In Re Beef Industry Antitrust Litigation, 542 F. Supp. 1122 (N.D. Tex. 1982)
June 14, 1982
United States District Court, N. D. Texas, Dallas Division.
*1123 *1124 Ben L. Krage, Kasmir, Willingham & Krage, Dallas, Tex., Lowell V. Summerhays, Robinson, Summerhays, Wells & Barnes, Salt Lake City, Utah, Robert R. Eidsmoe, Gleysteen, Harper, Kunze, Eidsmoe & Heidman, Sioux City, Iowa, Burt A. Braverman, Cole, Raywid & Braverman, Washington, D. C., Lex Hawkins, Hawkins & Norris, Des Moines, Iowa, John A. Cochrane, Cochrane & Bresnahan, St. Paul, Minn., Robert O. Vaughan, Vaughan, Hull, Marfisi & Miller, Elko, Nev., James W. Witherspoon, Witherspoon, Aikin & Langley, Hereford, Tex., Roger J. Allen, Strother, Davis & Levy, Dallas, Tex., Joseph M. Alioto, Law Offices of Joseph M. Alioto, San Francisco, Cal., C. R. Kit Bramblett, El Paso, Tex., Roland D. Saul, Saul, Smith & Davis, P. C., Hereford, Tex., Bill LaFont, LaFont, Tunnell, Formby, LaFont & Hamilton, Plainview, Tex., W. Randolph Elliott, Riddle & Brown, Dallas, Tex., for plaintiffs.
John D. Leech, Calfee, Halter & Griswold, Cleveland, Ohio, for First Nat. Stores, Inc.
David A. Rosen, Stein Rosen & Ohrenstein, New York City, B. Thomas McElroy, White, McElroy, White, Sides & Rector, Dallas, Tex., for Food Fair Stores, Inc.
Bernard Gordon, Finley, Kumble, Wagner, Heine, Underberg & Casey, Washington, D. C., for Giant Food, Inc.
Shepard Goldfein, Skadden, Arps, Slate, Meagher & Flom, New York City, William H. Tinsley, Johnson & Cravens, Dallas, Tex., for the Grand Union Co. Colonial Stores Inc.
*1125 Thomas F. Curnin, Cahill, Gordon & Reindel, New York City, for the Great Atlantic & Pac. Tea Co., Inc.
Theodore A. Groenke, McDermott, Will & Emery, Chicago, Ill., for Jewel Companies, Inc.
Alexander E. Bennett, Arnold & Porter, Washington, D. C., Wilson W. Herndon, Strasburger & Price, Dallas, Tex., for the Kroger Co.
C. Loring Jetton, Jr., Wilmer, Cutler & Pickering, Washington, D. C., Gregory Huffman, Thompson & Knight, Dallas, Tex., for Lucky Stores, Inc.
James F. Rill, Collier, Shannon, Rill & Scott, Washington, D. C., John L. Hauer, Akin, Gump, Hauer & Feld, Dallas, Tex., for the Nat. Ass'n of Food Chains.
Valerie A. Leopold, Kirkland & Ellis, Chicago, Ill., Lawrence G. Newman, John Shook, Newman, Shook & Newman, Dallas, Tex., for Nat. Provisioner, Inc.
Franklin P. Auwarter, Mayer, Brown & Platt, Chicago, Ill., for Nat. Tea Co.
Harold A. Segall, Gilbert, Segall & Young, New York City, for Pueblo Intern., Inc.
Les J. Weinstein, McKenna & Fitting, Los Angeles, Cal., for Ralphs Grocery Co.
Richard W. Odgers, Pillsbury, Madison & Sutro, San Francisco, Cal., W. B. West, III, Clark, West, Keller, Butler & Ellis, Dallas, Tex., for Safeway Stores, Inc.
Donald B. Holbrook, Jones, Waldo, Holbrook & McDonough, Salt Lake City, Utah, for Skaggs Companies, Inc. Skaggs-Albertson's.
Robert D. Paul, Goodwin, Procter & Hoar, Boston, Mass., Stanley E. Neely, Locke, Purnell, Boren, Laney & Neely, Dallas, Tex., for the Stop & Shop Companies, Inc.
Michael E. Bress, Dorsey, Windhorst, Hannaford, Whitney, & Halladay, Minneapolis, Minn., for Super Valu Stores, Inc.
Jerome G. Shapiro, Hughes Hubbard & Reed, New York City, T. L. Caudle, III, Coke & Coke, Dallas, Tex., for Supermarkets General Corp.
Timothy J. Sargent, Bodkin, McCarthy, Sargent & Smith, Los Angeles, Cal., for Thriftimart, Inc.
Ronald L. Olson, Munger, Tolles & Rickerhauser, Los Angeles, Cal., for Vons Grocery Co.
Kenneth Gordon, Tenzer Greenblatt Fallon & Kaplan, New York City, for Waldbaum Supermarkets, Inc.
Charles P. Pillans, III, Bedell, Bedell, Dittmar & Zehmer, Jacksonville, Fla., John K. DeLay, Jr., Storey, Armstrong, Steger & Martin, Dallas, Tex., for Winn-Dixie Stores, Inc.
Roger E. Podesta, Debevoise, Plimpton, Lyons & Gates, New York City, for Flavorland Industries, Inc.
Edward W. Rothe, Freeman, Rothe, Freeman & Salzman, Don H. Reuben, Carole K. Bellows, Reuben & Proctor, Chicago, Ill., for Iowa Beef Processors, Inc.
John C. Dods, Shook, Hardy & Bacon, Kansas City, Mo., for MBPXL Corp.
Morris Harrell, Marshall M. Searcy, Rain, Harrell, Emery, Young & Doke, Dallas, Tex., for Iowa Beef Processors, Estate of Currier Holman and MBPXL Corp.
Eugene M. Warlich, Doherty, Rumble & Butler, St. Paul, Minn., for Spencer Foods, Inc.
Wilson W. Herndon, Strasburger & Price, Dallas, Tex., for defendants in retail litigation.
Jess B. Hawley, Hawley, Troxell, Ennis & Hawley, Boise, Idaho, for Albertson's Inc.
Benjamin M. Quigg, Jr., Morgan, Lewis & Bockius, Philadelphia, Pa., Dudley Chambers, Jackson, Walker, Winstead, Cantwell & Miller, Dallas, Tex., for American Stores Co. f/k/a Acme Markets, Inc.
Roy Lewis Shults, Mitchell, Silberberg & Knupp, Los Angeles, Cal., for Arden-Mayfair, Inc.
*1126 Eugene Driker, Barris, Sott, Denn & Driker, Detroit, Mich., for Borman's, Inc.
E. Link Beck, Kemp, Smith, White, Duncan & Hammond, El Paso, Tex., John A. Gilliam, Jenkens & Gilchrist, Dallas, Tex., for the Circle K Corp.
PATRICK E. HIGGINBOTHAM, District Judge.
The task of applying the legal doctrine that forbids both offensive and defensive use of pass-on is presented a second time in this litigation. The prohibition was born of a familiar tension between finite proof and convenience and adjusted by the Supreme Court on the side of convenience. In this task, we are reminded that the relationship of antitrust law and economic theory continues to be a sometime thing suffering from both lack of commitment and an inability (or unwillingness) to communicate, each with the other. The choice of immediate over remote purchasers as preferred claimants, despite its purpose of avoiding undue complexity and speculation, thereby presumably aiding the treble damage suit, has produced a counterproductive result in the hands of the lower courts. The result is that remote plaintiffs claim to be ready to prove a pass-on completely confident that their proof can be sufficiently detailed without freeing the bogeyman of econometrics. Yet, ensnared by adherence to procedural regularity still adjusted for the finite, the goal of reducing complexity is frustrated by the effort spent in deciding if the proof would be unduly complex or speculative. In the long view, this short-term frustration may be the price for the clarity necessary to implement the goal. We will see.
I. Background of the Case
Cattle feeders who fatten and sell cattle to meat packers allege that retail grocery chains conspired to purchase beef from the packers at artificially depressed prices in violation of sections 1 and 2 of the Sherman Act, 15 U.S.C. §§ 1, 2. The feeders claim that such depressions in wholesale beef prices were reflected in the Yellow Sheet, a commercial daily beef price reporting service. They allege that the packers applied a fixed formula based on Yellow Sheet beef price quotations to compute the actual prices paid for cattle and that the entire alleged depression was "passed on" to the feeders and not suffered at all by the packers in a manner that functioned as an equivalent to a cost-plus contract. The pass-on was allegedly aided by the fact that cattle kept in feed lots had to be brought to market within a short period to prevent overfattening. The feeders maintain that this claimed inelastic supply of cattle provides the fixed quantity term of a cost-plus equivalency, see note 7 and accompanying text infra.
Feeders, plaintiffs, do not deal directly with the retailer defendants. Instead, they sell their cattle to packers who slaughter the cattle and convert them into beef and by-products. The packers then sell the beef (i) directly to food retailers, including defendants, or to others at the retail level (e.g., hotels, fast food chains, restaurants, and the government), and (ii) to other middlemen, such as beef jobbers, fabricators and other packers, who may process the beef further and in turn sell it to retailers. Packers do not sell by-products to the retailer defendants. A separate competitive market exists for the by-products.
The chain of distribution is as follows: *1127
That the feeders do not sell directly to the retailers poses a high hurdle to their suit against these "indirect" purchasers, a hurdle constructed by Hanover Shoe, Inc. v. United Shoe Machinery Corp., 392 U.S. 481, 88 S. Ct. 2224, 20 L. Ed. 2d 1231 (1968), and Illinois Brick Co. v. Illinois, 431 U.S. 720, 97 S. Ct. 2061, 52 L. Ed. 2d 707 (1976). The issue now before this court is whether the feeders can successfully tread the narrow path around this hurdle.
In Hanover Shoe, the Supreme Court decided that an antitrust defendant in a suit by a direct purchaser cannot assert a "passon" defense, claiming that the direct purchaser passed on any injury from illegal overcharges to its customer, the indirect purchaser. Concerned that the practical *1128 utility of private suits as an enforcing mechanism might be diminished by the complexity of proof and that indirect purchasers with "a tiny stake in the lawsuit" would not have the incentive to sue, the Court rejected the pass-on defense. 392 U.S. at 494, 88 S. Ct. at 2232.
Having rejected the defensive use of passing on with one exception, see page 1128 infra, the Court in Illinois Brick confronted its reciprocal and rejected the offensive use of a pass-on theory by an ultimate consumer against a remote seller. The Court noted that allowing offensive but not defensive use of pass-on would create a serious risk of multiple liability, 431 U.S. at 730, 97 S. Ct. at 2067, and that the offensive use of pass-on would present even more serious evidentiary complexities and uncertainties than its defensive use. Id. 431 U.S. at 732, 97 S. Ct. at 2067-68.
Both Hanover Shoe and Illinois Brick recognized a narrow exception to the pass-on bar: "when an overcharged buyer has a pre-existing `cost-plus' contract, thus making it easy to prove that he has not been damaged...." 392 U.S. at 481, 88 S. Ct. at 2226. See also 431 U.S. at 736, 97 S. Ct. at 2070. In that situation, the uncertainty *1129 of proof and the risk of multiple liability are demonstrably absent. Id.
Believing Illinois Brick controlled, Judge Taylor of this court before whom this MDL case was then pending dismissed the feeders' suits against the retailers. The court of appeals reversed, concluding that the complaints stated a claim within the "cost-plus" exception, or more precisely, that the feeders claimed a "functional equivalent" to a cost-plus contract. See In re Beef Antitrust Litigation, 600 F.2d 1148 (5th Cir. 1979), cert. denied, 449 U.S. 905, 101 S. Ct. 280, 66 L. Ed. 2d 137 (1980). It explained:
The complaints sufficiently allege that the impact of the retail chains' price changes upon the pricing decisions of the packers is determined in advance without regard to the interactions of supply and demand. The plaintiffs allege that the packers set the price of live cattle by strictly applying certain formulae to the Yellow Sheet or Safeway wholesale beef price. Under these allegations a plaintiff would be entitled, once he proved what the competitive wholesale price would have been for a given grade of beef in a given region at a given time, and once he established that the packer to whom he sold strictly applied a formula to the Yellow Sheet price for the particular sale, to damages in the amount of the difference between the price he actually received on that sale of fat cattle and the price he would have received absent price-fixing (computed by applying the packer's formula to the constructed competitive wholesale price). The packer's habitual use of predetermined formula would enable measurement of the effect on prices for fat cattle or changes in wholesale prices.... It is alleged that the intermediaries in these cases determine their purchase price for fat cattle by rigid application of formulae to the wholesale price for carcass beef. Assuming the allegations of the complaints to be true, a packer's absorption of a loss resulting directly from a decrease in the wholesale price would be detectable from a change in his purchasing formulae or from his paying a price for cattle that is above the price he would have paid had he adhered to his customary formulae. It is alleged, moreover, that the packers have no reason to depart from their purchasing formulae in the short run because short term supply is, by the very nature of cattle production, highly inelastic.
600 F.2d at 1165.
The court of appeals decision was made in the procedural context of reviewing a successful rule 12 frontal assault by the retailers. Unbridled by an evidentiary record, the feeders were most optimistic in their statements to the court of appeals of their intended proof. While generously credulous, the court of appeals invited the retailers to raise the pass-on issue in the more disciplined mode of a motion for summary judgment. Not surprisingly, the invitation has not been ignored. Equally anticipatable, the hoped-for proof of the feeders' counsel and reality have parted company.
II. Summary Judgment Standards and Burden of Proof
Following its statement that the "defendants will have the opportunity to raise [the pass-on issue] by summary judgment motion," the fifth circuit directed:
Given the strictures of Hanover Shoe and Illinois Brick, the district court may and should demand from the plaintiffs in each case a pretrial demonstration that they have definite and particularized proof that they will need to establish damages.
600 F.2d at 1166-67. The retailers assert and the feeders conceded at oral argument that under the fifth circuit opinion, the feeders "may proceed to trial only if they now demonstrate with certainty that the packers were immune from the interactions of supply and demand and purchased cattle in a manner, functionally equivalent to a cost-plus contract, that guaranteed in every instance that the entirety of any alleged depressions in wholesale beef prices was always passed back to plaintiffs." See Defendants' Summary Judgment Memorandum at 5. In particular, the retailers maintain that the fifth circuit required that the feeders prove (recast in the procedural context *1130 of rule 56, the retailers are entitled to prevail if they show there is no genuine issue of fact but that the feeders cannot show:) (a) that the packers rigidly applied predetermined fixed formulas to the Yellow Sheet beef price quotations or the Safeway wholesale beef price in purchasing cattle; (b) that the packers applied the formulas consistently and without deviation over a long period of time; (c) that the packers had no reason to depart from their purchasing formulae in the short term because fat cattle supply is highly inelastic; (d) that the packers did not absorb depressions in beef prices; (e) that the purchase prices were set without negotiation and without regard to the interactions of supply and demand; (f) that the difference between the price actually received and the price that would have been received without formulae pricing can be shown with adequate certainty; (g) that a change in wholesale beef quotations as reported on the Yellow Sheet had a precisely calculable and predetermined effect on the actual prices paid to the feeders; that is, once the Yellow Sheet price was known for any day, the price paid for cattle was completely determined; and (h) that there is detailed proof as to individual transactions.
The retailers argue that the fifth circuit's decision explicitly requires that the feeders' proof be clear, unequivocal, simple, non-inferential and direct, just as if a cost-plus contract existed covering each transaction in issue. Finally, the retailers remind this court that the fifth circuit emphasized that the teachings of Hanover Shoe and Illinois Brick reject proof presenting economic generalities or evidentiary complexities and uncertainties.
The feeders respond that the pass-on issue should be decided within "the context of traditional summary judgment determinations ... [and that] the burden is clearly defendants to show that no facts exist upon which a trier of fact could find for plaintiffs on the pass-on issue." See Plaintiffs' Memorandum in Opposition to Defendants' Motion for Summary Judgment at 7. It is hornbook law that the burden of the movant on a motion for summary judgment is a heavy one:
Before summary judgment will be granted it must be clear what the truth is and any doubt as to the existence of a genuine issue of material fact will be resolved against the movant. Because the burden is on the movant, the evidence presented to the court always is construed in favor of the party opposing the motion and he is given the benefit of all reasonable inferences that can be drawn from it. Finally, facts asserted by the party opposing the motion, if supported by affidavits or other evidentiary material, are regarded as true.
C. Wright and A. Miller, Federal Practice and Procedure: § 2727 (1973). Regarding the burden on the nonmovant, Wright and Miller explain:
... [I]f the movant makes out a prima facie case that would entitle him to a directed verdict if uncontroverted at trial, summary judgment will be granted unless the party opposing the motion offers *1131 some competent evidence that could be presented at trial showing that there is a genuine issue as to a material fact.... The burden is not a heavy one; he simply is required to show specific facts, as opposed to general allegations, that present a genuine issue worthy of trial.
The feeders are correct insofar as they conclude that traditional summary judgment burdens are applicable. See Lefrak v. Arabian American Oil Co., 487 F. Supp. 808, 815 (E.D.N.Y.1980) (movant had burden on motion contending Illinois Brick barred action by indirect purchaser).
Determining the burden of proof under Fed.R.Civ.Proc. 56 is aided by defining the precise ultimate issue posed by the motion. That definitional exercise is necessary because it provides the measure for the materiality of genuine fact issues. The ultimate issue presented here is an a fortiori result of Hanover Shoe and Illinois Brick. In Illinois Brick, the Court recounted the "principal basis for the decision in Hanover Shoe [which was] the Court's perception of the uncertainties and difficulties in analyzing price and output decisions `in the real economic world rather than an economist's hypothetical model....'" 431 U.S. at 731, 97 S. Ct. at 2067 (citing 392 U.S. at 492, 88 S.Ct. at 2231). It follows that the retailers must prove that no genuine issue of material fact exists with regard to the feeders' inability to present cost-plus equivalency proof in a non-complex manner. The feeders' responsive affidavits should be "definite and particularized," 600 F.2d at 1167, to show the simplicity of the passing-on proof.
Accepting all of the facts in the feeders' affidavits as true and giving them the benefit of all reasonable inferences, the retailers will have met their burden if they have shown and the feeders have not in a competent manner disputed that the proof of passing-on presents the evidentiary problems that Hanover Shoe and Illinois Brick sought to avoid. In determining if the feeders' proof is competent, the court notes that "only reasonable inferences may be drawn in [the feeders'] favor." Southway Theatres, Inc. v. Georgia Theatre Co., 672 F.2d 485, 493 (5th Cir. 1982).
III. The Complexity and Uncertainty of the Pass-On Theory
A. Factors Influencing the Pricing Decision
Unlike a cost-plus contract, a variety of factors influenced the pricing decisions made by packers: individual needs, competition and negotiation for cattle, estimation of yield and grade of cattle, conditions in the cattle markets, and the fluctuation of the by-product market. See Lefrak v. Arabian American Oil Co., 487 F. Supp. 808, 820-21 (E.D.N.Y.1980) (no rigid formula because a number of market variables affected costs). Stated otherwise, it is beyond reasonable factual dispute that the feeders cannot show that the pricing decisions of the packers are "... determined in advance without regard to the interactions of supply and demand," 600 F.2d at 1165, or that the "... habitual use of predetermined formula would enable measurement of the effect on prices for fat cattle or changes in wholesale prices." Id.
1. Packer's Individual Needs
In planning each day's cattle purchases, a packer considered his inventory and the number of cattle already purchased for delivery in the immediate future. His judgment about the adequacy of his cattle inventory directly affected the price he was willing to pay in the market. As the head buyer for IBP stated:
If we have a good inventory of cattle and we're not hard-pressed for numbers, then we'll negotiate tougher. If I tell them (our field buyers) we have to have cattle the next day, they will be easier to negotiate with. Walker Testimony at 519.
*1132 The packers' need to keep up kill line numbers thus affected pricing. Although the feeders' affidavits maintain that any variation in pricing occurred only on a "temporary basis," they acknowledge that there were occasions in which the packers' need to fill kill lines changed live cattle bids. See Affidavit of Brookhart at ¶ 18; Affidavit of Stallings at ¶ 18; Affidavit of Rudder at ¶ 19; Affidavit of Kirkeby at ¶ 18; and Affidavit of Sievers at 5. As one of the feeders' affidavits revealed: "[I]f ... they need cattle for their kill line, they will vary their bid up or down from the price quoted on the yellow sheet." See Affidavit of Vos at 1. Furthermore, packers generally had labor commitments that required them to pay weekly wages whether or not employees were working. The cost of labor was one of their most significant operating costs. Fixed costs inevitably and directly affected the packers' judgment about the prices they could offer for cattle and sometimes forced them to pay cattle prices that produced a loss when measured against anticipated revenues. As one packer explained in uncontradicted testimony:
One of the most important factors we considered when we determined the number of cattle to purchase and our initial offering prices is that once we opened our plants during a week we had to buy a minimum number of cattle in order to keep production facilities operating. Our union contracts required that if we brought employees to work at the beginning of a week, they were to be paid for 36 hours whether or not they worked. In addition, we needed to have cattle to maintain our relationships as a reliable supplier to customers of beef and byproducts. Such pressures frequently caused us to pay prices that were higher than current or projected dressed beef and by-product market conditions would warrant and resulted in running our operations at a loss. Affidavit of American Beef and Beefland Int'l at ¶ 12.
2. Competition and Negotiation for Cattle
The retailers have shown beyond genuine factual dispute that prices for fat cattle were, in some measure, affected by competitive bidding and individual negotiations with feeders. One packer explained:
The prices we pay for cattle are affected by the prices which our competitors are paying. Sunflower competes with approximately nine other packers in its procurement areas. If they are paying more for cattle than we determine should be our offering price, we will have to raise our price in order to be able to purchase cattle. Very often the competitive pressure has caused us to lose money because feeders "shop around" for the best available price. Affidavit of Sunflower at ¶ 18.
These competitive pressures caused packers to pay different prices for cattle during the course of a single buying day. As the head of cattle procurement for Swift stated:
It was not uncommon for the buying program to be changed before the day was over, usually by raising our target prices in response to competitive pressures in the market place. If as the day progressed we thought we could obtain the cattle we needed at a lower price, we would reduce the target price for some or all of our buyers. By the same token, if we were having trouble buying cattle, we would be forced to raise our target prices.... Swift Affidavit of Harvat at ¶ 8.
The feeders respond that any variance between prices offered by different packers was "almost always" less than one-half cent per pound. See Plaintiffs' Memorandum at 18. According to the feeders, any variance between live prices on the same pen of cattle offered by different packers can be explained either by an intervening movement of the Yellow Sheet price or by differences in estimating the quality or yield characteristics of the pen by the various packer-buyers, not by any variation from the Yellow Sheet formula. The feeders' own cattle buyer affidavit explains, however, that shortages of cattle caused packer buyers to disregard Yellow Sheet prices in bidding. See Affidavit of Gilbert at 2. In times of shortage, competition increased *1133 and packers had "reason to depart from their purchasing formulae in the short run...." 600 F.2d at 1165.
Furthermore, the price of a pen of cattle often depended on negotiation between packers and feeders as to the grade and yield of the cattle in the pen. With their assertion that the Yellow Sheet price established a ceiling or maximum price a packer would offer for live cattle, the feeders recognize that negotiation occurred below the alleged ceiling. As one of feeders' buyer affidavits stated, "If at any time we could buy cattle under the Yellow Sheet we got a pat on the back." Affidavit of Snodgrass at 1. (emphasis in original).
3. Estimating Cattle Characteristics
Because most cattle were sold to packers on a "live weight" basis, packers had to estimate these factors: (a) the number of pounds of beef that would be produced from the cattle, i.e., the dressing percentage; (b) the carcass weight range of the cattle; (c) the quality grade of beef that would be produced from the cattle; and (d) the yield grade of beef that would be produced from the cattle. As one packer explained:
At least 90 percent of the cattle that we bought were purchased on a "live weight" basis. Our buyers estimated the weight, dressing percentage, yield grade, and quality grade of the cattle before they made their bids. Even minor errors in these estimates significantly affected the revenues we actually received, and thus of profit or loss. Armour Affidavit ¶ 21.
To implement a perfect "pass-on," packers would have to know with certainty the way in which each pen of cattle would convert into beef. One of the retailers' packer witnesses estimated that an experienced buyer would
average within about ½ of 1% in their estimates of dressing percentage, and within about 5-10% in their estimates of how a pen of cattle would grade. On particular transactions, even an experienced buyer could be off as much as 4 percentage points in connection with the dressing percentage and as many as 20 percentage points in connection with yield grade. Incorrect estimates often caused errors in the price of individual pens of cattle as much [as] $5.00 per hundred weight. Affidavit of American Beef and Beefland Int'l at ¶ 19.
The feeders respond to the retailers' "estimation" argument with affidavits stating:
Cattle buyer's estimates of the various factors of quality grade and yield, average weight and dressing percentages are usually very close to correct when compared to the actual figures derived when cattle are slaughtered. Cattle buyers are fired if their [in]accuracy is not less than 1% of actual slaughter figures, and cattle buyers are given bonuses based on their ability to pay less for the cattle than the dressed beef is worth. Affidavit of Brookhart at ¶ 13 (emphasis supplied).
Accepting here, as I must, the feeders' version, I am still left with deviations that travel against Illinois Brick, because a cost-plus contract equivalency allows little room for guessing. Directly put, even the best eyes of buyers inject estimation into a claimed formulae pricing process, albeit skilled estimates.
4. Conditions in the Cattle Market
Retailers emphasize that supply and demand operated to move the cattle market, affecting packers' decisions about the prices at which they would attempt to buy cattle. According to retailers, packers considered: (a) the number of cattle received at terminal markets each day; (b) information from their buyers, cattlemen, the USDA, and wire services regarding availability of cattle; (c) competitive activity in the live cattle market; (d) prices paid for live cattle; and (e) the number of cattle sold. The Vice President for Cattle Procurement of John Morrell & Co. stressed the supply and demand factors he considered each day:
Estimated current cattle supplies constitute one of the most important factors in my pricing decisions. If there is an ample supply of cattle of the kind I want to *1134 buy, I am usually able to buy at a lower price than when cattle are scarce. In order to estimate realistic prices for the day, I must estimate supplies, which vary greatly by region and, at least in some areas, by time of year. I consider numerous sources of information to estimate supplies, including various market reports. The most important sources are the reports from my buyers as to supplies in their respective regions. Affidavit of Morrell at ¶ 7.
Approving the cost-plus equivalent as alleged in the feeders' complaints, the fifth circuit noted: "The complaints sufficiently allege that the impact of the retail chains' price changes upon the pricing decisions of packers is determined in advance without regard to the interactions of supply and demand." 600 F.2d at 1165. In reality, the interaction of supply and demand indisputably did affect the daily beef prices paid by packers. Despite an overall effort to suggest it was not great, the feeders' affidavits concede some effect of supply and demand. As one cattle buyer explained: "All prices were given to me on a dressed basis and I would take that times potential carcass yield to get the live price. Sometimes this would be on the yellow sheet price or it could be over or under the yellow sheet depending on demand for that day." Affidavit of Shima at 1. (emphasis supplied). Gordon Freie, a cattle feeder conceded:
The exception to the price being absolutely governed by the Yellow Sheet is in the instance of where there was a temporary shortage of cattle ready for market in a particular region, in which instance, the packer would temporarily bid higher for cattle. Affidavit of Freie at 2.
Suggesting that the effect of supply and demand was not great does not answer the retailers' assertion that a claimed "pushpull-click-click" formula was not at work.
5. The By-Product Market
The cattle sold to packers produces two products: the edible dressed beef carcass and the nonedible by-products such as hides, bone meal, blood, and intestines, known generally as "offal." While the feeders are alleging that there is no price competition for dressed carcasses, they acknowledge without qualification that competition exists in the offal market and that this competition causes the price for live cattle to vary. A number of their feedlot operator and cattle buyer affidavits make this concession. One stated: "If the market for the various components of offal is good enough, the packer will pay in excess of the top of the carcass price, in his efforts to bid for the offal.... If the offal market is down, then the packer will habitually bid something lower than wholesale carcass price, again to make sure he is covering slaughter costs." See Plaintiffs' Memorandum at 25; Affidavit of Sievers at 4-5. See also Affidavit of Brookhart at ¶ 17; Affidavit of Stallings at ¶ 17; Affidavit of Rudder at ¶ 18; Affidavit of Kirkeby at ¶ 18; Affidavit of Raskin at 1; Affidavit of Sand at 1.
If both beef and by-product anticipated revenues are considered in purchasing cattle, a competitive market in by-products effectively covers any alleged pass-on. At the least, the injection of an independent market makes detection of the completeness of pass-on most difficult. To prove a 100 percent pass-on of the alleged under-charge would involve tracing each sale of beef carcass and by-products to find the extent to which each packer absorbed losses in the sale of beef or by-products, or both. It would involve using regression studies or similar techniques. Such complicated and speculative evidence is forbidden by Hanover Shoe and Illinois Brick.
*1135 B. Empirical Data
The empirical studies presented by retailers demonstrate that the Yellow Sheet is not the index of a reasonably precise pricing formula. The court views the empirical data in the context of the motion for summary judgment; to raise a genuine issue of material fact, the plaintiff must respond with specific data, not conclusory assertions. See In re Municipal Bond Reporting Antitrust Litigation, 672 F.2d 436 (5th Cir. 1982).
1. The Dubuque Packing Company Studies
The first Dubuque study compares the initial prices the Dubuque Packing Company offered each morning at its Dubuque, Iowa plant for certain 800-900 pound (dress weight) choice steers with the prices of comparable steers reported on the Yellow Sheet for the period August 1971 through December 1977. The study reveals that the Yellow Sheet prices were not identical to those offered by Dubuque. Although the graphs plotting price movements suggest a general relationship between the two prices, no fixed relationship exists.
The feeders argue that the Dubuque study is irrelevant to most packer bidding because it involved a company selling a unique product, Kosher beef. Kosher cattle are heavier than regular fat cattle. The live animal weighs about 1200 pounds; dressed it weighs between 800 and 900 pounds (as compared to 600-700 pounds dressed weight for the normal steer). Sid Raskin, a former Kosher packer, stated that during the Jewish holidays Kosher front quarters could bring a $.05 per lb. premium and hind quarters could bring a $.04 per lb. premium. Affidavit of Raskin at 1. Robert Snodgrass, whose experience included employment as a buyer for Kosher packers, stated that "sometimes the packer could give $1.00 or $2.00 per hundred [weight] (dressed) over the [Yellow] sheet for the Kosher cattle." Affidavit of Snodgrass at 1. The feeders also attempted to discredit the Dubuque study with the affidavit of Glen L. Freie, a cattle feeder:
Animals yielding the heavy carcasses Mr. Gray [of Dubuque] states he was obtaining (800-900 pounds) would also yield a much larger percentage of offal, which would allow the Kosher packers to bid higher for the animal. I made a study of 9 selected months of Dubuque records ... and find that the price received for the offal itself, would generally account for the difference between the Yellow Sheet price and the price Mr. Gray was offering ... Affidavit of Freie at ¶ 7. (emphasis supplied).
The feeders' affidavits suggest that the Dubuque packing study is atypical and cannot be relied upon to support the motion for summary judgment, but retailers have performed an additional study using non-Kosher cattle. This study reached the following conclusions:
The Dubuque offering price and the Yellow Sheet quotation were the same on only 74 days out of 1731 days (4%).
The Dubuque offering price fluctuated between being higher than the Yellow Sheet quotation (68% of the days) and being lower than the Yellow Sheet quotation (28% of the days).
During each of 35 months, the Dubuque offering price moved both above and below the Yellow Sheet quotations.
The difference between the Dubuque offering price and the Yellow Sheet quotation almost always fluctuated during the course of a week. There was only one week during the entire six and one-half year period when the difference did not fluctuate. Defendants' Reply Memorandum at 20.
*1136 2. USDA Data
a. Comparison of Yellow Sheet and USDA Live Cattle Prices
To support their claim that there is no fixed relationship between cattle prices and Yellow Sheet quotations, retailers also offer a comparative study of Yellow Sheet quotations and official USDA live cattle price reports. The study was performed by Agri-Commodities, Inc. ("ACI"), an agribusiness management consulting, commodity information and commodity trading advisory service.
ACI used a computer file of carcass beef quotations as reported by the Yellow Sheet and a computer file of live cattle prices as reported by the USDA to compare changes in live cattle prices with changes in Yellow Sheet carcass quotations on the previous day for the period 1970-1979 for five areas: Interior Iowa, Peoria, Sioux City, Omaha, and East St. Louis.
Included in the comparison was every occasion on which there were two consecutive business days for which the Yellow Sheet reported a carcass beef quotation, and, on each of the two business days following the first day's Yellow Sheet quotations, the USDA reported a live cattle price.
The specific carcass beef item used for comparison purposes was choice steer carcasses, 600 to 700 pounds. The specific live slaughter cattle prices used are set out below:
(1) Interior Iowa From January 1, 1970 to September 25, 1972, the prices were for Choice slaughter steers, 1,100-1,300 pounds. From September 25, 1972 through 1979, the prices were for Choice slaughter steers, 1,050-1,150 pounds in early years, and 1,000-1,200 in later years.
(2) Peoria From January 1, 1970 to August 1, 1970, the prices used were Chicago prices for Choice slaughter steers, 1,100-1,300 pounds. From August 2, 1971 to September 25, 1972, the prices were Peoria prices for Choice slaughter steers, 1,100-1,300 pounds. From September 26, 1972 through 1979, the prices were for Choice 2-4 slaughter steers, 1,000-1,200 pounds.
(3) Sioux City, Omaha and East St. Louis From January 1, 1970 to September 25, 1972, the prices were for Choice slaughter steers, 1,100-1,300 pounds. From September 26, 1972 through 1979, the prices were for Choice 2-4 slaughter steers, 1,000-1,200 pounds.
The comparison for Interior Iowa is illustrative. It shows that during the period studied there were 1,991 occasions on which (i) the Yellow Sheet reported a carcass quotation for two consecutive business days and (ii) on each of the two days following the first day's Yellow Sheet quotation, the USDA reported a slaughter cattle price for Interior Iowa.
On 280 of the 1,991 occasions, or 14.1 percent, the Yellow Sheet and live cattle prices remained unchanged from one business day to the next. These 280 occasions occurred sporadically during the ten year period.
On the remaining 1,711 occasions for which there are Yellow Sheet quotations for consecutive business days followed by live cattle price reports for successive days, there was a change in the Yellow Sheet, or the live price, or both. On 800 occasions, or 40.2 percent, one price changed but the other did not move. On 243 occasions, or 12.2 percent of the 1991 occasions, the Yellow Sheet and the live cattle prices moved in opposite directions. On 668 occasions, or 33.5 percent, the Yellow Sheet quotation change was followed by a live cattle price change in the same direction, but the retailers emphasize that on only eleven occasions, or 0.5 percent of the total, was the Yellow Sheet carcass price change followed by a live cattle price change that was in the same direction and approximated the dressing percentage.
b. Comparison of Regional Markets' Prices for Live Cattle
The retailers note that the USDA data reveals that in different regions of the country, different prices were offered for *1137 live cattle for the same time period. The retailers compared the USDA's live cattle prices for Iowa with prices in East St. Louis and Omaha and found no consistent relation between cattle prices in these markets. See Defendants' Appendix C, Exhibit 2. These differences among geographical areas undercut the claim that Yellow Sheet pricing is a mechanical formula. The Yellow Sheet quoted a single nationwide price for each of the grades of live cattle and some other factors accounted for regional differences. See Defendants' Appendix B, Exhibit 1, Yellow Sheet dated September 25, 1974.
3. The Beef Pricing Report
Feeders refer to the USDA study published in December 1978 entitled "Beef Pricing Report" to support their claim of formulae pricing by packers. Feeders contend that the findings of the USDA study absolutely contradict the retailers' evidence (the affidavits of the meat packers) that shows that many factors other than the Yellow Sheet provide the basis for the purchase price of cattle. The USDA conclusions relied upon by feeders are that (a) "70% of all steer and heifer carlot carcass sales of the 35 plants [studied] were formula sales based on Yellow Sheet published price," Beef Pricing Report at 5; (b) 30,478 head or 19.9% [of all cattle purchased on a dressed weight basis] were bought on a formula based on a future Yellow Sheet published price (usually day of slaughter)," id. at 15; and "the study confirmed that the wholesale beef price quotation published by the Yellow Sheet is the principal guide used by the selected packers in issuing daily price instructions to cattle buyers," id. at 8. A review of the report places those statements in context.
The report focused on beef pricing activity in July, 1977 because that month showed more fluctuation than others up to the time of the study. Carlot sales of steer and heifer carcasses were chosen for the study because most pricing information is based upon such sales. USDA employees interviewed head buyers from 35 packing plants representing 26% of the federally inspected steer and heifer slaughter for all firms in the United States in 1976. These plants, 22 out of a total of 40 in the United States, were located in the major fed beef producing areas from the River Region comprised of North Dakota, South Dakota, Minnesota, Nebraska, Iowa, Kansas, and Missouriand the Southwest region, comprised of Texas and Oklahoma.
The principal purpose of the study was to investigate and report on meat pricing practices of packers and the meat price reporting practices of packers. Thus, the study focused on packer pricing to retailers, slaughterers and processors, not on livestock procurement practices.
In the investigation of packer procurement, only purchases made on a dressed weight basis were analyzed to determine how many were formula priced. Beef Pricing Report at 15. Those purchases constituted 25.4 percent of livestock procured for slaughter, live purchases and transfers from packers' feeding operations constituting the remaining 75.5 percent. Id. at 14. The study concluded that of a total of 152,961 head purchased in July 1977 on a dressed weight basis, 30,478 head or 19.9% were bought on a formula based on a future Yellow Sheet published price, usually day of slaughter.Id. at 15.
Dividing the number of all cattle slaughtered in July 1977, or 625,276, by the 30,478 head that were purchased by "formula" yields 4.9 percent. That is, 4.9 percent of the total slaughter was purchased on a dressed weight basis based on a future Yellow *1138 Sheet price.See Deposition of William V. Carroll at 85.
Based on this data compiled by the USDA interviewers regarding cattle purchased on a dressed weight basis, the study concluded that "the wholesale beef price quotation published by the yellow sheet is the principal guide used by the selected packers in issuing daily price instructions to cattle buyers." The report continued, however, that "[s]ome additional considerations in arriving at the exact bid price are operating costs, kill costs, by-product values, and the estimated quality grade and yeild [sic] for a particular lot of cattle." Beef Pricing Report at 8-9. The value of by-products was the second most important consideration. See Deposition of William V. Carroll at 142-43. Thus, the Yellow Sheet, though a main consideration, is only one element in the pricing decision. The Supreme Court excluded such cost-based "rules of thumb" when framing the narrow cost-plus exception because "[t]hese rules are not adhered to rigidly ... [T]he extent of the markup or allocation of costs is varied to reflect demand conditions...." Illinois Brick v. Illinois, 431 U.S. 720, 744, 97 S. Ct. 2061, 2074, 52 L. Ed. 2d 707 (1977). The court notes that the feeders told the fifth circuit that the pricing system was not a rule of thumb system.
The retailers' empirical data remains undisputed. The only contrary suggestions are conclusory assertions that the packers habitually adhered to the Yellow Sheet price. Such adherence does not inevitably result in a correspondence between the Yellow Sheet price and packers' bids because of the factors previously discussed. Hard and unanswered data showing the absence of such correspondence is a telling *1139 blow. Cattle buying is an estimation process; the packers estimate grade and yield of the cattle, their inventory needs, the price they will receive for the carcass several days after purchase, and the price they will receive for the by-product. Characterizing the use of the Yellow Sheet as a mechanical formula proves nothing. An habitual use of a "rule of thumb" by packers is nonetheless a rule of thumb.
C. Absorption of Price Depressions by Packers
Because of the varied factors that affect pricing, retailers have demonstrated that there is no genuine issue of material fact regarding whether the feeders can show that the alleged formula pricing prevented packers from sharing any illegally low price. The uncertainty precluded by Illinois Brick exists in this case. For example, the existence of separate markets for beef and by-products makes it difficult to determine (absent recourse to math models) the extent to which packers absorbed declines in beef prices by relying on by-product revenues. It cannot be demonstrated "whether (or to what extent)," 600 F.2d at 1166, when packers bought cattle, they absorbed the alleged depression in beef prices or rather absorbed a change in by-product prices. The silence of Judge Wisdom's opinion on the subject is understandable because neither the briefs on appeal nor the oral argument focused on the independent competitive market in by-products. The opinion's sensitivity to the general phenomenon of multiple costs makes that silence the more notable.
It specifically recognized that if a seller has multiple cost inputs, absorption could occur "without leaving any evidence" of its occurrence. "An intermediary, such as a contractor in Illinois Brick, who employs numerous inputs in assembling his final product might bear the larger brunt of an increase in the price of one of his inputs without leaving any evidence that he did so." 600 F.2d at 1165. It has on remand become clear that multiple sources of revenue make it difficult to determine with certainty what part of any beef price depression was absorbed by packers because undercharges could be absorbed without leaving any evidence.
Absorption of loss by packers would further be demonstrated by the very nature of the cattle market. Even if a Yellow Sheet price were used both by retailers and by packers, the packers purchased several days before selling beef to retailers and could not know with certainty what "price he will receive when he in turn sells," 600 F.2d at 1165 n.22, as the fifth circuit required. The feeders recognize this uncertainty: "The beef slaughterer will pay the same amount for the live animal as he expects to sell the dressed beef carcass for in the wholesale market." See Plaintiffs' Memorandum at 23-24 (emphasis supplied). If beef prices decline before packers sold, they would absorb part of any alleged depression. Feeders cannot demonstrate that "the formulae employed were such that a change in wholesale (beef) prices ... had a predetermined effect on [cattle prices]." 600 F.2d at 1165 n.21. That is, at best the Yellow Sheet provided today's price as a basis for predicting future carcass prices. The retailers have shown beyond dispute that the feeders cannot demonstrate that there is no risk of double recovery. Packers plainly absorbed some loss. At the least, a demonstration that they did not would require a forbidden excursion.
*1140 IV. Conclusion
Not surprisingly, the cost-plus exception of Hanover Shoe has been read differently by the lower courts. One description of this difference contrasts the decision of this circuit in this case with the decision of the third circuit in Mid-West Paper Products Co. v. Continental Group, Inc., 596 F.2d 573 (3d Cir. 1979). See A Legal and Economic Analysis of the Cost-Plus Contract Exception in Hanover Shoe and Illinois Brick, 47 U. of Chi.L.Rev. 743 (1980). There is a threshold issue that divides the fifth and third circuits. That question is whether there can be an equivalent to a cost-plus contract, given the objectives of Hanover Shoe and Illinois Brick. As earlier described, the Supreme Court's firm choice of immediate, over direct, purchasers had as its purpose the facilitation of the private damage suit as an antitrust enforcement mechanism. It saw tolerance of overly complex proof and dispersion of allowable plaintiffs as unacceptable impediments to that goal. Ironically, closing the door to the explanatory power and complexity of the tools of the econometrician was consistent with the teaching of basic economic incidence theory. That teaching is that in a monopsony case, apart from cost-plus contracts where the market effectively becomes internal to each contract, the immediate purchaser will not pass on the complete illegal burden absent a perfect model of inelastic supply.
The fifth circuit has decided that a pricing mechanism can be proved that functions in an equivalent manner to a cost-plus contract. Presumably, that equivalency must be demonstrated without resort to the kinds of proof and simplifying assumptions rejected by the very confinement of the exception to a cost-plus type contract. If not, the exception swallows the rule. The Supreme Court's choice of immediate over remote purchasers was perforce bottomed on the assumptions, with the exception noted, that immediate purchasers inevitably suffered injury or contrary proof of this likelihood was intolerably complex. It follows that to establish a cost-plus equivalent pricing mechanism a plaintiff must provide proof that the injury the mechanism inflicts upon the indirect purchaser is separate and distinct from the injury to the immediate purchaser. Otherwise stated, the plaintiff must present proof that the pass-on was complete. Stated more simply, equivalence requires proof in a sufficiently simple form that the immediate purchaser passed-on the complete illegal cost, or as here, the illegal cost reduction.
While it may well be that the fifth circuit opinion seems to assume that approximate inelasticity results in approximate passingon, a proposition that may be difficult to square with economic incidence theory, a careful reading of the fifth circuit opinion suggests that the gap between it and the third circuit is in actual effect not so wide, at least not as some commentators suggest. True, the fifth circuit was willing to allow equivalency, but its language was skeptical and the conditions for its demonstration are exacting. Moreover, its tolerance of the effort was born in a procedural context in which it had to indulge the proffered fantasy of the feeders. While squaring functional equivalence with the rulings of the Supreme Court gives me pause, I accept the institutional limits of my role and attempt to apply the decision of my court of appeals; not in a niggardly manner rooted in disagreement, but in full measure as the law of this case. All that has gone before has been that effort.
In conclusion, I point out that it is not necessary even to resort, as I have, to the studies and other data presentations to demonstrate that this market lacks the equivalence required by the fifth circuit, certainly by the Supreme Court.
The feeders urge that the industry pricing mechanism is virtually a non-judgmental mathematical calculation and is, accordingly, the functional equivalent of a cost-plus contract. The argument is constructed upon certain assumptions about the "market." In the attempt to prove a cost-plus equivalency, the feeders argue that fat cattle at market time must then be sold because their controlled diet prepares them *1141 for slaughter within a relatively small time window and feeders do not have the option of withholding them from the market. That is, the feeders argue that the supply of cattle is highly inelastic. The result, so it is argued, is that the market allows a functional equivalent of cost-plus contracts in the sense that the middleman's pricing decision need not weigh supply and demand. While not controlled by the contract between the buyer and seller, these decisions, they argue, are frozen by market forces. The argument continues that there was formulaic and mechanical pricing that allows a post-purchase reviewer to prove that the lowered cost was not shared between the packer and the feeder but was passed-on to the feeders, that a packer would not pay more for fat cattle than he could recoup for the carcass.
The seeming simplicity of this argument ignores the reality that the price of a steer has two components the carcass and the offal. How much of an illegal price reduction the packer passes on is in part a function of the offal market. If, for example, the carcass price is set at one hundred dollars, a price that represents a ten dollar illegal burden or lowered price, a packer in calculating the amount of his bid for the steer knows that if there is no change in the market by the time he buys, slaughters and resells, he will receive no more than one hundred dollars for the carcass (not the $110 of the free market), plus the expected revenue from the offal which we will assume will be $15, a total of $115. Subtracting his costs, assume 25%, plus a profit of 10%, he should bid no more than $74.75. The feeders' argument assumes both the mechanical (and therefore replicatable) application of such fixed numbers as well as such inelasticity in supply (in the short-run) of fat cattle that the flow through of the $10 was complete, and complete without adjustments in volume of sales.
The true facts are that the packers' calculations of anticipated profit have no such formulaic operation nor have feeders shown the inelasticity of the supply. That profit calculation has as a vital component (by the very contention of the feeders) an estimation of return upon the offal. It follows that the packers' bid for fat cattle, while it may well be calculated from an "on the average, relatively accurate" prediction of the return from the sale of the carcass, must accommodate the offal market forces as well as other changing costs. Putting aside that the market price might drop before he sells the carcass, difficult apportionment problems remain.
Finally, the fixity of the formula applied is further softened by the varying marginal costs of the packers. It is undisputed that the packers have high labor costs associated with plant costs and the other usual fixed costs. Fully allocated costs may well be the hypothesized 25%, but the marginal costs for a given pen of fat cattle may be sufficiently small that the packer bid may exceed the Yellow Sheet price. The question becomes, "How do we extricate by application of the feeders' simple arithmetical *1142 calculation, the undercharge?" Can we do so without resort to the complexity of marginal cost accounting? More directly, can it be done consistent with the teaching of Hanover Shoe and Illinois Brick? Can it be done on a transaction-by-transaction basis as distinguished from a post-hoc judgment that such occurred "on the average"? I think not. Instead, viewed most favorably to the feeders, this is a market whose middle level uses rule of thumb pricing. To find that in this market the packers' cattle purchases were the equivalent of a cost-plus contract requires an expansive construction of a narrowly based exception. That expansive view tears at the core of the remedial goals of Hanover Shoe and Illinois Brick and unnecessarily sets at odds a normative component of antitrust policy and economic theory. Why a court ought to consider doing so is unclear; that it ought not do so is beyond cavil; that I will not is here proved.
In accordance with the attached Memorandum, the motions for summary judgment filed by all retailers are GRANTED.NOTES
 In addition to the retail defendants, one defendant, Super Valu Stores, Inc., is a wholesale grocer; one, the National Association of Food Chains, is a national trade association, and one, the National Provisioner, publishes a beef industry price reporting publication, the Yellow Sheet.
 In some regions, the Safeway wholesale price is allegedly used as the basis for the formula.
 This case may be characterized as an inverted passing-on situation. That is, the feeders allege that price depressions, not price overcharges, have been passed on to them.
 In Hanover, the plaintiff, a shoe manufacturer that leased shoe machinery from defendant United, alleged that United had monopolized the shoe machinery industry by leasing, rather than selling, complicated shoe machinery. United claimed that Hanover suffered no injury because any illegal overcharge in the leasing was reflected in Hanover's charge for shoes it sold its customers.
 The Court stressed:
A wide range of factors influence a company's pricing policies. Normally the impact of a single change in the relevant conditions cannot be measured after the fact; indeed a businessman may be unable to state whether, had one fact been different (a single supply less expensive, general economic conditions more buoyant, or the labor market tighter, for example), he would have chosen a different price. Equally difficult to determine, in the real economic world rather than an economist's hypothetical model, is what effect a change in a company's price will have on its total sales. Finally, costs per unit for a different volume of total sales are hard to estimate. Even if it could be shown that the buyer raised his price in response to, and in the amount of, the overcharge and that his margin of profit and total sales had not thereafter declined, there would remain the nearly insuperable difficulty of demonstrating that the particular plaintiff could not or would not have raised his prices absent the overcharge or maintained the higher price had the overcharge been discontinued. Since establishing the applicability of the passing-on defense would require a convincing showing of each of these virtually unascertainable figures, the task would normally prove insurmountable. On the other hand, it is not unlikely that if the existence of the defense is generally confirmed, antitrust defendants will frequently seek to establish its applicability. Trebledamage actions would often require additional long and complicated proceedings involving massive evidence and complicated theories. 392 U.S. at 492-93, 88 S. Ct. at 2231. (footnote omitted).
 In Illinois Brick, the State of Illinois and 700 local government entities sued the manufacturer, Illinois Brick, alleging that illegal overcharges by the company were passed on to them by contractors who used the bricks in building state facilities.
 A cost-plus contract has these elements: (1) a pre-existing contract, (2) specifying a fixed quantity of products to be sold, (3) with the price per unit to be determined by an agreedupon markup to the seller's own cost. Recent Development, A Door in the Illinois Brick Wall A Functional Equivalent to the Cost-Plus Contract Exception, 33 Vand.L.Rev. 481 (1980). When parties agree to a cost-plus contract, the indirect purchaser's price can be easily determined arithmetically once the middleman's costs are known. The middleman is not at risk and neither profits nor losses when the cost to him changes. Rather, he increases or decreases the charge to his customer without ever absorbing a loss or gaining a profit.
Because the price is fully negotiated in advance of any transactions under a cost-plus contract, (i) the middleman need not make any judgments about supply and demand or other changeable conditions in the market; (ii) the middleman's price does not react to the pressures of competition; and (iii) the price does not change as the result of day-to-day negotiation with his customer. Finally, in a cost-plus contract, the relationship between the selling price and the costs to which it is tied will always remain fixed.
The Supreme Court allowed the pre-existing cost-plus exception because it did not give rise to difficulty of proof or damage tracing problems and ensured that the entire injury was passed on to the indirect claimant:
[T]he pre-existing cost-plus contract makes easy the normally complicated task of demonstrating that the overcharge has not been absorbed by the direct purchaser. 431 U.S. at 732 n.12, 97 S. Ct. at 2067 n.12.
 For each of the claims (a) through (h), retailers rely on specific language from the fifth circuit opinion: (a) & (b) "The packer's habitual use of predetermined formulae would enable measurement of the effect on prices for fat cattle of changes in wholesale prices," 600 F.2d at 1165; (c) "If allegations of structural inelasticity of short term supply and of rigid formula pricing do not bring a case within the cost plus exception for purposes of a motion to dismiss on the pleadings, then the cost plus exception is a narrow one indeed....," id. at 1166; (d) "[A] packer's absorption of a loss resulting from a decrease in the wholesale price would be detectable from a change in his purchasing formula....," id. at 1165; (e) "[T]he cattlemen ... are in no position to negotiate prices because they cannot, at least in the short term, withhold their product," id. at 1154; (f) "The [Supreme] Court's general rejection of the use of passing-on theory rested in part on its view that this issue of causation could not, except in rare cases, be resolved with adequate certainty," id. at 1165 n.22; (g) "We must assume, for purposes of the appeals ... that the formulae employed were such that a change in wholesale prices in a given region had a predetermined effect on the price offered feeders....," id. at 1165 n.21; and (h) The "proposed proof of damages ... will require detailed proof as to individual transactions," id. at 1166.
 The fifth circuit emphasized:
The cases as pleaded escape Illinois Brick only insofar as they avoid the use of general economic theory tracing the effect of the defendants' price decisions and only insofar as they can be proved, in the damages phase, with certainty.
600 F.2d at 1166.
 See also Affidavit of Gilbert at 2. "I can recall other instances of a real shortage of cattle in my area, and a shortage of cattle in the total area that supplied our plant with cattle to fill our kill requirements that we were told to disregard the `yellow sheet' and buy cattle the best way we could."
 The fifth circuit acknowledged that the "proposed proof of damages [in plaintiffs' complaints] is far from simple," but they accepted it as pleaded because it involved "a complexity born of quantity." 600 F.2d at 1166. The summary judgment record demonstrates that the proof of damages also presents a complexity born of quality. This complexity Hanover Shoe and Illinois Brick do not allow.
 The feeders' theory of packer pricing of Kosher beef does not follow from the Dubuque data because that data shows prices offered below the Yellow Sheet as well as above. Moreover, the feeders' theory on its face acknowledges considerable uncertainty and complexity in the pricing of cattle by Kosher packers.
 The USDA also collected data from 8 brokers, 3 market reporting services and 8 packers with activity in the fed cattle futures market holding 50 or more August contracts as of July 1, 1977. Beef Pricing Report at 14.
 The term "formula" for purposes of the study meant that the price was left open at the time of agreement to sell cattle and was established by the quotation of a meat reporting service, usually at a later date. See Deposition of William V. Carroll at 90. (Mr. Carroll is Division Director of the Packer and Poultry Division, Packers and Stockyards Administration, USDA. He collated the data from the investigation and wrote the Beef Marketing Report).
 Mr. Carroll, Division Director of the Packer and Poultry Division of the USDA, concluded that there was "no way" to "determine which of the cattle that were purchased on a carcass basis tied to a future yellow sheet price were sold later on as beef tied to a yellow sheet price." See Deposition of William V. Carroll at 88.
 The USDA interviewers did not question field buyers but only interviewed head buyers at the plant. See Deposition of William V. Carroll at 74.
 Mr. Carroll explained that 19.9 percent of the cattle purchased on a dressed weight basis were purchased on a formula based on the Yellow Sheet price not equal to the yellow sheet price. See Deposition of William V. Carroll at 99.
 The following are examples of conclusory assertions in the feeders' affidavits:
(1) "The Yellow Sheet is used as the basis for all trades between us and the packers, whether the trade is made on grade and yield or live weight." Affidavit of Hitch at 4.
(2) "Over 90% of the time, the price given to cattlemen by packers was absolutely controlled by the Yellow Sheet." Affidavit of Freie at 2.
(3) "[T]he price that I would use to pay for fat cattle was 9 times out of 10 obviously controlled by the movement of the wholesale beef market and or the `yellow sheet.'" Affidavit of Gilbert at 1-2.
(4) "The habitual practice of the packing industry is to match the wholesale price paid by the retailer or fabricator to the beef slaughterer, to the price paid by the slaughterer for the live animal.... [T]he packers habitually average out the costs of procuring the animal against the price received for the dressed beef carcass, and break even on the average." Affidavit of Sievers at 1-2.
(5) "[I]t always seemed to me that if there was a depressed market or continuously falling prices, that the only way a buyer would bid on my cattle was in the beef based on the yellow sheet." Affidavit of Bertram at 1.
(6) "It seems to me that it is getting harder all the time to get anything but a dressed beef bid from the packer buyers." Affidavit of Utesch at 1.
(7) "As a rule of thumb nowdays [sic], the buyer will bid you a dressed beef price the day he bids or the day of kill." Affidavit of Heiller at 1.
The one affidavit presenting specific data reported an ordinary least squares analysis of the "producer-packer margin" in percentages. The results indicated packers could maintain relatively constant margins without regard to the supply of cattle. The affiant concluded from the data that "packers `consistently' and `habitually' applied a formula which maintained relatively constant margins without regard to prevailing market forces." See Affidavit of Stuart at 5. By conducting such a study, the feeders have demonstrated that resort to complicated regression analysis is necessary to make their proof. Hanover Shoe and Illinois Brick forbid such analyses.
 The retailers' packer affiants emphasized that they paid prices exceeding Yellow Sheet quotations. See, e.g., Documents of H. C. Hitch, Jr. D. Ex. 1.
Feeders respond that packers did not absorb any part of the alleged depression in beef prices when they paid more than the alleged formula-derived price because the "packer temporarily cuts into his offal revenues if he needs animals." Plaintiffs' Memorandum at 25. Yet, whenever packers cut into any revenues because the alleged formula-derived prices were insufficient to allow them to buy cattle, they absorbed at least part of any alleged depression in beef prices.
Feeders further respond that packers suffered no loss because "on average" they recouped their costs. See Affidavit of Brookhart at ¶ 19-22; Affidavit of Stallings at ¶ 19-22; Affidavit of Rudder at ¶ 21-24. If packers averaged out gains and losses, however, they would absorb some loss on some days.
 The feeders have not presented a consistent theory of packer pricing that fits within the cost-plus equivalent exception. During oral argument on a discovery motion in their suit against the packers, the feeders argued that they need access to a study performed by the Commodities Future Trading Commission to help them prove that packers had "sold short" in the cattle futures market to bring down the price of live cattle. The feeders cannot consistently argue that they can prove in a noncomplex manner that packers served as passive conduits through which price depressions were passed and at the same time maintain that these packers manipulated the market to lower the price for cattle. According to the feeders' second theory, the packers no longer are conduits and the price bid for live cattle would be affected by market forces.
 Because the fifth circuit has suggested that the loss of volume in packer sales is not inconsistent with a cost-plus type of pass through, I will, for sake of this example, assume only a sufficient inelasticity as to allow the packer to gain short-term pricing leverage. It bears observation, however, that changes in the volume of sales and inelasticity are fundamentally at odds.
 A three week time frame is not insubstantial, especially when viewed against the packers' competing need to keep their kill lines moving.