Boise Cascade Corporation

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

IN THE MATTER OF BOISE CASCADE CORP.

Docket 9133. Interlocutory Order, November 1990 ORDER REISSUING FINAL ORDER This matter has been remanded to the Commission for further proceedings by the United States Court of Appeals for the District of Columbia Circuit. Boise Cascade Corp. v. FTC 837 F.2d 1127 (D. Cir. 1988), The remand directs the Commission to determine whether Boise Cascade Corporation s rebuttal evidence overcomes the inference of competitive injury previously drawn by the Commission. Having reviewed the record and considered the briefs and oral arguments of the parties, the Commission has determined that Boise Cascade Corporation has failed to overcome the inference of competitive injury established by the evidence. Accordingly, It is ordered That paragraphs I to V of the Commission s order dated February 11 , 1986 , are reissued. See Boise Cascade Corp., 107 FTC 76 , 224-25 (1986).

Chairman Steiger abstaining and Commissioner Owen not participating OPINION OF THE CO:vMISSION By CALVANI Commissioner:

* Prior to leaving the Commission, former Commissioner Calvani registered his vote in the affrmative for the Final Order and Opinion of the Commission in this matter. 1 The following abbreviations are used throughout this opinion. RPFF - Respondent Boise Cascade Corporation s Proposed Findings of Fact and Conclusions of Law.

CRRPFF - Complaint Counsel's Reply to Respondent s Proposed Findings of Fact and Conclusions of Law.

IDF - Initial Decision, February 1984 (cited by finding number). BOP - Brief of Petitioner Boise Cascade Corporation tothe United States Court of Appeals for the District of Columbia Circuit datcd August 15 , 1986. ERC - Brief for Respondent Federal Trade Commission to the United States Court of Appeals for the District of Columbia Circuit dated October 15, 1986. Bar Brief of Respondent dated May 18 , 1988. ROC - Brief of Complaint Counsel dated June 15 , 1988. RBR - Reply Brief of Respondent dated June 27 , 1988. - Transcript of July 28, 1988 hearing before the Commission on remand from the Cnited States Court of Appeals for the District of Columbia Circuit. BOISE CASCADE CORP. 957 956 Opinion 1. INTRODUCTION This case, arising under the Robinson-Patman Act ("Act" or " ), 15 U. C. 13 (1988), is before the Commission on remand from the United States Court of Appeals for the District of Columbia Circuit. Boise Cascade Corp. v. FTC 837 F.2d 1127 (D.C. Cir. 1988). At issue is whether evidence in the record overcomes the inference of injury that the Commission drew, relying upon FTC v. Morton Salt Co. 334 U. S. 37 (1948), from proof that manufacturers of office products engaged in substantial price discrimination between competing purchasers over time. Respondent Boise Cascade Corporation Boise ) offered specific evidence at trial about competing dealers to show the absence of any actual or reasonable possibilty of competitive injury. 837 F. 2d at 1144, 1148. The Commission found this evidence irrelevant to the legal issue, but the Court of Appeals, in a split decision, disagreed. (2) The court directed us to determine whether the rebuttal evidence overcomes the Morton Salt inference of injury. Courts and the Commission have in the past identified two types of competitive injury in R-P cases, each gauged by different standards. See, e. , Texaco Inc. v. Hasbrouck 110 S.Ct. 2535, 2543 n. 15 (1990). Primary line or seller level" injury is that which occurs at the level of competitors of the discriminating seller. Evaluation of such injury tends to focus on the use of market power as in cases brought under Section 2 of the Sherman Act. Boise Cascade Corp. 107 FTC 76 , 203 (1986). Secondary line or "buyer level" injury is that which occurs at the level of competing purchasers from the discriminating seller. Courts and the Commission have used a variety of devices to identify the existence of secondary line injury without explicitly defining it. See infra at pp. 8- , 15- 18. Since the disfavored dealers competed with Boise for sales to end-users, the injury at issue on remand is alleged to have occurred in the secondary line. IDF 96- I 00; 107 FTC at 105-07. In order to assess the probative value of Boise s rebuttal evidence we must understand the inference that it is directed at rebutting, the nature of secondary line injury itself. The three opinions from the Court of Appeals panel, together, send mixed signals about the court' readiness to jettison the traditional distinction between primary and 2 The majority opinion "steadfastlysecondary line competitive injury. 2 The opinion of the court was written by Judge Starr. Judge Williams joined in the opinion to form a majority, but wrote separately to express his doubt that the practices complained of could be found to be unlawful on remand. 837 F. 2d at 1148. Judge Mikva, writing in dissent, would have upheld the Commission order. 837 F. 2d at 1163.

Opinion 113 F.

adhered to traditional principles of Robinson-Patman law" in supporting the Commission s use of Morton Salt to establish a prima facie case of competitive injury. 837 F. 2d at 1146 n. 16. The court' s remand however, invites the Commission to blur the distinction between primary and secondary line analysis by directing us to weigh evidence that the Commission previously found irrelevant to the issue of secondary line injury. Nevertheless, we have examined this evidence on the assumption that it is relevant, but find that it is of insufficient probative force to rebut the Morton Salt inference. (3) A. General Background Boise entered the office products industry in 1964 through the acquisition of Associated Stationers Company and the Honolulu Paper 4 IDF 3 , 48. Boise is a dual distributor of office products; itCompany. operates as both a wholesaler (selling to retailers) and as a retailer (sellng to end-users). IDF 5. Boise received a "wholesaler" discount on purchases of office products from six selected manufacturers such that the prices it paid were from 5-33% less than the prices at which these manufacturers sold goods of like grade and quality to competing retail dealers. 107 FTC at 180, 182; IDF 96- 100 , 384. Boise received these discounts on all goods purchased from these manufacturers regardless of whether the specific goods were resold by Boise to other retailers or to end-users. 837 F. 2d at 1132-33. Thus, Boise obtained a wholesaler" discount on goods that it sold at retail in competition with retailers who purchased from the same manufacturers, but received no such favored treatment.

On April 23 , 1980 , the Commission issued an administrative complaint alleging that respondent violated Section 2(f) of the Robinson-Patman Act and Section 5 of the Federal Trade Commission Act, 15 U. C. 45 (1988). Administrative Law Judge Parker rendered the initial decision finding Boise in violation of both statutes on February 14, 1984. 107 FTC 76. Judge Parker made detailed factual findings in route to drawing the Morton Salt inference of competitive injury. These included the intensely competitive nature of the office products industry (IDF 408), the low net profits of the selected dealers (3-4%) (IDF 423-30), the fact that those profits were much lower than the price advantage Boise enjoyed (5-33%) (IDF 313 , 237) on its 3 We need not decide whether the examination of competitive injuryin primary line cases is identical to that under Section 7 of the Chilton Act, or Section 2 of the Sherman Act. Complaint counsel has conceded that it has " not established competitive injury in a primary line sense." Ir at 37. 4 The facts are set forth at length in the Initial Decision. ). ). BOISE CASCADE CORP. 959 956 Opinion substantial and sustained purchases from the six manufacturers (in excess of $10 million in 1979) (IDF 145 , 192, 239, 274, 324, 362), and the fact that Boise took advantage of its discounts to offer its customers better prices or services than those offered by its competitors. IDF 384-406. Judge Parker also concluded that successful efforts by the dealers to underprice Boise were accompanied by a significant impairment of the dealers' profits. 107 FTC at 182. Considering these facts, Judge Parker found that the effect of the substantial and sustained price discriminations favoring Boise "may be to destroy or prevent competition with the unfavored dealers. Id. Boise endeavored to overcome the inference of competitive injury by introducing evidence that the industry as a whole and the business of the disfavored dealers in particular had (4) flourished, that the lower prices Boise received were available to other dealers, that Boise lost accounts to the selected dealers, and that the resale functions Boise performed justified the discounts it received under the doctrine of Doubleday Co. 52 FTC 169 (1955) Doubleday Rejecting these arguments, Judge Parker concluded that the evidence of dealer and industry health showed only "apparent lack of effect on market structure (that) is not appropriate in a (secondary line) Robinson- Patman case. " 107 FTC at 183. Judge Parker further found inter alia that the discounts were not "available" to the disfavored dealers and could not be justified under the competing functional discount standards of either Doubleday or Mueller Co. 60 FTC 120 (1962), ajJd 323 F. 2d 44 (7th Cir. 1963), cert. denied 377 U.S. 923 (1964) Mueller Indeed, the disfavored dealers generally performed the same functions as those that Boise alleged justified its greater discount. IDF 70; 502- 514; 522.

The Commission adopted the findings and conclusions of the Administrative Law Judge. 107 FTC at 201. It rejected respondent' contentions that complaint counsel had failed to lay a foundation for invoking the Morton Salt inference, that actual injury must be demonstrated under sections 2(a) and 2(f) of Robinson-Patman, that competitive injury in a secondary- line case is identical to injury to competition under Section 7 of the Clayton Act, that respondent had a valid availability defense, and that the discounts were protected by the meeting competition and cost justification affirmative defenses. 107 FTC at 202- , 215-22. The Commission found that the inference permitted by Morton Salt established the causal connection between the price differences and the competitive injury. Id. at 208. Relying ), Opinion 113 F.

upon Falls City Industries, Inc. v. Vanco Beverage, Inc. 460 U. 428 (1983) Falls City the Commission observed that "in the absence of direct evidence of displaced sales, this inference may be overcome by evidence breaking the causal connection between a price differential and lost sales or profits. Id. at 435; 107 FTC at 206. We concluded, however, that "Boise does not adduce any such evidence; indeed, Boise does not address the causal connection at all. Instead, it cites evidence that competition ill the industry has not disappeared. 107 FTC at 208. Finally, the Commission viewed Boise s attempt to prove the absence of actual competitive injury as inconsistent with Falls City, 460 U. S. at 434- , which requires only that " reasonable possibilty that a price difference may hann competition be shown. " 107 FTC at 208.

B. The Court of Appeals Decision The Court of Appeals held that the Commission erred by failing to determine whether Boise s evidence demonstrated that "no injury or reasonable possibilty' of competitive injury existed. " 837 F.2d at 1144. While endorsing the Commission s (5) invocation of the Morton 5 the court rejected our view that under Falls City theSalt rule inference of injury may be overcome only by evidence that breaks the causal connection between a price differential and lost sales or profits. The court said that such a rule "defies both logic and the import of Morton Salt that the inference of injury is rebuttable; for if . . . there is no competitive injury (or reasonable possibilty of competitive injury) to begin with, then evidence breaking the causal connection is obviously impossible to adduce. Id. Therefore, the court concluded: 5 The court characterized theMarton Salt inference as "alive and well in the law " (837 F.2d at 1139) and concluded that the "inference was properly relied upon here to establish aprma fac case of competitive injury. Id. at 1146 n.16. The court stated, however, that " (tJhis case is simply not of the lineageMortonof Salt" (id.at n. l4.) and rejected the dissent' s notion that this is a "paradigmatic Robinson-Patman Act case, ld. at 1139 n. 14.

It is true that application of Robinson-Patman to dual distributors that receive functional disr.unts raises an array of complex legal and policy issues. It is also true, however, that courts and scholars have grappled with these issues for years in cases involving application of the Act to functional discounts.See, e. , Texruo lru. v. Ha.sbro1/ck 110 S.Ct. 2535 (1990); Calvani Purutioal Discounts Under the Robinson-Patman Act 17 Indus. & Com. L. Rev. 543, 555. 56 (1976); II E. Kintner & J. Bauer Federal Antitrut Law 22. 14 at 306- 15 (1983) and cases cited therein. In short, there is nothing trail-blazing about the Commission s case against Boise.

6 The parties dispute whether the remand requires the Commission to reconsider its decision that Boise failed to introduce evidence breaking the causal connection between the price discrimination and cumpetitive injury. \Vithout deciding this issue, we select the prudent approach of reexamining the probativeness of the (footnotecont' BOISE CASCADE CORP. 961 956 Opinion In reason, the inference can also be overcome by evidence showing an absence of injury within the meaning of Robinson-Patman. Id. Specifically, the court held that "the Commission s conclusion that Boise s dealer-specific evidence was irrelevant to the inference of competitive injury (was) wrong as a matter of law. Id. at 1144. The Commission s error was that it "simply failed to determine whether Boise s evidence demonstrated that no (6) injury or reasonable possibilty of competitive injury existed. Id. at 1144. "Specific substantial evidence of absence of competitive injury. . . is, in our view, sufficient to rebut what is, after all, only an inference. Id. The court dismissed four categories of evidence which, in its view the Commission erroneously ignored. First, the court referred to dealer-specific evidence supporting what it characterized as the Administrative Law Judge s "finding" that "the selected dealers were not wallowing in a hopeless or deteriorating environment." Id. Second, the court opined that the longstanding nature of the challenged discounts and their apparent lack of measurable competitive effects to date are relevant to determining whether a "reasonable possibility " of injury exists. Id. at 1145-56. Third, the court asserted that the Commission erred by "assuming" the existence of competitive injury without analysis id. at 1146 , noting the absence of evidence that Boise coerced more favorable discounts than those available to many other firms that met the objective definition of wholesaler. Id. at 1147. Finally, the court noted that since buyer liabilty under Section 2(f) is entirely derivative of seller liability under Section 2(a), the six manufacturers stand in violation of the R-P Act although they followed neutral, objective criteria in defining which purchasers were "wholesalers" for the purpose of qualifying for the discount. Significantly, the court declined to specify how the Commission should weigh or evaluate Boise s rebuttal evidence, choosing instead to defer to this agency s expertise. 837 F. 2d at 1145 n. I5. On March 29, 1988 the Commission voted to restore this matter to adjudicative status within the Commission upon issuance of the rebuttal evidence to determine whether it overcomes the inference of injury by either demonstrating an absence of actual injury (or reasonable possibility of competitive injury) or breaking the causaj connection. 7 In addition to the four categories of evidence discussed in the text, the court implied that the Commission erred when it "waved aside (as irrelevantJ substantial evidence (1) that competition among dealers generally was healthy, (2) that the selected dealers singled out for FTC examination were thriving, and (3) that this happy picture of prosperity was apparently unclouded by instances of diverted sales attributable to the challenged discounts," 837 F.2d at 1143- 44, Although we do not construe this language to mean that the Commission should have considered evidence of industry health unrelated to the specific dealers, we have fully evaluated all of the evidence cited by the court. Opinion 113 F.

court' s mandate. The Commission s order afforded the parties an opportunity to submit briefs "discussing whether there is evidence in the record sufficient to overcome the inference of competitive injury found by the Commission in its decision on February 11 , 1986. " The Commission s order further directed that (7) any briefs submitted by the parties address the following issues: (1) the correct theory of competitive injury" by which the Commission should evaluate respondent' s rebuttal case, and (2) the evidentiary effects of invoking and overcoming the inference of competitive injury. The order provided that if any party wishes that the Commission proceed in a different manner on remand, it should submit a motion within seven days from the date of service of the order. Neither party requested within the seven day period that the record be reopened for the submission of new evidence. Accordingly, we decide the case on the existing record in light of the opinion of the Court of Appeals. II. ANALYTIC FRAMEWORK A. The Development of the Morton Satt Inference Section 2(a) of the Robinson-Patman Act prohibits price discrimination that satisfies the jurisdictional prerequisites of the statute only if the effect of such discrimination:

may be substantially to lessen competition or to 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.

15 U. C. 13(a).

Complaint counsel bears the burden of persuasion on the element of competitive injury. See, e. , Sun Oil Co. 55 FTC 955, 976 (1959), rev d on other grounds 294 F. 2d 465 (5th Cir. 1961), rev d on other grounds 371 U. S. 505 (1963); General Foods Corp. 50 FTC 885 890 (1954); cf Falls City Indus. , Inc. v. Vanco Beverage, Inc. , 460 See generally S. at 434- 35 (burden on plaintiff in private action). Kintner & J. Bauer supra note 5 22. 3 at 251-54. Since liability under Section 2(f) is wholly derivative of liability under section 2(a), Great Atlantic Pacific Tea Co. v. FTC 440 U. S. 69, 76 (1976), the Commission must establish competitive injury when it proceeds under either provision of the Act.

The statutory language contains at least two standards for ), BOISE CASCADE CORP. 963 956 Opinion establishing the injury to competition element of an R-P cause of action. First, injury may be established by a substantial lessening of competition or tendency to create a monopoly in any (8) line commerce. 8 This is the test generally applied in primary line cases where competitive injury to the vigor of competition generally and not must be shown. See, e. , Borden Co. v.merely to the plaintiff-seller, FTC 381 F. 2d 175 , 178 (5th Cir. 1967). Second, the Act condemns price differences, the effects of which are to injure, destroy or prevent competition with a customer of the seller. This narrower standard applicable in secondary line cases, safeguards from competitive injury the seller s customers, as distinguished from competition generally. S. at 49; Foremost DairiesId.; see FTC v. Morton Salt Co. 334 U. 382 U.S. 959Inc. v. FTC 348 F. 2d 674, 678 (5th Cir. cert. denied (1965).

Conventional wisdom is that secondary line injury may be established in either of two ways. First, in the absence of evidence of substantial, continuous price discrimination, complaint counsel may establish the requisite injury by presenting a detailed "market analysis." Under such an analysis, the longer the duration of the discrimination, the more competitive the buyers' market, the lower the profit margins of firms in that market, the more homogeneous the product and the greater the similarity between the form in which the product is resold and the form in which it was purchased from the discriminating seller, the more likely the finding of competitive (9) injury. See generally E. Kintner & J. Bauer supra note 5 22. I3 at 304-06 and cases cited in accompanying footnotes; F. Rowe Price Discrimination Under the Robinson-Patman Act at 180- (1962).

8 Some commentators further disting-uish injury that may lessen competition generally from that which tends to create a monopoly. See, e. E. Kintner A Robinson-Patman Primer 21 (2d cd. 1979); H. Shniderman PI'ice Discrimination in Perspective 24 (1977). 9 This standard, added by the 1936 Robinson-Patman Amendments to Section 2 of the Claytn Act FTC v Morton Salt Co. 334 U.S. at 49.50 & n. , focussed on the "effects of a practice on competitive relationships among firms at the reseEer tevel." 107 FTC at 204. For this reason, we have characterized the concern of Robinson-Patman in secondary line cases a competition as fairness" and have concluded that Congress must have intended that this second formulation mean something other than injury to competition as understood in primary line cases and those brought under Section 7 of the Clayton Ad. ld. The Court of Appeals agreed that " it is fairness, as Congress perceives it. that Robinson-Patman is all about." 837 F.2d at 1146-47. Judge Williams, concurring, noted that the courts " have read the Ad to proved certain fairness goals, at the expense of competition." 837 F. 2d at 1149. See also Exxon Corp. v. Gorernor ofivIw-yland, 437 S. 117 , 132-133 (1978) (Act "retlect(sJ a policy choice favoring the interest in equal treatment of all customers over the interest in allowing seHer s freedom to make selective competitive decisions. 1D This "market analysis" approach to establishing secondary line injury must not be confused with a marketstrDcture" evaluation of injury to competition under other antitrust statutes such as Section 7 of the Clayton Act.

), Opinion 113 F. T.

The second method of establishing competitive injury is with evidence that a seller has charged competing customers substantially different prices over a prolonged period of time. See H. Shniderman supra note 8, at 34. In such instances, a long line of cases beginning with the Supreme Court' s decision in Morton Salt allow courts and the Commission to infer the requisite competitive injury under Robinson-Patman.

It would greatly handicap enforcement of the Act to require testimony to show that which we believe to be self-evident, namely, that there is a " reasonable possibility that competition may be adversely affected by a practice under which manufacturers and producers sell their goods to some customers substantially cheaper than they sell like goods to the competitors of these customers. This showing- in itself is sufficient to iustify our conclusion that the Commission s finding's of iniury were adequately supported by evidence.

334 U. S. at 50-51 (emphasis added).

This inferential method of establishing secondary line injury is arguably a particular application of the "market analysis " approach. Under the Morton Salt doctrine, an inference of injury may be drawn only where the price differential is substantial and continuous-two of the factors relevant to secondary injury market analysis. In that case for example, price differentials of up to 18.5% were in effect for decades. The courts of appeals have also permitted the inference to be drawn where the price differential, although small as an absolute percentage, was substantial relative to low profit margins in industries characterized by keen competition. Compare Purolator v. FTC 352 F.2d 874, 879- 80 (7th Cir. 1965), cert. denied 389 U. S. 1045 (4% price differential over number of years unlawful) and Foremost Dairies, Inc. v. FTC 348 F. 2d 674, 680 (5th Cir. cert. denied, 382 S. 959 (1965) (5% price differential over 2 years unlawful), with American Oil Co. v. FTC 325 F. 2d 101 (7th Cir. 1963), cert. denied 377 U.S. 954 (1964) (price differences of 25% to 75% sustained for only 17 days not unlawful). (10) In Morton Salt the Supreme Court reasoned that evidence that a seller s price differentials were "sufficient in amount to influence (competing customers J resale prices" was adequate in itself to support a finding of competitive injury. Id. at 47. Lower courts subsequently held that the inference may be drawn even where there is no difference in the resale prices of competing purchasers since the disfavored buyer may be injured by the impairment of his profits or a ), ), ), ), )). BOISE CASCADE CORP. 965 956 Opinion reduced ability to provide services that attract customers. See, e. Kroger Co. v. FTC 438 F. 2d 1372, 1378-79 (6th Cir. cert. denied 404 U. S. 871 (1971); Foremost Dairies, Inc. v. FTC 348 F. 2d at 680. Reduced profits leave the disfavored firm with "fewer funds available with which to advertise, make capital expenditures, and the like. Black Gold, Ltd. v. Rockwool Indus. , Inc. 729 F. 2d 676 , 681- (10th Cir. cert. denied 469 U. S. 854 (1984), quoting J. Truett Payne Co. v. Chrysler Motors Corp. 451 U. S. 557, 565 n.4 (1981). Although Morton Salt was later interpreted as having created only a "prima facie " inference of competitive injury, Falls City, 460 U. at 435, the cases often inferred such injury from little more than the bare existence of price differentials. Bargain Car Wash, Inc. v. Standard Oil Co. (Indiana), 466 F. 2d 1163 , 1174 (7th Cir. 1972) (Mr. Justice Clark, retired, sitting by designation). Several circuits upheld findings of competitive injury in the face of direct testimony by the alleged victims that they were not, in fact, injured. See, e. United Biscuit Co. of America v. FTC 350 F. 2d 615 , 621-22 (7th Cir. 1965), cert. denied 383 U.S. 926 (1966); Standard Motor Prods. , Inc. v. FTC 265 F.2d 674 (2d Cir. cert. denied 361 U. 826 (1959); Whitaker Cable Corp. v. FTC 239 F. 2d 252 (7th Cir. 1956), cert. denied 353 U. S. 938 (1957); E. Edelman Co. v. FTC 239 F. 2d 152 (7th Cir. 1956); see also National Dairy Prods. Corp. FTC 395 F.2d 517 522 (7th Cir. cert. denied 393 U. S. 977 (1968); Foremost Dairies, Inc. v. FTC 348 F.2d at 680; U. S. Dep t of Justice Report on the Robinson-Patman Act 12 (1977). In Moog Industries Inc. v. FTC 238 F. 2d 43 , 51 (9th Cir. 1956), ajfd per curiam, 355 S. 411 (1958), for example, the court refused to permit a witness by conclusion to deny a mathematical fact" (l1J -that he was competitively injured by paying more for like goods than his rivals paid. I' Cases such as these prompted influential commentators to opme 11 In 1. T1'Utt Payne the Supreme Court actually reserved the question of whether competitive injury may be established in the absence of a showing that the favored purchasers lowered their resal", prices. The outcome of this case, however, would not be altered by the adoption of such a rule since "Boise has used its advantage to underprice competitors on occasion or to offer bettn services than its competitors (F's 384- 406). " 107 FTC at 182.

12 The strength of the Morton Salt inference was illustrated by the opinion of the hearing f'xamincrSmiin Oil Co. 55 FTC at 962 , who explained' It seems self"evident that where a producer is selling a homogeneous product, such as salt, automotive parts or gasoline, where competition is extremely keen among retailers, and where margins of profit or rnarkupsaresrnall, a lower price to one or some ofsurh competing retailers not only " m ay " out must have the effect of substantially lessening competition. The Commission expressly adopted the findings and conclusions of the hearing examiner. 55 ITC at 981. , Opinion 113 F.

that Morton Salt created a virtually irrebuttable inference that price discrimination, particularly in highly competitive markets with low profit margins, was "per se" anticompetitive when the prerequisites for invoking the inference were present. See, e. E. Kintner & J. Bauer supra note 5 22. , at 300; ABA Antitrust Section Monograph 4 The Robinson-Patman Act: Policy and Law 101 (1980). Indeed, the Department of Justice reviewed the case law in 1977 and concluded (tJhe total effect of the majority of secondary line cases is to create a virtually irrebuttable presumption that any price discrimination is injurious to competition. " U. S. Dep t of Justice supra p. 10 , at 14.

The Supreme Court reaffirmed the viabilty of Morton Salt and described the showing respondent must make to overcome the inference of injury some thirty- five years later in Falls City: In Morton Salt this Court held that, for the purposes of (Section) 2(a), injury to competition is established prima facie by proof of a substantial price discrimination between competing purchasers over time r citations omitted). In the absence of direct evidence of displaced sales, this inference may be overcome by evidence breaking the causal connection between a price differential and lost sales or profits. Falls City, 460 U. S. at 435; see also Texaco Inv. v. Hasbrouck, 110 Ct. at 2544. In the instant case, the Court of Appeals agreed that the Commission properly relied upon Morton Salt to find a prima facie case of competitive injury. 837 F. 2d at 1146 n. 16. The issue on remand concerns the showing necessary to overcome this inference of injury. (12) It is difficult to distinguish between those cases in which the Morton Salt inference was deemed overcome and those in which the factual predicate for invoking the inference was not established in the first instance. 3 Many cases in which injury is not inferred from sustained 13 In Chrysler C1'edit Crn-p.I). 1. 7'rit Payne Co. 670 F.2d 575 (5th Cir. 1982), for example, the court found M01"ion Sail was "not relevant to the facts " of thQ case because plaintiff faj!ed to establish "substantial price differences granted to market leaders in a highly competitive market in which minor price differences significantly affected competitors' low profit margins. " 670 F.2d at 581. In finding for defendant. the court identified several reasons for plaintiffs injury unrelated to the price discrimination, including: (1) plaintiff was unable to obtain financing for his used car busir.ess and therefore had to sell his used cars at wholesale; (2) plaintiff chose to forego the new car market to some extent in pursuit of fleet sales; and (3) the amount of the discrimination was relatively small ($11.00 per car). Since1. T?"U/t Payn is the Drily case cited by the Supreme Court in Palls City to illustrate how the Morton Salt inference may be overcome by evidence breaking the causal connection, it js reasonable to assume the Supreme Court intended that such evidence would bc sufficient to overcome the inference even though Morton Salt was not invoked in the Fifth Circuit deci ion. See also Hichard Short Oil Co. v. Texaco. Inc., 799 F. 2d 415 , 421 (8th Cir. 1986) (after finding plaintiff failed to establish a prima facie case underMortO'n Salt court held in alternative that plaintiffs rnanagemer.t and questionable business practices caused alleged injury). , BOISE CASCADE CORP. 967 956 Opinion and substantial discriminations make no mention or only passing reference to Morton Salt. No inference of injury has been drawn in cases where the evidence showed plaintiff's own poor business judgment was the cause of his injury; 14 where the discrimination in issue related to the price of a component part which bore no correlation to the price of the finished product; 15 where a price differential reflected (13) nothing more than a consumer preference for a premium brand over a non premium brand; 16 or where the lowerpriced product was available from the same or an alternate source. See generally, ABA Antitrust Section supra p. II , at 99- 100. Together, these cases loosely illustrate the Falls City principle that the inference will not be drawn where the evidence breaks the causal connection between the price discrimination and any reasonable possibility of competitive injury. See generally, F. Lewis and R.L. Horstman, Competitive Injury Secondary/Tertiary Line, 53 Antitrust L.J. 891 , 898-903 (1985).

The competitive injury element of a Section 2(a) offense has two principal components: (1) the reasonable possibility of injury itself (hereinafter "injury" or " fact of injury); and (2) causation. See BOR at 21 , 32. The Morton Salt rule permits the trier of fact to infer the existence of both components. The Court of Appeals did not distinguish between these two aspects of the prima facie case in stating that the Morton Salt inference-as if it were only one inference-may be overcome either by demonstrating an absence of injury or by breaking the causal connection. The effect of the court' decision would be to permit respondent to overcome the "fact of injury" element of the competitive injury requirement of Section 2(a) 11 See, e. , Richa?'d Short Oil Co. v. Texw:o, Inc. 799 F. 2d at 421; cf Ommi:can Can Co. 1., Russel/1JiUe Canmng Co. 191 F.2d 38 , 60 (Bth Cir. 1951) (illness of plaintiffs manager, his preoccupation with preparing for litigation,difficult climatic conditions and change from a noncompetitive to a competitive market all affected plaintiffs business).

15 Minneapolis-HoneyweU Reqnlation Co. 1!. FTC 191 F.2d 786 , 791 (7th Cir. 1951),errrt. dismL sed, 344 S. 206 (1952): a.cordMarty s Floor Covering Co. v. GAFCorp. 604 F.2d 266 (4th Cir. 1979),cert. denied 444 1;. 8. 1017 (1980); Qitakej' Oats Co. 66 ITC 1131 (1964). J6 In such cases the price difference creates no competitive advantage to the recipient of the cheaper private brand product on which injury could be predicated. Borden Co. v. FTC, 381 F. 2d 175 , 181 (5th Cir. 1967).

17 See, e. , Tn- Valley Paehng Ass n V. FTC 329 F.2d 694 , 703- 04 (9th Cir. 1961);Hanson V. Pittslmrgh Plate Glass Indi(s., Inc. 482 F. 2d 220 , 227 (5th Cir. 1973), cert, denied 414 U.S, 1136 (1974); contm Fowler Mfg. CO. V. H. H. Gorlick 415 F. 2d 1248, 1253 (9th Cir. 1969), cel' t. denied 396 U. S. 1012 (1970); Pw' oiator Pmds, Inc. v. FTC, 352 F. 2d 874 , R82 (7th Cir. 1965), eert. denied 389 U. S. 1045 (1968). 18 Rowe states that a seller may escape liability by demonstrating the absence of a causal connection with evidence of any of the following: " (1) intervening economic factors intluencing a buycl" s resale activities, (2) addeci functions or offsetting costs by the low-price buyer, (3) the competitive inertia of his rivais, and (4) the availability of the goods at the lower price from another source. " Rowe supra p. 9, at 186. Opinion 113 F.

with evidence directed at breaking the causal connection between that injury and the challenged price difference. (14) We have found no authority, in the four decades since Morton Salt for the proposition that one does not incur this fact of secondary line injury despite paying substantially more than a competitor for goods of like grade and quality over a prolonged period in an industry with low profit margins and keen competition. Indeed, injury in this sense appears to be conclusively established in cases where Morton Salt was properly invoked in the first instance. As two commentators put it arguments attempting to show an absence of actual injury to competition, have. . . proved unsuccessful in the face of evidence that the price discriminations were substantial." E. Kintner & J. Bauer supra note 5 22. 13 at 302.

The presence of this " fact of injury, " however, does not mandate a finding of "competitive injury" within the meaning of Robinson- Patman. Falls City makes clear that in the absence of "direct evidence of displaced sales " respondent may escape liabilty by introducing evidence breaking the causal connection between the price discrimination and the injury. The corollary of the Falls City rule is that the Morton Salt inference of competitive injury is conclusiveregardless of evidence of dealer health-where direct evidence of displaced sales is adduced. The only difference between the rule conclusively inferring competitive injury under Morton Salt and the rule of inferring it subject to rebuttal is that in the former case the direct evidence of displaced sales supplies proof of causation independent of the inference. Since secondary line injury may be conclusively inferred where there is direct evidence of displaced sales Falls City suggests that this "fact of injury as distinguished from causation-is established by the same facts necessary to invoke Morton Salt in the first instance. Thus, a reading of the Court of Appeals decision to allow respondent to overcome the Morton Salt inference with evidence of the absence of this "fact of injury" would be inconsistent with the Supreme Court' s teaching in Falls City. B. The Standard for Overcoming the Morton Sale Inference The Court of Appeals held that the Morton Salt inference "can " be overcome by evidence showing an absence of competitive injury within the meaning of Robinson-Patman. " 837 F. 2d at 1144. Future interpretations of this phrase will ultimately determine whether the court' s decision wil lessen the evidentiary showing necessary to BOISE CASCADE CORP. 969 956 Opinion overcome the Morton Salt inference. The court eschewed articulating a standard for assessing secondary-line competitive injury despite Boise s urging that the court adopt a market-structure approach (15) comparable to that used in primary line cases or under other antitrust statutes. 837 F. 2d at 1145 n. I5. BOP at 27- There are at least four approaches to assessing secondary line injury that are discussed in the record or by the Court of Appeals. First is the view that secondary line injury should be evaluated by the same standards employed in merger analysis under section 7 of the Clayton Act or monopolization cases brought under section 2 of the Sherman Act. This approach-characterized by Judge Richard Posner as a "polar extreme" view-was taken by Boise below and, to some See BOR at 54-56. R. Posner Theextent, again on remand. Robinson-Patman Act 39 (1976). Whatever mixed signals may exist in the Court of Appeals s opinion, its decision cannot be read as expressly or impliedly reversing the Commission s previous rejection of this theory. 107 FTC at 202-04.

The second, and polar opposite view, would permit competitive injury to be conclusively inferred from a price difference. The premises of this theory is that any firm that (16) pays more than its rivals for goods that it is trying to resell in competition with them is supra p. 15, at 39. Althoughcompetitively disadvantaged. R. Posner some previous decisions by courts and the Commission have arguably approached adoption of this standard in secondary line cases, we do not today. 21 Such a rigid approach has been criticized because it Jg The court expressly adhered to the incipiency standard for measuring competitive injury, but did not otberv.'ise define the nature of secondary-line injury within the meaning of the Act. 837 F.2d at 1139 n. 1144. It acknowledged that the Commission rejected Boise s effort to " muddy the Robinson-Patman waters by implanting market-structure analysis in alien legal soil" 107 FTC at 202 n.3; 837 F. 2d at 1145 D. 15. The court stated that Judge Parker did not co- mingle "market-structure analysis appropriate under other provisions of the Clayton Act, such as Section 7 . . . but, as the Commission claims, off-target as to Robinson Patman s less global perspective. " 837 F.2d at 1145 n. 15. Similarly, the court observed that " (tJhis was no open-ended Sherman or Clayton Act case going broadly to the structure of the market or ' market dynamics Id. at 1146 n. 16. In any event, the court s "quarrel with the Commission" was that "the FTC pruceeded deliberately to ignore. ,. fact-specific evidence going directly to the Commission-annointed (sic) 'victims: name\y the selected dealers. ld. at 1146 n.16. The fact that the "target(sj" of Boises evidence were the selected dealers themselves and their specific circumstances in the competitive arena with Buise id. however, provides Ettie additional guidance as to the type of dea1cr-specific evidence that would demonstrate an absence of competitive injury.

20 At oral arR'ment, respondent's counsel initially argued that the Court of Appeals obliterated the distinction between primary and secondary line injury, then retreated from this position. Tr. at 16. Boise stated the Court of Appeals is telling us that the focus of the statute must be injury to the competitive process. " Tr. at 18- 19.

21 Foremost Duij-fes, Inc. 1i. FTC 348 F. 2d at 678, for example, has been interpreted as holding that threatened injury to an individual plaintiff that lacks the capacity to diminish competition generally is unlawful. See O'Connell v. Citms Bowl, Inc. 97 F.R.D. at 122 n.3 (without expressing an opinion as to the (footnutecont' Opinion 113 F.

reads the competitive injury requirement right out of the statute (id. and would condemn sporadic price discriminations that occur in the process of a market adjusting to a new equilibrium. Id. at 40. See also id. at 12-14.

The third approach, that Robinson-Patman condemns only those practices that constitute "economic price discrimination " is advanced by Judge Wiliams in the concurring opinion. Under this view, liability is appropriate only in those cases where the price difference "causes disfavored purchasers to bear a disproportionate share of the seller total costs of production. " 837 F.2d at 1149. Judge Wiliams argues that economic price discrimination could not occur in this case due to the absence of barriers which prevent arbitrage- resales by favored customers to disfavored ones. Id. at 1149-50. The cases hold, however, that the Robinson-Patman Act diverges from a law against economic price discrimination in two (17) fundamental ways. 22 First, discrimination under the Act is "merely a price difference" and is distinct from the economist's notion of economic price discrimination. Texaco Inc. v. Hasbrouck 110 S.Ct. at 2544; Falls City, 460 U.S. at 443 n. l0; FTC v. Anheuser-Busch, Inc. 363 U.S. 536 , 549 (1960). The Act does not proscribe economic price discrimination that exists where a seller charges two purchasers the same price although the seller s cost of supplying one of the customers is substantially lower than his cost of supplying the other. Sano Petroleum Corp. v. American Oil Co. 187 F. Supp. at 353-54; E. Kintner & J. Bauer supra note 5 21.7 at 167; R. Bork The Antitrust Paradox 391 (1978).

Second, the limited range of evidence that courts and the Commission have considered to support a cost justification defense has arguably led to the condemnation of practices which were cost justified in an economic sense. See R. Posner supra p. 15 , at 42. See also 837 F. 2d at 1149 (J. Wiliams, concurring) (Act condemns efficiency enhancing practices). Indeed, in Morton Salt itself, the correctness ofForemost). Arguably, such an approach is also inconsistent withFalls City to the extent that it conclusively infers that the price discrimination caused the disfavored purchaser s injury. 22 Economists teach that in order for a seller to profitably price discriminate by charging differem prices, the following three conditions must exist: (1) the seller must have monopoly power; (2) the seller must be able to segregate its customers into classes that have dther different price eJasticitjes of demand or varying reservation prices; and (3) opportunities for arbitrage must be constrained. F. Scherer lndustr-ial Market Stl"uctuTe and EconOlnW PeTfo1'name at 315 (2d ed. 1980). 23 The socia-political concerns that led Congress to enact Robinson-Patman arc we\! documented elsewhere and need not be repeated here. See, General Matrrs Corp. 103 FTC 641 , 693-96 (1983); E. Kintner & J. Bauer supra note 5 19. 1 et seq.; R Posner Sllpra p. 15 , at 25-29; F. Rowe supra p. 9, at 3-11. BOISE CASCADE CORP. 971 956 Opinion Seventh Circuit embraced a variant of this economic price discrimination theory only to be rebuffed by the Supreme Court. The fourth approach to secondary line injury is less a definition than a description of the conditions under which evidence of injury to competitors wil support a finding of liability absent an affirmative 25 It is consistent with (18) the Supreme Court' s view that thedefense. injury to1936 Amendment "was intended to justify a finding of competition by a showing of 'injury to the competitor victimized by the discrimination. Morton Salt 334 U. S. at 49; accord Hasbrouck v. Morton Salt it has beenTexaco, Inc. 842 F. 2d at 1041 (since permissible to infer harm to competition from evidence of harm to competitors. ) (emphasis in original).

Respondent proposes a narrow variation of this theory of competitive injury on remand. It argues that secondary line injury occurs only where a price differential, if continued, threatens the disfavored dealer s "existence" (Tr. at 18) or "competitive viabilty, thereby raising the concern of incipient injury to competition. " BOR at 56. But this test is more restrictive than that applied in Morton Salt which the Court rejected a defense premised on the small amount of sales for which table salt accounted relative to the total stock of the grocery store-purchasers. 334 U.S. at 49. The Court recognized that since a grocery store consists of many comparatively small items there is no possible way effectively to protect a grocery store from discriminatory prices except by applying the prohibitions of the Act to each individual article in the store. Id.

Our review of the case law and the arguments of the parties suggests the following definition-also a variation of the fourth approach. Secondary line competitive injury is that injury which flows from the practice of a seller charging two or more customers different 24 The Court of Appeals in Morton Salt defined unlawful price discrimination as follows: "It is the Art of unfairly, injuriously and prejudicially distinguishing between persons or objects, where economically speaking a sound and fair distinction does not exist." 162 F.2d 949 , 954-55 (7th Cir. 1947). 25 Cf JP Feeser, Inc. v. Serv- Portion, Inc., 7 Trade Reg. Rep. (CCH) (1990-2 Trade Cas. 124 at 173 n.10 (3d Cir. Aug. 2, 1990) ("(eJvidence of injury to a competitor may satisfy the component of competitive injury necessary to show a violation of the Robinson- Patman Ad" in a secondary line case); Hasbrow:k 1i. Texruo, Inc. 842 F. 2d 1034 , 1040 (9th Cir. 1988), aifd 110 S.Ct. 2535 (1990) (" (cJlearly, injury to competitors may be probative of hann to competition, although the weight to be attached to such evidence depends on its nature and on the nature of the challenged conduct." 26 In RifhaTd Short Oil Co. 1i. Texaco, bu:. 799 F.2d 415, 420 (8th Cir. 1986), the principal case upon which Boise relies for this proposition, the court found plaintiff failed to prove the elements necessary to invoke MOJ'ton Salt in the first instance.Id. at 420-21. Thus, the passage in Short that "injury to competition focusses on whether there has been a substantial impairment to the vigor or health of the contest for business is mere dictum. Moreover, it is dictum which the Eighth Circuit derived fromBrunswick Corp. v. P1(cb1o Bowl- Mat, Inc. 429 U. S. 477 , 488 (1977), a case brought under Section 7 of the Clayton Act which, as we previously have held, is not the controlling standard in this case. Opinion 113 F, prices for goods of like grade and quality, where the price differential is (1) substantial in amount 27 and (2) sustained in duration 28 (3) such that it (19) creates a reasonable possibility, 29 (4) that the favored purchaser would benefit from the lower price by either (a) lowering its resale prices in competition with the disfavored purchaser to attract customers 3O (b) using its increased gross margins to enhance its services which may attract customers, 31 or (c) enlarging its profit 32 margins over what they would have been without the discount. See supra at pp. 9- 11.

The Court of Appeals stated that the Morton Salt inference can be overcome by evidence showing an absence of injury within the meaning of Robinson-Patman. In our view, this would require the respondent to produce evidence that would preclude a finding that complaint counsel met its burden of proof on one or more of the four elements above as follows. 33 Evidence that negates a finding of either of the first two elements-that the price differences were (1) substantial, or (2) sustained-would overcome the Morton Salt inference. 34 Under those circumstances, the complaint should be dismissed unless complaint counsel makes its case under the more stringent market analysis approach. See supra at pp. 8-9 & note 10. The third element, causation, is (20) established through the Morton Salt inference from proof of the first two elements, and may be overcome, in the absence of direct evidence of displaced sales, with evidence breaking the causal connection between the price discrimination and injury. Finally, respondent may not overcome the Morton Salt inference with evidence negating one or two of the benefits that may inure to the favored purchaser as described in the preceding paragraph. Rather, the evidence would have to preclude the Commis- 27 Palls City, 460 U.S. at 435; Morion Salt 334 U. S. at 50- 51. Faits City, 460 U. S. at 435, Id.

30 M01"ton Salt 334 C. S. at 47.

3J See, e. , Kroger v. FTC, 438 F;2d at 1378- 79; Foremost Dairies, 1m:. v. FTC, 348 F. 2d at 680; Blad Gold, Ltd. v. Rockwoo/ 1111(8., Inc. 729 F.2d at 681-82, 32 See supra note 31. Increased profits confer a competitive advantage on the favored purchaser in several ways, such as by enabling to make greater capital expenditures than its competiwrs or to better weather a decline in the business cycle.

3:, Since the burden of persuasion on the issue of competitive injury ultimately rests with complaint counsel respondent cannot be required to show the absence of any element that goes to the question ofMorton Salt injury. Respondent should be deemed to have overcome the inference of injury if it has presented sufficient evidence frorr. which the trier of fact finds that the essential ele ments of injury haw not been established by a preponderance of the evidence.

Although Hobinson-Patrnan is not a law against economic price discrimination, these two elements operate to exclude from the grasp of the Act many price differences that would not constitute such discrimir.ation. See R. Posner wpm at p. 15 , at 39-40; Boise Cascade Cor. 107 FTC at 184. BOISE CASCADE CORP. 973 956 Opinion sion from finding a reasonable possibility that the price difference would enable the favored purchaser to enjoy any of the benefits previously identified: the ability to offer lower resale prices, to offer new or enhanced services, or to increase its profits. We recognize that such a standard will likely prove difficult for respondents to meet. This is no doubt why neither the court nor Boise cites a single case in which the Morton Salt inference, once properly invoked, was overcome on the ground that no " injury in fact" was demonstrated. C. Morton Salt Permits An Inference of Competitive Injury To Be Drawn from Predicate Facts Respondent argues that the Morton Salt inference" is in fact two presumptions one of injury and the other of causation-which may be overcome by introducing evidence sufficient to raise a genuine issue as to the existence of either presumed fact. BOR at 21, 32 and 9. Upon the production of such evidence, Boise suggests that the burden shifts back to complaint counsel to produce additional evidence sufficient to prove the previously presumed facts without the benefit of the presumptions. BOR at 30- 31. See IX Wigmore Evidence 92491 at 305 (Chadborn rev. 1981) (hereinafter "Wigmore supra Complaint counsel counters that Morton Salt does not create any rebuttable presumption, but merely permits the fact finder to (21) draw a permissive inference 37 of injury from proof of the facts that gave rise to the inference. Our view is that Morton Salt and its progeny stand for the proposition that proof of substantial price discrimination between competing purchasers over a competitively 35 A presumption oftenealled 0. rebuttable prescription " is "a ninfereneewhichthelawdirectstheltrierof facts to draw if it finds a given set of facts. Legme 1. Dann 544 F. 2d 1 , 5 n. 24 (D.C. Cir. 1976). It is a procedural rule which compels a factual conclusion in the absence of contrary evidence. Id. % This approach-commonly known as the " bursting bubble" theory-was pioneered by the late Professor Thayer and is the choice of leading scholars and most federa: courts today.Legille v. Da. 544 F.2d at 6see ger"erally H. Graham lIandbook of Federal Evidence 301. (2d cd. 1986); Fed. R. Evd. 30l. 31 "An inference is aconctusion as to the existence of a particular fact reached by considering other facts in the course of human reasoning. " M.H. Graham supra note 36, 301.7 at 108. An inference permits, but does not require, the trier of fact to deduce the existence of one fact from proof of another by using ordinary reasoning and logic. Id. See a/so E. Cleary, McCorrnick all Evidence !i342 at 965 (3d ed. 1984);Legille v Dann 544 F. 2d 1, 5 n.24; Bray v. Uni/ed States 306 F.2d 743, 747 (D.C. Cir. 1962). In some instances " inference may be so strong that no other conclusion may reasonably be reached. " M.H. Graham supra note 301.7 at 108. This is because of the "compeiling nature of the particular factual circumstances rather than that, as is the case with a presumption, a rule of law requires the conclusion to be drawn. Id. Assuming, however, that MOJ'ton Salt creates a rebuttable presumption . complaint counsel argues that the requisite competitive injury may be inferred from the evidence which gave rise to the presumption in the first instance. BOC at 22. The Commission "must lthenJ weigh the disputed rebuttal ev;dence to detennine the strength of any inference flowing therefrom, and then it must compare the strength of that inference against the strength of the MOTton Salt inference." BOC at 19 , ); , ). Opinion 113 F.

significant period of time is sufficient evidence from which a logical inference of competitive injury may be draw. The opinion of the Court of Appeals is consistent with our analysis. The court stated sustained and substantial price discrimination raises an inference but it manifestly does not create an irrebuttable presumption of competitive injury. " 837 F. 2d at 1144 (emphasis added). The court' s characterization of the Morton Salt rule as an inference, rather than a presumption, is supported by two score years of case law. 39 Indeed, a leading (22) Robinson-Patman Act scholar noted a quarter of a century ago that " Mr. Justice Black's opinion (in Morton Salts sanctions but does not compel conclusions of competitive injury from minimal evidence in other factual contexts. " F. Rowe supra p. 9 , at 135.

Against this plethora of authority, 40 Boise fails to identify a single case holding that the "inference " of injury established by Morton Salt and its progeny is a bursting-bubble type "presumption. " Instead, it relies upon language in Falls City, 460 U.S. at 435 , suggesting that the Morton Salt inference establishes only a "prima facie" case of competitive injury. Boise also points to language in the opinion of the Court of Appeals explaining that once complaint counsel successfully invoked the Morton Salt rule the burden therefore shifted to Boise to rebut the FTC's prima facie case by breaking, if it could, the causal connection. " 837 F. 2d at 1146 n. I6. Boise s review of the case law leads it to conclude that courts have "uniformly " applied the Morton Salt inference in practice as a legal presumption. BOR at 28- 29; RBR at 9.

Language in the cases referring to complaint counsel's obligation to present a "prima facie " case of injury does not establish that Morton Salt creates a rebuttable presumption of injury. The term "prima facie " is "often used in two senses and is therefore an ambiguous and 39 See, e. , Hasbrouck v. Te:raeo, Inc. 842 F. 2d at 1011 affd 110 S.Ct. at 2544; National Dairy Prods. Corp. v. FTC, 395 F. 2d al 521 (facts supporting invocation of Marlon Salt are "adequate to support the Commission s finding that the effect of National's price discriminations might be substantially to injure competition among retail stores Fonmosl Dai -ies, 1m:. v, FTC 348 F.2d at 679-81 (court must defer to the Commjssion "the task of drawing the inference of probable injury to competitors so long as that inference is supportd by a reasonable quantum of evidence in the record" In Rose Confections. Inc, v. A morosia Chocolate Co. 816 F. 2d 381 (8th Cir. 1987), the Eighth Circuil recognized thatMorton Salt permits the lrier offaclto draw an inference ofcompctitive injury, but does not require that such injury be found. The court upheld an instruction to the jury that it "may infer that such price differentia! lessened competition, " The court held that "the instruction as given adequately informed the jury that it need not draw, the Morton Sail inference.. . . Thus, the instruction permitted the jury to find that there was no jn jury to competition despite Rooe Confection Mm.ton Salt evidence. Jd. at 388-89. The Eighth Circuit expressly affrmed the district court' s refusal to instruct the jury that the M01.to-n Salt inference is rebuttable.Id, at 388. 40 See supra note 39.

BOISE CASCADE CORP. 975 956 Opinion often misleading term. " E. Cleary, supra note 37 342 at 965 n.4; see also Texas Dep t of Community Affairs v. Burdine 450 U. S. 248 254 n. 7. Burdine The phrase 'prima facie' case not only may denote the establishment of a legally mandatory, rebuttable presumption, but also may be used by courts to describe the plaintiff' s burden of producing enough evidence to permit the trier of fact to infer the fact at issue. "). Boise evidently ascribes only one meaning to itdescribe the evidentiary showing necessary to shift (23) the burden of production. RBR at 9. The term also may mean evidence simply sufficient to withstand a motion for a directed verdict. E. Cleary, supra note 37 342 at 965 n.4. Thus, the statement in Falls City that the Morton Salt inference establishes a "prima facie" case is inconclusive. In addition, language in the Court of Appeals' opinion stating that invocation of Morton Salt shifted to respondent the burden "to rebut the FTC's prima facie case " only points to the strength of the inference which, through logic and reasoning rather than by operation of law, may result in a finding against respondent in the absence of rebuttal evidence. See B. Graham supra note 36 301.7 at 108.

Even assuming, however, that Morton Salt creates "presumptions of injury and of causation and that respondent has rebutted those presumptions, the underlying inferences which may be drawn from the facts that gave rise to the presumptions remain. H. Graham supra note 36 30I.2 at 113; cf E. Cleary, supra note 37 344 at 975 (natural inference underlying presumption may be sufficient to take a case to the jury, despite the existence of contrary evidence and the resultant destruction of the presumption). There is no merit in Boise s suggestion that upon introduction of rebuttal evidence complaint counsel must produce additional evidence of competitive injury. BOR 30- 31. Indeed, the Title VII cases from which Boise argues by analogy hold that once a presumption of discriminatory intent has been rebutted, courts may continue to rely upon the evidence introduced by plaintiff to establish its prima facie case and upon inferences properly drawn therefrom in determining whether respondent' s explanation for the challenged conduct was "pretextual." Burdine 450 U.S. at 255 n. IO (plaintiff's initial evidence 41 Boisc s reliance on a series of age and sex dis('rimination cases that discuss the burden of proof once plaintiff makes a prima facie case of discriminatory treatment is wholly misplaced. BOR at 26- , 29-30 & n. See, e, , Elliott 1). Gr01(p Medj cal S1(Tg1:cal SeJ' 714 F. 2d 556 (5th Cif. 1983) and Love/ace 'f. Shenvin- W1Uio;ns Co. . 681 F. 2d 230 , 239- 40 (4th Cif. 1982) (cases under Age Discrimination in Employment Act, 29 C. 621 et seq.) Nothing in these cases augers in favor of oor construingMm' ton Salt to create a rebuttable presumption rather than to permit an inference of injury tu be drawn );

Opinion 113 F.

combined with effective cross-examination of the defendant, may 12 (24)suffice to discredit defendant' s explanation). In our view, the evidence of sustained and substantial price discrimination in the record is sufficient to permit us to infer the existence of competitive injury within the meaning of Robinson- Patman. This inference is bolstered by testimony that the selected dealers lost accounts to Boise because of its better prices or services and the evidence that Boise enjoyed greater net profits than dealers generally and the selected dealers in (25) particular. Respondent' rebuttal evidence is of insufficient probative value, for the reasons explained infra to overcome our drawing an inference of actual injury; afortiori said evidence does not establish any fact from which we may infer the absence of a reasonable possibility of competitive injury in this case.

III. ANALYSIS OF REBUTTAL EVIDENCE The Court of Appeals directed the Commission "to determine whether Boise s evidence demonstrated that no injury or ' reasonable possibility' of competitive injury existed. " 837 F.2d at 1144. The court identified four categories of evidence for us to consider on remand, but did not draw inferences from the facts previously found by the 42 The other cases Boise cites rlo not support its assertion that com plaintcounset' sevidence is insufficient to warrar1t an inference of competitive injury in the face of contrary evidence Boise produced. BOR at 28, 30- 31; RBR at 11. See, e, . Lovelacc!'. Shenc1n- Williams Co. 681 F. 2d at 230 239- 244-45 (4th (ir. 1982) (detailing the evidentiary effects of rebmtable presumptions operable in cases brought under Title VI and the Age Discrimination in Employment Act); Scott Medical Supply Co. 11.Bedsole S1l"gical Slcpplies, Inc., 488 2d 934 , 937- 38 (5th Cir. 1974) (inference of conspiracy cannot stand in the face of "substantial and uncontradicted evidence to the contrary Comfol.t Tmne Air Condi /i01l1 ng Co. Tmne Go. , 592 F. 2d 1373 1384 (5th Cir. 1979) (" overwhelming" evidence of business justification precludes finding of liability in the absence of "substantial" evidence). As explained in Part II info respondent' s evidence is underwhelming ir. the face of complaint counsel's substantial evidence of competitive injury. Equally misplaced is Boise s reliance on a line of cases counseling aliainst drawing inferences of conspiracy or anticompetitive intent from ambiguous evidence in the face of contrary business justification/independent action evidence in antitrust actions under the Sherman Act, 15 U, C. 1 et seq. (1982). BOR at 48- 50. See, e. RHS1ness Elecs. Corp. r. Sharp Elers. COI-p. 106 S. CL 1515 (1988); Mats1!shita Elec htdlls. CO. V. Zenith Radio Corp. 475 LS. 574 (1986); T.IV t;/ec.Snll. lne. Pac1j'1c Elee. Co-ntl oc/m' s Ass ' 809 F.2d 626 632 (9th Cir. 1987); Smith V. ,1I'orthem Mich. Hasps. , Inc. 703 F. 2d 942, 956 (6th Cir. 1983); Pano/ex Pipehne Co. V Philhps PetroleHln Co. 457 F. 2d 1279 , 1289 (5th Cir. 1972);H.L. Ha,yden Co. Siemens Medical Sys. J;1C.672 F. Supp, 724 , 733, 39 (S. y, 1987). The Morton Salt inference is a breed apart; its purpose is to simplify the burden on the Commission while roughly distinguishing between harmless and heinous price differences. We reject Boise s assertion that the reasoning of these cases applies here because the Supreme Court admonished against overbroad interpretations of R- P in A1damatie Canteen CO. V. FTC, 346 U. S. 61 , 63 (1953). Nothing into((omatJc Canteen permils us to read Robinson-Patmar. more narrowly than required by the fOllr corners of the Act. 837 F. 2d at 1158-59 (J. Mikva, dissenting). Boise s argument is sirr.ilal' to Morton Salt s assertion that the facts in that case were "equally reconcilable with an inference in favor of innocence of wrongdoing and the commission of the offense; thus, the inference of innocence should prevail 162 F.2d 949, 956 (7th Cir. 1947). The argument is no more persuasive :0 us on the facts of the case than it was to the Supreme Court on the facts of MortOiI Salt BOISE CASCADE CORP. 977 956 Opinion Commission. See Corn Prods. Co. v. FTC 324 U,S. 726, 739 (1945); FTC v. Algoma Lumber Co. 291 U. S. 67 , 73 (1934). We have reviewed the record in light of the court' s holding, and have carefully weighed the probative value of the evidence noted by the court against the Morton Salt inference. Pursuant to the framework set forth in the previous sections, Boise may overcome the inference of injury and prevail on remand by demonstrating that there is no reasonable possibilty that the discounts enabled Boise to lower its resale prices, provide enhanced services to attract customers, or increase its profits. As explained below, Boise s evidence is of insufficient probative value to overcome the inference of competitive injury based on the size and duration of the price differentials. A. Evidence of Dealer Health The court criticized the Commission for ignoring evidence that the selected dealers were not wallowing in a hopeless or deteriorating environment. " 837 F.2d at 1144. Specifically, the court directed the Commission to consider the following evidence: (1) the selected dealers' sales and gross profits increased by 22% during the period in question despite a recessionary economy, id. at 1135; and (2) evidence that "switches (of customers J to Boise apparently cannot be explained as sales diverted through (26) operation of the wholesaler discount." 837 F. 2d at 1145. 44 The court stated that the Commission "mischaracterizes Boise s evidence as going to market structure, rather than to 43 Throughout its briefs, respondent arg1€S that the Court of Appeals made factual findings which are now binding on the Commission. See, e. BOR at 21-22; HER at 6. Boise even asserts that the Court of Appeals found "an absence of competitive injury, an issue inextricably entwined with the remand itself. At orai argument, however, respondent virtually conceded that the Court of Appeais did not set aside any of the Commission s factual findings for want of substantial evidence.Te. at 8. Thus, we review the c,idence in the context of the opiniun of the court.

The court also pointed to evidence that credit ratings of the selected dealers were favorable "on the whole" and that those companies that did experience a change in their credit rating from 1977 to 1980 found that it "generally improved. " 837 F.2d at 1135. The court did not, however, direct the Commission to review this evidence of changing credit ratings on remand, The evidence supporting the contention that the dealers generally experienced improving credit ratings is very weak. Finding 432 quotes from the testimony of Boise s expert, Dr. Elzinga on this issue But overall, the a sessment of a number of the r.companies had not changed. Two or three, I think, had gotten worse, eight or nine lout of 23 total), I believe is the number you will find . that the credit ranking . was assessed by D & B to be up. IDr 432. Significantly, Dr. Elzinga saw credit reports for just over half of the selected dealers. He asked to see credit reports "for each" of the selected dealers, IDF432, but reviewed such reports for only thirteen dealers in August, 1981 and fourteen of the 23 selected dealers in January, 1983. Elzinga tr. 6267- 68. Moreover, Dr. Elzinga testified that he did not think that the credit report,s were broken down in a manner that would assist in evaluating the perfonnance of various segments (printing, retail . etc.) of the dealers ' businesses. ld. Thus his testimony cannot support a finding connecting the status of the dealers' credit to the health of their competitior. wit.h Boise.

Opinion 113 F.

the twenty-three selected dealers themselves, when Boise, in fairness Id. at 1145 , n. 15. On the other handadduced evidence as to both. the court noted that evidence that the median net profit of the dealers may have been lower than that of dealer-members of the National Office Products Association ("NOP A") as well as Boise s combined wholesaler/dealer profit also deserve attention. Id. at 1144. Each of these categories of evidence are discussed below. 1. Growth Evidence Following the court' s lead, Boise argues that evidence of the selected dealers' increasing sales and gross profits- that which we wil collectively refer to as "growth" evidence-demonstrates the general health of the selected dealers and overcomes the inference of injury. Although the court stated (27) this evidence "tends, if anything, to point to an absence of ' lost profits' of the sort described in Falls City as one possible manifestation of injury to competition the court expressly did " not decide the matter. " 837 F. 2d at 1144 (emphasis added). Indeed, the court noted only that the growth evidence "merited consideration by the Commission rather than the out-of-hand dismissal that it received. Id. Falls City is distinguishable. There the Court was concerned with evidence of lost profits from sales of one specific product-Falls City beer. 460 U. S. at 433 n.4, 436 n.7. Boise offered no evidence of the sales and profitability of the selected dealers on the products bought butfrom the six selected manufacturers at discriminatory prices, instead only evidence of the dealers' overall sales and profitability on all products, regardless of their source. Since the selected dealers resell between 3 500 (IDF 504) and 18 000 (IDF 510) products made by hundreds of manufacturers, their overall sales and gross profits are of limited value in assessing whether the dealers suffered competitive injury in connection with their purchase and resale of the specific products of the six manufacturers in question. Moreover, these aggregate sales and gross profit figures include data on sales of merchandise other than "office products" as the market has been defined in these proceedings.

Morton Salt and its progeny make clear that the Robinson-Patman 45 In maintaining that only gross (rather than net) profit figures are relevant to the issue of competitive injury, Boise' argues that the net profit figures for the dealers afe unreliable because they are based on data that includes sales and profilson products that are not defined as "office products" in this proceeding. BOR at 361'.. 10. Boise cannot have it both ways; if net profit figures are unreliable because they include products not at issue, then the ";r.crcascd" gross profit figures Boise cites should be disregarded tuo. BOISE CASCADE CORP. 979 956 Opinion Act must be applied to "each individual article " sold to competing purchasers at discriminatory prices. 334 U.S. at 49. It is of no import whether such goods constitute a major or minor portion of the dealer stock. Id. ; accord Gold Strike Stamp Co. v. Christenson 436 F. 791 797 (10th Cir. 1970). This reasoning applies here to dealers that sell thousands of office products to end-users just as it did in Morton Salt to grocery stores that sold thousands of products to consumers. The present case is much more like Morton Salt than Falls City. Boise attempts to distinguish Morton Salt on the ground that respondent in that case introduced no evidence of the absence of competitive injury. RBR at 23 n. 5. In fact, however, evidence was introduced in Morton Salt that showed "substantial increases in sales to all non-discount customers in all trade areas for the entire period (28) covered by the Commission s evidence " which the Court of Appeals credited as rebutting the inference of injury. 162 F. 2d at 957. Contrary to respondent' s suggestion Morton Salt imposes no obligation upon complaint counsel to present evidence of the selected dealers ' sales or profits solely on that portion of their business corresponding to the product market in this case. RBR at 22-23. Since Boise failed to demonstrate any connection between the selected dealers' overall growth and their performance in the relevant product market, the evidence of overall sales and gross profit growth is not persuasive. 46 Respondent states that it is "the dealers' overall competitive health and financial viability that is important to a determination of whether competition has been injured. " RBR at 23. The crux of this argument is that evidence of increasing sales by disfavored purchasers precludes a finding of competitive injury. BOR at 64-65; RBR at 21-22. Boise fails to appreciate, however, that the probative value of such evidence wil vary depending on whether a given case involves alleged primary or secondary line injury. Evidence that competitors of a discriminating seller experienced increased sales has been deemed significant in some primary line cases see Borden Co. v. FTC 381 2d 175, 179 (5th Cir. 1967); Minneapolis-Honeywell Regulator Co. v. FTC 191 F. 786 , 790 (7th Cir. 1951), cert. dismissed 344 U.S. 206 (1952), but evidence of new entry and sales volume growth by competitors of a favored buyer have been ignored by courts in assessing secondary line ;6 For example, although total saies by one of the dealers, A. Pomerantz & Co. , more than doubled from $14. 6 miJlionin 1977to$36. 3 million in 1980 the bulk of that increase was accounted for by a three-fold rise in sales by its furn:ture rJepartmcr,t from $8.3 to $26.4 million. Bertholdt tr. at 5891- 92. Thus, during this period furniture sales as a proportion of total sales for Pomerantz increased from nearly 57% to 73%. , .. 980 FEDERAL TRADE COMMISSIO DECISIONS Opinion 113 F.

injury. Cf FTC v. Morton Salt Co. 334 U. S. 37 rev 162 F.2d at 957 (injury found despite evidence of increased sales to non-discount customers); J.F Feeser, Inc. v. Serv- Portion, Inc. 7 Trade Reg. Rep. (CCH) (1990-2 Trade Cas. ) '169 124 at 64 168 (noting district court' s rejection of defendant's argument that overall competitive health of plaintiff and its market precluded finding of secondary line competitive injury). 47 (29) Further eroding respondent's claim that evidence of increasing sales and profits precludes a finding of secondary line injury are the cases in which such injury was found despite testimony from disfavored discrimination. Seedealers that they were not injured by the price supra at pp. 10- 11. These cases are consistent with Judge Parker earlier finding that the dealers would have earned even greater profits and experienced more growth had they enjoyed the same pricing that the six manufacturers afforded to Boise. 107 FTC at 184. Boise asserts that a finding of competitive injury on this basis would mere "speculation." RBR at 24. Relying upon the court' s statement that the growth evidence may point to " an absence of 'lost profits' of the sort described in Falls City as one possible manifestation of injury to competition " 837 F. 2d 1144 , RBR at 21 , Boise asserts that the sole purpose" of the Act is to prevent competitive injury. RBR at 24. Respondent argues that the dealers are " competitively healthy therefore the prospect that they might achieve even greater success" if accorded the favorable discounts "proves nothing. " RBR at 24.

We reject Boise s suggestion that dealers suffer no competitive injury within the meaning of Falls City unless their total profits are decreasing. In essence, respondent suggests the existence of some absolute standard of dealer health, above which there can be no finding of secondary line injury. Such an approach betrays the goals of the Act itself. Boise would have us ignore a fundamental principle of H The cases Boise cites for the proposition that evidence of an increase in plaintiffs sales vulume is inconsistent with a finding of competitive injury are inapposite.L01narIn IVlOlesale G)'oeery, Inc. v. Dieta GO!ll'met Foods, Inc. 824 F.2d 582, 597-98 (8th Cir. 1987), the court stressed that only primary line injury was in issue.Id. at 596. Equally misguided is respondent' s reliance uponR1'cW1' d Short Oil Co. v. Texaco, blc. 799 F. 2d 415 (8th Cir. 1986). In RichuJ'd Short the Eighth Circuit held plaintiff failed to establish a prima facie case of injury- that is, plaintiff failed to prescnt evidence of the amount, percentage and duration of the discrimination or the amount of competition the disfavored purchaser faced in the market. ld. at 420- 21. In the alternative, the court found that even if plaintiff " had been able to e tabli h competitive injury, " such injury was caused by defendant's own mismanagement, thereby breaking tbe causal cormection.Jd. at 421. For the same reasons, Boise can take no solace fromChrysler' Credit COT'p.l'. .J Tnu lt Payne, Co" 670 F. 2d 575 (5th Cir 1982). The Fifth Circuit statedMorton Salt. and its progeny is not relevant to the facts of this case and went on to determine that plaintiff failed to demonstrate any of the elements necessary to invoke the Morton Sail rule. It!. at 581 BOISE CASCADE CORP. 981 956 Opinion business-that money not gained is money lost. A finding that the dealers would have been more successful had they received the greater discounts is not "speculation " but is consistent with the evidence in Morton Salt itself. There complaint counsel demonstrated competitive injury in part by asking customer witnesses hypothetical questions about whether each customer s business would be "affected" if it either paid or charged a higher price for goods than its competitor. 162 F.2d at 956. Thus, we find Boise s growth evidence is of (30) insufficient probative value to overcome the Morton Salt inference of injury. 48 (31) 18 Complaint counsel further maintains that the gro\Vih evidence should be given litte weight because it is misleading and unreliable." Specifically, complaint counsel argues the following: (1) the growth figures overstate real g-rowth because they fail to adjust for inflation; (2) the figures were not offered in conjunction with any industry average growth figures so that it cannot be determined whether or not the selected dealers kept up with the average dealer growth nationwide; (3) there is at least one "obvious inconsistency" in underlying data for one of the selected dealers from which the average growth figures are calculated, allegedly casting doubt upon the reliability ofthe figures as a whole; (4) the statistics, which include data for only 1977- 1980, focus on an improperly narrow time frame since significant competitive injury allegedly occurred shortly after Boise s entry into given markets well before the chosen time period; and (5) gross profits are an inferior measure of dealer success. BOC at 27-30.

We need not reach these issues on remand since we find that the rebuttal evidence, if reliable, would be insufficient to overcome the inference of injury for thc reasons stated in the text. In the interest of judicial economy, however, we note there is more merit in some of complaint counsel's arguments than in others. First, while there is testimony in the record that the gross profit figures did not adjust for inflation, there is insufficient evidence to support a finding as to what the relevant measure of inflation should be in this case. Therefore, we do not reject the growth evidence on this basis. Second, we agree that the record is devoid of evidence of probative industry-wide growth standards with which the selected dealers may be compared. Boise argues on remand, apparently for the first time, that the XOPA dealer survey df'monstrates that the average reporting- membf'rs growth rate was 15% , less than the 22% average of the selected dealers. Boise explained on remand that the 15% figure is derived by multiplying average sales per employee of the reporting NOPA dealers by the average number of employees of those dealers, and then calculating average growth from 1978 through 1980. Tr, at 62. Initially, we note that Boise s estimated NOPA growth rate is the average of the growth rates so calculated for only two years, 1978- 1979 and 1979- 1980, whereas the selected dealers' actual average growth rates also included data for the 1977- 78 period. ;\ioreover,we find no evidence in the record supporting the reliability of conducting such a calculation fur the purpose of deriving meaningfu: average NOPA-member dealer growth rates. In addition the KOPA growth rates, like those for the selected dealers, fail to distinguish between sales of office products as defined in this procef'ding and sales from other aspects of the dealers ' businesses. Third, complaint counse: made the same argument to Judge Parker about inconsistencies in the calculations of the selected dealers ' gross sales and profits (CRRPFF at 67), who nevenheless found their average growth to be 22% per year. Complaint counsel has offered no reason why we should modify our previous adoption of that finding Fourth, we need not decide whether the " baseline" year for cxamining the effects of the price discriminations is, as compiaint counsel argues, the year in which Boise first entered each market (BOC at 28), or the first year in which the discounts were introduced in each market.Cf RBR at 26-27, In this case, the six manufacturers had granted the wholesaler discounts to Boise or its predecessors in most markets before the first year for which growth evidence was introduced.See iY'franote 63. Thus, even if growth evidence did have some probative value in secondary line cases, this value would be diminished where the evidence presented failed to include economically significant periods of time both before and afur injury was likely to have begun as a result of the discriminatory discounts. Fifth, our previous finding that net profits arc "the true measure of dealer success " 107 F7C at 182 n, does not cast doubt on thc accuracy of the gross profit data which we are specifically directed by the Court of (footnoteeont' 982 FEDERAL TRADE COMMISSION DECISIO:\S Opinion 113 F.

2. Displaced Sales a. Relation to the Morton Salt Inference The court focussed on other evidence "relevant to resolution of this controversy and unusual in the context of a R-P proceeding-a virtually complete absence of sales lost to Boise by the selected dealers traceable to the price differential caused by the wholesale discounts to Boise. " 837 F.2d at 1135. The court stated "the evidence, as found by the AU, failed to demonstrate 'displaced sales,' another form the Falls City Court indicated injury could take. " 837 F. 2d at 1145. In support of this statement, the court pointed to four factors: (I) evidence that it was "not uncommon " for accounts to switch; (2) while dealers lost accounts to Boise, Boise also lost accounts to the dealers and was unable to meet competitors' prices in some instances; (3) the fact that the number of accounts that switched to Boise was "quite small" and the proportion to dealers' total (32) accounts was strikingly low; " and (4) what the court termed "the key point; None of the selected dealers who lost accounts in whole or in part to Boise were able to conclude that losses were due to the different prices charged them and Boise by the six manufacturers (citations omitted). Id. at 1146. It concluded the switches to Boise apparently cannot be explained by operation of the wholesale discount. " Id. at 1145 (italics in original; emphasis added).

Preliminarily, we reconcile the court' s statement that the evidence failed to demonstrate "displaced sales " with its more cautious observation that accounts "apparently" did not switch to Boise because of the wholesale discount. In light of the evidence that dealers lost accounts to Boise because of the latter s lower prices, better services, or both (IDF 384), and the fact that the court declined to specify how the Commission should weigh or evaluate the evidence on remand (837 F. 2d at 1145 n. 15), we read the opinion to mean that the evidence, as found by the AU, did not establish "direct evidence of displaced sales " the Falls City formula for conclusively invoking Morton Salt. (Emphasis added.

Assuming, however, that the evidence does not establish "displaced sales by direct evidence or otherwise-this showing, alone or in conjunction with other record evidence, would fail to overcome the Appeals to examinc. :-everthelcss, the evidence of increasi .g gross profits fails to demonstrate an absence of actuai inJul'yforthe reasons set forth in the text. BOISE CASCADE CORP. 983 956 Opinion inference of injury. The Administrative Law Judge correctly held that diversion of trade to the favored customer need not be proved. " 107 FTC at 181; see Bargain Car Wash, Inc. v. Standard Oil Co., 466 2d at 1174 (proof of loss of specific sales not required). In describing displaced sales as "another" form in which competitive injury may be manifested, the Court of Appeals implicitly recognized that injury may be found in the absence of such evidence. Injury may also take the form of the favored purchaser s use of its price advantage to increase its profits or enhance its services. See supra at pp. 10 , 19-20. 49 (33) More importantly, the Court of Appeals confirmed that proof of actual harm to competition in any form is not required. Boise Cascade Corp. v. FTC 837 F. 2d at 1139 n. 14. Under the court' s decision evidence of the absence of displaced sales is relevant to whether the inference of competitive injury has been overcome to the extent such evidence either (a) demonstrates an absence of actual injury from which an inference of the absence of competitive injury within the meaning of R-P can be drawn, or (b) breaks the causal connection between any injury and the price differential. The inference would be overcome by probative evidence that would preclude the fact-finder from concluding that Boise s price advantage enabled it to lower its resale prices, offer improved services, or reap greater profits. As we explain below, no such rebuttal has been made. b. The Evidence Pertaining To Displaced Sales Fails To Overcome The Morton Salt Inference.

The Court of Appeals directs us to consider four types of evidence pertaining to displaced sales. First, the court observes that account shifting was "not uncommon" but rather was the "order of the day. 837 F.2d at 1145. Frequent account shifting between competitors may suggest that competition for the business of those accounts " healthy," but, like the growth evidence, does not preclude a finding of secondary line injury.

Boise makes the related argument that while account shifting may be common industry-wide, the relatively small amount of account switching in this case suggests an absence of injury. Boise points to the court' s observation that " (tJhis stabilty of dealer accounts is of 49 The introduction of new services by the favored purchaser would be expected to cause service-sensitive customers to switch their business to the favored purchaser. At some point, however, the market would substantially adjust to these new conditions, and most customers sensitive to the better services offered by the favored dealers would have already swit.ched to them. :.loreuver, extensive switching may not be manifested where the favored purchaser uses some or all of its price advantage to widen its net profit margins, a resuit that may later redound to that competitor s advantage. See supra note 32. Opinion 113 F.

especial note in an industry where switching of accounts is common. 837 F.2d at 1135; RBR at 33. Such a phenomenon may be explained however, by dealers electing not to compete aggressively with Boise for some customers because of the substantial cost advantage enjoyed by Boise as a result of its favorable discounts (IDF 76), or by the market previously adjusting to the effects of the price discrimination. See supra note 49.

Second, the court directs us to consider that this "phenomenon of lost accounts" was "a two way street " noting Boise lost accounts to the selected dealers and at times Boise was unable to meet the selected dealers ' prices. Id. at 1136. The court characterized the considerable evidence" that Boise lost sales to the selected dealers as s attempted showing of diverted (fJurther diluting the Commission 50 (34)sales. Id. at 1136.

While the loss of some accounts from Boise to the selected dealers may "dilute" complaint counsel's showing of "diverted" sales, it does not defeat a finding of liability. Such a showing may limit respondent' s exposure to damages in a private action under Section 4 of the Clayton Act, but would not absolve Boise from liability under 2(a). See Falls City Indus. , Inc. v. Vanco Beverage, Inc. 460 U.S. at 437 (respondent liable to the extent plaintiff's injury was attributable to price difference, even if other factors accounted for most of the plaintiff's lost sales); Rose Confections, Inc. v. Ambrosia Chocolate Co. 816 F. 2d 381 , 387 (3d Cir. 1987). Similarly, a finding that disfavored dealers were injured within the meaning of Robinson- Patman is not overcome by evidence that they obtained some accounts from the favored purchaser.

A myriad of factors may explain Boise s loss of accounts to the selected dealers or the abilty of dealers to underprice Boise in certain instances. The evidence shows that Boise established above average return on investment standards for itself (CX 44E; RPFF 389) and targeted customers interested in Boise s high service-oriented marketing approach. See, e. tr. at 5186 5221-22; CX 85J; RPFF 339. addition, Boise may have incurred greater costs (other than cost of goods sold) and thereby operated less efficiently than the selected 50 The court did not distinguish between "diverted" and "displaced" sales in its opinion. 0) The point is ilustrated by Boise s use of price lists (" net priccrs ) which include four differer.t price levels-known as column 5 through column 8 pricing--which Boise quoted to accounts for given products. Willingham Dep. ex 974 , pp. 144-45; Barnett ir. 5789-90. The price level Boise quoted accounts varied based on a number of factors, including the prices offered by competing dealers (Kupp. ir. 5197), the mutual interest of Boise and the account in doing business together (RX 239B), or the amount of services Boise offered to the account. Kurr. tr. 5186, 5221-22; Bode tr. 5247 , BOISE CASCADE CORP. 985 956 Opinion dealers. See RPFF 383-84 (dealers' lower inventory costs); RPFF 388 (dealers' greater purchasing flexibility); RPFF 389 and Wolnhofer Dep. CX 672, pp. 38-39 (dealers' lower overhead, wage and fringebenefit expenses than Boise). Indeed, the Administrative Law Judge found (tJhe extent of (the dealers J success in the face of these price discriminations is a testimony to their business acumen and establishes that they are as efficient, and perhaps more efficient, than Boise. " 107 FTC at 184.

While respondent minimizes the 162 specific accounts the dealers lost to Boise (id. at 1135), the figure takes on greater significance when compared to the no more than 60 accounts that the Administrative Law Judge specifically identified as either having been lost by Boise to other dealers or having received lower price quotes from dealers in competition with Boise. (35) IDF 409-421. 52 Ironically, just as complaint counsel argues that the 162 specific accounts were merely examples of those that the selected dealers lost to Boise respondent emphasizes the existence of other accounts Boise lost which were not identified in Judge Parker s opinion. IDF 422. Were we to treat the accounts specifically identified as only examples of the total number of accounts Boise and the selected dealers actually lost to one another, there would be no reason to conclude that Boise share of the actual lost accounts would be larger. Indeed, it was probably easier for Boise to readily identify accounts it had lost than for complaint counsel to track down accounts lost by more than 20 different dealers. 51 Thus, the specific accounts (36) identified in the 52 Interestingly, each party challenges the reliabilty of those portions of Judge Parker s findings about lost accounts which are not favorable to its respective position. For example, complaint counsel argues that the Administrative Law Judge mistakenly found Boise "lost" certain accounts which it did not in fact service in the first instance or which it later regained. BOC at 47" 51; but see RBR at 43-44. Similarly, Boise argues that many of the 162 so-called ' lost accounts' were never the dealers ' accounts, were never lost by the dealers were not lost to Boise Cascade or wcrc regained by the dealers. " BOR at 18; Imi see BOC at 54 n. 27. The Adminiotrative Law ,Judge made specific findings concerning the lost accounts evidence which we have previously adopted.

.\:1 Although the court cited the Commission s brief to the Court of Appeals as the source of the 162 lost accounts figure, 837 F. 2d at 1135, the Commission s brief actually referred to that number as " Boise s own highly conservative tabulation. See BRC at 32; RPFf' 310. Both the dealers (see Kelly tr. 4000; Wilhelmi tr. 3557; Cashman tr. 2607; Lake tr. 3035; B. Jones tr. 3426) and Boise s own salespeople (see, e. Bode tr. 5263- , 5267: Kupp tr. 5223 5230- 310)testifiedthatotheraccounts switched from the selected dealers to Boise. The evidence is insufficient, however, for us to determine just how many accounts switched from dealers to Boise, either in absolute terms or as a percentage of the total number of accounts that switched in both directions.

M Complaint counsel' s comparative difficulty in marshalling lost accounts evidence is illustrated by its cffort to introduce testimony at trial about accounts the dealers lost that were not identified in response to Boise interrogatories. ,Jones tr. 3426;see also Mack tr. 3667. In one instance, Boise objected to a dealer-witness testimony about ar. account that was not brought to complaint counsel' s attention until the night before the wtness testified. Jones tr. 3428. :-o ruling on the objection was made, but complaint counsel elected not to pursue the line 0: questioning Opinion 113 F. T.

record may well overstate the number of accounts that Boise lost to the dealers relative to those that it gained. The third factor the court cited was the "quite small" number and strikingly low" proportion of accounts that switched to Boise. Specifically, Boise estimates that the 162 accounts identified as having been lost by the selected dealers to Boise account for less than .4% of the dealers ' 42 000 accounts. BOR at 18; RBR at 33. Respondent claims this "dearth of evidence regarding lost accounts led the ALJ to conclude that complaint counsel's attempt to establish displaced sales was a very fruitless endeavor." Tr. 4554; RBR at 33. As previously noted, evidence in the record supports complaint counsel's assertion that the 162 specific accounts are merely examples of the accounts lost by the selected dealers to Boise. In any event Boise does not explain how the fact that only a small number of accounts switched shows the absence of actual injury or breaks the causal connection between the price discrimination and the alleged injury. 55 Since complaint counsel (37) has no obligation to establish displaced sales it is Boise that must produce probative evidence to overcome the Morton Salt inference. In essence, the court' s reference to the small number of accounts that switched is a restatement of its initial observation that account switching was relatively infrequent and fails to overcome the inference of injury for the reasons previously set forth. See supra at p. 33. 57 (38) 55 The context of the statement suggests a very different meaning than that which Boise ascribes to it, Judge Parker was urging counsel for both parties to enter into a stipulation that Boise took some customers from the selected dealers and vice versa. Hoise, asserting that the government had not yet proved competitive injury, took the suggestjon under advisement. Tr. 4554-55. The import of this exchange was that it was fruitless for both parties to fill the record with examples of account switching-nothing more. 5r, Complaint counsel argues that the 162 accounts specifically identified as lost by the dealers to Boise included some of those dealers ' largest accounts. BOC at 54 , 57; IDF 391 , 392 , 394 , 398, 400. Respondent notes that complaint counsel identified only five such "large" accounts, and questions whether the loss of two of those accounts was causally related to the challenged discounts. HER at 34-35 & n. 12. WI; are satisfied that the evidence supports Judge Parker s previous findings that at least some of the accounts sper.fically identified as lost by the dealers were among their Jargest. The message1"-fortUltof Salt that Robinson Patman guards disfavored purchasers against price discrimination affecting even a minor portion of their stock. This is of particular importance in this industry where it is not ullcommon for a dealer to derive a disproportionately large share of its revenue from a small number of accounts.See, e. Cushman tr. 2601; Applebaum tr. 1334-1335; Crompton tr. 723; Elzinga tr. 6258. 57 Respondent points to two additional aspects of the displaced sales evidence which it asserts demonstrates the absence of competitive injury. First, it relies upon the testimony of its expert, Professor EJzinga, that "YOll simply cannot look at that evidence and conclude that injury to competition has occurred or is occurringhere. BOR at 41; RBR 38; 837 F. 2d at 1136; IDF 433. Professor Elzinga, however, did not address whether the evidence demonstrated the existence of secondary line injury rather than "injury to competition" as that term is used in other antitrust contexts.

Second, Boi e argues that there is no evidence that injury has occurred in any other form in which it may be manifested; therefore, the dearth of displaced sales overcomcs the inference of injury. It asserts that the governmentoffcrsspecolation, not evidence, that profits ofdis favored dealers were impaired by the discounts and that such speculation i unfounded given that the selected dcaler gross sales and profit grew at the (fODlnotecont' , BOISE CASCADE CORP. 987 956 Opinion The final point emphasized by the court was that none of the selected dealers testified that they lost accounts to Boise because they paid higher prices than Boise paid on the products of the six manufacturers. The court concluded (iJn short the switches to Boise apparently cannot be explained as sales diverted through operation of the wholesale discount. " Id. at 1146.

The selected dealers' inabilty to provide such testimony is not significant. As a practical matter, dealers would be unable to testify credibly on the subject without extensive access to Boise s cost and markup data broken down by manufacturers and customers. To place this type of evidence in the record, the dealers would have to respond to questions concerning the price Boise itself pays for products purchased from specific manufacturers. The dealers would also have to understand Boise s operations so thoroughly as to testify that no explanation other than the difference in cost of goods sold accounts for the differences in prices charged by Boise and the selected dealers on those products. Dealers are unlikely to have such intimate knowledge of their competitors' costs. Indeed, the dealers' (39) acquisition of such knowledge may raise serious questions about their compliance with the antitrust laws.

Respondent argues that accounts switch among dealers for a variety of reasons unrelated to the price differences. Boise initially same rate. RBR at 36. It further states there is no evidence to support complaint counsel's assertion that dealers may have been injured because Boise was able to enlarge its profis or enhance its services. Boise argues that the latter point is only relevant if Boise s enhanced services "led to saics diversion, a fact which the Court explicitly found not to exist," RBR at 37. Finally, respondent asserts enhancement of its profits or services "are not elements of competitive injur)' under the Art." HER at 37. Boise misstates the iaw and stretches the fact.. Secondary line injury may be established where a price difference creates a reasonable possibility that the favored purchaser will enhance its profits or services relative to its disfavored competitor. Evidence of sales and gross profits increasing at the same rate does not preclude a finding of impaired profits as a form of competitive injury,see infm at pp. 42-43, particularly where the evidence relates to a period long after the discount structure was put into place. Contrary to Boise s assertions. there is substantial evidence that Boise earned greater net profits than not only the selected dealers, but than the average for dealers as reported in the NOPA and Robert Morri, A,sociates dealer studies.See 11Ifm note ,9. Respondent's statement that there is "no evidence suggesting that dealers did not or cOllld not offer the same consumer services" that Boise offered disequallyspecious. In support of this proposition, Boise cites testimony of the purchasing director of the Delaware Trust Company, which actualiy supports complaint counsel's position. Hitchens tr. at 5829-32. Mr. Hitchens explained that Boise Cascade attempted to interest him in Boise s " stock control program. " Thereafter, Hitchens solicited expressions of interest in supplying the stock control program from the three suppliers Delaware Trust had used for years, H&H Stationery, George D. Hanby Company (" GDR") and Hugh A. George Company HAG"). GDH and HAG were two of the selected dealers in this proceeding. Hitchens tr. at 5830. H&H and Hugh A. George indicated that "they would not be interested in getting involved in that type of program or being able to offer that kind of a service to their customers. Id. A GDH representative stated his company could provide such a service but that it would "take a little time" to put it together.Id. at 5831. Aftr a while Mr. Hitchens decided that he did not want to be a guinea pig for Hanhy and decided to go with Boise Cascade. ld. at 5832. The example in no way undennines the Administrative Law Judge s finding that Boise used its price advantage to offer better services than its competitors offered. 107 FTC at 182. Opinion 113 F.

emphasized that " (tJhe most prominent reason" for accounts switching among vendors was the change in employment of sales representatives that serviced said accounts (BOR at 41), but later conceded that it "has never claimed that the change in employment of sales representatives explains all or most of the 162 so-called lost accounts. " RBR at 43. In its reply brief, Boise summarized its view as follows:

The evidence demonstrates that the change in employment of saes representatives is just one of many reasons, unrelated to the functional discount, that explain the shifting of accounts. If account shifting was the result of an anticompetitive pricing practice, it would move only in one direction, from the dealers to Boise. The Court and the AU explicitly found that this was not the case. Indeed, the Court held that "the phenomenon of lost accounts was very much a two-way street, " 837 F.2d at 1145 (citation to Appendix omitted). RBR at 43.'8 Boise erroneously assumes that if accounts shifted as the result of an anticompetitive pricing practice, they would move in only one direction, from the dealers to Boise. In fact, dealers may have lost some accounts to Boise because of the cost advantage resulting from the favorable discounts while others may have been lost for other reasons. On the other hand, there was no evidence that accounts that shifted from Boise to the dealers were aided by any "anticompetitive pricing practice " favorable to the dealers. In short, while account shifts resulting from the discriminatory pricing would have moved in only one direction, other accounts may have shifted in both directions for reasons wholly unrelated to the challenged pricing practices. Boise s argument is not persuasive. (40) 3. Comparative Net Profit Data The court pointed to evidence that the selected dealers' median net profit before taxes as a percent of sales (2. 3%- 5% for 1976 to 1981) may have been lower than that of dealer-members of NOPA (3. 0%- 9% for 1967 to 1980), and lower still than Boise s combined wholesaler/dealer net profit (3. 1%- 6% from 1976 to 1979). 837 F. at 1I44. Acknowledging "comparative problems " the court counseled 58 Dr. Elzinga admitted that accounts do not always switch because of salesmen changing positions, and that. " pr:ce and service has (sic) something to do with it." !DF 434. Even f cha ges :n the employment of salesmen were the principal reason for account shifting among dealers, there is evidence that Boise may have attracted salesmen fl"m competing dealers by offering them higher wages and better benefits. Welnhofer Dep. ex 672 p. 38 BOISE CASCADE CORP. 989 956 Opinion that "this apparent discrepancy in profitabilty obviously deserved consideration." 837 F.2d at 1144 (emphasis in original). The Administrative Law Judge expressly referred to this net profit data in concluding, " (cJonsidering these (and others facts, the only possible inference is that the effect of the substantial and sustained price discriminations favoring Boise may be to destroy or prevent competition with the favored dealers. " 107 FTC at 182 , 201; see also IDF 423-30. Although the Commission previously (41) adopted this 59 Rather than focus on whether the net profi evidence overcomes the inference of injury, the parties vigorously dispute whether the evidence cited by the Court 0f Appeals is reliable or the best indication of dealer success. Boise maintains that the dealers' gross profits, rather than net profits, are the relevant inquiry, and faults complaint counsel fof making assertions that Boise claims were previously rejected by each body that has reviewed the evidence. RBR at 27. In fact, it is Boise that makes the same arguments previously rejected by the Administrative Law Judge and the Commission. Boise again asserts that net profit data for closely held dealers is unreliable because the dealers have tax incentives for artificially lowering their bottom line. BOR at 36 n.lO; RBR at 29- 30. This is the very argument that was dismissed by Judge Parker who found, " net profits are the true measure of dealer success. . and I reject Boise s claim that only gross profits of private€ly-heid companies should be considered. " 107 FTC at 182. Most importantly, the Court of Appeals directs us to consider net, not gross, profit figures for comparative purposes. Boise further asserts that no reliable inferences regarding the effect of the challenged discounts can be drawn from a comparison of the net profit data for the XOPA dealers and the selected dealers. RBR at 30. Boise asserts the NOPA figures are unreliable because (1) they were derived from summaries prepared by !\OPA staff from data submitted in a voluntary survey; (2) many surveys submitt€d were incomplete but nonetheless included in the study; (3) Boise s expert concluded the survey was not valid or reliable; and (4) a study entitled Dealer Operating Results published in an industry publication Office Produds Dealer reported dealer net profits at 14.3%. BOR at 36 n. lO. Respondent further questions the probative value of the profitability figures cited by the court concerning Boise itself because they reflect Boise s sales to both endusers and to dealers, as well as sales of merchandise that were excluded from this proceeding. Id. Interestingly, while Boise challenges the 1\OPA survey results on the one hand, it relies upon them on the other to argue that the leading NOPA dealers had median net profits that were comparable to or greater than Roise s from 1978 to 1980 (RBR at 31), and that the average :\OPA growth rate was below that of the selected dealers. See ill(m note 62 and accompanying text. We are unpersuaded that we have erred in adopting Judge Parker s findings. First, the reliability of the NOPA net profit figures is supportd by evidence that such data is used by dealers in approaching banks for loans (Tr. 1813) and that the Dealer Operating Results reflect " a pretty good overall sample of the industry and not just the dealers who report. IOF 424. Significantly, whatever deficiencies may exist in the NOPA survey do not appear to have undermined the reliability of its net profit results, which are substantially corroborated by the results of a Robert Morris Associates Annual Study.Compare IOF 428 , 429 with IOF 425. The consistency between the Robert Morris and !\OPA surveys for the years they have in common are illustrated by the following chart (which also shows that Boise s profits were generally higher, and the selected dealers' generally lower, than the Morris and NOPA results): PROt1TS BEFORE TAXES 1976 1977 1978 1979 1980 1981 Boise Net Profit (IDF87 5lZ12) 1\OPA' Dealer Operating Results (CX 355 (footnotecont' g Opinion 113 F.

conclusion from Judge Parker s initial decision, we take this opportunity to address Boise s specific concerns on remand. The net profit evidence does not demonstrate the absence of any actual or reasonable possibility of competitive injury. It does not preclude a finding that Boise enjoyed any of the benefits of favorable pricing from which courts have inferred competitive injury. See supra at pp. 10- , 19-20. If anything, this evidence tends to demonstrate that Boise benefited from the discriminatory discounts at the expense of the dealers.

Unable to contend that the net profit data alone weighs against a finding of injury, 60 respondent argues that the data, examined in the context of the selected dealers' increasing sales and gross profits breaks the causal connection between any substandard performance in the selected dealers' net profits and (431 the challenged discounts. First, Boise states that evidence that 18 of 23 selected dealers experienced increasing sales and profits at the same rate from 1977 through 1980 (IDF 431) demonstrates "the dealers clearly did not achieve their sales growth at the expense of profits." RBR at 28. The argument is without merit. That gross sales and profits increased in parallel implies nothing about the effects of the discriminatory prices Robert Morris Associates Annual Statement Studies (CX 2300;

Rowe tr. 7113) 3.'% Selected Dealer Financial Statements or Tax Returns (CX 2301) See SOC 32, 39. 2.4% Secot1d, by citing the testimony of complaint r.ounsel's expert who found NOPA' s Dealer Operating Results to be reliable (lDF 426 , 427), Juclg-e Parker apparently rejected the opinion of Boise s expert. Third, the testimony of Ronald Rowe, complaint wunsel's accounting expert, raised serious questions about the methodology of the Office Products Dealer study, including the fact that it reportedaverage not median deaier net profits. Obviously, Judge Parker found theOPD survey results less reliable than NOPA's when he found net profits " in the industry are low (from 3-4%), " 107 FTC at 182 citing IDF 423-430, To the extent the court' s net profit figures include sales by Boise to dealers, they may understate Boise s profits on sales to endusers, the sales for which Boise competes with the selected dealers.See, e. Bertholdt tr. 6029- 30; CX- 1380 1381 (In Camera: 1979 pre-tax net profit on sales to dealers was 0-8%; pre-tax net profit on sales to commerciai accounts that same year was 6. 1%). We recognize that the net profit data suffers from the same defect as Boise s gross profis and sales figures--overinclusion. Thus, while this evidence alone might not support a finding of competitive injury in the relevant product market, it also cannot overcome the MOJ.ton Salt inference of injury. 60 Boise does not contend that a comparison of the net profit data supports its case. It maintains simply that no reliable inference regarding the competitive effect of the functional discount can be drawn from a comparison of the NOPA net profit data and the net profits of the selected dealers" (RBR at 30), and that a comparison of Boise Cascadc s net profits to lthe selectedl dealer(s l net profits. . is equally unavailing. RBR at 31 BOISE CASCADE CORP. 991 956 Opinion unless the period studied includes years both before and after the discounts were initially granted. Otherwise, these parallel increases in sales and profits would be of no more probative value than evidence of growth in either measure independent of the other. Since many of the challenged discounts at issue in this case were put into effect prior to 1977, the first year of parallel rises in gross sales and profits cited by respondent, the evidence fails to overcome the Morton Salt inference. See irira note 63.

Even assuming that the evidence showed the selected dealers overall gross sales and profits increased at the same rate both before and after Boise began to receive the favorable discounts, such evidence, alone, would not overcome the inference of injury. As explained earlier, rebuttal evidence that is not product-specific is of relatively little probative value since some dealers earned significant revenues from the sale of products outside of the defined market. Morton Salt 334 u. S. at 49; see supra at pp. 27-28. Thus, higher sales and profits in one segment of a dealer s business may have offset poorer performance in other segments. In addition, evidence of increasing parallel gross sales and profits by dealers both before and after the introduction of the favorable discounts does not rule out the possibility that these increases would have been even greater but for Boise s receipt of the discriminatory discounts. Boise s attempt to use growth evidence under the guise of net profit data to break the causal connection fails in the face of direct evidence that the dealers did, in fact, lower their prices and margins in order to retain accounts for which they competed with Boise. IDF 405. The ALl concluded:

The dealers have also underpriced Boise at times (F. ' s 409-22), but they have had to do so in the face of its substantia! price advantage, so that when a dealer lured a customer away from Boise, its success was accompanied by a signifieant impairment of profits.

107 FTC at 182. Other dealers refrained from competing against Boise for certain accounts because they could not compete on price. IDF 406.

Second, Boise argues that evidence that the selected dealers increased their sales and gross profits at a greater rate than their net profits breaks the causal connection between the price (44) discrimination and competitive injury. Judge Parker found the selected dealers experienced, on average, a 22% increase in sales and gross profits Opinion 113 F.

from 1977 through 1980. IDF 431. Net profits over the same period however, ranged from 2. 4% to 3. 2% and, while drifting downward, did not do so without interruption. See supra note 59. Boise argues that net profit does not provide a true picture of dealer success because its calculation requires that many expenses other than cost of goods sold be deducted from sales. As Boise puts it, a "dealer s management of these expenses may suggest the extent of his business acumen, but it says nothing about the potential competitive impact of the functional discount." RBR at 28 (emphasis added).

Although Boise s approach might simplify the task of courts and the Commission in evaluating rebuttal evidence, a firm may experience declining net profits while sales are rising for many reasons. Thus that experience alone would not preclude a finding that the challenged discounts substantially contributed to the disfavored dealer s injury. A firm faced with greater costs of goods sold than its rival, with resale prices and all other costs equal, would realize lower (gross and net) profit margins than its rival. In the long run, however, the disfavored firm might decide to make operational or strategic changes in order to remain competitive. It might, for example, explore ways to reduce its operational costs and increase its efficiency by making capital investments or other expenditures. Alternatively, it may attempt to maintain market share by providing additional consumer services (at greater costs) in pursuit of an upscale clientele. This approach would reduce the firm s net profits unless it could pass the increased costs through in the form of (45) higher prices without an offsetting loss in unit sales. Under either scenario (and surely there are others) the disfavored firm is gambling that additional costs incurred today will win it loyal customers for tomorrow. These added costs would have the effect of reducing net profits without necessarily triggering a corresponding decline in sales.

The analysis is not altered by the fact that the selected dealers also experienced growth in gross profits while their net profits declined. 61 The parties' respective positions are illustrated by the example of Kelly Co. , whose gross profit increased from 27.97% in 1979 to 29.06% in 1980 while its net profit decreased from 2.54% to 0.47% during- the same period. Complaint counsel maintains that the example demonstrat€s that net profits are a better measure of dealer performance. BOC at 29. Boise counters that an examination of Kelly Company s financial statements shows that the decline in net profits was occas:oned by increased "offcc and "occupancy" expCr1SCS unrelated to the challenged discounts. The financial statements upon which Boise relies, however, are merely exhibits which were marked for identification but were never offered into evidence. They are not part of the record of ';these proceedings, and are entitled to no weight on remand.Cf Commission Order dated September 6 . 1988 (granting in part and denying in part Respondent' s Motion for Leave to File Attached Supplemental IRtter). :.1moreover. even if the financial statements were part of the record they could not alone establish that the decrease in net profits was unrelated to the challenged discounts for the reasons set forth in the text. BOISE CASCADE CORP. 993 956 Opinion Gross profits wil rise as long as prices increase relative to costs of goods sold. This rise may be the result of many factors such as increased demand resulting from growth in the white collar sector of the economy or inflation. Here, it may also reflect increased sales of products that were not the subject of these proceedings. In short, the evidence is too ambiguous to overcome the inference of injury. Third, Boise argues that the challenged discounts should have affected all dealers equally; therefore, evidence that the selected dealers earned lower net profits than the dealers as a whole "leads only to the conclusion that complaint counsel attempted to select the least profitable dealers in the hope that such dealers would help support complaint counsel's theory of competitive injury." RBR at 30. In essence, the argument is that complaint counsel chose the selected dealers because of their inefficiency, and that such inefficiency breaks the causal connection between the price discrimination and the alleged resulting injury. The argument assumes, without proof, that all dealers are otherwise identical, devoting the same proportion of their respective businesses to the sale of office products in relation to the sale of other products excluded from consideration in this case. It also assumes that the net profit data showing that the selected dealers earned lower profits than the NOP A dealers on all sales is a reliable surrogate for the selected dealers' respective net profits on sales of only those products that are the subject of these proceedings. In any event, the more pertinent comparison is between Boise and the selected dealers, not between the selected dealers and the NOP dealers. Boise fared significantly better.

Even if complaint counsel' s case did focus on dealers who were less profitable than most with respect to sales of the products in issue their low profitability is stil consistent with an inference of injury. These relatively low profits may suggest the disfavored dealers injuries were caused in part by other factors, but does not preclude the reasonable inference that the price discrimination caused a material part of the dealers' injury. "It would be enough with respect to causation if the defendant ' materially contributed' to plaintiffs injury . . . or 'substantially contributed notwithstanding other factors contributed also. Bargain Car Wash, Inc. v. Standard Oil Co. (Indiana), 466 F. 2d at 1174 , (46) quoting Perma Life Muffers, Inc. v. Int' l Parts Corp. 392 U. S. 134 , 143 (1968) (J. White, concurring); see also Zenith Radio Corp. v. Hazeltine Research Inc. 395 U. S. at 100 , 114 n. 9 (1969); Hasbrouck v. Texaco, Inc. 842 F. 2d at 1042 994 FEDERAL TRADE COMMISSIO DECISIONS Opinion 113 F.

(iJt is enough that the illegality is shown to be a material cause " of plaintiff' s injury to recover under Section 4 of the Clayton Act). Fourth, Boise argues that its net profits were comparable to those of the leading industrial stationers (IDF 87, 837 F. 2d at 1135), and that the top 25% of the dealers in the NOP A survey had greater median net profits than Boise for the years 1978 to 1980. RBR at 31. Assuming that these assertions are fully supportable by the record they are of litte probative value. The following colloquy at oral argument on remand was instructive:

Commission: Counsel, what is your explanation for taking the upper 25 percent, as opposed to the entire average, the range? Respondent' s Counsel: I think the point is this. If, in fact, the discount had any impact, it impacted all of the NOPA dealers in the same way. (47) This shows that Boise was doing less well than the more successful dealers. , in fact, the discount had any competitive impact at all, one would expect to see these figures reversed, that Boise would do better than even the best, even the most successful dealers.

Tr. at 65.

This evidence shows only that Boise fared no better than leading dealers. As Boise acknowledges, there are many reasons which may reduce a firm s net profit other than paying more for inventory. RBR at 28-29. Since Boise had lower costs of goods sold than the selected dealers one would expect Boise to have greater gross and net profits than the other dealers if all other costs were equal. But Boise may have had greater costs (i. been less efficient) than the leading dealers, which would account for Boise s lower net profits. Thus, this evidence does not overcome the inference that competitive injury resulted from the discounts and, like the other comparative net profit evidence, is insufficient to overcome the Morton Salt inference. 62 While the NOPA surveys do report that the highest 25% of dealers in profits earned higher median net profits than Boise for 1978- 1980, there is reason to question the evidence that the Icadingindustnal stationers earned profits comparable to those earned by Boise. Boise relies upon language in the opinion of the Court of Appeals that "(tJhis return on sales, as Boise s 1979-83 Business Plan explained, was comparable to the leading industrial stationers." The court cited lDF 87, which in turn cites only the Business Plan itself, ex 48B, which provides in pertinent part:

Larj?e industrial stationers At the present time, accurate information regarding competitive performance statistics is unavailable. We do fee!, however, that our projected inventory turns and return on sales for the period of the plan are comparable to the leaders in our industry.

Whatever significance may attach to this evidence is flirther eruded by the fact that the term " contract stationer " apparently used interchangeably with the term "indc:strial stationer " has "no generally accepted meaning in tr.t office products industry." IDF 44. BOISE CASCADE CORP. 995 956 Opinion B. Evidence of the Longevity of the Discounts The Court of Appeals recognized that the Falls City Court looked to the longstanding nature of a price difference as giving rise to Morton Salt' inference of injury, but read that case to mean only that longterm price discrimination is more likely to lead to injury than shortterm or sporadic discrimination. Id. at 1146, n. l? The court reasoned that "a price difference with the potential for causing injury wil eventually result in measurable effects, such as diverted sales or lost profits. " 837 F.2d at 1146 n. 17. Relying upon Matsushita Electrical Industrial Co. v. Zenith Radio Corp. 475 U.S. 574 (1986), the court observed that the "more time that passes without such effects showing up, the more likely it becomes that the price difference does not in fact pose a threat of competitive injury. Id. While acknowledging that the "longstanding nature of the discounts does not preclude the possibility" that they pose a "threat of injury to competition in the future " 837 F. 2d at 1146, the court stated one would reasonably expect to have seen competitive injury manifest itself by this time " the form of some dragging down. Id. As yet, the court observed there appears to be no such damper. Id.

Certainly, it may be argued as a general proposition that the longer a price difference prevails without visible manifestation of competitive injury, the less likely such injury is to occur in the future. But such is not this case. The record contains evidence that the selected dealers have (48) experienced " dragging down." We have previously discussed the evidence that both the selected dealers and dealers in general earned lower average net profits than did Boise for the years 1977 through 1980. See supra at pp. 40-42. Accounts of substantial size were lost by the dealers to Boise during that period. See supra note 56. Judge Parker found the selected dealers often had to lower their prices to compete with Boise or refrained from competing with Boise for a number of accounts. IDF 405, 406. These indicia of injury are particularly significant in light of evidence suggesting that the selected dealers may have operated more efficiently than Boise in some ways. See supra at p. 34. There are, however, limits to applying the principles of Matsushita a Sherman Act predatory pricing case, to the problems of price discrimination under Robinson-Patman. Secondary line Robinson-Patman injury is distinct from injury to competition under the Sherman Act. Xleasurable competitive effects may be manifested within a relatively short time following the initiation of the price discrimination Opinion 113 F.

in some industries, while it may take longer in others. Moreover, the profit and sales data produced in this case for the selected dealers is overbroad see supra at pp. 27-28 & note 59, and therefore is of litte value in assessing the effects of even recently introduced price discrimination. In short, there is reason to doubt whether Boise rebuttal evidence, which focuses on a period some 40 years after respondent asserts the discounts began, can be causally related to either the presence or absence of secondary line injury under the Robinson-Patman Act.

Thus, it is unnecessary for us to determine whether the relevant period during which the discounts have been in effect is 40 years, as 63 or the time since Boise first (49) received therespondent asserts discounts in given geographic markets as complaint counsel intimates. 64 Indeed, in Morton Salt itself, the challenged carload discount had been offered by the Morton Salt Company since 1922. Morton Salt Co. v. FTC 162 F. 2d 949, 953 (7th Cir. 1947). It was a full 22 years later when the Commission issued its order imposing liability, and an additional four years until the Supreme Court made that order stick.

The evidence of competitive injury in this case is as strong as that which supported the Supreme Court' s finding of liability in Morton Salt more than two decades after that challenged discount program began. Thus, even assuming that the complaint in this matter was issued some forty years after the hybrid discounts first went into effect, this fact does not overcome our finding of competitive injury. 63 BOR at 58. The six manufacturers whose sales are the subject of this litigation actually initiated the challenged discount programs at various times over the past forty years. See, e. IDF 116 117 (Boise or predecessors received favorable wholesaler discount on all Rediform business forms since the late 1950' s or early 1960' , and on Hecordplate since 1977); IDF 171 (Boise received favorable wholesaler discount from Sheaffer Eaton as long as twenty-five year employee of Sheaffer Eaton had known Boise); IDF 226, 230 (Boise received favorable wholesaler discount on visible equipment and on insulated fies from Victor Systems and Equipment division of Kardex Systems, Inc. , since 1979); !DF 263 (Boise received favorable wholesaler discount from Boorum & Pease for almost 20 years); !DF 302 (Bates offered wholesaler discount since the I940s and Boise received it as long as it had been a customer); !DF 346 (Masterproducts gave Boise the favorable discount for at least 10 years prior to issuance of the Initial Decision). 64 BOC at 61. For example, Boise entered the Massachusetts area by acquiring Dennis Offce Supply, a commercial office products dealer that did not engage in any wholesaling, on March 21 , 1978. Rodman tr. at 4435-36. Similarly, Boise entered Phoenix in 1978 by acquiring PBSW (Starr tr. at 5154), which.h previously made relatively few sales ' 0 other dealers.Id. at 5161-62; RPFf 87. Boise entered Salt Lake City by acquiring PembrukE's Company, a local office supplies dealer (Vieistich tr. at 4934), in 1974. Pembroke did not previously sell to other ciealers.Jd. at 4940; RPFF 88. It entered San Francisco through the acquisition of SOlithern California Stationers in approximately 1966. ex 977 at 26-27. Prior to being acquired by Boise Southerr. California Stationers did not engage in wholesaling. Edquist tr. at 5075-76. BOISE CASCADE CORP. 997 956 Opinion C. A Finding of Liability is Consistent with the Purpose of P and is Not Averted by an Availability Defense. The court criticized the Commission for failing "to inform" its application of Morton Salt with the purposes of Robinson-Patman by assuming, without analysis, the existence of competitive injury. " 837 2d at 1146. Specifically, the court stated the Act is directed toward halting the practice of singling out large purchasers for "disparately favorable treatment " a practice which the court found no evidence of in this case. Id. The majority emphasized four points: (1) an absence of evidence that the six manufacturers sold to Boise on more favorable terms than they sold to smaller wholesalers; (2) lack of evidence that manufacturers framed the definitions of "wholesaler" to favor Boise method of doing business; (3) a dearth of evidence that Boise was a power buyer" or coerced the manufacturers into granting it favorable treatment in any way; and (4) no evidence of cabal between Boise and the six manufacturers to give Boise an additional price break. Id. at 1147. We address each of the court' s concerns below. (50) There is little value in focusing on the availability of the wholesale discount to distributors who meet each manufacturer s objective definition of a "wholesaler. " As the Ninth Circuit recently stated in Hasbrouck v. Texaco, Inc.

That all wholesalers were offered the same discount would be an appropriate defense in a case where the L selected dealers J and the other customers of the defendant were all wholesalers performing at the same level in the chain of distribution. Here, however, only the other customers are wholesalers; the Lsclccted dealersJ are retailers who are further down the chain of distribution. 842 F.2d at 1039. The favoritism that pervades this case is not the preference of one wholesaler over another, but of Boise, in its capacity as a dealer, over other dealers who could not purchase at the substantial wholesale discount. Boise purchased products from the six manufacturers at the "wholesale discount" that it resold to end-users in competition with the selected dealers.

Respondent' s " availabilty defense" was previously rejected by the Commission. 107 FTC at 215- 17. The premise of the defense is that there is no competitive injury "where the discounts are generally and practically available to competitors of the favored customer. . . . " 837 2d at 1130. Boisc argues that the fact that some distributors have Opinion 113 F.

chosen to accept certain risks in order to qualify for the wholesale discount, while the selected dealers have not, does not render the functional discounts "practically unavailable." BOR at 60. Boise suggests the discounts are not unavailable because the selected dealers need only to become dual distributors so as to qualify as wholesalers " under the manufacturers' definitions, which generally required that a distributor resell at least 20% of its office products to 65 We have previously held, however, thatdealers. BOR at 9- 44. discounts are not practically available where a buyer must alter his purchasing status (e. from a retailer as to a dual distributor) before receiving them. 107 FTC at 215; Dayton Rubber Co. 66 FTC 423 470- 71 (1964), rev d on other grounds sub nom. , Dayco Corp. v. FTC, 66 The Court of Appeals did not362 (51) F.2d 180 (6th Cir. 1966). question our reliance on Dayton Rubber as the law on availabilty. Indeed, it noted that " (dJue to the understandable reluctance of dealers to buy from their competitors, dealers rarely move into the class of dual distributors. " 837 F. 2d at 1131. The discount cannot be deemed " practically available and hence lawful-solely because any distributor who makes 20% of its sales to dealers would qualify as a "wholesaler. " Under Boise s theory, a manufacturer (either independently or in reliance upon a different standard adopted by WSA) could decide that a distributor must make 50% or more of its sales to dealers in order to qualify for a wholesaler discount. See Golder Tr. 1280. Obviously, at some point an objective standard that would require the selected dealers, in order to qualify for the wholesaler discount, to make a very large portion of their total sales to other dealers, would render such a discount practically unavailable to them.

The Commission and the Administrative Law Judge also rejected the availability defense on the ground that "dealers do, in fact perform wholesale functions, but do not receive the functional discounts that Boise enjoys" even when the dealers specifically request the discounts from the manufacturers. 107 FTC at 215- 16; 65 The Court of Appeals stated that manufacturers "typically" consider membership in the \Vholcsale Stationers Association ("WSA") as the first hurdle in qualifying for the discount, but that manufacturers differ in whatever additional services they require distributor stoperforminordertoqualify. Only about 50 of 100 wholesalers in the United States, however, actually were members of WSA. Seltzer Tr. 6587-88. 66 The wholesale discounts in this case were far less available than the "carload" discount in Morton Salt which small buyers were able to obtain by aggregating their purchases. In fact, less than one tenth of one percent of .\orton Salt s total salt business did not benefit from the carload-lot discount. 334 U.S. at 60 (J. Jackson, dissenting).

61 The fact that discounts are not practically available to ail customers, however, does not itself establish liability. Comp:aint counsei must establish all essential clements of a Robinson-Patman Act violatior,. BOISE CASCADE CORP. 999 956 Opinion accord 107 FTC at 184-86. The Court of Appeals observed, however that while marketing functions Boise performs (in its capacity as a dealer) for manufacturers through its virtually "autonomous " distribution centers are similar to those performed by other dealers in the industry, the Administrative Law Judge "did not expressly find that any of the individual dealers provide the quality and quantity of services that Boise does. " 837 F.2d at 1134 n.7. Nevertheless, we do not see a material distinction in the quality or quantity of functions performed by Boise and the selected dealers for the manufacturers. Complaint counsel' s expert witness, Dr. Nevin, agreed that "to the extent that Boise performs functions for manufacturers which are not performed by dealers, it is entitled to a lower price (Tr. 6989-90); however, according to Dr. Nevin (52) this situation does not exist in the present case (F. 522). " IDF 502 (Emphasis added). Dr. Nevin further testified that the "dealers could not obtain the discounts even if they performed the same functions as wholesalers (Tr. 6938-39). IDF 454. Again, Finding 522 supports the view that Boise and the selected dealers performed the same relevant functions for Rpurposes:

Based on his review of portions of the record, Dr. Nevin "Basically. . . found the dealers performing the same marketing functions that Boise performs their sales to commercial users " a conclusion with which I agree (F.'s 70 , 503 14).

(Emphasis added) IDF 522.

We do not understand Dr. Nevin s use of the word "same" to mean same-type but materially different" in comparing the marketing functions performed by Boise and other dealers. Our interpretation is bolstered by Boise s own expert, Dr. Elzinga, who agreed with the following statement:

On the other hand, to say that two buyers who perform the same services and buy in the same quantities should, on grounds of efficiency, receive different discounts solely because they resell to different customers also is nonsense (Tr. 6400).

IDF 519. The record does not support the hypothesis that the quality and quantity of the services provided by the selected dealers as a whole for the manufacturers materially differed from those provided by Boise. In short, the challenged discounts were mere subterfuges to Opinion 113 F.

avoid the restrictions of Robinson- Patman. See Texaco Inc. v. Hasbrouck 110 S.Ct. at 2546-47.

Whether or not the manufacturers specifically "framed" their definition of "wholesaler" to benefit Boise or other dual distributors the fact is that Boise receives the favorable discounts while other dealers do not largely because of its status as a large volume purchaser. In 1980 , for example, Boise received the "wholesaler discount on all its purchases of the products from the manufacturers in issue, even though seven of the eight distribution centers involved in this case made fewer than 20 percent of their sales to other dealers-the minimum criterion the court identified for receiving the wholesale (53) discount. 837 F. 2d at 1134 n.7; IDF 61. On this basis only six of the eight centers would have independently qualified for the discounts in 1980. IDF 61. Thus, while the proscriptions of Robinson-Patman are of general applicability and are not limited to large purchasers with market power Falls City, 460 U. S. at 436, the evidence in this case shows that Boise qualified for the favorable discounts by virtue of its aggregate wholesale sales throughout the country, rather than at each virtually autonomous distribution center. Finally, there is no requirement that a buyer have monopsony power or that discriminatory discounts be granted in secret in order to violate Robinson-Patman. We have previously acknowledged the limits of economic theory in Robinson- Patman jurisprudence. See supra at pp. 16- 17. In Morton Salt the Court of Appeals emphasized- . . . it is not contended that any of said discounts or the price system which embraces them all was formulated to give any particular customer or customers . . . a cost advantage over others who qualify therefor for quantity purchases. It is not claimed that there has been anything secret, local, special or personal about any such quantity discount as offered by petitioner to any of its customers. "

162 F.2d at 953. Nevertheless, the Supreme Court found liability. This holding is consistent with the language of Section 2(f) itself, which makes it unlawful for any person to "knowingly induce or receive unlawful discounts, and contemplates no exception for open and notorious price discriminations that are otherwise unlawful. 68 Of course, even jf Boise actualiy perfomed "certain functions, assuming all risks, investment, and costs :nvolved" to qualify for a legitimate functional discount on the goods itpu rchased as a wholesaler, it could not c:aim a func:oral discount on the goods it purchased for sale as a retailer.See Texaco Inc. 1). Hasbrouck, 110 Ct. at 2546- 47.

BOISE CASCADE CORP. 1001 956 Opinion D. The Manufacturers ' Discount Based on " Objective Criteria May Be the Basis of Section 2(f Liability The Court of Appeals deemed ironic that the six manufacturers now stand condemned as price discriminators" although they followed neutral, objective criteria in determining what buyers are to be considered wholesalers. " 837 F.2d at 1147. The court warned that the Commission was marching toward a regime of "price uniformity and rigidity " in " an all-out attack on uniform wholesale prices to dual distributors. Id.

Given the multitude of manufacturers involved and Boise s position as a leading buyer, it was both appropriate and a more efficient use of limited public resources for the Commission to proceed against Boise. Moreover, since most of Boise s distribution centers in issue would not have qualified for the wholesale discount based upon their own operations, the manufacturers' discounts appear to have been more neutral" on paper than they were in practice. Thus, we condemn the practice (54) of manufacturers sellng to Boise s distribution centers which are essentially autonomous dealers (837 F.2d at 1134 n. 7), at a substantial price advantage relative to the prices available to similarly situated dealers, over a sustained period of time, solely because Boise itself would qualify for the wholesale discount based on its aggregate sales.

IV. ORDER Paragraphs I through V of the Commission s order dated February , 1986 wil be reissued in the following form: This matter has been remanded to the Commission for further proceedings by the United States Court of Appeals for the District of Columbia Circuit. Boise Cascade Corp. v. FTC 837 F.2d 1127 (D. Cir. 1988). The remand directs the Commission to determine whether Boise Cascade Corporation s rebuttal evidence overcomes the inference of competition injury previously drawn by the Commission. Having reviewed the record and considered the briefs and oral arguments of the parties, the Commission has determined that Boise Cascade Corporation has failed to overcome the inference of competitive injury established by the evidence. Accordingly, It is ordered That Paragraphs I to V of the Commission s order dated February 11 , 1986 , are reissued. See Boise Cascade Corp. , 107 FTC 76 , 224-25 (1986).

, Concurring Opinion 113 F.

V. CONCLUSION Having reviewed the record, and particularly the evidence cited by the Court of Appeals, we hold that Boise has failed to overcome the inference of competitive injury established by the evidence that Boise purchased office products at sustained and substantial discriminatory prices from the manufacturers over a competitively significant period of time.

CO:\CTJRRING OPINION OF COMMISSIONER MARY L. AZCUENAGA This case is before the Commission on remand from the Court of Appeals directing the Commission to consider evidence that we previously deemed irrelevant to the issue of competitive injury. I agree with the majority that the respondent Boise Cascade Corporation Boise ) has failed to rebut the previously drawn inference of competitive injury and concur in the resulting conclusion that Boise knowingly received discriminatory prices in violation of Section 2(f) of the Robinson-Patman Act.

I do not join the opinion of the majority. To the extent that the majority may believe that the three opinions from the Court of Appeals fail to establish an adequate context within which to carry out this remand, we perhaps should have sought further clarification or review. To the extent that the analytical framework set forth by the majority may be inconsistent with the opinion of the court, that argument is beyond the scope of the remand. To the extent that I can discern new standards in the opinion of the majority, I disagree. For example, the majority advances the proposition that "a reading of the Court of Appeals decision to allow respondent to overcome the Morton Salt inference with evidence of the absence of th(eJ ' fact of injury' Las distinguished from causationJ would be inconsistent with the Supreme Court' s teaching in Falls City. Slip op. at 14. This proposition seems at odds with the plain statement of the Court of Appeals that the Morton Salt inference of injury can be overcome by evidence ''' breaking the causal connection between a price differential and lost sales or profits'" as well as " by evidence showing an absence of competitive injury. " 832 F.2d at 1144 (emphasis omitted). If "there is no competitive injury (or reasonable possibility of competitive injury) to begin with " the Court of Appeals said then evidence breaking the causal connection is obviously impossible to adduce. Id. Nor do I endorse the evidentiary standard, as I understand it, that ) BOISE CASCADE CORP. 1003 956 Concurring Opinion the majority would impose on Boise. According to the majority, Boise may prevail on remand if the evidence would "preclude the Commission from finding a reasonable possibility" that Boise benefited from the preferential discounts. Slip op. at 20. Preclusion is a demanding standard of proof. In addition, instead of focusing on evidence concerning absence of competitive injury, a critical element of a statutory violation and the matter at issue on remand, the majority proposed standard would require Boise to prove the absence of benefits from a discriminatory price. The majority is closer to the mark when it suggests that Boise need show only that "the essential elements of (competitive) injury have not been established by a preponderance of the evidence. " Slip op. at 19 n. 33. (21 said that the Morton Salt inference of The Court of Appeals competitive injury drawn from proof of substantial, sustained price differences can be overcome by " (s)pecific, substantial evidence of the absence of competitive injury" and directed us on remand to "sift and weigh Boise s evidence of absence of injury" to determine whether it demonstrates that there is no competitive injury (or reasonable possibilty of competitive injury. Boise Cascade Corp. v. FTC, 837 2d 1127 , 1144 (D. C. Cir. 1988), reversing and remanding Boise Cascade Corp. 107 FTC 76 (1986). Having accepted the remand and forgone further attempts to obtain review of the issue of which evidence is relevant to rebut the inference of injury, our task on remand is narrow.

relies on Morton Salt and its For the most part, the majority progeny to conclude that the evidence identified by the Court of Appeals is "not probative" of the lack of injury. The Court of Appeals required us on remand to examine the evidence, not to explain again why we believe that under the case law the evidence proffered by Boise does not matter. Nowhere in the opinion does the majority say the that the evidence does not show what Boise says it shows. Instead, conclusions of the majority regarding the evidence come perilously close to repeating the Commission s original position that the evidence is irrelevant but substituting the phrase "not probative" for "irrelevant. " Because relevance usually is defined by reference to probative value-relevant evidence is evidence that tends to make the existence of any fact more or less probable than it would otherwise be, Fed. R. Evid. 401; McCormick on Evidence 9185 (3d ed. 1984)-the approach of the majority may beg the question.

1004 FEDERAL TRADE COMMISSION DECISIOXS Concurring Opinion 113 F.

The Court of Appeals directed us to consider evidence proffered by Boise of "high profitability and financial health on the part of the selected dealers" and of the absence of diverted sales attributable to the wholesale discount granted to Boise. 837 F. 2d at 1144-45. Boise asserts that the "evidence established that the selected dealers enjoyed increasing sales and profits and ' a virtually complete absence of sales lost to Boise' . . . thereby precluding a finding of competitive injury. " (3) R.R.B. at 20-21. The majority concludes that Boise evidence is of insufficient probative value to overcome the inference of competitive injury. Slip op. at 25. As I read the record, the evidence does not show what Boise says it shows.

Injury within the meaning of the Robinson-Patman Act may be found although the businesses of the disfavored purchasers remain financially viable and the favored purchaser does not win all accounts from its disfavored competitors. The Act in secondary line cases does not require the destruction of competition, nor are diverted sales a necessary element of injury in a government injunction action, and complaint counsel expressly disclaimed any intent to make their case , 1984).under these standards. Oral argument tr. at 62-65 (June 20 The opinions of the Court of Appeals do not appear to require application of these standards but rather consideration in the existing context of "whether Boise s evidence demonstrated that no injury or reasonable possibility' of competitive injury existed. " 837 F.2d at 1144. An examination of Boise s evidence shows that it is consistent with the Commission s inference of competitive injury. The theory of competitive injury in this case is "that the competitive opportunities of certain merchants were injured when they had to pay . . . substantially more for their goods than their competitors had to pay. FTC v. Morton Salt Co. 334 U. S. 37 , 46- 47 (1948). Boise, a 2 obtains wholesale discountssubstantial reseller of office products from office product manufacturers on goods that Boise resells directly ) The following abbreviations are used in this concurring opinion: Initial Decision Finding I.D.F. RLE. Respondent s Initial Brief (May 18 , 1988) R.R.B Respondent's Reply Brief Complaint Counsel's Answering Brief RX Respondent' s Exhibit Complaint Counsel' s Exhibit. 2 Boiseexs combined sales ($346. 6 milion in 1980 , LD.F. 58 in camera) make it the largest distributor of offce products in the United States, LD.F. . , and it is Doe of the two largest wholesale distributors ($196. milion in 1980 LD.F. .58 in camera) in the United States. LD. F. 53. ) ( BOISE CASCADE CORP. 1005 956 Concurring Opinion to large commercial and industrial end users in competition with dealers that also buy directly from manufacturers and sell to the same (4) accounts. The direct-buying dealers ' pay the manufacturers 5% to 33% more for office products than Boise pays for the same products. 107 FTC at 182. The complaint challenges as unlawfully discriminatory the wholesale discounts that Boise obtains on goods resold to end users in competition with the direct-buying dealers; it does not challenge the wholesale discounts to the extent that Boise resells as a wholesaler.

Boise does not dispute that it competed with the direct-buying , Boisedealers. I.D.F. 96-100. In its 1976 and 1977 business plans specifically targeted its sales effort at the large customers served by in camera; CX44H in camera.the direct-buying dealers. CX34B Although Boise received wholesale discounts on all its purchases from the six selected manufacturers, seven of the eight Boise distribution centers examined in this matter made more than 80% of their sales directly to end users. 5 The direct-buying dealers, because of the outset at ahigher price they had to pay, were placed at the competitive disadvantage with respect to Boise. A. Sales and Profits Boise asserts that the direct-buying dealers were "competitively healthy and financially prosperous " R. B. at 4 , citing testimony that the sales of the direct-buying dealers increased more than 22% annually between 1977 and 1980 and claiming that this growth was unusual, particularly in light of (5) the recessionary economy. B. at 21. 6 To support its claim that the growth of the selected dealers was "unusual " Boise asserted on remand that "Boise Cascade s growth rate was 19 percent " oral argument tr. at 63, and that dealer-members of the National Office Products Association NOPA" experienced an average annual growth rate of 15%. 3 To prove a vio:alion, complaint counsel presented evidence concerning twenty-three dealers that also buy directly from manufacturers and compete with Boise for sales to large commercial end users. LD. F. 88 & 96. The selected dealers are referred to throughout this opinion as "direct. buying dealers, 4 Boise piarmed "selective expansion of its direct sales efforts to compete for business against dealers who buy a significant percent of their requirements from manufacturers 4% of the dealers do more than 00Q 000 of volume. .. but buy only 15% of their requirements from wholesalers ). Boise limited its direct or "commercial" sales to large end users "to minimize conflicts" with dealer-customers that buy from Boise and other wholesalers. CX34B n camera; see also CX37D in camera. s Ofthe eight, only Boise s center in Philadelphia, with 47. 1 % ohales to end users, made less than 80% of its 19fW sales directlv to end users. LD.F. 61 in camera. (; Contrary to B ise s argument see R.R.B. at 21 , the Administrative Law Judge did not find that these assertions were true but only that two of Boise s witnesses, Mr. Bertholdt and Dr. Elzinga, had so testified. See !.D, F. 431" V..

Concurring Opinion 113 F. B. at 26. These growth figures do not stand up on examination. According to Boise s 1981-1985 Business Plan, its commercial business (in competition with the direct-buying dealers) grew at an average annual rate of 26.4% in 1977- 1979. CX5IO in camera (data reprinted at page 6 below). 8 In addition, the 15% NOPA growth rate asserted by Boise in its brief on remand is lower than the NOP growth rate that appears in the record in Boise s business planning documents. 9 (6) Boise s Business Plan for 1981- 1985 provides an additional context for comparing the 22% rate of growth that Boise claims the directbuying dealers enjoyed between 1977-1980. The following statistics showing year-to-year growth are taken from that plan ;!O NOPA Boise Dealers Direct Sales 1977 17. 23.

1978 18. 37.

1979 21.3 18.

These data indicate, assuming, as Boise asserts, a 22% rate of growth for the direct-buying dealers, that Boise s growth was in the same range and even somewhat higher (three-year average of 26.4%). In Boise s estimation, office product dealers in general (three-year average of 19. , based on the above column for NOPA dealers) fell short of Boise s performance. See also CX40J- in camera (showing ational Office Products Assoriation, a major ifJdustry trade association, the members of which are offce product manufacturers, wholesalers and dealers, compiles and publishes annually data concerning dealer operations. LD. F. 423.

B The document cited by Boise s counse! to show a 19% growth rate for Boise does not show data for Boise commercial sales but rather shows that Boise s overall sales grew 21. 4% in 1977 , 26. 2% in 1978 and 17.3% in 1979 (an average of 21.6%). CX51Z1 in camera. Boise estirnatf'd in August 1980 that its overall average annual rate ofgrawth for 1976- 1980 would be 19. 26%, but this does not appear to be a simple average and the method of calculation is not explained. In addition, thisestima tewasbascdinpartonsalesfor1976 forwhich we lack figures, for purposes of comparison, for the direct- buying dealers (in 1976, Boise s sales increased 17.3%) and 1980, for which Boise lacked actual data. g Boise s business plans cited I\OPA growth rates of 18% in 1978 , 21.3% in 1979 and 12. 2% for 11 months in 1980. CX510 in camera; CX43H in camera, The NOPA growth rate of 15% that Boise asserted in its brief was nut ca,culated by KOPA but was "determined (presumably by Boise or its counsels by multiplying the average sales per employee, which is in the OPA data, by the total number of employees of an average dealer " allegedly also in the NOPA data, Oral argument tl" at 62. The NOPA surveys in the record, however do not show "the total number of employees of an average dealer " as counsel claimed, but instead a "typical personnel structure" This is not necessarily the same as the total number of employees, and Boise makes no showing that the two are eq'Jivalent.

10 CX510 n camera, Although the majority states that "the record is devoid of evidence of probative industry-w:de growth standards " slip op. at 30 nab, these data in Boise s 5-year plan provide evidence on gromh tr.at undermines Boise s claim that the selected dealers' growth was "unusua!" during a " recessionary pel' lor:.

, , BOISE CASCADE CORP. 1007 956 Concurring Opinion Boise s growth in sales exceeding NOPA growth). ll This evidence dealerstends to show that office product dealers and the direct-buying did not fare as well as Boise, which is consistent with the inference of sustained price discrimination.competitive injury from substantial, Boise s own documents also belie its claim that the growth of the selected dealers was unusual "in light of the recessionary economy, See LD.F. 431. In addition to the growth figures cited above, Boise business documents indicate that the office products industry was not adversely affected by a recession in the years for which growth figures were presented. Boise s 1980-1984 Business Plan (dated August 1979) stated that " (tJhe growth of white collar employment continues to be the stimulus for office supply consumption growth. According to Boise, in the preceding six years the dealer to consumer growth of the industry has been at a compounded rate of 14.6 percent; wholesale growth has been growing at a 21.6 percent rate over the same period. Over the longer term " Boise "anticipated that these growth rates wil be similar. CX50E in camera. One year later (7) in August 1980, Boise stated that "(iJndustry growth " CX510 in camera.continues to look attractive. In addition to the growth data, the record contains net profit data for the direct-buying dealers, for Boise, for NOP A dealer-members and for retail offce product dealers (calculated by Robert Morris Associates). These data also indicate that the direct-buying dealers and office product dealers in general were not as profitable as Boise: 1976 1977 1978 1979 1980 1981 Boise Dealers 13 2.4 NOPA Morris As the Court of Appeals pointed out these statistics pose obvious comparative problems " 837 F.2d at 1144 , because they were prepared by different persons for different periods. Despite these differences, the data for the direct-buying dealers, the NOP A dealers and the Morris study are similar. The NOP A data and the Morris 11 Boise described the KOPA report of "year-ta-year sales growth of dealers to final onsumers" as "the most significant" measure. CX401 in camenL. 12 CX51Z1 in camera (estimated figures fur 1980 and 1981). 13 CX2301 (data for the direct- buying dealers, as calculated by complaint counsel's expert accountant). CX356 at 4; CX355 at 5.

CX23QO Concurring Opinion 113 F.

study have been used for more than 50 years by the industry and by lending institutions for comparative purposes, and Boise regularly cited NOP A data in its business planning documents. These are significant indicia of their reliabilty. See LD.F. 427 & 429; 107 FTC at 182 & n. 15. In general, the data show that the net profits of the direct-buying dealers and of office product dealers declined between 1976 and 1980. During the same period, Boise s net profit increased steadily, and, except for 1976 , Boise s net profit always exceeded those of both the direct-buying dealers and NOPA dealer-members. This is consistent with Boise s belief that its "growth has exceeded NOPA growth" and that its "sales growth on commercial sales has exceeded industry rates. " CX40I in camera. The net profit data also are consistent with the inference of competitive injury from discriminatory prices. (8) The Court of Appeals suggested that the Commission should consider the " apparent discrepancy in profitability" between the NOP A dealer-members who were surveyed and the direct-buying dealers and " analyze the relationship of Boise s higher net profit (which, of course, reflected both its sales as a wholesaler and a dealer) to that of the selected dealers. " 837 F. 2d at 1144-45. The slight difference between the profits of office product dealers as shown in the NOP A survey and the net profits of the direct-buying dealers (calculated by complaint counsel's expert accountant) may be explained by the fact that the NOP A survey includes dealers that sell through retail, walk-in stores: retail sales usually are made at full list price while sales to commercial end users usually are made at some discount from list price. Boise s higher net profits are entirely consistent with its favorable prices from office product manufacturers: Boise s cost-of-goods-sold was lower than that of the selected dealers.

B. Diversion of Sales Evidence that sales have been diverted from the disfavored to the favored customer as a result of discriminatory prices usually is 16 Additional data in the record also are consistent with this conclusion: For example, the return on investment for :-OPA dealers was 20.35% in 1977 , 20. 55% in 1978, 22.77% in 1979 and 19. 10% in 1980. CX355 at 5; CX356 at 4; CX357 at 4. Boise s return on investment was consistently higher: 22.8% in 1977 24.4% in 1978 , 28.8% in 1979 and 31.9% in 1980. CX51Z1 in camera; CX43H in camera. :7 Although Boise enjoyed substantial price advantages, it worked to improve its margins "by better buying and pricing" and " more effective purchasing. . . to reduce material costs and keep the majority of the savings as profit." CX44H & 44Z37 1:n cam.era, Boise s eentral purchasing organization " implementledJ bTfOSS margin improvement g-oals (andl. . more effective vendor negotiations." CX37J j cu,mem. g , BOISE CASCADE CORP. 1009 956 Concurring Opinion required in private cases under Section 4 of the Clayton Act to support a claim for damages. 18 As the Court of Appeals pointed out, diversion of sales is "another form " that competitive injury could take. 837 F. at 1145. A showing of diverted sales is not necessary to support injunctive relief under the Robinson-Patman Act, because the Act does not require that the discriminatory prices in fact have harmed competition. It is sufficient for a finding of liability and the imposition of injunctive relief "that there is a reasonable possibility " that the 12; seeprice discrimination may harm competition. 837 F.2d at 1139 n. 107 FTC at 181-82. The distinction between actual injury and the reasonable possibility of injury is (9) important, because this case was not tried on an " actual injury" theory. Complaint counsel made no claim to have shown direct evidence of competitive injury, oral argument tr. at 65 (June 20, 1984), and the Commission in 1986 expressly rejected Boise s claim that actual injury must be proved. 107 FTC at 208.

The record does support the specific findings of the Administrative Law Judge (subsequently adopted by the Commission, 107 FTC at 201) that accounts switched from the direct-buying dealers to Boise because of Boise s lower prices, better service, or a combination of these factors" and that some of the direct-buying dealers believed based on their experience, that they could not economically compete against Boise and, therefore, did not bid for some business. LD.F. 384 & 406. The Administrative Law Judge concluded that the directbuying dealers were able to underbid Boise only "in the face of (a substantial price disadvantage " and their success "was accompanied by a significant impairment of profits." 107 FTC at 182. Boise in its business planning documents suggested that the directbuying dealers were able to compete for sales to large commercial accounts-despite Boise s price advantage-because they were more efficient than Boise: their " overheads are generally lower than Boise Cascade, hence the tendency to accept lower average gross margins than (Boise J can tolerate. " Boise also believed that the direct-buying dealers might have lower standards for return on investment than supra. Boise s formerBoise did. " CX44E in camera; see note 16 18 See, e, J. Truett Payne Co. v, Chrysler Iotors Corp.. 451 U. S. 557 , 562 (1981). As the majority points out, the finding of the Administrative Law Judge that the direct-buying dealers CQu1d not conclude that thelosses wefe due to the discriminatory prices obtained by Boise, LD.F. 407 quoted at 837 F.2d at 1115, is not surprising, because the dealers were not in a position to know Boise s costs, Boise bids or why an account shifted to Boise.

20 Boise focm;ed on larger commercial accounts, choosing not to compew for smaller aCl'ounts to minimize conflicts with IT"any small dealers who support who:esa\ers and to improve the profiabilty of such accounts (foot!',oteeofJ Concurring Opinion 113 F.

general manager testified that Boise likely had higher wage and pension costs than the direct-buying dealers and that Boise "may elect to have better facilities" than the dealers. CX672ZI2- 13 in camera; CX44H in camera. In addition, Boise did not regularly exploit its price advantage by offering lower prices. Instead, Boise believed that (s)ales are more responsive to service and in-stock condition than to price. " CX44Z19 in camera. This evidence is entirely consistent with the Administrative Law (10) Judge s conclusion that the direct-buying dealers faced a su bstantial price disadvantage in competition with Boise and were able to compete with Boise only with "a significant impairment of profits.

The alleged longevity of the wholesale discount in the office products industry is a related point. The Court of Appeals suggested that" a price difference with the potential for causing injury will eventually result in measurable effects, such as diverted sales or lost profits. " 837 F.2d at 1146 n. 17. Boise asserts that the issue in this case is "the competitive effect of the manufacturers ' decades-old practice of voluntarily providing the functional discount to dual distributors" and that any competitive injury must be judged on the basis of a "forty year track record" of functional discounts. R.R.B. at 39-40. But the discriminatory prices challenged here are of more recent vintage: Boise did not begin obtaining a wholesale discount for direct sales to commercial accounts in competition with the directbuying dealers until after its "acquisition program in 1964- 1966. CX44A in camera.

In addition, Boise s recognition that its competitors for direct sales see CX44Ewere direct-buying dealers, not other wholesalers 21camera seems to suggest that the "practice of voluntarily providing the functional discount to dual distributors" may not be "decades old" but may have originated with Boise. 22 By the middle to late 1970' the period examined in the record, Boise was still in the process of e.defining and implementing its dual distribution policy. 23 See, CX34B n camem, and to eliminate " unprofitable accounts. " CX37D in cmnem; see also CX54A. fn camera; CX78V in came-r; CX88Z24 in camera; CXIOIJ in camera. 21 Boise estimated that it was in 1978 the largest seller to the commercial market. Boise identified directbuying dealers, not other wholesalers, as its competitors for direct sales to commercial cnd users.See CX400 w camera 22 Boise apparently was the only one of the " Big 5" wholesalers to pursue a national dual distribution policy. Boise described Champion and S.P. Richards as " pure wholesalers " CX45E- in camera and Zellerbach maintained a "tradition of wholcsaie. " CX40F hI camera. Although Boise asserted that United is a dual distributor, CX45E in camera Boise did not identify any non-wholesale sales for United, CX51Z34 in cu.mem and there is no showing that Cnited obtained a wholesale discount on products sold directly to end llsers. 2:1 The complaint in this case issued in April 1980 BOISE CASCADE CORP. 1011 956 Coneul'l'ing Opinion CX37D in camera; CX54A- in camera. The Court of Appeals raised the concern that Boise may not have been singled out for "disparately favorable treatment." 837 F. 2d at 1147. The record evidence suggests that Boise was indeed so favored as compared to the direct-buying dealers with which it competes. (11) II.

The Comt of Appeals expressed two additional concerns: that Boise is not a "power buyer exerting its muscle to win additional pricing concessions" and that the manufacturers, by implication if Boise violated Section 2(f) of thc Act, have "run afoul of Robinson-Patman even though they have followed neutral, objective criteria" in granting wholesale discounts. 837 F.2d at 114 7. The record evidence leaves the first concern largely unanswered. Although Boise is one of the largest purchasers and resellers in the offce products industry, complaint counsel made no attempt to prove that Boise has power in any market believing that such a showing was legally unnecessary. Oral argument tr. at 48 (June 20 , 1984). \ 1f this were held to be a necessary element in a Section 2(f) case, the record would not sustain a finding of liability. With respect to the concern about the liability of the manufacturers, because the record shows that Boise obtained discriminatory prices on goods sold in competition with the direct-buying dealers, 1 agree with the majority that "the manufacturers' discounts appcar to have been more ' neutral' on paper than they were in practice. " Slip op. at 53.

III.

The evidence proffered by Boise is consistent with the Commission previous conclusion that the discriminatory prices obtained by Boise caused, or reasonably might cause, competitive injury. " 837 F.2d at 1129. Therefore, I agree with the conclusion and the order of the majority.

ClJRRI G OPII\JO OF COMMISSJONER ANDREW J. STRF:NIO , JR. I wish to emphasize that in voting to approve this Opinion and cl T!w S'.!l1'1l Court l as said that " I,lJ thougn ()nc('rrlS about ,)10 eXfessivf' mal" kct power" of larg-e plll' ('il ISl' C'-Io prinwl' i:y l'cspol1siiJle fot" passage of t1w Robil1son-Patman Art the Art is ' of gt'nrn, aIJpiic' a!Jili; y alia p1'ohibi:s d:sl'l'il1ini :ons gencmlly Falls Ci!!JhldIiS!I"'fS, hlf. c, (luilro llrl' eruyr Jill' 160 US 4:28. -136 (l9S3).

Concurring Opinion 113 F.

Order, my deliberations were confined solely to the single issue remanded by the Court of Appeals, namely whether Boise Cascade Corporation ("Boise ) has overcome the inference of competitive injury. On this specific question, my answer is in the negative for the reasons explained at length in the Commission s Opinion. In brief after taking into account all the evidence presented by Boise and applying the standard set forth in footnote 33 of the Opinion, I find that the essential elements of injury have been established by a preponderance of the evidence.

I also agree with the additional reasons articulated by Commissioner Azcuenaga for supporting the Order. Some of these grounds include her analysis of Boise s sales and profits data, her comparison of that data with the sales and profits data of National Office Products Association dealers generally and the 23 direct-buying dealers in particular, and her explication of why there is a reasonable possibility that the disfavored dealers suffer competitive harm as a result of the price differentials.

EW ENGLA'iD MOTOR RATE BUREAU , I"IC. 1013 1013 :Modifying Ordcl'