatory (RBr. 35-36). In the context of this case, this proposition is not controllng. Such a view overlooks the fact that a large multiplant firm which prices products below fully allocated costs for an extended period of time in a local market either knows or should know that single plant firms in that local market may be irreparably injured or become so Hconditioned" that they become !!passive" competitors. This conditioning would be especially true in the wholesale bread business because no wholesaler can charge a higher wholesale price for white bread than the lowest wholesale price prevailing in the market for a comparable product without losing a substantial share of its white bread sales.

Continental contends that its pricing practices were reasonable responses to competitive conditions engendered by low retail prices established by captive bakers. It contends that the alternative to below cost pricing is withdrawal from the market (See R Reply Br. at 1) Continental's argument might be rephrased to mean that Continental should be free to engage in any conduct that wil ensure to it a share of the market suffcient to permit Continental to operate at optimum effciency. Approval-of such a philosophy would negate the principal purpose of the antitrust laws, which is to prevent the restructuring of an industry to accommodate the most powerful member.

Continental also contends that the question of whether it sold its bread products below cost (whether below average variable cost or below fully allocated costs) should be determined on the basis of its costs of selling all products in its line to a particular grocery customer. In this respect, Continental points out that private label bread is sold with the expectation that the wholesale baker, by obtaining the next best position on the bread rack, wil increase its sales of (93) advertised label bread. Continental suggests that it would never enter into a private label agreement unless it expected to make a profit on the total amount of business done with any customer. The record shows that Continental does not control the "mix" of products purchased by the grocer. Many factors, including the retail price spread between private label and advertised label products determine the relative quantities of each label sold. The grocer appears to negotiate private label prices separately and usually makes his decision to enter into a private label arrangement with the wholesale baker on the basis of the wholesale price of the most popular size Initial Decision 104 F. loaf or loaves of white bread. Although Continental' s contention as to measuring costs on the basis of an entire line might have some merit in a different industry setting, the peculiarities ofthe bread industry make such treatment inappropriate. The record shows that Continental continued selling private label to certain chain stores, although the amount of advertised label also sold could not possibly make the total sale profitable on a fully allocated cost basis. Unreasonable Pricing Practices In its recent DuPontdecision the Commission set forth criteria by which to measure whether a firm s competitive behavior is reasonable, notwithstanding the anticompetitive effects that might result from such behavior (Current Binder) 3 CCH Trade Reg. Rep. n 21770 (1980) (96 F. C. 653). Aside from the fact that Continental's pricing practices were ilegal under Section 2(a) of the Robinson-Patman Act its practices do not appear reasonable under the stated criteria. For example, below cost pricing cannot be considered to be an ordinary marketing practice and it is certainly not profitably or economically rational in itself. Such practices do not result in improved product performance and it is doubtful whether such below cost pricing would be effective in gaining and in holding market share for a firm which did not have substantial economic resources. In the market setting where the barriers of entry into wholesale baking are high, the anticompetitiveness of the challenged practices are exclusionary and disciplinary. In this respect, the effects of the practices are not reversible. In my opinion, Continental's behavior was not !treasonable within the framework of a "rule of reason" test. Price Leadership Continental points out that it has no power to raise its prices without regard to its competitors' pricing, and that the absence of such power demonstrates that it has no monopoly power, and, in the circumstances of the bread industry, cannot obtain such power. Although it cannot raise prices unilaterally, Continental is considered to be the price leader in many (94) marketing areas, in that historically, the other wholesalers will usually follow Continental' lead in raising prices. It is not clear whether such leadership is the result of prolonged periods of low wholesale bread pricing, and the necessity to regain profitability, or an automatic response to Continental' s moves in order to forestall any disciplinary action. In my opinion, such price leadership demonstrates Continental's pricing power. In my opinion, Continental's predatory conduct is clearly !!intended" to maintain or acquire such pricing power. The price leadership theme appears often in Continental' s internal docu- 280 Initial Decision ments and appears to be the principal motivating force behind Continental' s conduct challenged in this proceeding. Alleged Threats Complaint counsel contend that respondents' predatory intent is demonstrated by certain other actions taken by Continental' s personnel against its wholesale baker competitors. For example, in the TCTA (Minneapolis), Continental's offcials allegedly made threatening statements to Zinsmaster and Creamy Crust (CCPF 9-82; CC Mem. 97). In the Los Angeles marketing area, certain threats were allegedly made to the Gordons (CCPF 11-58; CC Mem. 122). These accounts are not supported by suffcient non-hearsay evidence to be considered as part of the pricing transactions that are challenged as predatory in nature. However, such events do indicate that some of Continental's wholesale baker competitors were apprehensive about Continental' s possible actions in their marketing areas. Market Strategy One of complaint counsel's main contentions is that respondents have adopted a strategy of delaying wholesale price increases in order to hasten the exit of their weaker wholesale baker competitors. Complaint counsel cite this "hold-the-line-on-prices" strategy as the indicia of respondents' predatory intent.

Respondents claim that they never adopted such a strategy. They contend that the McKinsey report merely suggested a further study into the feasibility of such a strategy and they deny that Continental's offcials ever discussed such a strategy.

The record shows that the effects of such a strategy were discussed among ITT, Continental and McKinsey personnel during the meetings that were part of the study conducted by McKinsey during 1971 (SeeCX 28S, T). Denial of this by respondents belies the management sophistication which is respondents' principal asset. Moreover, there can be no doubt that such pricing would be injurious to Continental's wholesale baker competitors. The record also shows that Continental did maintain low discriminatory, below cost prices on certain (95) products, usually on private label products, over relatively long periods of time in certain marketing areas.

It appears, however, that respondents adopted market strategies responsive to each individual market situation. Although their pricing moves were designed to take advantage of the weaknesses oftheir wholesale baker competitors, it cannot be concluded on this record that such decisions were part of some over-all "hold-the-line-onprices" strategy. Nevertheless, the fact that respondents were aware Initial Decision 104 F. of the result of their below cost pricing supplies the "predatory intent" necessary in an attempt to monopolize case. Market Shares Finally, Continental argues that in, the circumstances ofthe bread industry, there is no possibility that it could be successful in obtaining a monopoly in any local market. They point out that the evidence pertaining to the six geographic areas which were the subject of detailed analysis in this case demonstrate that it has not been able to obtain a leading share of any market, much less a monopolist's share in any of those markets.

In my opinion, Continental' s actual success or failure of obtaining very high shares of the wholesale baking business in any of these six markets is not important. There is no doubt that Continental has improved substantially its position in the wholesale white bread market in several areas, namely, Denver, Los Angeles, Northern California, and Cleveland. In the latter three, Continental, for all intents and purposes, shares the entire wholesale produced white bread market with one of its multiplant wholesale baker competitors. In Los Angeles and Cleveland, it shares the wholesale baked bread market with Interstate and in Northern California it shares the market with Campbell-Taggart.

High concentration of market shares ofthe wholesale baked white bread market in a local area does not appear to be procompetitive in the bread industry. Wholesale bakers with established consumer franchises are able to sell advertised white bread at higher prices than other identical products. Maintaining such consumer franchise through advertising creates substantial barriers to entry into the wholesale white bread market because ofthe limited shelf space available. It is entirely possible that several wholesale bakers with a highly concentrated share of the white bread market can, through observance of price leadership and through control of the source of private label bread available to grocers, obtain higher prices for their white bread products, increase their sales of more profitable advertised label and variety breads, and thus make higher profits. It appears that Continental's strategy in certain markets in which they share the market about" equally with another multistate wholesale baker was to emphasize sales of advertised label bread following the events challenged by (96) complaint counsel and the withdrawal of some of its smaller wholesale baker competitors from the market. The Northern and Southern California and Denver markets reflect Continental' s reemphasis on the higher priced advertised label bread. Moreover, Continental has been a defendant in private treble damage actions in each ofthese six markets and it may well be that as a INTRNATIONAL TELEPHONE & TELEGRAH CORP., ET AL. o10v 280 Initial Decision result respondents' conduct since 1974 has been less aggressive in these areas. It should also be pointed out that respondents' production capacity available to service a particular market is limited. Although a multiplant firm such as Continental is more flexible because it can make certain interplant shipments, it does not appear that Continental has the production capacity to take more than a 50% share of any ofthe six market areas. In fact, with the exception of Pittsburgh and Washington, D. , Continental' s market share of any major wholesale white bread market does not appear to be more than 50% (See 4(131); see also CCPF 6-232). In my opinion, Continental engaged in practices which contravened the general prohibition against attempting to monopolize. Such conduct violates Section 5 of the Federal Trade Commission Act. If respondents are not prohibited from engaging in such predatory practices as sellng below cost and discriminatory pricing and if the market power or potential market power they have obtained in certain markets is not controlled or reduced, the wholesale baking industry may continue to become more concentrated and less competitive. ITT's Responsibility for Continental's Behavior Respondent ITT is fully responsible for Continental's conduct which is the subject of this proceeding and which took place after the merger ofITT and Continental in 1968. It is not disputed that ITT had knowledge of, participated in and ratified Continental' s management decisions and policies designed to assure that Continental would contribute its share to ITT's profits (See 626, 634-35 Woodward; 885-896 Stolle; CXs 113A, 25A-C; seeCX 20). ITT benefited directly from Continental's business practices, and it should be named in any order issued in this proceeding. See PF Collier Son Corp. v. Federal Trade Commission 427 F. 2d 261 (6th Cir. 1970), cert. denied 400 U.S. 926. CONCLUSIONS OF LAW 1. The Federal Trade Commission has jurisdiction over the subject matter of each of the counts of the complaint in this proceeding. 2. The Federal Trade Commission has jurisdiction of respondents. (97) 3. Respondent International Telephone & Telegraph Corp. is engaged in interstate commerce as "commerce" is defined in the Federal Trade Commission Act.

4. Respondent ITT Continental Baking Company, Inc., is engaged in the sale and shipment of bread in interstate commerce as "commerce" is defined in the Federal Trade Commission Act. 5. The baking, sale and distribution of white pan bread and bread Initial Decision 104 F. type rolls by wholesale bakers is an appropriate relevant product market in which to evaluate the conduct of respondent ITT Continental Baking Company, Inc.

6. The respondents' acts and practices as hereinabove found, hindered, lessened, eliminated, injured, destroyed or foreclosed actual and potential competition in wholesale baking and increased the probability that respondents will attain monopoly power in the whole. sale baking industry in local geographic markets. 7. The respondents' acts and practices, as hereinabove found, are in violation of Section 5 ofthe Federal Trade Commission Act, as amend- , and of Section 2(a) of the Clayton Act, as amended. 8. The proceeding is in the public interest. REMEDY In the order fied with their proposed findings, complaint counsel seek provisions that: (1) would require the divestiture by respondents offive bakeries (Beverly Hils, Sacramento, Minneapolis, Akron, and Salt Lake City, Utah); (2) would require compulsory trademark licensing of respondents' trade names to these new wholesale bakers; (3) would prevent respondents from impairing the viability ofthe bakeries to be divested; (4) would reinstate the moratorium on acquisitions which expired in 1973; (5) would prohibit sales of bread products below cost; (6) would prohibit price discriminations in the sale of bread products; and (7) would require respondents to maintain certain records pertaining to its costs (SeeCC Mem. 174-181; CCPF 5--0-- 66).

Respondents contend that there is no record support for the proposed divestitures because the record in Docket 9000 does not detail the present competitive posture or financial condition of the five bakeries selected by complaint counsel for divestiture (R Reply Br. 66-72). Respondents argue that the proposed compulsory licensing makes no sense. They point out that Continental would be eliminated from any market in which it could not use its brand name (advertised label) because two bakers could not use the same brand name in any one market. Respondents also point out that the provision of the order that prohibits Continental from hiring personnel employed by the (98) bakeries selected for divestiture is unfair to the employees and would, in any event, eliminate Continental from the markets involved (R Reply Br. 70-71). Respondents also argue that the Commission does not have the authority to order divestiture in a case brought pursuant to Section 5 of the Federal Trade Commission Act which does not involve ilegal acquisitions.

Respondents challenge complaint counsel' s proposed ban on future INTERNATIONAL TELEPHONE & TELEGRAH CORP., ET AL. 387 280 Initial Decision acquisitions, noting that the record does not contain evidence of any ilegal acquisition since the Commission, in 1973, refused to extend a 1962 order which prohibited such acquisitions. Respondents argue that the provision of complaint counsel' s order that prohibits Continental from selling any of its bread products below fully allocated costs is not only unprecedented but would put Continental out of business. For example, if a competitor established prices above its marginal costs but below Continental's fully allocated costs, the competitor would be able to take away Continental's business.

Respondents argue that the prohibition on price discrimination would eliminate their private and secondary label sales because such products would have to be priced at the same level as its advertised label. They also point out that the provision, as written, would not permit the use ofthe "meeting competition" or other statutory exceptions.

It is well settled that the Commission has wide discretion in its choice of a remedy adequate to cope with the unlawful practices in which respondents were found to be engaged. Federal Trade Commission v. Ruberoid Co. 343 U.s. 470 (1952). That discretion will not be overturned by the reviewing courts unless the remedy has no reasonable relation to such practices. Ruberoid Co. 343 U. S. at 473; See Jacob Siegel Co. v. Federal Trade Commission 327 U.S. 608, 611-613 (1946).

In my opinion, the Commission has the authority to order divestiture to break up a monopoly or to restore competition that might have been lessened as a result of illegal conduct, even where no acquisition has been challenged. See Federal Trade Commission v. Dean Foods Co. 384 U.S. 597 , 606 n.4 (1966); L.G. Balfour Co. v. Federal Trade Commission, 422 F.2d 1 , 23 (7th Cir. 1971); International Boxing Club v. United States 358 U.S. 242 (1959). Otherwise, the Commission could not fulfi the Congressional mandate "to prevent. . . unfair methods of competition" in cases involving illegal monopolization or attempts to monopolize.

Notwithstanding the general rules concerning the Commission discretion and authority in issuing orders, the provisions of any order must be realistic and the consequences of a respondent' s compliance therewith must be compatible with the result the order is intended to accomplish. See L. G. (99) Balfour 442 F. 2d at 23; compare Colgate Palmolive Co. v. Federal Trade Commission, 310 F.2d 89 (1st Cir. 1962) with Federal Trade Commission v. Colgate Palmolive Co. 380 U.S. 374 (1965), reversing Colgate Palmolive Co. v. Federal Trade Commission 326 F.2d 517 (1st Cir. 1963).

DoonMopnts' objections to the divestiture provisions of the order Initial Decision 104 F. raise serious questions as to whether the order wil accomplish complaint counsel's purpose. As I understand complaint counsel's position, they intend to establish five independent single plant baking companies in order to restore effective competition in four local markets by reducing the level of concentration of market shares among wholesale bread bakers. In this respect, complaint counsel appear to envision that Continental wil stil be one of the wholesale baker competitors in each market; they point out that Continental wil have at least one bakery in each market viz. DiCarlo in Southern California; San Francisco and Oakland in Northern California; Rochester in Southern Minnesota; Columbus and Dayton in Ohio; and Ogden in Utah.

I agree with complaint counsel that the viabilty ofthese new bakeries would depend on large part on each one having a strong consumer franchise. Compulsory licensing of Continental's established advertised brand name, Wonder, would accomplish this. But, on the basis of this record, respondents are correct in pointing out that grocers would not purchase the same brand product (Wonder) from two different wholesale bakers. And if Continental was successful in developing a consumer franchise for another brand, for example Home Pride, the newly created bakers could, under the provisions ofthe order, require Continental to grant them a license to use that brand name also. It should be pointed out that the major wholesale baker competitors in four of the markets in which complaint counsel's newly created wholesale bakers will operate are multiplant corporations, Interstate in Los Angeles and Cleveland, American in Minneapolis, and Campbell- Taggart in Sacramento. The record indicates that Interstate and Campbell-Taggart also discriminated in price and sold bread products below their fully allocated costs. While it should not be presumed that any wholesaler wil engage in illegal practices, the record facts do point out how vulnerable complaint counsel's newly created one- plant wholesale bakers wil be to the competition of major multiplant competitors, who apparently will not be under conduct prohibitions such as those contained in the proposed order in this case. I am not certain that the newly created companies would be any more competitive than Inglis, Gordon, Prosser or Laub proved to be. In spite of the illegal practices in which respondents were found to be engaged Continental does provide competition that checks the market power of those other multiplant corporations with which Continental now shares a monopoly in these marketing areas. (100) Although the legal and factual issues in this case turn on the importance of white pan bread to wholesale bakers, as well as to retail grocers and consumers, it is also clear that successful wholesale bakers sell a complete line of bread products, including variety bread. INTERNATIONAL TELEPHONE & TELEGRAPH CORP., ET AL. 389 280 Initial Decision and muffns. The record shows that during the relevant time periods during which the challenged practices took place, many of Con tin ental' s bakeries specialized in producing particular products. For example, the Minneapolis bakery produced mainly white bread and white bread buns and rolls, whereas the Rochester bakery produced mostly variety breads. The Akron bakery produced only white bread and was supplied with variety breads from other Continental bakeries in the Detroit region. Although the Beverly Hils bakery appears to have had the capability of producing both white pan bread and variety breads, it was supplied with buns and rolls and certain variety breads from the DiCarlo and the Salt Lake City bakeries. In my opinion, the viability of complaint counsel's newly created wholesale bakers must depend in large part on their capability of producing, or purchasing for resale at wholesale, a full line of bread products. I agree with respondents that the record does not contain specific information about the current competitive posture or financial condition of the bakeries targeted for divestiture. One of complaint counsel' s expert witnesses who recommended divestiture as a possible remedy in this case, agreed that the feasibility of any divestiture would depend upon a very extensive examination of the bakeries to be divested in terms of production, finance and economics (8083 Boyle). Although such a record diffciency could be remedied by reopening the record in this case, a greater problem exists. It can be anticipated that any divestiture that may be ordered by theCommission wil not actually take place until the mid-1980' , because any such order wil not have any force and effect until it becomes final following Commission decision and court review. Moreover, during this period of time, respondents wil not be subject to the ancillary prohibitions of the order relating to transferring personnel, holding separate the operation ofthe bakeries, and preserving the viability of the bakeries as such separate entities.

Although it is impossible to predict what changes in facts and circumstances may take place during this interim, it is highly unlikely that complaint counsel, by creating five new independent one-plant wholesale bakers, can accomplish their intended purpose of restoring the competition that was lessened by the exit of independent wholesalers from the various markets. I am also of the opinion that the targeting of individual plants for divestiture at this time is not appropriate. Ifultimately, such a style of divestiture is deemed appropriate the designation ofthe individual plants to be divested should be made on the basis of the competitive realities and financial conditions that exist at the time of the divestiture. (101) During the hearings, complaint counsel's expert witnesses also proposed the divestiture of groups of Continental bakeries in geo- Initial Decision 104 F. graphic proximity to one another, They also recommended that ITT be required to divest Continental.

It is not clear why complaint counsel have abandoned these alternative forms of divestiture. In my opinion, the practices of sellng below cost and discriminating in price over relatively long periods of time in selected marketing areas for the purpose of driving competitors out of the market or "conditioning" them to accept and follow Continental's price leadership was only possible due to respondents ' vast economic resources with which they- subsidized the losses incurred by their individual bakeries which were engaged in such practices. The only way to ensure that such economic power is not misused in the future is to eliminate the source of the power which appears to engender such anticompetitive practices. Complaint counsel appear to rely upon such a rationale regarding the proposal to have Continental divest its Salt Lake City plant.

The Administrative Law Judge recommends to the Commission that it consider ordering ITT's divestiture of Continental and the restructuring of Continental, with or without divestiture, into financially autonomous regional wholesale bakery units. This restructuring of Continental should overcome the problems inherent in situations where individual bakeries do not have the capability of producing a full line of bread products, a fact which makes the divestiture of individual plants impracticable, With the possible exception of some transshipments of products from the Braun bakery in Pittsburgh to the Youngstown and Akron bakeries, there does not appear to have been any serious product interdependence between Continental' s regions.

Respondents' objections to the provision ofthe order banning future acquisitions are well taken. The record in this case does not support any finding that respondents have ilegally acquired any bakery or bakery assets during the time periods involved or in the marketing areas which were the subject of this proceeding. In my opinion, such a ban bears no reasonable relationship to the ilegal practices in which respondents were found to be engaged. The paragraph contained in complaint counsel's order that prohibits price discriminations is directed to differences in price between competing customers. It would not appear to cover so-called territorial price discriminations as found to have been implemented in the Minneapolis and Rochester marketing areas. In this respect, the price discrimination prohibition does not appear to be directed at preventing direct subsidization of below cost prices in one marketing area through higher prices in a second marketing area. Moreover, the proposed price discrimination paragraph could be interpreted, as respondents contend, to prohibit (102) outright any INTERNATIONAL TELEPHONE & TELEGRAH CORP., ET AL. 280 Initial Decision sale of private label or secondary label at any wholesale price lower than the wholesale price of the same size loaf of advertised label bread. The record in this case shows that such a price difference although discriminatory, may not have an adverse effect on competition if the lower wholesale price for private label bread or secondary label bread is above Continental' s competitors' average total costs of producing the same product, including a margin of profit, and where the resulting retail price spread between private label and advertised label bread is not extreme.

Complaint counsel take the position that their price discrimination provision in the proposed order "prohibits unlawful price discriminations. " In my opinion, a separate paragraph covering private label and secondary label bread should be included in the order which adopts some ofthe statutory language relating to competitive effects. Such an all inclusive provision would clarify the ambiguities that appear in the version proposed by complaint counsel. In addition, a paragraph wil be added prohibiting territorial price discriminations resulting in prices on advertised label bread that undercut the prevailng market price for advertised label bread and injure competition between competing wholesale bakers.

The "meeting competition" and other statutory exceptions or defenses are implicit in any price discrimination order, unless specific determination is made that certain conduct does not meet the requirements of the statute, such as sellng below fully allocated costs. The absolute ban on sales below Continental's fully allocated costs appears to be unreasonably harsh. Although respondents must expect some "fencing in " the order should not place them in an impossible business position. See Federal Trade Commission v. National Lead Co., 352 U.S. 419, 431 (1957). The principal vice found in this case is the predatory use of sales below fully allocated costs to take business away from competing wholesale bakers in selected markets. In my opinion, sales below Continental' s fully allocated costs should be permitted only in situations where competing wholesale bakers are actually selling below Continental's fully allocated costs and where Continental affords a price below fully allocated costs to retain the business of a specific customer.

For the above reasons, the provisions ofthe order relating to divestiture, compulsory licensing and future acquisitions wil be deleted from complaint counsel' s proposed order. The price discrimination and sales below cost provisions wil be modified to conform with th, views expressed in this section of the initial decision. (103) Initial Decision 104 F. ORDER For purposes of this order, the term respondents refers to respondents International Telephone & Telegraph Corp. and ITT Continental Baking Company, Inc., and the successors, assigns, offcers, directors agents, representatives, employees, subsidiaries, and affliates of either of them and the term bread refers to the products encompassed by the Bureau of Census' Standard Industrial Classification Codes 20511 and 20512 (1973), commonly known as bread and bread type rolls.

It ordered That respondents, directly or indirectly, in connection with the sale of bread in or affecting commerce, as Hcommerce" is defined in the amended Clayton Act, shall cease and desist from discriminating, directly or indirectly, in the price of advertised label bread between purchasers in the same line of coinmerce, by granting a price reduction from the established market price in any market where respondents are in competition with any other wholesale baker if the effect ofthe lower price may be to substantially injure competition between respondents and their competitors engaged in the wholesale baking business in the market in which the price reduction is granted unless it proportionally reduces its prices everywhere on the same advertised label bread products.

It is further ordered That respondents, directly or (104) indirectly, in connection with the production, marketing and sale of bread in or affecting commerce, as "commerce" is defined in the amended Clayton Act, shall forthwith cease and desist from discriminating in the price of any advertised label bread product and the price of any private label or secondary label bread product of like grade and quality 'y selling to any purchaser at a higher price than the price charged ny other purchaser who, in fact, competes with the purchaser paying 'te higher price and where the effect of such discrimination may be Ibstantially to lessen competition between respondents and their 'mpetitors engaged in the wholesale baking business or between spondents' customers paying the higher price and respondents ' cus- TIers paying the lower price.

INTERNATIONAL TELEPHONE & TELEGRAH CORP., ET AL. ;J";J 280 Initial Decision It is further ordered, That respondents, directly or indirectly, in connection with the production, marketing and sale of bread in or affecting commerce, as "commerce" is defined in the Federal Trade Commission Act, shall forthwith cease and desist from sellng any bread product below their fully allocated costs; provided that nothing herein contained shall prevent respondents from sellng below their fully allocated costs in order to retain an established volume'ofbusiness with specific customers and where competing wholesale bakers are actually selling identical products at the price at which respondents sell such products below their fully allocated costs. (105) It is further ordered That respondents shall maintain, at all times records of its cost of producing and distributing each bread product. Such records shall be kept up to date so that no such record is more than six months old. The costs of manufacturing, sellng and distribution and overhead shall be reflected in such records, so that the fully allocated cost of each product can be ascertained. The fully allocated, incremental, average variable or marginal cost reflected in any such records shall be binding upon respondents in any enforcement proceeding involving violation of Paragraphs I, II and III of this Order. In the event the costs of any bread product is at issue in an enforcement proceeding involving violations of Paragraphs I, II and III of this Order and respondent cannot produce such a cost record of a product at issue, respondents may not object to the use ofa cost record pertaining to an advertised label ofthe same grade and quality as a product at issue to show the costs of a product at issue. Such cost records shall be kept for ten years.

It is further ordered That respondents shall, within sixty (60) days after the date of service ofthis Order submit in writing to the Federal Trade Commission a verified report (106) setting forth in detail the manner and form in which respondents intend to comply, are complying or have complied with this Order.

It is further ordered That respondents notify the Commission at least thirty (30) days prior to any proposed change in the corporate , . Opinion 104 F.

respondents such as dissolution, assignment or sale resulting in the emergence of a successor corporation, the creation or dissolution of subsidiaries or of any other change in the corporation which may affect compliance obligations arising out of the Order. OPINION OF THE COMMISSION By DOUGLAs, Commissioner:

Introduction The complaint in this matter alleges that International Telephone and Telegraph Corporation, through its wholly owned subsidiary ITT Continental Baking Company, Inc. (hereinafter referred to as "Continental"), violated Section 5 of the Federal Trade Commission Act, 15 C. 45 (by violating Section 2 of the Sherman Act, 15 U.s.C. 2), and Section 2(a) of the Clayton Act, as amended by the Robinson-Patman Act, 15 U. C. 13(a).I Count I of the complaint alleges that Continental engaged in the following practices, inter alia: (1) Sales of bread (at pricesJ below. - . cost or at predatory prices for substantial periods of time in various geographic markets; (2) (2) Subsidization of sales below cost or at predatory prices in various geographic markets by sales at higher prices in Jess competitive geographic markets; (3) Discriminations in price, directly or indirectly, between purchasers of bread of like grade and quality.

Count I concludes by alleging that these practices represented an attempt by Continental to monopolize and injure competition in the wholesale baking industry in relevant geographic markets in violation of Section 5 of the Federal Trade Commission Act. ! The following abbreviations are used in this opinion: initial decision page number IDF initial decision finding number Tc. - transcript oftfJstimotly page number - complaint cOllosfJl's exhibit number CAP - complainl counsel's appeal brief CAB - complaint counsel's answeriog brief CRB - complaint counsel's reply brief CMF - complaint counsel's memorandum supporting proposed findings of fact and cooclusions of law CPF - complaint counsel's propo,j(d finding of fact number - respondent's exhibit number RAP respondent's appeal brief RAB respondent's answering brief RRB - respotldf'nt's reply brief RMF respondent's memorandum supporting proposed findings of fact and conclusions of law RPF - respondr.nl's proposed findings of fact number For the purposes of the complaint bread" is defined to encompass "white pan bread and bread type rolls and related products, INTERNATIONAL TELEPHONE & TELEGRAPH CORP., ET AL. i)Vv 280 Opinion Count II of the complaint alleges inter alia that Continental discriminated in price, directly or indirectly, between different purchasers of bread, by selling bread of like grade and quality to some of such purchasers at substantially higher prices than to other of such purchasers.

Count II concludes by alleging that these practices violated Section 2(a) of the Clayton Act, as amended by the Robinson-Patman Act. The Administrative Law Judge (ALJ) determined that Continental sold white bread products at predatory and discriminatory prices for significant periods of time in five relevant geographic markets: Cleveland, Denver, Northern California and Western Nevada, Southern California, and the Twin (3) Cities of Minneapolis and St. Paul.4 The ALJ concluded that Continental had attempted to monopolize white bread product sales in each of those markets; should be held liable for primary-line price discrimination in each of those markets; and should be held liable for secondary-line price discrimination in the Cleveland and Northern California/Western Nevada markets. The ALJ issued an order that would prohibit Continental (1) from discriminating in the sale of advertised label and private label bread products, where the effect ofthe discrimination might be to substantially injure or lessen competition among competing sellers, or to substantially lessen competition" among competing buyers; and (2) from sellng such products at prices below fully allocated costs, with certain exceptions.

On appeal, Continental argues that it did not attempt to monopolize the relevant markets or engage in illegal price discrimination, and that the complaint in this matter should be dismissed. Complaint counsel argue that the findings and order of the ALJ should be sustained, and that in addition the Commission should (I) require Contilocated nental to divest its "plants, (4) equipment and other assets" in Sacramento and Beverly Hils, California; Akron, Ohio; Minneapolis, Minnesota; and Salt Lake City, Utah; and (2) prohibit Continental for ten years, from acquiring any bread manufacturing facility without first securing Commission approval.

The complaint raises important questions as to the extent to which the Commission should act to prevent firms from selling at allegedly 3 The complaint contains a variety of other allegations, but the allegations described above represent the gravamen of the complaint and sub ume all its other allegations 4 The ALJ concluded that Continental did not attempt to monopolize a sixth relevant geographic market in Seattle, Washington. Complaint COlJnsel have not appealed that detennination , Primary !idc discrimination cases under the Clayton Act address injl.lry to competition among rivalellers while secondary line discrimination cases address injury to competition among favored and disfavored buyers. 6 The order would permit sales at prices below fl.11y allocated coo;t in order to "retain an established volume of business with specific customers," where "competing wholesale bakers arc actually selling identical products" at thesameprin\ u"u FEDERAL TRADE COMMISSION DECISIONS Qpinion - 104 F.

predatory or discriminatory prices. As a general proposition, firms should be accorded the discretion to set prices at whatever levels they choose in response to competitive conditions, since permitting that flexibility is most likely to maximize consumer welfare. Section 2 of the Sherman Act and Section 2(a) of the Clayton Act are intended inter alia, to prevent pricing conduct that injures competition and therefore reduces consumer welfare. However, as we noted in General F(Jods, overly broad efforts to apply these standards may sometimes chil "the rivalry that is the essence of dynamic competition" by discouraging aggressive price and non-price competition.7 Therefore the legal and economic standard for evaluating allegedly predatory or discriminatory conduct should carefully distinguish the structural conditions and behavioral patterns that are likely to improve competitive performance from those that are likely to injure competition. We have concluded that Continental' s conduct did not violate either the Federal Trade Commission Act or the Clayton Act, and have therefore determined to dismiss the complaint in this matter in all respects. (6) I. INDUSTRY CHARACTERISTICS The International Telephone & Telegraph Corporation is a large corporation whose $10.2 bilion in sales placed it ninth among domestic corporations in 1973. IDF 1. Continental, its wholly owned subsidiary, manufactures and distributes a wide variety of food products, including bread, cake, snacks such as potato chips, and frozen prepared foods. In 1972, Continental's total net sales amounted to $865 milion. IDF 2. Continental has been the largest bread and snack cake baker in the United States since 1924; in 1970, its $375 milion in sales of these products represented 12.4 percent of total industry sales nationwide." In 1973 its bread sales amounted to approximately $445 millon; by 1977, they had increased in nominal terms to $609 million.9 Continental's largest competitors in wholesale baking include Campbell-Taggart, American Baking Company, Interstate Brands Corporation, Flowers Bakery, and Metz Baking Company. IDF 10. In addition, it confronts substantial competition from the large captive )akeries that many retail grocery chains operate to produce most of :he bread they sell in their retail stores. See IDF 19. Wholesale bakers typically produce white bread, hamburger buns nd hot dog rolls, as well as variety breads such as whole (7) wheat , and pumpernickel breads. IDF 20. Economies of scale are impor- , Generul Foork Corp.. Trade Reg. Rep. (CCII) TI22,142 (Aprij 6, 1984), al 22 973. (103 F. C. 204 (1981)1 , IDF 4; I1T Crmti/!f!nlnl Bnking Co. 84 F.T.C. 1319, 1395and n.4 (1974). IDF 3, 43. In constant 1973 dollars (as adjusted by the prodcwer price index for total finished goods), Continens bread sales actually declined to $129 milion in 1977.Sfe Economic Report of/he Pre ident(February 1 , 1984), :85.

.... . , lil J..I. J.U 280 Opinion tant in the baking industry; moreover, wholesale bakers must operate their bakeries at approximately eighty percent of maximum production capacity (encompassing two full production shifts and some overtime) in order to maximize production effciency.'o The costs of producing white bread are generally uniform among bakeries. However, chain store bakers enjoy a distributional cost advantage that is attributable to differences in delivery methods.'! Most wholesale bakers sell bread primarily to retail grocery stores; their other bread customers include restaurants, institutions, and hotels.'2 Continental and most other wholesale bakers market an advertised label" line of bread products sold under a brand name that is usually owned and promoted by the wholesaler. Continental' principal advertised brand is "Wonder Bread " it accounts for most of its white bread sales. IDF 29. Continental and most other wholesalers also sell "private label" bread products-usually including the most popular white bread (8) loaf sizes, and hamburger and hot dog rollsthat are packaged for sale under the private labels of retail grocery store chains. IDF 30. Finally, some wholesalers sell "secondary label" bread to independent grocers that are too small to have their own private label programs.'3 Nearly all remaining bread that is sold in retail outlets is "captive label" bread; that is, it is produced by the retail grocery chain that sells it. In 1971, captive and private label bread accounted for thirty-six percent oftotal white bread sales in the United States; by 1977, that percentage had climbed to approximately fifty percent. IDF 32, 38. Most retail grocery stores, including chains with their own bakeries, will carry limited quantities of the advertised label breads of each of the wholesale bakers active in a given area. IDF 34.

The "quality, nutrition, palatibilty and physical features" of white bread are relatively similar from one brand to another.'4 The wholesale and retail prices of advertised label bread are typically higher than those of private, captive, controlled, and secondary label breads largely because wholesale bakers have been able to create consumer franchises for their (9) advertised label bread products. IDF 35, 36. However, as the size of this price differential increases, advertised 10 IDI- 23; Shaus, Tr. 11204-5, 11223-.24; HeeJakacki, Tr. 10295, 10300-02. 11 IDF 27. Retail chain stores use semitraillJrs, driven by hourly rate Teamsters, to deliver their bread to the doors" oftheir retail outle!$, for she1fpJacement by their OW!) employees. By contrast, wholesale bakers generally use more highly paid "routesalesmen Teamsters who carn a salary pluscoommission to distribute their bread products to retail grocery stores and rcstaurants, and place the bread on the shelves. IDF 25-27. IDF 24. For example, restaurants and institutional accounts represent approximately thirteen percent of Continental's total route sales.Id.

13 IDF 31. A wholesaler may also sell bread to a given retailer under a label that the wholesaler controls, but to which the reu:ti1lJr is accorded exclusive license in a particular ..rea. This bread is referred to as "controlled labe)" bread because the label is ultimately controlled by the wholesaler rather than by the retailer./d. H IDF 35. This dm$ not mean, however, that aU white bread should be considered to be identical. There are of course some quality; nutrition, palatability, and physical difTIJrences among brands. Opinion 104 F.

label bread loses more and more sales to private, captive, controlled and secondary label breads. Conversely, as the size ofthe diferential declines, these other types oflabels lose more and more sales to advertised label breads. IDF 36.

The bread industry has undergone some very substantial changes over the last twenty years, and an earlier Commission case involving Continental produced some useful evidence as to competitive conditions during the time period at issue in this case. In 1962, the Commission issued a consent order against Continental that inter alia prohibited it from acquiring any interest in, or the assets of, any concern engaged in the production and sale of bread and bread-type rolls without Commission approval. In 1968, International Telephone & Telegraph acquired Continental as a wholly-owned subsidiary.!6 On April 27, 1972 the Commission issued an order requiring Continental to show cause why the Commission should not extend the acquisition ban for an additional five years.!' On November 26 , 1974 the same day it issued the complaint in this matter-the Commission reversed the decision of the (10) Administrative Law Judge to extend the ban on acquisitions without Commission approval and dismissed the order to show cause.!8 The opinion of the Administrative Law Judge in the earlier case provides a useful overview of economic conditions in the bread baking industry between 1963 and 1973. The ALJ found that the number of plants and producers in the industry had declined substantially during that period, and suggested a number of explanations for the industry s problems:

At least since the beginning of 1973, 90 percent of all wholesale bakeries have been operating at a loss. The president of Quality Bakers of America estimated that about 50 percent of his cooperative members are running in the red and that the industry is in a state of disaster. Wards went through a radical reorganization. Interstate was almost acquired by Beatrice Foods. Most wholesalers are closing planlc;. Many smaller bakers are going out uf business.

The wholesale bakers' problems have been caused, in large part, by the activities and growth of chain bakeries, costly lahor contracts, technological improvements that have created overcapacity, improved highways permitting larger areas to be served by a single plant, high ingredient, selling and distribution costs, problems arising from Federal price controls in effect since Aug. 1971 , and the decline in per capita consumption of bread. Of particular current impact is the tremendous (11) increase in the cost of flour following the shipment of flour to Russia in 1972 and the corresponding short 15 Continental Bakinr ., 60 F. C. 1183, 1193-94 (1962) (consent order). 06 ITT Continentul Baking Co. 84 F. C. 1349, 1393-94 (1974) 1d.at1394 .8 ITT Continental Baking Co 84 F. C. 1349, 1400. Although it is not completely clear from the record, the Commission may have i31ued the complaint in this matter becau.se it was fiot Sllre-as of 1974-"whether competition in the various local afid regional markets ill the country was increasing, decreasing, Dr holding steady. . . and whether any such changes were attributable to merger activity "or to smne other eme or causes. Id. 1394-95.

..

IN'l.t1''1UJ.vn.. ..

280 Opinion supply here coupled with the inability to r coup co ts b cause- pricecontrols. These factors are generally found, to varying extents, in all markets, whether or not lit Continental is present.

In its opinion dismissing the show cause order, the Commission confirmed the ALJ's assessment of the economic condition of the bread baking industry:

The number of bread producers has been declining sharply for many years. . . While a substantial number of these departing firms were acquired by the larger members of the industry in the pre-1964 period, a major factor here has clearly been a series of technological changes in the industry that have significantly increased the minimum effcient plant size. As more and more firms have sought to get their per-unit costs down to the minimum level by incorporating the newer and lower-cost technology in their plants, the productive capacity of the industry has sharply outstripped the growth in consumer demand. The result is that the industry has been suffering from chronic excess capacity for many years, a situation that of course further intensifies the com petitive struggle (12) for volume.20 The smaller firms, especially those with older plants and equipment, have thus found themselves operating not only high-cost facilities but operating them at less than full capacity. Not being able to match the lower costs and prices of their more technologically advanced rivals, many of these smaller firms have been forced to either close their doors or sell out to other firms in the industry. In short, during the 1963-1973 period, the bread baking industry confronted substantial and chronic excess capacity created largely by technological improvements; high ingredient, sellng and distribution costs; Federal price controls that became effective in August, 1971; and a decline in per capita bread consumption. These conditions existed to varying degrees in all parts of the country, regardless of the presence or absence of Continental in any particular market. It therefore not surprising that a number of firms left the industry during this period, and that a large number of other firms suffered losses; indeed, it would be surprising if a period of such widespread economic and technological change did not" produce precisely those effects that are the gravamen of the complaint in this matter. (13) The record evidence developed since the complaint was issued provides no reason to conclude that any of the economic conditions the Commission described in its 1974 opinion should be attributed to Continental' s pricing behavior in any ofthe relevant geographic mar- 19 ld at 1381 (Initiai Decision) (citatioos omitted)- The ALJ concluded that the acquisition restriction ill the 1962 order should be extended until Aprij 13, 1977, He based this cOllcJusioll principally UpOIl the fact that "iodu.strywide concentration" had increased nationwide since 1962, despite his determination that the relevant geographic markets were regional or local, and the absence of evidence that Continental's alleged acquisitions had increased concentration in any relevant regional or local market.ld. at 1387- . The Commission reversed because of the absence of evidence of injury to competition in the relevant local bread markets.ld. at 1399. The Commission noted, for example, that in 1958 the industry was reportedly operating at only forty ptlrcent t. sixty percent of capacity-ld. at ) 396 n. 12. The ALJ in this case concluded, by contrast, that there was "no chronic excess capacityin the bread baking industry du.ing the relevant periods" ID at 19. However, the record evidence does not support that conclusion.See Metz, Tr. l0223 Jakacki, Tr. 10273-76; CX 1355; RX 1000; RX 1101- 21 lit Continental Baking Co 81 F. C. at 1396 and n. 12. :!,:! ); ); ); ); !, ); ), Opinion 104 F.

kets. The Commission has therefore determined to dismiss the complaint in this matter in all respects. A more detailed discussion ofthe attempted monopolization and price discrimination allegations follows.

II. SECTION 2 OF THE SHERMAN ACT: ATTEMPTED MONOPOLIZATION In its classic formulation of the offense of attempted monopolization, the Supreme Court determined:

Where acts are not suffcient in themselves to produce a result which the law seeks to prevent for instance, the monopoly-hut require further acts in addition to the mere forces of nature to bring that result to pass, an intent to bring it to pass is necessary in order to produce a dangerous probability that it wil happen. Somewhat more recently, the Commission has described the elements of the offense in the following fashion: (14) . (TJhe attempt offense includes three principal elements: (1) specific intent to control prices or destroy competition, (2) exclusionary or anticompetitive conduct, and (3) a dangerous probability of success.

This portion of the opinion discusses each of these elements in some detail.

A. Specific Intent Proving the first element ofthe attempted monopolization offense requires establishing a specific intent to control prices or otherwise injure competition.24 However, the Commission has (15) emphasized that the specific intent element depends importantly upon the nature of the conduct that a firm employs pursuant to that intent: As a general matter, it seems unwise to find that a firm has the requisite specific intent for anticipating the exclusionary consequences of successful competitive behavior 22 Swift Co. v. United States 196 lI.S. 375, 396 (1905). E./ DIl Punt de Nemours Co. 96 F, C. 653, 725 (1980); flccord, General Foods Corp., supra note 7, H22 142 at 22,973; D.E. Rogers Associates. Inc.v. Gardner-Denuer Co. 718 F.2d 1431 , 1435 (6th Gir. 1983),cert. denied, US.LW. :l886 (U,S, June 12, 1984) (No. 83-1698);Wiliam Ingli., Sons Baking Co. v. lit Continental Baking Co. 668 F.2d 1014, 1027 (9th Cir. 1981)cerl. denied 103 Ct. 58 (1982);United States v, Dairymen, Inc. 660 F. 192, 194 (6th Gir. 1981);N"rlhew;lem Telephone Co. v. American Telephone Telegraph La., 651 !,'2d 76, 85 (2d Gir. 1981),cert. denied 455 DB- 943 (1982); Chil/icotheSand Gravelv. Martin Mariel/a Corp. 615 F.2d 427 430 (7th Cir. 1980);California Computer Products. Inc. IBM Corp. 613 F.2d 727, 736 (9th Cir. 1979);Paci(ic Engineering Production Co. v. Kerr.McGee Corp. 551 2d 790, 791 (IOthCir.1977); Merit Motors, Inc. v. Chrysler Corp. 417 F:Su.pp. 263, 269-70 (D. C. 1976), a(rd, 569 F.2d 666 (D.C. Cir. 1977). Times Picayune Publishing Co. Uniled States 345 U.S, 594, 626 (1953) ("a specific intent to destroy competition or bund monopoly E. Rogers Associates, Inc. v. Gardner-Denver Co. 718 F.2d at 1435 ("specific intent to monopolize William Inglis&SonsBaking CO. V. ITTContinenl/l1 Raking CO. 668 2d at 1027 ("specific intent to control prices or destroy competition Northeastern Telephone CO. V. American Telf!phlJne Telegraph CQ., 651 2d '-t 85 (" specific intent to monopolize Chillicothe Sand Gravelv. Marlin. Marietta Corp- 6 I:! F. lit 430 ("specific intent to control prices or destroy competition United Sta.tf!Sv. Empire Ga.B Corp. 537 F.2d 296 302 (8th Cir. 1976) ("an intent to control prices or to restrict competition unreasonably cerl. denied 429 U'.8, 1122(1977) 280 Opinion which leads or may lead to a monopoly, so long as that behavior is reasonable. To suggest otherwise would he to proscribe all acts in which firms conjure up some thoughts of achieving monopoly irrespective of the actual character of the means employed to gain that end.

In short, although evidence of a specific intent to injure or destroy competition can be helpful in establishing liability for attempted monopolization, its utility depends crucially upon the nature of the conduct at issue. Any successful business strategy wil injure competitors to some degree; it satisfies the specific intent requirement only if it contemplates doing so by means of anticompetitive conduct. As the Commission recently stated, the specific intent element is not satisfied by ambitious and aggressive plans to compete, even with the goal of taking business from competitors or vanquishing a troublesome rival. The (16) antitrust laws provide no protection from such designs, where the means to effectuate them amount to no more than vigorous competition. B. Anticompetitive Conduct Proving the second element of the attempted monopolization offense requires a thorough evaluation of the conduct that the firm involved employed. In Dupont the Commission determined that essentially three criteria should be considered in conducting such an evaluation: (1) whether firms without substantial market power would find the conduct at issue to be profitable or economically rational; (2) whether the conduct improves product performance; and (3) whether industry conditions such as high entry barriers are likely to mitigate or accentuate any anticompetitive effects of the conduct. When properly defined, predatory pricing satisfies these criteria, because it is highly unlikely that firms without substantial market power wil find it either profitable or otherwise economically rational; it is highly unlikely to improve product performance; and whether it wil prove to be successful is largely a function of a variety of structural industry characteristics. Of course, a determination that a given firm has sold at predatory prices for a significant period of time does not in and of itself establish liability under the Sherman Act or the Clayton Act. It satisfies the "specific intent" and (17) "anticompetitive conduct" components of the attempted monopolization offense but does not, without more, satisfy the "dangerous probability of :1 E.I. Du Pont de Nemours& Co. 96 F. C. 653, 727 (1980); accord, e. , Geneml Foods Corp., supmnote, ,:22, 142 at 22 974; D.E, Rogers Associates, Inc.v. Gardner-Denver Co. 718 F.2d at 1435; Wiliams Inglis Suns Baking Cu. v. /1TContinental Boking Co. 668 F.2d at 1028, 1031 n. 18; Lektro- Vend Corp.v. Venda Co., 660 F-Zd 255, 273 (7th Cir. 1981);Buffolo Crmrier Express, Inc. v. Buffalo. Evening News, Inc. 601 F.2d 48, 54 (2d Cir. 1979);Hayes V. Solomon 597 F.2d 958, 977 (5th Cir. 1979),cat. denied,444 U.s. 1078 (1980). Geneml Foods Corp_, supmnote 22, 142 at 22 974 (citations omitted). 27 E. I. Du.pont 96 F. C. 653, 738-39 (1980). Opinion 104 F.

success" requirement. It also satisfies the Hpredatory pricing" requirement for primary-line liability under Section 2(a) of the Clayton Act, but does not satisfy the jurisdictional requirements of that section.

The diffcult question is of course how to define predatory pricing. It is crucially important to distinguish prices that are perfectly consistent with competitive behavior from prices that are not. Price is the central nervous system" of the economy, 28 and vigorous and healthy competition engenders economic effciency which redounds to the benefit of consumers. By contrast, overly zealous efforts to prevent sales at prices below cost are likely to reduce competition and increase prices.29 Therefore, we must carefully avoid adopting a predatory pricing rule that wil deter legitimately competitive pricing conduct. The significant likelihood of injury to competition from erroneously prohibiting low prices should be contrasted with the low likelihood that an unrestrictive rule (18) wil erroneously permit anticompetitive conduct, given the low likelihood that any predatory strategy will prove successful. As Philip Areeda and Donald Turner have pointed out proven cases of predatory pricing have been extremely rare. ... That predatory pricing seems highly unlikely does not necessarily mean that there should be no antitrust rules against it. But it does suggest that extreme care be taken in formulating such rules lest the threat of litigation. . materially deter legitimate, competitive pricing. An ideal predatory pricing rule must therefore satisfy two criteria. First, it must distinguish predatory intent from competitive intent; that is, it must distinguish pricing behavior that is very likely intended to injure competition from pricing behavior that could very well be directed toward perfectly legitimate competitive objectives. Sales at prices below average variable cost-as properly defined-for a significant period of time may well satisfy this requirement. They are more likely intended to injure competition than to achieve legitimate competitive objectives because they do not cover any fixed costs of operation, and do not cover all of the variable costs of operation. The. firm that sells at such prices consequently loses (19) more money continuing to operate than by shutting down altogether. Sustained United States v. Socuny- Vacuum Oil Co. 310 U.s. 150, 224-26 n. 59 (1940). 29 The National Recovery Administration s effort to enforce its code against "destructive price cutting" provides a good example of this latter phenomenon- An economist associated with the program and a contemporary Brookings Institution study separately concluded that extensive enforcement ofthe code tended to reduce competition, to raise prices, and, in some cases, to injure small competitors that could not capitalize upon their greater ef1ciency. See L. Lyon et al., The National Rer:oery Administration: An Analysis And An Approisal604-5 620-21 (1935); C. Roos,NRA Economic Planning 249, 25 9, 275, 407--8, 416 (1937). 3U P. Arecda and D. Turner, IIIAntitrust Law152 (1978). At least two commentator.' have argued that predation is so rare that it should be completely ignored, in order to avoid deterring legitimate competitive pricing. R. Bork The Antitrw;t ParauI149-155 (1978); J.asterbrook PredatoryStmtegies and Cou.nterstrategies 48 U. Chi. L- Rev. g, 280 Opinion sales at such prices can therefore b presumed to be intended to injllre competition.

An ideal predatory pricing rule must secondly distinguish pricing behavior that is likely to injure competition in the generality of cases from pricing behavior that is not. The antitrust laws focus upon preserving or enhancing consumer welfare by preserving or enhancing competition. One effect of healthy competition is to redirect production and sales from less effcient firms to more effcient rivals. Therefore, one logically ought to determine the pricing level that is likely to force equally effcient firms to shut down, with the effect of-injuring competition. 32 A price that forces an equally effcient (20) firm to sell at a price below its own average variable costs for a significant period oftime satisfies this criterion. Because sales at such prices do not even cover the variable costs of operation, an equally effcient firm wil ordinarily shut down completely rather than continue to operate. Since its shutdown is induced not by competitive conditions but rather by anticompetitive conduct on the part of the predatory firm, it injures competition and therefore worsens consumer welfare. The most appropriate predatory pricing rule wil therefore satisfy two separate requirements: (I) a predatory intent requirement-sales at prices below average variable cost for a significant period of time; and (2) a predatory conduct requirement-sales at prices that are likely over a significant period of time to force equally effcient firms to shut down completely. Sales at prices below the average variable costs of an alleged predator for a significant period of time wil presumptively satisfy both of these criteria. The Commission has therefore concluded that it should rely upon average variable cost to distinguish presumptively legitimate prices from presumptively predatory prices, in the following fashion: (1) Sales at prices that equal or exceed average variable cost should be strongly, often conclusively, presumed to be legal. This presumption could possibly be rebutted in some circumstances (21) by a strong showing that sales at such prices were not, absent the effect of the alleged predation, consistent with profit maximization or loss minimi- 31 This analysis of predatory intent focuses upon average variable cost rather than marginal cost to distingush predatory irom competitive objectives.Compare, e. P. Areeda and D. Tumer, III Antitrust Law 153 (1978). That is because a firm wil produce at an output at which marginal cost equals price (average revenue) only Ulder perfectly competitive conditions. Any firm with some degree of market power-that is, any firm that confronts a downward.sloping demand cure, and is therefore to some degree a price makerrather than a pricetoker-wil seek to maximize profits or minimize losses by seJJng at the point at which (declining) marginal revenue equals marginal costs, rather than at the point at which (8 constant)price equ.als marginal cost. 32 Two firms are "equally effcient" jf each firm minimizes its average total costs at the same level. Scherer Predatory Pricing and the Sherman Act: A Comment 89 IIar. L. Rev- 868, 872 n. 10 (1976); Transamerico Computer Co. v- IBM Corp. 481 F.Supp. 965, 991, 992 (N.D. Cal 1979), affd, 698 F.2d 1377 (9th Cir. 1983),cert. denied 104 S.Ct. 370 (1983);seer. Posner Antitrust Law: An Economic Perspeccive188-9. When ths is the case, each finn can produce a given quantity of output with precisely the same quantity of resources. :! ), ., Opinion 104 F.

zation.

(2) Sales at prices below average variable cost for a significant period of time should be rebuttably presumed to be anticompetitive.34 This presumption could be rebutted by showing, for example, that the sales at issue (1) were of products that were obsolete, perishable, or otherwise subject to rapid value deterioration; (2) were at introductory prices designed to induce trial; or (3) were (22) made to avoid losing goodwil, such as product loyalty, that it would be very costly to regain were the firm to shut down and then reopen later. (3) Sales at prices that equal or exceed average total cost should be conclusively presumed to be legitimate.36 Sales at such prices cannot by definition exclude equally effcient firms. They may, of course exclude less effcient firms, but that is often an effect of vigorous and healthy competition. (23) Two important considerations should be addressed in conjunction with this standard. First, although it is diffcult to precisely determine the duration of a "significant" time period, it should be suffciently long to make it likely that sales at prices below average variable cost could in fact force equally effcient firms to exit. Discontinuous or episodic instances of sales at such prices are unlikely to satisfy this standard, because they may very well be nothing more 33 General Foods Corp., supra note 7, TI22,142 al22,975-76 (citations omitted). Complaint counsel argue that this standard should not apply "when predatory pricing can be cross-subsidized currently from profits in the predator more secure markets" CAB at 16. However, even lfa firm earns greater profits in some markets than in others sales at prices equal to or greater than average variable cost whl not make aenae as a predatory strategy because they wiil not force equally effcient finns from those markets, and subsequent recoupment of I05s!l in those markets wil therefore !lot be pOilible. Complaint counsel a130 suggest that the legitimacy of 8Ilcs at prices below full (untJ cost" should vary with the degree to which, and the sales volume and duration for which, such prices fall below full cost. GAB at 26-27. Apart from its conflct with the strong presumptive lef"Tjtimacy of prices equal to or greater than average variable cost, the "rule of reason" standard complaint counel propose wowd he very difcult to apply in practice, with very uncertin results, and we therefore decline to adoptit. 34 General Foorh Corp., supra note 7, 1122,142 at 22 975-76 (citations omitted). The concept of "average variable costs" IIhowd probably encompass "reasonably anticipated" average variable costs, to protect saes made in anticipation that costs wil fall. P. Areeda and D. Turner,. III Antitrust Law 154, 174 (1978). :1 Buffolo Courier E:zpressv. Buffalo Evening News 601 F.2d 48, 55 (2d Gir. 1979);General Foods Corp., supro. Dole 7, TI22,142!:t 22,975; P. Areeda, AntitrustLawSupplement121, 151 157- 58(1982); seep. Areeda and D. Turer, II Antitrust Law 176-78 (1978).

Most court have now concluded that sales at prices equal to or greater than average variable cost should be rebuttably presumed to be legitimate, and that SIiea at prices below average variable cost should be rebuttably presumed to be predatory. E. Rogers Associates, Inc. v. Gardner-Denver Co 718 F.2d 1431, 1437 (6th Cir. 1983), Cfrt. denied 52 D. W. 3886 (U.s. June 12, 1984) (No. 83-1698);Transamerico. Computer Co. v. IBM Corp., 698 2d 1377, 1386 (9th Cir. 1983);Wiliam Inglis Sons Baking Co. ITTContinental Baking Co. 66B 2d 1014,- 1035-6 (9th Cir. 1981),Cfrt. denied 103 8.Ct. 58 (1982);Stlperttlr(. Inc. v- Monsanto Co. 660 F.2d 1275, 1281 (8th Cir. 1981); O.Hummel Co. v. Ferro CrJrp.659 F.2d 340, 352-3 (3d Cir. 1981) (dictum);Northeastern Tel. Co. AT&T Co. 651 F.2d 76, 88 (2d Gir. 1981), Cfrt.denied 455 U.S. 943 (1982);International Air Industries, Inc. American Excelsior Coo, 517 F.2d 714, 724 (5th Cir. 1975),cerf. denied 424 U.S. 943 (1976);see Chillicuthe Sand & Gravel Ca. v- Martin Marietta Corp. 615 F.2d 427, 432 (7th Cir. 1980);Pacific Engineering Prod. Ca. Kerr.McGee Corp. 551 F.2d 790, 797 (loth Cir. cert. denied 434 U.s. 879 (1977). For two good reviews of the relevant cases see Calvani aud Lyuch Predatary Pricing Under the Rubin.rJn-Patman and Sherman Acts: An. Introduction 51 Antitrut L.J. 375 (1982); Hurwitz artd Kovacic,Judicial Analysis of Predatiun: The Emerging Trnrh 35 Vand. L. Rev. 63 (1982).

Arthur S. Langenderfe", 1m:. v. E. Johnson Co. 1984-1 Trade Cas. (CCH) TI65905 (6th Gir. 1984), at 67 865; Barry Wright Corp. v. 11TGrinnell Corp. 724 F.2d 227, 231, 235-6 (1st Cir. 1983); P. Areeda and D. Turner, II Antitrust Law 169-70 (1978).

....

- n - a..

280 Opinion than responses to fluctuating demand and in any event are unlikely to be suffciently long-lived to exclude equally effcient firms. Second, the diffculty of identifying the costs that should be treated as variable in any given situation should not be underestimated. As the Commission pointed out in General Foods Corp. it wil often be diffcult to allocate joint variable production, distribution, or promotional costs among different brands of a given product, among different products, or among different geographic areas.3? Moreover, it wil frequently be diffcult to distinguish the investment or fixed cost component (24) of promotional expenses from the variable cost or current expense component. As the Commission has indicated: Promotional outlays or reduced prices that cause current accounting losses may represent an investment in long-lived information and goodwil that wil payoff with enhanced future revenues. Ifso, the investment component should be amortized over the life cycle for which respondent expecte it to endure. With these considerations in mind, we nevertheless believe that relying upon average variable cost represents the most economically sensible and predictable predation standard available. In her concurring and dissenting opinion, Commissioner Bailey argues that the Commission should instead rely upon industrywide capacity utilization levels to determine whether average variable cost or average total cost should be used to distinguish competitive from predatory pricing. More particularly, she argues that prices that equal or exceed average variable cost may nevertheless be predatory if they fall below average total cost in a market characterized by a high level of capacity utilzation. Under her approach, a plaintiff(or (25) complaint counsel) apparently could shift the burden of justifying sales at prices below average total cost to the defendant or respondent by establishing that the industry involved does not confront substantial excess capacity.

There are a number of diffculties with this approach that would make it less desirable as a predatory pricing standard. First, it might very well deter competitive pricing behavior, because there are a variety of legitimate competitive reasons for prices that fall below average total cost but equal or exceed average variable cost. In particular, prices within this range may represent a perfectly legitimate 37 General Foods Corp., supra note 7, 142 at 22,975; accord, P. Areeda, Antitrust Law Supplement 146 (1982). 3! General Foods Curp. t22 142 at 22,975. Professors Areeda a!'d Turner have argued that all cost.! except capital costs, taxes unaffected by output, and depreciation should he rebuttably presumed to be variable. p, Areeda and D. Turer, In Antitrust Law173 (1978); P. Areeda Antitrust Low Supplement 147 (1982). Although thit approach may be a useful starting point in some respects, we would Eke to emphasize that the investment or fixed cost component of promotional expenses should always be treated as a fixed cost. Of course, as the time period relevant to evaluating an allegedly predatory strategy lengthens more types of costa should be characterized as "variable. In the long run, all costs are varable; at that point, average variable cost and average total cost are therefore equivalent.

Opinion 104 F.

loss-minimizing response to excess capacity at the firm level. As the Ninth Circuit Court of Appeals indicated recently: Pricing below average total cost may be a legitimate means of minimizing losses, particularly when the firm is "temporarily" experiencing "excess capacity" in its productive facilities. . . . Prices below the average total cost of production, but above the average variable cost, may represent a legitimate means of minimizing losses during the period of inadequate demand. Such a price will be suffcient to recover the variable costs of production and at least some portion ufthe firm s fixed coststhose costs that would remain even if(26) the firm ceased production. To discontinue production under these circumstances would increase losses because even that portion of its total fixed costs would be lost.

Moreover, sales at prices within this range are not likely to represent an effective predatory strategy. As long as prices equal or exceed average variable cost, they wil not force an equally effcient firm with comparable variable costs to shut down. In fact, an equally effcient rival wil ordinarily have a strong incentive to continue production, since prices at that level cover at least the variable costs of production and possibly some fixed costs as well. Second, for a variety of reasons, measuring the imprecise construct average total cost" is relatively more diffcult than measuring average variable cost. The accounting measure of average total cost during any particular period will not necessarily reflect the long-term value of the underlying investment. Under conditions of competitive equilibrium, the discounted present value of revenue contributions above variable cost that a given investment will produce over its useful life ("quasi-rents ) will equal its capital cost. However, the accounting value of revenue over cost may vary considerably over time, causing accounting "profits" in some years (when price (27) exceeds the accounting measure of average total cost) and accounting nlosses" in others (when price falls below the accounting measure of average total cost). Moreover, average total costs, as measured by accounting data, include arbitrary measures of depreciation (since there is no correct" depreciation schedule) and measures of capital cost which are both "embedded" (i. reflecting past interest rates and historic costs of capital equipment) and incomplete (not including the opportunity cost of equity). Furthermore, where production takes place jointly with other products, allocating fixed costs among these products must be arbitrary, making the ultimate evaluation of average total costs similarly arbitrary.

Third, relying upon industrywide capacity utilization levels to de- J" William Inglis Sons Baking Co. v fit Continental Baking Co 668 F.2d 1014 , 1035 (9th Cir. 1981),cert. denied 103 S.Ct. 58 (1982) ( ;t:tions ointted). Of cou.rse, as the Court points out, sales at such prices over the long ru "wiu not justify renewal of investment at the previouslevel. " /d.at 1035. However, the rule we have adopted """nt_ fnr nhC'nomenrm because alj costs-jncJuding investment.are variable in tbe long run. INTERNATIONAL TELEPHONE & TELEGRAH CORP., ET AL. 407 280 Opinion termine when the average total cost standard should be used would not be practicable or appropriate. The concept of "capacity" envisions a discontinuity in the production function which does not typically exist, or at best can be measured against rapidly rising marginal cost only with imprecision. As a consequence, any measure of capacity represents only a very imprecise surrogate for actual capacity. Moreover, predatory pricing requires an assessment of the intent of the alleged predator, and industrywde capacity utilization levels are not really relevant to that assessment. Indeed, in many industries, such as electrical power generation, effciency mandates some continuous level of "excess" capacity in order to permit rapid responses-in the form of production levels substantially greater than normal-to seasonal or otherwise cyclical peaks in demand. (28) The fact that an alleged predator may have recently increased its individual capacity is not likely to be any more helpful. There are after all, a variety of legitimate competitive reasons for plant expansion, and if a given expansion reduces costs, concurrent or subsequent price reductions to reach optimal utilization levels may be a perfectly logical and competitive corollary.4o Furthermore, if additions to capacity must be made in large "lumps" because of technological constraints, large expansions wil increase fixed costs so that competitively effcient prices may at times fall below average total cost, even though industrywide capacity utilzation levels may be high. The final diffculty with Commissioner Bailey s approach is that it would reduce business certainty, because businesses would need to determine the level and significance of industrywide capacity utilization before deciding whether to follow the average variable cost or the average total cost standard. As a result, many businesses might simply decide to avoid sellng at prices below average total cost altogeth- , regardless of competitive conditions. In conjunction with the other problems outlined above, this factor confirms the preferability of a (29) standard that focuses upon average variable cost alone, rather than upon capacity utilization levels and average total or fully allocated cost.

The foregoing discussion explains how the Commission s predatory pricing standard fits into the attempted monopolization offense. However, the Commission has taken the position that in attempted monopolization cases, the dangerous probabilty of success element should be evaluated "before proceeding to the other two elements. 42 That tD i.:l. Du.pont tk NemOUTS Co. 96 F. C. 653, 747--8 (1980). 4! It would be particularly inappropriate to rely upon average tow cost to disting1ish competitive from predato ry pricing in the baking indUf try bm:ause-as we have noted supru.-the industry confronted chronic and substatia! excess capacity in most ifnot all relevant geographic markets throu.gllOUt the time period at issue in this case. See pp. 10-12 !Jpra.

4Z General Foods Corp., supmtlote 7, 22, 142 at 22,977; E. 1 Dupont de Nemours Co. 96 F. C. 653, 725-26 (1980).

y, ), Opinion 104 F.

approach was particularly appropriate in General Foods because an accurate evaluation of the extensive price and cost information in that case would have required a very complex and time-consuming evaluation of a variety of relevant data. In the attempted monopolization part (30) of this case, the dangerous probabilty of success element can be evaluated fairly readily. We therefore proceed to discuss that element.

C. Dangerous Probability Of Success Proving the attempted monopolization offense finally requires establishing that the conduct at issue created a dangerous probabilty that the firm involved would acquire "the power to control price or exclude competition" in the relevant market(s).44 This section discusses the legal and economic standards for making such a determination, and then applies them to the facts of this case. 1. Relevant Markets The first step in evaluating the dangerous probabilty element is delineating the relevant product and geographic markets within which monopoly power may be acquired. As the Supreme Court has indicated with respect to product market definition: (31) To establish monopolization or attempt to monopolize a part of trade or commerce under 2 ufthe Sherman Act, (one must) appraise. . exclusionary power. . . in terms of the relevant market for the product involved. Without a definition of that market there is no way to measure (a firm s) ability to lessen or destroy competition. Most courts have taken the same position with respect to geographic 1'3 M our diec\ission of the Clayton Act below indicates, i! showing of predatory pricing is an essntial el!'ment ofa primary line price diBcriminatiOD case, and the treat for predatory pricing under the Shennan Act should apply in Clayton Act cases as well. An evaluation of the price and cost data that complant CO\JJ1.sf:l relied upon in their effort to establish !iabjjjty isreqltred to determine whether Continenta violated the Clayton Act in all five of the relevant geographic markets.

4( General Food GJrp., supra note 7, 1\22 142 at 22,976. (5 In an analogous cont.ext, the Supreme Court has repeatedly indicated that "Determnation of the relevant product and geographic markets is ' a neces.ry predicate' to deciding whether a merger contravelJes the Clayton Act." United Statesv. Marine Bar/corporation 418 U.S. 602, 618 (1974),quoting United Statesv. Dupont Co. 353 U.S. 586, 593 (1957).

Walker Process Equipment, Inc. Y. Food Machinery Chemical Co. 382 U.S. 172, 177 (1965);accord, Super Turf, Inc.v. Monsanto Co. 660 F.2d 1275, 1283 (8th Cir. 1981);Spectrofuge Corp. v. Beckman IfUtruments, Inc, 575 F.2d256 276 (5th Cjr. 1978);UnitedStatesv. Empire Gw; Corp. 537 F.2d 296, 298-99 (8th Cir. 1976), certdenied, 429 U.S. 1122 (1977); Coleman Motor CO. Y. Chryler Corp. 525 F.2d 1338, 1348-9 (3d Cir. 1975);George R. Whitten, Jr., Inc. Padock Pool Bldrs. lnc. 508 2d 5(7, 550 (1st Cir. crrt. denied, 421 U.S. 1004 (1974);Mullisv. ARGO Petroleum Corp.,502 F.2d 290, 295 (7th Cir. 1974); Merit Moturs, Inc. v. Chrysler Corp. 417 F.Supp. at 269-70. Only the Ninth Circuit has suggested that market definition may Dot he a prerequisite to attempted monopolization Jiahi!ty, although its most recent decisions require a showing of a "specific intent to control prices or destroy competition with respect to a part of commerce" it) order to establish attempted monopolization.Compare 17'1nsllmerica Computer CO. V. IBM Corp. 698 F.2d at 1382; William Inglis Sons Baking Co. 117 Continental Baking Co.,668 F.2d at 1027 with Greyhound Computer Corp. V. IBM S59 2d 488, 504 (9th Cir. 1977J, cert. denied 434 U.S. 1040 (1978); Lessigv.Tidewater Oil Co. 327 F.2d 459, 474-75 (9th Cir. 1964),cert. denied 377 U.S. 993 (1964) 280 Opinion 47 (32)market definition.

Reliable measures of supply and demand elasticities provide the most accurate estimates of relevant markets. However, it is ordinarily quite diffcult to measure cross-elasticities of supply and demand accurately.49 Therefore, it is usually necessary to consider other factors that can serve as useful surrogates for cross-elasticity data. In the case of geographic market definition, these factors may include the extent of different price changes and patterns from region to region; the level of barriers to trade flow between regions (including high transportation costs relative to product value); the degree of product shipping from one region to another (i. transshipment); and the perceptions of competition from distant firms on the part of industry members.

In the case of product market definition, these factors may include whether the products and services have suffciently distinctive uses and characteristics; whether industry firms routinely monitor each other s actions and calculate and adjust their own prices (at least in part) on the basis of other firins' prices; the extent to which consumers consider various categories of sellers . . . as substitutes; and whether a sizeable price (33) disparity between the different types of . . . sellers. . . persists over time for equivalent amounts of comparable goods and services. a. Relevant Geographic Market The parties have agreed that the relevant geographic markets in this industry are localized, and the record evidence supports that conclusion. A given bakery can provide fresh bread for only a relatively small area-swept by a radius of one hundred to two hundred miles-and competitive conditions tend to differ from one area to Therefore, the five areas in which Continental's conduct isanother.52 .7 E. , Super Turf. Inc.v. Monsanto Co., supra, 660 f.2d at 1283; Spectrofuge Corp. v. Beckman Irwtruments Inc., 575 F.2d at 276;United Statesv. Empire Gas Carp. 537 F.2d 296, 298-99 (8th Cir. 1976), cert. denied 429 U. 1122 (1977);Caleman Motor Co. v. Chrysler Corp. 525 F.2d 1338, 1348 (3d Cir. 1975);George R. Whitten, Inc. 42J U.S. 1004 (1974);Mullisv. ARCOPadk Pool Builders, Inc. 508 F.2d 547, 550 (1st Cir. 1974), cert. denied Petroleum Carp. 502 Jo'2d 290, 295 (7th Cir. 1974). B Grand Union Co. 3 Trade Reg. Rep. (CCH) TI22,050 (July 18, 1983), at 22 702-3 l102 F. C. 812J; accord, General Foods Corp., supra note 7, TI22,142 at 22 977-78; Beatrice Foods Co. 101 F. C. 733, 829-0 (1983) (Douglas, Corni8fioner, and Miler, Chairman, concurring); FTStatemen.t On Horiiontal Mergers(June, 1982), reprinted in 42 Antitrust & Trade Reg. Rep. Special Supplement (June 17, 1982) (hereinafter cited liSFT' Statement), at (5-15)-S-16).

.9 Grand Union Co. TI22,050 at 703.

50 ld. at 22 703-4; accord, FT Statement, supra note 48, at S-16; Ju/!tice Department Merger Guidelines(June 14, 1984), reprinted in46 Antitrust & Trade Reg. Rep. Special Supplement (June 14, 1984) (hereinaftr cited as DOJGuidelines), at S-.

51 Grand Union CD. TI22 050 at 22 703; acrord, FT Stutement, supra note 48, at 5-15; no.l Guidelines, supra Dote at 5-2, The Justice Departent indicates that the test of the viabilty ofa proapective market should be whether a hypothetical monopolist would find it profitable to impose a "small hut signficant and nontranaitory price increase, ordinarily to be approximated by II five percent price increase lasting oneDOyear.Guidelines supra at (S-2)-(S-).

52 IDF 28; accord, 11TContinental Baking CD. 84 F, C. 1349, 1397 (1974). Continental agrees that the relcvaDt geographic markets are local, but argues that the complaint only aleges a notionwide predatory campaign. However, paragraph 17 of the complaint alleges a plan "to achieve dominance. . in all relevant geographic (footnote cont' Opiaiorr 104 F.

at issue (the metropolitan areas of Denver, the Twin Cities of Minneapolis and St. Paul, Los Angeles-(34)Ventura (covering most of Southern California), and Cleveland, and the larger Northern California/Western Nevada area) should each be treated as a separate geographic market.

b. Relevant Product Market The Administrative Law Judge concluded that the "baking, sale and distribution of white pan bread and bread type rolls by wholesale bakers" should be treated as the relevant product market. ID at 97. However, the relevant product market should also include all white pan bread and bread type rolls produced by the captive bakeries of chain grocery stores. The record does not contain any direct evidence as to the cross-elasticities of demand and supply between white bread products produced by wholesale bakers and white bread products produced by captive bakers. However, on the demand side the record evidence does establish a high degree of competition between these two sources of white bread products. 54 The AU himself pointed out that "As the retail price of captive bread decreases, noncaptive chains " ID at 9. (35)demand lower wholesale prices for private label bread. He also noted that "private, controlled, and secondary label breads (all of which are produced by wholesale bakers) are used by chain stores which do not have their own bakeries to compete at the retail level with the captives." IDF 36. A number of industry witnesses agreed that captive and wholesale bakers compete vigorously with one another and that their respective brands compete vigorously at the retail level 55 and complaint counsel's own expert agreed that captives and wholesalers "compete with each other." Walsh, Tr. 8356; but see Tr. 8360. Furthermore, the AU indicated: Most grocers including chains who have their own bakery wil usually carry a limited quantity of the advertised labels of all the wholesale bakers sellng in a particular marketing area in order to meet consumer demand. market.;" paragraph 18 alltJges an intent to "attain monopolies. . . in one or more relevant markets;" and paragraph HJ(h) alleges that Continental's practices wil "riJncrease the probability that respondents wil attain a monopoly in the wholesale baking industry in each and aU relevant geographic markets. 03 The Northern California/Western Kevada market, consisting of California north of Bakersfield and south of Eureka, and the weswro Nevada communities of Rena, Carson City, and Lake Tahoe is considerably larger than a single metropolitan area. However, prices throughout the region appear to be relatively unform, and the region encompasses the service areas of a number of wholesale bakers who ship product. throughout the region. Continenta' s counsel stipulated the area as a relevant geographic market. CPF 12-14 through 12-21. The principal issue with respect to demand is whether a small change in the price of captive label white bread would induce asigJificant and like-signed change in the quantity of wholesale baker white bread that is demanded, and vice-versa. FT Statement, supra note 48, at 3-15;Justice Department Merger Guidelines, supra, note 50, at (S-2).

5, Murray, Tr. 8858; Vander Giessen, Tr. 10954-57; Schaus, Tr. 11225-28; Metz, Tr. 10208-14; Nissen, Tr. 10138-9; Jakacki, Tr. 10271-73.

(, . ..

INTERNATIONAL TELEPHONE 8. TELEGRAPH CORP., ET AL. 411 280 Opinion IDF 34. In short, captive and wholesaler bread products compete vigorously at both the wholesale and the retail level. On the supply side, the principal issue is whether a small change in the price of wholesale baker white bread wil induce a significant and opposite change in the quantity of white bread that captive bakers supply, and vice-versa.56 That is clearly the case here. Captive bakers could readily divert production to retail grocers in response to an increase in wholesale baker (36) prices. Moreover, some! retail chains could switch from purchasing white bread from wholesale bakers to producing it themselves, in their own captive bakeries. When considered together, these factors all suggest that the relevant product market should include white bread products produced by both wholesale and captive bakers. That would be consistent with the 1984 Justice Department Merger Guidelines which expressly conclude that captive production should be included in the relevant product market in merger cases when a "small but significant and nontransitory" price increase is likely to induce vertically integrated firms to increase production ofthe relevant product, either for outside sales or to increase their own downstream sales. 57 It would also be consistent with previous cases in which the Commission itself has recognized that wholesale and captive bakers compete with one another.

2. Likelihood That Continental Would Acquire Monopoly Power Having delineated the relevant product and geographic markets the next step is to determine whether the conduct at issue created a dangerous probability that Continental would acquire monopoly power within those markets. A successful predatory strategy depends on the following scenario: once competition has been injured, the predator wil be able to raise (37) prices to supracompetitive levels long enough to recoup losses incurred during the predatory period and to earn greater overall profis than would have been possible from pursuing a competitive strategy.59 A number of factors-including in particular the market power of the prospective predator, and the height of barriers to entry-affect the degree to which a given course of predatory action can be expected to result in successful monopolization.

Of FT Sta.tement, supra note 48, at 8-15; DOJ Guidelines, supra. note 50, at 8-. 51 DOJ Guidelines, supra note 50, at 8-.

Bakers of Wa. hingtDn, Inc. 64 F. C. 1079, 1124-25 (1964),affd sub nom. SafewayStores, lne. f! al. v. FT, 366 F.2d 795 (9th Cir. 1966),ccrt. denied 386 US. 932 (1967); See Flowers Industries, Inc. Docket No- 9148 (1980) (complaint). (102 F. C. 1700 (1983)j 59 Joskow and Klevorick A Framework For Ana.lyzing P..eda/oryPricing Policy, 89 Yale L. J. 213, 219-20 (1979).

GO The tenns "market power" and "monopoly power" are oftn treated as synonymous from an economic perspective. However, the term "market power" is used here to describe Ii whole continuwn along which the power hA"'Il!1iog with the complete absence of market power at one end and emting with Opinion 104 F.

Relevant determinants of the market power of a prospective predator in this regard include its absolute and relative market shares, and those of competing firms; "the strength and capacity of current competitors; the potential for entry; the historic intensity of competition; and the impact of the legal or natural environment."61 A firm with a large absolute share of sales in a given market wil ceteris paribus find it easier to execute a successful predatory strategy than a smaller firm. Most courts have determined that market share ranging from forty percent to (38) sixty percent prior to the commencement of a predatory strategy ordinarily must be established in order to prove the requisite dangerous probability of successful monopolization. However, as the Supreme Court has recognizd, focusing exclusively upon absolute market share percentages can produce a distorted view of actual market power:

Obviously no magic inheres in numbers; "the relevant effect of percentage command of a market varies with the setting in which that factor is placed," . . . obviously, if a producer controlling an even lesser share than (the 40% share the defendant in the cas controlled) is ringed by numerous smaller satellites together accounting for (39) the rest, his mastery of the market is greater than were he facing fierce rivalry of other large sellers.

In this case, the record evidence indicates that it is highly unlikely that Continental could have acquired monopoly power in any of the relevant markets. As noted above, absolute and relative market share data can provide important evidence on this issue. The record does contain some useful market share data for three of the five relevant geographic markets (Denver, the Twin Cities, and Northern California/Western Nevada), and some less useful market share data for a fourth geographic market (Cleveland). These data are summarized below: (40) 6! General Fauds C""p., supra note 7 142 at 22 976. 62 E. , Times-Picayu.ne Pu.b/i.hing Co.v. United Stotes 345 U.S. 594, 612-13 (i953) and n.33 (40%); Nifty Foods Corp. v. Great At/untie Pacific Tea Co. 614 F.2d 832, 841 (2d Cir. 1980) (a share of 54.5% not sufcient when share fell to 33% withill five years);United Statesv. Empire Ga Co., 537 F.2d 296, 305 (8th Cir. 1976) (50%liot necessrily sufcient, pa.rtcwarJy in light ofunreliabilty of data),ccrt. rknied 429 U.S. 1122 (1977);Kearney & Treckcr Corp.v. Giddings Lewis, /nc. 452 F.2d 579, 598 (7th Cir. 1971) (33%),ccrt. denied 405 U.S. 1066 (1972); uktro- Vend Corp. V. Venda Co. 500 F.Supp. 332, 356 (N.D. Il 1980) (not even 50% market shar in itslf suffcient), affd,660 2d 255 (7th Cir. 1981),cert. denied 102 S.Ct. 1277 (1982);Outboard Marine Corp. V. Pczetel 461 F.Supp. 384, 410 (D. Del. 1978) (35%);W.L. Gore Associalesv. Carlisle Corp. 381 Supp. 680, 704 (D. Del. 1974) (40-0%), o.frd in part and rev'd in part on other grounds,529 F.2d 614 (3d Cir- 1976);Bowl America Inc. V. Fair Lanes, Inc., 299 F.Sltpp. 1080, 1091, 1094-5 (D. Md. 1969) (in two markets: 15.5%; 33%); Diamond International Corp.v. Wolterhoefer 289 F.Supp. 550, 578 (D. Mrl 1968) (share of 51 % insufcient); General FlJds Corp., supronotc, TI22, 142 at 22 976; E./. du Pontrk Nemours CO. 96 C. 653, 725-26 0.. 16 (1980) (30% might in some circltmstances be sufcient), 6.1 Time. Picayune Publishing CO. V. United State$ 345 U.s. 594, 612-13 and n.33 (1953),qu.oting United States v. Columbia Steel Co. 334 U.S. 495, 528 (1948); o.r:rd, American TobQCCO Co. v. United States,328 U.S. 781, 796 (1946); Pacific Cot Agricultum! Export Association V. Sunkist Growers, Inc. 526 F.2d 1196, 1204-5 (9th Cir. 1975), cert. denied, 425 U.S. 959 (1976);Som United Statesv. Grinnel Corp. 384 U.S. 563, 571 (1966). .._ , , INRNATIONAL TELEPHONE & TELEGRAPH CORP.

280 Opinion- Sales of White Bread Products By Wholesale Bakers and Captive Bakers In Four Markets Denver: 196364 Twin Cities: 196655 Firm Share Firm Share 1. King Soopers 50. 1. Red Owl and 30. and Safeway National Tea Captive Bakeries Captive Bakeries 2. In-Store Bakeries 20.

of Other Firms 2. Continental 17. 3. American 15. 3. Old Homestead 14. 4. Continental 15. 4. Campbell-Taggart 5. Zinsmasters 10. 5. Interstate 6. Pan 0 Gold 7. Creamy Crust 5% (41) N, Cal.W. Nev.: 197266 Cleveland: 197167 Firm Share Firm Share 1, Campbell. T a9gart 34. 1. Pick N Pay 50. and Fisher Fazio 2. Continental 25. 2. Interstate 3. American 21. 3. Ward 4. Safeway and Lucky 15.0%. 4. American 5. Inglis 5. Continental The foregoing data indicate that Continental accounted for less than twenty percent of total white bread sales in the Denver and Twin Cities markets, and considerably less than thirty (42) percent of total white bread sales in the Northern California/Western Nevada market, at the respective points in time at which it began its allegedly predatory campaigns in each of those markets.6 Those shares are 64 IDF 60, 61. The market share for the captive bakeries of Safe way and King Soopsr" is a minimum, since the ALJ indicate that "finn volwne of white bread sales, the bakery operations of King Soopers and Safeway were larger than the combined volume ofalllhe wholesale baker " IDF 61 (citation omitted). Therefore, the wholesale baker market shares the ALJ cites must be reducsd by at least fifty percent. 6S IDF 111, 112. Since captive and in-store bakeries accounted for fifty percent of total white bread sales, the wholesale baker market shares the ALJ cites must be reduced by fifty percent. 66 IDF 240, 242. These data are very approximate. The ALJ found that Safeway and Lucky together accounted for 17.8% of total white bread sales in the San Francisco/Oakland area and 11. 7% of tota white bread &ties in the Sacramento area. IDF 242. Presumably, thsse figures are for their captive bakeries only, since wholesaler sales are listed separately. Since the first area is considerably larger than the second, 15% may he a roughly accurate estimate of Safe way s and Lucky s combined sales throughout the Northern Calforna market. The share data for wholesalers that the ALJ cites should therelore be reduced by at least fiften percent, and the Chart reflect,q that corredion. However, the wholesaer shares probaWy should be reduced even furher to account for the captive bakery sales of Alpha Beta, another large retaij chain. Unfortunately, the ALJ was apparently not able to sciu.r market share data for Alpha Beta.

61 IDF 296, 297. The ALJ indicates that among wholesalers, Continental was tied for third with American behind Interstate and Ward- IDF 296. Since the captive bakeries of Pick N Pay and Fischer Fazio alone accounted for fifty percent oHotal white bread sales to grocery stores and Kroger, A&P, and Lawsonsalso operated their own captive bakeries, Continental's share must not have been very large. Ofcou., it probably increased substahtially when Continental became Pick N Pay s supplier of private label white bread. In 1980, when both wholesaler and captive bakers are included, Continenta ra!)ked second. stil behind Interstate but ahead of Fisher-Fazio. IDF 342. M By 1967, Continental had increased its share of total white bread saes in the Denver market to 25 6 percent at most. See IDF 88. This figue Bsgumes that captive bakers stil accounted for at least fifty percent of tota whit! QC"mntion seems reasonable, because wholesaler while bread dollar sales did not increase a Opinion 104 F.

considerably below the forty percent absolute market share that the courts have generally concluded must be shown ceteris paribus before a dangerous probability of success can be established. That conclusion is strengthened when Continental's share levels are considered in relative terms. In the Denver and Twin Cities markets Continental confronted very strong competition from substantially larger captive bakery operations, as well as strong competition from one or more wholesalers of comparable size in each market. In the Northern California/Western Nevada market, Continental confronted strong competition from a substantially larger wholesaler, a wholesaler of comparable size, and three large retail chains with significant captive bakeries.

The foregoing data also indicate that Continental accounted for a significantly smaller absolute share of sales in the Cleveland market (at least before Pick N Pay closed its captive bakery and began buying private label bread from Continental). As a consequence, it seems likely that Continental's absolute share of sales was considerably below the forty percent absolute market share that the courts have generally concluded must be (43) shown ceteris paribus, before a dangerous probability of success can be established. Moreover, Continental confronted strong competition from two larger wholesalers and one wholesaler of comparable size.

The record apparently does not contain any evidence with which Continental's approximate share of total white bread sales in the Southern California market can be calculated. However, the ALJ did conclude that in 1974, after its allegedly predatory conduct had end- , Continental and Interstate each accounted for about forty percent of wholesaler white bread sales. IDF 225. These absolute shares would in all likelihood be reduced considerably if the white bread sales of captive bakeries-including those of Safe way, Ralph' , Yon, Lucky, Alpha Beta, Albertson, and Certified, all large retail grocery chains -were included. See Vander Giessen, Tr. 10954; IDF 189. Moreover it is important to remember that forty percent is the share of wholesale baker sales that Continental secured after its allegedly predatory conduct, and it is no larger than Interstate s share of sales; that ,ardly represents successful monopolization. Therefore, it is highly mlikely that Continental's actual share of total white bread sales, ,efore or after its allegedly predatory conduct, satisfied the standard at most courts have endorsed.

As we noted earlier, other industry characteristics may help deterine whether a given absolute and relative share of the relevant arket creates a dangerous probability of monopolization. The abnce of substantial entry barriers, (44) ceteris paribus, is particularly ely to eliminate that probability. Barriers to entry must be sub- INTERNATIONAL TELEPHONE & TELEGRAPH CORP., ET AL. 415 280 Opinion stantial for a predatory strategy to succeed. Otherwise, when the predator attempts to raise prices to supracompetitive levels after the predation period, new firms wil enter and/or terminated firms will reenter and force the predator to lower its prices to competitive Ievels.69 The Commission has defined entry barriers as "substantial non-recurring outlays" that raise the entry costs of potential entrants suffciently high to permit incumbent firms to restrict output and raise prices for a significant period of time.7o Governmental restrictions may also create partial or absolute barriers to entry. The record evidence indicates that barriers to entry into the bread baking industry are not particularly high. Three potential barriers to entry have been suggested in this industry. First, complaint counsel argue that the diffculty of acquiring retail shelf space for new bread products makes entry diffcult. CAB at 21. This does not appear likely, because it assumes that retail grocers cannot easily adjust the size of their bakery sections to include promising new brands, and that any given prospective entrant wil not be able to secure shelf space in any retail grocery outlets. The record does not contain any significant evidence to support these assumptions. (45) Second, the ALJ has suggested that making expenditures associated with developing a franchise for advertised label bread represents a barrier to entry. It is true that product differentiation may in some circumstances represent such a barrier. However, as noted above, the Commission has determined that product differentiation should be treated as a barrier to entry only if it makes new entry contingent upon non-recurring outlays substantial enough to permit incumbent firms to restrict output and raise prices for a significant period of time.71 In this case, that condition is not satisfied because private label and captive label breads-which do not ordinarily require more than minimal promotional support-have been and continue to be both successful and a significant constraint upon advertised label bread sales.

Third, the ALJ has suggested that "the propensity of the large multi-plant companies to engage in below cost pricing" represents a barrier to entry. ID at 92. However, there is little in the record to suggest that multiplant economies of scale, or any other technological factors, for that matter, deter entry. See, e. Jakacki, Tr. 10282. In any event, allegations of predatory pricing represent the central issue in this case, and we conclude in Part IILD. below that Continental's pricing behavior did not violate the Commission s predatory pricing standard. (46) 69 Joskow and Klevorick, supra note 59, at 227. )Q General Food Corp. , supranote 142at22 981 citing&. Posner Antitrust Law:An EWllomic Perspectiu€ 59 (1976). Relevant entry ('o t.- may include production, distrihutional and promotional expenses. I "'''ods Corp., supronote, TI22 142 at 22 981. Opinion 104 F.

The market share data developed above, in conjunction with the absence of any countervailing industry characteristics, persuade us that even if Continental had made sales at prices below average variable cost in some areas for a significant period of time, such pricing behavior could not have created a dangerous probability that Continental would acquire monopoly power in any ofthe relevant markets. Since establishing a dangerous probability of success is an essential element ofthe attempted monopolization offense, the Commission has therefore determined to reverse the ALJ and dismiss that portion of the complaint. (47) II. SECTION 2(a) OF THE CLAYTON ACT Section 2(a) of the Clayton Act, as amended by the Robinson-Patman Act, provides That it shall be unlawful for any person engaged in commerce, in the course of such commerce, either directly or indirectly, to discriminate in price between different purchasers of commodities of like grade and quality, where either or any of the purchases involved in such discrimination are in commerce, where such commodities are sold for use, consumption, or resale within the United States or any Territory thereof or the District of Columbia or any insular possession or other place under the jurisdiction of the United States, and where the efiect of such discrimination may be substantially to lessen competition or tend to create a monopoly in any line of commerce, or to injure, destroy, or prevent competition with any person who either grants or knowingly receives the benefit of such discrimination, or with customers of either of them. In this case, Count II of the complaint alleges that Continental discriminated in price, directly or indirectly, between different purchasers of bread, by sellng bread oflike grade and quality to some of such purchasrs at substantially higher prices than to other of such purchasers.

This allegation has evolved into two more specific allegations: (1) that Continental sold its advertised label bread, secondary label bread, and/ or private label bread at different prices to some purchasers than to others in all five relevant geographic (48) markets; and (2) that Continental sold its private label hot dog and hamburger buns at higher prices to some purchasers than to others in the Cleveland market. This part of the opinion evaluates these allegations. 7Z 15 U. C. 13(8). Section 2(a) also includes a CQlt justification defense, and several other miscellaneous provi. ons.

In General Food Corp. the Commission concluded that it did !lot need to determine whether General Foods nduct satisfied the jurisdictional and predatory pricing components of a Section 2(a) violation because it was nahle to find any prospect of injury to competition from the events described in ltheJ record"Genera.l Food 'rp. , supra note 7 22,142 at 22 988 (emphasis added). The Commission detennined that the relevant markets saue were larger than the sales districts that complaint counsel had alleged, and that because complaint counsel (footnot... ""- INRNATIONAL TELEPHONE & TELEGRAPH CORP., ET AL. 417 280 Opinion A. Preliminary Jurisdictional Issues Like other parts of the Clayton Act, Section 2(a) includes a variety of jurisdictional requirements that must be satisfied before the substantive prohibitions of the subsection apply. One commentator has summarized these jurisdictional requirements in the following fashion; (49) In order to bring the substantive portions of the Act into play, there must'be (1) two or more consummated sales, (2) reasonably close in point of time, (3) of commodities, (4) oflike grade and quality, (5) with a difference in price, (6) by the same seller, (7) to two or more different purchasers, (8) for use, consumption, or resale within the United States or any territory thereof,. (9) which may result in competitive injury. Further more, (10) the "commerce" requirement must be satisfied. All ten of these jurisdictional elements must be met in order to invoke the power of the Federal Trade Commission or the court to consider the lawfulness of pricing transactions. Seven of these requirements-all except the "like grade and quality, commerce, and competitive injury requirements-are not in dispute and the evidence supports concluding that they have been satisfied in each of the five relevant geographic markets. In each of these markets, Continental made at least two sales-reasonably close in timeof commodities at different prices to two or more different purchasers for use, consumption, or resale within the United States." At certain times from 1964 through 1969 in the Denver market, Continental sold private label bread from its Denver bakery (under the "Tender Crust" label) to Associated Grocers at-prices that were typically between one cent and three cents lower per one-pound loaf than the prices it charged other purchasers for its advertised label Wonder Bread (50) products. 5 At certain times from 1966 through 1968 in the Twin Cities market, Continental sold Wonder Bread, secondary label bread (under the "Wonder Country Style" label) and private label bread from its Minneapolis bakery at different prices to different purchashad not alleged a.y Jarger markets the Commission was "colltrained on review from finding that respondeot attempted to monopoli:ze some larger market. Id at 22 982 and n.56. Hence, even if General Foods had controlled one hundred percent of saes in one of its Baes distrclB, the Commisson concluded that it would not have posssd monopoly power, becaus firms outaide the district would have constrained that power.Id at 22 983. Moreover the Commission found that competition in the ground coffee industry was "healthy and viualy invurnerabJe to the asuJta of any one firm. . . .n and concluded: "Healthy competition does not violate the Rohinsorl-Patman Act. Id. at 22,98S9. In short, the Comm!lon concluded that there was noreasonable possibilty that price differences charged by General Foods could injure competition. Id. at 22,988, citing Falls City Industries, Inr:v. Vanca Beverage, Inc. 103 S.Ct. 1282, 1288 (1983). 11' this ca, by contrast, the reJevantproduct and geographic markets are well defined, and the reord evidence does not permt WI to conclude that competition within! these markets is so vigoroWl that no reasnable possibilty of har to competition from discriminatory pricing could be esta lished.

71 E. Kintner A Robinson.-Patman Primer35 (2d ed. 1979); accord, L. Sullvan, Handbook of the Law of Antitrust 679-(1977).

"16 ID at 82; IDF 66, 67, 83. Continental' s contracts with Assodated Grocers durng this period required its Tender Crst wholesae price to be at least one cent lower than its Wonder Bread whoJe.se price. IDF 78. :

Opinion 104 F.

ers.76 At certain times from 1970 through 1973 in the Southern California market, Continental sold private label bread to several retail grocery chains at prices lower than those it charged other purchasers for Wonder Bread. IDF 217. At certain times between 1972 and 1974 in the Northern California market, Continental sold private label bread from its San Francisco Bay Area bakeries to several retail grocery chains, at lower prices than it charged other purchasers for Wonder Bread.77 Finally, during 1973 and 1974 in the Cleveland market, Continental sold private label bread from its Akron bakery to Pick N Pay (hereinafter "PNP"), a large local retail grocery chain, at prices that were lower than those it charged other purchasers for Wonder Bread from its Akron bakery. IDF 311, 328, 329, 338. At certain times during that period, it also charged PNP lower prices for private label hot dog and hamburger buns than it charged other purchasers. IDF 311, 332, 339, 353, 356. (51) As the foregoing analysis suggests, Continental charged different prices for private label and advertised label bread products in each of the five relevant geographic markets, and its conduct consequently s The threesatisfies seven of the ten jurisdictional requirements. disputed jurisdictional issues are discussed below. B. The Like Grade and Quality Requirement One might reasonably argue that private label and advertised label bread products are not "commodities of like grade and quality" because the advertised label bread products consistently sell at a higher price than their physically identical private label counterparts. See IDF 36. That persistent price differential indicates that consumers perceive some differences in quality between advertised label and private label bread products, suggesting in turn that the cross-elasticity of demand between the two may not be extremely high. Moreover although the cross-elasticity of supply between private label and advertised label bread is high at the production level, it is probably considerably lower at the distribution level because the sale of advertised label bread typically requires substantial (52) promotional efforts not needed to sell private label bread. However, the Supreme IDF 106, 118, 120, 121 , 125, 128-130.

'1 IOF 231, 233, 249, 258, 263, 272-278. These data exclude November, 1972-av.uary, 1973, whet! Continen:ta' bakeries were dosed by a labor etrike. IDF 260; CPF 12-79. 76 Continental argues that in primary Jine cases "the lower price must be shown to have been subsidized, made possible, by the higher price. " RAB at 30. However, the Supreme Cour has indicated that a price discrimination is merely Ii price difference " without incorporating Ii subsidization requirement.FrCv. Anheuser-Rusch, Inc., 363 U.S. 536, 549 (1960); accord, William Inglis Sons Baking Co. v. 11' Continental Baking Co., 668 F.2d at 1040. 79 This does not mean that private Jabel and advertsed label bread should not be trate as part of the same relevant product market, for purposes afthe Shermap Act. The filet that two items command different prices does not nece8rily establish that they are different product.. The key issue is whether crosslasticities of demand and supply are sufciently high to make the prices ofthe tWQ items substatialy interdependent. The record evidence indicates that cro lasticitie!: of demand and supply between advertsed label and private label bread products are sufciently high to make them part of the sae product market. Indeed, the AU pointe out. (footn"t-",.ng,g, ), ), ._, ., . 280 Opinion Court has determined that physically identical private label and advertised label products-regardless of distributional differencesshould be treated as commodities of "like grade and quality" under the Clayton Act. In FTCv. Borden the Court rejected Borden s claim that private label and Borden-brand evaporated milk-Dtherwise indistinguishable-should be treated as different "grades" because the Borden brand regularly sold at a higher price than the private label (53) brand. The Court determined that since the two products were physically and chemically identical " they were of like grage and quality, despite their persistently different prices.S! The Commission has therefore determined to treat advertised label and private label bread products as commodities of "like grade and quality. C. The Commerce Requirement The commerce requirement can be broken down into three distinct segments. First, the !!person" involved must be "engaged in commerce." Second, the price discrimination alleged must occur in the course of such commerce.82 Third, at least one of the sales that create the alleged price discrimination at issue must be "in interstate commerce,"83 although it does not matter whether the sale at the higher price or the sale at the lower (54) price moves across a state line. B' If the third requirement can be satisfied, then the first two requirements must necessarily also be satisfied.

The courts have interpreted the commerce requirement differently in primary line and secondary line Section 2(a) cases. In primary line cases, it is suffcient to establish that the seller made sales at different prices in two or more states, even though the areas in which the discriminatory sales were made are not part of the same geographic market. In Moorev. Mead' s Fine Bread, the Supreme Court noted that the defendant sold bread it manufactured in New Mexico to A large retail price spread between captive and private label bread, on the one hand, and the advertised label bread, on the other, will result in a loss of sales of advertised label bread. If this price spread is reduced the captive and private label products wil lose sales and there wil be an increase in saes of advertised label brelld. IDB' 36; accord NiBSen, Tr. 10139; Metz, Tr. 10239-40. In short, as the price differential increa es or declines, the quantity of private label bread products demanded wiJ respectively increase or decline. However, it is important to recognize that advertised label bread and privllte label bread arcidenticalnot products; they are differentiated in the minds of consumers to a significant degree. 80 FTCv. Burden Co. 383 U.s. 637, 640, 645-6 (1966). IcLat 638, 640. Real difference in quality may ofcour place products in different "grades. Icl at 644 n. H1.Standard Oil Ct!. v. FTC 340 U.S. 231 , 236-8 (1951). fj Cul(Oil Corp. v. CoppPaving Co. 419 U.s 186, 195 (1974). By contrast, the Federal Trade Commission Act applies to unfair methods of competition " in Or u((ectin/:commerce: 15 U. C. 45(a)(1) (emphasis added). Moorev. Mead's Fine Bread Co., 348 U.S. 115, 119 (1955); William Inglis Sons Raking Co. v. I1TContinental Baking CO. 668 2d 1014, 1043-4 (9th Cir. cat. denied 103 S.Ct. 57 (1982). We do not need to determine whether de minimis interstate sales are suffcient to satisfy this jursdictional requirement because Continental' s sales at different prices were significantin each of the relevant markets. Compare William Inglis Sons Baking Co. lit Continental Baking Co. 668 F.2d at 1011 and n.54 with Food Rasket, Inc. v. Albertson s Inc. 383 F.2d 785 788 (10th Cir. 1967).

8. See, e. , Lehrman v. Gul(Oil Corp. 464 F.2d 26, 36-7 (5th Cir. cert. denied 409 U.S. 1077 (1972); iquilux Ga Serus., Inc. v. Tropical Gas Co. 303 F.supp. 114, 416-17 (D. 1'. R. 1969). ), Opinion 104 F.

customers in Texas at a price higher than the price it charged to certain customers in New Mexico.s6 As a consequence, the Court determined that the primary line discrimination alleged in the complaint satisfied the (55) commerce requirement of Section 2(a),6 A number of courts have more recently taken the position, however that the discriminatory sales at issue must originate from the same plant. In Willard Dairyv. National Dairy, the Sixth Circuit concluded that Section 2(a) did not apply to allegedly predatory sales in Willard, Ohio by an interstate defendant because the plant that the defendant used to supply the Wilard area was used only for intrastate sales; the defendant did not sell wholesale milk from that plant in any other state.6S The court pointed out that the fact that the defendant also made interstate shipments from otherthan its Shelby, Ohio plant to areas in which the plaintiff did not engage in business is immaterial to the issue in the case.89 (56) These cases establish that in order to satisfy the commerce requirement in a primary line case, the defendant that makes the allegedly discriminatory sales at issue in one state must also make sales of the same product from the same plant in other states. In this case, the commerce requirement is satisfied for primary line purposes-in each of the relevant geographic markets. In the Denver market, Continental's Denver bakery produced both the private label bread that it sold to Associated Grocers (with members in five different states) and the advertised label Wonder Bread that it sold to retailers in Colorado, Wyoming, and Nebraska. IDF 53-55, 57. In the Twin Cities market, Continental's Minneapolis bakery produced both the private label bread sold to three large chain stores and the Wonder Bread and Wonder Country Style bread sold to other retailers in Minnesota and Wisconsin. IDF 103, 106. It also supplied the Continental bakery in Rochester with bakery products that were in turn shipped to other Minnesota and Wisconsin communities. IDF 103-104. In the Southern California market, Continental's Beverly Hils and DiCarlo (San Pedro) bakeries produced both the private 86 Moorev. Mea.d's Fine Bread Co. 348 U.S. at 116--17. R7 Respondent argues that the commerce requirement cannot be satisfied because the discriminatory saes allegedly injured competition only within the relevant geographic market. . RAP at 31. However, that is not the standard the Supreme Court has adopted in primary line cases. It is enough simply to show that some ofthe sales from a given plant were made at different prices in different states RR Willard Dairy Corp-v. National Dairy Product$ Corp. 309 F.2d 943, 946 (6th Cir, 1962),cert, denied 373 D. 934 (1963).

39 Id.; accurd, Borden Co. v. FTC 339 F.2d 953, 955 (7th Cir. 1964);Beatrice Foods Co. 76 F. C. 719, 822 (1969), af(d sub nom- Kroger Co. v- FTG 438 F.2d 1372 (6th Cir. cer!. denied, 404 U.S. 871 (1971). 90 This suggests that an inter t.te firm may be able to avoid the primary line proscriptions of Section 2(a) by using a local plant that sells its products solely intrastate. Kintner supra note 74, at 95;see Bacon v. Texaco, Inc., 503 F.2d 946, 948-9 (5th Cir. 1974)(per curiam), cert. denied 420 U.S. 1005 (1975);Kanev. Martin Paint Stores T_.. IQ7A'J 'T ,-D r,, Irrl-T\ U7!i 9f\ IS n NY, 1974\. "t. 97- 914-15. ), 280 Opinion label bread sold to large retail chains and-the advertised label bread. sold to other retailers, including (57) bread that was sold and delivered-through Continental's San Diego bakery-to customers in Arizona. IDF 177, 179, 201, 217. In the Northern California/Western Nevada market, Continental' s San Francisco Bay Area bakeries produced both the private label bread sold to several chain grocery stores (some with stores in both California and Nevada) and the advertised label bread sold to other retailers. 91 Finally, in the Cleveland market Continental's Akron bakery produced both the private label bread sold to PNP in the Cleveland area, and the Wonder Bread and VI onder Country Style Bread sold to customers in both Ohio and (58) Pennsylvania.92 As a consequence, the commerce requirement for primary line jurisdiction is satisfied in all five relevant geographic markets. (59) The commerce requirement for secondary line cases is more restrictive. In Mayer Pauingv. General Dynamics the Seventh Circuit concluded that in determining whether "either or any" discriminatory sale is in commerce, only the purchases of the allegedly injured purchaser and its competitors can be considered. The Court pointed out that to the extent customers on different sides of a boundary line do not compete with each other, no adverse competitive effects on the customer level can ensue from the supplis price variations.

The Court distinguished this conclusion from the Supreme Court ap- 9L IDF 229, 231-33;uccard, William Inglis Sons Baking Co. v. lit C(JT!tinentul Baking Co.,668 F.2d at 1044. Continental operated bakeries in San Francisco, Oakland and Sacramento. However, their output is considered collectively OOulle each bakery supplied bread producl to-and transshipped hread products through--ach of the others. IDF 231 921DF 285-287, 347; CPF 13-126, 13-143 through 13-145, 13-147, 13-148. Although Continental did not ship Wonder Bread directly from the Akron bakery to retailers in other states, the Akron bakery supplied the Y Ollgstown bakery with Wonder bread on a daily basis, and YOllgstown itJ turn sold that bread daily to retailern in both Ohio and Pennsylvana. As long as the practical economic continuity of the interstate sales are unbTOken that is, the goods are shipped in s completely unchanged state, with no hiatus for repacking or warehousing-sales such as theBe satisfy the "ineomrerce" reqwrement. Compare Belli. tonv. Texaco, Inc. 455 F.2d 175, 178-0 (loth Cir. (interntate shipment of crude oj) not sufcient to place sales of gasoline refined from crude oil in interstate commerce), cert. denied 408 U.S. 928 (1972); MT., Inc. v. House of Sohel, 197!J2 Trade Cas. 817 (N.D. Cal. 1979) ("practical economic continuity" not maintained where defendants ordered liquor for gei1eral inventory, rather than to meet the needs or ordern of specific customers)with Foremost Dairies, Inc.v. prc, 348 F.2d 674, 677-78 (5th Cir. cert. denied 382 S. 959 (1965) (milk passed in "steady flow" from Colorado fsnns through Santa Fe processing phmt to retail grocery stores in Albuquerque). I am indebted to Commissioner Bailey for this point. 93 The Commission does pot accept the ALJ'.' conclusion that all of Continental' s sales-including the allegedly discrimirmtory sales--are in commerce because all local grcery chains negotiate their bread purchases with a corporation located in New York." ID at 80. As the Commssion has indicated previously, "Interstate negotiation . . . alone is insufcient to fulll the commerce requirement of the Rohinson-Patman Act. . Beatrice FO(JrLCo. 76 F. C- at 822;accord, Borden Co. v. J.TC 339 F.2d 953, 955 (7th Cir- 1964). gi Mayer Paving Asphalt CD. v. General Dynamics Corp. 486 F.2d 763, 769 (7th Cir. 1973),cart. denied 414 U.s. 1146 (1974);accord, McGrJtfinv. Sun Oil Cv., 539 2d 1245, 1248 (10th Cir. 1976);Myersv. Shell Oil CD., 96 Supp. 670, 675-76 (S.D. Cal. 1951);Beatrice Foods Co., 76 F, C- at 822. Mayer Paving Asphalt Co. v- General Dynamics Corp. 486 F.2d at 770 quotingF. Rowe Price Discrimination Under the Robinson-Patman Act 179 (1962). Opinion 104 F.

proach in primary-line cases by noting that injury to competition among sellers may arise if the discriminating seller can subsidize its lower prices for product from a give.n plant in one state with higher prices for product from that same plant in other states. By contrast injury to competition among buyers may arise only if the favored and disfavored buyer(s) compete with one another in a relevant geographic market. In short, secondary line injury cannot be establishedbecause the jurisdictional commerce requirement cannot be satisfied in any relevant secondary line geographic market that does not cross (60) state lines, because by definition favored buyers in that market do not compete with disfavored buyers in other markets. In this case, allegations of secondary line violations of Section 2(a) extend only to the Northern California/Western Nevada market and the Cleveland market. Of course, the relevant geographic markets for secondary line retail grocery sales are considerably smaller than the relevant geographic markets for primary line wholesale and captive bakery sales. The Commission has used Standard Metropolitan Statistical Areas (SMSAs) as a good approximation of relevant geographic markets in a number of recent retail grocery store merger cases.98 At least one retail grocery market in the Northern California/Western Nevada area satisfies the commerce requirement, because disfavored retailers in Reno and Sparks, Nevada that purchased Wonder bread products from Continental arguably competed with favored retailers in California that purchased private label bread products from (61) Continental.99 However, the second market, the Cleveland retail grocery market, does not satisfy the secondary-line commerce requirement, because it does not include any areas outside Ohio. The Cleveland SMSA consists of Cuyahoga, Lake, Medina and Geauga Counties, all of which lie within the confines of the state of Ohio. CPF 13-32. Moreover, the market area of PNP, the retail grocer that allegedly received lower prices, apparently does not extend beyond greater Cleveland, so that PNP does not compete with any retailers located outside Ohio. RRB at (62) 28; see CX 992-E. Therefore, the allegation of secondary line discrimination in Cleveland must be dismissed because it does not satisfy the "in commerce" jurisdictional 90 lei 97 Falls City v- Vaneo is not inconsistent with this conclusion- There, the District Court had concluded that although the two buyers involved could not sell to the same retailers, they nevertheloo98 competed with one another becaus the relevant geographic market at retail induded both the Indiana county and the Ke1Jtucky county involved. Falls City Industries, Inc. v. VaneD Beverage, Inc. 103 S.Ct- 1282, 1287 (1983,. 9I g., Grand Union Co., 3 Trade Reg. Rep.(CCm 050 (July 18, 1983), at 22 708; Albertson's, Inc. 97 F. 343 345 (1981) (allegation in complaint issued with consent);Godfrey Co., 97 F. C. 456, 458 (1981) (alegation in complaint issued with consent).

CPF 12-125 through 12-129; CPF 12-187 through 12-201. The relevant retail grocery market arguably includes both the Nevada communities of Reno, Sparlu, and Lake Tahoe and the California Lake Tahoe area. ), 280 Opinion requirement. log D. The Injury To Competition Requirement.

In order to establish the injury to competition component of a Section 2(a) violation, the Commission must establish that the price discrimination at issue may substantially "lessen competition or tend to create a monopoly in any line of commerce." 15 U. C. 13(a). More precisely, this implies that the Commission must establish a "reasonable possibility that a (63) price difference may harm competition."101 The injury to competition standards in primary line and seeondary line cases differ, however, and we must therefore discuss the allegations of primary line and secondary line injury separately. 1. Primary Line Injury In order to establish primary-line injury, complaint counsel must establish that the price differences at issue either injured or threatened to injure competition at the primary level; that is, among competing sellers.!o2 That showing may be effected either (1) by means of a detailed market analysis establishing that the discrimination at issue actually injured competition or (2) by establishing predatory intent, from which competitive injury may be inferred.103 Under the first approach, it is not enough simply to establish that individual sellers have been (64) injured, or that some competitors have left the 104market. Direct evidence that competition among sellers has been 100 Since Section 5 of the Federal Trade Commission Act can be used to jil inadvertnt gaps irl the coverage of the Robinson-Patman Act-to an extent consistent with the "spirit" of that Act-it might be possible for the Commssion to apply the more e1'paIlsive jurisdictional coverage ofprirnary line cases to secondary line cases. There are at least two reasons not to take that approach. First, the more restrictive approach is more consistent with economic theory in this case. Grocery retailers compete with one another in SMSA markets, and a pair of discriminatory prices can therefore injure disfavored retailers only ifsales at both prices in the pair are made in the same SMSA. Second and more generally, the Commission has concluded that unike the Sherman and Clayton Acts, the underlying goal of the Robinson-/'abnan Act is the protection of competitors, not competition. Accordingly, the "spirit" theory mWit be applied with great caution in the context of cases brought under that Act.

General Molars Corp. 3 Trade Reg. Rep. (CCH) 16fi (June 21, 1981), at 23 023. 1103 FTC. 641J 101 Fal/s City Industries, Inc. v. Vanco Beverage Co. 103 S.Ct. 1282, 1288 (1983); oCNJrd, Corn Products Refining Co. v. FTC 324 U.S. 726, 742 (1945). Complaint counsel need not show that the discriminations alleged actually injured competition.J. Truelt Poyne Co. v. Chrysler Motors Corp.,451 U.S. 557, 562 (1981). DO!lble JI Plastics, Inc. v. Son(Jco Products Co. 5 Trade Reg. Rep. (CCH) 949 (3d Cir. April 20, 1984), at 102; 0. Hommel CO. V. Ferro Corp. 659 F.2d 340, 346 (3d Cir. 1981),ccrt. denied 455 U.S. 1017 (1982). !03 Double H Plasticsv. Sonoco 65,949 at 68.102; E. Rogers Associates, Inc. v. Gordner-Denuer Co., 718 F. at 1439; 0.Hommel Co. v. Ferro Corp. 659 F.2d at 347;Pacific Engineerin,; Prod CO. V. Kerr-McGee Corp. 551 2d 790, 798 (10their), cert. denied 434 UK 977 (1977); Nation'll Dairy Products Corp.Y. FTC 412 F.2d 605 612-13 (7th Cir. 1969);Beatrice Foods Co., 76 F. C. at 799. m' Wiliam In,;lis Sons Baking CO. Y. 11TContinental Baking Co. 668 F.2d at 1042 (the fact that the plaintiff suffered losses and eyentl.ally ceased operations is not enough to establish a Section 2(a) violation);HommelO. CO. Y. Ferro Corp.659F.2d at347; Interno.tionolAir Industries, Inc.V. AmericanExcelsiorCo. 517 2d 714, 721- (5their. 1975),Cf'rt dl'nied 424 U.S. 943 (1976); Anheuser-Bu. , Inc. V. FTC 289 2d 835, 840 (7th Cir. 1961) (other competitors losing market share does not demonstrate injury);General Foods Corp., supra note 7 142 at 22 989 and n.76 (Commission Opinion), 22,992 (Bailey, Commissioner, concurrng); Beatrice floodsCo., 76 F. G 719, 800 (1969), ofrd sub nom. Kroger Co. y. FTC 438 F2d 1372 (6th Cir. cert. denied 404 U.S. 871 (1971);Lloyd A. Fry Roofing Co. 68 F. C. 217, 260 (1965),affd, 371 F,2d 277 (7th Cir. 1966). ., Opinion 114F.

injuredJ05 must be developed,!o6 Under the second approach, in lieu of direct evidence of anticompetitive effects, the Commission must establish injury to competition by rebuttable inference from predatory intent, which can in turn be established either directly or by rebuttable inference from predatory (65) conduct,!O? When predatory intent or predatory pricing are relied upon to establish primary line injury to competition in violation of Section 2(a), the standards that govern the intent and conduct components of the attempted monopolyization offense should apply. lob For example, the Ninth Circuit has concluded that where (66) a price differential threatens a primary line injury, . .. Section 2 of the Sherman Act . . . and Section 2(a) of the Clayton Act. . . are directed at the same economic evil and have the same substantive content. . ,109 The record does not contain any direct evidence suggesting that Continental' s behavior injured competition. It is true that a number of bakeries, including Laub in Cleveland, Old Homestead in Denver and Inglis in northern and southern California exited their respective markets at some point during the corresponding relevant time period. However, there is little evidence to the effect that these exits were occasioned by Continental's behavior, rather than by poor management, ineffciency, an overall reduction in the demand for white bread, or other unfortunate but nevertheless perfectly competitive explanations. Moreover, even ifthe exit of one or more ofthese bakers could be attributed to the conduct of Continental, that would not be suffcient to establish the requisite injury to competition unless Continental's behavior could be characterized as predatory or otherwise anticompetitive. Vigorous, legitimate competition forces less effcient 10S The "substantial je e(ling of competition" component ora Section 2(a) offense m;iY he somewhat easier to satisfy than the "dangerous probabilty ofsuccessful monopoliz tjon " requirement for proving attempted monopoly. ization under the Sherman Act.Wiliam Inglis Sons Baking Co. v. 11TContinental Baking Co 668 F.2d at 1042. However, whefJ noshowing of predatory conduct can be made it is highly unlikely that the Section 2(30) injury tu competition requirement can be "atisfied by direct evidence.Id.; Anheuser-Bust'h v. FTC, 289 F.2d at 843--4. 106 l"or an example where the Commission determined that no reasonable possibility of injury to competition existed, eenoted 73 supra.

ID1 O Hommel Co. v, Ferro Corp. 659 F.2d at 347; 'onich Brol'.v. American Dist;Filing Co. 570 F.2d 848, 855 (9th Cir, 1977).cert, denied 439 U.s, 829 (1978);Reotrice Food. Co. 76 F. C. 719, 799-800 (1969),afrd 438 F. 1372 (6th Cil),cert, denied 404 U.s. 871 (1971). As the Ninth Circuit has poinood out, however direct evidence of intent alone can be ambiguous and misleading. . . EspedaHy misleading here is the inveterate tendency ofsa.!es execu.tives to brag to their superiors about their competitive prowess, often using metaphors of coercion t.hatare compelling evidence of predatory intent to the naive. Any doctrine that relies upon proof of intent is going to hc applied erratically at. best Wiliam Inglis Sons Baking Co. v. 117' Continental Baking Co. 718 F.2d at 1028 and n.6 (emphasis added). 100 E, ROKers A. 8aciates, Inc. v. Gardner-Denver Co. 718 F.2d at 1439; William Inglis Son. Baking Co 117 Continental B(!king Co. 668 F.2d at 1041-42; O. Hommel Co. v. Ferro Corp. 659 F 2d at 348; Janich Bros. Inc v. American Distilling (;0. 570 2d 848, 855 (9th Cir. 1977), cal. denied 439 S. 829 (1978); Pacific Engineering Production Co v. Kerr.McGee Corp. 551 F.2d 790, 798-99 (10th Cir.),cert. denied 134 U.s, 879 (1977); International Air Induslriesv. AmericlZ/! Excelsior Co. 517 F.2d 714, 720 n. lO (5th Cir. 1975),cert. denied, 414 U.S. 975(1976).

ltil Janich Bros., Inc. V. American Distilling Co. 570 F.2d at 855. ), 280 Opinion firms to exit constantly; that in fact is a frequent consequence "r vigorous competition.

We must therefore determine whether the requisite injury to competition may be inferred from Continental's pricing behavior. We have already determined that sales at prices equal to or greater than average variable cost will be strongly (67) presumed to be legitimate and that sales at prices below average variable cost wil be presumed to be predatory. This standard should apply to defining predation under Section 2(a) of the Clayton Act, as well as to satisfying the specific intent and anticompetitive conduct components of the attempted monopolization offense under Section 2 of the Sherman Act. Therefore, sales at prices that equal or exceed average variable cost cannot ordinarily satisfy the predation element of primary line injury to competition under Section 2(a). 110 The first step in such an analysis must necessarily be to determine the contours of the product that Continental allegedly sold at prices below average variable cost. Professor Areeda has recently argued that in determining whether one should consider the prices and costs . associated with a single product: or those associated with an entire line of products, the crucial issue should be the contours of the product or line that rival firms can sell. When competing firms can sell the same line of products that the alleged predator can sell, selling a single product in that line at prices below average variable cost cannot exclude equally effcient competing firms, which can match those prices as long as overall revenues exceed the variable costs (68) associated with the product line as a whole.11 In this case, the instrument of predation should be defined at least broadly enough to include the full line of private label white pan bread products that Continental sold in each market. As Continental has pointed out (rJetailers do not buy, and wholesalers do not sell, simply one item in a line."112 As a consequence, if a wholesaler wishes to supply private label bread to a given retailer, it typically must supply an entire line of white bread products, including the most popular loaf sizes and hamburger and hot dog rolls. The record evidence does not establish that competing bakers could not, like Continental, produce an entire line of private label white pan bread products. Therefore, to the extent that an entire line of white bread constitutes the minimum product offering necessary for a baker to stay in business, only prices no E. ROf:ers Associates, Inc. v. Gardner-Denver Co. 718 F.2d at 1439; 0.Hommel Co. v. Ferro Corp. 659 F. at 349-50; Pacific Enf:ineering Prod. Co.v. Kerr-McGee Corp. 551 F.2d 790, 797-98 (10th Cir. cert. denied, 434 U.S. 879 (1977);see International Air Industries, Inc. v. American Excelsior Co. 517 F.2d 714, 724 (5th Cir. 1975), cert. denied, 424 U.S. 943 (1976).

II P. Areeda Antitrust Law Supplement 145-46 (1982). m RAP at 33, citing Biechner, Tr. 3553-4, Jones, Tr. 3700. Opinion 104 F.

that cause revenue to fall below variable cost for the entire line could conceivably force equally effcient competitors out of business. The case law supports considering at least the whole line of private label white bread sales in the aggregate. In Janich v. American Distilling, the Ninth Circuit considered allegations that the defendant had attempted to monopolize the sale of private label gin and vodka in California, and had sold gin and (69) vodka oflike grade and quality at discriminatory prices,113 The Court of Appeals determined that whether the defendant sold gin and vodka at predatory rices should be determined by considering its prices and costs for its full line of gin and vodka products, rather than simply its half gallon sizes of gin and vodka,114 The Court indicated that It)he product must be such that ifpredatorily priced, rivals are likely to be driven out of the market or excluded, allowing the firm to raise prices. .115 A strong argument can be made, however, that the instrument of predation in this case should be defined more broadly, to include both advertised label and private label white pan bread products. The relevant product market includes both advertised label and private label bread products, because the cross-elasticity of supply between them is relatively high, and the cross-elasticity of demand between them is at least somewhat significant. Moreover, the Supreme Court determination that physically identical private label and advertised label products should be considered to be of "like grade and quality requires (70) us to treat advertised label and private label white pan bread as commodities of like grade and quality in satisfying that. jurisdictional requirement of the Clayton Act. Defining the instrument of predation to include both advertised label and private label white pan bread products would be consistent with the determination that the instrument of predation should encompass the product or line of products that rival firms can sell. The record evidence indicates that almost all of the major wholesale bakers in the five relevant geographic markets, including the firms that exited those markets during the relevant time periods (Old Homestead in Denver, Prosser and Gordon in Southern California, and Inglis in Northern California/Western Nevada), marketed an advertised label white pan bread product, and nearly as many marketed ItJ Janich Bros., Inc. v. Americon Distilling Co. 570 F.2d 848, 852 (9th Cir. 1977), cert. denietl 439 s. 829 (1978). 114 Id at 856; accord, ILC Peripheral. v. International Business Machines 458 F.Supp- 423, 433 (N.D. Cal. 1978), affd 636 F.2d 1188 (9th Cir. 1980),cere. denied 452 U.S. 972 (1981). The opinion in Janich does not indicate whether the defendant !lold branded gin and vodka, as well as private label gin and vodka, in California. 115 Janich Bros., Inc. v. American Di. tilling Co. 570 F.2d at 856. The Court did recogni e that in some cases " given size might be so significant that a chain retailer would select the overall line on the basis of that size. 280 Opinion private label bread as well.116 The record evidence confirms that-a wholesaler that produces private label bread for a given retailer almost always makes substantial sales of its advertised label bread as a part of the deal. Continental argues that increased advertised label sales are an essential ingredient of any private label transaction; that it sells "a (71) bread line of various items at different prices under two labels " and assesses profitabilty on a full line basis; and that the composite price of the entire line" is the relevant price)!7 In Denver, for example, as complaint counsel themselves indicate, Continental expected to make substantial Wonder bread sales through Associated Grocers retailers as a consequence of its Tender Crust private label program for those grocers. CAB at 54 citing L. Johnson Tr. 7201, CX 1522 D-E. Therefore, it may be perfectly logical for a firm like Continental to sell private label and advertised label bread as a package, offsetting lower prices for and profits from private label bread with higher prices for and profits from greater advertised label bread sales. In fact, Continental argues that ifit were to refuse to sell private label products at prices below full cost, its advertised label sales would decline. RAP at 26. As a result, even ifprices for Continental's private label bread did not always cover the variable costs associated with private label sales, that arguably should not be considered predatory unless those losses were sutIciently high to push overall revenues-including advertised label revenues-below variable cost.

At least one recent Court of Appeals opinion has taken this approach. In Pierce Packing Co. v. John Morrell Co. the Ninth Circuit considered allegations that the defendant had attempted to monopolize the Montana "pork and pork products" market by sellng "pork loins, private label bacon and private label (72) frankfurters" at prices below cost.118 At trial, the district court judge had permitted the defendant to introduce evidence concerning "sales of unrelated products" to "show that (defendant' s) total Montana sales were profitable and that the relevant products were not sold below marginal COSt."119 The Court of Appeals concluded that the evidence was properly admitted because it 1!6 IDF 56-8, 107-110, 181-182, 185-186, 234-238, 288. The only possble exceptions may have been Laub in Cleveland, which made 58% ofiw sales to restaurants and also sold many specialty breads, and Ward in Cleveland. IDF 289-290. A:y wholesaler that can produce advertsed label bread can produce private label bread just as easily. The wholesale bakers that exited therefore could have marketed both private label and advertised label bread just as Continental did. It would be more diffcult for an exclusively private label baker to begin marketing advertised label bread as well, because of the time and promotional costs associated with developing an advertsed label. I..ub' s strong emphasis upon restaurant and specialty les makes it uwikely--ontrary to Corrsaioner Bailey s suggestion-that Continentsl'!! prices to retail grocers were "s major csuse" of Laub's exit from the Cleveland market.

11 RAP at 33 citing Biechner, Tr. 3553-4; Jones, Tr. 3700; Nissn, Tr. 10145-8; Inglis, Tr. 3786; Dierker, Yr. 9790-91.

119 Pierce Pocking Co v. John Morrell Co. 633 F.2d 1362, 1363 (9th Cir. 1980) !l91d at 1364.

428 FEDERAL TRADE- COMMISSION DECISIONS Opinion 104 F.

made the fact of the profitability of(defendant' s) Montana operations, both generally .and with respect to pork and the particular pork products involved in this case, more probable than it would have been without the evidence,120 In short, the Court of Appeals sustained the consideration of the profitability of "pork and pork products " rather than simply pricecost comparisons for pork loin and private label bacon and frankfurters, in determining whether predatory pricing had in fact occurred. The same principles arguably should be applied here. Continental's overall revenues from and variable costs (73) associated with its sales of all white pan bread products-both advertised label and private label-should quite logically be considered in determining whether it sold products at prices below average variable cost)2! The record evidence in this case does not satisfy the test for predatory pricing that the Commission has adopted, whether or not advertised label and private label sales are considered together. As indicated supra sales at prices that equal or exceed average variable cost wil be presumed to be legitimate, while sales at prices below average variable cost will be presumed to be predatory. The Administrative Law Judge determined that Continental sold certain white bread products at prices below average variable cost on a number of occasions. However, to make that determination, the ALJ assumed that Continental's variable costs were uniformly and persistently (74) equal to eighty percent of its "total allocated costs." IDF 45. There are two important diffculties with this assumption. First, it is unrealistic to assume that Continental's variable costs were always equal to eighty percent of its total allocated costs, throughout the period of time under consideration in this case, and in each ofthe five separate relevant geographic markets. Second, the ALJ derived the eighty percent figure by comparing Continental' s variable costs in 1971 with its revenues in 1971 , rather than with its total costs. IDF 45; see 262C-D. Since revenues frequently differ from total costs, there is no way of knowing whether the eighty percent figure accurately reflects the relationship between Continental's variable and total allocated costs. For these reasons, the Commission has determined not to rely 120 Id.

) III the collateral Northern California private litigation, neither the District CuurtOOT the Cour of Appeals expre ly addressed the question of whether advertised label ano private label bread soles should be considered separately or together.W!/iam Inglis Sons Buking Co. v. lit Conlinental Baking Co., 461 F.Supp. 410 (N. Cal. 1978),rev d on other grounds 668 F.2d 1014 (9th Cir. 1982),cert. denied 103 S.Ct. 57 (1982) The Commission has in the past taken the position that private label and branded label sales should not be aggregated to determine whether sales of private label milk at lower prices than advertised label milk to a retailer represented unlawful secondaryJine discrimination. Beatrice Foods Co., 76 F. C. 719, 805-07 (1969),a(fd sub nom. Kroger Co. v. FTC, 438 F.2d 1372, 1379 (6th Cir. 1971),cert. denied 404 U.S. 871 (1971). However, the Commission noted that the key issue is whether the favored retailer is given a competitive advtlntage over competing retailers as a consequence of being able to purchase the cheaper private label milk.Beatrice Fr;r;dsCo. supra 76 F. C. at 806. That position. arguably should not apply when disfavored retailers may also purchase private label products at the same price.Borden Cr;. v. FTC, 381 F. d 175, 180 (5th Cir. 1967) INTRNATIONAL TELEPHONE & TELEGRAPH CORP., ET AL. 429 280 Opinion upon the (75) "eighty percent" assumption that the ALJ developed. The following subsections apply more traditional evaluative techniques to Continental's pricing conduct in each of the five relevant geographic markets,!22 a. Denver Market Prior to 1964, the AU determined that in the Denver market, wholesale bakers did not compete on the basis afthe wholesale or retail prices of their products. . . . There were meetings or communications between them prior to any price move, which all the wholesale bakers took at the same time. IDF 63. Largely as a consequence ofthis arrangement, no wholesaler sold private label bread. IDF 65. However, in early 1964, after three years of effort, Associated Grocers-a cooperative of independent retail grocery stores together accounting for four percent of retail grocery sales in the Denver area-induced Interstate and Continental to negotiate the development of a private label program. IDF 67, 68. In August, 1964 Continental entered into a written agreement to supply (76) Tender Crust private label bread-including one pound expanded loaf, 1.25 pound round top, 1.25 pound sandwich bread, and hamburger and hot dog buns-to members of Five States, a cooperative company that Associated Grocers organized for its private label program. IDF 69-72. The agreement provided inter alia, that in the event that Continental reduced its advertised label prices below their June, 1964 level, it would reduce the corresponding Tender Crust price to at least one cent below the advertised label price,!23 Continental expected a price of$.175 for its one pound Tender Crust loaf to generate revenues forty-six percent above variable cost, because its "cost to the doors of its plant" including direct costs plus "other manufacturing, operating, office/' and " packers -would be nine and one-half cents per one pound loaf. Continental expected its other private label products to make similar revenue contributions. CX 1525 A, C-K. Continental retained the Tender Crust private label business until the middle of 1978, and then reacquired it in late 1979. IDF 94. Complaint counsel argue that Continental sold Tender Crust at prices below average variable cost at various times between 1965 and 122'lhe AU' s assumption a;; to the relationship between Continental's variable and total aJloeate custs does not in any event establish any significantly lengthy instances of sales at prices below average variable cost, as the discussion in each subsection indicates.

In conducting our analysis, we emphasize the evidence that complaint counsel presented, because complaint counsel bear the burden of proving sales at prices below average variable cost, and if the evidence they adduce is not adequate, then 11 finding of saes at such prices cannot be sustained. The cost evidence in the record might be considerably better if Continental had supplied certain subpoenaed accounting forms in a timely faf\hion, or if complaint counsel had heed wiling to acclJpt Continental' s later tender of the forms after the entry of a sam:tions order against it. The sanctions order is discussd in Part IV,infra. 12.1 IDY 73, 78. In 1967, this pre!\cribed differential was increased to three cents. IDF 00. ), Opiiiion 104 F.

1967. CPF 8-191, 8-192. As support for that position, they argue that (1) Continental' s total costs for producing Tender Crust in the third quartr of 1967 amounted to $.1659 per one-pound loaf; (2) eighty percent of that figure is $.1327; (3) in 1965, Continental sold 10 000 loaves of Tender Crust to (77) one grocer for $.125 per loaf; (4) from June, 1965 to February, 1966, Continental sold Tender Crust to another grocer for $.125 per loaf; and (5) in the fall of1967, Continental sold Tender Crust for $.1305 per loaf. CPF 6-191 , 8-192; see CAB at 28. The ALJ relied upon the foregoing data and cost assumptions to conclude that Continental sold Tender Crust at prices below average variable cost in November, 1967 and March, 1968. IDF 82, 83, 85. These data cannot support a finding that Continental sold Tender Crust at prices below average variable cost for a significant period of time. As we indicated supra, we cannot assume that Continental's variable costs were always eighty percent of its total costs, whether in 1967 or otherwse; a more detailed depiction of actual average variable costs must be used. That is particularly important where, as here, the respondent argues that its variable costs were considerably lower. Continental argues that its variable costs per one pound loaf amounted to only $.095 per loaf at the time it entered the (78) Tender Crust contract in July, 1964,IZ4 and the prices at issue exceeded that figure by an estimated three cents to eight cents per loaf during the cited time periods. CX 1728. Moreover, the sales cover one pound Tender Crust bread alone; there is no evidence that if all Tender Crust private label products were included-including other loaf sizes and hamburger and hot dog buns-revenues would not have exceeded costs by an even greater degree. Furthermore, these data all relate to private label sales alone. There is apparently no evidence in the record that-when private label and advertised label sales are aggregated-Continental made sales at prices below average (79) variable cost. In the absence of any other evidence that Continental sold white bread at prices below average variable cost in Denver, the Commission has determined to dismiss the Section 2(a) count as to that market.1 12- RAP at 20, 50, citingCX 1525; RPF 123. ex 1728, which sets forth the expemrs atOciated with saleaofofie pound loaves of Wonder bread and Tender Crust from the first quarr of 1964 through the last quarr of 1967 indicates that the "coat to doors" assciated with a one-pound Joaf of Tender Crust increasd from $.09 in the third quartr of 1967, when the private label contract with Asociated Grocers began, t. as much as $.10 in the their quarr of 1966, and then declined to $.098 in the fourth quartr of 1967. ex I72SC, 1728Y, 1728Z-. Complant count argue that ex 1728 sholtd be disregarded because it is "contradicted by the regular busine records of CEC" and became some of the costs used were estimated as percentages of list price. CAB at 61--2. However, the "regular business records" caver overall plant operations, while ex 1728 covers ot1y one paw1dWonder bread and Tender Crt bread. Moreaver, most of the costs employed were actu.al; omy a few were estimated on the basis of percentages of list prices. ex 1722H, R- Y, Z-21 through Z-1. 121 In the private suit against Continental il: DCl:ver the Court of Appeals sustained Ii finding ofliabijjty under Section 2(a) cfthe Claytn Act. Continental Baking Co. v. Old Homestead Bre Co., 476 F.2d 97 (10th Cir. cert. lhnied, 414 U.8- 975 (1973). However, the Cour noted that the record contained evidence of "saes below cost" for only "a very short period" and therefore appears to have rel.ed primarily upon Continental's construction of (footnote cont' INRNATIONAL TELEPHONE & TELEGRAH CORP., ET AL. 4'H 280 Opinion b. Twin Cities Market In this market, complaint counsel argue that Continental sold Wonder roundtop and sandwich breads from its Minneapolis bakery at prices below average variable cost for one and one-half months in the middle of1967. CPF 9-98; see CAB at 28. As support for that position they point out that during that period (1) Continental's fully allocated costs of manufacturing and distributing those products were $.2512 and $.2513 per loaf respectively; (2) eighty percent of $.2512 is $.201; and (3) Contbental sold Wonder roundtop and sandwich breads for 20 per loaf during the one and one-half month period. CPF 9-96 98; see CAB at 28. Complaint counsel also argue that Continental sold Pantry Pride (a controlled label bread) at a price below average variable cost for two months in late 1967. CPF 9- 111; see CAB at 28. As support for that position, they (80) point out that during that period: (1) $. 156 per loafrepresented eighty percent of Continental's fully allocated costs; and (2) Continental sold Pantry Pride for $.1519 per loaf. The ALJ relied upon these data to conclude that Continental sold Wonder round top and sandwich bread at prices below average variable cost for one and one-half months in 1967, and sold Pantry Pride to a Minneapolis retailer at prices below average variable cost for two months in 1967. IDF 128, 131. However; these calculations also rely upon the "eighty percent" assumption, and there apparently is no other record evidence that Continental sold any white bread products at prices below average variable cost in the Twin Cities market. The Commission has therefore determined to dismiss the Section 2(a) count as to that market.

c. Southern California Market In this market, Continental' s bakeries included plants in Beverly Hils, San Pedro, and San Diego. IDF 175-177. Complaint counsel argue that Continental sold a number of private label white pan bread products from the Beverly Hils bakery at prices below average variable cost in the Southern California market for a four-week period ending in August, 1973. As support for this position, complaint counsel cite Continental's own " reckoning" in an internal study that its total gross sales" were less than its (81) "total variable costs. 126 However, it is important to note that the study covered a large numa large new plant in 1962 to sustan the :requisite finding of predatory ld.intent.at lO4-5. As we have noted supra there arc a variety ofperfecuy legitimate reasons for constructing new, more effcient plants and theD sellng at prices suffciently low to maximize production effciency and minimize unit costs We therefore decline to adopt the conclusion of the Court of Appeals 126 CPF 11-46 citingCX 441F ("McCoy lib. white ), Z-170 ("8&8 Reg ), Z-171 ("8&8 Buttermilk"), Z-188 Jordano 8 pk. Bun ), Z-198 ("Stop !) Go 8 pk. Dog ), Z-199 ("Mayfresh 8 pk. Dog ). Continenta's calculations were apparently made as par of "a special, one.time 'profit by variety ' study" of the Beverly Hils bakery in August, 1973. RPF 325.

Opinion 104 F.

ber of varieties of bread products, and the fact that revenues from six ofthem may not have covered all variable costs during one four-week period cannot therefore support a finding that Continental sold its complete white bread private label line, or its complete white bread advertised label line, at prices below average variable cost during that period, Continental argues that its study indicates that it sold every brand and size of white pan bread from the Beverly Hils bakery, of which there were several dozen, at prices higher than "incremental costs," with the exception of one that represented less than one percent of the sales of the bakery. RPF 325, citingCX 441E. Continental argues in addition that "when the profits of Wonder white bread are included, the total white bread sales of the Beverly Hils bakery were profitable on a full cost basis. Id., citingCX 441. Continental argues further that its standard cost reports for the bakery for June, 1973 through January, 1974 show a sixty percent gross profit margin for advertised label bread, and a forty percent gross profit margin for private label bread, during that period.1 The AU apparently concluded that the record evidence was not (82) suffciently strong to establish that Continental made any sales in this market at prices below average variable cost. See IDF 219, 222. On balance, the record evidence does not support concluding that Continental engaged in predatory pricing in the Southern California market. Complaint counsel cite only four weeks of sales at prices below average variable cost, and the Commission predatory pricing standard requires a showing of sales at prices below average variable cost for a significant period oftime to create a presumption ofpredatory pricing. A single isolated month of sales at such prices cannot satisfy that standard. Moreover, as in the other markets, the evidence that complaint counsel have developed focuses upon different "types of private label white pan bread-in this case, sales of only one or a few private label bread varieties-rather than focusing upon all private label bread products collectively. Furthermore, the record evidence does not establish that Continental made any white pan bread sales in the Southern California market at prices below average variable cost, when both advertised label and private label bread sales are aggregated. The Commission has therefore determined to dismiss the Section 2(a) count as to this market. (83) d. Northern California/Western Nevada Market Continental served this market from three bakeries located in San rancisco, Sacramento and Oakland; the first two produced variety nd white pan breads and cake products, while the Oakland bakery coducedjust variety and white pan breads. IDF 231. Complaint coun- ': RI'F 326 citingCX 433-Z31 , ex 435C, ex 436D, ex 437C, ex 438C, ex 439G. INRNATIONAL TELEPHONE & TELEGRAPH CORP., ET AL. 433 280 Opinion sel argue that Continental sold certain products at prices below average variable cost in this market.l More particularly, complaint counsel argue that Continental sold private label "cello" (one pound expanded" load at prices below average variable cost (1) for four weeks, ending on July 22, 1972 from its Oakland and San Francisco bakeries; (2) for four weeks, ending on August 19, 1972 from its Sacramento bakery; (3) for five weeks, ending March 30, 1974 from its Oakland bakery; (4) for four weeks, ending February 23, 1974 from its San Francisco bakery; (5) for five weeks, ending June 9 1974 from its Oakland bakery; (6)'for four weeks, ending April 27, 1974 from its San Francisco bakery; and (7) .for five weeks, (84) ending September 28 1974 from its Oakland bakery.l29 Continental admits in response that it sustained "bookkeeping losses on a full cost basis in 1971-74 because of federal price controls, the labor strike and substantial cost escalations. . ." RPF 394. However, Continental argues that its plants always had a "positive cash flow" during that period; that in every quarter from 1972 through 1975 its private label price "substantially exceeded the marginal cost of producing and sellng those items. . . . (by about 35%);" and private label products contributed profits of$2. milion during those years.l30 (85) The ALJ apparently did not evaluate or rely upon the foregoing evidence. Instead, he simply assumed that "Continental's (total) costs were comparable to American " in 1973; assumed that Continental's average variable costs were eighty percent of American s total costs; and therefore concluded that Continental's $.172 price for its private label bread during the July, 1972 - August, 1973 period was lower than its average variable cost during that period. IDF 249, 254. The study that complaint counsel rely upon to establish sales at prices below average variable cost-CX 30l0-is helpful, but it suffers from two major deficiencies. First, the data for 16 ounce white round top bread, 24 ounce white round top bread, and hamburger and hot dog buns should be aggregated, in order to determine whether revenues from sales of those products collectively fell below variable costs 111 CAB at 28, citingCPF 12-144, 12-146, 12-149, 12-150; CAB at 7l 72, citingCX 3010, IDF 252, 254, 255. 12!CPF 12-144, 12.-146, citingCX 3010A; seeCAB at 71; RX 3010; Diener, Tr. 12227, 12230-78. CompJaintcoUDeJ also argue thatContinenlaJ sold a number ofjndividual bread varieties at prices below average variable coat during the five week period ending on September 9, 1973. CPF 12--147, 12-148, 12-150 citingCX 733. However, as we have noted su.prQ.,rnflggregatioD beyond the level of private label white pan bread and advertised Jabel white pan bread is not appropriate. Moreover, the variable costs RSlociated with allofthe individual bread varietiell in ex 733 cited by complaint counsel exceeded revenues from sales of those varieties by ony $1,207 during the cited period. By way of contrast, the collective revenues from sales of the third variety listed in CX 733-and ex 733 covers Ii large uwnber ofvarietie xceeded the total variable costs associated with sallJs of that variety by $9 027. See ex 733C. Complaint cOlise! criticize Continental's treatment of certn costs as fixed and others as varable for certin other varieties in CX 733. CPF 12-149. However, they do not provide any indication of whether any wles at prices blJlow average variable cost would be shown if all private label white pan bread were aggregated and all advertised label white pan bread were aggregated. 130 RPF 394-'396 citingRX 125, RX 3010-A. Continental argues that its "marginal cost" for northern California sales amounted to about IJleven cents per loaf at the time it lowered its price to $. 172 in 1972. OpiIiio 104 F.T:C.

in any ofthe relevant time periods. As we noted supra, the instrument of predation should at least include the aggregated sales of all private label white pan bread products. It is not clear from the record whether these three products alone accounted for all of Continental' s private label line; ifit also sold other private label products from its Bay Area plants, then the costs and revenues associated with those sales should be aggregated as well. In any event, when sales of the three identified products are aggregated, one finds that collective revenues from sales of these products exceeded variable costs as defined by complaint counsel in all three of the 1972 time periods during which (86) predatory pricing allegedly occurred.1 Of the remaining five periods that complaint counsel cite,. the data for the two periods in the second quarter of 1974 are less useful because they do not account for the revenues and costs associated with sales of hot dog and hamburger buns during those periods. Since hot dog and hamburger buns account for a substantial portion ofthe private label revenues from each of the three Bay Area plants, it is possible that total private label revenues might actually have exceeded variable costs as defined during one or both of those periods had the revenues and costs associated with hamburger and hot dog bun sales been included. That leaves three time periods-a five-week period ending in March, 1974 concerning the Oakland bakery; a four-week period ending in February, 1974 concerning the San Francisco bakery; and a five-week period ending in September, 1974 concerning the Oakland bakery-during which aggregate private label revenues may have fallen below variable costs as complaint counsel have defined them. There are, however, four significant impediments to basing a finding of liability upon these data. First, they assume that a variety of costs should be treated as completely variable, when at least portions of them probably should be treated as fixed, particularly over the very short periods of (87) time at issue.1 In particular, at least portions of the costs associated with such items as "local plant vehicular ex- " Uhealth and wel-pense " Htractor trailer equipment " ttvacation pay, fare pension fund," ttsales supervisors' salaries " Hgarage labor," and executive" should very probably be treated as fixed costs, particular- !31 Sales of theBe products from the Oakland, San Francisco, and Slicramento bakeries respectively generated cvenues of$1,012, $118, and $813 over varable costs as complaint counsel have defined them during these period. 132 The study asumed that the following items are variable costs: vehkle, accident, sundres, fuel, tranport bor, local plant vehicular expense, outsde expresa and freight, operating labor, packert, electrc power, repair! nd renewal, telephone, telegraph, jantor supplies, rent, lease cas, tractr trailer equipment, vacation pay, illnei' lIyments, health and welfare pension fud, tranport carriers, cartns and sellng tapes, s.es managers, saes ens' salaries and commissons, saes supervsors' salaries, other route labor, sales contets and prizes, special Ibilty and retroadjuatments, payroll taes, garage jabor, executive, supplies, repairs, tires. By cOlltrast, Contillen" I argues that when a bakery has unused capacity, only six items should be treated as variable: the cost of grdients, the cost of the wrapper, saesmell S commssions, direct productiolliahor, oven fuel, aud pan grease. e first three items account for more than ninety percent of variable cost. RAP at 19-20, citing Dierker, Tr. l7- , 9727, 9760 INTERNATIONAL TELEPHONE & TELEGRAH CORP., ET AL. qui. 280 Opinion ly in connection with the four or five week periods the data cover. It is unclear whether revenues would stil fall below variable costs in any ofthese three periods if these adjustments were made. The second problem is that the cited time -periods are too short and episodic to satisfy the Commission standard for predatory pricing. Third, as we have indicated supra it would be perfectly logical to aggregate branded and private label sales in this industry to evaluate alleged predatory pricing, and the record evidence does not establish any such sales when Continental's Northern California/Western Nevada sales are aggregated in that fashion. (88) Fourth and finally, Continental may assert as a complete defense in response to the foregoing data that those sales simply represented a good faith effort to meet the competition of competing firms. The Supreme Court recently determined that " territorial price differences that are in fact responses to competitive conditions" satisfy the requirements of the meeting competition defense)33 In short, if a seller has a good reason to believe that competing firms are charging lower prices in a particular market, it may respond with comparably low prices on a territorial basis, rather than on a customer-by-customer basis)34 The Court determined that this standard would be satisfied by showing that a reasonable and prudent businessman would believe that the lower price he charged was generally available from his competitors throughout the territory and throughout the period in which he made the lower price available.135 The Court also determined that the defense could be asserted even if the prices at issue were offered to secure new customers, rather than to retain old customers)36 The record evidence establishes that the prices noted above were offered as part of a good faith effort to meet the competition of competing sellers. In early 1972, Inglis began (89) sellng one pound private label bread for seventeen cents per loaf; Campbell-Taggart took the Wentz private label account from Continental by offering one pound bread for seventeen cents per loaf; and American lowered its private label price to $.172 per loaf after discovering that Campbell-Taggart had adopted $.172 as its market-wide price, and had offered that price to Food Fair, one of American s private label customers)37 American then offered that price to Mayfair Stores, a large Continental private label account, to cover its private label purchases throughout the 13. Palls City Industries, Inc. v. Vaneo Beverage, Inc. 103 S.Ct. 1282, 1295 (1983). 134/d.at1295-97.

lJ.Id.at1297.

ljl fliat 1294-95.

137 RAP at57 citingRX 30-2, RX 34-4; McGinley, Tr. 3975-813, 3985-6; ex 647; Temkin, Tr. 4297- , 4305; Frelink, Tr. 3380-2 Inion 104 F.

Northern California/Western Nevada market. RAP at 58 citingRX 93; Temkin, Tr. 4296-99. After learning of this offer, and that the same price had been offered to 7-Eleven Stores, the largest private label buyer in the area, and to Cala Markets, Continental lowered its private label price throughout the market to $.172.1 That price persisted until August, 1973; the market price then increased to $. in September, 1973. RAP at 59 citingCX 780. In its opinion in the collateral private litigation between Inglis and Continental in North. ern California, the Ninth Circuit Court of Appeals accepted this account and concluded: (90) This reveals much about the way the market for private label bread operated during the complaint period. Although the record does not always identify which of Continental' s competitors initiated a price reduction, it does reveal how a price reduction to one customer quickly became the prevailing market price. Each bakery ignored this new price at its peril. Because the number of private label accounts was not large, buyers for those accounts communicated with each other and with all of the wholesale baker. ies. Price reductions to one account did not long remain secret. We conclude, therefore that there was a reasonable basis for Continental's assumption that a new price offered to one of its accounts by a competitor, which Continenta did verify, would become available, with or without its assistance, to all of its existing and prospective customers. The weight or the evidence supports Continental' s claim that its price reductions were goo faith responses to competition. There was no need for Continental to verify that each customer had actually received a competitor s offer of an equally low price, as it did in the case of advertised bread. 139 The record evidence before us supports this conclusion. Complaint counsel argue that American s offers did not justify Continental's marketwide price reduction because the offer to Mayfair covered only the Bay Area; Cala Markets operated stores only in San Francisco; and the Regional Vice-President who lowered Continental's price was not certain of any offer to 7-11 (91) Stores" before he lowered the price.1 The ALJ appears to have relied primarily upon the first point to reject the meeting competition defense in his decision. See IDF 250. However, in Falls City the Supreme Court made it clear that the defense should be sustained when it is "reasonable and prudent" to believe that the lower price offered is generally available from competing firms in the relevant market during the relevant time period; universal availability and complete certainty are not required. Price competition was so intense in the Northern California 136 IDF 249; RAP at 58; Frielink, Tr. 3223-24, 3294-97, 3379-3. The ALJ suggests that no other wholesalers sold rivate Jabel bread at this price throughout the market. IDF 250. However, Continental argues that Campbellfaggart and American also quoted a single private label price to an their customers. RAP at 59 n.l, citingMcGinJey, r. 3975-0, IJ9 Wiliam Frielink,Inglis Sons BakingTr.Co. v. ITTContinental3381--2.Hakin,; Co. 668 F.2d lit 1047. The Court neverteless versed the district court' s judg:edt notwithstanding the verdict on the issue, becau. although it believed its mclllion to be themastreasonable conclusion, it could not say that it wasonlyreasonablethe oondu.aion pos.ible. /. at 1047--8. It remanded that issue,inter alia,to the di.atrict cour for a new trial. 140 CAB at 69-70 citing McGin!ey, Tr. 3980, Frie1ink, Tr. 3229, 3383. _._ _. _._ . .. .... 280 Opinion market during the relevant period that Continental could reasonably have believed that the $.172 price its competitors were offering was generally available throughout the market, and that it would lose its private label accounts if it did not match those competing offers immediately.'4! The Commission has therefore determined to dismiss the primary line count in the Northern California/Western Nevada market.

e. Cleveland Market The allegations of anti competitive conduct in the Cleveland market arise primarily from a contract Continental negotiated in 1972 to supply private label bread products to Pick N Pay, a local grocery chain that had theretofore operated its own captive bakery. IDF 305. The ALJ found that during this period: (92) There was excess bread capacity that could service the Cleveland market, especially Continental' s Akron bakery which was operating at about 50% of capacity, one and 114 shift. . , .

IDF 300. As a consequence, the Akron bakery was generating pre-tax losses of over $3 000 per week, and Continental was seriously considering closing it. IDF 305. Continental believed that if it began to supply private label bread to PNP, it could convert its Akron bakery losses into a "substantial profit " and realize a profit net of "fully allocated costs" of $240 000 or more per year.'42 Moreover, Continental expected to increase its sale of branded label products to PNP; in fact, its sale of Wonder bread products to PNP increased by $20 000 per week after it began to supply private label products. IDF 317. The contract became effective in July, 1973. CAB at 74. However, the cost of providing private label bread to PNP turned out to be much higher than Continental had expected. IDF 319. As a consequence, Continental renegotiated its contract with PNP to secure higher prices in September, 1974. IDF 328-332.

Complaint counsel argue that Continental sold private label bread to PNP at prices below average variable cost "on both a single product and total (private label) business basis." CAB at 31. As support for that position, they argue that (1) in the (93) four week August, 1973 accounting period Continental's revenues from its total private label sales to PNP fell below the variable costs associated with those 141 McGinJey, Tr. 3976; Johnson, Th. 3892. I am indebted to Commiasiooer Bailey for her assistace in resolving the meeting competition issue.

142 IDF 305, 315. This figure was derived by convertng the weekly estimate of at least $6 000 in profits in IDF 315 into an annual profits estimate, and then subtracting $70 000 in annllal "proof of perfonnance" payments. Continental paid $210,000 to PNP from 1973 throllgh the first three months of 1976, or approximately $70 000 per year, to "reimburse Pick N Pay for promotions of private label products commensurate with the estimated book value of the Pick N Pay bakery:' IDF 313 Opinion 104 F.

overall sales of all products to PNPsales,!43 and its revenues from its fell below the variable costs associated with those sales;!44 (2) in the five-week December, 1973 accounting period, Continental' s revenues from its total private label sales to PNP fell below the variable costs and (3) in the four week May, 1974associated with those sales;!45 accounting period, Continental's revenues from its total private label sales to PNP fell $16,600 below its variable costs.!46 In short, complaint counsel argue that (94) when Continental's total private label white pan bread product sales to PNP are considered, its revenues from those sales did not cover its variable costs in July, 1973, December, 1973 and May, 1974. 147 Continental argues in response that "The profitabilty ofthe PNP private label account was never significantly below break-even on a variable cost basis." RPF 479. Continental argues, for example, that (1) in August, 1973 private label sales showed a "weekly profit" of $690; (2) in December, 1973, private label sales showed a "variable profit" of $286 per week; (3) in December, 1973 and January, 1974 the whole line of private label bread and buns produced an incremental profit of $658 per week; (4) in May, 1974 the Pick N Pay "account as a whole broke even on an incremental basis. . . 148 and (5) under the contract renegotiated with PNP in 1974, private label sales alone generated profits in the first three months of (95) 1976, in July, 1976 and in August, 1977,149 In short, Continental argues that its revenues from the PNP account usually covered the variable costs associated with that account. Continental admits that some of its documents from the contract renegotiation period show that it incurred full cost and some variable cost losses when its private label sales to PNP are considered in isolation. RPF 480. However, Continental argues that the relevant measure should be whether Continental earned greater total profits (or incurred fewer losses) on both advertised label and private label sales as a consequence ofthe PNP private label contract. 143 CAB at 86, citingCX 800B, ex B83C. ex B83C actually shows that revenues exceeded variable costs by $690. However, complaint COUIcl argue that $5,384-/52 of the $70 000 annual promotional payment,ahould be treated as a varable cost and subtracted from that figue. CAB at 86. !U CAB at 85-86, citingCX 800B; CPF 13-205, 13-207. 140 CAB at 86 citingCX 2639T. ex 2639T actually shows that revenues exceeded variblc costs by $286, but complaint counsel argue that 730-/52 of the $70 000 allual promotional payment-should he treated 8S variable cost and subtracted from that figue. CAB at 86. 146 CPF 13-262, 13-263, 13-266, 13-267 citingCX 2608B, ex 2610, ex 2636C, ex 2665A. 141 Complaint counsel also argue that Continenta sold certain individual varieties of bread to PNP at prices below average variable cost or "total direct CQst" or "cost to doors" at various times throughout the relevant period. CPF 13-265, citingCX 26392- (two varieties; however, a third variety, one pound white bread was priced signficantly uboveaverage variable cost); CPF 13-266, 13-267 citingCX 911Z-4 (one variety), CX 915F, r (three varieties), CX 2677 A, O (three varieties); CAB at 86;citingCX 2632 (two vareties), ex 26392- (two vareties); CAB at 87 citingCX 2678 (fve varieties; however, total revenues OD all varieties exceeded tota cost to doors by $8 838). We have aleady noted, however, that revenues and costs should at Ii minimum be aggrgated across the entire private label or advertsed label Hnc.

,.s RPF 479 citingCX 883C, C'X 2639R, CX 2639A-Q, and ex 2636- 1.9 RPF 481 citingCX 829, CX 2616-A, ex 2615, CX 2658, (,'X 2668, CX 2661, CX 2660, RX 309, Schmdt, Tr. 11189 11191, 11192 ....&....._ .I"'I..

280 Opinion RPF 480. That measure would account for profits and losses attrib!ltilble to advertised label sales, as well as to private label sales. The AU considered these arguments and reached the following conclusions: (1) the Akron bakery sustained a $16 000 loss on its PNP private label contract in the four-week period ending on August 25 1973;!50 (2) in December, 1973, Continental's private label prices to PNP covered incremental costs for "8 Pack" hamburger and hot dog buns and one pound white bread loaves, but did not cover incremental costs for 20 oz. and 24 oz. white bread loaves;!5! (3) in May, 1974 Continental "basically broke even on the total (PNP) busines " on an incremental basis, losing $16 000 (96) on PNP private label and making $15 100 on sales of branded products to PNP;!52 (4) in January, 1976 Continental sold its 24 oz. "Giant Sandwich" loaf and "8 Pack" hamburger buns at prices below incremental costs;!53 (5) during the first three months of 1976, Continental sold "three other bun varieties" under the contract at prices below nincremental costs; 154 and (6) in July, 1976 Continental's private label sales under the contract produced losses on both a "full load" and a "variable" basis,155 In short, the AU found that Continental's revenues from its overall private label sales to PNP fell below the variable costs attributable to those sales only in August, 1973, May, 1974, and July, 1976. The foregoing summaries prepared by the parties and by the do not provide the kind of detailed itemization ofthe costs that should respectively be treated as fixed and variable during the relevant time period that is needed to determine whether Continental actually sold white pan bread products at prices below average variable cost for a significant period of time in the Cleveland market. In any event complaint counsel identify only three non-consecutive months in which Continental' s revenues (97) from total private label sales to PNP may not have covered the variable costs associated with those sales. The AU does not endorse complaint counsel' s conclusion as to December, 1973 and identifies only one additional month over two years later in which such an imbalance may have occurred. These discontinuous and short-lived instances cannot satisfy the Commission standard for predatory pricing.

In her concurring and dissenting opinion, Commissioner Bailey discusses her own analysis of prices and costs in the Cleveland market. Commissioner Bailey s more detailed price-cost comparisons. do 150 IDF ,120 citing ex BOO-B, ex 882, (,'X 2628 1 IDF 322, citingCX 2632-A-F, CX 2639- \2 IDF 324, Quoting Breines, ex 2608-8. Ths figue does not account for the $70 000 anual payment for "proof of performance" IDF 324 citinf!CX 2693-A-B. However, it apparently does account for private label variety, premium, and sweet good sales, as well as white bread sales. \53 IDF 334,citing ex 2669-A.

1M IDF 335, citingCX 2668-B 106 IDF 336 citingCX 2661-C.

Opinion 104 F.

not, however, affect the conclusion that Continental's pricing behavior did not violate the Commission standard. In conducting her analysis, Commissioner Bailey treated almost all sellng and distribution costs as variable.1 In addition, she treated Continental's reimbursement of PNP's private label promotional expenses as a variable cost.1 Finally, she relied upon the same documents-but not the eighty percent assumption-that the ALJ relied upon in making his cost calculations.1 Commissioner Bailey found that Continental sold white pan bread to PNP under the private label contract at prices below average variable cost during the (98) months of June, 1974 (87.0% of average variable cost) and July, 1976 (99.3% of average variable cost).159 Commissioner Bailey s calculations provide a helpful comparison of prices and costs under the PNP private label contract. However-like the calculations of complaint counsel and the ALJ-they do not create a presumption that Continental violated the Commission standard. They reveal only two months-over two years apart-when Continental's revenues from its private label white pan bread sales to PNP under the contract failed to cover variable costs, and the percentage point shortfall in the second month was very small. Two widelyseparated months of sales at prices below average variable cost are simply not suffcient to satisfy the "significant period of time" requirement of the Commission predation standard. The allocation of expenses between fixed and variable costs in Commissioner Bailey s analysis creates additional diffculties. In particular, at least a portion of Continental's reimbursement to PNP of promotional expenses associated with its (99) private label bread sales should be treated as a fixed rather than a variable cost. Continental believed that "any baker who secured (PNP's private label) business would either have to purchase (PNP's captive) bakery assets or compensate Pick N Pay for their book value. "160 Instead of simply paying for the assets, however, the parties agreed that Continental would pay $210 000, an equivalent amount, to PNP over the July, 1973-April IOUConcurring tmd dissenting opinion ofCommiasioner Bailey at 12 citingGase. Tr. 9421-22, RX 309, ex 2661. 157 Id.at 12-13 and n. 16.

158 ld at 14 citing IDF 320-125, 347.

158 Jd. at 14. Commissioner Bailey also found that Continental collectively soJdai/of its private label products to PNP (inc!udil1g variety, premium and sweet goods, as we!! as white pan bread) at prices below average variable cost during four months (two of which were consecutive) in 1973 and 1974. However, given their differences, variety, premium and swed goods probably should not be included in the same relevant product market as white pan bread produds. Moreover, such product and white pan bread do not in any event appear to be goods oflike grade and ql\ality," since they arenot physically identical Conlrary to Commissioner Bailey s assertion (id at 20),the record evidence does not indicate that Continental sold private label bread producl to PNP at prices below fully allocated cost for four years. As she herself points out, the document.s she relies upon "do not. detaij every month of the four year period, but sWJlnarize performance at irregular intervals.. Id at 11 n. 15.

100 RAP at 64-65 citingCX 2683, Vail, Tr. 9968, 9971-72. A number of other retailers had insisted upon and received similar commitment. when they switched from captive baking to purchasing private label bread from wholesalers. Id at 65 n, INTERNATIONALT.tLJ:.tt1Vl .l al J.J.:u.."""w. 280 Opinion 1976 period when PNP presented proof of promotional efforts sucnadvertising.!6! Nevertheless, even if the payments are treated as entirely promotional, at least a proportion should be treated as fixed costs, because of the investment component of promotional efforts that we have described supra. Treating some of the reimbursement to PNP in that fashion would of course reduce Continental's average variable costs to some degree, although the record evidence does not establish the percentage ofthe payment that should be excluded from variable costs. It seems likely, however, that the difference would at least be substantial enough to place Continental's revenues from private label sales to PNP in July, 1976 above variable costs, leaving only the month of June, 1974 as an instance of sales at prices below average variable cost. (100) Apart from these considerations, it is important to recognize that the evidence discussed above relates only to private label sales white pan bread to PNP. It does not indicate whether the findings of sales at prices below average variable cost would persist if Continental's sales of branded white pan bread products were also considered. If the latter sales were considered, then the $16 000 loss on private label sales to PNP in May, 1974 would be countered almost completely by Continental's $15 100 in revenue above incremental costs from sales of branded products to PNP, yielding a net loss below incremental costs of only $900. IDF 324. Adopting the same approach with respect to the second instance in July, 1976 might produce the same result, although the ALJ did not include a finding as to Continental' revenues from branded products sold to PNP during that month. In addition, it should be noted that the price cost comparisons focus only upon Continental's sales to PNP; they do not represent an aggregation of costs and revenues associated with private label sales from the Akron plant to all private label buyers.!62 Including private label sales to firms other than PNP might very well indicate that revenues from all private label sales from the Akron plant exceeded the variable (101) costs associated with those sales throughout the relevant period. For all of these reasons, the Commission has therefore determined to dismiss the Section 2(a) primary line count in the Cleveland market.

2. Secondary Line Injury The Commission must similarly establish a "reasonable possibility that a price difference may harm competition" in order to establish secondary line liability; that is, that the price discriminations alleged Id. at 65 citing" Knwitz, Tr. 5435-31, Vail, Tr 9971 9977. \01 By contrast, the price-cost comparisons developed for the NorthertJ California/Wester/; Nevada market focus upon aU private lahf!l bread sales from the Oakland, San Francisco, or Sa.cmmento plallt. ), ), Opinion 104 F.

may have injured competition among competing buyers.'63 The Supreme Court has indicated that substantial price differentials between competing purchasers over time are suffcient to give rise to an inference of injury to competition.164 However, the only remaining price differential at issue in the secondary line aspect of this case concerns a price disparity between advertised label bread (Wonder bread) and private label bread in the Northern California/Western Nevada market. The Supreme Court has recognized that in these circumstances the difference in wholesale prices may be accounted for by the existence of a consumer preference for the advertised label product and that preference "should receive due legal recognition in determining (102) whether an injury to competition has occurred.'65 As a consequence, it is not enough to establish in this case that Continental sold private label bread to a favored retailer at a lower price than the price at which it sold Wonder Bread to competing retailers. Such a difference may simply reflect a consumer preference for Wonder Bread at the retail level, and hence be perfectly legitimate, because the private label bread has a limited consumer franchise while the advertised brand is largely pre-sold through national advertising.'66 The record evidence indicates that the price Continental charged four grocery retailers for private . label white bread products ranged from 25 percent to 38 percent lower than the price it charged two other retailers for Wonder bread between. January, 1972 and December; 1974, These differentials are substantial, but there is no reason to believe that consumer preferences and the substantially greater costs associated with promoting Wonder bread do not largely or completely account for them. Moreover, as we indicated supra the prices that Continental charged for private label white pan bread products in Northern California were in any event simply a good faith effort to meet (103) the competition of rival bread manufacturers. The Commission has therefore determined to dismiss the secondary line count in the Northern California/Western Nevada market.

16. Falls Cityv. Vanco 103 Ct. at 1288. 164 ld. at 1289; accord, Beatrice Focx Co.,76 F. C. 719, 801 (1969), o.rrd sub /Zorn The Kroger Co. v. 438 2d 1372 (6th Cir. ccrt. tknied, 404 US. 871 (I97l), citing United Biscuit Co. of Americav. 350 F.2d 615 (7th Cir. 1965),rert. denied 383 U.s 926 (1966) and Foremr;st Dairies, fne. v. FTC, 348 F.2d 674 (5th Cir. cert. denied 382 U.S. 959 (1965).

165 F7v The Borden Co. 383 U.S. 637, 646 (1966); accord Borden Co. . FT 381 F.2a175, 180-1 (5th Cir. 1967).

100 Borden Co. v. FTC, 381 F.2d 175: 18Q1 (5th Cir. 1967); Beatrice Food: Co" 76 F. C. at 80S-9; Fr Policy With Respect to Anticoropetitive Practices In the Marketing ofG!l5oline, 3 Trade Reg. Rep. (CCll) TIlO373 (1967). The AW concluded, without citing any evidentiary support, that some of the differentials were Ja.rger than any differential that could be attrbuted to coW!umer preferences. ID at 88. In the absence of any evidence to support that wnclu.ion, we cannot. accept it . !! lil L:"' 280 Opinion IV. SANCTIONS IMPOSED BY THE Administrative-LAW JUDGE One final issue must now be -addressed, relating to the imposition of discovery sallctions against Continental by the first ALJ to preside over this case. Resolving this issue is very important to preserving the integrity of the adjudicative process, and the Commission wishes to strongly reaffrm the power of Administrative Law Judges to impose such sanctions, as appropriate, in adjudicative proceedings. The relevant facts are as follows. In February, 1976, at the request of complaint counsel, ALJ Harry R. Hinkes issued a subpoena to Continental for certain documentary material.B7 Shortly theteafter IB8 In lateContinental fied a motion to quash or limit the subpoena. March the ALJ denied that motion, but after an April prehearing conference he deferred Continental's obligation to comply with (104) the subpoena until J uly,169 After several subsequent months of negotiation, complaint counsel reduced the scope of certain subpoena specifications on which compliance had been delayed. At a prehearing conference in October the ALJ ordered Continental to comply with the modified subpoena, and to submit a progress report on its compliance by December 10, 1976. Tr. 180-81. On that date counsel for Continental advised the ALJ that it would not comply with the portions ofthe modified subpoena requiring the production of four types of accounting forms, Forms 430 (cake portion only), 521, 526 and 528.170 Form 430 reveals the volume of production and sale of each variety of bread and cake in units, weight and dollars. Forms 521 , 526 and 528 provide, in differing degrees of aggregation, the expenses associated with baking groups of products, such as all bread, all cake, or all bread and cake combined. In conjunction with "recipes" that Continental did provide-showing how much of each ingredient and how much labor and baking time are used for each variety omplaint counsel believed that these forms would permit the (105) allocation of Hall bread all cake," or !!all bread and cake" costs to different bread varieties on the basis of units, weight, and value. Continental argued that the subpoenaed forms were not relevant to 11;7 Mr- Hinkes served as the AI. for this case until July, 1977. At that point, he asked to be relieved of his responsibilities for the cae becaus he would shorty be presiding over the trialKellQggin Co. Notice To Daniel H. Hallom, Chief Administrative Law Judge (July 26, 1977). Ths cage was then assgned to Miles J. Brown. Order Substituting Admirustrative Law Judge (Augit 2, 1977). 1GB Motion To Limit Or QUlish Subpoena Duces Tecum (March 4, 1976). J09 Order Denying Respondent' s Motion To Quash Complaint Coum,el's Subpoena of February 20, 1976, And Setting Prehearing Conference (March 18, 1976); Order Reschedulng Retur On Complaint Counsel's Subpoena Duces Tecum Directed to m.continenta (April 110 Memorandum In Support of Complaint COImseJ's19,Application1976).For Sanctions Under Rule 3. 38 (December 29 1976), at 6-12.

!7 Id at 9. Complaint counsel defined a "variety" of bread-such as "one pound white pan private label" or " ounce hearh rye loafn to be "8 product which is identified separately on IT"Continenta' s detailed accounting recotds." Complaint Counsel's Answer to Respondent' s Motion For Oral Argument Ard Leave To File Additional Memorandum (April 20, 1977), at 2 n.

...

Opinion 104 F.

this case because, it alleged, the Commission had not endorsed and probably would not endorse complaint counsel's "current efforts to develop cost figures for individual bread varieties. . . Continental argued that, during the relevant period, it had only kept "cost and profit and loss records on the basis of the full bread product line (as distinguished from the cake line and the sweet goods line). . . . " that complaint counsel had previously requested and received those records; and that those records show that in the various bakeries Continental' s overall bread sales were profitable most ofthe time and that when they showed a loss it was only on a fully allocated basis which nevertheless contributed to overall plant profit.173 (106) Continental also argued that because it "did not keep cost figures by variety" during the relevant period, the fact that a given variety of bread might have been sold at prices below cost could not establish that those sales were made with the requisite predatory intent,14 In late December, 1976 complaint counsel applied to the ALJ for the imposition of a number of discovery sanctions-including adverse inferences and evidentiary restrictions-against Continental. March 22, 1977 ALJ Hinkes entered the following order: It is ordered That complaint counsel are deemed to have established that ITT Continental has sold bread below cost no matter how cost is measured in the following geographic markeis during the following time periods: Akron, Ohio................ "..- 1970 through 1974: Beverly Hills, California .w',..,'..... 1967 through 1974; Denver, Colorado ",....,'... 1964 through 1968; Minneapolis, Minnesota. 1964 through 1970; Oakland, California .. 1967 through 1974; Rochester, Minnesota '... 1964 throogh 1970; Sacramento, California. 1967 through 1974: San Francisco, Calif... 1967 through 1974; San Pedro, California .""., 1967 through 1974: Seattle, Washingtn.. 1968 through 1973; and Youngstown, Ohio 1970 through 1974, It i,qfurther ordered That ITI Continental may not introduce into evidence or otherwise rely upon the documents which ITT Continental has failed to produce. (1071 It is further ordered That ITT Continental may not object on the grounds that the withheld documents are better evidence, to complaint counsel' s introduction and use of other relevant material and reliable evidence that I'I'T Continental made sales below ri Letter from John H. Schafer, Counsel for Respondent, to the Honorable Harry R Hinkes (December 10, 1976), at 2-3.

ld. at 3. Continental argued that when the Commission issued the complaint in this case, it only had the overall bread line" cost and profit evidence before it. Request For Pre-Hearing Conference and Alternative Reql!est For wOlve To File Further Memonmdum (Febrl!OIry 4, 1977), at :: Letter from John H. Schafer, supra note J72, at 4 175 Complaint Counsel's Application For Sanctions Under Rule 3.38 (Dec. 29, 1976) 280 Opinion cost in the geographic markets and for the.time. periods.specified.1 On March 23, the ALJ permitted Continental to apply to the Commission for a review" of his order on the narrow policy question of whether it would be more appropriate for the Commission to seek District Court enforcement of the subpoena at issue than for the ALJ to issue the sanctions order.!77 Continental fied its appeal shortly thereafter. In explaining its earlier refusal to supply the subpoenaed documents, Continental once again argued (1) that the documents were not relevant because the Commission had relied only upon "full bread line figures" when it issued the complaint; and (2) that the documents could not in any event establish that Continental (108) knowingly sold below cost with intent to injure and destroy competition and competitors" because it did not know what it!; individual variety costs were when it marketed the products.!78 On June 29, 1977 the Commission denied Continental's appeal, concluding that the issue certified did not warrant interlocutory review.!79 Shortly thereafter, on July 8, 1977 Continental notified the ALJ that it would provide all of the documents at issue; however complaint counsel argued in response that Continental's offer should be declined. Continental nevertheless collected the responsive documents and tendered them to complaint counsel; on September 9, 1977 complaint counsel refused to accept them.!80 In September Continental moved to set aside the sanctions order, with the understanding that it would provide the subpoenaed documents. In November the ALJ denied the motion.!8! Continental again moved for rescission of the sanctions order in July, 1979, at the conclusion ofthe case in (109) chief, noting that complaint counsel could present the evidence during their rebuttal case and that Continental would waive "any right to present rebuttal evidence on the matter. !82 Complaint counsel 116 Order Imposing Sanctions (March 22, 1977), at 2 (citations omitted). The order defines "bread" to include white pan bread, bread type rolls, and related products, and describes each "geographic market" as "the location of an I'l Continental bread plant" a.d the area flurrol1ndingit. Jd. nn. Order AJJowing lit.Continental' s Appeal Of Order of February 16, 1977 (March 23, 1977). The ALJ did Dot certify the questions of the validity of Section 3.38 of the Commssion Rules of Practice or the propriety of the underlying subpoena.

\78 Application For Review Of Ruling Of Administrative Law Judge (March 3D, 1977) at 3-. Continental also argued that compjiaD.ce with the subpoena would be costly and timf!consumng. Complaint counsel later indicated that simply analyzing the data in the subpoenaed forms would require "a minimwn of a year" after the material had been supplied. Complaint Counsel's Response To Show Cause' Order (Sept. 29 , 1977), at 2 n. 179 Order Denying Appjication For Review (June 29, 1977), at 2. 18( Motion To Set Aside Orders Granting Application And Imposition Of Sanctions (Sept. 30, 1977), at 2; Answer To Respondent's Motion To Set Aside Orders (Oct. 14 , 1977), at 4. IR' Order Denying Motion To Set Aside Orders Granting Application For And Imposition of Sanctions Under Section 3.38 of the Commission lI Rules of Practice (November 7, 1977). The ALJ later denied Continental' subsequent motion for an interlocutory appeal of his order. Order Denying Respondents' Request For Interlocutory AppeaJ (November 28, 1977) IR2 Respondent.' Motion To Resch:Jd " Order Imposing Sanctions" Of March 22, 1977 (July 16, 1979), at 10-11. Inion 104 F.

once again opposed the motion, and the AU subsequently denied it.1 One last incident involving the documents covered by the sanctions order occurred the following year. In March, 1980, after Continental had concluded its defense case, complaint counsel served an extensive subpoena on Continental to secure evidence for its rebuttal case. Specification 32 of that subpoena sought documents "suffcient to show the Company s profits or losses on white pan bread alone" for most of Continental's bakeries during the 1969- 1979 period.1 Although Continental moved to quash the subpoena, it offered the documents covered by the sanctions order to partially comply with Specification 32, and delivered them to complaint counsel on March 1980. Complaint counsel refused to accept the documents and later returned them, arguing that the documents were not responsive because they related to costs rather than profitabilty, and that (110) it would be patently absurd for us inadvertently to abandon the adverse inferences upon which we have relied for four years. We did not intend to do so. Moreover, if Mr. Wachter s critique can only be met at that price, we shall forego the luxury.185 Complaint counsel simultaneously fied an application to modify the subpoena by waiving production of the documents covered by the sanctions order.1 On April 15, 1980 the AU granted Continental' motion to quash in part and struck most of the subpoena specifications, including Specification 32.1 The current significance of the adverse inferences that the sanctions created is somewhat unclear. The AU briefly described the sanctions order in his opinion, but did not rely upon any inferences derived from that order in reaching his decision. See ID at 3-4. In their appeal brief, complaint counsel refer to the adverse inferences in the sanctions order as "an alternative ground of support for the conclusion that respondents sold below average variable cost." CAP at 6 n.4. In their answering brief, complaint counsel argue that "The adverse inferences in this (111) case, CCPF 2- , are a suffcient ground for finding predatory conduct and intent. 188 Finally, in the oral argument before the Commission, complaint counsel indicated: At this stage of the case we rely on one element of that (sanctions) order, that Continen- Order Denyi!Jg Respondents' Motion To Rescind " Order ImpowDg Sanctions" Of March 22, 1977 (September 28, 1979).

1M Complaint Counsel's Application For ISluancc Of A Subpoena Duces Tecum To lit Continental (March 7 1980) (Instructiolls), at 17.

IB. Motion To Modify Order Of March 22, 1977 And To Admit RX 125 (March 28, 1980), at 2-; CompJaillt Counsel' s Opposition To Motion To Quash March 1980 Subpoena (April 7, 1980), at 15-16; Response To Motion To Modify Order Of March 22, 1977, And To Admt. RX 125 (April 9, 1980), at 3-. 1M Application To Modify March 1980 Subpoena To ITI Continental Baking Company, Inc- (April 7, 1980), at 187 Order Limiting Complaint COllel's Discovery As To Their Casein-Rebuttal (Aprii 14 1980). 188 CAB at n. l; see also iliat 28 (relying upon CPF 2-7 inter alia in an effort to show sales at pricet; bcJow average varable cost), 43.

), . . 1.1 .0. .n.u. .LTJ:1\l"H\TIUl'lft lJ: rnVl"H:... .IJ:Ll.&'I.I\. 280 Opinion ta should not be allowed to rely on documents they did not produce to (complaint) counsel. On that ground the cost study Mr. Schafer attempted to introduce (RX 125), (was J properly held out, only certin portions were allowed in. Judge Brown did not rely on the adverse inferences relied upon by Judge (Hinkes).89 In short, although complaint counsel relied extensively upon the adverse inferences prescribed by the sanctions order in their briefs to the ALJ, they appear to have reduced that reliance to some degree during the appeal phase of this proceeding. The adverse inference rule upon which the sanctions order in this case relies provides that when a party has relevant evidence within hi control which he fails to produce, that failure gives rise to an inference that the evidence is unfavorable to him,190 The power of federal courts to impose discovery sanctions such as adverse inferences pursuant to Rule 37(b) ofthe Federal Rules of Civil Procedure is well established. Most courts have sustained an analogous power on the part of federal administrative agencies (112) to draw adverse inferences from the failure to produce relevant evidence and the Commission has adopted that position with respect to its own adjudications.!9! If the evidence at issue has been subpoenaed and the party involved declines to comply with the subpoena in order to suppress the evidence, that may strengthen the adverse inference to be drawn from the failure to provide the subpoenaed material.192 By contrast, the inference may be weakened by a strong alternative explanation for the failure to provide subpoenaed material, such as an effort to avoid the public disclosure oftrade secrets for which one or (113) more patents are pending.!93 Administrative agencies may also prohibit an entity that deliberately withholds relevant subpoena- 189 Oral Arguent Transcript at 33-4.

110 International Union (UA W)v. NLRB 459 1o'2d 1329, 1336 (D.C. Cir. 1972). 191 R. Mallory Co. v. NLRB 400 F.2d 956, 959 (7th Cir. 1968);NLRBv. A.P. W. Products Co. 316 F-2d B 903-4 (2d Cir. 1963);NLRBv. Wallick 198 F.2d 477 483 (3d Cir. 1952);NLRBv. Remington Rand, Inc. 94 F. 862, 868 (2d Cir.), cert. denied 304 US. 576 (1938);Market Development Corp. 95 C. 100 223-27 (1980) (dictum); Amerirun Medical Association 94 F. C. 701 , 1027-29 (1979), affd, 638 F.2d 443 (1980),affd by an equallydiuided Court per cu.riam 455 U.S. 676 (1982).But see NLRB v. International Medication Systems, Ltd. 640 F.2d 1110 1115-16 (9th Cir. 1981).

This conclusion is not inconsistent with the more general principle that administrative agencies cannot compel obedence to compulsory process by imposing fines or imprisonment. ICCv. Brimson 154 U-8 447, 485 (1894). Discovery sanctions such as adverse inferences are considerably different from the imposition offines or imprison. ment through the contempt power because they do not actually compel the production of subpoenaed material. They simply give the subpoenaed party the option of either complying with the subpoena or facing the adoption of adverse inferences. If the party choooos the latter course, it will not suffer any injury unless an order is subsequently entered against it on the basis of the inferences involved, and it can secure judicial review of the propriety of the inferences on appeal.

192 International Union (UA W)v. NLRB 459 F.2d at 1338.But see NLRBv. International Medication Systems, Ltd. 640 F.2d at 1115 n.

193 Evis Mfg. Co. v- FT 287 F.2d 831, 842-7 (9th Cir. cfl-t. denied; 368 U.S. 824 (1961); but see Charles Of The Ritz Dist. Corp.v. FTC 143 F.2d 676, 678 (2d Cjr. 1944). Opinion 104 F.

ed material from relying upon that material in its own presentations.!9' The central purpose of these sanctions is to "maintain the integrity of the hearing process. !95 Rule 3.38(b) of the Commission Rules of Practice provides in relevant part that an Administrative Law Judge or the Commission may impose sanctions such as those imposed in this case when a party fails to comply with (1) a subpoena, (2) an order for the production of documents, or (3) an order issued by the Commission or an AU as " ruling upon a motion concerning such an order or subpoena or upon an appeal from such a ruling. . ." The Rule indicates that the Commission or an AU (114) may impose these sanctions to permit the "resolution of relevant issues and disposition of the proceeding without unnecessary delay. . ." The Commission has developed some more specific principles to help determine when one or more ofthese sanctions should be applied. In American Medical Association, the Commission stated:

Application ufthe adverse inference rule may only be made when the party s failure to produce documentary or other evidence is not adequately explained. Thus, the adverse inference rule makes the conduct of the person withholding the material an evidentiary fact in and of itself. The resulting inference may be strong or weak, depending on the person s conduct and the surrounding circumstances. For example, an inference drawn against a respondent offering a weak explanation for its refusal to produce relevant evidence will be stronger than an inference drawn against a respondent providing a more plausible explanation. 196 The Commission concluded that the Administrative Law Judge had properly adopted an adverse inference against the respondent for failing to supply certain subpoenaed materials. The Commission noted that it was highly unlikely that the AMA's jurisdictional challenge to the subpoena-directed at its principal defense-would succeed, and that the AMA, despite its jurisdictional doubts, had nevertheless complied with every other subpoena issued in the case. In conjunction with the absence of "a strong (115) explanation for noncompliance " these facts persuaded the Commission to conclude NLRRv. Cll Sprague Son Co. 4281".2d 938, 942 (18t Cir. 1970). The Court noted that this conclusion might not have been valid if the company had taken the po ition "that all the information sought by the subpoena wa irrelevant. Id. However, the firm had admitted that the requested information was relevant, and had offered no justification for failing t.o comply with the subpoena.Id. Administrative ag!oncie may also decline to permit a firm that withholds subpoenaed materia! from later producing econdary evidence to prove what could have been conclusively established if the subpoena had been honored. NLRBv. American Art Industries, Inc" 415 F.2d 1223, 1229 10 (5th Cir. 1969),citing Bannon Mills, 146 N.LR.B. 611 (1964);Imt see NI.RBv. International Medication Systems, Ltd. 640 F.2d 1110, 1115-16 (9their. 1981) 195 NI.RRv. American Art Industries, Inc. 415 F.2d 1223, 1230 (5th Cir. 1969);accord, NLRBv. C.R Sprague & SOl! Co. 428 F.2d 938, 942 (1st Cir. 1970); International Union (UA W)v. NLRB 459 F.2d 1329, 1338 (D.C. Cir. 1972) 1% American Medical Ass 94 F. C. at 1027; aC,-Jrd, Markel Development Corp. 95 F, C. at 226 (dictwn) _... .. .. ..

280 Opinion 197that the adverse inferences had been properly drawn. More recently, in Grand Union the Commission elaborated upon its earlier analysis to conclude that sanctions under Rule 3.38 should be imposed only if (1) production ofthe requested material has been mandated by a subpoena or specific discovery order issued by an or the Commission and directed at the party (or its offcer or agent) from whom the material is sought; (2) the party s failure to comply is unjustified; and (3) the sanction imposed "is reasonable in light of the material withheld and the purposes of Rule 3.38(b). 198 The COp:mission noted with respect to the third requirement that "(a)n adverse ruling is a severe sanction to be imposed only in extraordinary circumstances. 199 The Commission agreed with the AU that adverse inferences against complaint counsel would have been inappropriate because the delay in furnishing certain information to the respondents had been the product of a misunderstanding, and complaint counsel had made a good faith effort to disclose all requested data to the respondents as they became aware of it.2oO (116) The foregoing principles have led the Commission to conclude that it should not rely upon the adverse inferences that the AU prescribed in his sanctions order in this case. In this sort of situation, Rule 3. should be interpreted to permit the party that fails to supply the required documents to tender them within a reasonable period oftime following the issuance of an order imposing sanctions. Prior to that time, a party that elects to contest portions or all of an order for the production of documents or other materials does not know whether the AU or the Commission will in fact impose some or all of the available sanctions, modify the terms of the subpoena or order, or instead apply to a district court for enforcement. This approach would be consistent with the procedure adopted in International Union. There, the Court of Appeals confronted a respondent in an NLRB proceeding that had for seven years refused to provide clearly relevant documents in response to an NLRB subpoena. The Court of Appeals nevertheless directed the NLRB to draw an adverse inference from that failure to produce only if the respondent failed to produce the documents at issue within thirty days after the entry of the Court's order.201 The court indicated that in order to be absolutely certain that no miscarriage of justice occurs, we think the company should be given one last chance to come forward with the documents. Now that the consequences of suppression have been made abundantly clear, surely Gyro- 191 American Medical Ass 94 F. C. at 1028. J98 Grand Uni,,, Co_ 3 Trade Reg. Rep. (CCH) 1122050 (July 18, 1983) at 22 731. Idat22 732 200 Id. at 22 730-22,731- 1m International Union (UA W) NLRB. 459 F2d OIt 1347--8. Opinion 104 F.

dyne will produce the rehiring records if they are in any way exculpatory. If the company stil prefers (117J suppression--ven at the price of having its cost-cutting defense stricken-then the tenor of the documents wil be obvious to all. By permitting Gyrodyne a last chance to come forward with the documents, however we do not mean to suggest that the proceedings may be delayed indefinitely while Gyrodyne ponders its decision, We have seen quite enough pondering-and not nearly )Ugh deciding-already. Therefore, the Board should allow the company 30 days to produce the rehiring records. If, by the end of that time the company has still not come forward with the evidence, the consequences outlined above should swiftly follow. 202 In this case, the ALJ simply imposed the sanctions at issue without giving Continental the alternative of tendering the disputed documents within a limited period of time-thirty days would probably have been a useful maximum-after the finalization of his order. The ALJ then certified for appeal to the Commission the policy question of whether the Commission should rely upon his adverse inferences or should instead seek federal district court enforcement ofthe subpoena at issue. Once the Commission determined that interlocutory appeal of the ALJ's order was not appropriate, and remanded the issue to the ALJ, Continental immediately tendered the disputed documents. Nevertheless, the ALJ refused to withdraw the adverse inferences entered earlier, and instead permitted complaint counsel to refuse to accept the disputed documents. The ALJ should have permitted Continental to tender the disputed documents within thirty days after the Commission denied Continental's appeal, and should have withdrawn the adverse inferences once Continental did (118) SO.203 That approach would have provided a better resolution of the cost issue than the sanctions order, and the purpose of Rule 3.38(b) is after all to induce parties to supply subpoenaed material. We should note that if the ALJ had prescribed an additional time period within which to tender the subpoenaed documents, and Continental had refused to supply them within that time period, then reliance upon the ALJ's sanctions order-to (119) determine in particular that Continental sold bread at prices below average variable cost-would have been entirely appropriate.204 202 Id.at 1348.

20.1 This is not to suggest that an ALJ order adopting discovery sanctions must always be certified for appeal to the Commission. Th Commission Rules give ALJs the authority to determine the prop r scop of discovery orders, and to impose sanctions for failure to comply with such orders when appropriate; c rtification will be apprppriate only in unusual circumstaces- If an ALJ does not certfy the imposition of discovery sanctions to the Commis.ion for appeal, then the party involved would be required to provide the subpoenaed materials within thirty days or some other pos.ibly shortr prescribed period afr the entry of the ALJ's sanctions order in order to have the sanctions withdrawn.

Continental' s explanawoD!: for its behavior, which are described above, might conceivably weaken the strength of the inferences to be drawn from Continental's failure to produce the subpoenaed docwnents to some degree. However, we need not resolve that issue becaus of our detennnation that the procedure for imposing the sactions should have been modified. We do note that ALJs are quite capable of resolving relevance questions relating to discovery orders, and we urge them to do so as expeditiously lis poasible. 20 Permitting a short period such as thirty days within which required documents may be tendered afi r the entry ofa sanctions order need not necesarily delay Commission adjudicative proceedings. The COmnS8ion Rules (footnote cont'd) &.

INRNATIONALTELEPHON1' ;r:,V'.... 280 Separate Statement The discovery sanctions set forth in Rule 3.38(b) represent a legitimate and necessary procedure, and the Commission wil vigorously apply them when necessary tOJ'remedy an unjustified failure to comply with a valid subpoena or other discovery order. The Commission has, however, determined for procedural reasons not to rely upon the adverse inferences in the ALJ' s sanctions order in this case. V. CONCLUSION The Commission has concluded that Continental did not violate Section 5 of the Federal Trade Commission Act or Section 2(a) of the Clayton Act. The Commission has therefore determined to dismiss the complaint in this matter in all respects.

COMMISSIONER PATRICIA P. BAILEY CONCURRING IN PART AND DISSENTING IN PART After I presented the draft of this opinion to the Commission for consideration, it became clear that the Commission was unanimous in its view as to the disposition of most of the case. That is, regardless of the cost standard used to define "predatory pricing," those parts of the case involving St. Paul/Minneapolis, Denver, Northern California and Southern California, should be dismissed for failure to establish a violation of either Section 2 of the Sherman Act or Section 2(a) of the Robinson-Patman Act.

A majority of the Commission, however, disagreed with the cost standard contained in my draft and therefore disagreed also with that section of the draft involving Cleveland because the standard presented resulted in a finding of Robinson-Patman primary line liability in that market.

Thus, I concur in the opinion of the Commission in this case with respect to the dismissal of all Sherman Acts charges and all Robinson- Patman charges outside of Cleveland, although I do not necessarily ascribe to various modifications as to nuance and emphasis in those portions of the opinion. In particular, as I stated in connection with the final decision in General Foods Corporation, Docket 9085 (103 C. 204 (1983)), I do not agree that product differentiation is only rarely an entry barrier. Nor do I see the necessity for a lengthy discussion of national market trends when the focus of the case is on local markets. (2) The crux of my dissent concerns the question of how to distinguish curently permit parties to fie motions to quash i:ubpoenas aod other discovery orders, and delay comphance with those orders until the motions have been. ruled upon- If an ALJ felt such a procedure to be appropriate, he or she could, consistent with the rules, prescribe sanctions for the faiJw-e to produce subpoenaed materials within a prescribed period at the same time that he or she denies a motion to quash the subpoena or other order at issue. . Commiasioner Pertchuk join.s in this separate statement Separate Statement 104 F. a predatory price from a legitimate competitive price. The majority approach is too rigid and, as we are dealing with a stil developing and controversial area of law, their approach is prematurely strict. Predatory Pricing Few issues in antitrust law have produced such a gallmaufry! of economic theory and legal precedent. Since 1975 there has been " virtual explosion in the legal and economic literature dealing with predatory pricing. " Brodley & Hay, Predatory Pricing: Competing Economic Theories and the Evolution of Legal Standards, 66 Cornell L. Rev. 738, 740 (1981). At least nine different economic theories for detecting predation have been advanced " and the courts have been In theseboth selective and (3) idiomatic in applying these tests. circumstances it is neither fruitless nor presumptious for the Commission to forge its own rule. However, that rule should be a flexible cautious one, capable of assimilating new learning on the subject and avoiding excessive leniency or harshness to either plaintiffs or defendants. The majority s approach, it seems to me, is less an analytical tool than a conclusory statement which can fairly be characterized as follows: price discriminations are either harmless or justified, thus price predation does not exist. I cannot share their confidence on this point; nor do I believe there would be such an outpouring of academic and judicial debate if the issue were all that clear. Accordingly, my approach, described more fully below, would be a phased series of structural and firm-specific inquiries, incorporating a cost-price benchmark for legality which varies (4) depending on the circumstances ofthe case. While I agree with the majority that prices l "This is Que oftlte greatest GaJiy-maufries that ever I saw; but it WkiS intended a8.m Antidote against Plague Salmon l'harm- (1678) The semina! disr.ssion recommended 11 short-fun marginal cost pricing rule using average variable cost as a practical surrogate for marginal cost. Areeda & Turner Predatory Pricing and Related Practices o(Sectiai'Under the Sherman Act 8f! IIarv. L.Rev- 697 (1975)- This proposal was challenged for disregarding the risk that a dominant finn can 8llct:essfut!y pursu,- fl strategy of sacrificing short-term profit for long-term benefits in order to exclude actual! or threatened competition. Scherer,Predatory Pridng and the Sherman Acl: A Comment Harv. L. Rev. 869 (1976) (offering an economic model for testing cost based rules and suggesting a broad rule-ofreason approach). Other commentators proposed long-run pricing rules t.hat emphasized different cost factors. R Posner Antitrw;l Low: An Economil. Per. pective 184 - 196 (1976) (presumptively condemning sales below average total cost with intent to exclude a competitor);.loskow & KJevorick A Framework (or Analyzing Predatory Pricing Policy, 89 Yale L.J. 213 (1979) (proposing a two-tier test: only ifmonopoJistic conditions exist in the market may pricing below the ilverage variable cost be conclusively ilegal, or pricing below average total cost be' presumptively ilegal under specified conditions). Other economists recommend approaches focusing on output changes or the timing of price cuts. WiHiamson, Predatory Pricing, Strategic and Welfare Anulysis 87 Yale LJ. 284 (1977) (barring dominant. firms from expanding output or selling below fuB cost to forestall entry); Baumol,Qu.asi- Permanence of Price Redu.actions: A Policy of Prevention o( Predatory Pricing,89 Yale L.J. I (1979) (barring price incn"..ses by a domiJ:ant firm for a specified period after its price cuts drive competitors from the market). Although not proposing a specific legal standard, two commentators have drawn attention to the prerequisites for succe5Sfb! entiy"deterrance conduct. Salop, Strategic Entry Deterrance 69 Am. Econ. Rev. 335 (1979); Spence Entry, Capacity, Investment and Olin:opolistic Pricing,8 Been. Econ. 5:34 (1977). And finally, at !ea"tonecommenta tor has argued that there should be no standard at al!, since no problem exists. R. Bark The Antitrust Paradox 154(1978).

a Hurwitz & Kovadc Judicifll Analy. is of Predation: the Emerging Trends 35 V..nderbilt I,. Rev. 63 (1982) g 280 Separate Statement above average total cost (ATC) are legal r disagree strongly with theoften conclusive" presumption oflegality they assign to prices below ATC but above average variable cost (A VC). As my analysis of the Cleveland market demonstrates, I believe such prices can be predatory, particularly as they approach the A VC line and if they continue for some time. On the other hand, in some market conditions prices in the zone between ATC and AVC can be legitimate. Therefore looking at the facts of each case rather than relying on near-conclusive presumptions is, for me, the only responsible way to decide the issue. Finally I would attach a much stronger presumption of illegality to prices below A VC than does the majority; I would limit the number of excuses for pricing at that level, and I suspect I would find the conduct to have anticompetitive effects after a much briefer predatory pricing incident than the majority. Aside from the use of near conclusive presumptions, the majority tests are no more effcient than the one I propose: cost definitions must be made under either. We are in agreement on the propriety of the "leap-frog" analytic technique as announced in General Foods Corp, D. 9085 (103 F. C. 204 (1983)). That is, we all agree oil avoiding the time and resource-consuming quagmire of cost-based pricing rules if easier preliminary inquiries reveal that. below cost pricing either could not result in successful predation or is shielded by a legal defense. Therefore, I would first examine competition in the alleged market to see whether (5) and what kind of predation is possible. The existence of entry barriers and the strength of respondent's market power are significant factors. Also important are the level of capacity in the market and duration of the alleged predatory incident. I believe that the relevant measure of capacity utilization is that of the market and not that of the respondent because, in order to predate, a firm must always have some excess capacity. Zerbe and Coop- An Empirical and Theoretical Comparison of Alternate Predation Rules 61 Texas L. Rev. 655, 682 (1982). Otherwise, it cannot serve its rival's former customers when exit is induced. Thus, finding that the respondent has excess capacity may not be exculpatory. However, if capacity utilization is very low throughout the market, competitive market conditions may have forced respondent to price at or below its short term marginal costs in a desperate effort to avoid the even greater losses of temporarily closing or leaving the market altogether.' On the other hand, where the market does not face substantial excess capacity, pricing below marginal cost begins to look suspect, because competition should force prices to at least that level. (Areeda . See, e. Williamson supra, 87 Yale L. , 284 (1977);Willam Inglis Sons Baking Co. v. ITT Continental Baking Co. 461 F.Supp. 410, 418-19 (N.D. Cal 1978),orrd inputs and rev d in part 668 F.2d 1014 (9th Cir. 1981), crrl. denied 103 s.n. 57 (1982); ILC Peripherals Lea. inl! Corp. v. IBM Corp. 458 F,Supp. 423 (1978), o.trd per dlriom sub nom. Memorex Corp. v. IBM Corp. 636 F.2d (9th Cir. 1981),rert denied 452 U.s. 972 (1981) ), 454 FEDERAL TRAJ:E COML\ISSION DECISIONS Separate Statement 104 F. & Turner supra 88 Harv. L. Rev. at 702) The inference is that prices were lowered, not in response to competition, but rather in anticipation of their destructive (6) effect upon competitors and consequent enhanced market position of respondent.

Having established the competitive setting, I would then determine the relevant measure of cost. There is a general consensus that pricing below marginal cost gives rise to a presumption Of ilegality. There is much argument, however, on what accounting definition of cost is the proper evidentiary surrogate for that elusive economic benchmark, which is not recorded on a firm s business records. Some courts and commentators have suggested that a company s prices be compared to its average total cost (ATC);6 others have suggested average variable cost (A VC);7 stil others have suggested a middle course. In my view, no one cost standard is always appropriate; (7) rather, the market setting dictates the choice. Price below A TC can be predatory where there is a high level of capacity utilization in the market, and pricing below A VC is presumptively predatory. For me, the presumption against legitimate prices below A VC is very strong,9 but could be rebutted by a showing of excess market capacity as discussed above. Furthermore, I would take the duration of the alleged predatory incident into consideration. When the more lenient ATC standard is used the low prices must endure for some significant period of time. As the price level approaches A VC, however, the scope of harmful duration may be shortened. When price falls below A VC an even shorter span of low pricing may be deemed potentially harmful. Of course, the presumptions of harm to competition derived from the level and duration of the price reduction ultimately must be tested against any evidence the record may contain about actual impact of respondent' conduct upon competition.!o Thus, if it is clear from a preliminary examination of the record that the market continued to function competitively after the alleged predatory incident, the case may be See generally,Areeds & Turner, supra, 88 Har. L. Rev. at 712, 733. 6 Posner, supra; Trcmsamerica Computer Co. v. IBM Corp. 698 F.2d 1377 (9th Cir. 1983) (pricesaboueATC are not per se!awful, hut plaintiff must prove predation by clear and convincing evidence),eer!. denied 104 S.CL 370 (1983); bu.t see Barry Wright Corp. u. 11TGrinnell Corp., 19BQ-1 Trade Cas. 862 (D. Mass. 1981)affd1984-1 Trade Cas. n65 787 (1st Cir. 1983) (prices above ATC conchL'Iively lawful) 1 Areeda & Turner supra; Northeastern Telephone Co. v. A T&7' Co. 651 F.2d 76 (2d Cir. 1981), crr!.denied, 455 S. 943 (1982);International Air Industries Inc. v. Amerioon Excelsior Co., 517 F.2d 714 (5th Gir. 1975),cert. denied 439 U.S. 829 (1978).

R Zerbe and Cooper supra (compare prices to ATC unle!l excess capacity exisb!; in that case, compare to A VC); Joskow and Klevoric supra' and compare Willam Inglis & Sons Baking Co. v. IT Continental Baking Co., supra (price below ATC is predatory ifaccompanied by other proofofpredatory intent) within! Communications Corp. v. AT&T, 708 :('2d 1081 (7th Cii-. cert. denied, 104 S.Ct. 234 (1983) (conclusive presumption of legality for prices exceeding long-run incrementa WElts; very little weight attached to subjective evidence of intent). 9 Areeda & Turner wou.d make it a conclusive presumption, 88 Harv. L. Rev. at 733, as do Joskow and Klevoric under specified market conditions, 89 Yale L.J. at 252 lQ Post-predation evidence is not a necessry element of a predation case, but often exisb!, given the slow process of antitrust litigation, as in this case. Where it appear in the record, it should be considered. INRNATIONAL TELEPHONE & TELEGRAH CORP., ET AL. 4()() 280 Separate Statementdismissed without tracing the elaborate steps of the cost-price quadrile.

I agree with the majority that the issue of pricing below cost reached in only one of the five markets examined in this (8) case. The Sherman Act counts cannot survive in any city, once a market definition including captive bakers is accepted. I agree that the Robinson- Patman counts are dismissed because of a valid meeting competition defense in Northern California, and lack of data from which to generate accurate cost-price comparisons in Southern California ' and St. Paul/Minneapolis. In these last two markets I would add my own conclusion that there has been a demonstrable lack of anticompetitive effects. In both markets the sum total of competitors remained practically unchanged after the alleged predatory incidents. In the Denver market I would dismiss the Robinson-Patman count on the grounds of a valid cost-justification defense. Virtually the only cost data in the record is contained in an accounting study prepared by Continental for the Old Homestead litigation. That study shows that for the last eight weeks ofl967 Continental priced the one pound Tender Crust bread loafbelow average variable cost. Setting aside the questions of whether the one pound loaf is an adequate vehicle for predation and whether an eight week period is suffciently long for effective predation, and assuming, arguendo, that the study is entirely free from methodological error ll the case for predatory (9) intent must stil fail. Whatever else it may prove, the Old Homestead study clearly shows that the difference in price between Tender Crust and Wonder products was cost justified.

Advertising expenses are a specific line item on the cost study: Wonder Bread had known advertising expenses, while Tender Crust, as a private label brand, had none. The Old Homestead cost study consistently shows (1964-1969) that costs of advertising Wonder exceeded the price difference between Tender Crust and Wonder, even when discounts are included in the calculation,12 The study also shows other specific costs which are generally higher for Wonder than 1J Complaint. counseillre in the anomaJo1.s position of urging that ex 1728, which shows sales above fully allocated costs for 1964 through October, 1967, be disregalded becaus offaulty methodology--xcept as it pertns to the last eight weeks of 1967, when sales below average variable cost are shown. (CCAB, 62) In brief, complaint counsel contend that respondent erred in allocating production costs between Tender Crust and Wonder Bread on the basis ufsale!! price. (CX 1722U-X, Z) For purposes oftm!! allocation, the saes price was 8!!umed to be the same for Wonder as for Tender Crust, in recognition of the fact that the two Jabels surround identical products. Complaint counsel wou.ld have allocated costs "in proporton of unts, weights aud values (Memorandum in Support of Complaint Counsel' s Application for Sanctions Under Rule 3. , December 29, 1976 p. 9) For Robinson-Patman purposes we need not decide between thel allocation methods, since both sides apparently agree that Wonder and Tender Crust should have identical amounts of allocable costs assigned to them, under any allocation method. Thus all non-specifically allocahle expenses cancel out and can bl! bypassed when evaluating the cost justification defense, which rests on specific expenses, as described above 12 The ALJ incorrectly statl!d that CX 1728 docs not show dil\ounts on Tender Crst. (lDF 83; ID at 86) It does show such discounts, on the line headed "Other Sellng Expenses." (CX 1722Z-) Separate StatNlu: 104 F. Tender Crust, such as returns and route labor;13 but the high advertising costs alone can establish a complete cost justification defense. Indeed, it (10) is clear that throughout the relevant period Continental made more money on Tender Crust, or at least lost less in loss periods, than it did on the Wonder label. Since the price differentials were cost-justified, I would dismiss the Robinson-Patman case. This leaves Cleveland as the only market in which to demonstrate our differing approaches. My analysis is as follows. Cleveland In this market between 1973 and 1977 ITT-Continental allegedly captured and kept, by means of below cost sales, a large private label white pan bread account, thus causing primary line competitive dislocations which were stil observable in 1980. The preconditions for predation were certainly present in the Cleveland area in the early 1970's. The record shows no new entry between 1970 and 1980, a decade which reaches significantly before and beyond the alleged predatory conduct. On the eve of the incident almost no baker serving the market had excess production capacity. Milbrook and Ward' , the first and second-ranked wholesale bakers were both running at least two full shifts a day. (IDFs 290, 308) Other bakers were similarly at full capacity, according to several witnesses, including Joseph Signore, Continental's then-Regional Vice President. (Bateman Tr. 5775; Gase Tr. 9375-76; Signore Tr. 9989- 10051) The exception was Continental, whose white pan bread plant at Akron was operating at only 50%-60% of capacity (11/4 shilts). (IDF 300) (11) Joseph Signore, the chief architect of Continental's drive to secure the aforementioned private label contract, recognized that Continental' s unique under-capacity situation could be used not only to win the private label account, but also to make Continental the "dominate (sic) factor on the market." (CX 2683B).

In these circumstances, I would infer predatory intent and effect from sales below fully allocated cost (F AC)14 even if sales were not below average variable cost. The place I find such sales is in the private label contract which Continental negotiated with Pick' Pay (PNP), a major grocery store chain in the Cleveland area. That contract was signed on July 13, 1973, and amended September 25, 1974. (IDFs 311, 328) The record contains a variety of data (cost studies J;j The full rack service offered with Wonder Bread included pick-up orstalc bread; no pick up of returns was offered in the private label program. (CX 1722Z-) Because of a union contract, bakers paid lower commssions \0 route salesmen on private label bread than they did on advertised label bread. (CX 1722Z-3) !4 Fully allocated cost, someUmes called full cost, is uSlod as a surrogate for average total cost in this case because ontinenta!'s records did not show ATC, an economic cO!Kept which in e3Sence is FAC plus a nor:al return on 'vestmenL FAC is thus a more lenient proxy for marginal cost than is ATe. (Arceda & Tuner have noted that orma! rdurn on investment is "a figure usually not determinable with any precision" 88 Harv. L RtJv. at 709) g., g., INTERNATIONAL TELEPHONE & TELEGRAPH CORP., ET AL. 40' 280 Separate Stateml!nt analyses and monthly sales reports) by which the profitability of the contract may be tracked from its inception through August 1977.1 Nevertheless there are further definitional questions about the PNP contract which must be answered before one can determine if white pan bread was sold below cost. The written agreement (12) (CX 803) was not the full extent of Continental's obligations. There were also a variety of "side bar" agreements to lease PNP' s trucks, racks and dollies and rent the PNP warehouse for the early months of the contract. Continental negotiators made such obligations contemporaneously with the formal, written contract and Continental honored those obligations. (lDFs 313, 314) Therefore, where the Continental internal cost studies assign these costs to the account (e. CX 2680) I have included these costs in my calculations. A second issue concerns what costs should be considered variable under the PNP contract. The appeal briefs set up quite a conflct on this point, with complaint counsel arguing that virtually all selling and distribution costs are variable, and respondents' counsel asserting that sales commissions are the only truly variable sellng expense. However, Continental's internal cost studies belie the theories of respondents' counsel. Uniformly these studies, supported by testimony of Continental's employees, describe almost all sellng and distribution costs as variable. (See, e. Gase Tr. 9421-22; RX 309; CX 2661). Accordingly, I have taken the Continental's variable cost calculations as given, and have not subtracted out such items as sales management and vehicular costs, as respondents' counsel advocate. I have, however, included in my cost calculations one variable not shown in the primary Continental cost records. As noted, Continental made many auxiliary, verbal commitments to the July 13 1973, written agreement with PNP. (lDFs 313, 314) One (13) was to reimburse PNP for promotions of private label products. (lDF 313). While these payments were known to the Continental management level, they were not disclosed to the accountant who prepared the line profit studies. (Vail Tr. 9971- , 9981-82; Schmidt Tr. 11163--4; CX 2626A , 0) Accordingly these studies lack that item. While respondents counsel made no attempt to argue that the promotional payments are a fixed cost1 indeed, the payments are classic examples ofa variable cost, since they fluctuate directly with changes in output-they never- '5 The doruments do 110t detail every month of the four year period, but summarize performance at irregular intervals, providing the nine data points which are referenced in the tabllJ!! infra. l'i Continental may have had the option of incurring these costs in a lump sum, one-time fix..d form as a purchase ofPNP' s bakery assets. (Vail Tr. 9968, 9972). The promotional payment obligation was subject to an outer limit ofthc estimated book value oftha!!e aSHcts. (lDF 3t3) However, the fact that these CO!\ts could have been structured differently is specl1lative and irrelevant; no doubt other terms of the contract would have been different if Continental had committed to an upfront payment of $210 000. In the contract as performed, that sum was stretched over three years, and conditioner! to bread output. Its effect on the cost ofbotb the total contract and the white pan bread line was variable.

Separate tatement - 104 F. theless made no effort to correct the incomplete contemporaneous variable profit calculations. Such adjustments can be made, since the amount of promotional payments is known. (CX 2682; Breines Tr. 11096-11100) They should be made, since the amount is significant: approximately $210 000 between July 1973 and April 1976. (IDF 313) I have made those adjustments,!7 (14) This brings me to the relevant cost calculations. It should be emphasized that they are based on the same documents which the.AU used to reach his cost conclusions (IDFs 320-325, 347). My review, however had to be more precise inasmuch as I neither accept his ruling that average variable cost (A VC) always amounted to 80% of fully allocated cost (F AC); nor would I find liability on the mere fact of sales below F AC. The degree by which costs of either type exceed prices must be closely observed.

My calculations are set forth in the following tables: Table 1 Price as a Percent of Average Variable Cost Month(s) & Year for White Pan Bread June 1974 87.0"/0 January 1976 101. January March 1976 101. July 1976 99. August 1 977 110. Table 2 Price as a Percent of Fully Allocated Cost Month(s) & Year r White Pan Bread December 1973 Price below FAC on a per-unit basis for each white pan bread product ranging from 99.99% of FAG to 91.0% of FAG exclusive of promotion payments May 1974 81. June 1974 77. December t 975 82.4 % January 1976 81. January-March 1976 81. July 1976 79. August 1977 779% (15) 17 The promotional payments apparently were in support of the white pan bread products only. Therefore, ariving at the cost figures for white pail bread products I attributed al of the promotional payments to thoae products, 280 Separate Statement Thus, between 1973 and 1977 Continental persistently priced far below FAC on.the white pan bread items in the PNP contract. For four years the white pan bread price was consistently about 20% less than fully allocated cost. This deep a cut below F AC often approached the A VC level and twice fell below the A VC level. White pan bread was sold at such a loss that the entire private label contract never made a profit on a FAC basis during this time. IS (lDF 347; Gase Tr. 9402) (16) The inference of predation raised by cost data is confirmed by a survey of the marketplace before and after Continental won the PNP contract. In 1971 there were five strong independent bakers in the Cleveland market, plus a scattering of smaller bakers. The five top companies were of roughly equal strength, and there was no price leader among them. (IDF 299) Far from being the leader of the pack Continental shared third rank with American. (IDF 296) By 1980 Continental shared dominance of the market with Interstate. Those two were the acknowledged price leaders. (IDF 342) Laub, one of the top five firms in 1970 left the market completely after losing its PNP shelf space to Continental. (IDF 318, 341) American r able 3 Price as a Percent of Average Variable cost lor Ihe Private month(s) & Year Label Account August 1973 98. December 1973 98. May 1974 92. June 1974 89. January 1976 106. January-March 1976 1D3-7% July 1976 101.0% Augusl1977 108. Table 4 Price as a Percent 01 Fully Allocated Cost for the Private Months(sj & Year Label Account August 1973 98. December 1973 Price below FAC on a per-uni! basis tor all bullwo low-volume varieties therefore price below FAC for entire private label account May 1974 86. June 1974 79. Deceber 1974 84. January 1976 83. January-March 1976 82. July 1976 79. August 1977 , , p.

Separate Statement l04F.

Continental's erstwhile head- to-head competitor, had slipped to sixth place, behind even Nickels and Schwebel, bakers which Continental's regional vice president assessed as Hnot strong factors, grocery-wise (Gase Tr. 9533; IDF 342) The market is clearly less competitive in 1980 than it was in 1970.

Having determined that the PNP contract was predatory for as much as four years, I turn to the question of whether Continental has any recognizable defenses.

The tortured history of Continental's negotiations for the PNP private label account is ably set forth by the AU at IDFs 305-314. My reading of the record, which consists mainly of testimony of persons involved in both sides of the negotiations, convinces me that the correctly concluded that Continental has no "meeting competition defense under Section 2(b) of the (17) Robinson-Patman Act,!9 (ID 85) This is abundantly clear with regard to the renegotiated 1974 contract, as there is not the slightest evidence in the record that Continental believed its offers, which were still far below F AC, to be in good faith response to any competing offers. As for the original July 1973 terms, it appears that the prices were decreased and contractual obligations increased several times in the early part of the year significantly after competing bidders' offers had lapsed and after Vail Continental' s vice president in charge of national accounts, became confident that Continental would become PNP' s supplier of private label bread. (IDFs 309, 310; CXs 803, 809, 839, 884) A second defense which respondents raise in Cleveland is the argument that Laub was not harmed by the effects of the PNP contract. In other words, they argue that the causes of Laub's demise were business problems unrelated to Continental's low cost sales. My examination ofthe record convinces me that Continental' s conduct, though not the sole cause, was a major cause of Laub's closing. In the early 1970's Laub had been losing significant (18) restaurant business and some small grocery accounts, often to Continental;2! but its overall business, especially the grocery side, was definitely operational. It had a 'fully automated, very effcient plant, an aggressive sales force, and the label rights to a nationally-recognized label Sunbeam bread". (Stonbraker Tr. 5529, 5563-66 5598; Bronczek Tr. 5710- 1" It should be noted that, in Cleveland, any meeting competition defense islimited to the competing bids for the PNP contract. It is not an "area-wide" defense such as the Commission considers in the Northern California market.

20 PNP originally was interei\tcd in dock delivery to its warehouse and received several bids on this proposition. (Kravitz Tr. 5394-95) However, when PNP changed jL terms to store drop those bids were not renewed- (Kravit 5393- 5408; Schwebel Tr. 5817; Bogolrnony Tr. 5869; Bateman Tr. 5777 ex 884) Signore had only the most general beUefthat other companies might be in the running for the contract; after 1972 he was not aware of any spedfic competitive offers. (Signore Tr. 9993-9994, 10031) 2\ Complaint cOWlsel devote oome time to CBC's "potshotting" of Lauh restaurant accOlmt8; the Initia! Decision also notes CRe's inroads here. (IDF 301-303) However, since the record contains PO indication that these ,;les were won by predatory means, I have not considered these practices to be part of the case .

IN'lJ:ltNATIU1'll\ T.I.Lrn.Vl'UCI QI ,I.LU.LW1 .1.. VV.Lw.. 280 Separate Statement 11) In 1973 Laub had approximately 54 grocery routes, which com' pares favorably wjth the 68 routes of Interstate, the market's then leading wholesale baker. (IDF 288; Stonbraker Tr. 5588) In 1973, PNP was Laub's largest customer, and Laub's bread enjoyed the largest share of the branded portion of PNP shelf space. (Stonbraker Tr. 5590-91) As a result of the Continental contract Laub' s all-important white pan bread sales to PNP declined drastically. Laub' s white pan bread products were simply edged off the shelf by Continental's private label and advertised brands. (IDFs 317 , 318) Moreover, the loss of general exposure to consumers through the PNP stores also hurt Laub's sales through other outlets. (Stonbraker Tr. 5605-5606) The loss of volume associated with exile from the PNP stores had an immediate effect: Laub was forced to consolidate delivery routes, but even that cost-cutting measure was not BIough to save the company and within six months the bakery had shut down. "You just can t go on when your volume is not there. " (Stonbraker Tr. 5608) (19) I think this chain of causality is fairly strong, and it becomes more so when we note that Interstate, though considerably larger and healthier than Laub, also suffered from losing .PNP shelf space to Continental. (Meehan Tr. 5341) Clearly, the PNP account would be very important to any supplier, and its loss could be the final straw to a smaller bakery such as Laub.

Of course, actual injury and permanent loss of sales need not be proven to show a violation ofthe Robinson-Patman Act; but such facts are convincing evidence of a violation. National Dairy Products Corp. v. FTC, 412 F.2d 605 (7th Cir. 1969). Here Continental' s long, deep cuts in the price of white pan bread create such a possibilty of harm to competition that a violation of the act must be found, absent some showing that the probable effect did not take place. Respondents have not made such a showing. To the contrary: all the evidence in the record points to a market much less competitive now than it was a decade ago; with Continental's dominance unchallenged by either new entrants or existing competitors. Moreover, the loss of at least one independent baker seems directly related to Continental's predatory pricing between 1973 and 1977. Accordingly, I would have found that Continental's discriminatory prices on white pan bread in the Cleveland market from 1973 to 1977 caused primary line injury in violation of Section 2(a) of the Robinson-Patman Act. Conclusion The difference between my views on predation and that of the Commission majority was sketched out in my partial dissent to the General Foods opinion and earlier in my dissent to the decision (20) , Final Order 104 F.

not to seek certiori in the Borden (Realemon) matter. The majority opts for a series of assumptions that places the danger zone well below A VC ("properly defined", of course). It is inconceivable to me that any firm could fail to show prices safely above this line, given the wealth of acceptable excuses listed by the majority, not to mention the requirement of a "significant," wholly continuous period oflow prices. (Apparently four years-the time of below cost sales in Clevelandnot "significant" eIJough).

The approach I have outlined is assailed principally because it is subject to accounting ledgerdemain. To this I answer, so is any test where the definition of cost is at issue.22 I do freely admit to one of the criticisms leveled at my approacl: by the majority: it does not foster as much industry certainty as their A VC test. Certainly, my approach would require a modicum of structural and firm-specific inquiry. Nevertheless, I believe it is a practical, workable standard. In contrast, the majority s A VC test gives near absolute business certainty after one reading: in the words of Cole Porter Anything goes." It would be simpler, and surely a great saving of everybody time, if the Commission today had simply announced that it does not believe predatory pricing exists.

FINAL ORDER This matter has be".. heard by the Commission upon the appeals of complaint counsel and respondent from the initial decision and upon briefs and oral argument in support of and in opposition to the appeals. For the reasons stated in the accompanying Opinion, the Commission has determined to reverse the initial decision. Respondent' s appeal is granted and complaint counsel's appeal is denied. Accordingly, It is ordered That the complaint is dismissed. Commissioners Pertschuk and Bailey dissented in part and concurred in part.

22 For example, the m3jority would amortze promotional and advertsing expenses over a product' s goodwill life cycle (presumably established by the promotor s restimony as to his fondest expectations). If this isn t an arbitrary variable, fraught with accounting peril, what is DIAMOND CRYSTAL SALT CO. "0"

463 Modifying_Order