Federal Trade Commission v. Arch Coal, Inc. , 329 F. Supp. 2d 109 ( 2004 )

329 F. Supp. 2d 109 (2004)

FEDERAL TRADE COMMISSION, Plaintiff,
v.
ARCH COAL, INC., et al., Defendants.
State of Missouri, et al., Plaintiffs,
v.
Arch Coal, Inc., et al., Defendants.

Nos. CIV.A.04-0534 JDB, CIV.A.9409535 JDB.

United States District Court, District of Columbia.

August 16, 2004.

Rhett Rudolph Krulla, Federal Trade Commission, Bureau of Competition, Marc I. Alvarez, Federal Trade Commission, Washington, DC, Counsel for plaintiff Federal Trade Commission.

Anne E. Schneider, Office of the Attorney General, State of Missouri, Jefferson City, MO, Counsel for plaintiff States and State of Missouri.

Bradford J. Phelps, Office of the Attorney General, Little Rock, AR, Counsel for plaintiff State of Arkansas.

Karl R. Hansen, Office of the Kansas Attorney General, Topeka, KS, Counsel for plaintiff State of Kansas.

Robert W. Pratt, Illinois Office of the Attorney General, Chicago, IL, Counsel for plaintiff State of Illinois.

Thomas J. Miller, Iowa Department of Justice, Layne M. Lindebak, Iowa Department of Justice, Des Moines, IA, Counsel for plaintiff State of Iowa.

Rebecca Fisher, Office of the Attorney General of Texas, Austin, TX, Counsel for plaintiff State of Texas.

Stephen Weissman, Howrey Simon Arnold & White, LLP, William Bradford Reynolds, Howrey Simon Arnold & White, LLP, Washington, DC, Counsel for defendant Arch Coal, Inc.

Charles Edward Bachman, O'Melveny & Myers, LLP, New York, NY, Richard G. Parker, O'Melveny & Myers LLP, Washington, DC, Counsel for defendants New Vulcan Coal Holdings, LLC and Triton Coal Company, LLC.

Kenneth George Starling, Piper Rudnick LLP, Washington, DC, Counsel for movant Peter Kiewit Sons, Inc.

MEMORANDUM OPINION

BATES, District Judge.

Coal is the primary fuel that produces electric power for residential and business consumers across the United States. It is mined in various regions across the country, in either surface or underground mining operations, after which the coal is transported by rail, truck or barge to electrical generating plants. One-third of the coal produced annually in the United States — over 360 million tons — is produced from large-scale surface mining operations in the Southern Powder River Basin ("SPRB") region of Wyoming. Seven companies operate fourteen mines in the SPRB at this time.

In May of 2003, Arch Coal, Inc. ("Arch"), the owner and operator of two SPRB mines (Black Thunder and Coal Creek) as well as other mining operations across the United States, and New Vulcan Coal Holdings, LLC ("New Vulcan"), the owner of two SPRB mines (North Rochelle and Buckskin), which it operates through its subsidiary Triton Coal Company, LLC ("Triton"), entered into a merger and purchase agreement under which Arch would acquire Triton and its two SPRB mines. Hart-Scott-Rodino pre-merger notification was provided to the Federal Trade Commission ("FTC"), which in August 2003 requested additional information from Arch and New Vulcan. Arch subsequently informed the FTC that it intended to divest one of the acquired mines (Buckskin) to Peter Kiewit Sons, Inc. ("Kiewit"), a large company with some mining interests outside the SPRB, and in January 2004 a firm asset purchase agreement was entered by Arch and Kiewit.

After a nine-month review, the FTC voted on March 30, 2004, to commence this action seeking to enjoin the proposed acquisition under Section 13(b) of the Federal Trade Commission Act, 15 U.S.C. § 53(b), as violative of Section 7 of the Clayton Act, 15 U.S.C. § 18. The FTC seeks to preliminarily enjoin Arch's proposed acquisition of Triton until an administrative FTC proceeding challenging the transaction under Section 7 can be completed. A parallel suit was also filed on April 1, 2004, by the States of Missouri, Arkansas, Kansas, Illinois, Iowa, and Texas (the "States") seeking both preliminary and permanent injunctive relief. The two actions were consolidated by this Court on April 21, 2004.^[1]

The Court held a two-week trial commencing on June 28, 2004, during which it heard from more than twenty witnesses and received hundreds of exhibits, many of them lengthy, including deposition and affidavit testimony of several additional witnesses. The parties have submitted well over 700 pages of post-hearing proposed findings of fact and briefs. The Court has reviewed that substantial body of evidence and argument in assessing the FTC's challenge to the proposed acquisition of the North Rochelle and Buckskin mines by Arch, and the simultaneous transfer of Buckskin to Kiewit, and the probable effect of those transactions on competition in the SPRB.^[2] The case is complex, and represents an attempt by the FTC to enjoin transactions that do not reduce the number of competitors and only modestly increase the concentration in what has been a very competitive market. Moreover, the case rests on a novel FTC theory of likely future "tacit coordination" among competitors to restrict production, as opposed to direct coordination of prices. In the end, the Court concludes that the FTC and the States have not met their burden under Section 7 of the Clayton Act and (for the FTC) Section 13(b) of the FTC Act to show a likelihood that the challenged transactions will substantially lessen competition in the SPRB. The requested preliminary injunctive relief will therefore be denied.^[3]

I. APPLICABLE LAW

Section 7 of the Clayton Act, 15 U.S.C. § 18, prohibits a merger between two companies "where in any line of commerce or in any activity affecting commerce in any section of the country, the effect of such acquisition ... may be substantially to lessen competition, or tend to create a monopoly." The Supreme Court has observed that Section 7 "deals in probabilities, not certainties." United States v. General Dynamics Corp., 415 U.S. 486, 505, 94 S. Ct. 1186, 39 L. Ed. 2d 530 (1974); see also Brown Shoe Co. v. United States, 370 U.S. 294, 323, 82 S. Ct. 1502, 8 L. Ed. 2d 510 (1962); United States v. El Paso Natural Gas Co., 376 U.S. 651, 658, 84 S. Ct. 1044, 12 L. Ed. 2d 12 (1964). As defendants also correctly stress, however, "Section 7 deals in probabilities not ephemeral possibilities." FTC v. Tenet Health Care Corp., 186 F.3d 1045, 1051 (8th Cir.1999). To warrant injunctive relief under the Clayton Act, the challenged acquisition must be likely substantially to lessen competition. Although certainty is not required, Section 7 does demand that a plaintiff demonstrate that the substantial lessening of competition will be "sufficiently probable and imminent" to warrant relief. United States v. Marine Bancorporation, 418 U.S. 602, 618, 94 S. Ct. 2856, 41 L. Ed. 2d 978 (1974); see United States v. Baker Hughes, Inc., 908 F.2d 981, 984 (D.C.Cir.1990) ("Section 7 involves probabilities, not certainties or possibilities.").

Congress has empowered the Federal Trade Commission to seek preliminary injunctive relief preventing parties from consummating a merger until the FTC has had an opportunity to adjudicate the merger's legality. Section 13(b) of the FTC Act "provides for the grant of a preliminary injunction where such action would be in the public interest — as determined by a weighing of the equities and a consideration of the Commission's likelihood of success on the merits." FTC v. H.J. Heinz Co., 246 F.3d 708, 714 (D.C.Cir.2001); see 15 U.S.C. § 53(b). The FTC "is not required to prove, nor is the court required to find, that the proposed merger would in fact violate Section 7 of the Clayton Act." FTC v. Staples, Inc., 970 F. Supp. 1066, 1070 (D.D.C.1997) (citations omitted); see FTC v. University Health, Inc., 938 F.2d 1206, 1218 (11th Cir.1991) (court's task is to make preliminary assessment of impact on competition). Rather, the FTC "need only show that there is a ``reasonable probability' that the Acquisition may substantially lessen competition." Staples, 970 F.Supp. at 1072. Proof of actual anticompetitive effects is not required; instead, the FTC must show "an appreciable danger" of future coordinated interaction based on a "predictive judgment." Heinz, 246 F.3d at 719 (quoting Hosp. Corp. of Am. v. FTC, 807 F.2d 1381, 1389 (7th Cir.1986)). While proof of prior cooperative behavior is relevant, it is not a necessary element of likely future coordination in violation of Section 7. The FTC, then, must raise "questions going to the merits so serious, substantial, difficult and doubtful as to make them fair ground for thorough investigation, study, deliberation and determination by the FTC in the first instance and ultimately by the Court of Appeals." Id. at 714-15 (citations omitted).

Given the stakes, the FTC's burden is not insubstantial, and "[a] showing of a fair or tenable chance of success on the merits will not suffice for injunctive relief." Tenet Health Care, 186 F.3d at 1051; see Fruehauf Corp. v. FTC, 603 F.2d 345, 351 (2d Cir.1979) ("mere possibility" will not justify preliminary injunction). Because the public interest in effective enforcement of the antitrust laws is of primary importance, a showing of likely success on the merits will presumptively warrant an injunction. See Heinz, 246 F.3d at 726; University Health, 938 F.2d at 1225; Staples, 970 F.Supp. at 1091. Conversely, absent a likelihood of success on the merits, equities alone will not justify an injunction. See FTC v. PPG Indus., Inc., 798 F.2d 1500, 1508 (D.C.Cir.1986).

This Circuit has articulated an analytical approach by which the FTC may establish a Section 7 violation. First, the FTC must show that the merger would produce a firm controlling an undue share of the relevant market and would result in a significant increase in the concentration of the market. Heinz, 246 F.3d at 715 (citations omitted). "Such a showing establishes a presumption that the merger will substantially lessen competition." Id. (citing Baker Hughes, 908 F.2d at 982). Defendants can then rebut the presumption by producing evidence that market-share statistics produce an inaccurate account of the merger's probable effects on competition in the relevant market. Id.; see also United States v. Citizens & S. Nat'l Bank, 422 U.S. 86, 120, 95 S. Ct. 2099, 45 L. Ed. 2d 41 (1975). "If the defendant successfully rebuts the presumption [of illegality], the burden of producing additional evidence of anticompetitive effect shifts to the government, and merges with the ultimate burden of persuasion which remains with the government at all times." Heinz, 246 F.3d at 715 (quoting Baker Hughes, 908 F.2d at 983). Accordingly, plaintiffs have the burden on every element of their Section 7 challenge, and a failure of proof in any respect will mean the transaction should not be enjoined.

The Supreme Court has observed that "``only examination of the particular market — its structure, history, and probable future — can provide the appropriate setting for judging the probable anticompetitive effects of the merger.'" General Dynamics, 415 U.S. at 498, 94 S. Ct. 1186 (quoting Brown Shoe, 370 U.S. at 322 n. 28, 82 S. Ct. 1502). Hence, antitrust theory and speculation cannot trump facts, and even Section 13(b) cases must be resolved on the basis of the record evidence relating to the market and its probable future. See Eastman Kodak Co. v. Image Technical Services, 504 U.S. 451, 460-67, 112 S. Ct. 2072, 119 L. Ed. 2d 265 (1992); FTC v. Weyerhaeuser Co., 665 F.2d 1072, 1080 (D.C.Cir.1981) ("Hearings on preliminary injunctions [under Section 13(b)] necessarily look to the future and decisions must rest on comparative, tentative assessments of the course of events if the injunction is issued, and if it is not."); Verizon Communications, Inc. v. Law Offices of Curtis V. Trinko, ___ U.S. ___, ___, 124 S. Ct. 872, 882, 157 L. Ed. 2d 823 (2004) (focus on actual rather than theoretical guards against false condemnation that may chill conduct that antitrust laws are designed to protect). Analysis of the likely competitive effects of a merger requires determinations of (1) the relevant product market in which to assess the transaction, (2) the geographic market in which to assess the transaction, and (3) the transaction's probable effect on competition in the relevant product and geographic markets. See Marine Bancorporation, 418 U.S. at 618-23, 94 S. Ct. 2856; General Dynamics, 415 U.S. at 510-11, 94 S. Ct. 1186.

II. ANALYSIS

A. The Coal Industry

Coal is the base fuel for over fifty percent of the electricity generated in the United States. Bales Tr. (6/21 afternoon) at 33:7-8.^[4] It is mined from underground or surface mining operations, depending on the location and reserves, then delivered by truck, barge, or rail to generating sites where it is typically stored and then fed into boilers to generate electricity. Id. at 33:6-13. The coal available in the United States falls into three general categories: lignite, subbituminous, and bituminous. Subbituminous coal has a Btu^[5] content that typically ranges from 8000 to 9500 Btus per pound, and is principally located in Montana and Wyoming. It has become the fuel of choice for a large number of electric utility companies in the United States. Id. at 33:14-25.

The Powder River Basin ("PRB") refers to a large coal-bearing geological formation located in Montana and Wyoming. DX 9007 at 001. Coal in the Southern Powder River Basin ("SPRB") of northeastern Wyoming is located close to the surface, and can be mined by large-scale surface operations as opposed to underground mining techniques. Lang Tr. (6/24 afternoon) at 46:6-49:2; see DX 1024 at 24648; DX 9000. The estimated reserves of coal in the SPRB exceed 7.8 billion tons. DX 9005 at 002. Currently, approximately 360 million tons of coal are produced annually in the SPRB, which represents approximately one-third of the coal produced annually in the United States. FTC Compl. ¶ 13; States Compl. ¶ 15.

SPRB coal is known for its low sulfur content. Bales Tr. (6/21 afternoon) at 35:7-9. Sulfur content first became a critical factor when the Clean Air Act was passed in 1970. Id. at 35:2-9. SPRB coal is the most economical source of fuel that complies with Clean Air Act sulfur limitations. Id. SPRB coal also has a moderately high heat content and low ash content. Heat content (measured in Btu) affects the amount of energy a user can extract from a unit of coal; ash content affects the grindability of the coal and the performance of air quality equipment, and thus customers prefer coal with low ash content. The SPRB provides a critical source of plentiful, inexpensive coal having a strong combination of these important characteristics. PX 0021 ¶ 8 (Rea Decl.); PX 0042 ¶¶ 4-11; PX 0012 ¶¶ 5, 11-15 (Panzarino Decl.); PX 0010 ¶¶ 3-7 (Flippin Decl.).

Coal resources in the SPRB are controlled primarily by the federal government, and the vast majority of SPRB production is from lands subject to federal coal leases, with a small amount from state and private lands. DX XXXX-XXXXX. The SPRB mines can be divided into three tiers based on coal quality, heat content, and mine location. Bales Tr. (6/21 afternoon) at 40:1-41:2; PX 0312 at 018; DX 1024 at 24648. Tier 1 mines typically produce a high Btu (8600-8900) coal and include the Antelope, Black Thunder, Jacobs Ranch, North Antelope/Rochelle, and North Rochelle mines. Bales Tr. (6/21 afternoon) at 40:2-7; PX 0011 at 013-014; PX 0012 ¶ 5 (Panzarino Decl.); DX 1024 at 24650. Tier 2 mines produce coals ranging from 8300 to 8550 Btu, and include Belle Ayr, Caballo/North Caballo, Coal Creek, and the Cordero Rojo complex. PX 0772 at 36, 66; PX 0011 at 013-014; PX 0012 ¶ 5 (Panzarino Decl.); DX 1024 at 24650. Tier 3 mines produce relatively low Btu coal (7900-8450) and include the Buckskin, Dry Fork, Eagle Butte, Fort Union, Rawhide, and Wyodak mines. PX 0011 at 013-014; PX 0012 ¶ 5 (Panzarino Decl.); DX 1024 at 24650.

Seven companies currently operate the fourteen mines in the SPRB. DX 1024 at 24670. Four companies, each operating a Tier 1 mine, are considered the major producers of SPRB coal: Arch, Triton, Kennecott Energy Co. ("Kennecott"), and Peabody Holding Co. ("Peabody"). Arch operates the Black Thunder and Coal Creek mines; Triton operates the North Rochelle and Buckskin mines; Kennecott operates the Antelope, Jacobs Ranch and Cordero-Rojo mines; and Peabody operates the North Antelope/Rochelle, Caballo, and Rawhide mines. PX 5675 at 002. RAG American ("RAG") is another significant producer in the SPRB, but it only operates mines in Tiers 2 and 3 (Belle Ayr and Eagle Butte). PX 3308 at 002. Two small mining entities, Western Fuels and Wyodak, generally do not compete for business in the region and, therefore, are not recognized by most customers as feasible supply alternatives to the five larger producers. PX 0010 ¶¶ 8-9 (Flippin Decl.); PX 0042 ¶ 10 (Luksan Decl.); PX 0022 ¶ 3 (Stolwyk Decl.); PX 0012 ¶ 14 (Panzarino Decl.).

Virtually all SPRB coal is purchased by electric power companies for use in their coal-fired steam generating units. FTC Compl. ¶ 13; States Compl. ¶ 15. There are approximately 150 generating plants that presently consume coal from the SPRB, DX 0086 ¶ 21, located throughout the United States ranging from Montana to Texas, and Washington to New York, DX 9007 at 004. SPRB coal is shipped via railroad for the most part, and the SPRB is served by two railroads: the Union Pacific ("UP") and the Burlington Northern/Santa Fe ("BNSF"). Shalah Tr. (6/25 afternoon) at 28:13-18. The "joint line" is a railroad line that is jointly owned by BNSF and the UP. It is dispatched by the BNSF and serves the SPRB mines south of Gillette, Wyoming: Antelope, North Antelope/Rochelle, North Rochelle, Black Thunder, Jacobs Ranch, Coal Creek, Cordero-Rojo, Belle Ayr, and Caballo. Id. at 22:2-6. The "single line" is a railroad line owned and operated by the BNSF, which serves the mines north and east of Gillette: Buckskin, Eagle Butte, Dry Fork, Fort Union, and Wyodak. Id. at 22:7-20.

The SPRB is the fastest growing coal producing region in the United States, with demand projected to grow from a present level of over 360 million tons per year to over 500 million tons per year during the next ten years. PX 8605 at 005-006; DX 1024 at 24756. Industry analysts attribute the expectation for growth in demand for SPRB coal to low-cost production and compliance with recent environmental legislation. PX 0123 at 002-003. The SPRB coal industry has experienced some consolidation in recent years, as the number of SPRB producers has decreased. PX 8605 at 005-006.

B. The Coal Market

The definition of the relevant market is necessary to identify that area of trade within which a defendant allegedly has acquired or will acquire an illegal monopolistic or oligopolistic position. See Marine Bancorporation, 418 U.S. at 618, 94 S. Ct. 2856 (determination of relevant market "necessary predicate" to deciding whether Section 7 violation). A relevant market has two dimensions. First, the "relevant product market" identifies the product and services with which the defendants' products compete. Second, the "relevant geographic market" identifies the geographic area in which the defendant competes in marketing its products or service. The FTC bears the burden of proof and persuasion in defining the relevant market. United States v. Sungard Data Sys., 172 F. Supp. 2d 172, 182-83 (D.D.C.2001).

1. Relevant Product Market

In determining relevant product markets, courts have traditionally emphasized two factors: "the reasonable interchangeability of use [by consumers] and the cross-elasticity of demand between the product itself and substitutes for it." Brown Shoe, 370 U.S. at 325, 82 S. Ct. 1502; see also United States v. E.I. du Pont de Nemours & Co., 351 U.S. 377, 395, 76 S. Ct. 994, 100 L. Ed. 1264 (1956). The general question is "whether two products can be used for the same purpose, and if so, whether and to what extent purchasers are willing to substitute one for the other." Staples, 970 F.Supp. at 1074 (quoting Hayden Publ'g Co. v. Cox Broad. Corp., 730 F.2d 64, 70 n. 8 (2d Cir.1984)). Relevant markets will generally include producers who, given product similarity, have the ability to take significant business from each other. See SmithKline Corp. v. Eli Lilly & Co., 575 F.2d 1056, 1063 (3d Cir.1978); Staples, 970 F.Supp. at 1075.

Determining interchangeability is relatively straightforward. Courts compare the use or function of defendant's product with other products, and assess "the degree to which buyers are willing to substitute those similar products for the [test] product." Sungard Data, 172 F.Supp.2d at 182. If consumers can substitute the use of one for the other, then the products in question will be deemed "functionally interchangeable." See, e.g., Staples, 970 F.Supp. at 1074 (finding that office supplies sold by an "office superstore" like Staples are functionally interchangeable with office supplies sold at "mass merchandisers" like Wal-Mart); E.I. du Pont de Nemours, 351 U.S. at 399, 76 S. Ct. 994 (recognizing "functional interchangeability" between cellophane and other flexible wrappings). Courts will generally include functionally interchangeable products in the same product market unless factors other than use indicate that they are not actually part of the same market. See, e.g., United States v. Archer-Daniels-Midland Co., 866 F.2d 242, 246 (8th Cir.1988) (even though beet sugars and high-fructose corn sugars were functionally interchangeable, they did not belong to the same product market because government price support for beet sugars meant that prices for corn sugars could be raised substantially without feeling the competitive impact of beet sugar prices).

After finding interchangeability, courts look for cross-elasticity of demand. If a slight decrease in the price of product A causes a considerable number of customers of product B to switch to A, that would indicate that a cross-elasticity of demand exists between A and B and that they compete in the same product market. See E.I. du Pont de Nemours, 351 U.S. at 400, 76 S. Ct. 994. In defining relevant product markets, courts should "exclude any other product to which, within reasonable variations in price, only a limited number of buyers will turn." Times-Picayune Publ'g Co. v. United States, 345 U.S. 594, 612 n. 31, 73 S. Ct. 872, 97 L. Ed. 1277 (1953). The FTC's Horizontal Merger Guidelines set forth an analytical framework for considering the issues of interchangeability and cross-elasticity of demand by defining a product market as "a product or group of products such that a hypothetical profit-maximizing firm that was the only present and future seller of those products (``monopolist') likely would impose at least a ``small but significant and nontransitory' increase in price" (SSNIP). U.S. Dep't of Justice & Fed'l Trade Comm'n, Horizontal Merger Guidelines § 1.11 (1992), as revised (1997) (hereinafter "Merger Guidelines § ______"). In most contexts, the FTC considers a price increase of five percent to constitute a SSNIP. Id. Courts have also relied on several "practical indicia" to help identify the relevant product market.^[6] These indicia include: industry or public recognition of a 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 venders. Brown Shoe, 370 U.S. at 325, 82 S. Ct. 1502. Examination of such factors is helpful to augment the assessment of interchangeability and cross-elasticity of demand when determining the relevant product market.

Relevant market analysis is based on the "narrowest market" principle. Guerin-Calvert Tr. (6/30 afternoon) at 44:5-7. The analysis begins by examining the most narrowly-defined product or group of products sold by the merging firms to ascertain if the evidence and data support the conclusion that this product or group of products constitutes a relevant market. If not, the analysis shifts to the next broadest product grouping to test whether that is a relevant market. This process continues until a relevant market is identified. Id. at 44:8-17. In a relevant market analysis, the hypothetical monopolist test is generally used to identify whether a small but significant price increase for the product tested would be profitable.^[7] Morris Tr. (6/23 afternoon) at 48:24-49:9; Guerin-Calvert Tr. (6/30 afternoon) at 44:24-45:11.

The parties agree that SPRB coal is a relevant product market within which to examine the competitive effects of the proposed transactions. DFF 57; PFF 60. Plaintiffs' expert concluded, after applying the hypothetical monopolist test, that the relevant market is no broader than SPRB coal. Morris Tr. (6/23 afternoon) at 43:11-25; Morris Tr. (6/24 morning) at 56. Plaintiffs also argue that 8800 Btu SPRB coal is a distinct and narrower relevant product market within which to assess the proposed transaction. Defendants' expert, applying the same test, concluded that the relevant market is no narrower than SPRB coal. Guerin-Calvert Tr. (6/30 afternoon) at 40:13-15. However, defendants have not set forth a viable argument, supported by evidence, for consideration of all PRB coal as a broader relevant market. The Court, therefore, rejects the proposition that the entire PRB coal market is the relevant product market for analysis of the proposed transactions.

With respect to plaintiffs' contention that the relevant product market is narrower than SPRB coal, plaintiffs' expert did state his belief that 8800 Btu SPRB coal is a distinct relevant market, but did so only very reluctantly when pressed by the Court. Dr. Morris first stated that he was unable "to state with a reasonable degree of certainty that 8800 Btu coal is a relevant antitrust market," PX 4300 ¶ 39; Morris Tr. (6/23 afternoon) at 44:19-21, then testified that "the relevant market ... could be as narrow as 8800 Btu coal," Morris Tr. (6/24 morning) at 56:9-15. Later, he opined that "8800 Btu coal is likely to be a relevant market," and finally, when prompted by the Court, asserted that 8800 Btu coal is a separate relevant market. Id. at 57:16-58:7. Nonetheless, the clear impression left by Dr. Morris is that he is not entirely sure that 8800 Btu SPRB coal is a relevant product market, and that he would prefer to undertake further analysis before so concluding. The Court thus finds unconvincing his reluctant conclusion that 8800 Btu SPRB coal is a relevant product market.

In order for 8800 Btu coal to be considered a separate relevant market, plaintiffs must show that 8800 Btu coal is not interchangeable with 8400 Btu coal. The testimony of Dr. Morris does not satisfy that burden, because he does not account for the substantial evidence regarding the purchasing practices of utilities, which establishes that 8800 and 8400 Btu coal are substitutable. In both trial testimony and depositions, virtually all the utilities acknowledged that they can and do purchase and consume both 8800 and 8400 Btu coal, and that they actively solicit and consider both in their coal bidding procedures. Guerin-Calvert Tr. (6/30 afternoon) at 57:18-58:5, 60:5-11; DX 0086 at ¶ 22; DX 2006 at ¶¶ 19-20; Holloway Tr. (6/22 afternoon closed) at 77:24-78:14 (OG & E's facilities were designed to burn, and have burned coals ranging from 8400 to 8800 Btu); Freund Tr. (6/21 afternoon) at 80:25-81:2 ("Q: Are you able to burn both 8,400 and 8,800 SPRB coal at your facilities? A: Yes we are."); Orme Tr. (6/22 morning) at 13:18-14:9; Kelly Tr. (6/22 morning) at 56:17-20; Herndon Tr. (6/22 afternoon) at 23:13-18; Carr Tr. (6/23 afternoon) at 10:17-20; Thede Dep. at 43:11-22; Rackers Dep. at 105:16-106:22; Flippin Dep. at 105:25-107:3 (LCRA has purchased both 8400 and 8800 coal in the past five years); McGowan Dep. at 155:3-10, 167:4-9, 180:21-24; Stolwyk Tr. (6/23 morning) at 67:20-24; DX 0919; DX 0920; DX 0921; DX 0926; Werner Tr. (6/23 morning) at 8:19-9:10, 26:9-11; Lapplander Tr. (7/7 morning) at 8:7-18; Stuchal Tr. (7/7 morning) at 67:21-22; DX 0919 at 0022359; DX 0853 at XXXXXXX-XXX; DX 0951 at 0023232; DX 0953 at XXXXXXX-XX; DX 0955 at 0023399; DX 0897; DX 0898; DX 0899; DX 0716 at 0018525. Some customers did indicate that they prefer 8800 Btu coal to 8400 Btu coal, but the evidence also shows that customers having that preference nonetheless can use and have used other Btu coals, and benefit from the competition between 8800 and 8400 Btu coal. Kelly Tr. (6/22 morning) at 48:1-5, 62:10-13; Stuchal Tr. (7/7 morning) at 73:21-74:3 (the Nebraska Public Power District's plant managers prefer 8800 Btu coal but purchase 8400 to 8800 Btu coal depending on which coal has the best evaluated price); Guerin-Calvert Tr. (6/30 afternoon) at 73:15-20. Plaintiffs do not dispute that for most SPRB customers "purchasing decisions [between 8800 Btu and 8400 Btu SPRB coal] are based on economics." Bales Tr. (6/21 afternoon) at 42:17-43:21. The evidence also shows, however, that some customers can only purchase either 8800 or 8400 Btu SPRB coal, but not both, regardless of the economics. Rahm Tr. (6/21 morning) at 89:7-90:11; Orme Tr. (6/22 morning) at 7:11-20.^[8] The Supreme Court has stated that the "[d]etermination of the competitive market for commodities depends on how different from one another are the offered commodities in character or use, how far buyers will go to substitute one commodity for another." E.I. du Pont de Nemours, 351 U.S. at 393, 76 S. Ct. 994. In determining interchangeability, therefore, the Court must consider the degree to which buyers treat the products as interchangeable, but need not find that all buyers will substitute one commodity for another.

Plaintiffs contend that defendants, in arguing against 8800 Btu coal as a relevant market, have failed to consider both the extent to which buyers are unwilling to substitute between 8800 and 8400 Btu coal and the extent of the price increase that would be necessary before significant substitution would occur. Pls. Post-Hearing Reply Br. at 20. The burden, however, is squarely on plaintiffs to establish that 8800 Btu coal is a separate relevant market. See Sungard Data, 172 F.Supp.2d at 182-83. Based on the reluctance of plaintiffs' own expert to conclude that 8800 Btu SPRB coal is a separate relevant market, and the evidence of significant interchangeability between 8800 Btu and 8400 Btu coal, the Court declines to find that 8800 Btu coal is a separate relevant market. Both Dr. Morris and Ms. Guerin-Calvert concluded, after applying the hypothetical monopolist test and critical loss analysis, that SPRB coal is a relevant market. Plainly, SPRB coal is considered a unique and preferred product by customers. See Kelly Tr. (6/22 morning) at 41:21-25; Holloway Tr. (6/22 afternoon) at 76:1-7; Werner Tr. (6/22 afternoon) at 116:1-22. The Court therefore concludes that the relevant product market is no broader and no narrower than SPRB coal. Defendants' half-hearted argument for a market of all PRB coal is totally unpersuasive, and plaintiffs have not carried their burden of establishing that 8800 Btu SPRB coal is a distinct relevant product market.

2. Relevant Geographic Market

The relevant geographic market in which to examine the effects of a merger is "the region in which the seller operates, and to which the purchaser can practicably turn for supplies." FTC v. Cardinal Health, Inc., 12 F. Supp. 2d 34, 49 (D.D.C.1998) (citing Tampa Elec. Co. v. Nashville Coal Co., 365 U.S. 320, 81 S. Ct. 623, 5 L. Ed. 2d 580 (1961)). The Supreme Court has emphasized that the relevant geographic market must both "correspond to the commercial realities of the industry and be economically significant." Brown Shoe, 370 U.S. at 336-37, 82 S. Ct. 1502 (internal citations omitted). The Merger Guidelines also provide guidance for determining the relevant geographic market. The geographic market should be delineated as "a region such that a hypothetical monopolist that was the only present or future producer of the relevant product at locations in that region would profitably impose at least a ``small but significant and nontransitory' increase in price, holding constant the terms of sale for all products produced elsewhere." Merger Guidelines § 1.21. If buyers would respond to the SSNIP by shifting to products produced outside the proposed geographic market, and this shift were sufficient to render the SSNIP unprofitable, then the proposed geographic market would be too narrow. Id.

As with the relevant product market, the relevant geographic market depends on interchangeability and cross-elasticity of demand. Here, however, because the relevant product market is defined in geographic terms as SPRB coal, which is produced and sold in that region, the product market and geographic market analysis are really the same. See Morris Tr. (6/23 afternoon) at 67:14-22. The parties agree that the relevant geographic market has the same scope as the relevant product market — the SPRB — and hence the Court concludes that the SPRB is the relevant geographic market.

C. Market Concentration

As recognized by the D.C. Circuit in Heinz, the theory of merger law is that in a market with few rivals, firms are able to coordinate behavior, "either by overt collusion or implicit understanding," to restrict output and achieve anticompetitive profits. 246 F.3d at 715; see also PPG Indus., 798 F.2d at 1503. The FTC has the burden of showing that the challenged transactions will result in "undue concentration in the market." Baker Hughes, 908 F.2d at 982. Increases in concentration exceeding certain levels raise a likelihood of "interdependent anticompetitive conduct." Heinz, 246 F.3d at 715-16 (quoting General Dynamics, 415 U.S. at 497, 94 S. Ct. 1186).

Market concentration is a function of the number of firms in a market and their respective market shares. Merger Guidelines § 1.5. Market shares are calculated by the Department of Justice and the FTC using the best indicator of firms' future competitive significance. Dollar sales, shipments, and unit sales can be used to calculate market shares, depending on the nature of the firms and their products. Physical capacity or reserves generally will be used as an indicator of firms' future competitive significance if these are the measures that most effectively distinguish those firms. Id. § 1.41. Assessing the coal industry in the late 1960's, which was then dominated by long-term supply contracts, the Supreme Court in General Dynamics rejected statistical evidence of coal production as a valid measure of competitive power in the market: "A more significant indicator of a company's power effectively to compete with other companies lies in the state of a company's uncommitted reserves of recoverable coal." 415 U.S. at 502, 94 S. Ct. 1186.

Other things being equal, market concentration affects the likelihood that one firm, or a small group of firms, could successfully exercise market power. Merger Guidelines § 2.0. The Herfindahl-Hirschman Index ("HHI") of market concentration is used by the FTC to interpret market data. The HHI is calculated by summing the squares of the individual market shares of all the participants in the market. See Heinz, 246 F.3d at 716 n. 9.

The spectrum of market concentration as measured by the HHI is divided into three regions. A market with an HHI of less than 1000 is "unconcentrated," a market with an HHI between 1000 and 1800 is "moderately concentrated," and a market with an HHI above 1800 is "highly concentrated." Merger Guidelines § 1.5. An increase in HHI of greater than 100 points in a post-merger moderately concentrated market potentially raises significant competitive concerns. Likewise, an increase in the HHI of 50 points or more in a post-merger highly concentrated market may raise significant competitive concerns. It is presumed that mergers producing an increase in HHI of greater than 100 points in a highly concentrated market are likely to create or enhance market power or facilitate its exercise. Merger Guidelines § 1.51; see generally Heinz, 246 F.3d at 716 & n. 9. Other things being equal, cases falling just above and just below a threshold present comparable competitive issues. Merger Guidelines § 1.5.

If HHI figures are sufficiently large, they will establish a prima facie case of an anticompetitive merger. See Heinz, 246 F.3d at 716; Baker Hughes, 908 F.2d at 982-83 & n. 3. Thus, where the HHI calculation potentially raises significant competitive concerns or indicates a merger to be presumptively illegal, the factors set forth in Sections 2 through 5 of the Merger Guidelines become relevant in undertaking a more comprehensive and holistic assessment of whether the proposed merger is likely to create or enhance market power or facilitate its exercise. These factors broadly include: the potential adverse competitive effects of a merger, entry analysis, efficiencies, and failure and exiting assets.

1. Competitors in the SPRB

There are currently five significant producers of SPRB coal: Peabody, Kennecott, Arch, RAG, and Triton. PX 5675 at 1; PX 1040 at 1; Morris Tr. (6/23 afternoon) at 44:6-7; Morris Tr. (6/24 morning) at 16:12-15. Post-merger, there will still be five significant producers of SPRB coal, with Kiewit replacing Triton as an SPRB producing entity. The percentages of the firms' market shares will change, to be sure, as Arch will acquire the North Rochelle mine and Kiewit will take over only Triton's Buckskin mine. However, Arch will remain third among the five producers. PX 5675 at 3-4.

Based on the HHI calculation for the current SPRB coal market, it is highly concentrated. Morris Tr. (6/23 morning) at 43:11-44:7, 67:23-68:9. Whether market concentration is measured in terms of practical capacity, loadout capacity, production, or reserves, the post-merger market remains highly concentrated. PX 5675 at 3.^[9] The post-merger increase in HHI ranges from 49 points to 224 points, depending on which measure is used to calculate market concentration. Set forth in the table below is an abridged summary of the different market concentration measurements derived from PX 5675 at 3, based on Arch's acquisition of the North Rochelle mine and Kiewit's acquisition of Buckskin:

SPRB Coal Market Concentration (HHI)
-------------------------------------------------------------------------------
HHI             Practical Capacity   Loadout Capacity    Production   Reserves
-------------------------------------------------------------------------------
Current Market           2152              2068              2201       2054
-------------------------------------------------------------------------------
Post-Merger              2346              2292              2365       2103
-------------------------------------------------------------------------------
Post-Merger
Increase                193               224               163        49
-------------------------------------------------------------------------------

2. Proper Basis to Calculate Concentration

Market shares are calculated by the Department of Justice and the FTC using the best indicator of firms' future competitive significance. In General Dynamics, the Supreme Court stated that evidence of market shares and market concentration should provide a "proper picture of a company's future ability to compete." 415 U.S. at 501, 94 S. Ct. 1186. Defendants urge the Court to use reserves as the basis for calculating concentration, while the FTC argues that the Court should look to other factors as well, and to loadout capacity in particular, to measure concentration.

Reserves are a valid measure of market concentration, albeit an imperfect one. According to defendants' expert, the conclusion that recoverable reserves are the best measure of a producer's future capability is consistent with the economic literature on exhaustible resources. Guerin-Calvert Tr. (6/30 afternoon) at 97:3-15. At current production rates, Triton's North Rochelle mine, for example, has only seven-and-a-half years of reserves and has long-term contracts already in place that commit a significant percentage of the mine's production over that period. Hake Tr. (6/29 closed) at 28:19-20; 29:2-8; DX 9005 at 002; DX 0274; DX 0281. The Supreme Court in General Dynamics rejected statistical evidence of coal production as a valid measure of competitive power in the market in favor of a company's uncommitted reserves of recoverable coal. The Court reasoned that

[t]he bulk of the coal produced is delivered under long-term requirement contracts, and such sales thus do not represent the exercise of competitive power but rather the obligation to fulfill previously negotiated contracts at a previously fixed price. The focus of competition in a given time frame is not on the disposition of coal already produced but on the procurement of new long-term supply contracts.... A more significant indicator of a company's power effectively to compete with other companies lies in the state of a company's uncommitted reserves of recoverable coal.

415 U.S. at 502, 94 S. Ct. 1186.

Here, however, the reserves data defendants urge the Court to use include both committed and uncommitted reserves of coal. Closing Argument Tr. (7/20) at 37:17-21. Those are the reserves that plaintiffs present as well. Id. at 37:20-21. Furthermore, the long-term contracts under which coal was sold before 1974, when General Dynamics was decided, extended for many years and mines effectively could not obtain additional reserves. Morris Tr. (6/23 afternoon) at 74:23-75:8. The current contracting trend in the SPRB is towards contracts of shorter duration, PX 8623 at 001, often lasting three years or less, Morris Tr. (6/23 afternoon) at 75:9-10; PX 8611 at 021 ¶ 68 (Morris W.D.); see also Stuchal Tr. (7/7 morning) at 74:9-14, although they can also be longer. Rahm Tr. (6/21 afternoon) at 7:25-10:1 (Westar has supply agreements through 2020 for most of its coal and other supply agreements through 2007 and 2009); Kelly Tr. (6/22 morning) at 44:2-10; PX 8606 at 3-4; Lien Dep. at 195:1-14 (RAG has signed recent contracts of 2 to 3 year duration and utilities sometimes ask for longer terms); Herndon Tr. (6/22 afternoon) at 5:10-6:23, 42:21-43:5; PX 8604 at 3; Stuchal Tr. (7/7 morning) at 74:9-14 (NPPD's typical coal purchase agreements are 3 to 5 years in length; NPPD has 9-year contracts currently). Contracts that last over three years, which still exist, often include re-opener clauses, PX 4613 at 003, 004-007; Guerin-Calvert Tr. (7/1) at 67:14-17, that allow a customer at certain points over the course of the contract period to solicit competing bids from other suppliers and afford the existing supplier an opportunity to match the competitor's bid. Morris Tr. (6/24 morning) at 48:7-16. Furthermore, in the SPRB today, mines have the ability to acquire more reserves. Morris Tr. (6/23 afternoon) at 75:19-21; PX 8611 at 017 ¶ 57 (Morris W.D.). When SPRB mines need to acquire more reserves, they apply to the Bureau of Land Management ("BLM") to lease additional reserves. Id. at 73:16-75:21. All the major mine owners in the SPRB today have applications (known as "LBAs") pending with the BLM to acquire additional reserves. Id. at 75:15-18; see PX 8446; PX 8447. Today in the SPRB, mines can even make sales for periods beyond the life of the mine based on current reserves because they are able to acquire these additional reserves. Morris Tr. (6/24 morning) at 89:25-90:13. These considerations, in combination, undermine exclusive reliance on reserves to measure market concentration.

Nevertheless, in other contexts the federal antitrust agencies still espouse the view that reserves provide the proper measure for determining concentration levels in the coal industry. See DX 0033 at 000295 (Federal Trade Commission Staff Report on the Structure of the Nation's Coal Industry) (concluding that "a company's holdings of coal reserves are an important indicator of its present and future competitive position, and coal reserve concentration ratios are a very important measure of market power"); DX 0034 (United States Department of Justice, Competition in the Coal Industry). In 1996, in connection with amendments to the Hart-Scott-Rodino Act, the FTC recognized that, in assessing acquisitions in the coal industry, concentration levels are impacted by firms' reserves. See DX 0035 (commenting that "holdings of [coal] reserves ... are widely dispersed, and individual acquisitions have had minimal effect on concentration"). The FTC points out, however, that when using reserves to measure concentration, the DOJ has stated that a market share of 15 percent or more is prima facie inconsistent with the antitrust laws; Arch currently has 16 percent of SPRB reserves and acquiring North Rochelle would increase Arch's SPRB reserves to almost 20 percent. See PX 8611-018 ¶ 59 (Morris W.D.).

Notwithstanding the Supreme Court's reliance on reserves in General Dynamics, exclusive reliance on that measure to determine market concentration in the SPRB today does not appear warranted. The present SPRB coal market is different from the relevant coal market in General Dynamics — particularly in the use of more short-term contracts.^[10] Moreover, the parties have presented statistics based on all reserves, but only uncommitted reserves were considered in General Dynamics and are particularly relevant to assessing future competitiveness. The greater availability of additional reserves through LBAs has also not been fully accounted for in the parties' presentations based on present reserves. Hence, although the Court will look primarily to market concentration as measured by reserves, it will also consider other measures. Plaintiffs agree that it is appropriate to do so. See Closing Argument Tr. (7/20) 53:16-54:1. Those other measures can also be expected to provide some indication of the projected increase in market concentration resulting from the challenged transactions.

Loadout capacity is a helpful measure of market concentration, but it is likewise imperfect. As discussed above, plaintiffs argue that loadout capacity is the most appropriate measure of future competitiveness and thus market concentration because expanding a mine's capacity can be an expensive, slow, and difficult process. See PX 0347 at 002 (expanding Arch's Black Thunder mine by an additional 10 million tons would cost $55 million and take two years); PX 8302 at 202:10-204:20 (Lien Dep.); PX 0016 at ¶ 10 (Conn Decl.) (to expand RAG's Eagle Butte mine beyond its current capacity of 28 million tons would require the addition of another rail loop, the installation of crushers, conveyor, and silos and would cost approximately $45 to $50 million and take up to two years). But it has been demonstrated that loadout capacity can also be expanded fairly quickly and inexpensively. Arch steadily expanded Black Thunder's capacity from 50 million to 70 million tons per year without any significant capital expenditure, Leer Tr. (6/28 morning) at 20:17-21:16; Lang Tr. (6/24 afternoon) at 60:17-62:10, and RAG expanded loadout capacity by 6 million tons between 2000 and 2002, DX 2006 at ¶ 25. Moreover, as a static measure that does not reflect future capacity, Guerin-Calvert Tr. (6/30 afternoon) at 93:12-95:3, loadout capacity does not take into account the constraints that limited reserves may place on a mine's ability to compete in the future. For example, regardless of the loadout capacity at the North Rochelle mine, given the present rate of production, it currently has reserves of only approximately seven-and-a-half years, which will make North Rochelle less of a competitive player in the future unless it acquires additional reserves.^[11] Still, loadout capacity does provide some insight as to competitiveness, and thus is of some value in measuring market concentration.

Production and practical capacity are less useful measures, although they can still be informative. Both are, like loadout capacity, static measures that are not stable predictors of a producer's ability to deliver in the future in the form of sales or contracts. Guerin-Calvert Tr. (6/30 afternoon) at 88:7-21; Lang Tr. (6/24 afternoon) at 62:11-63:19. The Supreme Court specifically rejected past production figures as a measure of future ability to compete in General Dynamics: "[i]n a market where the availability and price of coal are set by long-term contracts rather than immediate or short-term purchases and sales, reserves rather than past production are the best measure of a company's ability to compete." 415 U.S. at 502, 94 S. Ct. 1186. And practical capacity is not generally as sound a predictor of future competitiveness in the SPRB coal market as is loadout capacity.

Because these different measures of concentration are all informative to some degree, yet remain imperfect indicators of future ability to compete, the Court has considered all of the measures in assessing whether plaintiffs have established a prima facie case of an anticompetitive merger. See Heinz, 246 F.3d at 716; Baker Hughes, 908 F.2d at 982-83 & n. 3. Plaintiffs agree with that approach, see PFF ¶ 239, which is also consistent with the general admonition that all aspects of the relevant market should be examined to determine whether a proposed transaction is likely to produce any anticompetitive effects.

3. Post-Merger Increase in HHI

The Court concludes, however, that reserves remain the best of these imperfect measures, and that using reserves as the primary tool to assess concentration and calculate post-merger increases in HHI is also consistent with the Supreme Court's analysis in General Dynamics. The reserves data provided to the Court, see PX 5675 at 3, establish that the current market concentration (HHI) is 2054, and that post-merger it will be 2103, for an increase in HHI of 49. According to the Merger Guidelines, an increase in HHI of 50 points or more in a post-merger highly concentrated market raises "significant competitive concerns." Merger Guidelines § 1.51(c). Although the HHI increase calculated on the basis of reserves is only 49, the Merger Guidelines state that: "the numerical divisions suggest greater precision than is possible with the available economic tools and information. Other things being equal, cases falling just above and just below a threshold present comparable competitive issues." Id. § 1.5. Based on reserves, then, the proposed transaction may raise significant competitive concerns — although just barely.

Loadout capacity is a novel but fairly logical basis for calculating market concentration and assessing future competitiveness because it does act as a constraint, although perhaps not as rigid a constraint as plaintiffs have asserted, on a coal company's ability to compete for future contracts. The post-merger increase in HHI based on loadout capacity is 224. In a highly concentrated post-merger market, according to the Merger Guidelines, with an increase in HHI above 100 it is "presumed that ... [the merger is] likely to create or enhance market power or facilitate its exercise." Id. § 1.51(c). The post-merger increases in HHI calculated based on production and practical capacity are 163 and 193 respectively, both of which also trigger a presumption of the likelihood of creating or enhancing market power or facilitating its exercise.

The various measures of market concentration in the SPRB presented by the parties thus reflect an increase in HHI ranging from 49 to 224. Considering all these measures of market concentration, therefore, at a minimum the proposed transactions raise significant competitive concerns and if, as the Court believes may be appropriate, one departs from a strictly reserves-based approach adopted in General Dynamics because of changes that have occurred in the coal market over the last thirty years, then there may even be a presumption of an anticompetitive increase in market power. Ignoring altogether the other measures of market concentration in favor of an exclusively reserves-based assessment seems unwarranted. The FTC has, therefore, satisfied its prima facie case burden.

Nevertheless, it is important to note that this case is not one in which the post-merger increase in HHI produces an overwhelming statistical case for the likely creation or enhancement of anticompetitive market power. Indeed, the single best available measure of market concentration — reserves — produces an increase in HHI of only 49, which is actually below the level for significant concern in the highly concentrated SPRB market. The measure plaintiffs urge — loadout capacity — only produces an HHI increase of 224. Such HHI increases are far below those typical of antitrust challenges brought by the FTC and DOJ. For example, in Heinz the HHI increase was 510 based on a pre-merger HHI of 4775. See 246 F.3d at 716. In Baker Hughes the HHI increase was 1425, from 2878 pre-merger to 4303 post-merger. See 908 F.2d at 983 n. 3. In Staples the average HHI increase in the several markets under consideration was 2715. See 970 F.Supp. at 1082. And in FTC v. Libbey, Inc., 211 F. Supp. 2d 34, 51 (D.D.C.2002), the impact of the original merger agreement (used by the court for its analysis) was an HHI increase of 1052. All of these levels of HHI increase dwarf even the highest increase arguably present here.^[12] Indeed, between 1999 and 2003, only twenty-six merger challenges out of 1,263 (two percent) occurred in markets with comparable concentration levels to those argued here. See DX 0833 (Federal Trade Comm'n & United States Dep't of Justice, Merger Challenges Data, Fiscal Years 1999-2003 (Dec. 18, 2003)).

Thus, although the FTC has satisfied its prima facie case burden, the FTC's prima facie case is not strong. Certainly less of a showing is required from defendants to rebut a less-than-compelling prima facie case. See Heinz, 246 F.3d at 725; Baker Hughes, 908 F.2d at 991 ("[t]he more compelling the prima facie case, the more evidence the defendant must present to rebut it successfully"). Even assuming that the FTC's showing of an increase in HHI, and thus market concentration, warrants a presumption that the transactions will lessen competition, defendants have pointed out the shortcomings of statistics based on capacity or production, rather than on reserves, in providing the best assessment of the proposed merger's likely future effect on competition. Defendants have, therefore, successfully rebutted the presumption that the merger will substantially lessen competition and the Court will proceed to examine the issue of the likely competitive effects of the proposed merger in the relevant market, for which plaintiffs bear the ultimate burden of persuasion. Heinz, 246 F.3d at 715 (citing Baker Hughes, 908 F.2d at 983). As discussed below, an analysis of the SPRB market confirms that defendants have produced sufficient evidence to further rebut the FTC's prima facie case and that, ultimately, plaintiffs have not carried their burden of persuasion. See Baker Hughes, 908 F.2d at 985 (rebuttal evidence may include factors relating to competition in relevant market and financial or market weakness of acquired company.)^[13]

C. Likelihood of Coordinated Interaction In This Market

In General Dynamics, the Supreme Court cautioned that, although significant, statistics concerning market share and concentration are "not conclusive indicators of anticompetitive effects." 415 U.S. at 498, 94 S. Ct. 1186; see also Brown Shoe, 370 U.S. at 321-22, 82 S. Ct. 1502; Heinz, 246 F.3d at 717 n. 12. Indeed, this Circuit has cautioned against relying too heavily on a statistical case of market concentration alone, and that instead a broad analysis of the market to determine any effects on competition is required. "That the government can establish a prima facie case through evidence on only one factor, market concentration, does not negate the breadth of this analysis. Evidence of market concentration simply provides a convenient starting point for a broader inquiry into future competitiveness." Baker Hughes, 908 F.2d at 984; see Merger Guidelines § 2.0. Judge Posner has described the need to examine the evidence relating to competition in the market as follows:

... the Supreme Court, echoed by the lower courts, has said repeatedly that the economic concept of competition, rather than any desire to preserve rivals as such, is the lodestar that shall guide the contemporary application of the antitrust laws.... Applied to cases brought under Section 7, this principle requires the district court ... to make a judgment whether the challenged acquisition is likely to hurt consumers, as by making it easier for the firms in the market to collude, expressly or tacitly, and thereby force price above or farther above the competitive level.

Hosp. Corp. of Am., 807 F.2d at 1386.

Plaintiffs may seek to show that a merger will diminish competition by showing that it will facilitate coordinated interaction. That is, in fact, the thrust of plaintiffs' case here. The Merger Guidelines define coordinated interaction as "actions by a group of firms that are profitable for each of them only as a result of the accommodating reactions of the others. This behavior includes tacit or express collusion, and may or may not be lawful in and of itself." Merger Guidelines § 2.1; see also Brooke Group Ltd. v. Brown & Williamson Tobacco Corp., 509 U.S. 209, 227, 113 S. Ct. 2578, 125 L. Ed. 2d 168 (1993) (tacit coordination is a "process, not in itself unlawful, by which firms in a concentrated market might in effect share monopoly power, setting their prices at a profit-maximizing, supracompetitive level by recognizing their shared economic interests and their interdependence with respect to price and output decisions").^[14] Indeed, antitrust policy seeks particularly to inhibit "the creation or reinforcement by merger of ... oligopolistic market structures in which tacit coordination can occur." Heinz, 246 F.3d at 725 (quoting 4 Phillip E. Areeda, Herbert Hovenkamp & John L. Solow, Antitrust Law ¶ 901b2, at 9 (rev. ed.1998)).

A market is conducive to tacit coordination, then, where producers recognize their "shared economic interests and their interdependence with respect to price and output decisions." Brooke Group, 509 U.S. at 227, 113 S. Ct. 2578. Successful coordination requires two factors: (1) reaching terms of coordination that are profitable to the firms involved and (2) an ability to detect and punish deviations that would undermine the coordinated interaction. Merger Guidelines § 2.1. Coordination need not be complex or complete — "[i]nstead, the terms of coordination may be imperfect and incomplete ... and still result in significant competitive harm." Id. at § 2.11. The Merger Guidelines provide, moreover, that the punishment of deviation will not always be direct and specific: "Credible punishment ... may not need to be any more complex than temporary abandonment of the terms of coordination by other firms in the market." Id. at § 2.12. But "where detection or punishment is likely to be slow, incentives to deviate are enhanced and coordinated interaction is unlikely to be successful." Id.

1. The FTC's Novel Theory

The FTC brings this action for a preliminary injunction under the theory that the mechanism of tacit coordination that is most strongly supported by the evidence is a form of output restriction in which the major coal producers in the SPRB market would constrain their production so that increases in supply would lag behind increases in demand, thereby creating an upward pressure on price. See, e.g., Pls. Post-Hearing Br. at 8. What this means is that the FTC must show projected future tacit coordination, which itself may not be illegal, which is speculative and difficult to prove, and for which there are few if any precedents.

As defendants have noted, prior coordinated effects challenges to mergers based on alleged output coordination have invariably been accompanied by a coordinated effects theory grounded on price coordination. See, e.g., Univ. Health, 938 F.2d at 1219; FTC v. Elders Grain, Inc., 868 F.2d 901, 905-06 (7th Cir.1989); Hosp. Corp. of Am. v. FTC, 807 F.2d 1381, 1387 (7th Cir.1986); FTC v. Bass Bros. Enters., Inc., 1984-1 Trade Cas. (CCH) ¶ 66,041 at 68,612, 1984 WL 355 (N.D.Ohio 1984). Similarly, although not a merger case, In re High Fructose Corn Syrup, 295 F.3d 651 (7th Cir.2002), involved a coordinated effects theory where the product was highly standardized, completely fungible, and did not vary from producer to producer. See 295 F.3d at 656-57; In re High Fructose Corn Syrup, 156 F. Supp. 2d 1017, 1021 (C.D.Ill.2001), rev'd mem., 295 F.3d 651 (7th Cir.2002), cert. denied, 537 U.S. 1188, 123 S. Ct. 1251, 154 L. Ed. 2d 1019 (2003). There, coordination among producers needed to take account of only one fact — price — see 295 F.3d at 657, and price transparency was virtually complete, with producers publishing prices in advance of their becoming effective, see 156 F.Supp. at 1037. Deviations from coordination were immediately detectable, the other producers could quickly deprive the cheater of any benefit, and there was evidence of a history of anticompetitive behavior in the market. 295 F.3d at 666. As discussed further below, none of these factors is present here. SPRB coal has a wide range of heat level (8800 to 8400 Btu), quality (e.g., sulfur and ash content) and ultimately price, information regarding prices is far from transparent, there is no historical evidence of actual price or output coordination, and the SPRB market does not lend itself to ready detection and punishment of deviations from coordination. Moreover, the number of competitors in the SPRB pre- and post-merger remains the same.

The novel approach taken by the FTC in this case makes its burden to establish anticompetitive effects in the post-merger SPRB market more difficult. In support of their case, plaintiffs cannot simply seek to show that the proposed transactions will decrease production in the SPRB market. As plaintiffs have noted, a coordinated restriction in production will not necessarily decrease total SPRB coal production, but rather, will likely cause the output of SPRB coal to be "less than it would have been absent the [a]cquisition." Pls. Post-Hearing Br. at 8. Instead, plaintiffs must establish that the proposed transactions will increase the risk of coordinated output restriction and decrease the likelihood of deviation in the SPRB market. This requires a sophisticated attempt to show a developing propensity towards this form of tacit coordination in the SPRB market supported by an ability of the SPRB firms to monitor each other's behavior.

Plaintiffs thus seek to demonstrate that the proposed transactions will substantially increase the risk that SPRB producers will engage in coordinated output-constraining behavior because the transactions will: increase the gains from coordination; place most of the excess capacity in the hands of firms most likely to restrict production; eliminate a significant, independent competitor (Triton); substantially weaken the post-merger competitive fringe (RAG and Kiewit); and generally enhance Arch's ability to lead the industry toward an anticompetitive outcome. The Court will proceed by examining the competitive state of the SPRB market today, determining whether coordinated interaction is feasible and, if so, whether there is evidence that actual or tacit coordination has occurred, and then examining the structure and dynamics of the SPRB market, the competitive strength of Triton, and the likely roles that RAG and Kiewit would play in a post-merger market. This broad analysis of the SPRB market is necessary to determine whether the FTC has carried its burden to persuade the Court that the proposed transactions increase the risk of coordinated interaction that will likely substantially lessen competition in the SPRB market.

2. Competition in the Current Market

An oligopolist market is one having a modest to a relatively few number of players and is distinguished from either perfect competition or monopoly. Guerin-Calvert Tr. (6/30 afternoon) at 113:20-114:18. The SPRB coal market, with five significant producers, is an oligopoly. A non-cooperative oligopolist market (or non-coordinated oligopolist market) is recognized in the economic literature to be a market characterized by competitive outcomes even though there are relatively few firms. Guerin-Calvert Tr. (6/30 afternoon) at 113:20-114:18. The evidence shows that the SPRB market is currently a competitive non-cooperative oligopolist market.

SPRB producers appear to follow different competitive strategies depending on their own economic self-interest. DX 9004 at 003; DX 0154; DX 0155; DX 0158. Numerous witnesses from electric utilities testified that the SPRB marketplace presently is "competitive" and that recent bid processes have been "competitive." Rahm Tr. (6/21 afternoon) at 5:12-18, 11:5-7; Holloway Tr. (6/22 afternoon) at 88:5-8; Luksan Tr. (7/7 morning) at 42:5-8; Stuchal Tr. (7/7 morning) at 67:14-20; PX 8309 at 164:12-15 (Panzarino Dep.); PX 8303 at 136:22-137:2 (Richards Dep.); PX 8302 at 208:22-209:1 (Lien Dep.)("The Powder River Basin is extremely competitive."). When a utility issues a Request for Proposal ("RFP"), all five major SPRB producers typically respond. Bales Tr. (6/21 afternoon) at 62:14-63:5. Responses by SPRB producers to RFPs evidence the absence of any coordinated behavior. As one customer testified, there is almost always one "surprise" SPRB producer — invariably a different one — that offers a price lower than expected as a result of unique circumstances facing that particular producer at the time of the bid. Bales Tr. (6/21 afternoon) at 63:15-64:3; see PX 8601 at 8. Customers have the ability to solicit bids from more than the five main SPRB producers, and they often do, soliciting and receiving spot bids from brokers/traders. Herndon Tr. (6/22 afternoon) at 16:3-18:5; DX 0951 at 023232; DX 0953 at 233390; Luksan Tr. (7/7 morning) at 42:1-4; Stuchal Tr. (7/7 morning) at 64:24-65:3; PX 8307 at 107:12-109:15 (Rackers Dep.). Defendants' economic expert concluded, based on her thorough analysis of the SPRB market, that the market is competitive today. Guerin-Calvert Tr. (6/30 afternoon) at 109:12-14; DX 0086 ¶ 36. The FTC's economic expert was not asked to analyze the current state of competition in the SPRB and testified that he therefore had no opinion on the competitiveness of the SPRB today. Morris Tr. (6/24 morning) at 58:22-25. Hence, the FTC does not contend that the current SPRB market is not competitive because there has been actual coordination on price or supply by producers in the SPRB market in the past.

a. The 2001 Price Spike

In 2001, spot prices for SPRB coal spiked. See Bales Tr. (6/21 afternoon) at 44:21-46:11; Rahm Tr. (6/21 afternoon) at 17:23-18:2; DX 0935 at 2 ¶ 5. Rather than being an indication of anticompetitive coordination, however, the 2001 price spike was a short-term phenomenon that is largely explained by several factors other than tacit coordination in the marketplace, including reduced stockpiles, severe weather conditions, and higher prices for natural gas, all of which unexpectedly increased the demand for coal. Guerin-Calvert Tr. (7/1) at 13:5-24; Rahm Tr. (6/21 afternoon) at 17:22-20:7; Luksan Tr. (7/7 morning) at 39:1-40:6; DX 0935 at 2. Buyers reacted to these various factors by resorting to panic buying, primarily in the over-the-counter market, which drove up spot prices. Rahm Tr. (6/21 afternoon) at 19:4-9; DX 0086 ¶ 11. Coal prices in other regional markets across the country also rose in 2001 for the same reasons. Prices for SPRB coal are well below prices for coal from other regions and have remained stable over the last several years following the 2001 spike in spot prices. See PX 4309.

Utility stockpiles were unusually low in 2001. Rahm Tr. (6/21 afternoon) at 18:25-19:23; PX 8303 at 243:15-245:3 (Richards Dep.); DX 0935 at 2. Weather also played a significant role in the spike; the summer and fall of 2000 were mild, Bales Tr. (6/21 afternoon) at 46:19-47:9; PX 8306 at 256:21-257:12 (McGowan Dep.), but the winter of 2000-2001 was unusually cold, causing overutilization of coal reserves and leaving inventories depleted going into the spring and summer of 2001, Rahm Tr. (6/21 afternoon) at 18:23-20:1; DX 0935 at 2.

The price spike was also preceded by a run-up in the prices of alternative fuels. DFF 112. There was a shortage of supply for eastern coal due in part to several mine closures and bankruptcies. Rahm Tr. (6/21 afternoon) at 18:6-19; 19:17-19; Freund Tr. (6/21 afternoon) at 111:13-15. Demand and prices for coal increased in all regions, with eastern coal prices increasing even more than SPRB prices. Luksan Tr. (7/7 morning) at 38:16-21; PX 8306 at 256:21-257:12 (McGowan Dep.). Natural gas prices also increased in the winter of 2000-2001. Rahm Tr. (6/21 afternoon) at 20:2-7; Luksan Tr. (7/7 morning) at 39:21-23; PX 8302 at 58:19-59:12 (Lien Dep.); DX 0935 at 2. At the time, California was experiencing an energy crisis. Luksan Tr. (7/7 morning) at 39:14-16; DX 0935 at 2. The price adjustment for SO2 emission allowances also increased during this time period. PX 8306 at 257:24-258:4, 258:21-259:21 (McGowan Dep.).

These factors explain the 2001 price spike without any anticompetitive or coordinated activity by SPRB producers. SPRB spot prices following the 2001 price spike have declined to near pre-spike levels, and have shown no signs of climbing again. See DX 9007 at 013-014; PX 1510; PX 1511; PX 1514. This is consistent with the pattern for prices for coal in other regions. See PX 4309. Moreover, to the extent a difference can be seen in the spot prices before and after the spike, increased costs largely account for those differences. Guerin-Calvert Tr. (7/1) at 13:25-14:9; 14:22-15:1; Luksan Tr. (7/7 morning) at 40:7-10. A comparison of price levels in other regions shows that, in general, prices in the SPRB are at the same level as they would be if they had increased at the rate of other coal regions. Guerin-Calvert Tr. (7/1) at 14:10-21.

b. Reduced Output and Mine Closures

Plaintiffs contend that SPRB producers have been moving in the direction of industry-wide production discipline since 1999-2000. According to plaintiffs, Peabody and Kennecott have, at various times, recognized the utility of constraining production in order to obtain improved pricing, and have issued public announcements stating that they curtailed production or delayed expansion because current prices did not provide adequate returns. However, defendants have offered evidence of credible, legitimate reasons for the reduced output among SPRB producers in 1999-2000, including the closure of the Cordero-Rojo mine by Kennecott, the closure of the Rawhide mine by Peabody, and Arch's closure of the Coal Creek mine.

Arch idled its Coal Creek mine beginning in mid-2000. Lang Tr. (6/24 afternoon) at 69:5-10. Coal Creek had been closed by its previous owners more than once, and was losing money when the mine was acquired by Arch. Lang Tr. (6/24 afternoon) at 68:5-12. Arch tried to make the mine profitable by expanding production but that effort failed. Leer Tr. (6/28 morning) at 21:17-22:6; Lang Tr. (6/24 afternoon) at 67:13-68:4, 70:3-12. Arch managers and its Board of Directors carefully considered the options available for Coal Creek. DX 1066; DX 0803; DX 0782; DX 0783. Mr. Lang and his staff prepared detailed economic analyses regarding idling Coal Creek. See DX 0774; DX 0088; DX 1067. At the time that Coal Creek was struggling, Arch had implemented several changes to the Black Thunder mine that increased the efficiency of that mine and expanded its production. Lang Tr. (6/24 afternoon) at 68:13-69:4. Entering 2000, Arch was exceeding its production operations at Black Thunder. Id. at 71:17-72:4; DX 0783 at 020276. Arch management determined that it could supply the volume of coal committed by contract for the balance of the year for both mines from the Black Thunder mine alone. Lang Tr. (6/24 afternoon) at 72:22-73:8; DX 0783.

Ultimately, the Court finds Arch's decision to idle the Coal Creek mine in 2000 was prompted by legitimate business reasons, not anticompetitive motives. Many of plaintiffs' witnesses testified that they understood the Coal Creek closing to be for legitimate business reasons and did not regard it as an attempt at coordination. Freund Tr. (6/21 afternoon) at 113:24-114:7; Bales Tr. (6/21 afternoon) at 48:5-10; PX 8302 at 72:22-73:4; 151:10-154:17, 160:20-22 (Lien Dep.); Carr Tr. (6/23 afternoon) at 18:14-17; Warner Tr. (6/23 morning) at 31:11-19, 34:10-14; PX 8304 at 111:16-112:12 (Flippin Dep.). The few customers who speculated otherwise acknowledged that they had no knowledge regarding the financial performance of the mine and were unable to say how Coal Creek's closure affected prices and production at other mines. Orme Tr. (6/22 morning) at 33:6-34:1.

In 1999, Kennecott reduced production at its Cordero-Rojo mine due to geological problems. PX 7141 at 15:4-16:10 (Randen Dep.); DX 1024 at 024799. The decision was an independent one based on Kennecott's operational needs and circumstances. PX 7141 at 32:18-24 (Randen Dep.). In April 1999, Peabody idled its Rawhide mine due to production failures and a lack of customer demand. DX 1024 at 024836. At that time, Peabody switched its Rawhide customers to its North Antelope or Caballo mines and continued to serve them without any shortfall in production. Id.

Thus, the evidence shows that there were independent, legitimate business reasons for the mine closures by Arch, Kennecott and Peabody in 1999 and 2000. Like the factors that credibly explain the price spike of 2001, these reasonable explanations for the mine closures by the "big three" SPRB producers preclude the reduced output during this period from serving as evidence of coordinated or anticompetitive activity in the SPRB market.

c. Increased Production Over Time

SPRB production has increased consistently over the past ten years even though the number of SPRB producers has declined.^[15] DX 9007 at 016; PX 4335 at 001; Carr Tr. (6/23 afternoon) at 23:24-24:12; Shalah Tr. (6/25 afternoon) at 35:9-25; Luksan Tr. (7/7 morning) at 47:8-15; Kelly Tr. (6/22 morning) at 63:20-25. Defendants' expert examined the competitive environment and found that the decisions made by the different producers as to production levels, the contracts on which they bid, and the costs they elected to incur were independently derived based on each producer's particular business strategy. See Guerin-Calvert Tr. (7/1) at 12:18-13:4. For example, Arch's "market driven" strategy was adopted to minimize costs without sacrificing customer contract needs, and the closure of Coal Creek was an independent decision based on a legitimate business assessment. See Leer Tr. (6/28 morning) at 25:13-18, 26:17-22, 27:17-28:1, 50:2-6; Lang Tr. (6/24 afternoon) at 80:4-12; Freund Tr. (6/21 afternoon) at 113:24-114:7; see also Bales Tr. (6/21 afternoon) at 48:5-10 (decision was independent of other mine closures and intended to cut costs); PX 8302 at 72:22-73:4, 151:10-154:17, 160:20-22 (Lien Dep.)(market would need to improve before opening Coal Creek would make sense; SPRB producers were losing money in 2000 and needed to change business practices); Carr Tr. (6/23 afternoon) at 18:14-17 (closure not due to agreement among competitors); Warner Tr. (6/23 morning) at 31:11-19, 34:10-14 ("we did not view [closure] as a signal to the market at the time"); PX 8304 at 111:16-112:12 (Flippin Dep.)(in mid-2000, LCRA was informed by an Arch representative that the Coal Creek mine was being idled due to high operating costs).^[16]

Peabody's business strategy for the last several years has been actively to expand production at its several mines without taking into account the decisions and activities of the other SPRB producers. DX 0154; Rahm Tr. (6/21 afternoon) at 15:19-16:5; DX 0150 at 5533. Kennecott brought production at the Cordero-Rojo mine back to near pre-2000 levels, when it temporarily reduced the mine's operations, and is also expanding production at its Antelope and Jacobs Ranch mines. PX 7141 at 11:5-24, 13:20-24, 15:4-16:10, 16:19-17:17, 19:17-25, 20:10-21:10, and 32:18-24 (Randen Dep.); DX 1024 at 024799; Guerin-Calvert Tr. (7/1) at 12:4-17; PX 0090 at 2. RAG also claims to be poised to expand and it pays no attention to either the actual production or the announcements relating to production of its competitors. PX 8302 at 37:2-8, 42:19-22, 314:12-13; 229:15-230:1, 180:16-185:3, 42:12-18, 61:18-62:15, 316:22-317:7, 317:11-318:4, 85:3-87:22, 328:2-18 (Lien Dep.); DX 0163.

Based on the sealed bid system in place for SPRB coal contracts, the views of customers who solicit bids from the SPRB producers, the conclusions of the experts, the fact that legitimate reasons other than the existence of production coordination account for the 2001 price spike and for the decisions of certain producers to close mines between 1999 and 2000, and the fact that production has been consistently increasing in the SPRB, the Court finds that the SPRB market is currently competitive. Indeed, it is fair to say that the SPEB has in the past operated as a very competitive market for the sale of coal to electric utility customers. The next question, however, is whether, notwithstanding the historical and present competitiveness of the SPRB market, coordinated interaction in the form of tacit output reduction is likely to occur in this market as a result of the proposed transactions.

3. Possibility of Coordinated Interaction

a. Interest in Production Discipline

Producers in the SPRB have certainly evinced some past interest in price or production discipline. On April 25, 2000, Irl Engelhardt, Chairman and CEO of Peabody, gave a speech before the Western Coal Transportation Association, a meeting attended by SPRB coal producers and customers, in which he remarked on disciplining production in the coal industry. PX 0658 at 001. Engelhardt noted that "[o]ne example [of approaches in the coal industry] is making capital investments to improve productivity and lower costs. Nothing wrong here. Lower costs mean higher margins, right? They do unless the incremental production that results contributes to an oversupply situation." PX 0658 at 002. Engelhardt then commented that "[i]f coal producers would use growth in returns as their performance metric, we believe more discipline would be applied to investments that would otherwise lead to oversupply situations." Id. Engelhardt detailed the actions that Peabody was taking to address oversupply in the market:

Peabody is focusing on profitability and high return investments in the Powder River Basin. Here are some recent steps that they have taken:

• In early 1999, Peabody suspended the 10-million-ton-per-year Rawhide Mine, one of the most productive mines in the United States;

• Also in 1999, the company delayed a 30-million-ton-per-year capacity expansion at North Antelope/Rochelle until margins generate the proper returns; and

• In April 2000, it idled a truck/shovel fleet at Caballo, producing 8 million tons per year, until market conditions improve.

Id. at 004.

A month later, on May 23, 2000, Steven Leer, Chairman and CEO of Arch, addressed the Western Coal Council's 2000 Spring Coal Forum, attended by SPRB coal producers and customers. PX 0253; PX 0795; PX 0603. Leer posed the question "What can we do about oversupply?" His answer was "Produce less coal." In identifying a response to low coal prices, he provided the following information:

Subliminal Messages

• If you produce it, they will buy it

• Outcome: Prices have suffered

• Solution: Produce less coal.

PX 0253 at 013; PX 0795 at 012; PX 0603 at 012; Leer Tr. (6/28 morning) at 84:20-21 (citing DX 1069). Leer also identified "produce less coal" as the solution to low prices resulting from the evaporating export market and huge stockpiles. PX 0253 at 013; PX 0795 at 012; PX 0603 at 012; Leer Tr. (6/28 morning) at 84:20-21 (citing DX 1069).

Plaintiffs view these statements from SPRB producers as strong evidence of the type of production coordination that is likely if Arch's acquisition of Triton is allowed. Defendants have explained, however, that Leer's comments are simply an articulation of Arch's "market driven" business strategy, under which Arch will restrict its production when it believes that, due to oversupply, it cannot obtain returns it considers adequate. Leer Tr. (6/28 morning) at 62:21-63:7; 86:11-15. This approach is consistent with the accepted business objective of obtaining an adequate rate of return to fund expansion. Nonetheless, statements of the type made by Leer and Englehardt in 2000 are indicative of possible producer coordination to limit production, and warrant close scrutiny in an assessment of the likelihood of anticompetitive coordination in the SPRB market.

b. Feasibility of Coordinated Interaction

There is evidence that coordination in the SPRB market is feasible. The differences that distinguish coal produced at one SPRB mine from that at another SPRB mine, such as Btu content, sulfur content, moisture and ash content, are similar to differences that distinguish crude oil produced from different wells. Guerin-Calvert Tr. (7/1) at 95:5-23. Standard adjustments are made in pricing to account for any specific differences that do exist in the coal from different SPRB mines. Orme Tr. (6/22 morning) at 20:4-17; Bales (6/21 afternoon) at 41:9-22; Freund Tr. (6/21 afternoon) at 102:4-103:22. Furthermore, plaintiffs' expert concluded that the demand for SPRB coal is inelastic, i.e., the elasticity of demand is less than one. Morris Tr. (6/23 afternoon) at 50:21-51:14. This means that a modest price increase in the highly concentrated SPRB market would be very profitable to producers because it would increase revenues, and therefore profits, even before taking into account the additional profits that would be realized from reductions in total costs as a result of any reduction in output. Id. at 51:6-14.

Barriers to entry into the SPRB coal market increase the likelihood of coordinated interaction. Morris Tr. (6/23 afternoon) at 117:15-21. Certainly there are appreciable start-up costs associated with becoming an SPRB coal producer. The small and frequent transactions for SPRB coal also increase the likelihood of coordinated interaction, decrease the incentive to deviate from coordinated interaction, and increase the likelihood that deviations from coordinated interaction will be quickly detected. Morris Tr. (6/24 morning) at 6:13-8:21, 49:5-24. A typical transaction size in the SPRB coal market is less than one percent of the total market. Id. at 6:23-24.

Key market information relating to the other competitors in the SPRB coal market is available from numerous sources, which would theoretically permit the sharing of information among producers. These sources include: trade reports and conferences, PX 5614 at 001, industry analysts and consultants who publish reports containing annual production, production capacity, and cost-of-production by mine information for the SPRB coal market, DX 1024 at 24662, 24698; PX 0012 ¶¶ 22-23 (Panzarino Decl.); governmental filings such as the Form 423 monthly reports required by the Federal Energy Regulatory Commission ("FERC") stating the quantity and quality of coal purchased and the delivered price for each source, PX 1424 at 007 at 012; PX 0012 ¶ 23 (Panzarino Decl.); PX 0005 ¶ 16 (Rackers Decl.); coal company announcements that inform the public on market conditions, production costs, mine productivity, and whether a company is gaining an adequate return for its coal, PX 1821 at 001; PX 1035 at 021; PX 1408 at 014-016; PX 0097 at 020; PX 5614 at 001; PX 7141; the bidding process which, even though sealed under confidentiality provisions, nevertheless allows some information to be transmitted to producers from customers regarding how their bids compared to other bids from producers, PX 0118; PX 0161; PX 1433 at 015; PX 3307 at 005; PX 3316 at 004; and merger and joint venture negotiations which may allow for limited transfer of certain competitive information between producers, PX 0118 at 001; PX 0161; PX 0609 at 002; PX 0660 at 004; PX 0633 at 001; PX 1405 at 003.

Given a stated interest by some SPRB producers in production discipline, these general features of the SPRB market would not appear to preclude coordinated interaction having anticompetitive effects. However, even though these factors and conditions make post-merger coordinated activity to limit production in the SPRB market feasible, whether anticompetitive coordination is likely requires closer examination of such factors as the past history of coordinated interaction, the SPRB market structure and dynamics, and the roles of "fringe" or "maverick" producers.

c. Existence of Actual Coordinated Interaction

There is insufficient evidence to conclude that express or even tacit coordination has taken place in the SPRB market. Traditional factors that industrial organization economists consider when assessing the susceptibility of a market to coordinated interaction include whether producers recognize their mutual interest in competing less aggressively and whether producers with incentives to compete less aggressively communicate their intentions to one another. PX 8611 ¶ 62. Plaintiffs' expert concluded that he would need to do additional analysis before he could offer conclusive testimony on whether coordinated interaction is occurring in the SPRB coal market. Morris Tr. (6/24 morning) at 15:3-20. Plaintiffs have not produced sufficient evidence that such coordination to limit production has actually occurred.

Based on a review of the evidence over time, it is unlikely that coordination has taken place in the SPRB, especially since the evidence through which the FTC attempts to show the existence of coordination is focused primarily on 2000 and 2001. The lynchpin of the FTC's position is the comments and actions of Arch with respect to "production discipline." Through 1999, Arch sold coal on an incremental basis, which meant it sold coal for anything more than the cost of producing it. Leer Tr. (6/28 morning) at 57:15-58:5. Under what Arch calls its "market driven" approach, however, Arch will restrict its production when it believes that, due to oversupply, it cannot obtain returns that it considers adequate. Leer Tr. (6/28 morning) at 62:21-63:7; 86:11-15. On May 17, 2000, Arch announced that it planned to reduce production at its Coal Creek mine by as much as 10 million tons annually. PX 0214 at 001. Arch announced that it would no longer expand capacity to keep pace with growing demand for SPRB coal until SPRB price and margin increased to an acceptable level. Id. at 002. Arch informed the industry that "moves such as the one we are taking today ... should have a positive impact on prices," in light of the "supply/demand fundamentals" for SPRB coal. Id.

On May 23, 2000, Arch's CEO Steven Leer drew attention to the fact that "Arch has been conscientious" and illustrated the statement by observing that Arch was idling the Coal Creek mine and had limited expansion at Black Thunder. PX 0253 at 014; PX 0795 at 013; PX 0603 at 013. Jon Kelly of Tuco (a utility) recalled that he found Leer's comments disturbing because there were representatives from Triton, Peabody, and Kennecott in the room. PX 8606 ¶ 12 (Kelly W.D.); PX 5303. The speech was reported by the trade press to the coal industry as calling for reductions in coal production. See PX 5680 at 003. A year later, on April 17, 2001, Leer gave an address at the Western Coal Transportation Association in Santa Fe, New Mexico, during which he presented a slide depicting the following information:

What's different today?

Supply/demand balance

Southern PRB

• Fewer producers, so greater potential for discipline

• Loadout and rail capacity create constraints

• Adding incremental production more expensive.

PX 0656 at 015. The next year, Arch announced in a March 17, 2002 press release that it had recently cut production by seven percent. See PX 0602 at 001. Arch announced the production cuts in 2002 despite the fact that the cuts would have a negative impact on earnings: "We are committed to being a market-driven producer. We believe it would be a mistake to sell coal into an oversupplied market, at prices that will not provide an adequate return." Id.

Although Arch made public announcements about cutting production and its commitment to being a market-driven producer, other SPRB producers chose not to follow Arch's strategy. Other producers may have contemplated following suit, but there is no evidence that any of them chose to adopt Arch's approach. Indeed, between 2001 and 2003, Arch itself believed that it detected indications that its competitors (Kennecott, Peabody and Triton) were selling coal at incremental costs, and thus not on the same basis that Arch was advocating. PX 0090 at 002 (Kennecott); PX 0165 at 002 (Vulcan/Triton); PX 0105 (Peabody). Leer testified that public statements by various SPRB producers regarding their mining operations were made to advise the investor community and customers of independent and unilateral business decisions being taken by each company in response to general supply and demand conditions and to the unique business challenges faced by each company — not to signal any sort of coordinated activity. Leer Tr. (6/28 morning) at 8:20-9:10, 42:15-22. Consistently, SPRB producers have denied that there has been any effort among suppliers to coordinate production output. PX 8302 at 149:4-14 (Lien Dep.); PX 7141 at 32:3-24 (Randen Dep.); Leer Tr. (6/28 morning) at 52:19-53:10; Hake Tr. (6/29 morning) at 10:21-11:22. Over the past ten years, SPRB production has expanded considerably even though the number of SPRB producers has declined. DX 9007 at 016; PX 4335 at 001; Carr Tr. (6/23 afternoon) at 23:24-24:12; Shalah Tr. (6/25 afternoon) at 35:9-25; Luksan Tr. (7/7 morning) at 47:8-15; Kelly Tr. (6/22 morning) at 63:20-25.

Furthermore, there is no evidence that Arch sought to "punish" the producers who declined to restrict production, even if it had the means to do so. According to defendants' expert, public announcements about production made by Peabody, Kennecott, and Arch did not trigger a coordinated output reduction by coal producers, Guerin-Calvert Tr. (7/1) at 9:10-25, and were instead followed by enhanced output in the SPRB market, id. at 10:20-11:2; DX 9007 at 031; see also DX 9007 at 022-030. The totality of the evidence, then, establishes that although production restrictions were advocated and even practiced by Arch during 2000-2002, and broader coordination by SPRB producers to limit supply was feasible, no express or tacit coordination to limit production has actually occurred among the major SPRB coal producers.

d. Market Structure and Dynamics

That observed conclusion is consistent with an assessment of the SPRB coal market. The structure and dynamics of the SPRB market may permit coordination, but do not make coordination likely. While barriers to entry into the SPRB market exist, PX 8611 ¶ 119 (Morris W.D.), and such barriers may facilitate the creation or enhancement of market power or its exercise, Merger Guidelines § 3.0, a substantial number of firms actively compete in the marketplace. Furthermore, heterogeneity of products and producers limit or impede the ability of firms to reach terms of coordination. See Merger Guidelines § 2.11. The evidence establishes that products in the SPRB market are heterogeneous; SPRB coal is different from one mine to another, Rahm Tr. (6/21 afternoon) at 12:13-15, 22:8-23:1; Freund Tr. (6/21 afternoon) at 103:9-16; DX 0853 at 7; Orme Tr. (6/22 morning) at 19:16-22; DX 0199 at 000342, and the SPRB mines and coal companies differ in many important respects, including their production costs, cost structures, contractual commitments, level of reserves, and financial viability, Guerin-Calvert Tr. (7/1) at 6:21-7:3, 6:25-7:10; DX 1024 at 0024728; PX 4353 at 001.

It is true that industry publications make some market information available among producers. However, the information published in those sources is limited, imperfect, and largely unreliable and untimely. Public data on coal pricing, capacity, and production levels are historical, not particularly comprehensive, and tend to lag behind the market by several months, if not more. Guerin-Calvert Tr. (7/1) at 15:8-21; DX 2006 at ¶ 38. MSHA data is a good example. It provides on a quarterly or annual aggregated basis a snapshot of historic shipments, but provides no information about future shipments, long-term mine commitments, or the nature of any contracts a supplier may have with utilities, let alone the nature or date of the specific contract under which the reported shipments were made. Guerin-Calvert Tr. (7/1) at 15:22-16:12; Lapplander Tr. (7/7 morning) at 10:11-19; DX 0935 at ¶ 4. FERC 423 reports suffer similar deficiencies in that they, too, report only historic shipments, and the data only become available months after the reported shipments were made. Hake Tr. (6/29 open) at 13:1-8; Guerin-Calvert Tr. (7/1) at 16:13-20; Lapplander Tr. (7/7 morning) at 10:9-17; Stuchal Tr. (7/7 morning) at 66:3-6; PX 1255 at 152:20-153:9 (Pettibone I.H.); PX 8302 at 122:8-12 (Lien Dep.); PX 8303 at 194:10-195:13 (Richards Dep.); PX 8307 at 247:6-247:22 (Rackers Dep.); PX 8309 at 221:4-221:17 (Panzarino Dep.). Moreover, the shipments reflected in FERC reports are (like MSHA data) tied to contracts that were entered into years earlier. Carr Tr. (6/23 afternoon) at 11:15-25; Hake Tr. (6/29 open) at 13:1-8. Thus, if a supplier signs a contract today for deliveries commencing in 2005, FERC data will not be reported until several months after the first shipment is sent in 2005. Guerin-Calvert Tr. (7/1) at 16:21-17:4.^[17] FERC data are reported on a delivered basis (i.e., the combined cost of the coal and transportation) and do not break out the cost of the coal alone. Hake Tr. (6/29 open) at 13:1-12; Kelly Tr. (6/22 morning) at 40:24-41:8; Carr Tr. (6/23 afternoon) at 11:7-14; PX 8302 at 122:3-7 (Lien Dep.); Stuchal Tr. (7/7 morning) at 66:1-7. Because utilities treat their transportation costs as confidential, Herndon Tr. (6/22 afternoon) at 25:19-24; Kelly Tr. (6/22 morning) at 40:5-8, 75:15-19, it is impossible accurately to determine the cost of the coal at the mine-mouth.

FERC statistics (and MSHA data) also do not address a number of critical items and provide no information on most of the competitively significant terms and conditions of coal contracts. Guerin-Calvert Tr. (7/1) at 17:2-4. For example, FERC data do not capture the various terms and conditions of coal contracts that affect the overall value of the contract to the producer. PX 8302 at 114:10-115:14, 117:8-16, 119:7-122:7 (Lien Dep.)("I would say that the FERC price represents the — the cost that the utility has incurred by purchasing and transporting the coal. I don't think anything else would be in there."). Also, unregulated utilities (representing twenty percent of SPRB customers) are not even required to report their data, and hence will not show up in the FERC reports. Hake Tr. (6/29 open) at 13:1-12; Guerin-Calvert Tr. (7/1) at 17:5-7. For these many reasons, it simply is not possible to estimate current coal prices by looking at FERC (or MSHA) data. Hake Tr. (6/29 open) at 13:1-21; Freund Tr. (6/21 afternoon) at 107:17-21 ("it won't tie to what they would be bidding today"); Stuchal Tr. (7/7 morning) at 65:25-66:6; PX 8307 at 248:6-9 (Rackers Dep.)(stating that he would not look to FERC data to determine what market prices should currently be).^[18]

Available spot market data (OTC data) also are of only marginal usefulness. OTC data are based on a few transactions, primarily coal purchases for delivery within one year, and do not include sales made pursuant to long-term contracts. Leer Tr. (6/28 morning) at 35:23-36:4; Hake Tr. (6/29 open) at 13:13-21; Hake Tr. (6/30 closed) at 9:20-10:2; Kelly Tr. (6/22 morning) at 65:8-22; PX 8306 at 261:2-24 (McGowan Dep.)(stating only 10% of coal trades are reported). Producers and customers alike testified that they would not be able to estimate coal prices for long-term contracts by looking at current OTC prices. Hake Tr. (6/29 open) at 13:13-21; Hake Tr. (6/30 closed) at 9:20-10:2; Freund Tr. (6/21 afternoon) at 107:22-108:1; Orme Tr. (6/22 morning) at 26:13-19; Carr Tr. (6/23 afternoon) at 12:1-20; PX 8307 at 125:20-127:16 (Rackers Dep.). Moreover, OTC prices do not dictate the prices producers bid. PX 7741 at 43:5-19 (Pettibone Dep.). In fact, at the time of the price spike in 2001, there was little change in the negotiated price for long-term coal contracts. See DX 9007 at 013-014.

A market is conducive to tacit coordination where producers recognize their "shared economic interests and their interdependence with respect to price and output decisions." Brooke Group, 509 U.S. at 227, 113 S. Ct. 2578. Successful coordination requires both that firms reach terms of coordination that are profitable and that they be able detect and punish deviations from the coordinated interaction. See Merger Guidelines § 2.1. In order for producers to be able to coordinate production, they would need a reliable reference point to attain agreement as to a lag in production. Supply and demand estimates in this marketplace, however, have been historically inaccurate and uncertain.

Demand for SPRB coal is not predictable either in the short- or the long-term. Guerin-Calvert Tr. (7/1) at 37:19-38:10; DX 9008A. The two largest demand drivers for coal consumption are the weather and the economy, and neither can accurately be predicted. Hake Tr. (6/29 open) at 11:6-13; Kelly Tr. (6/22 morning) at 66:10-20. Unexpected weather and changes in economic conditions can result in utilities delaying receipt of coal under contract, advancing receipt of coal, choosing to increase or decrease inventories, or buying and selling coal on the spot market to meet immediate, unanticipated demand. PX 8304 at 159:3-14 (Flippin Dep.).

The opaqueness of utility inventories also makes it difficult to predict the demand for SPRB coal from year to year. When utilities draw down inventories, they defer coal purchases, resulting in reduced demand; conversely, when utilities build up inventories, they drive up demand. Utilities do not disclose their inventory levels to suppliers. Luksan Tr. (7/7 morning) at 38:6-15. There are thus many different forecasts of future inventories, none of which is very accurate. Hake Tr. (6/29 open) at 11:7-12:1. As a result, predictions of future demand for coal vary dramatically, with no real consensus as to what future usage of electricity might be. See Guerin-Calvert Tr. (7/1) at 22:25-23:9, 25:5-16; Leer Tr. (6/28 morning) at 38:18-22; Lapplander Tr. (7/7 morning) at 12:12-20; DX 0848 at 0021359. Customers are no better at projecting demand. As Mr. Bales testified, he takes his own company's estimates of coal consumption "with a grain of salt." Bales Tr. (6/21 afternoon) at 71:19-72:1. Demand estimates, particularly as they relate to coal prices, "would be pure speculation." Orme Tr. (6/22 morning) at 31:16-21; see also Carr Tr. (6/23 afternoon) at 22:20-23:7; DX 0506 at 3 (coal prices "are extremely difficult to predict"); DX 0937 at ¶ 20 ("future coal pricing is difficult to predict."). Long-term, variable contracts further complicate efforts to forecast the demand for coal from year to year. Guerin-Calvert Tr. (6/30 afternoon) at 109:21-110:1; DX 0086 at ¶ 49.

Similarly, supply is hard to predict because suppliers and other would-be sources of supply side intelligence seldom publicize reliable statistics regarding production levels and decisions. Most public statements made by suppliers contain little meaningful information about actual market conditions or the likely actions of market participants. For instance, RAG does not make public its decisions regarding pricing or production and does not reveal its uncommitted tonnage or its proprietary mine models. PX 8302 at 46:22-47:15, 137:17-20, 259:14-260:2, 291:14-292:14 (Lien Dep.). Similarly, Triton does not discuss its committed and uncommitted sales with third parties. PX 7785 at 202:17-24 (Wilkerson Dep.). Buckskin's mine manager has never seen committed or uncommitted sales figures for Arch or Peabody. Id. at 202:25-203:6. Kennecott, likewise, testified that the limited "hearsay" about other mines is generally related to safety issues, not production decisions. PX 7141 at 35:6-17 (Randen Dep.). Arch, too, shares such information with no producer, and receives it from none. Leer Tr. (6/28 morning) at 50:2-6, 50:25-51:5; Lang Tr. (6/24 afternoon) at 80:4-7.

Statements made in analyst calls and at industry conferences, or contained in financial reports, are of little informative value as they tend to be infrequent, not systematic, and purposefully vague and incomplete. When coal producers make statements as to their uncommitted future tons, they "very rarely" provide data specific to the SPRB but rather provide their position company-wide. PX 8302 at 256:16-257:2 (Lien Dep.); Leer Tr. (6/28 morning) at 31:10-32:4; DX 1068; DX 0972. As MidAmerican's representative explained, "they're pretty clever at hiding how much of it's PRB versus eastern coal." Freund Tr. (6/21 afternoon) at 106:24-107:4. In addition, the statements usually differ greatly from actual performance. For example, publicly available reports do not accurately reflect RAG's expansion plans or mining costs. PX 8302 at 34:1-7, 34:19-36:11 (Lien Dep.). And as one utility explained, "I've heard producers tell me they're going to increase production in the past and then ultimately don't." Freund Tr. (6/21 afternoon) at 115:3-4.^[19]

The emphasis on sealed bids and confidentiality is an important aspect of the market structure and dynamics that would frustrate coordination among producers. Utilities purchase the great majority of their coal through the issuance of an RFP. Kelly Tr. (6/22 morning) at 55:3-8; Carr Tr. (6/23 morning) at 81:19-22; Luksan Tr. (7/7 morning) at 35:15-19; Stuchal Tr. (7/7 morning) at 64:12-15; PX 8306 at 214:8-13 (McGowan Dep.). The RFP bidding process is conducted on a strictly confidential basis. Rahm Tr. (6/21 afternoon) at 6:25-7:14; Bales Tr. (6/21 afternoon) at 63:6-14; Kelley Tr. (6/22 morning) at 55:3-12; Herndon Tr. (6/22 afternoon) at 13:20-14:16; Werner Tr. (6/23 Morning) at 19:16-22; Lapplander Tr. (7/7 morning) at 10:6-8; Stuchal Tr. (7/7 morning) at 65:12-24; Luksan Tr. (7/7 morning) at 10:6-8; DX 0940; PX 8302 at 106:21-107:17 (Lien Dep.); PX 8303 36:17-19, 37:3-6, 37:10-38:2, 38:10-39:8 (Richards Dep.); PX 8306 215:21-216:10 (McGowan Dep.); PX 8307 at 20:17-20, 83:2-6 (Rackers Dep.); PX 8304 at 169:2-17 (Flippin Dep.); PX 8309 at 160:2-22 (Panzarino Dep.). Hence, producers are not aware of who else is bidding or the bids of the other producers. Rahm Tr. (6/21 afternoon) at 7:6-11; Bales Tr. (6/21 afternoon) at 63:6-14; Kelly Tr. (6/22 morning) at 55:25-56:5; Werner Tr. (6/23 morning) at 19:23-20:2; Luksan Tr. (7/7 morning) at 37:11-19; Stuchal Tr. (7/7 morning) at 65:12-24; PX 8302 at 160:10-20 (Lien Dep.); DX 0001; DX 1038; DX 0935. Triton, like other SPRB producers, has never told another producer what it bid; nor has it been told by another producer what that company bid. Hake Tr. (6/29 open) at 12:17-25. Confidential bidding gives an incentive to producers to offer a low price because "the vendors have the understanding that those offers that they turned in will not be viewed by their competition." Kelly Tr. (6/22 morning) at 55:20-24; DX 0026.

Even after the bidding process is completed and there is a winner, utilities typically do not reveal the successful bidder, the winning bid price, exactly how far off the other bidders were, the ranking of the bids or how the utility determined the winner. Bales Tr. (6/21 afternoon) at 64:4-6; Freund Tr. (6/21 afternoon) at 105:11-15;PX 8302 at 19:18-20:3 (Lien Dep.); PX 8306 223:12-16 (McGowan Dep.); Luksan Tr. (7/7 morning) at 37:20-24. Confidentiality provisions typically included in coal supply contracts prohibit the winner from revealing the terms as well. Hake Tr. (6/29 open) at 12:17-21; Rahm Tr. (6/21 afternoon) at 7:15-21; Stuchal Tr. (7/7 morning) at 66:25-67:9; PX 8303 at 234:1-237:15 (Richards Dep.); PX 8307 at 83:7-84:3 (Rackers Dep.); DX 0016; DX 0896 ¶ 18. The few references to specific contracts in the trade press are highly generalized and do not include specific contract terms. Guerin-Calvert Tr. (7/1) at 18:24-19:2.

Utilities are careful not to divulge other information that might be useful to suppliers. For example, utilities treat their transportation costs as confidential. Herndon Tr. (6/22 afternoon) at 25:19-24; Kelly Tr. (6/22 morning) at 41:5-8, 75:15-19; PX 8309 at 29:4-20 (Panzarino Dep.). They also keep their inventory levels confidential. Luksan Tr. (7/7 morning) at 38:6-15; DX 0523 at 1 ("little real time data exists relative to utility coal inventory"). They frequently do not publicize the results of test burns. Herndon Tr. (6/22 afternoon) at 26:23-27:7. Utilities divulge even less information today than they did in the past because, as a result of deregulation, virtually all utilities compete with each other, and they therefore are particularly intent on guarding information about how much coal they are purchasing and at what price. PX 1255 at 149:22-151:2 (Pettibone I.H.). Utilities generally solicit bids without disclosing in the RFP the particular plant for which they are purchasing the coal so as to protect themselves against price discrimination. Carr Tr. (6/23 afternoon) at 6:1-7; PX 8306 at 232:21-233:15 (McGowan Dep.). They may also wait to decide the plant to which they send the coal until after seeing the bids or even after the contract is signed. Carr Tr. (6/23 afternoon) at 6:8-19; Freund Tr. (6/21 afternoon) at 104:22-105:4.

Tacit agreement would also be difficult to coordinate in this marketplace because the terms of agreement would be hard to communicate between producers, even though tacit agreement only requires producers to adopt a uniform strategy that is consistent with less aggressive competition. Moreover, there is no effective mechanism in the SPRB to discipline any producer that would deviate from the terms of coordination. Plaintiffs' economic expert posited no theory for punitive discipline among producers. Due to the nature of the confidential bidding and contracting process that gives producers incentives to submit aggressive bids to capture long term contracts, cheating would not be detected until well after the fact, if ever, and any punishment would come well after the fact as well. Guerin-Calvert Tr. (6/30 afternoon) at 42:19-23. Such delays in detection or punishment generally mean that deviations are likely and that coordinated interaction is unlikely to succeed. See Merger Guidelines § 2.12.

Finally, plaintiffs rely heavily on testimony from utility customers expressing concern about the increased consolidation in the SPRB market that will result from the proposed transaction. Herndon Tr. (6/22 morning) at 74:9-16 ("we think that a continuation of a consolidation of the number of suppliers in the Powder River Basin is not beneficial to the customers who use that coal"); PX 0030 ¶ 9 (Herndon Decl.); Holloway Tr. (6/22 afternoon) at 54:18-25 ("My biggest concern is that I will have one less person to do contract business with to purchase coal from in the Southern Powder River Basin" and that "concentration of ownership in the PRB coal market would translate to higher prices."); Bales Tr. (6/21 afternoon) at 55:5-13 (concern about "the concentration of market power, the consolidation and the fear that I had that by having one less competitor in the market we would see prices rise"); Freund Tr. (6/21 afternoon) at 96:4-24 ("My biggest concern just ties to fewer suppliers. It's my belief that as you reduce the number of suppliers, it will have a negative impact on price...."); Werner Tr. (6/22 afternoon) at 122:4-25; Orme Tr. (6/22 morning) at 12:13-22; PX 0003 at 013-014 (Richards Decl.); PX 0010 ¶¶ 11-12 (Flippin Decl.); PX 0021 ¶¶ 14-17 (Rea Decl.); PX 0012 ¶¶ 24-26 (Panzarino Decl.); PX 0005 ¶¶ 21-24 (McGowan Decl.); PX 0036 ¶¶ 17-18 (Kotara Decl.). In many contexts, however, antitrust authorities do not accord great weight to the subjective views of customers in the market. See 2A Areeda, et al., Antitrust Law ¶ 538b, at 239 ("[l]east reliable [evidence] is ``subjective' testimony by customers ... Though not irrelevant, such statements are often unreliable."); FTC. v. Owens-Illinois, Inc., 681 F. Supp. 27, 38 (D.D.C.), vacated as moot, 850 F.2d 694 (D.C.Cir.1988) ("opinions of purchasers must be viewed in light of their actual behavior"). Furthermore, while the Court does not doubt the sincerity of the anxiety expressed by SPRB customers, the substance of the concern articulated by the customers is little more than a truism of economics: a decrease in the number of suppliers may lead to a decrease in the level of competition in the market. Customers do not, of course, have the expertise to state what will happen in the SPRB market, and none have attempted to do so. The Court therefore concludes that the concern of some customers in the SPRB market that the transactions will lessen competition is not a persuasive indication that coordination among SPRB producers is more likely to occur.

Having concluded that the structure and dynamics of the SPRB market are not conducive to an increased likelihood of tacit coordination as a result of the proposed transaction, the Court proceeds to a consideration of whether there will be an increased likelihood of coordination because a significant and independent competitor will be eliminated from the SPRB market.

e. Triton as a Market "Maverick"

"``An important consideration when analyzing possible anticompetitive effects' is whether the acquisition ``would result in the elimination of a particularly aggressive competitor in a highly concentrated market ....'" Libbey, 211 F.Supp.2d at 47 (quoting Staples, 970 F.Supp. at 1083). For purposes of a coordinated effects analysis, the Merger Guidelines define a "maverick" firm as one possessing a greater economic incentive to deviate from the terms of coordination than those of its rivals. Merger Guidelines § 2.12. FTC officials have noted that, in the context of an auction market, to be a maverick a firm must "consistently compete aggressively when it bids, causing other firms to bid more aggressively when it is present." David T. Scheffman & Mary Coleman, Quantitative Analyses of Potential Competitive Effects from a Merger, June 9, 2003 at 6, available at www.ftc.gov/be/quantmergeranalysis.pdf. The "loss of a firm that does not behave as a maverick is unlikely to lead to increased coordination." William J. Kolasky, Deputy Assistant Attorney General, Antitrust Division, United States Dep't of Justice, Coordinated Effects in Merger Review: From Dead Frenchmen to Beautiful Minds and Mavericks, Address Before the ABA Section of Antitrust Law Spring Meeting (Apr. 24, 2002), available at htt:// www.usdoj.gov/atr/public/speeches/11050.htm (citing Jonathan B. Baker, former Director, Bureau of Economics, FTC, Mavericks, Mergers, and Exclusion: Proving Coordinated Competitive Effects Under the Antitrust Laws, 77 N.Y.U.L.Rev. 135 (2002)).

The evidence here does not support the proposition that Triton is, or will likely become, a maverick in the SPRB market. Triton is not presently a maverick in the market, particularly not over the last two to three years. Triton's North Rochelle mine is one of the highest cost mines in the SPRB. DX 1024 at 24708-24709; Hake Tr. (6/29 closed) at 51:20-22. As a consequence, Triton has been forced to adopt a "last mine standing" sales strategy for its North Rochelle coal. Hake Tr. (6/29 closed) at 50:23-51:5, 51:20-23. Pursuant to this strategy, Triton bids its North Rochelle coal at a price that covers its cost plus a profit and waits for the market to come to that price as other mines in the SPRB sell out. Hake Tr. (6/29 closed) at 50:23-51:10. ("We have to wait until others sell out of their production until we get the price which would cover our cost at a profit."). This strategy is driven for the most part by Triton's debt financing obligations, which require Triton to obtain a sufficient return on its coal sales to meet bank commitments as they come due. Id. at 22:18-23:6. Thus, Triton would rarely deviate from the "last mine standing" strategy. Id. at 52:1-13; see also DX 0510 (Triton "intends to hold to its strategy, which is to be ``the last man standing"' despite market weakness).

Because of these circumstances, Triton is wholly indifferent to competitors' production levels or their likely uncommitted tonnage in pricing its North Rochelle coal. Hake Tr. (6/29 closed) at 51:15-25. Triton's goal is not to increase its market share by pricing under its competitors. Hake Tr. (6/29 open) at 9:18-10:2. Rather, Triton seeks to cover its cost and make a profit on each sale by waiting out the competition and obtaining the highest price possible. Hake Tr. (6/29 open) at 9:18-10:2, 50:23-51:10, 86:5-87:5; PX 0623; DX 0644 at 17417 ("Triton's marketing strategy ... is to hold its unsold tonnage as long as possible"). Given North Rochelle's high cost structure, therefore, Triton has been unable in recent years to be at all competitive on contract bids. Guerin-Calvert Tr. (7/1) at 33:17-25; see also DX 9005 at 003. The result, the Court concludes, is that Triton does not lead or even influence pricing in the market, does not compete aggressively, and does not have a history of bidding on contracts consistent with the behavior of a maverick in the SPRB market.

Nor does Triton's excess production or loadout capacity provide a significant constraint on the market because Triton is unable effectively to deploy that capacity at a reasonable cost. Guerin-Calvert Tr. (7/1) at 34:1-5. If Triton remains an independent company, its continuing high production costs will eliminate all prospects that it can become a more competitive bidder in the marketplace in the future, let alone operate as a maverick. Hake Tr. (6/29 closed) at 57:20-24; Guerin-Calvert Tr. (7/1) at 33:7-16; DX 2006 ¶ 36. The testimony of utility customer witnesses established that, in bid after bid, Triton was not competitive and often finished "dead last" in terms of competitiveness for the contract. Rahm Tr. (6/21 afternoon) at 11:20-24; Bales, Tr. (6/21 afternoon) at 50:5-13; 64:24-65:4, 67:6-8; Kelly Tr. (6/22 morning) at 62:14-20, 63:2-6, 70:14-24; DX 0107; PX 8606 at 8; Herndon Tr. (6/22 afternoon) at 19:18-23:11; DX 0955 at 23399; DX 0953 at 23389; Stolwyk Tr. (6/23 morning) at 61:23-62:21; 77:3-6; DX 0919; Lapplander Tr. (7/7 morning) at 16:7-9; Stuchal Tr. (7/7 morning) at 76:14-18, 77:1-6. There is no convincing evidence that the situation would change in the future given Triton's high costs and very weak financial condition, which is discussed further in Section II-E infra. Triton is the only SPRB producer not to increase production over the past five years, and its financial condition and other weaknesses, including the high strip ratio at North Rochelle and its need for extensive loans and increased reserves (the satisfaction of which are highly questionable), make it unlikely that Triton will become any more competitive in the marketplace than it is right now. It has no pricing strategy available to it that would increase competition in the marketplace and no other likely buyer for Triton has been identified. See discussion in Section II-E infra.

4. RAG and Kiewit in the Post-Merger Market

It is defendants' burden to show that the "fringe" of the SPRB market would expand sufficiently to defeat any merger-induced price increase. See Heinz, 246 F.3d at 715. RAG and Kiewit fall in the category of fringe producers in the SPRB market. Both are strong companies with credible plans to expand production significantly and lessen the risk of coordination in the market. The record evidence establishes that they could, and likely would, expand production to meet any production constraint by the major SPRB producers.

Given sufficient future demand, Kiewit intends to increase production at Buckskin by several million tons per year. DX 1089. Kiewit will promptly begin the permitting process required to effect that expansion if the proposed transaction is consummated and Kiewit will reach out to customers to determine demand for this new coal on "day one" after the acquisition goes through. Id. at 52:3-18.

In valuing the Buckskin mine, Kiewit used the same sophisticated modeling techniques the company uses for all of its major investments. These models showed that increasing production by several million tons was "very, very attractive" financially. Id. at 49:2-51:5; DX 2011. Mr. Grewcock testified that he was "very pleased" to see the results of these models, and that the company "absolutely" intends to pursue the avenue of significant expansion once the acquisition is complete. Grewcock Tr. (6/28 afternoon) at 50:21-23, 52:3-7. Kiewit believes that it could achieve this expansion, which would increase Buckskin's loadout capacity, within two years of the acquisition. Id. at 52:19-24; DX 1089.

Kiewit sees the acquisition of additional reserves through the West Hay Creek LBA as a further significant expansion opportunity. Grewcock Tr. (6/28 afternoon) at 51:6-23. The West Hay Creek LBA offers 150 million recoverable tons of coal, which will last 7.5 years at current production rates, and is not near any other existing mines. The acquisition of this lease would allow Kiewit to continue to operate a very low cost coal mine because of the favorable overburden ratios in the West Hay Creek area. Kiewit also plans to upgrade Buckskin's loadout, plant and equipment, which will significantly improve Buckskin's productivity and facilitate future expansion. Id. at 54:16-55:11; DX 1089 at 3. With Buckskin's increased production, overall production within the SPRB would be virtually the same as it is today even if Arch were to refuse to mine a single ton of North Rochelle's uncommitted coal. This expansion would also represent twice the projected estimated growth for SPRB coal of 7.7 million tons per year. See Morris Tr. (6/24 afternoon) at 16:13-18:14. Whether Kiewit operates Buckskin with or without the West Hay Creek reserves, Kiewit's plans to upgrade Buckskin's loadout, plant, and equipment will likely significantly improve Buckskin's productivity and facilitate future expansion, allowing Buckskin to become a more active competitor in the SPRB market.

RAG, under its new management, is a viable competitor that also intends to expand its operations in the SPRB. RAG currently has detailed plans to expand its output from 44 to 60 million tons per year by January 1, 2006. PX 8302 at 37:2-8, 42:19-22, 314:12-13 (Lien Dep.); DX 0163. Referring to its intention to expand, RAG noted in a recent business document that

We are positioned and ready to act swiftly. Belle Ayr is the PRB's sleeping giant which has been written off by our competition and the outside consultants reports, such as Hill because of our issues in 2000-2001. They flat line us going forward for the next ten years and they are as wrong as they can be.

DX 0162 at 2. According to its plans, RAG intends to expand Belle Ayr from 19 to 30 million tons and Eagle Butte from 25 million to 30 million tons without requiring "a major capital investment." PX 8302 at 29:22-30:6, 37:2-8; 59:20-60:1, 314:6-13 (Lien Dep.); DX 0162; DX 0163. This expanded output represents more than the entire uncommitted output of the North Rochelle mine. Hake Tr. (6/29 closed) at 52:19-22. The only regulatory prerequisite for RAG's expansion is the need to submit a new mine plan to the Wyoming Department of Environmental Quality ("DEQ"), which RAG does not consider to be an obstacle. PX 8302 at 128:14-130:8 (Lien Dep.). This is because RAG's mines are "fully permitted, with respect to having gone through all the archeological, wildlife, vegetation, air quality, all those issues." Id. at 183:8-16. RAG has already engaged in "preliminary discussions" with the DEQ regarding its expansion plans. Id. at 229:15-230:1. As a result, RAG expects that it would receive approval to expand its mines to 60 million tons of annual production within a year, even if subjected to a formal review by the state. Id. at 180:16-185:3. Indeed, RAG could accelerate the date for its planned expansion to a date earlier than January 2006. Id. at 187:4-189:15, 215:14-216:9, 225:1-12. "The time frames are relatively short. We're not talking about having to develop a new mine. That's essentially expanding existing operations that have had a successful performance record for three decades." Id. at 232:13-17. RAG has sufficient reserves to implement this plan and "the company views the plans positively." Id. at 42:15-43:14, 130:9-20, 232:5-12 (quotation), 263:2-5.

RAG's expansion plan derives from its overall strategy in the SPRB. RAG designs its mines to have the ability to expand quickly as market opportunities develop. Id. at 49:19-50:1, 52:5-9. Its SPRB mines thus can rapidly expand to their permitted capacity of 70-80 million tons of annual production. Id. at 92:11-93:16. Mr. Lien testified that he would be comfortable ordering equipment and signing contracts even before a new permit is approved, id. at 302:7-303:4, and that RAG has the ability to shift assets and workers between its mines for immediate increases in production, id. at 81:21-83:3. Thus, RAG can and would expand production in response to a price increase. Id. at 42:12-18, 61:18-62:15, 316:22-317:7, 317:11-318:4. RAG has a demonstrated history of expanding production in response to market demand. In 2001, RAG responded to higher demand by bidding for contracts and expanding production. Id. at 85:3-87:22. RAG was able to expand production in 2001 even though the equipment needed was not in the budget. Id. at 172:9-176:22. Consummation of the proposed transaction would not affect RAG's decision to expand if attractive coal prices were available. Id. at 328:2-18.

The presence in the marketplace of RAG and Kiewit, strong companies with concrete plans to expand production significantly, lessens the risk of anticompetitive coordination in the SPRB market. Defendants have shown that the post-merger fringe capacity in the SPRB would be more than sufficient to absorb any increase in demand caused by any production lag coordinated by the "big three" producers — Peabody, Kennecott, and Arch — over the next three years. The profitability model of defendants' expert economist establishes that, based on a published demand forecast, available fringe expansion would be sufficient to defeat any output restriction, thereby resulting in no demand shortfall or price increase in the post-merger scenario. Guerin Calvert Tr. (7/1) at 28:4-24; DX 9015. Hence, fringe expansion by Kiewit and RAG would be enough to cover a demand shortfall (assuming coordinated output limitation by the three large producers) and defeat any price increase. Id. RAG and Kiewit would both be better able to play the role of maverick in the post-merger market than would Triton if no merger occurred, even though neither will possess 8800 Btu coal, which tends to lead prices in the SPRB.

* * * * * *

The Court therefore concludes that the SPRB market is competitive today. Although producers have shown an interest in production discipline and coordination is feasible, there is no evidence that express or tacit coordination has occurred in the SPRB market. Furthermore, an examination of the structure and dynamics of the market, including the significant number of competitors, the heterogeneity of SPRB coal and the mines and firms that produce the coal, the imperfection of pricing information available in the market, the difficulty in obtaining accurate or current supply and demand information, the emphasis on confidentiality in the bidding system, and the difficulty firms would face in identifying and punishing cheaters, leads the Court to conclude that this market is not one in which coordination would be likely if Arch acquires North Rochelle and Kiewit acquires Buckskin. Moreover, Triton is not, nor is it likely to become, a maverick in the SPRB market and, therefore, the proposed transactions will not make coordination more likely by eliminating Triton from the marketplace. Finally, the evidence also establishes that the fringe in the SPRB (consisting of RAG and Kiewit) will be a viable constraint on producer coordination in the post-merger SPRB market.

D. Efficiencies

The Supreme Court has not sanctioned the use of an efficiencies defense in a case brought under Section 7 of the Clayton Act. See FTC v. Procter & Gamble Co., 386 U.S. 568, 580, 87 S. Ct. 1224, 18 L. Ed. 2d 303 (1967) ("Possible economies cannot be used as a defense to illegality. Congress was aware that some mergers which lessen competition may also result in economies but it struck the balance in favor of protecting competition."). However, "the trend among lower courts is to recognize the defense." Heinz, 246 F.3d at 720 (citing cases). In Univ. Health, 938 F.2d at 1222, the court held that "a defendant may rebut the government's prima facie case with evidence showing that the merger would create significant efficiencies in the relevant market." Similarly, in Staples the court considered the defendant's claimed efficiencies, even though it ultimately decided that they were not sufficient to rebut the FTC's case. 970 F. Supp. at 1090. Likewise, this Circuit in Heinz examined (but rejected) evidence of efficiencies, requiring "proof of extraordinary efficiencies" in light of the highly concentrated market there. See 246 F.3d at 720-21.

In 1997, the Department of Justice and the FTC revised their Horizontal Merger Guidelines to recognize that "mergers have the potential to generate significant efficiencies by permitting a better utilization of existing assets, enabling the combined firm to achieve lower costs in producing a given quantity and quality than either firm could have achieved without the proposed transaction." Merger Guidelines § 4. Hence, "[e]fficiencies generated through merger can enhance the merged firm's ability and incentive to compete, which may result in lower prices, improved quality, enhanced service, or new products." Id. However, the government will only consider those efficiencies that are merger-specific and verifiable by reasonable means. Id. The higher the market concentration present in a given case, moreover, the more extraordinary the efficiencies that a defendant must show. See Univ. Health, 938 F.2d at 1223; Merger Guidelines § 4 ("[e]fficiencies almost never justify a merger to a monopoly or near-monopoly"). When there are high market concentration levels, "the court must undertake a rigorous analysis of the kinds of efficiencies being urged by the parties in order to ensure that those ``efficiencies' represent more than mere speculation and promises about post-merger behavior." Heinz, 246 F.3d at 721. However, even where evidence of efficiencies in the relevant market will not support an outright defense to an anticompetitive merger, such evidence is relevant to the competitive effects analysis of the market required to determine whether the proposed transaction will substantially lessen competition. See Tenet Health Care, 186 F.3d at 1054; Staples, 970 F.Supp. at 1088.

Defendants claim that the proposed transaction will realize between $130 and $140 million in efficiencies over the 2004-2008 period. Mr. Lang, President and General Manager of Thunder Basin Coal Company, was admitted as an expert in these proceedings on the subjects of surface mining operations, mine engineering, and mine planning. Lang Tr. (6/24 afternoon) at 107:11-14. He prepared a series of mine plans, including a 2004 budget and a five-year plan, based on a combination of the North Rochelle and Black Thunder mines, as well as with Black Thunder as a stand-alone mine. Lang Tr. (6/25 morning) at 10:25-11:21; DX 1123; DX 1124; DX 2004. In Mr. Lang's supplemental report, he claims that the acquisition will realize $134.8 million in savings over the 2004-2008 time period. DX 1111 at 25758. Of this amount, $27.4 million is general and administrative expenses, which Mr. Lang himself acknowledges is not merger-specific because "another coal company" without an adjacent mine could achieve it. Lang Tr. (6/25 morning) at 18:11-16. This leaves $107.4 million in claimed merger-specific savings from the combination. See DX 1111 at 25758.

Even as to that remaining amount, however, defendants have not made a strong case on efficiencies. Plaintiffs have systematically pointed out deficiencies in defendants' estimates of efficiencies and shown that defendants have not been able to quantify with precision the savings netted by the proposed transaction. Some of the efficiencies identified by defendants are not merger-specific while others are undercut or reduced on the basis of the evidence.

For example, Arch claims that it can recover 8.9 million tons of upper seam coal at North Rochelle for a five-year efficiency of $15.2 million. DX 1111 at 25773. Triton, historically, has chosen not to recover that coal because of the thin seam and poor quality of the coal. However, the quality has improved substantially as Triton has moved further into the North Roundup lease area, PX 7875 at 115, and Triton is now mining it, see Kostic Tr. (6/25 afternoon) at 82. Accordingly, Triton intends to recover all that coal. PX 7875-031 at 115-117 (Deppe Dep.). Mr. Lang concedes that if Triton were to recover this upper seam coal at North Rochelle on its own accord, independent of the merger, then the alleged efficiency would not exist and its value would be zero. Lang Tr. (6/25 morning) at 63-64.

Arch also claims that it will save $11.5 million over the 2004-2008 time period and recover an additional 3.8 million tons of coal by moving a dragline from the current Black Thunder west pit to the adjacent North Roundup lease area of the North Rochelle mine. The relocated dragline would operate in its new location for a two-year period, running in an east-west orientation, before returning to the west pit. It would operate in addition to Triton's existing north-south oriented dragline. However, Arch concedes that Triton will recover this coal on its own accord, independent of the merger — just not as quickly as Arch would be able to in the combined operation. Lang Tr. (6/25 morning) at 66; Kostic Tr. (6/25 afternoon) at 84-85; PX 4405 at 133-34 (Lang Dep.); Painter Tr. (7/7 afternoon) at 33. Arch did not attempt to adjust the value of the savings from reconfiguring Triton's dragline pits to reflect the fact that Triton would likely mine this coal at a later time. Lang Tr. (6/25 morning) at 65-66; Kostic Tr. (6/25 afternoon) at 99. The correct value of this savings, therefore, is simply the time value of money, which would be only a fraction of the $11.5 million claimed by Arch. PX 4501 ¶ 62 (Painter Report). This saving is therefore overstated. Id.; see Painter Tr. (7/7 afternoon) at 33. Moreover, there is no indication that actual sales will increase to reflect the additional production resulting from this efficiency. To the contrary, the combined output of Black Thunder and North Rochelle is projected at a constant 93.5 million tons, which is merely the sum of the pre-merger planned output of the mines. DX 1123.

Arch claims that it received quotes for insurance to cover Triton's property, including the risk of business interruption, at no additional cost to it, and thus that savings would be achieved. Arch alleges that this efficiency flows from its size, diversified portfolio of assets and excellent safety record. DX 0770 at 19715 ¶ 8 (Lang Report). Only a fraction of these savings are merger specific, however. Insuring Triton's property under Arch's policy would change the coverage terms of the insurance, and had Triton accepted those same terms, it would be able to realize over 80% of those savings on its own. Painter Tr. (7/7 afternoon) at 36. Arch's claimed savings, therefore, represent merely the assumption of the additional risk of self-insurance. Id. at 36:19-23. Moreover, another potential purchaser of Triton might be able to achieve the same savings, and therefore they are not merger specific. Id. at 37.

Arch also claims a one-time $3.5 million inventory reduction from the combination of Arch and Triton based on the elimination of North Rochelle's repair parts inventory. DX 0770 at 19715 ¶ 10. But this claim is really just a conversion of an inventory asset to a cash asset. Painter Tr. (7/7 afternoon) at 38. Mr. Lang also claims carrying cost savings of $1.6 million for the period 2004-2008, for the alleged reduction of duplicate parts inventory from the combination of Arch and Triton. DX 0770 at 19776 (Lang Report). These savings derive from the elimination of Triton's entire stock of repair parts, but no evidence was submitted comparing the pre-merger repair parts inventories of Arch and Triton and, therefore, there is insufficient evidence that the acquisition will enable Arch to eliminate all of North Rochelle's supply inventory. PX 4656 at 007 ¶ 16 (Painter W.D.).

Arch further claims that $26.2 million in savings will flow from the elimination of one tractor, one drill, one grader, and one loader as a result of the merger. The claimed savings are derived from more fully utilizing existing capital equipment. Mr. Lang concedes that he estimated equipment needs based in part on his subjective opinion. Lang Tr. (6/25 morning) at 75; PX 1009 at 34 (Lang I.H.); PX 4501 at 012 ¶ 24 (Painter Report). Mr. Kostic testified that a quantitative study could have been done in three weeks, but it was not. Kostic Tr. (6/25 afternoon) at 68:6-15. The claimed savings amount, therefore, is questionable.

According to Arch, another $18.8 million in savings can be effected over five years because three fewer haul trucks will be needed than would be required had the North Rochelle mine remained with Triton. However, no documents support Mr. Lang's estimates regarding these savings. See PX 1009 at 37, 42. The document on which Mr. Lang's estimate purports to be based, the combined mine plan, contains no haul truck detail. DX 1123; Painter Tr. (7/7 afternoon) at 44:11-14. Another plan (which assumes a close three months later) does contain such detail, PX 4654 at 37, but predicts a significantly larger number of trucks for 2006-2008 than the number estimated by Mr. Lang. Compare DX 770 at 19766 with PX 4654 at 37. No explanation has been provided as to why pushing the closing date back three months would significantly change the savings calculation.

The Court concludes that most — perhaps $100 million — of the purported savings from the acquisition originally claimed by Mr. Lang have been called into question as either non-existent or overstated. Hence, defendants' efficiencies defense cannot by itself defeat plaintiffs' claim of anticompetitive effects resulting from the transactions. It is apparent, nonetheless, that some efficiencies will naturally result from the transactions, particularly in the area of lower costs because North Rochelle is adjacent to Black Thunder, which Arch already owns and operates. It also appears that some savings from dragline efficiencies, inventory reduction, and equipment sharing between the two mines will be realized. The realized efficiencies are more likely to be in the $35 to $50 million, rather than the $130 to $140 million, range over the five year period from 2004 through 2008.

Such savings do not, the Court concludes, provide a complete defense to plaintiffs' prima facie case, and defendants have failed to carry their burden of proving a specific amount of efficiencies or showing the impact of the alleged savings on competitiveness in the market. Nevertheless, even if the savings are neither as great as defendants have claimed nor capable of precise quantification based on the evidence presented by defendants, it is apparent that combining the adjacent Black Thunder and North Rochelle mines will inevitably allow Arch to achieve some measure of lower costs and higher productivity. The existence of such efficiencies, therefore, remains relevant to an assessment of the post-merger market and the potential benefits to consumers from cost reductions and increased competition. Efficiencies resulting from the transactions, then, provide some limited additional evidence to rebut the claim of post-merger anticompetitive effects.

E. Failing Firm Defense

A "failing" or even "flailing" company defense has evolved from the Supreme Court's decision in General Dynamics. There, the Court observed that the appropriate focus under Section 7 is on the merging firm's "probable future ability to compete — rather than ... [its] past production." 415 U.S. at 503, 94 S. Ct. 1186. As crystallized here by the FTC, under the defense "a presumption of illegality based on market concentration alone can be rebutted if defendants can prove that the acquired firm's current market shares overstate its future competitive significance due to its weak financial condition." FTC Post-Hearing Br. at 42. A weak financial condition, or limited reserves, may mean that a company will be a far less significant competitor than current market share, or production statistics, appear to indicate. See General Dynamics, 415 U.S. at 503, 94 S. Ct. 1186; Dr. Pepper/Seven-Up Cos. v. FTC, 991 F.2d 859, 864-65 (D.C.Cir.1993). Following General Dynamics, several courts have relied on the weak and worsening position of the proposed acquired company as a significant factor in declining to enjoin a proposed merger. See, e.g., Lektro-Vend Corp. v. Vendo Co., 660 F.2d 255, 276 (7th Cir.1981); FTC v. Nat'l Tea Co., 603 F.2d 694, 699-700 (8th Cir.1979).

However, the evidence of financial or other weakness must genuinely undercut the statistical showing of anticompetitive market concentration. As the Eleventh Circuit has noted:

[W]e view General Dynamics as standing for the unremarkable proposition that a defendant may rebut the government's prima facie case by showing that the government's market share statistics overstate the acquired firm's ability to compete in the future and that, discounting the acquired firm's market share to take this into account, the merger would not substantially lessen competition. The weakness of the acquired firm is only relevant if the defendant demonstrates that this weakness undermines the predictive value of the government's market share statistics.

Univ. Health, 938 F.2d at 1221 (citations omitted; emphasis added); see also FTC v. Warner Communications, Inc., 742 F.2d 1156, 1164 (9th Cir.1984). Indeed, "[f]inancial weakness, while perhaps relevant in some cases, is probably the weakest ground of all for justifying a merger," and "certainly cannot be the primary justification" for permitting one. Kaiser Aluminum & Chemical Corp. v. FTC, 652 F.2d 1324, 1339, 1341 (7th Cir.1981). Thus, financial difficulties "are relevant only where they indicate that market shares would decline in the future and by enough to bring the merger below the threshold of presumptive illegality." 4 Areeda, et al., Antitrust Law ¶ 963(a)(3), at 13.

A merger is not likely to create or enhance market power or facilitate its exercise if the following circumstances are met: 1) the allegedly failing firm would be unable to meet its financial obligations in the near future; 2) it would not be able to reorganize successfully under Chapter 11 of the Bankruptcy Act; 3) it has made unsuccessful good-faith efforts to elicit reasonable alternative offers of acquisition of the assets of the failing firm that would both keep its tangible and intangible assets in the relevant market and pose a less severe danger to competition than does the proposed merger; and 4) absent the acquisition, the assets of the failing firm would exit the relevant market. Merger Guidelines § 5.1. Defendants concede that Triton's financial condition is not so weak as to support a failing company defense under these requirements. See Defs. Post-Hearing Reply Br. at 13. Instead, they offer evidence of Triton's shaky financial status and future prospects as part of the overall record that must be examined in determining whether substantial anticompetitive effects are likely from the transactions. See Weyerhaeuser Co., 665 F.2d at 1080 ("Hearings on preliminary injunctions [under Section 13(b)] necessarily look to the future and decisions must rest on comparative, tentative assessments of the course of events if the injunction is issued, and if it is not."). They stress that "``only examination of the particular market — its structure, history, and probable future — can provide the appropriate setting for judging the probable anticompetitive effects of the merger.'" General Dynamics, 415 U.S. at 498, 94 S. Ct. 1186 (quoting Brown Shoe, 370 U.S. at 322 n. 28, 82 S. Ct. 1502). Even if not weakened or failing to the point of constituting a full defense, then, defendants contend that the financial condition of the prospective acquired company may reveal that its future competitive significance in the relevant market has been overstated.

The FTC contends that Triton is a strong company, citing Triton's showings of positive EBITDA (earnings before interest, taxes, depreciation, depletion, and amortization), the viability of both of its mines in the present marketplace, and its plans to continue seeking refinancing options and acquiring additional reserves should the proposed transaction be enjoined. The FTC points out that Triton's mines have posted "operating profits" in each of the past two years. In 2002, for example, Triton had EBITDA of $53.4 million and profits from operations of $31.4 million. PX 0881 at 004. In 2003, Triton's EBITDA was $51 million and its operating income was $23.2 million. Hake Tr. (6/29 afternoon) at 61:25-62:12; PX 0882 at 003. Triton's first quarter 2004 consolidated earnings reveal a consolidated EBITDA of $9.9 million and operating income of $5 million. PX 0902 at 010; Hake Tr. (6/29 afternoon) at 65:12-23.

Nevertheless, EBITDA is not cash and, for a mining company, ignores a significant factor affecting a mine's financial performance — depreciation/depletion. Hake Tr. (6/29 closed) at 18:15-20:03. Unlike other corporate settings where depreciation is primarily an accounting and tax proposition and does not affect the actual productive capacity of the subject asset, depletion at a mining company directly reduces its future earnings capacity. Id. at 19:17-20:03. Once depleted, reserves are gone and if not replaced will terminate a mine's operation; thus, any earnings measure such as EBITDA that does not account for depletion seriously misstates the financial performance of a mining company such as Triton. Id. at 18:15-20:03. A mining company with positive EBITDA but negative net income is not sustainable for the long-term. Hake Tr. (6/30 open) at 29:10-14. Since its inception in 1998, Triton has consistently lost money; total losses to date are between $100 and $150 million. Hake Tr. (6/29 closed) at 15:17-16:24; DX 9002; DX 1093. The only year in which Triton made money, 2002, resulted from contracts signed during the 2001 price spike. Hake Tr. (6/29 closed) at 16:13-18. Without the benefit of the long-term higher price contracts it signed during the 2001 price spike, Triton would have lost much more. Id. at 49:10-50:4.

Still, the FTC contends that Triton's mines are viable in the market today and will likely remain so over the upcoming years. The "viability" of the mines, however, is not inconsistent with the difficulties Triton faces in owning and operating its mines. The mines may continue to operate, but Triton may become less and less of an active competitor in the SPRB market. For example, Buckskin currently produces approximately 20 million tons, but its output will soon begin to decline unless the mine receives additional capital. Hake Tr. (6/29 closed) at 29:19-20. Buckskin's mining operations are moving away from the loadout facility and will eventually move to a higher overburden area. Id. at 29:21-30:8. Without additional trucks and an overburden loader, which Triton cannot afford, annual production will drop to 16-18 million tons. Id. at 29:21-30:16, 57:2-6; see also PX 7785 at 205:21-208:13.

Triten's North Rochelle mine is one of the highest cost mines in the SPRB. DX 1024 at 24708-24709; Hake Tr. (6/29 closed) at 51:20-22. The cash cost to mine a ton of North Rochelle coal today is significantly greater than the current spot market price for that coal ($5 to $5.50 per ton). Hake Tr. (6/29 closed) at 31:2-7; DX 9005. As North Rochelle's mining operations continue to move away from its loadout facility, its costs will continue to increase even though the overburden ratio may improve somewhat over the next few years. Hake Tr. (6/29 closed) at 83:19-84:11.

North Rochelle is also currently facing a significant shortfall in sales. The mine's budgeted production for 2004 is just 24 million tons. Id. at 20:4-7. In the first six months of 2004, North Rochelle's sales were two million tons under budget. Id. at 20:8-10, 21:4-6. That volume cannot be made up later in the year because of the capacity of the mine and its current equipment. Id. at 21:7-15. North Rochelle also has a few million tons of coal as yet unsold for the second half of the year, and Triton is not optimistic about selling that remaining tonnage in the current marketplace. Id. at 21:16-18, 22:2-5. If North Rochelle ends the year under budget, Triton's budgeted net income for 2004 will turn into a loss. Id. at 21:19-22:1; DX 1092 at 25623. North Rochelle faces a similar sales shortfall for 2005. Hake Tr. (6/29 closed) at 52:14-22. Even these 14 million tons are uncertain, because two contracts representing 5.2 million tons in 2005 production are currently undergoing reopeners, which could result in the termination of the contracts or reduced revenues going forward. Id. at 29:9-13, 50:5-22.

Although Triton has represented that it is seeking additional funds for capital expenditures and reserve lease acquisitions, PX 7859 at 11:2-8, 82:2-13 (Hake Dep.); Hake Tr. (6/30 afternoon) at 12:12-24, Triton faces substantial hurdles in obtaining those additional funds. North Rochelle will exhaust its reserves in seven and a half years and, absent refinancing, Triton will be unable to acquire additional reserves. DFF 192. To date, Triton's prospects for refinancing are questionable. Hake Tr. (6/29 closed) at 56:18-57:6, 58:3-13; DX 0506 at 5. Triton has nominated an LBA contiguous to the North Rochelle mine called West Roundup and an LBA contiguous to Buckskin called West Hay Creek. Hake Tr. (6/29 closed) at 31:23-32:12. The BLM will put these leases up for bid at the end of 2004. Id. at 32:13-20. Triton estimates that it would need over $50 million in cash the first year to bid and win both LBAs. Id. at 32:21-34:5. But Triton does not have the money to acquire these LBAs and would need to obtain additional capital, either through further borrowings or equity infusions, in order to do so. Id. at 34:6-15, 58:3-6.

Because of Triton's credit profile, it cannot obtain bank financing and must instead look to the junk bond markets for financing. DFF 203. Triton has approached two financial institutions to discuss such an offering. Hake Tr. (6/29 closed) at 34:16-35:1. To float a junk bond offering, however, Triton would need to have a Moody's rating above a CCC, which it has never been able to achieve. Id. at 35:2-23. Despite ongoing discussions, Triton has received no commitment for a refinancing. Id. at 35:24-36:1. Triton also has no commitment from its investors for additional capital to fund the LBAs. Id. at 36:2-7. Even if Triton were to obtain the requisite financing, there is no guarantee that it would win the West Roundup LBA, and even if Triton were to acquire West Roundup, its production costs would be substantially more than for the original North Rochelle reserves. Hake Tr. (6/29 closed) at 39:21-40:16; DX 0506 at 4. At the end of the second quarter of 2004, Triton failed to satisfy the fixed charge ratio, which is a covenant in Triton's current loan agreement that measures the ability to repay that loan out of current cash flow. Hake Tr. (6/29 closed) at 46:22-47:16; DX 1092 at 25624.

Although the FTC insists that other buyers exist for Triton's mines, the prospects for identifying and securing another buyer are dim. Over the past three years, Triton has engaged in a comprehensive search for a buyer of its two mines, retaining Rothschild, Inc., the preeminent investment banking firm in the coal industry. Oehlberg Tr. (6/28 afternoon) at 89:2-12, 91:23-92:5. Vulcan had engaged Rothschild in late 2000 or early 2001 to create a transaction that would permit the investors to exit the investment in whole or in part. Id. at 94:24-95:25. After discussions with the four parties submitting final bids, Vulcan selected the most attractive proposal because it created the most value to the Vulcan investors. Id. at 105:2-106:3, 151:23-152:5. Despite a four-to-five month effort, the necessary capital could not be raised and the transaction was terminated in July 2002. Id. at 107:7-108:4.

Vulcan re-engaged Rothschild in the fall of 2002 for a similar effort. Oehlberg Tr. (6/28 afternoon) at 109:8-25, 152:18-153:1. The firm contacted potentially interested parties, provided an information memorandum to companies that signed a confidentiality agreement, solicited preliminary indications of interest, and opened a data room. Id. at 110:1-116:3. Ultimately, three companies submitted binding proposals: Arch, Peabody and Kiewit. Id. at 116:4-11; Hake Tr. (6/29 closed) at 48:13-20. Peabody and Arch proposed to acquire all of Triton; Kiewit proposed to acquire Buckskin only. Oehlberg Tr. (6/28 afternoon) at 116:4-18; Hake Tr. (6/29 closed) at 48:13-20.

For both sales efforts, Rothschild developed a list of, and then contacted, every potential purchaser worldwide, including SPRB coal companies, other coal companies, utilities and foreign mining companies and private equity investors. Oehlberg Tr. (6/28 afternoon) at 96:1-99:3, 99:4-15, 108:11-109:1, 110:1-114:19, 115:10-12, 146:8-147:8; DX 0941. Triton's interest in a sale or refinancing was also publicized in various coal industry publications. Oehlberg Tr. (6/28 afternoon) at 108:17-109:1, 120:25-122:7. As a result of Rothschild's extensive contact efforts, "it's highly unlikely" that any potentially interested party was unaware of the opportunity to acquire Triton from 2001 to 2003. Id. at 108:5-109:7, 122:14-24, 167:16-20.

Although not a failing firm in the technical sense, Triton is plainly a relatively weak competitor in the current SPRB market, with no convincing prospects for improvement. The evidence establishes that it faces high costs, has low reserves, has at best uncertain prospects for loans or new reserves, is in a weakened financial condition, and has no realistic prospects for other buyers. The evidence also suggests that Kiewit, as owner and operator of the Buckskin mine, will be a stronger competitive force in a post-merger market than Triton has been or will be if no merger occurs. See Section II-C-4 supra at XX. Although defendants cannot avail themselves of a failing firm defense to defeat the FTC's antitrust challenge, Triton's weak competitive status remains relevant to an examination of whether substantial anticompetitive effects are likely from the transactions. The Court concludes that based on the evidence before it, plaintiffs' claims of Triton's past and future competitive significance in the SPRB market has been far overstated.

F. Plaintiffs Have Not Met Their Burden on the Likelihood of Success

To repeat, in this action under Section 7 of the Clayton Act and Section 13(b) of the FTA Act, plaintiffs bear the burden of showing a "reasonable probability" that the challenged transactions may substantially lessen competition, but they need not prove an actual Section 7 violation — instead, they must raise such "serious, substantial, difficult and doubtful" questions going to the merits that further investigation and deliberation by the FTC and the court of appeals are warranted. Heinz, 246 F.3d at 714-15. Once plaintiffs have established a prima facie case that there will be a significant increase in market concentration, close examination of the particular market — here the SPRB — is required to assess the likely anticompetitive effects of the challenged transactions. Id.; Baker Hughes, 908 F.2d at 982-83.

The Court has found that plaintiffs have shown a prima facie case, but that defendants have successfully rebutted, in two ways, the resulting presumption that the transactions will substantially lessen competition. First, plaintiffs' statistical case of increased market concentration is far from compelling, and indeed is much weaker than other FTC antitrust challenges, when, consistent with General Dynamics and as urged by defendants, reserves are used to calculate the HHI increase that reflects the likely increase in SPRB market concentration. This case, then, begins as a fairly weak prima facie case, which requires less of a rebuttal showing by defendants. See Heinz, 246 F.3d at 725; Baker Hughes, 908 F.2d at 991. Second, and more importantly, the considerable evidence before this Court with respect to the particulars of the SPRB coal market confirms that it is unlikely that the challenged transactions will substantially lessen competition, and thus rebuts plaintiffs' marginal prima facie case of increased market concentration and the accompanying presumption that the transactions will substantially lessen competition in the SPRB.

Among the conclusions reasonably drawn from an examination of the record evidence relating to the operation of the SPRB coal market are the following. Although the tacit coordination on production by major SPRB producers feared by the FTC is feasible, and some producers have indicated an interest in greater production discipline in the market, the SPRB has been (and remains) a competitive market with no historical evidence of actual express or tacit anticompetitive coordination. Production in the SPRB has increased steadily over the past decade. Neither the 2001 price spike (which is explained by factors other than tacit coordination to limit production) nor occasional mine closures in 1999 and 2000 (which are explained by legitimate, independent business reasons) constitute evidence of anticompetitive coordination among SPRB coal producers. When Arch made statements in 2000 supporting production discipline in the SPRB and thereafter curtailed its coal production, other coal producers did not adopt Arch's approach, and instead continued to produce and sell coal at incremental cost. The absence of evidence of tacit coordination to limit production is fully consistent with the structure and dynamics of the SPRB market, in which sufficient supply and pricing information to facilitate coordination is generally not available, and which is dependent upon an extremely competitive sealed bid contracting process. These features of the SPRB market make tacit agreement difficult to coordinate and demonstrate both that it would be hard to identify deviations from any agreement to limit production and that there is no effective mechanism in the SPRB to discipline producers who deviate from the terms of coordination.

Although defendants have not presented sufficient evidence to support a full defense based either on efficiencies from the proposed transactions or on Triton's weak financial and competitive condition, these are relevant considerations supporting the conclusion that the transactions are not likely to substantially lessen competition in the SPRB. Defendants' evidence on efficiencies does not sustain the claimed savings in excess of $100 million, but does establish that some cost reductions, primarily related to the adjacent positions of the Black Thunder and North Rochelle mines, would benefit competition in the SPRB. Triton has not been, and is not likely to be, a maverick in the SPRB market, given its non-aggressive "last mine standing" bidding strategy, high production costs, and extremely weak financial condition. Although not technically a "failing" company, it is unlikely to be a strong competitor in the SPRB if the transactions are enjoined, having no concrete prospects for a needed cash infusion or a new buyer. Both RAG and Kiewit, on the other hand, would be strong "fringe" competitors in a post-merger SPRB market. Kiewit promises to expand production at Buckskin and compete vigorously, and has the financial strength and mining experience to do so, and RAG also pledges to expand production under its new management consistent with its strategy to expand quickly as market opportunities develop. Together, their fringe expansion is more than enough to cover any demand shortfall and defeat any price increase assuming coordinated output limitation by the three large SPRB producers, and each is better positioned to play the role of a maverick competitor in the SPRB market than Triton would be.

Based on these and the other findings and conclusions discussed above, the Court concludes that defendants have successfully rebutted plaintiffs' fairly weak prima facie case of a Section 7 violation. Plaintiffs have not shown a "reasonable probability" that the challenged transactions may substantially lessen competition in the SPRB. Moreover, the Court concludes based on the record evidence that plaintiffs have not raised such serious and substantial questions going to the merits of the Section 7 claim that a preliminary injunction is warranted under Section 13(b) of the FTA Act to permit further review by the FTC. In short, plaintiffs are not likely to succeed on the merits of their claim of a Clayton Act violation based on the novel theory of prospective tacit coordination on production limits.

G. Weighing the Equities

Section 13(b) "provides for the grant of a preliminary injunction where such action would be in the public interest — as determined by a weighing of the equities and a consideration of the Commission's likelihood of success on the merits." Heinz, 246 F.3d at 714; see 15 U.S.C. § 53(b). The primary public interest favoring preliminary injunctive relief in a Section 13(b) case, which Congress specifically contemplated, is the effective enforcement of the antitrust laws. Heinz, 246 F.3d at 726. Absent a likelihood of success on the merits, however, equities alone will not justify an injunction. See PPG Indus., Inc., 798 F.2d at 1508. Therefore, here the requested preliminary injunction must be denied for that reason alone.

In any event, the equities here do not support enjoining the transactions in the public interest. Plaintiffs argue that a preliminary injunction would protect the public from antitrust injury in two ways: first, an injunction would ensure that an effective remedy is possible at the end of the case, and second, it would protect the public against the exercise of market power by the combined firms in the interim, until the permanent injunction proceeding before the FTC is decided. Plaintiffs point out that the merger will be difficult, if not impossible, to undo and that a preliminary injunction would prevent "scrambling the eggs" — the transfer of trade secrets, degradation of the acquired firm's capabilities, combination of the Black Thunder and North Rochelle mines, and any other actions whose results would be difficult to reverse. Absent a preliminary injunction, Arch will sell Buckskin to Kiewit, which will be difficult to recover, Leer Tr. (6/28 morning) at 14:25-15:2, and will integrate Triton's North Rochelle mine with Arch's Black Thunder mine, which may be impossible to reverse after time passes. See DX 1111 at 757, 765, 767, 769, 770-71, 775-77. The FTC also argues that the public interest in the enforcement of the antitrust laws will be served by enjoining the proposed transaction because public utilities pass on fuel input increases to their utility customers. See PX 0007 at 9 ¶ 17 (Rahm Decl.) ("Any increase in the price of Wyoming PRB coal would immediately impact our electricity customers because we would need to pass on any price increase to end users of electricity generated with Wyoming coal."). According to the FTC, an injunction will benefit all the utilities and consumers of electricity who rely on SPRB coal to provide electric power.

The Court is skeptical that much harm would be incurred by denying the preliminary injunction, however, given the modest level of the projected increase in market concentration and the brief period (presumably less than a year) before FTC proceedings are completed. If this Court issues a preliminary injunction, on the other hand, Arch and Triton will abandon the transaction rather than undergo an administrative proceeding, and any cost savings and output enhancements that the transactions will create will be lost. Hake Tr. (6/29 open) at 13:22-14:17 (Triton will abandon); Leer Tr. (6/28 morning) at 14:17-24 (it is "very likely" that Arch will abandon). Defendants therefore argue that a preliminary injunction would effectively deprive the market of pro-competitive savings, efficiencies, and a new strong competitor. Based on the evidence presented, the Court concludes that the equities, which cannot alone justify an injunction, do not weigh strongly in favor of plaintiffs.

CONCLUSION

For all the reasons stated above, the requests for a preliminary injunction by the FTC and the States in these consolidated cases will be denied. A separate order has been issued on this date.

ORDER

Upon consideration of the motions for preliminary injunction filed in these consolidated cases by plaintiffs Federal Trade Commission ("FTC") and the States of Missouri, Arkansas, Kansas, Illinois, Iowa, and Texas ("States") (together "plaintiffs") against the merger proposed by defendants Arch Coal, Inc., New Vulcan Holdings Co., LLC, and Triton Coal Co., LLC ("defendants"), the evidence and argument in support and in opposition submitted to the Court through briefing and at the evidentiary hearing in this matter, which commenced on June 21, 2004, and the entire record herein, and upon consideration of the renewed motion of defendants to consolidate the preliminary and permanent injunction proceedings, it is this 13th day of August, 2004, for the reasons stated in the Court's Memorandum Opinion issued on this date, hereby

ORDERED that plaintiffs' request for a preliminary injunction is DENIED; and it is further

ORDERED that defendants' motion to consolidate the preliminary and permanent injunction proceedings, renewed by defendants at the close of the hearing on July 7, 2004, is DENIED.

NOTES

[1] Hereinafter, the plaintiffs in the two actions will collectively be referred to as "plaintiffs" or as "the FTC," unless specific reference to the States is necessary. The defendants in these actions are Arch, Triton and New Vulcan (collectively "defendants"), while Kiewit is participating only as an amicus curiae.

[2] On July 7, 2004, the Court issued a decision concluding that Arch's sale of the Buckskin mine to Kiewit would be considered as part of the challenged transactions along with Arch's acquisition of Triton.

[3] The renewed motion of defendants to consolidate the request for preliminary and permanent injunctive relief will also be denied at this time.

This opinion was provided to the parties in final form on August 13, 2004, but public release was delayed to ensure that no confidential business information that had been submitted under seal was being released. Based on defendants' input, some minor adjustments to the opinion have been made.

[4] Citations to the trial record are in the following form: for testimony — "[Witness] Tr. (date) at page: line"; for plaintiffs' exhibits — "PX ____"; for defendants' exhibits — "DX ____"; for plaintiffs' proposed findings of fact — "PFF ____"; for defendants' proposed findings of fact — "DFF ____." The following abbreviations are employed in citations: "Decl." for declaration, "Dep." for deposition, "I.H." for investigational hearing, and "W.D." for written direct testimony.

[5] A British Thermal Unit (Btu) is the amount of heat required to raise the temperature of one pound of water one degree Fahrenheit.

[6] In Brown Shoe, the Supreme Court found that a relevant product market could exist within a larger market. Although courts have sometimes referred to this as a "submarket," the term "submarket" can be used interchangeably with "relevant product market" and the analysis is the same regardless of which term is used. Staples, 970 F.Supp. at 1080 n. 11.

[7] The test assumes a five to ten percent price increase on the identified test product in the hands of a hypothetical monopolist, and examines whether, in such circumstances, there would be a sufficient diversion or loss of sales to producers of another product to make that price increase unprofitable. Morris Tr. (6/23 afternoon) at 49:2-9; Guerin-Calvert Tr. (6/30 afternoon) at 44:24-45:16. If so, the market is too narrowly delineated. The market is appropriately defined when the price increase would be profitable to the hypothetical monopolist. Morris Tr. (6/23 afternoon) at 48:24-49:9; Guerin-Calvert Tr. (6/30 afternoon) at 44:24-45:11. To measure the volume of diverted sales needed to make a particular price increase unprofitable under the hypothetical monopolist test, economists generally perform a critical loss analysis. Morris Tr. (6/23 afternoon) at 49:16-50:11; Guerin-Calvert Tr. (6/30 afternoon) at 46:20-47:1; DX 0086 ¶ 28 n. 24. The critical loss analysis is mathematically informative of the percentage of lost sales needed to defeat the hypothetical price increase. Guerin-Calvert Tr. (6/30 afternoon) at 47:2-5.

[8] Specific boilers may be constructed or configured to burn 8800 rather than 8400 coal, or vice versa.

[9] Capacity may be an appropriate measure of market share because expanding a mine's capacity can be expensive, slow, and difficult. PX 0347 at 002; PX 8302 at 202:10-204:20 (Lien Dep.); PX 0016 ¶ 10 (Conn Decl.). Practical capacity is considered by customers to be the measurement used to determine a mine's ability to supply them with coal. PX 8601 ¶ 8 (Bales W.D.). It refers to the capacity available given the current physical equipment at the mines and the expected overburdens in mine plans. PX 8611 at 15. Loadout capacity refers to the capacity to process coal that has been removed from the pit, crush it, accumulate it for shipment, and load it onto trains. PX 8611 at 15. Current production (whether by mine or by mine operator) may not be a stable predictor of future competitiveness, or a producer's ability to deliver in the future in the form of sales or contracts. Guerin-Calvert Tr. (6/30 afternoon) at 88:7-21; Lang Tr. (6/24 afternoon) at 62:11-63:19. Producers recognize that current production can lead to mistaken perceptions regarding a mine or mine operator's competitive significance. DX 0162.

[10] The continued use of longer-term contracts in the SPRB market does, however, support the use of reserves as one important measure of market concentration.

[11] The uncommitted reserves at North Rochelle are even less, since a significant percent of reserves are already committed through long-term contracts. Hake Tr. (6/29 closed) at 28:19-20; 29:2-8; DX 9005 at 002; DX 0274; DX 0281.

[12] So, too, do the pre- and post-merger market concentration figures present in those cases far exceed the figures for the SPRB market, which barely place the SPRB market in the "highly concentrated" market range.

[13] In Heinz the court noted that rebuttal evidence may take the form of nonstatistical market evidence raising doubts about the persuasive quality of the FTC's statistics or, as was the case in General Dynamics, economic circumstances undermining the predictive validity of the government's statistical presentation. See 246 F.3d at 715 n. 7. Here, defendants have produced both types of rebuttal evidence — as in General Dynamics, an alternative statistical assessment based on reserves undermines plaintiffs' prima facie case, and in addition a close examination of competition in the relevant market (both presently and post- merger) undermines the claim of significant anticompetitive effects.

[14] The FTC has stressed that its challenge here, based on concerns that the major firms in the SPRB may coordinate to restrict output, is founded on the premise that output would be less than it would be absent the merger. FTC Post-Hearing Reply Brief at 6 n. 8; see Staples, 970 F.Supp. at 1082 n. 14.

[15] SPRB production has increased from 230 million tons ten years ago to 360 million tons today. Rahm Tr. (6/21 afternoon) at 16:24-17:21; PX 8306 at 123:7-18 (McGowan Dep.); PX 8304 at 213:24-214:3 (Flippin Dep.); see PX 1040 (production increases over the last five years).

[16] Arch is unlikely to coordinate in reducing production if it would jeopardize the reopening of its Coal Creek mine. Arch is required under its federal lease to resume production at Coal Creek by 2007, Lang Tr. (6/24 afternoon) at 76:5-12; DX 0803 at 0020521, and stands to lose a very substantial investment if it forfeits that lease, DX 2001. Based on current mine planning, Arch anticipates restarting production at Coal Creek before the 2007 deadline. DX 2002. Arch intends to produce enough coal at Coal Creek before the end of 2007 to recoup much of its investment. Lang Tr. (6/24 afternoon) at 75:15-76:4, 91:31-92:2; DX 2000. Mr. Lang testified that he is confident that Coal Creek will again be up and running before the November 2007 forfeiture date. Id.

[17] Pricing data that is even three months old is considered "stale" in the industry. Freund Tr. (6/21 afternoon) at 107:8-13.

[18] COALdat data are no more useful, relying heavily on FERC data and, while attempting to derive some FOB mine prices from delivered prices, suffering from the same shortcomings as FERC data. Guerin-Calvert Tr. (7/1) at 17:8-19. Information that mine operators provide to the railroads for planning purposes concerns short-term shipments and provides no insight into a mine's future contracting plans. Guerin-Calvert Tr. (7/1) at 17:20-18:8. The resulting data from the National Coal Transportation Association (total tonnage and total trains for the upcoming month) is provided on an aggregated basis — no mine specific data are available to customers or producers. Lapplander Tr. (7/7 morning) at 10:20-12:1.

[19] Information from coal industry news services varies from source-to-source and is frequently unreliable. As one customer put it, the coal rags are of limited usefulness because they are known to "have bollixed up some of [their] stories." PX 8307 at 101:1-6 (Rackers Dep.). As explained above, industry consultants vary in their assessments of supply and demand. Hake Tr. (6/29 open) at 11:6-13; PX 8302 at 35:3-36:1 (Lien Dep.). Information provided by consulting firms, forecasting supply and demand, are similarly not regarded as reliable. PX 8307 at 250:14-18 (Rackers Dep.).