United States v. Engelhard Corp.

970 F. Supp. 1463 | M.D. Ga. | 1997

970 F. Supp. 1463 (1997)

UNITED STATES of America, Plaintiff,
v.
ENGELHARD CORP., et al., Defendants.

No. 6:95-cv-45 (WLS).

United States District Court, M.D. Georgia, Thomasville Division.

April 3, 1997.

Hugh Randolph Aderhold, Jr., Macon, GA, Angela L. Hughes, Washington, DC, for U.S.

Howard Jerome Strickland, Macon, GA, William T. Lifland, New York City, for Engelhard Corp.

Robert Russell Gunn, II, Macon, GA, George M. Chester, Jr., Robert A. Long, Jr., for Floridin Co., U.S. Borax, Inc., U.S. Silica, Inc.

Robert B. Langstaff, Albany, GA, Thomas Q. Langstaff, Albany, GA, for ITC, Inc.

Keith C. Hennessee, Dallas, TX, Laurence K. Gustafson, Dallas, TX, for Dresser Industries, Baroid Drilling Fluids, Inc.

AMENDED ORDER^[1]

SANDS, District Judge.

Background

Defendants Engelhard Corporation ("Engelhard") and Floridin Corporation ("Floridin") are manufacturers and distributors of attapulgite clay. Floridin is a wholly owned subsidiary of Defendant U.S. Silica Company, and U.S. Silica is a wholly owned subsidiary of Defendant U.S. Borax, Inc. Floridin Answer ¶ 9. Engelhard owns attapulgite clay reserves in southwestern Georgia and northwestern Florida and operates an attapulgite clay processing plant in Attapulgus, Georgia. Engelhard Answer ¶ 8. Floridin also owns attapulgite reserves in similar localities of Georgia and Florida, and Floridin operates an attapulgite clay processing plant in Quincy, Florida. Floridin Answer ¶ 9.

Attapulgite clay is a naturally occurring mineral that in the United States is only found along the Georgia-Florida border. Tr. 7 at 86:13-86:16.^[2] Attapulgite clay is also found in various foreign countries, including Spain, China, Ukraine, Senegal, India, Australia, Turkey, France and South Africa. Tr. 7 at 86:13-86:21. There are two general types of attapulgite clay: gellant quality attapulgite ("GQA") and sorbent quality attapulgite ("sorbent clay"). Tr. 7 at 86:6-86:12. Crude GQA and crude sorbent clay are normally found together in attapulgite reserves, Tr. 7 at 88:6-88:16, with sorbent quality crude usually closer to the surface. Tr. 7 at 92:24-93-12. The two products are usually mined together, separated at the mine site, and delivered for separate processing at a processing facility. Tr. 7 at 88:17-89:16. Once processed, sorbent clay is used for its absorption qualities in products such as kitty litter. Tr. 7 at 88:4-88:12, 86:2-86:9. On the other hand, processed GQA is used as a thickener and suspension agent in such end use applications as suspension fertilizers, animal feeds, paints, drilling fluids, asbestos-free asphalt roof coatings, tape joint compounds, and molecular sieves, among other things. Tr. 7 at 86:2-86:12; PX 349; PX 350; PX 362.

At the present time, only three companies mine, process and distribute GQA in the United States: Engelhard, Floridin, and Milwhite Company ("Milwhite"), which also operates its processing facility in Attapulgus, Georgia. A fourth company, Oil-Dri Company ("Oil-Dri") operates a processing facility in Ochlocknee, Georgia in which it used to process GQA until 1993, when it discontinued such processing. Tr. 3 at 212:18-213:2. Of the three presently competing forces in the United States GQA market, Engelhard and Floridin are the largest competitors, with each holding a market share slightly above 40 percent. Milwhite controls the remaining 15 percent or so of the GQA market. Tr. 6 at 40:24-41:25.

In 1993, U.S. Silica decided to leave the attapulgite business. Tr. 7 at 120:20-120:22; DX 1121.^[3] U.S. Silica offered to sell Floridin's assets, including the Quincy plant, its associated reserves and the intangible assets of its sorbent and gellant businesses.^[4] Tr. 7 at 120:18-120:22. In order to increase profitability, Engelhard expressed interest in purchasing Floridin's more modern processing facility in Quincy, as well as its sorbent clay business. Tr. 7 at 121:2-121:11; 135:15-169:2. Engelhard and Floridin had previously attempted an acquisition in 1987 that included Floridin's GQA business, but the Department of Justice notified both parties that it would challenge the proposed transaction under the antitrust laws insofar as it resulted in the merger of the two companies' GQA businesses.^[5] DX 1205. Thus, the Defendants structured the present transaction, such that Engelhard would acquire Floridin's Quincy plant, associated reserves, and sorbent clay business, while a third-party, ITC Corporation ("ITC"), would purchase Floridin's "GQA business." Tr. 7 at 121:2-121:11; DX 1116. Specifically, ITC would be purchasing Floridin's customer lists, know-how and other intangible assets of Floridin's GQA business. Tr. 2 at 155:11-155:15. In addition, Engelhard and ITC would enter into a joint venture agreement under which Engelhard would supply ITC with GQA at cost, and the two companies would share the Quincy processing facility, but operate and compete as independent manufacturers and distributors. See DX 1115.

The Department of Justice determined that the proposed acquisitions would not result in two new, independent competitors, but rather that the transaction would reduce the number of competitors in the U.S. gel clay market from three to two. Accordingly, Plaintiff brought this action pursuant to § 7 of the Clayton Act, 15 U.S.C. § 18. Plaintiff seeks a permanent injunction against the proposed Engelhard-Floridin-ITC transaction described above.

Discussion

Section 7 of the Clayton Act provides in pertinent part:

No corporation engaged in commerce shall acquire, directly or indirectly, the whole or any part of the stock or other share capital ... of another corporation engaged also in commerce, where in any line of commerce in any section of the country, the effect of such acquisition may be substantially to lessen competition, or to tend to create a monopoly.

15 U.S.C. § 18.

"The first step in any Section 7 case is to determine the relevant product market." United States v. Gillette Co., 828 F. Supp. 78 (D.D.C.1993). See U.S. Anchor Mfg., Inc. v. Rule Industries, Inc., 7 F.3d 986, 994 (11th Cir.1993). Indeed, in order to make a meaningful determination about market power and the potential for a substantial lessening of competition as the result of a given transaction, both a relevant product market and a relevant geographic market for that product must be defined. Brown Shoe Co. v. United States, 370 U.S. 294, 324, 82 S. Ct. 1502, 1523, 8 L. Ed. 2d 510 (1962). In order to prevail, the Plaintiff must carry the burdens of proof and persuasion regarding market definition. FTC v. University Health, Inc., 938 F.2d 1206, 1217-18 (11th Cir.1991); United States v. Baker Hughes, Inc., 908 F.2d 981, 982-83 (D.C.Cir.1990); Kaiser Aluminum & Chem. Corp. v. FTC, 652 F.2d 1324, 1340 (7th Cir.1981); United States v. Idaho First Nat'l Bank, 315 F. Supp. 261, 265 (D.Idaho 1970) (stating that government must show by a preponderance of the evidence a reasonable probability of lessening of competition within a relevant market).

In the present case, the government argues that the relevant product market is gel quality attapulgite clay ("GQA") and the relevant geographic market is the United States. The Court shall first address the relevant product market.

While all products compete at some general level, in three landmark cases, the Supreme Court recognized that not every possible substitute for a given product falls within the parameters of an antitrust relevant market.

For every product, substitutes exist. But a relevant market cannot meaningfully encompass that infinite range. The circle must be drawn narrowly to exclude any other product to which, within reasonable variations in price, only a limited number of buyers will turn; in technical terms, products whose "cross-elasticities of demand" are small.

Times-Picayune Publishing Co. v. United States, 345 U.S. 594, 612 n. 31, 73 S. Ct. 872, 882 n. 31, 97 L. Ed. 1277 (1953).^[6]

In considering what is the relevant market for determining the control of price and competition, no more definite rule can be declared than that commodities, reasonably interchangeable by consumers for the same purposes make up that "part of the trade or commerce," monopolization of which may be illegal.

United States v. E.I. du Pont de Nemours & Co., 351 U.S. 377, 395, 76 S. Ct. 994, 1007, 100 L. Ed. 1264 (1956).

The outer boundaries of a product market are determined by the reasonable interchangeability of use or the cross-elasticity of demand between the product itself and substitutes for it. However, within this broad market, well-defined submarkets may exist which, in themselves, constitute product markets for antitrust purposes. The boundaries of such a submarket may be determined by examining such practical indicia as industry or public recognition of the submarket as a separate economic entity, the product's peculiar characteristics and uses, unique production facilities, distinct customers, distinct prices, sensitivity to price changes, and specialized vendors.

Brown Shoe Co. v. United States, 370 U.S. 294, 325, 82 S. Ct. 1502, 1523-24, 8 L. Ed. 2d 510 (1962) (citations omitted). Over the last half century, courts have found a variety of ways to express the concepts of "cross-elasticity of demand" and "reasonable interchangeability." In the Eleventh Circuit's most recent pronouncement on the subject, the Court of Appeals stated:

Defining a relevant product market is primarily a process of describing those groups of producers which, because of the similarity of their products, have the ability — actual or potential — to take significant amounts of business away from each other.

U.S. Anchor Mfg., Inc. v. Rule Industries, Inc., 7 F.3d 986, 995 (11th Cir.1993) (quotation marks omitted).

Within this framework, the Plaintiff has strenuously argued that pursuant to § 1.11 of the U.S. Department of Justice and Federal Trade Commission Horizontal Merger Guidelines (1992), when determining the boundaries of a relevant product market, "the relevant inquiry is whether there exist substitutes to which a consumer would switch in response to a small but significant (i.e., 5 or 10 percent) price increase in the product in question (here, gel clay)." Post-Trial Submission of the United States at 8.^[7] Under this test, if such a 5 or 10 percent price increase, "would not cause a sufficient number of customers to switch to other products to make the price increase unprofitable, that product ... is a relevant product market or line of commerce under Section 7 of the Clayton Act." Id. at 5.

This "5%-10%" test was critical to the testimony of Plaintiff's relevant market experts and formed the foundation of Plaintiff's case-in-chief. See e.g., Tr. Vol 6 at 101:2-4 (Dr. Bodisch stating, "If the customers told me they would not switch away from attapulgite in response to a 10 percent price increase, I have excluded the alternative from the market"); 23:21-23:24; 26:12-26:16; 31:10-31:12; 68:2-68:6; 70:5-70:8; 74:15-74:20. However, notwithstanding Plaintiff's position that application of this test is dispositive in this case, it must be noted that:

In 1982, when the Department of Justice first announced the five-percent test, the Attorney General noted that the Merger Guidelines merely serve the purpose of determining what mergers the department will challenge. United States Department of Justice, Merger Guidelines, 1982 (Explanation and Summary) reprinted in 4 TRADE REGULATION REPORTS (CCH) ¶ 13,102, at 20,528.... Certainly, the Guidelines are not binding on the courts....

Olin Corp. v. F.T.C., 986 F.2d 1295, 1300 (9th Cir.1993). See F.T.C. v. PPG Industries, Inc., 798 F.2d 1500, 1503 n. 4 (D.C.Cir.1986) ("Department of Justice Guidelines offer a useful illustration ... but are by no means to be considered binding on the court"). On the other hand, the Court recognizes that in many cases faithful application of the 5%-10% test will, in fact, result in an accurate description of the relevant product market. Thus, analysis must begin by determining whether Plaintiff's application of the 5%-10% test provides an accurate picture of the relevant product market for the instant action. For the following reasons, the Court finds that it does not.

As discussed above, attapulgite gel clay is generally used as a thickener or suspension agent. The evidence shows that in most of its applications GQA forms only a small percentage, anywhere from 1/10th of a percent to about 10 percent, of the total cost of the final product in which it is an ingredient. On average, it appears that GQA comprises 5 percent or less of the overall cost of the final products in which it is a part. In addition, each final product in which GQA is used has been specially formulated to achieve some particular end-use requirement. The evidence further shows that most, if not all, of the products in which GQA is used would have to undergo some degree of reformulation in order achieve the same end-use requirements without or with less GQA.

In light of these factors, the results of the 5%-10% test may not provide the same information that they would for a direct, end-use product like pens or shoes. One must ask whether a decision not to substitute another product in place of GQA in the face of a 5 to 10 percent rise in the cost of GQA is indicative of a lack of alternate competitive products within the relevant market or simply a matter of triviality for the purchaser? For example, in a product in which GQA comprises 5 percent of the overall cost of the product, a 5 percent rise in the cost of GQA will only result in a .25 percent increase in the overall cost of the product. In other words, in a $100 product, a 5 percent increase in the cost of GQA would result in a 25 cent increase in the overall product cost.

By the Plaintiff's own standards, a relevant market would consist of a product for which a sole producer could institute a small but significant price increase without losing enough customers to substitute products so as to make that increase unprofitable. While "significant" in that paradigm is clearly meant to be looked at from the viewpoint of the seller, in the present case, market realities dictate that significance must also be examined from the point of view of the purchaser. The "varying circumstances of each case" determine the cross-elasticity of demand, duPont, supra, 351 U.S. at 395, 76 S.Ct. at 1007, and in this case, the significance of GQA price increases to purchasers must be considered in order to determine whether a negative response to the 5%-10% test reveals meaningful information about product competition or simply gives insight into market complacency.

When the cost of reformulation is also taken into consideration, the Plaintiff's 5%-10% test betrays itself. The record is replete with examples of customers who purchased attapulgite exclusively from Engelhard or exclusively from Floridin. In many of these cases, simply changing GQA suppliers would require product testing and potential reformulation. See e.g., Tr. 2 at 51:24. In light of the low cost of GQA as a percentage of their overall product cost and the potential for significant qualification costs, some customers stated that they would not switch from Engelhard GQA to Floridin GQA, or vice versa, if their present supplier raised the price by 5 or 10 percent. See Tr. Vol. 5 at 126:15-126:18; [Director of Purchasing for a pesticide manufacturer] Dep. at 34:9-34:23. Pursuant to the Plaintiff's formalistic application of the 5%-10% test, such information would require a finding that Engelhard GQA and Floridin GQA constituted separate product markets. Such a conclusion would reflect the notion that only physically identical products could be in the same market. The Supreme Court has rejected such a position. See duPont, supra, 351 U.S. at 394, 76 S.Ct. at 1006-07.

Plaintiff's expert witness, Dr. Frederick Warren-Boulton, discussed in his rebuttal testimony the importance of being unimportant. Tr. 11 at 18:16-18:17. The Court does not dispute the fact that a low-cost item in a high-priced, specialized formulation enjoys a certain amount of protection against cost-cutting substitutions, since the potential benefits of the minimal cost savings will often be outweighed by the fear of distorting the quality of the final product. To the same degree that this must be kept in mind as a potential source of inhibition to the free flow of competition, so too must it be acknowledged in formulating a realistic test to determine the boundaries of the relevant product market. "Indeed, the Department of Justice itself acknowledges that a higher percent increase in price is appropriate in determining the relevant product market in certain cases." Olin, supra, 986 F.2d at 1302 (citing United States Department of Justice, Merger Guidelines — 1984 § 2.11 n. 7 ("For example, a larger increase may be appropriate[] if the `price' to be increased is a tariff or commission that constitutes a small fraction of the price of the product being transported or sold")). Understanding the circumstances under which most GQA purchasers operate, the answer to the 5%-10% question posed by the Plaintiff in this case almost seems predetermined. Thus, as applied by the Plaintiff, the test tells very little about competing products and actual competition to GQA.^[8]

Plaintiff's application of the 5%-10% test suffers from another shortcoming. The Plaintiff only made inquiries with current purchasers of GQA. Tr. 7 at 73:6-73:13; Tr. 11 at 90:10-90:19. In so doing, Plaintiff ignored perhaps the best source of information relevant to the product market inquiry, to wit: those customers who have already switched to an alternate product. While Dr. Bodisch correctly stated that such switches away from GQA may have been motivated for reasons totally irrelevant to the product market inquiry, Tr. 7 at 73:16-73:17, so too may they have been the result of price competition in combination with functional interchangeability. So again, the Court is precluded from viewing meaningful actions of consumers in the marketplace.

Perhaps more importantly, by only interviewing current GQA users about products that currently employ GQA, the Plaintiff's 5%-10% test presupposes that competition can only exist at the post-formulation stage, when GQA has already been chosen as an ingredient in an end-use product. Again, this forces the customer to account for qualification and reformulation costs when responding to the 5%-10% test, which raises the problems discussed above. Further, the Supreme Court contradicted such a position in United States v. Continental Can Co., 378 U.S. 441, 84 S. Ct. 1738, 12 L. Ed. 2d 953 (1964). In finding that glass containers and metal containers competed in a relevant product market for § 7 purposes, the Supreme Court stated:

Interchangeability of use and cross-elasticity of demand are not to be used to obscure competition but to recognize competition where, in fact, competition exists.... Thus, though the interchangeability of use may not be so complete and the cross-elasticity of demand not so immediate as in the case of most intraindustry mergers, there is over the long run the kind of customer response to innovation and other competitive stimuli that brings the competition between these two industries within § 7's competition-preserving proscriptions. ... That there are price differentials between the two products or that the demand for one is not particularly or immediately responsive to changes in the price of the other are relevant matters but not determinative of the product market issue.

Id. at 453-455, 84 S.Ct. at 1745-1746. The Supreme Court so held notwithstanding the fact that:

[T]he machinery necessary to pack in glass is different from that employed when cans are used [or] that a particular user of cans or glass may pack in only one or the other container and does not shift back and forth from day to day as price and other factors might make desirable....

Id. at 450, 84 S.Ct. at 1743. Thus, the Supreme Court contemplated that competition at the reformulation or manufacturing facility restructuring phase is relevant to the product market issue. While the Continental Can decision makes no reference to a 5%-10% test, the distinct possibility that a glass bottler might not restructure its bottling facility to accommodate canning in response to a 5 or 10 percent increase in the cost of glass is not beyond serious question. Again, this does not mean, ipso facto, that simply because a product can be used in the same end-use functions as GQA that it is in the same relevant market. Rather, it highlights the failure of the 5%-10% 0% test to account for the possibility that purchasers and potential purchasers of GQA could opt to use a substitute substance for the same function when creating a new product or retooling an old one. Such formulation stage competition could very well serve as a restraint against anticompetitive price increases by forcing GQA producers to price their products competitively or price themselves out of the market completely.^[9] While this may or may not be the case, the breadth and scope of the Plaintiff's 5%-10% inquiry in this action leaves no way of knowing the answer with any reasonable certainty.

While the 5%-10% test may be very useful for the purposes of examining the hundreds of proposed transactions that the Department of Justice must evaluate yearly and may readily define the product market in many of those instances, its application to the facts in the instant case is of little use in establishing a relevant product market. Indeed, based solely on this evidence, the Plaintiff would not even meet its burden of proof on this issue. However, the Plaintiff introduced additional evidence on the issue of relevant product market during the trial of this case. The analysis must now focus on whether the full record establishes by a preponderance of the evidence that the relevant market for § 7 purposes is gel quality attapulgite clay produced in the United States. For the following reasons, the Court finds that it does not.

A. Evidence of Substitution in Response to a Price Increase Above 5 to 10 Percent Was Unreasonably Discounted

As previously discussed, the Plaintiff relied heavily on the Merger Guidelines 5%-10% test in concluding that GQA formed a relevant market for antitrust purposes. As a result, the Plaintiff discounted evidence in the record of substitution, or at least initiation of a substitution process, by GQA consumers in response to different levels of "small but significant price increases." In fact, several customers indicated that substitution would occur at price increase levels at or slightly above those proposed by the Plaintiff.

Of three GQA customers in the animal feed industry interviewed by Dr. Bodisch^[10], one manufacturer indicated that he would switch from GQA to gums if GQA prices rose approximately 15%. Tr. 6 at 111:13-111:24; DX 1135 at DJ-A4-02168. A second manufacturer indicated that a rise of 10 to 20 percent in GQA prices would initiate a search for substitutes. Tr. 6 at 112:23-113:4. The third interviewee, a raw materials supplier indicated that switches away from GQA and toward increased gum usage would occur if the price of GQA went up by $10 per ton. Tr. 6 at 110:114-111:1. Even at the low end of GQA prices, around $140 per ton, such a per ton cost increase would only amount to about a 7 percent change in price.

Another customer, who uses GQA in a "sound deadening" end-use product, testified in a CID deposition that his employer, [a chemical company], would likely switch to a substitute product if the price of GQA rose 10 percent, or perhaps even less. [Purchasing Manager for a chemical company] CID Dep. at 38:2-38:5.

One of Dr. Bodisch's interviewees, a GQA customer in the tape joint compound industry, stated that in the face of a 5 to 10 percent increase in GQA prices he might look into reformulation although customer satisfaction would be the number one concern. Tr. 6 at 117:4-117:16. Dr. Bodisch's reaction to this particular customer response exemplified the rigidity of the Plaintiff's approach to defining the relevant market.

Q: You told us, sir, that nobody would look at reformulation at that level of increase. Did I miss [sic] hear you?

A: I think I said nobody would switch. Nobody told me definitively they would switch.

Tr. 6 at 117:17-117:20 (emphasis added). Thus, even if substitution was contemplated in response to minimal price increases, Dr. Bodisch would not consider further inquiry if a cold-called interviewee was not willing to definitively commit to technical reformulation at a 5 or 10 percent increase.^[11]

Other customers indicated that switching would occur at slightly higher levels. Asphalt roof coating manufacturer Jack McClellan, of DeWitt Products, testified that a 25 percent price increase would initiate a reformulation process at his company. McClellan CID Dep. at 50. (10% increase in final product cost or 25% increase in GQA cost). Robert Knauer of Benjamin & Moore Company, a paint manufacturer, testified that substitution would occur if GQA prices rose somewhere between 25 and 50 percent. Tr. 2 at 15:4-15:7. Two other paint manufacturers interviewed by Dr. Bodisch similarly indicated that a 25 to 50 percent increase in the price of GQA would be required to initiate substitution. See DX 1133 at DJ-A1-00001; DX 1133 at DJ-A1-00018. As to these and other paint manufacturers interviewed in preparation of his expert report, Dr. Bodisch testified:

They told me that for an excessive price increase, defined as maybe 25 percent or so, they might start looking [sic] alternatives; but with respect to the relevant antitrust question for market definition, consistently the paint manufacturers told me they would not go to alternatives for gel clay in response to a 10 percent price increase.

Tr. 6 at 74:15-74:20. In fact, Dr. Bodisch stated that a lot of the customers he interviewed in the paint industry indicated that they would switch to, or look for, substitute products if GQA prices rose in the neighborhood of 25 to 30 percent. Tr. 6 at 78:11.

Of course some customers indicated that price increases would have to be even larger to induce substitution. A raw material distributor in the suspension fertilizer industry indicated to Dr. Bodisch that a 40 to 50 percent GQA price increase would be required to induce substitution in his industry. Tr. 6 at 103:-103:15; DX 1135 at DJ-A4-02167. A paint manufacturer indicated to Dr. Bodisch that it would take a "gigantic" price increase to prompt reformulation. DX 1133 at DJ-A1-00005. Jim Nattier, of Baroid Drilling Fluids Company, testified at trial that there is no price increase level that would induce his company to substitute for GQA in salt water drilling muds due to the belief that there are no available substitutes for GQA in such products. Tr. 4 at 109:18-109:25. Allen Harrison of UOP, a molecular seive producer, stated that similar conditions prevail in the molecular seive industry. Tr. 2 at 59:9-59:12.

Thus, the evidence shows that at least some GQA customers would respond to a price increase of less than 20 percent by switching to or searching for substitute products. In addition, a greater number would switch away in response to price increases between 25 and 50 percent. Again, in light of the small role that GQA plays in most end-use products, such facially compelling numbers lose some of their force. See e.g., Tr. 6 at 78:18-78:22 (A paint industry consumer "guesses it would take at least a 25 to 50 percent price increase in gel clay before he would look for substitutes because it is only three-tenths of 1 percent of the cost of his product which translates into about one cent on a gallon of paint").

Even considering the customers who indicated that a GQA price rise would have to be gigantic to induce substitution, the Court finds that the evidence does not paint as clear a picture as the Plaintiff would suggest. The evidence appears less conclusive still when account is taken of the numerous customers who indicated that no substitution would occur at a 5 to 10 percent price increase, yet never pinpointed a price increase level at which substitution would occur. These problems are compounded by the fact that Dr. Bodisch drew his conclusions without ever ascertaining the full size of the individual end-use markets he was analyzing. For example, in relation to paints, Dr. Bodisch guessed that the number of paint manufacturers in the United States numbered over one hundred, Tr. 6 at 73:18, yet admitted that he had interviewed only six to eight paint manufacturers in preparing his expert report. Tr. 6 at 74:7. Dr. Bodisch could not conclusively identify the numbers of companies using thickeners in animal feeds, asphalt coatings, suspension fertilizers or tape joint compounds. Tr. 7 at 8:1-8:23. While knowledge of the "all thickeners" market is not a requirement, Dr. Bodisch's testimony indicated a lack of awareness of even the total number of GQA customers in those markets. In light of the variety of responses by the fairly narrow group of customers interviewed about a hypothetical price increase, a deeper look into these product markets was required.

Perhaps if no customers had been willing to substitute at any level near 10 percent, then that "cushion" would bolster the reliability of the 5%-10% test. However, that was not the case, and by ignoring evidence of substitution in response to small price changes, though greater than 10 percent, the relevant question — is there cross-elasticity of demand between GQA and other products — once again appears approached but unanswered. Without greater detail, it remains guesswork for the Court to determine whether no reasonably interchangeable, competitive products exist or whether such products exist but would not be turned to immediately by current GQA users due simply to triviality or the specter of reformulation costs.

B. Evidence of Customers' Use of or Ability to Incorporate Substitute Products Was Unreasonably Discounted

The major applications for GQA are in suspension fertilizers, animal feeds, paints, tape joint compounds, asphalt roof coatings, drilling fluids, and molecular sieves. Tr. 7 at 103:9-103:19; DX 1109 at 4.

The record suggests that substitute products are either currently used or available for use by GQA customers in several of the main end-use applications of GQA. The Court shall evaluate the evidence in this regard on a market-by-market basis.

1. Suspension Fertilizers

a. Testimony

Ted Walden, Manager of the Fertilizer Division of United Suppliers in Elora Iowa, testified by deposition on July 13, 1995. United Suppliers makes phosphate and potash suspension fertilizer and uses GQA in both. Walden Dep. 9:9-9:17. United Suppliers has purchased GQA exclusively from Engelhard for the past 11 years, Walden Dep. at 10:13. While Mr. Walden has heard of other companies using substitute products in suspension fertilizers, he has no direct knowledge of the same. Walden Dep. at 30:7-30:19. Mr. Walden stated that an increase in the price of GQA would likely result in a shift to production of dry fertilizers, which require no GQA, as opposed to an effort to reformulate. Walden Dep. at 31:17.

b. Customer Calls by Dr. Bodisch

In preparing his expert report, Dr. Bodisch interviewed four GQA customers in the suspension fertilizer industry. One of these customers was a distributor of raw materials to suspension fertilizer producers, and this customer indicated that 30 to 35 percent of suspension fertilizers could use sepiolite as a complete or partial substitute for GQA depending on the acidity and location of the end-use product. Tr. 6 at 102:16-103:4; DX 1135 at DJ-A4-02167. The same distributor indicated that at least one of his customers used sodium bentonite in their liquid fertilizers. Tr. 6 at 103:5-103:7; DX 1135 at DJ-A4-02167. However, this distributor further stated that it would take a 40 or 50 percent increase in the price of GQA to induce substitution. Tr. 6 at 103:9-103:15.

Another customer indicated that sodium bentonite as well as a new polymer might work as substitutes for GQA. Tr. 6 at 104:13-104:21; DX 1136 at DJ-A4-02244. However, this customer also indicated that he knew little about the polymer and that sodium bentonite did not work as well as GQA. Tr. 6 at 104:15-104:21; DX 1136 at DJ-A4-02244. Dr. Bodisch indicated that this customer also referred to the use of gums and xanthum in suspension fertilizers. Tr. 6 at 105:2-105:3. In relation to these potential substitute products, the ensuing discourse revealed some of the problems previously discussed in this opinion.

Q: Would it make any difference to the (sic) your analysis how many people were actually using the product?

A: Using which product?

Q: Gums or xanthum, for example?

A: Only if a price increase in gel clay would induce substitution to that product. The fact that somebody else is using it does not necessarily suggest the gel clay producers could not raise price.

Q: The fact that they are using it does suggests [sic] that at some point they have made the decision that it is the best product for them to use, price and other factors taken into account, does it not, sir?

A: Yes, it does, for that product; but that decision has already been determined. The issue here is if gel clay prices go up, would somebody who is now using xanthum gums switch to gel clay, not somebody who has already made the decision, for whatever reason, to use xanthum gums. That's a different question. What I am trying to identify here is what happens in response to a price increase in gel clay.

Tr. 6 105:4-105:23. Dr. Bodisch more specifically explained that the 5%-10% test dictated which alternatives needed further investigation.

Q: Let me ask you the question of whether you would regard xanthum and attapulgite as reasonably interchangeable if, in fact, a customer having considered both, were the [sic] choose xanthum?

A: They might be substitutes but not in the context of a 5 or 10 percent price increase in gel clay.

Tr. 6 106:17-106:22.

Another suspension fertilizer industry customer interviewed by Dr. Bodisch indicated that he is presently considering switching from GQA to sodium bentonite, and that he was also considering a new polymer as a substitute for GOA. Tr. 6 at 107:14-107:25; DX 1136 at DJ-A4-02241.

The final customer in the suspension fertilizer industry interviewed by Dr. Bodisch stated that he used GQA in all his products, was unfamiliar with sepiolite or bentonite, and that he had tried a xanthum gum product with limited success. DX 1133 at DJ-A1-00014.

In light of these responses, Dr. Bodisch again looked to the 5%-10% test:

Q: The answers you got, at least from this group of customers, were that other products were actually in use, were they not, but that they didn't answer your particular question of what would happen in response to a 10 percent price increase?

A: That's right, with respect to this customer.

Tr. 6 at 108:10-108:14.^[12]

Thus, notwithstanding considerable evidence about current use or potential use of alternative products, Plaintiff saw no need to conduct a further inquiry in this consumer market because switching would not occur at the 5 to 10 percent range.

2. Animal Feeds

a. Testimony

No witnesses from the animal feed industry testified at trial or by way of deposition.

b. Customer Calls by Dr. Bodisch

Dr. Bodisch interviewed a distributor of raw materials to animal feed manufacturers who stated that both GQA and sepiolite could be used in animal feeds, with sepiolite capturing the majority of the sales in the northwestern region of the United States. Tr. 6 at 109:21-109:23; DX 1135 at DJ-A4-02166. This distributor further indicated that sepiolite was cheaper than GQA, and that several large companies were currently using it in formulations. Tr. 6 at 109:24-110:2. However, Dr. Bodisch testified that at least some of those companies had switched back to GQA due to poor performance, notwithstanding cheaper prices. Tr. 6 at 110:2-110:7. This distributor further indicated that gums are used for purposes similar to GQA by some manufacturers, and that he thought some would switch towards increased gum usage if the price of GQA went up by $10 per ton. Tr. 6 at 110:14-111:1.

An interview with a manufacturer of animal feeds revealed that this GQA customer only used xanthum gum in liquid feeds, and that if the price of GQA went up by approximately 15% he would switch from GQA to gums in those applications where GQA is used. Tr. (Bodisch) Vol. 6 at 111:13-111:24; DX 1135 at DJ-A4-02168.

A second animal feed manufacturer stated that he used both GQA and sepiolite in his formulations, and that the two products are essentially "one to one." Tr. 6 at 111:25-112:2. In addition, Dr. Bodisch testified about another company using a substance called halocite is imported from New Zealand. Tr. 6 112:14-112:19; DX 1133 at DJ-A1-00019. The interviewee indicated that a price rise of 10 to 20 percent in GQA would initiate a search for substitutes, including the imported material from New Zealand. Tr. 6 at 112:23-113:4.

Thus, in the animal feed industry, substitute products have either been used, are currently being used, or appear available for purposes of substitution.^[13]

3. Tape Joint Compounds

a. Testimony

Pat Hudgens, purchasing manager for Georgia Pacific's Gypsum Division, testified by way of deposition, taken July 6, 1995. Georgia Pacific uses GQA in a variety tape joint compounds. Hudgens Dep. at 7:24-8:2. Mr. Hudgens testified that Georgia Pacific has looked at alternatives in the past, including bentonite, but has not found a substitute for GQA. Hudgens Dep. at 12:1-12:13. He added that he was unaware of any direct substitute for GQA. Hudgens Dep. at 11:21-11:22. Mr. Hudgens acknowledged that his company has really never had a need to seriously evaluate alternatives to GQA. Hudgens Dep. at 42:9-42:10.

b. Customer Calls by Dr. Bodisch

Dr. Bodisch interviewed one manufacturer of tape joint compound who stated that there are no substitutes for GQA in his products and that he would not switch away from GQA in the face of a 10 percent price increase. Tr. 6 at 115:12-115:18. DX 1136 at DJ-A4-02251.

A second tape joint manufacturer, however, told Dr. Bodisch that his company currently uses liquid associative thickeners as a substitute for GQA in about 10% of its tape joint compounds, and that the two different formulations are actually marketed and sold as the same product. Tr. 6 at 115:19-116:20; DX 1136 at DJ-A4-02260. The same customer also stated that his company had looked at sepiolite and found it to perform more efficiently than GQA, but that the sepiolite was only cost efficient as a complement to GQA rather than as a substitute. DX 1136 at DJ-A4-02260.

A third customer discussed the use of sepiolite in combination with GQA in the its west coast plants, and that substitution to sepiolite would be a possibility in response to a GQA price increase, so long as customer satisfaction could be maintained. Tr. 6 at 117:4-117:16.

Thus, the record indicates that substitutes are in fact used or available to replace GQA, either partially or fully, in tape joint compound applications. In fact, in the case of liquid associative thickeners, the substitution has been so complete that the substitute-bearing product and the GQA-bearing product are marketed as one in the same.^[14]

c. Documentary Evidence

American Colloid Company issued a document touting "Accu-Gel F Volclay" as a replacement to GQA "in the production of asbestos-free compounds requiring a thixotrope to improve workability and spreadability." DX 1156. The document specifically refers to use in tape joint compounds, indicates that the Volclay is a cost effective substitute to GQA. Id.

An Engelhard customer call report from 1991 indicated that one customer was "embarking on new technology with an `Attapulgite free' joint compound. This involves the use of organic thickeners and air float Kaolin. This produces a viscosity stable compound at a lower price than Attapulgite." DX 1250.

4. Paints

a. Testimony

Robert Knauer, a consultant for Benjamin Moore & Company, a paint manufacture, testified during the trial of this action. Mr. Knauer stated that Benjamin Moore uses GQA in some but not all of its solvent-based paints. Tr. 2 at 7:8-7:9. For those paints in which GQA is an ingredient, Benjamin Moore considers qualified products from Engelhard and Floridin to be "complete alternates," interchangeable "on a pound-for-pound basis ... without any change in the formulas." Tr. 2 at 12:18-12:21. Mr. Knauer stated that he knew of no complete alternative for GQA in the paints which use GQA. While Benjamin Moore also manufactures water-based and latex paints, it does not use GQA in those formulations. Tr. 2 at 7:22.

Mr. Knauer identified several thickeners that are used in Benjamin Moore's solvent paints which use no GQA, including: Natrosol Plus, Tr. 2 at 27:11; wax based thickeners, Tr. 2 at 28:2; Viscotrol, Tr. 2 at 28:22; castor oil based products, Tr. 2 at 29:1-29:4; and Aerosil, Tr. 2 at 29:22. However, Mr. Knauer did not indicate that these products could replace GQA. On the other hand, Mr. Knauer did acknowledge that GQA and Bentone SD-1 can compete at the formulation level for use in solvent-based paints. Tr. 2 at 43:16-43:20 and Tr. 2 at 45:15-45:17.

James Depew, Vice President of Purchasing for Glidden Paint Company also testified during trial. Without stating which types of paints in which GQA is used by Glidden, Mr. Depew stated that Glidden purchases about 800 tons of GQA annually. On direct examination, Mr. Depew confirmed what he had previously written in a letter, DX 1028, stating that if an increase in GQA price had an unfavorable affect on any of Glidden's products, the company would be able to reformulate to eliminate or reduce GQA usage. Tr. 5 at 113:6-113:21. Mr. Depew also testified that, to his knowledge, substitutes for GQA were presently available, Tr. 5 at 113:25, although he was unable to identify any on the stand. Tr. 5 at 127:4-127:11. As GQA comprises about 1/10th of a percent of Glidden's overall raw material costs, Tr. 5 at 114:9-114:23, Mr. Depew did not believe that a 5 percent price increase would lead to substitution. Tr. 5 at 116:21-116:23.

John McMahon, Jr., of Majemac company, a wholesale distributor to the paints and plastics industries, testified by deposition, taken July 13, 1995. Mr. McMahon stated that Majemac supplies 167 paint customers. McMahon Dep. at 5:19-5:21. He further testified that GQA is normally used in waterbased paints, and that cellulosic thickeners and organoclays are largely used in solvent or oil based paints. McMahon Dep. at 8:23-9:3. However, he stated that cellulosic thickeners and organoclays could compete with GQA in water-based formulations, McMahon Dep. at 8:14-8:18, and that customers could offset increased GQA prices by adding more cellulosics and organoclays while reducing GQA usage. McMahon Dep. at 10:14-10:17. In fact, Mr. McMahon believed that GQA could likely be completely replaced, perhaps by cellulosics, but that reformulation would be required. McMahon Dep. at 22:19-22:23.

b. Customer Calls by Dr. Bodisch

One interviewee contacted by Dr. Bodisch stated that GQA was used in water based paints, and that his company would pass on to customers a 5 percent price increase in GQA prices. DX 1133 at DJ-A1-00001.

A second GQA purchaser told Dr. Bodisch that his company used GQA as a thickener in latex paints, and indicated that it would take a "gigantic" price increase to prompt reformulation. DX 1133 at DJ-A1-00005.

A third paint manufacturer indicated that there are several substances that can work well with GQA in water based and latex paints, but did not discuss any substitutes. DX 1133 at DJ-A1-00010. This customer did mention that bentones can be used in some solvent paints instead of GQA. Id.

Another customer stated that he used GQA in latex paints, and he was unaware of substitutes for GQA. DX 1133 at DJ-A1-00018. However, this customer also noted that there had never been a need to look for substitutes, and that a 25 to 50 percent price increase would be required to prompt such a search. Id.

Another customer, a distributor, called by Dr. Bodisch revealed that at least one paint company was using some type of montmorillionite product as a substitute for GQA in water based paints. DX 1133 at DJ-A1-00021. The interviewee indicated that he did not believe that GQA competitors knew about this use of montmorillionite, and that he hoped to gain 25 percent of the thixotropic market for paints with this product in the next five years. Id.

Thus, of three witnesses from the paint industry called to testify at trial or by deposition, two believed that GQA competed with and could be partially or fully replaced by alternative thickeners in response to a price increase. While the third witness did not know of alternatives for GQA in his products where GQA had already been incorporated, he did acknowledge that GQA and Bentone SD-1 compete at the formulation levels. On the other hand, while one of Dr. Bodisch's customer calls revealed some use of montmorillionite as a paint thickener, the remainder generally indicated a lack of competitive alternatives to GQA.

A close reading of the testimony and customer call notes reveals an important piece of information. On the one hand, James Depew, John McMahon, and all the customers called indicated that GQA is only used in water-based or latex paints. On the other hand, Robert Knauer, who was called as a witness for the Plaintiff in its case-in-chief, testified that Benjamin Moore only uses GQA in its solvent-based paints, and does not use GQA at all in its water-based and latex paints. Tr. 2 at 7:8-7:22. Mr. Knauer testified that in latex paints Benjamin Moore uses cellulosic and associative thickeners. Tr. 2 at 11:24-12:14.

One of the Plaintiff's main contentions in this action has been that while many thickeners exist, they cannot and do not compete with GQA in those applications that use GQA. The premise of this argument is that specific applications require very specific results, and thus small variations in end-use requirements can be the enough to preclude head-to-head competition, notwithstanding simultaneous use in the same general market or specific application. However, this argument is dealt a terrible blow when the government's own witnesses testify that some companies make all of their solvent paints without GQA, while others only use GQA in such formulations, and some companies make all of their water-based and latex paints without GQA, while others use only GQA in such formulas. The logical conclusion is that either form of paint can be manufactured at competitive prices either with or without GQA, and that the reluctance of companies to change formulas is not based upon a lack of competitive alternatives within the relevant market, but rather is the result of complacency due to the trivial cost of attapulgite in most paint formulations.^[15]

c. Documentary Evidence

In assessing its entrance into the fine grind market, Floridin analyzed competition in several end-use applications. See DX 1251. In evaluating the paint market, Floridin accounted for a variety of thickeners which presented different competitive problems or advantages for GQA. These thickeners included cellulosics, associative thickeners, hectorite, organoclays, and fumed silica. See id. at FL 72 000097-FL 72 000098.

5. Drilling Fluids

a. Testimony

At trial, Jim Nattier, Vice President of Industrial Products for Baroid Drilling Fluids Company, testified about the use of GQA in the drilling fluids industry. In the most basic terms, drilling fluids serve as a lubricant protecting the drill during the process of drilling for oil and gas. Mr. Nattier explained that there are three basic classifications of drilling fluid-fresh water, salt water and oil based. Tr. 4 at 92:10-92:16. These fluids are not interchangeable, but rather customized for particular drilling operations. Tr. 4 at 92:25. GQA is the primary constituent of salt water drilling fluids, 4 at 93:8-93:12, while bentonite is generally used for fresh water drilling fluids. Tr. 4 at 94:24-94:25. In oil based drilling fluids, both chemically altered attapulgite and chemically altered bentonite are used. Tr. 4 at 95:1-95:2. Mr. Nattier stated that the chemically modified agents cannot be substituted for GQA in salt water drilling muds, Tr. 4 at 95:5, and that he knows of no substitute for GQA in salt water drilling fluids. Mr. Nattier also explained that because drilling an oil or gas well is a several million dollar venture in which the cost of the drilling fluids are minimal, that formulation decisions are driven mainly by performance rather than cost. Tr. 4 at 100:12-100:22. In fact, Mr. Nattier stated that only a lack of GQA or the failure of GQA to perform in a particular application would cause Baroid to switch to something else. Tr. 4 at 109:18-109:25. While a foreign affiliate of Baroid uses Senegalese attapulgite in salt water drilling fluids, Baroid has never used foreign attapulgite in its salt water drilling fluids due to poor quality. Tr. 4 at 113:9-118:14.

Ray Somers, Director of Procurement for Baroid Drilling Fluids, testified at trial. Mr. Somers indicated that Baroid purchased clay from Floridin and Milwhite, but not Engelhard. Tr. 4 at 128:2-128:16. Other than indicating concern about losing one of three potential suppliers of GQA, Mr. Somers had little to add to Mr. Nattier's testimony on the issue of relevant markets in the drilling fluids industry.

During a Civil Investigative Demand deposition, taken May 8, 1995, David L. Ruff, General Manager of Floridin discussed drilling fluids. Mr. Ruff stated that about 10 to 15 percent of Floridin's business is in drilling fluids ("drilling muds"). Ruff CID Dep. at 113:2-113:4. Mr. Ruff testified that Floridin had in fact lost a salt water drilling mud customer to a synthetic-alternative. Ruff CID Dep. at 110:9-10:22. This loss was discussed in a Floridin interoffice memorandum regarding its drilling mud customer, Coastal Caisson Corporation. See DX 1226. The memo discusses the loss of approximately 300 to 400 tons per year of Florigel H-Y to Coastal Caisson as a result of a switch to a product called "Super Gel." Id.

This Super Gel is a polymer which is used in fresh and salt water and one five gallon pale[sic] has about the same efficiency as a ton of H-Y. It's cost per five gallon is about the same as a ton of H-Y.... Other drilling firms in the area such as Smith Drilling state they too use Super Gel when required but they also use Florigel because it's main advantage is that it is not harmed or degraded by high acid conditions or a very high pH at the drill site.

Id. See also DX 1258 (interoffice memo discussing loss of salt water drilling fluid business to pre-hydrated bentonite). In his deposition testimony, Mr. Ruff added that recent technological advances have allowed customers to switch to treated sodium bentonite in salt water drilling mud applications. Ruff CID Dep. at 112:18-116:9. See DX 1179 at 2 ("Reports claim salt water mud clays (Attapulgite) have been increasing prices at a much higher rate than Sodium Bentonite and that it is now more economical to pre-hydrate Sodium Bentonite than to use Attapulgite").

b. Customer Calls by Dr. Bodisch

One GQA customer called by Dr. Bodisch was a distributor of raw materials to the drilling mud industry. This distributor indicated that an increase in the cost of GQA would likely result in a shift to sepiolite or treated bentonite. DX 1133 at DJ-A1-00003. Dr. Bodisch noted, however, that this contact was not an engineer and thus not certain about possible technical barriers to such reformulations. Id.

A drilling mud manufacturer who uses GQA in all his salt water drilling muds indicated to Dr. Bodisch that both polymers and sepiolite are alternatives to GQA in the drilling mud industry. DX 1133 at DJ-A1-00006. This manufacturer noted, however, that polymers are very expensive, and he did not expand on the use of sepiolite except to say that it is the "closest" to GQA. Id.

The trial testimony of both witnesses for Baroid Drilling was perfectly clear — there are no substitute for GQA in salt water drilling mud applications. However, the customers called by Dr. Bodisch seemed to contradict that position, suggesting that sepiolite, polymers or treated bentonite are possible alternatives to GQA. Mr. Ruff also discussed the use of treated sodium bentonite in salt water drilling fluid applications, and this testimony was is supported by historical data.^[16]

Thus again, based on the strong testimony of witnesses from one company, the Plaintiff seemed willing to overlook evidence of potential substitutes that did not fit neatly into the Merger Guidelines product market test. In addition, the testimony of Jim Nattier makes clear that the search for substitute products is not a high priority due to the minor cost of drilling fluids as a part of extremely expensive drilling ventures. Thus, the question of lack of alternatives versus market complacency again arises, but due to the Plaintiff's reliance on the 5%-10% test, probative questions in this regard are neither asked nor answered.

6. Asphalt Roof Coatings

As a general matter, all witnesses agreed that asbestos was the product of choice for asphalt roof coatings. However, due to environmental and insurance coverage concerns, the industry, in large part, has been forced to turn to alternative thickeners.

a. Testimony

Vanessa Whitmore, a raw material buyer for Tremco Incorporated's roofing division testified at trial about the use of GQA in Tremco's roofing felts and roofing adhesives. Tremco began using GQA to replace asbestos as a thickener. Tr. 4 at 152:17-152:18. Ms. Whitmore stated that Tremco uses four thickeners for its roofing products — Bentone-SD2, Tixogel-VZ, Minugel-AR and Attagel 36. Tr. 4 at 158:21-158:24. Bentone-SD2 and Tixogel-VZ are interchangeable with each other for Tremco's purposes, as are Minugel-AR and Attagel 36. Tr. 4 at 159:1; 159:24-159:25. GQA is used in Tremco's solvent-based adhesives, while Tixogel-VZ and Bentone-SD2 are used in solvent-free adhesives. Tr. 4 at 161:16-161:18. Ms. Whitmore testified that Tremco would not substitute for GQA in response to a 5 percent price increase because it would not be worth the trouble of reformulation. Tr. 4 at 157:10-157:21.^[17]

Carl Defani, of Gardner Asphalt, a roof coating manufacturer, testified by Civil Investigative Demand deposition, taken June 6, 1995. Gardner uses GQA in its roofing cements. Defani CID Dep. at 9:21. Gardner has tested many substitutes, but has never marketed a product including these substitutes. Defani CID Dep. at 15:25-16:3. Mr. Defani believed that bentones would make a satisfactory product but not at a competitive price. Defani CID Dep. at 18:12-18:13. Sepiolite has also been tested with unsatisfactory results. Defani CID Dep. at 18:16-18:17. At the present time, Mr. Defani is not aware of any substitutes for GQA.

Robert A. Bair, of ALCO-NVC, a roof coating manufacturer, testified by deposition, taken July 14, 1995. Mr. Bair testified that there are substitutes for GQA in asphalt roof coatings, Bair Dep. at 9:2-9:5, stating, "We have looked at all different types of raw materials. Could we substitute a bentonite clay or a ball clay? Yes. It really wouldn't make the best product though." Bair Dep. at 9:10-9:13. Mr. Bair stated that ALCO-NVC has formulas using no GQA which it could produce presently. Bair Dep. at 11:22-11:24. GQA is a minor cost in ALCO-NVC's final product, so a 5 to 10 percent increase would likely not initiate a substitution of GQA in its roof coatings.

Jack McClellan, of DeWitt Products, manufacturers of asphalt roof coatings, testified by Civil Investigative Demand deposition, taken May 23, 1995. When asked what DeWitt could replace GQA with, Mr. McClellan responded:

Attapulgus — there are different channels that we can take. There are Bentonite clays that we can look at and possibly replace, if they were cost effective. There is Sepiolite, and I'm not sure if that's a clay or not. I'm not scientifically well endowed to know whether Sepiolite is a clay or not, but it is an ingredient that has properties which can swell and make a gel system in an asphalt product. There are precipitated silicas, fume silicas, and there might be some other things that we have investigated but I just can't recall at this time, that you can use for gelling purposes. I take that back I do remember and recall that there are a couple of liquid ingredients that we probably can look at, that will swell upon heat activation, that may work.

McClellan CID Dep. at 18:11-19:2. Mr. McClellan explained that after a decision was made to stop using asbestos, various substitute products were tested, and that sepiolite was the second choice behind GQA. McClellan CID Dep. at 25:19-25:20. Sepiolite performed identically to GQA in DeWitt's tests, however, due to higher shipping costs for sepiolite, GQA was chosen. McClellan CID Dep. at 25:22-26:8. Mr. McClellan stated that if sepiolite were available, a replacement product could be produced "within a couple of weeks." McClellan CID Dep. at 27:25. Mr. McClellan added that he would look to reformulation if the price of his finished product went up by 10 percent, however, it would probably take an increase of about 25 percent in the cost GQA to initiate a reformulation effort. When asked about the proposed transaction between Engelhard and Floridin, Mr. McClellan answered:

My opinion is that I think competition is good, and the limit on competition is bad. But what is competition? Competition could be Attapuigus clay. The competition is also asbestos. The competition is also Bentonite clay. Competition is sepiolite, fume silica, precipitated silica. Those are all competitive things that can make decisions in the marketplace.

McClellan CID Dep. at 86:9-86:16.

b. Customer Calls by Dr. Bodisch

The Court could not recognize which, if any, of Dr. Bodisch's customer calls related to the asphalt roof coatings industry.

Once again, the record contains compelling evidence that competitive, substitute products are presently available in the marketplace. While the evidence is by no means conclusive, it certainly suggests that a fuller inquiry was necessary to clearly define the market. However, because substitution for the most part would occur only in response to a price increase above 10 percent, no such inquiry was made.^[18]

7. Molecular Sieves

a. Testimony

Allen R. Harrison testified at trial about UOP's use of GQA in molecular sieves. Mr. Harrison stated that GQA is used in about 70 to 75 percent of UOP's molecular sieves. Tr. 2 at 56:21. He added that in sieves which use GQA, no product could substitute for the GQA. Tr. 2 at 58:13. UOP has purchased all of its GQA from Floridin for over a decade. Tr. 2 at 50:15-50:22. Prior to switching to Floridin, UOP had purchased exclusively from Engelhard. Tr. 2 at 50:22. Mr. Harrison testified that UOP has tested other products in the past, such as GQA from Milwhite, Tr. 2 at 53:17-53:25, Chinese GQA, Tr. 2 at 55:8-55:13, and sepiolite from Tolsa, Tr. 2 at 57:23-58:5, but all have failed in testing. UOP uses Kaolin in sieves which do not use GQA, Tr. 2 at 56:23, but Mr. Harrison believed that Kaolin could never be a substitute for GQA in sieves that use GQA. Tr. 2 at 57:3.

Roger Allen Nash, also of UOP, testified at trial and essentially affirmed the positions taken by Mr. Harrison.

b. Customer Calls by Dr. Bodisch

The only GQA customer in the molecular sieve industry called by Dr. Bodisch was Roger Nash. As stated, Mr. Nash's trial testimony reflected that of Mr. Harrison. The customer call report provides no additional information.

Thus, the evidence of record, though limited to the experiences of one company, indicates that there are no competitive substitutes for GQA in the molecular sieve market.^[19]

8. Other Applications

The applications discussed above are the most common applications for GQA. However, other applications exist, and the Plaintiff took a Civil Investigative Demand deposition on June 1, 1995 of a customer using GQA for such an application. [This deponent] is the Purchasing Manager at [a chemical company]. [Purchasing Manager] CID Dep. at 7:4. [This chemical company] uses GQA in two sound-deadening applications. One is a [confidential] spray and the other is a [confidential] spray, each meant to reduce the transfer of noise through the particular [surface]. [Purchasing Manager] CID Dep. at 19:21-20:5. [The Purchasing Manager] testified that there are four alternatives for GQA in [his company's] applications — CabO-Sil, Bentonite, Claytone, and a fourth product from Milwhite (most likely another brand name of GQA) — and that each can be a complete substitute. [Purchasing Manager] CID Dep. at 20:22-22:14. Indeed, [the Purchasing Manager] believed that reformulation could be done within one day, one week at the most. [Purchasing Manager] CID Dep. at 16:10-13. [The Purchasing Manager] further stated that [his company] would likely make a switch if the price of GQA rose by 1 percent or perhaps even less. [Purchasing Manager] CID Dep. at 38:2-38:5.

[The Director of Purchasing for a pesticide manufacturer], testified by deposition on July 11, 1995. [His company] makes and markets pesticides, [Purchasing Director] Dep. at 7:6-7:9, and uses GQA to provide viscosity and suspension in its products. [Purchasing Director] Dep. at 8:10-8:13. [This company] purchases exclusively from Engelhard, and [the Purchasing Director] was not even familiar with Floridin or Milwhite. [Purchasing Director] Dep. at 20:15-20:24. [The Purchasing Director] stated that there are no substitutes for GQA in [his company's] pesticide formulations, except by selling the active ingredients in separate containers for mixture by the end users. By doing this, the need for Suspension is removed. [Purchasing Director] Dep. at 23:23-24:21. However, the market preference is for premixed products. [Purchasing Director] Dep. at 25:13-25:4. Another reason that [this company] would likely not look for alternatives to GQA is that each new formulation must be approved by the EPA. [Purchasing Director] Dep. at 26:1-26:12. Thus, [this company] would not reformulate if the price of GQA rose by 5 to 10 percent. However, neither would [this company] switch suppliers of GQA at those price increase levels. [Purchasing Director] Dep. at 34:9-34:23.

These two depositions do not provide tremendous insight into the GQA market. However, the testimony reflects the trend of all the evidence of record, to wit: that an examination of the GQA market reveals some ability on the part of GQA consumers to switch to substitute products in response to an increase: in the price of GQA. So, what are the exact parameters of this marketplace power, or in other words, what is the cross-elasticity of demand between GQA and substitute products? This question cannot be answered on the basis of the evidence presented in this case, because the Plaintiff chose an arbitrary, and non-probative, cutoff point beyond which it considered any cross-elasticity of demand be too inelastic to suggest a relevant market.

C. Competition from Sepiolite Was Unreasonably Discounted

In 1987, Engelhard sought to acquire Floridin. At that time, the Department of Justice indicated that it would challenge the proposed acquisition, contending that the relevant market which the transaction would have affected was the market for hormite clays, which includes both GQA and sepiolite. DX 1205. In the present case, the Plaintiff still recognizes that "some [GQA] customers, primarily located west of the Rocky Mountains, have product requirements that enable them to use sepiolite instead of gel clay." GX 283 at ¶ 13. However, when analyzing the current proposed transaction, the Plaintiff determined that "[t]he large majority of gel clay customers, for performance reasons and because of reformulation and transportation costs, would not switch to sepiolite in response to an increase in the price of gel clay. Therefore ... sepiolite is properly excluded for the gel clay market." Id. As previously shown, many GQA customers questioned about increased GQA prices referenced sepiolite as a potential substitute product. In light of that evidence as well as the following evidence of direct competition between sepiolite and GQA, the Court finds that, at the very least, a further investigation of sepiolite as a part of the relevant market was required in this case.

There are two major producers of sepiolite in the world. In the United States, the sole producer of sepiolite is IMV Corporation, a subsidiary of Defendant U.S. Silica. The other major source of sepiolite Tolsa S.A., is in Spain. Tosla's sepiolite is imported into the United States and distributed exclusively by the J.M. Huber Corporation. Through J.M. Huber, Tolsa has introduced its sepiolite into the United States market, under the trade name "Pangel," as a direct substitute for GQA.

In a manual entitled "Sepiolite: Choice of the Future," J.M. Huber announced that "[s]epiolite is an excellent replacement for attapulgite, especially as a rheological control agent and as a bodying agent." DX 1036. See also, DX 1046 (same language in an advertisement); DX 1038 at EC047-00165 (internal document from Tolsa, showing applications for sepiolite in many of GQA's major end-use applications); DX 1187a (IMV product bulletin stating that industrial grade sepiolite "may replace attapulgite in drywall tape joint cements, asphalt roof coatings, caulks, sealants, etc."); DX 1187b (IMV product bulletin stating that fine grind sepiolites can be used in latex paints and roof coatings, among other things); DX 1187d (IMV product bulletin stating that Sepiogel A is a highly effective gelling and suspending agent in a wide range of fertilizer formulations); DX 1187e (IMV product bulletin stating that Sepiotone can be used in replacing asbestos in asphalt roof coatings and automotive undercoatings, among other things); DX 1252 (IMV brochure stating that sepiolite can be used in fresh and salt water drilling fluids, suspension fertilizers, feed suspensions, and asphalt roof coatings).

In an internal memorandum entitled "Pangel Sepiolite v. Attagel Thickeners," Engelhard recognized the competition from Huber/Tolsa, stating, "J.M. Huber Corporation has become the exclusive North American distributor of Tolsa's sepiolite from Spain. They are targeting Engelhard's specialty ATTAGEL thickener markets." DX 1037. The Plaintiff has been quick to point out, however, that the same memo states:

We compete with Tolsa in Europe and, to the best of our knowledge, have never lost an ATTAGEL customer to PANG EL. Their products have not successfully competed with ATTAGEL products for the following reasons.

Id. (emphasis in original). The memo attributes Attagel's success to performance advantages, lack of motivation to change, and insignificant cost savings due to the small percentage that either Attagel or Pangel would amount to in an end-use formulation. Id. The memo also discusses sepiolite produced in the United States stating, "Occasionally, we have seen attempts by IMV (Industrial Minerals Venture) of Nevada to try and displace ATTAGEL thickeners with their sepiolite products, but they have always been unsuccessful."^[20]Id.

However, the fact that competition has only produced limited success for a competitor, at current market prices, does not mean that competition does not exist nor that competition will not grow if market prices change. Increases in the cost of GQA are exactly the type of "motivation" that consumers need to search for substitute products, and each increase will proportionally raise the level of cost savings that sepiolite usage can bring to a customer.

Certainly, Engelhard did not perceive sepiolite as a non-competitor when it set forth an "action plan" stating:

Distributors should continue to pay close attention to this new threat to our specialty ATTAGEL thickeners. The danger will be at our truckload accounts. Please report any sepiolite activities to your local sales representatives immediately.

Id. See also DX 1076 at Bates # EC064-00279 (wherein Engelhard lists as competitors: Floridin, Milwhite, and Huber/Tolsa); DX 1127 (also 1077) at 41 (Engelhard strategic plan, 1994-1998, wherein defending "Attagel specialty business against new Huber/Tolsa entry in the U.S." is listed as a "critical success factor").

As a sister corporation to IMV, Floridin's attitude towards the success of sepiolite was different than Engelhard's. In fact, Floridin attempted to convert some of its West Coast GQA customers to sepiolite in order to share with these customers the reduced transportation costs. Goodell CID Dep. at 161:16-162:17. However, Floridin had little success in this endeavor due to performance problems with sepiolite. Goodell CID Dep. at 162:16-165:5. Obviously, such testimony indicates the inability of sepiolite to compete with GQA.

However, Engelhard has witnessed competition from sepiolite. After a visit to Ashland Chemical Company, Engelhard issued a memo stating:

Ashland is a large manufacturer of foundry mold coatings for the automotive and other industries. They are currently using some Attagel 50 in their coatings. They recently started to test products called Pangel HV and Sepiogel UF. They are finding better gelling and anti-settling performance with these products, especially the Pangel HV.

DX 1064. This is not inconsistent with Dr. Bodisch's testimony that he spoke with Ashland and was informed that use of the Tolsa sepiolite was still in the testing phase. See Tr. 6 at 89:18-89:25, and it indicates head-to-head competition between the products.

While Engelhard appears to have considered Huber/Tolsa's attempts to take Attagel business as unsuccessful in 1995 see DX 1082 at 24 ("Milwhite is entering the specialty Attagel market with some modest success while Huber has failed with the Tolsa products"), it simultaneously acknowledged the loss of business to sepiolite from Nevada. See id. at 24, 26 ("IMV is taking market share in the fertilizer and animal feed markets with low pricing," and "IMV recaptured several large animal feed accounts from Engelhard this year at very low pricing"). See also DX 1249 at EC088-00032 (notice to Engelhard from a distributor that United Coatings, a paint manufacturer, had "dropped the Attagel 40 for Sepiogel UF").

Another telling indication of Engelhard's view of sepiolite as a not-in-kind competitor is the fact that Engelhard did not seek to purchase IMV as a part of the currently proposed transaction with Floridin and ITC. Explaining why Engelhard should not pursue a bid for IMV as a part of a proposed Floridin acquisition, an inter-office memorandum stated, "The IMV products ... have been carefully reviewed, and believed to be marginally competitive with attapulgite-based products, and therefore of little value, offering Engelhard no operating or market synergies." DX 1117 at 2.

Once again, the record indicates historical and on-going competition between sepiolite and attapulgite. While the Plaintiff did not consider the level of competition significant for purposes of the relevant market issue, the Court finds that, particularly in light of the history of head-to-head competition between the products, this market should have been more carefully studied. The record simply does not sufficiently illuminate the Court as to the real size of the sepiolite market, or its potential for growth in response to a small but significant GQA price increase above 5 to 10 percent.

D. Dr. Warren-Boulton's Relevant Market Analysis Was Not Reliable

Although the Plaintiff had at the beginning of this action specifically stated that Dr. Frederick Warren-Boulton would not testify on the issue of relevant market, the Court allowed Plaintiff to present Dr. Warren-Boulton as a rebuttal witness on this issue. However, Dr. Warren-Boulton's analysis on product market suffered from two major defects. First, Dr. Warren-Boulton, for the most part, simply relied on Dr. Bodisch's findings under the Merger Guidelines test in determining the relevant market. See Tr. 11 at 71:19-71:22; 73:6-73:13. In fact, while Dr. Warren-Boulton testified that he is capable of performing cross-elasticity of demand studies, he did not do so in this case because, in his opinion, the failure of any customer to respond to a 5 to 10 percent increase in the price of GQA answered the cross-elasticity question. Tr. 11 at 74:11-75:2; 76:22-76:25. In fact, Dr. Warren-Boulton's devotion to the 5%-10% test was so strong, that when presented with the Eleventh Circuit's definition of a relevant market in U.S. Anchor, supra, 7 F.3d at 995, he responded, "It is simply not the way I would characterize the merger guidelines." Tr. 11 at 99:12-99:13.

Dr. Warren-Boulton's relative price comparisons also fail to give a clear definition to the relevant market. To begin with, the analyses suffer from an admitted lack of data points. Tr. 11 at 79:5-79:14. Further, Dr. Warren-Boulton's analysis of relative sales prices among products was based upon total sales prices for each product studied in all applications, regardless of whether the sales were in competitive uses to GQA. See Tr. 11 at 114:7-115:3. As discussed throughout this opinion, the market in which gel clay competes is one in which special end-use requirements generally necessitate special formulations, and in some of these formulations one competitive product may be useful in place of GQA whereas in another formulation the same substitute thickener would completely ruin the end-use product.^[21] Thus, competition in this type of market is not necessarily defined by across-the-board product to product competition, but rather by the competition that a combination of products may provide in various applications. The ultimate question must ask whether within each of the panoply of GQA applications there are reasonably interchangeable alternatives with a sufficient cross-elasticity of demand, so that on the whole, a hypothetical GQA monopolist would be precluded from exercising unlawful market power. Nevertheless, Dr. Warren-Boulton analyzed the relative prices of substitute products in both competitive and noncompetitive applications, explaining, "[W]hy that would be appropriate, given the inquiry, is it is my understanding that the, [sic] and the argument made by the economists for the defendants is that all CMC is in the market, and, therefore, the average price for all CMC would appear to be the relevant test." Tr. 11 at 114:23-115:3. The problem is that it is not the Defendant's burden to establish the relevant market. Although Dr. Warren-Boulton's test may handily rebut the proposition that competition exists among all thickeners in all uses, a proposition that is wholly unsupported by the record, it does little to reveal where competition among GQA and other thickeners truly does exist.

In the final analysis, Dr. Warren-Boulton's testimony does little more than to retread the 5%-10% ground that Dr. Bodish plowed during the Plaintiff's case-in-chief. Again, the Court is moved no closer to a clear view of competition in the thickeners market.

Conclusion

The Merger Guidelines 5%-10% test is an inaccurate barometer of cross-elasticity of demand as to the facts presented in this case. Further, Plaintiff's steadfast application of the test as the foundation of it market definition analysis resulted in a pervasive failure to acknowledge relevant information throughout the evidence of record. For the reasons discussed herein, the Court holds that the entirety of the evidence does not weigh in favor of a finding that the relevant product market is gel quality attapulgite. Indeed, the evidence is insufficient to make any reasonable judgment as to what the relevant product market is. Thus, Plaintiff has failed to meet its ultimate burden of persuasion as to an essential element of its case, see University Health, supra, 938 F.2d at 1217-18; Kaiser Aluminum, supra 652 F.2d at 1340, and as Plaintiff's expert Dr. Bodisch testified, "If the market is incorrectly defined, the market shares would have no meaning." Tr. 6 at 71:8-71:9. Accordingly, Plaintiff's request for a permanent injunction should be, and hereby is, DENIED.

NOTES

[1] This order amends the Court's final decision in this action, entered under seal on March 10, 1997, and serves as the public version of such decision. A substantial volume of confidential business information was introduced during the trial of this action. By agreement of the parties and the Court, some of this evidence was filed under seal. Thus, the Court's final decision in this action was also entered under seal on March 10, 1997, in order to allow the parties an opportunity to review the decision and inform the Court if any confidential information had been included in the text of the opinion. After conferring on this issue, the parties informed the Court that the following portions of the Court's original decision potentially jeopardized the confidentiality of proprietary business information of nonparties to this action: page 9, line 19; page 38, lines 3-15; page 14, lines 8-10; page 37, lines 12-20; page 38 lines 1-2; and footnote 9, lines 9-10. In this amended order, the Court has modified those portions so as to delete specific references to proprietary information without changing the substance or meaning of the text of the original order. To the extent that this amended order includes changes from the Court's original order, the same have been denoted by brackets and bold lettering at: page 9, line 17; page 14, lines 4-6; page 36, line 21; page 37, lines 1-20; page 38, lines 1-7; and footnote 11, lines 11-15. The original decision of the Court, entered March 10, 1997, shall remain under seal and may only be viewed with prior approval of the Court.

[2] References to the trial transcript shall be denoted as follows: Tr. (Volume Number) at (Page Number): (Line Number).

[3] This citation is an amendment from the Court's original order. The Court mistakenly miscited this fact to Tr. 7 at 135:5-135:14; Goodell CID Dep. at 100:19-101:5.

[4] Floridin's assets also include International Minerals Venture ("IMV"), a producer of sepiolite and organoclays in Amargosa Valley, Nevada. This asset is not a part of the challenged transaction.

[5] Both in 1987 and today, the Plaintiff contends that "there are no antitrust issues with respect to sorbent clay," as there is vigorous competition between sorbent clay and non-attapulgite products. PX 283 ¶¶ 10, 45.

[6] Some cases cited in this opinion involve the Sherman Act. The standards applied in Sherman Act cases for defining the relevant product market are the same as in cases involving Section 7 of the Clayton Act. United States v. Grinnell Corp., 384 U.S. 563, 573, 86 S. Ct. 1698, 1705, 16 L. Ed. 2d 778 (1966).

[7] Due to the volume of both the testimony and documentary evidence put into the record during nearly three weeks of trial in this case, the Court permitted the parties to submit post-trial briefs in addition to oral closing arguments.

[8] Dr. Bodisch did ask some customers if they would switch at a higher percentage price increase. Dr. Bodisch's finding in this regard shall be discussed below. However, as a great deal of Dr. Bodisch's conclusions about the relevant product market arose in large part from the 5%-10% inquiry, the Court believes that such findings must be singularly analyzed.

[9] Although many of the potential substitute products are much more expensive than GQA, they are more efficient in terms of use, so "their cost effectiveness is equal or many times better than attapulgite." Tr. 7 at 106:15-106:24.

[10] Dr. Bodisch testified that his expert report was based on many sources of information including approximately 25 to 30 interviews he personally conducted with GQA customers in a variety of industrial markets. Dr. Bodisch expert report, however, does not specifically reference all of these interviews. As a result, the Court has endeavored to evaluate as many of these interviews as possible based upon Dr. Bodisch's trial testimony as well as his handwritten notes from the interviews. As a result, it is possible that certain of these interviews will not be touched upon.

[11] Plaintiff has objected to the admission of the Kline Survey, DX 1130, into evidence. This survey, conducted by a professional research company on behalf of the Defendants, was simply a series of interviews with GQA customers and distributors, not terribly unlike Dr. Bodisch's customer calls. While the Court finds that the Kline Survey passes evidentiary muster for purposes of admissibility, it recognizes that there are many shortcomings in this study that raise significant questions about its reliability. For that reason, the Court has opted not to rely on the Kline Survey in a substantive way, but rather to simply reference it by footnotes, when appropriate, to show that the results fairly well mirrored the findings of Dr. Bodisch and witnesses' testimony.

In that regard, the Kline Survey customer call reports showed several customers that indicated that switching, or initiation of a search for substitute products, would occur at the 10 to 20 percent price increase level. See DX 1130 at [a suspension fertilizer company] (10%-15%); [another suspension fertilizer company] (10%); [a third suspension fertilizer company] (10%); [a tape joint compound company] (10%). Another tape joint manufacturer interviewed for the Kline Survey indicated that switching would not occur without at least a 25% price increase for GQA. See DX 1130 [another tape joint compound company].

[12] Kline Survey customer call responses in the suspension fertilizer industry were similar to those found by Dr. Bodisch. A sepiolite supplier indicated that sepiolite could substitute for GQA in suspension fertilizers and had better colloidal properties in this respect. DX 1130 at IMV (interviewee reported that PMS company purchased sepiolite for use in suspension fertilizer, but Dr. Bodisch testified that PMS had discontinued use due to poor performance, notwithstanding 20 percent lower prices than GQA. Tr. 6 at 27:14-28:8). Other interviewees discussed the use of bentonite, sepiolite, polymers, cellulosic gums and other clays as potential substitutes to GQA. See DX 1130 at United Suppliers; Nutra-Flo; Terra Chemical International; and United Agra. On the other hand, several interviewees were unaware of substitute products. See DX 1130 at MFA Exchange; Noland Feed & Fertilizer; and Old Mill Grain & Seed ("[GOA] is the only thing you can use").

[13] Again, the Kline Survey customer calls reflected the findings of Dr. Bodisch's interviews. Of two interviewees in this industrial market, one indicated that xanthum and guar gums, as well as bentonite were alternatives to GQA, see DX 1130 at PM Ag Products, and the other stated that his company, which currently relies upon GQA, could change to other types of thickeners relatively easily. See DX 1130 at Golden Sun Feeds.

[14] Again, the Kline Survey interviews contained responses similar to those obtained by Dr. Bodisch. One customer indicated that many substitutes are available and that his company chooses raw materials, for example GQA, sepiolite or bentonite, depending on the geographic location of the manufacturing facility. DX 1130 at National Gypsum. Three other customers similarly indicated a variety of thickeners that could substitute for GQA, with one having already abandoned the use of GQA in favor of hydroxymethyl and hydroxypropyl cellulose. See DX 1130 at Hamilton Materials; Domtar Gypsum; and Tool World.

[15] Again the Kline Survey customer call reports were generally reflective of the testimony and interviews taken by Dr. Bodisch. One customer stated that cellulosics and associative thickeners are used in water-based paints, but that there were no reasonable alternatives to GQA in latex paints. DX 1130 at Union Carbide Corporation. The others interviewed suggested synthetic, associative and cellulosic thickeners, as well as bentonite, as possible alternatives to GQA. See DX 1130 at Benjamin Moore & Co. (2); Carolina Coatings; and Passono Paint.

[16] In the case of drilling fluids, the Kline Survey only reached two raw materials suppliers, neither of which is involved in actual commercial drilling ventures. A sepiolite supplier indicated that sepiolite is a substitute for GQA in salt water drilling fluids. DX 1130 at IMV (This company is of course a subsidiary of Floridin's parent corporation, U.S. Silica). The other interviewee, a bentonite supplier, discussed some alternative products but the call report leaves unclear whether these products could be used in salt water drilling fluids or only as a substitute for treated GQA in oil-based drilling fluids. The Court does not find these interviews particularly useful, but again they do not contradict the evidence adduced from testimony and Dr. Bodisch's interviews.

[17] It should be noted that in a declaration, given April 20, 1995, Ms. Whitmore stated that "if the price of gel clay were to rise by more than 10 percent, we would attempt to either reformulate, which is timely, costly, and not always possible, or find another source of gel clay." GX 264 at ¶ 5.

[18] The Kline Survey included customer calls to many GQA customers in the roof coating industry. The Court does not believe that it is necessary to address each customer call individually, as the responses were generally similar to each other. For the most part, the customers indicated that GQA was the industry standard for solvent-based coatings and that bentonite is used in water-based coating. See e.g., DX 1130 at Fields & ATCO. Many customers noted, however, that water based coatings had limited applications during the winter in cold weather areas of the country due to "freeze-thaw stability" problems. See e.g., DX 1130 at ALCO-NAV. One customer stated that the unavailability of GQA "would be a big problem for the industry" DX 1130 at Gardner Asphalt. Another said that bentonites and different types of clay were alternatives to GQA, but that the final product must still compete with asbestos-based roof coatings and thus these substitutes are not cost competitive. DX 1130 at Gibson-Homans. Most customers seemed to reflect the same position, indicating that alternatives would be found if GQA became unavailable but that economic competitiveness was the major problem with substitute products. See e.g., DX, 11301 at Gulf States Asphalt; Karnak Company; and Monsey Products.

Several customers, however, did indicate that solvent-based asphalt coating could be competitively produced without GQA. See e.g., DX 1130 at FBC Chemical. One customer stated that his company currently produces solvent-based, asbestos free roof coatings without the use of GQA. DX 1130 at Kol-Tar. Another customer indicated that GQA is the product of choice, but that if GQA became too expensive, he has a alternative formula "on the shelf." DX 1130 at Roof Top Partners. He further stated that the price was about the same as the GQA formula, and that any reduction in quality could be overcome if put to the test. Id.

Again, the survey interviews mirror the conclusions drawn from the trial and deposition testimony relating to this industrial market, to wit: reliance on GQA by some customers, but the indication that substitutes are available if customers needed them.

[19] The Kline Survey did not consider molecular sieves within the scope of its inquiry about GQA. See DX 1130, Draft Report at p. 1.

[20] In the case of suspension fertilizer customer PMS, Engelhard did not even attempt to retain the customer through price competition with sepiolite, relying instead on the judgment that PMS would return to GQA for quality reasons. See PX 68 at EC121-00056 ("let [IMV] have this business and see if PMS can tolerate the poor performance of Sepiolite versus Attapulgite"). At trial, Dr. Bodisch testified that PMS had, in fact, switched back to GQA due to poor performance of sepiolite, notwithstanding the 20 percent higher cost of GQA. Tr. 6 at 27:14-28:8.

[21] In this type of market, a simple hypothetical shows the flaw of Dr. Warren-Boulton's relative price analysis. Suppose, for example, that Product A competes with only Product 1 in Application X, and Product A competes only with Product 2 in Application Y, and Product A competes only with Product 3 in Application Z, and Products 1, 2, and 3 all compete vigorously against each other in a variety of other applications in which Product A is a complete non-competitor. If the "non-Product A" applications comprise a much larger portion of Products 1, 2, and 3's overall sales than do the "Product A" applications, a relative price analysis of each product in all of their applications could show a high correlation between Products 1, 2, and 3, and a low correlation between those products and Product A, notwithstanding vigorous price competition between Product A and the other products in the applications where they do compete.