This document is available in two formats: this web page (for browsing content) and PDF (comparable to original document formatting). To view the PDF you will need Acrobat Reader, which may be downloaded from the Adobe site. For an official signed copy, please contact the Antitrust Documents Group.








Civil Action No.: 99-1180-JTM



Plaintiff United States submits this Reply in Support of its Motion for Reconsideration of the Court's September 28, 2000 Order Granting Defendants' Motion to Review Magistrate Judge Humphreys' Decision of June 13, 2000 ("September 28 Order"). The issue presented here is not whether American agrees with Professor Hovenkamp's analysis of relevant costs -- clearly it does not. Instead, the issue is whether Professor Hovenkamp has consistently maintained that the ability of an airline to shift aircraft from one market to another means that aircraft costs are variable (or "avoidable") on a route where predation is alleged. The answer, based on Professor Hovenkamp's writings since at least 1986, and confirmed by his declaration, is clearly "yes."

American prevailed on its Motion to Review, and the United States no longer seeks the return of the document which led to filing of that motion; the United States merely seeks a correction to the public record. Professor Hovenkamp's declaration establishes the existence of the error in the record and, other than claiming that the declaration is "ambiguous," American does not dispute the contents of the declaration. American's refusal to accept the sworn statements of Professor Hovenkamp concerning the meaning of the treatise which he authored serves no purpose but to prolong the false impression created by American in its Motion to Review and supporting papers.


  1. Reconsideration is Necessary to Correct a Factual Error Contained in the Court's Order

"It is the general rule that a motion to reconsider is the opportunity for the court to correct manifest errors of law or of fact and to review newly discovered evidence." Taliaferro v. Kansas City, Kansas, 128 F.R.D. 675, 677 (D. Kan. 1989). "A motion to reconsider is proper when the court has obviously misapprehended a party's position, the facts or the law, or has decided issues outside of those presented in the original motion." Voelkel v. General Motors Corp., 846 F. Supp. 1482, 1483 (D. Kan. 1994). The United States properly filed its Motion for Reconsideration to provide the Court with an opportunity to correct the following factually- erroneous statement: "that after writing the letter and consulting with the government, Hovenkamp revised the treatise, adding a footnote which suggested for the first time that the entire cost of an airplane (in addition to depreciation and obsolescence) should be considered a variable cost." September 28 Order at 2. Consistent with the standard applicable to motions for reconsideration, the United States simply seeks to correct a "manifest error of fact," now part of the public record, which implies that Professor Hovenkamp revised his position following his consultation with the United States.

    1. American Promoted and Continues to Foster this Error through Misstatements Concerning Professor Hovenkamp's Antitrust Law Treatise

The error that is the subject of this motion was promoted by a series of misstatements in American's Reply Memorandum in Support of American's Motion to Review Magistrate Judge Humphreys' June 13, 2000 Order Granting Plaintiff's Motion for Protective Order ("July 26 Reply"). The July 26 Reply, a pleading to which the United States had no opportunity to respond, contained the following mischaracterizations of various editions of and supplements to Professor Hovenkamp's treatise, 3 P. Areeda & H. Hovenkamp, Antitrust Law:

  • "While still retained as the DOJ's 'litigative expert,' [Professor Hovenkamp] revised the next supplement to Antitrust Law, to suggest, for the first time, that the entire cost of owning an aircraft should be considered a variable cost rather than a fixed cost." July 26 Reply at 1 (emphasis in original).
  • "It is only in his 1999 Supplement that [Professor Hovenkamp] adds a footnote that for the first time asserts that the ownership cost or cost of aircraft capital should be considered to be variable because an airplane can be moved from one route to another." July 26 Reply at 4, n.2.
  • "Professor Hovenkamp's 1999 Supplement is the first place he ever expressed the view that aircraft ownership cost is a variable cost because the plane can be moved from one route to another." July 26 Reply at 3.
  • "The idea that the cost of owning a plane is variable or 'avoidable' because you can move the plane from one route to another does not appear in the 1986 version of Antitrust Law or in any version or supplement until 1999." July 26 Reply at 3.
  • "In the 1986 version of his treatise, Professor Hovenkamp offered three different types of alleged variable costs associated with the operation of an aircraft: (1) depreciation; (2) obsolescence; and (3) opportunity costs in the form of revenues that could have been earned on another route. (Citation omitted.) He included no cost of ownership or capital cost because a plane could be redeployed from one route to another." July 26 Reply at 3, n. 2.
  • "By 1996, Professor Hovenkamp had abandoned the idea that opportunity costs in the form of foregone revenue should be properly considered as variable costs in evaluating claims of predation." July 26 Reply at 4, n.2.

The most obvious misstatement contained in the July 26 Reply (and unfortunately the one adopted by the Court in the September 28 Order) claims: "It is only in his 1999 Supplement that [Professor Hovenkamp] adds a footnote that for the first time asserts that the ownership cost or cost of aircraft capital should be considered to be variable because an airplane can be moved from one route to another." In fact, the relevant text of ¶740 of the 1999 Supplement contains the following language: "When the question is whether the airline's rate structure as a whole is below relevant costs, then the airplane itself must be considered a variable cost item. First, it is subject to use depreciation in proportion to the number of hours it flies. Second, its cost is avoidable because it can readily be transferred to another market if not needed in the market in question." As Professor Hovenkamp explained in his declaration, he used the fact that aircraft can be shifted or transferred from one route to another to explain why in some circumstances aircraft cost should be defined as variable; just as in his earlier statements he referenced depreciation and opportunity costs to explain why investment in shifted aircraft is variable. See Appendix, Ex. 1, Hovenkamp Decl. at 4:¶10.

The consistency between the view expressed by Professor Hovenkamp concerning the appropriate characterization of aircraft costs in the 1999 Supplement to Antitrust Law and the view expressed by Professor Areeda and Professor Hovenkamp as early as 1986 is attested to by Professor Hovenkamp's declaration, id. at 2:¶5, and by the following excerpts from the 1986 and 1999 Supplements:

1986(1) 1999(2)
"As another example in which a superficial definition of the relevant variable costs would be wrong, suppose that an airline shifts planes to a route on which a rival complains of predatory pricing." "Second, its cost is avoidable because it can readily be transferred to another market if not needed in the market in question."
"Put another way, the defendant expanded capacity on the route in question in order to lower his price there, the cost of doing so -- in the form of additional capacity -- is relevant to the appraisal of his price there." "When the question is whether the airline's rate structure as a whole is below relevant costs, then the airplane itself must be considered a variable cost item."

American supports its argument that Professor Hovenkamp's pre-1997 reference to "opportunity cost" differs from his 1999 reference to aircraft costs by stating: "This [opportunity cost] is quite different from the capital cost of the aircraft." Opposition at 5. American does not explain what difference it perceives, either in magnitude or concept. In fact, the question is, when an airline shifts an aircraft onto a route for allegedly predatory reasons, whether the value of the aircraft should be considered, or ignored. If considered, its value may be determined by looking to its capital costs or its opportunity costs (i.e., what else it could be used for).(3) "Capital cost," as such, is not a term used by Professor Hovenkamp in ¶740 of his 1999 Supplement, and American does not define it. The reference in the treatise to aircraft costs follows a discussion concerning "avoidable" costs -- costs which should be considered variable if the alleged predator could have avoided incurring those costs by not taking the allegedly predatory action. As Professor Hovenkamp explains in his declaration, he employed the term "avoidable" in 1999 because it was used in an article by Professor William Baumol, Predation and the Logic of the Average Variable Cost Test, 39 J.L.&Econ. (1996). Appendix, Ex. 1, Hovenkamp Decl. 4:¶10.(4)

    1. American Mistakenly Implies that Professor Hovenkamp Revised His Views Regarding the Variable Nature of Aircraft Costs in Reaction to Continental Airlines v. American Airlines

American claims that Professor Hovenkamp "expressly rejected the suggestion to treat opportunity costs . . . as short run variable costs for purposes of the average variable cost test" in his 1996 Revised Edition of Antitrust Law. Memorandum in Opposition to Plaintiff's Motion for Reconsideration ("Opposition") at p. 5. American further claims that Professor Hovenkamp's alleged "rejection" of opportunity costs arose from his agreement with the decision of the United States District Court for the Southern District of Texas in Continental Airlines v. American Airlines, 824 F. Supp. 689, 701 (S.D. Tex. 1993). After quoting Professor Hovenkamp's reference to the Continental Airlines decision, American states: "In view of this discussion, it was not surprising that Professor Hovenkamp's 1996 Revised Edition dropped the footnote about airline variable costs that the treatise had carried since 1986 . . . ." Opposition at p. 6.

This attempt to imply that Professor Hovenkamp dropped the footnote because of the Continental Airlines decision is simply wrong. First, Professor Hovenkamp did not drop "the footnote about airline variable costs" when he first referenced the Continental Airlines decision in his 1994 Supplement to Antitrust Law; he retained the footnote in the 1994, 1995 and 1996 Supplements.(5) Second, the argument is at odds with Professor Hovenkamp's sworn statement. Appendix, Ex. 1, Hovenkamp Decl. 3:¶ 7. As Professor Hovenkamp explains, space limitations and not any revision to his opinion, led to the dropping of the footnote.(6)

  1. American Improperly Uses its Opposition to Advance Misleading Arguments Concerning the Legal Standards Applicable in This Case

Rather than accept Professor Hovenkamp's declaration and permit a renowned scholar to correct a misstatement in the public record concerning his authoritative treatise, American uses this discovery motion to make misleading arguments concerning the legal standards applicable in this case. Not content to argue the merits of its opposition, American takes the opportunity in footnote 2 of its opposition to argue to the Court that "opportunity costs should not be considered a variable cost for the purposes of the legal test for predation."

In support of its claim regarding opportunity costs, American quotes deposition testimony of Professor Maher, one of the testifying experts retained by the United States, incompletely and out of context. While acknowledging the practical difficulties in many attempts to measure opportunity costs, Professor Maher further testified that "American provides measures of opportunity costs and understands measures of opportunity costs in its system." See Appendix, Ex. 6, Maher Dep. Tr. at 103:10 - 12. Professor Maher's testimony as a whole does not support American's argument that "[o]pportunity costs are inherently so ephemeral and speculative that their inclusion as a cost would render the average variable costs test unworkable as a practical matter." Opposition at p. 6, no. 2.

Even more surprising is the inconsistency between American's argument that and its method of doing business. American apparently insists that if it shifts airplanes to a route to drive out a low cost competitor, those aircraft should be considered costless to American in analyzing whether its actions were below cost. Yet, American's own internal accounting system, used in making decisions to add or subtract aircraft to a route, carefully takes account of the ownership or opportunity costs of aircraft -- probably because it makes eminent sense to do so, and because they wouldn't stay in business for long if they acted as if aircraft are free. In other words, American's argument -- that the "foregone revenues" or "opportunity costs" of shifting aircraft from one route to another should be ignored under a price-cost test -- contrasts sharply with its own business practice and with the benchmarks that American uses to evaluate the economic impact of exiting a hub or a route. See Appendix, Ex. 7, Response to Interrogatory 4 of CID No. 15982 (filed under seal). Testifying experts retained by American, moreover, appear to believe that it is appropriate to consider opportunity costs.(7) According to Professor Baumol, "[e]conomists agree that the type of sacrifice that they call 'opportunity cost' is a legitimate part of any cost calculation. Indeed, they (including myself) assert that any cost calculation that totally ignores the opportunity cost component is likely to be illegitimate." William Baumol, Predation and the Logic of the Average Variable Cost Test, 39 J.L.&Econ. 49, 69 (1996).

American's use of this discovery issue to argue its substantive case is even more obvious in footnote 3 of its Opposition, in which American previews its motion for summary judgment. American claims that the issue of whether aircraft ownership costs should be treated as variable or fixed in applying the average variable cost test will not make any difference in this case "because, even if aircraft ownership costs are treated as variable costs . . . , American's revenues on each of the routes in question during the alleged period of predation still comfortably exceeded American's average variable costs . . . ." Suffice it to say that there is significant dispute on the issue of relevant variable (avoidable) costs and American faces substantial contrary evidence in the record. Indeed, in 1993, American's Senior Vice President for Marketing, Michael Gunn, wrote to Congressman Dan Glickman and explained that American was disappointed with the results of its "low-fare pricing test in the Dallas/Fort Worth - Wichita market." See Appendix, Exhibit 12, April 5, 1993, Letter from Mr. Gunn to Congressman Glickman. Mr. Gunn's letter includes a chart showing that the revenue per passenger for January and February of 1993 on the DFW - ICT route was $93 and $94 respectively. According to the letter, "the DFW-ICT flights have consistently lost money. ... In fact, the January and February revenues in this market dropped below variable costs." It is difficult to understand how American can claim that a $60 average price in 1996 was above variable cost, when American told Congressman Glickman (with a copy to the Wichita Airport Authority) in 1993 that a $93 fare resulted in revenues below variable cost.

  1. The Court Should Consider Professor Hovenkamp's Declaration in Deciding this Motion

American argues that the "Hovenkamp declaration may not be presented to the Court at this late date or serve as the basis for modifying the Court's prior findings" because the United States had prior opportunities "to offer evidence to support its claim" that Professor Hovenkamp did not change his views of aircraft costs between 1986 and 1999. Prior to the September 28 Order, the United States was arguing the issues of whether the deliberative process privilege applied to the Hovenkamp Letter and whether it had waived that privilege by producing the letter to American. The United States had already produced declarations to support its positions regarding the applicability of the privilege and the inadvertence of production in its briefing of these issues before Magistrate Judge Humphreys. The United States responded to American's collateral arguments regarding the alleged change in Professor Hovenkamp's views regarding aircraft costs by producing the 1986 and 1996 Supplements to Antitrust Law. In light of the clear language contained in supplements to Antitrust Law dating back at least as far as 1986, the United States believed that the treatise excerpts themselves constituted the best evidence to support the argument that Professor Hovenkamp's views on aircraft costs pre-dated Professor Hovenkamp's consulting relationship with the Department of Justice.

Had the Court simply granted American's Motion to Review and overruled Magistrate Judge Humphreys' issuance of a protective order, the United States would not have filed this Motion to Reconsider. After Professor Hovenkamp received a copy of the September 28 Order adopting the misstatements made by American in its reply brief, however, Professor Hovenkamp volunteered to prepare a declaration in order to correct the mistaken impression created by the Court's order. See Declaration of Roger Fones. American's attempt to exclude Professor Hovenkamp's declaration, when the United States no longer seeks the return of the Hovenkamp Letter and seeks solely to correct a factual error in the September 28 Order, does a disservice to this Court, in its role as seeker of the truth, and to Professor Hovenkamp, acknowledged by the Court as the editor of the leading treatise in the field of antitrust law.

Respectfully submitted on the 17th day of November

Plaintiff United States

By: __________/"s"/__________
Rebekah J. French
601 D. Street N.W., Suite 1700
Washington, D.C. 20530
Tel. (202) 307-5779
Fax (202) 353-8856


1. P. Areeda & H. Hovenkamp, Antitrust Law ¶714.6, n.46 (Supp. 1986). See Appendix, Ex. 2.

2. P. Areeda & H. Hovenkamp, Antitrust Law ¶740( Supp. 1999). See Appendix, Ex. 3.

3. Normally, these two cost measures would be closely related, because the value of a business asset is usually a function of what it can be used for.

4. American does not contend that this characterization is improper, perhaps because it has retained Professor Baumol as an expert in this case.

5. Compare P. Areeda & H. Hovenkamp, Antitrust Law ¶714.6, n.51 and ¶715.2f (Supp. 1994, 1996). See Appendix, Ex. 4, and Ex. 5.

6. While it may be true that Professor Hovenkamp accepted the reasoning used in the Continental Airlines decision as it related to the incremental cost of supplying a single seat on an already-scheduled flight, he distinguishes between such a case and a situation where an airline shifts an entire aircraft from one route to another. 3 P. Areeda & H. Hovenkamp, Antitrust Law ¶740 (Supp. 1999).

7. See Appendix, Ex. 8, Kalt Dep. Tr. at 330: 2 - 10 (redacted from publicly filed version of brief.); Ex. 9, Baumol Dep. Tr. 274: 14 - 23 (redacted from publicly filed version of brief.); Ex. 10, Ordover Initial Expert Report at 26: ¶ 53 (redacted from publicly filed version of brief.).