Free Proposed Findings of Uncontroverted Fact - District Court of Federal Claims - federal

Case 1:03-cv-00288-EJD

Document 29

Filed 05/30/2006

Page 1 of 17

IN THE UNITED STATES COURT OF FEDERAL CLAIMS CHEVRON U.S.A. INC., And TEXACO INC., And TEXACO DOWNSTREAM LLC, Plaintiffs, v. THE UNITED STATES, Defendant. ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) )

No. 03-288C (Chief Judge Damich)

PLAINTIFFS' PROPOSED FINDINGS OF FACT Pursuant to RCFC 56(h)(2), Plaintiffs, Chevron U.S.A. Inc., Texaco Inc., and Texaco Downstream LLC (collectively, "Chevron"), respectfully submit their Proposed Findings of Fact. I. THE PARTIES 1. DESC1 is the principal purchaser of military fuel for the Department of Defense

("DOD"). It is the single largest purchaser of fuel in the world.2 As relevant here, DESC principally purchased unique types of military jet fuel. Second Amended Complaint ¶ 8; App. 651-1074.3

DESC previously was known as the Defense Fuel Supply Center ("DFSC"). For simplicity, "DESC" is used throughout.
2

1

Dimensions, Defense Logistics Agency News Magazine (1999 Almanac ed.), available at http//www.dla.mil/dimensions/almanac/desc.htm. Chevron Products Company, Aviation Fuels Technical Review 13 (2000) (military "maintain[s] separate specifications for jet fuel [because of] operational and logistical differences between the military and civilian systems and the additional demands high-performance jet fighter engines place on the fuel").

3

Case 1:03-cv-00288-EJD

Document 29

Filed 05/30/2006

Page 2 of 17

2.

Chevron was a supplier of military fuel to DESC. Chevron maintained long-term

supply relationships with DESC and committed substantial, long-term capacity to refining military fuel. App. 651-1074. DESC purchased military fuel from Chevron, as it did from essentially all military fuel suppliers, under long-term contracts typically lasting one year. Second Amended Complaint ¶ 8; App. 651-1074. During the period relevant here, Chevron and DESC entered 39 contracts providing for the purchase of different types of military fuel at different locations in the United States. Second Amended Complaint ¶¶ 34-36; App. 651-1074. II. DESC'S MILITARY FUEL PRICES AND ITS STATED INTENT TO PAY FAIR MARKET VALUE FOR FUEL 3. DESC's military fuel contracts contained adjustable prices that permitted DESC

to set military fuel prices monthly or weekly during the contract term. App. 651-1074. 4. As relevant to DESC's motion, DESC set fuel prices under its military fuel

contracts using a compilation of petroleum sales data published by the Department of Energy ("DOE") known as the Petroleum Marketing Monthly ("PMM").4 App. 711-14. The PMM reports a monthly average sales figure for specified fuels, such as jet fuel or diesel fuel, for five regions known as Petroleum Administration for Defense Districts ("PADDs"). Barrett Ref. Corp. v. United States, 42 Fed. Cl. 128, 130 n.6, 131 (1998), aff'd in part and rev'd in part, 242 F.3d 1055 (Fed. Cir. 2001). As the PMM moved upward or downward each month, DESC's fuel prices also moved upward or downward. App. 711-14.5 5. DESC's Deputy Director of Contracting, Edward Biddle, stated in a 1993

memorandum that the purpose of DESC's adjustable prices was to "ensure[] that the sale will be
4

DESC also used other compilations of sales data to set fuel prices. However, in its motion, DESC offers no evidence concerning these other compilations or their relationship to the market for military fuel. Because DESC fails to meet its burden to offer any evidence concerning these other compilations, or otherwise move with respect to them, Chevron does not address them specifically in its response and limits its discussion to the PMM. The PMM is designed for Congressional use in developing broad energy policy. Barrett, 42 Fed. Cl. at 131.

5

2

Case 1:03-cv-00288-EJD

Document 29

Filed 05/30/2006

Page 3 of 17

at the market price, which is a fair and reasonable price. Not only is the government protected, but the contractor is protected by knowing that its price will always be related to the price it could obtain in the commercial marketplace." App. 4 (emphasis added). 6. DESC's Director of the Office of Market Research and Analysis, Lawrence Ervin,

testified similarly before this Court in 1998 that DESC used adjustable prices because: "[W]e wanted our prices to reflect, as closely as possible, market levels for the same or similar commercial fuels in the marketplace. . . . We wanted the price to go up and also down when the market prices of similar fuel went [up or] down." App. 14; see also App. 62 ("the EPA [economic price adjustment] reference selected is [intended to be] reflective of changes in commercial market prices"); App. 69 ("the EPA reference is designed to adjust the award price in concert with changes in going market price levels;" "the D[E]SC Office of Market Research and Analysis endeavors to select the reference that most accurately reflects movements of going market prices"). 7. Consistent with this purpose, DESC's pricing clause is titled "Economic Price

Adjustment ­ Published Market Price." App. 9. Among its various iterations over time, the clause states that "[t]he market price [used to set military fuel prices] is derived from quotes, assessments, or sales prices in the market place" and further provides that military fuel prices are "to reflect market conditions." App. 9 (emphasis added).6 It is therefore unsurprising that in Barrett the United States Court of Appeals for the Federal Circuit expressly found that DESC's adjustable pricing clause embodied a promise to "pay at least fair market value" for fuel. 242 F.3d at 1060.

See also App. 87 (Mr. Ervin stating that, pursuant to DESC's pricing clause, "replacement of the escalation reference is called for if the reference `consistently fails to reflect market value'"); App. 85 (Mr. Ervin stating that through its pricing clause "DESC assumes the risk of higher product prices").

6

3

Case 1:03-cv-00288-EJD

Document 29

Filed 05/30/2006

Page 4 of 17

III.

DESC'S CONFLICTING INTERNAL ASSESSMENTS AND PUBLIC STATEMENTS ABOUT ITS MILITARY FUEL PRICES 8. Outwardly, DESC maintained that the PMM was the "gold standard" for

measuring fair market value. DESC wrote to its parent organization, the Defense Logistics Agency ("DLA"), that its PMM "references are virtually unimpeachable from an accuracy standpoint, since they don't just `track' the market ­ they are the market." App. 98 (emphasis added).7 In response to a supplier's objection to the PMM, DESC similarly wrote that the PMM "consistently reflects actual market prices," App. 96, and to another objecting supplier DESC wrote that the PMM is "a reliable indicator of price movements in the petroleum products market." App. 91. As late as 1998, DESC's Mr. Ervin testified before this Court that the PMM is "the benchmark against which we . . . compare everything domestically." App. 29. 9. DESC's statements to DLA, its suppliers and this Court stand in stark contrast to

DESC's expert economists' assessment of the PMM. In 1986, DESC's then-Director of the Office of Market Research and Analysis, Mr. Christopher Lee, undertook an assessment of the PMM to compare its performance with commonly used commercial price references for fuel, such as Platts and OPIS (which DLA subsequently directed DESC to use in place of the PMM). Referring to Platts and OPIS as "other references,"8 Mr. Lee, who holds a Doctorate in Economics,9 stated: "Yet our analysis of the movement of the PMM as compared to other references as used for interim price adjustments indicates that the PMM may have moves as
7

DESC's public espousal of the PMM is unsurprising, given DESC's acknowledgment that "DOE normally will work with us to see that our needs are satisfied as we are one of the primary users of PMM." App. 100.

In the 1986 assessment, Mr. Lee uses the phrase "other references as used for interim price adjustments." App. 103. DESC used its interim price adjustments to set prices on an interim basis at the time fuel was delivered. App. 120. DESC set final prices when the PMM was published three months after the fact ­ i.e., three months after the fuel was delivered. App. 121. DESC's interim references, as identified in its adjustable pricing clause, were Platts and OPIS. App. 120.
9

8

App. 111.

4

Case 1:03-cv-00288-EJD

Document 29

Filed 05/30/2006

Page 5 of 17

much as one or two cents per gallon out of step with such interim references." App. 103 (emphasis added). In contrast, DESC required its suppliers to submit bids to the one tenthousandth of a cent ($0.000001). See, e.g., App. 865 (reflecting a bid price of $0.535660). 10. In 1987, DESC's Office of Market Research and Analysis undertook another

study "comparing PMM versus Platts and OPIS." App. 134. That study supported using the commercial price references Platts and OPIS to make price adjustments ­ not the PMM. App. 134. When DLA ultimately learned of DESC's 1987 study ­ albeit seven years later ­ DLA directed DESC to stop using the PMM to set prices and to use Platts and OPIS instead. 11. Mr. Ervin explains how DESC's 1987 study led to DLA's directive to stop using

the PMM in 1994: I was sitting in a room when the decision happened and it was made in response to some pressure from headquarters to do things differently. So the analysis is based on, basically is the analysis in the '87 study. Some people who have read the '87 studies [stated] that they would draw different conclusions than what was presented at the end of the study, certainly a close call. Analysis would have supported that as well so it's really the '87 study which is the basis for that move [in 1994 from PMM to Platts and OPIS]. App. 134.10 12. The 1987 study was written by the Office of Market Research and Analysis with

the objective of justifying continued use of its "gold standard," the PMM. App. 595-602.11 While the all-important statistical analysis behind the study has not yet been produced, the admissions against interest in the narrative of the study are compelling. DESC's expert

10

DESC proffered Mr. Ervin as a witness in response to a Notice of Deposition served under Rule 30(b)(6) in La Gloria Oil & Gas Co. v. United States, No. 02-465C (Fed. Cl.). App. 141. The notice directed DESC to designate a witness to address the existence of documents requested in the plaintiff's requests for production of documents. App. 141. Because, pursuant to the terms of the notice, the deposition was limited to the existence of responsive documents, the plaintiff's ability to inquire into the substance of the conclusions of the documents was limited. App. 141. That inquiry must await further discovery. While this document is attached to DESC's motion, DESC does not specifically identify the document as the 1987 study referred to by Mr. Ervin in his deposition. Final confirmation of this fact awaits further discovery.

11

5

Case 1:03-cv-00288-EJD

Document 29

Filed 05/30/2006

Page 6 of 17

economists admit that using Platts and OPIS to set prices instead of the PMM "would cost DESC money" and concede that "[t]here is market risk associated with the move away from escalation on actual sales [PMM] particularly with respect to jet kerosene [military jet fuel]." App. 599; App. 602.12 DESC's economists additionally recount in the 1987 study how its suppliers had complained that, as a result of DESC's use of the PMM to set prices after fuel is delivered, "`we never know if we have made or lost money when you (D[E]SC) lift a cargo.'" App. 597 (emphasis added); see also App. 595 (cover memorandum to the 1987 study stating with respect to the proposed change in using the PMM, "The issue has been raised several times in recent years, formally and informally, by contractors"). 13. Once DLA directed DESC to stop using the PMM in 1994 and to set prices using

Platts and OPIS instead, DESC made the extraordinary admission that prices it previously had set using the PMM so departed from market value that they could not be compared to prices set based on Platts and OPIS. DESC stated: "Direct contract price comparisons are not possible due to the switch from the monthly interim-final (PMM) escalation method to a weekly market price escalation basis [using Platts and OPIS]." App. 144 (emphasis added).13 DESC's admission is unsurprising, given that DESC had concluded elsewhere that changing from the PMM to commercial references, such as Platts and OPIS, "could significantly influence contract prices," App. 98, and further found, in response to a proposal from one of its largest suppliers to

12

DESC elsewhere has asserted that its statements about the PMM's relationship to the marketplace are, in reality, criticism of DESC's then-interim price references Platts and OPIS and not criticisms of the PMM at all. DESC's assertion is nothing short of startling, given that DESC currently uses Platts and OPIS to set fuel prices.

13

For each procurement, DESC prepares a "Price Analysis Memorandum" to determine its price objectives during contract negotiations. See, e.g., App. 144, 150, 154, 157, 163. In determining its price objectives for the procurement, DESC typically considers, under the heading "Price History," the price it is paying under its current, existing contracts. See, e.g., App. 144, 151, 155, 157, 163, 167-68. However, as noted above, when DESC stopped using the PMM and switched to using Platts and OPIS instead, it stated that its usual practice of making price comparisons with its existing contracts was "not possible." App. 144.

6

Case 1:03-cv-00288-EJD

Document 29

Filed 05/30/2006

Page 7 of 17

stop using the PMM, that its "PMM JP-5" jet fuel prices were ten cents a gallon higher, then ten cents lower, and yet again ten cents higher than Platts ­ all within the same year. App. 182.14 IV. THE NON-NEGOTIABILITY OF DESC'S MILITARY FUEL PRICES 14. Despite DESC's admission that "comparisons are not possible" between the

prices it set using the PMM and those set using commercial market references such as Platts and OPIS, DESC had mandated that "[a]ll offerors are required to accept a method of price adjustment based on . . . the Petroleum Marketing Monthly (PMM)." App. 195; see also App. 199 (stating that DESC "would not consider" an offer unless objection to the PMM was withdrawn). As Mr. Ervin testified before this Court, DESC used its "leverage as a buyer" as the sole purchaser of military fuel to "improve the competition" for DESC's business by requiring use of the PMM. App. 17. 15. Accordingly, even though suppliers objected to DESC's use of the PMM as

"coerced," App. 201; App. 204, the use of the PMM to set prices was "not negotiable as a matter of policy." App. 205. DESC's procurement authorizations expressly stated under the heading "POTENTIAL PROBLEMS" that "common escalation [using the PMM] is non-negotiable." App. 206-278; Navajo Ref. Co., L.P. v. United States, 58 Fed. Cl. 200, 214 (2003) ("DESC explicitly stated in its pre-negotiation briefing memoranda that it deemed its price adjustment clause to be non-negotiable"); see also Barrett Ref. Corp. v. United States, 45 Fed. Cl. 166, 171 (1999) (supplier "had no control over use of PMM EPA clause, and indeed attempted, unsuccessfully, to persuade the agency not to use it"), aff'd in part and rev'd in part, 242 F.3d 1055 (Fed. Cir. 2001). Evidencing the futility of objecting to DESC's pricing clause, even after a supplier filed a claim in 1988 challenging DESC's use of the PMM, DESC continued to use the
14

The assessments of the PMM by DESC's expert economists, discussed above, together with DLA's direction to stop using the PMM based on DESC's 1987 study, draw into serious question the credibility of Mr. Ervin's subsequent cross-examination testimony in this Court in 1998 that he never identified any "problem with using the PMM" to set military fuel prices other than "billing" issues. App. 36-37.

7

Case 1:03-cv-00288-EJD

Document 29

Filed 05/30/2006

Page 8 of 17

PMM for six more years, until it was directed to use Platts and OPIS instead. MAPCO Alaska Petroleum, Inc. v. United States, 27 Fed. Cl. 205, 407 (1992). V. THE FAILURE OF DESC'S MILITARY FUEL PRICES TO REFLECT FAIR MARKET VALUE 16. While Chevron knew that DESC's use of its adjustable pricing clause and the

PMM were non-negotiable, what Chevron did not know was that DOE had not designed the PMM to be used to set fuel prices or to reflect market value. As Dr. John Cook, Director of the Petroleum Marketing Division of the Department of Energy's Energy Information Administration (which publishes the PMM), has since testified and as this Court found: "[T]he PMM is not intended to be an index that is used for setting prices." Barrett, 42 Fed. Cl. at 131. Thus, while Chevron "entered the fuel contracts with the intent that the prices DESC set would reflect fair market value[,]" it could not have been more wrong. App. 638. See also App. 640.15 17. An assessment of the PMM as an established market price is set forth in the

Expert Report of Joseph Kalt16 and Peter Killen.17 In their report, Kalt and Killen conclude that DESC's "PMM-based adjustment clauses failed to provide an effective measure of established market prices or the fair market value of military fuels." App. 611.

15

Chevron has submitted the declarations of two former employees, Timothy Taigen and Kenneth Johnson, with respect to the issue of Chevron's intent in entering the contracts. App. 638-41. For companies and time periods not covered by their declarations, discovery of additional former employees pursuant to Chevron's Rule 56 Motion will be required. However, for purposes of this Opposition, the three companies which executed fuel supply contracts with DESC will be generally referred to as Chevron rather than individually.

Joseph Kalt is the Ford Foundation Professor of International Political Economy at the John F. Kennedy School of Government, Harvard University. App. 609-10.
17

16

Peter Killen is a Senior Advisor in the Houston offices of the firm of Muse Stancil & Company, a global consulting firm specializing in the energy industry. He holds a B.S. in Chemical Engineering and is a Registered Professional Engineer. App. 610.

8

Case 1:03-cv-00288-EJD

Document 29

Filed 05/30/2006

Page 9 of 17

18.

The following chart, drawn from Kalt and Killen's report, dramatically illustrates

the PMM's departure from the marketplace:
Figure 3 PMM ADJUSTMENT FAILS TO TRACK THE MARKET: DIFFERENCES BETWEEN MONTHLY CHANGES IN MARKET-BASED ESCALATOR AND PMM-BASED ESCALATOR ARE UNSTABLE IN MAGNITUDE AND DIRECTION February 1983 ­ March 1995
15¢

10¢

5¢

Cents/Gallon

0¢ 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995

-5¢

-10¢

19.

The chart illustrates how many cents per gallon the marketplace changed more or

less than the PMM during the years DESC used the PMM. App. 634; App. 625.18 If the change in the PMM was equal to the change in the marketplace, the bars in the chart would be zero or nonexistent. Thus, as the chart shows, in July of 1990 the marketplace increased by 13 cents a gallon more than the PMM. Indeed, not only did the change in the PMM depart substantially
18

As Kalt and Killen explain, it is the change in price levels of the PMM and the marketplace, rather than the absolute prices levels themselves, that are relevant here, because "the PMM is used by the DESC to index changes in price levels, rather than the price levels themselves." App. 624 (emphasis original).

9

Case 1:03-cv-00288-EJD

Document 29

Filed 05/30/2006

Page 10 of 17

from the change in marketplace, but also the PMM's relationship to the marketplace was fundamentally unstable, inasmuch as the change in the PMM substantially understated the change in the market one month and then substantially overstated the change in the market in the next. As Kalt and Killen explain: "The fact that we see significant bars indicates that the relationship between the movements of the two series [the PMM and marketplace] is notably unstable." App. 625. Thus, as Kalt and Killen demonstrate, the chart shows the PMM's relationship with the marketplace to be "extremely volatile," with the change in the PMM varying from the change in the market place from approximately -7 cents to approximately +13 cents during the relevant period. App. 625; App. 634.19 20. Equally compelling, Kalt and Killen explain that the PMM and the marketplace

"frequently actually move in different directions [e.g., the market goes up and the PMM goes down] in a starkly erratic fashion." App. 625. 21. The next chart, again drawn from Kalt and Killen's report, illustrates on an annual

basis how much the change in the PMM departed from the change in the marketplace:

19

As Kalt and Killen further explain, the fact the PMM overstated the change in market does not mean that Chevron was overpaid. Given the PMM's random relationship with the market, "the winning bidder could readily [and erroneously] forecast that the Defendant's PMM index in any particular contract period would turn out to yield payments in the next contract period in excess of fair market value ­ pushing the optimistic bidder's base price bid downward." App. 613; see also App. 627-31.

10

Case 1:03-cv-00288-EJD

Document 29

Filed 05/30/2006

Page 11 of 17

Figure 6 THE AVERAGE MONTHLY DIFFERENCE BETWEEN CHANGES IN MARKET-BASED AND PMM-BASED ESCALATORS OVER THE COURSE OF A CONTRACT IS HIGHLY UNSTABLE
6¢

5¢ 1986 Contract 4¢ 1988 Contract 1989 Contract 1991 Contract

3¢

1985 Contract*

2¢

1¢ Cents/Gallon

0¢ 1990 Contract -1¢ 1984 Contract -2¢ 1992 Contract 1993 Contract 1994 Contract*

-3¢

1987 Contract

-4¢
*Note: 1985 Contract assumed to have same reference date as 1986 contract; 1994 Contract assumed to have same reference date as 1993 Contract.

22.

The height of the bars shows for each year how many cents per gallon, on

average, the change in the PMM departed from the change in the marketplace. App. 637; App. 626. Once again, not only is the amount of the error significant (in 1986 the PMM understated the marketplace on average 4 cents a gallon), but both the amount and the direction of the error are erratic (in 1987 the PMM overstated the market by 3 cents a gallon). 23. This volatile nature of the PMM is significant, because it made it essentially

impossible for a bidder to predict correctly how the PMM would relate to the marketplace and, therefore, to bid in a way that would provide some meaningful protection from the PMM. As Kalt and Killen explain: "The deviations between changes in the Market-based escalator and the PMM-based escalator are quite volatile . . . This volatility confounds rational prediction by 11

Case 1:03-cv-00288-EJD

Document 29

Filed 05/30/2006

Page 12 of 17

which a rational bidder might reliably adjust bid prices so as to ensure receipt of no less than market value under a given period's fuel contract." App. 626. 24. Kalt and Killen state that there are a number of reasons why the PMM does not

reflect the marketplace. Among them, Kalt and Killen explain, "the PMM data were not designed or intended to be used as an index." App. 611. They also show that the PMM suffers from a statistical flaw known as an "index number problem." App. 616-19. This statistical flaw, Kalt and Killen state, results in the PMM erroneously reporting changes in the volume of fuel sold as changes in the price of fuel. Because of the PMM's index number problem, Kalt and Killen conclude: PMM data do not contain accurate information needed to understand how the prices of specific products, such as a particular grade of military jet fuel, may have changed over time. An index that relies on these data, such as the DESC's PMM-based escalator, is readily susceptible to situations in which the index reports a price increase (or decrease) even though the actual market value of the fuel has declined (or risen) in the marketplace. App. 617-18.20 25. Significantly, Kalt and Killen point out that, unlike DOE here, other federal

agencies that publish indexes intended to be used as price indexes go to great lengths to ensure that the indexes do not suffer an index number problem. Kalt and Killen state: It is important to note here that statisticians at government agencies and in other organizations often go to great lengths to address the "index number" problem when constructing indices such as the PPI and the CPI, which are used to reflect the underlying changes in price levels for specific products used by producers and consumers. Each of these indices is constructed using fixed weights from period to period such that short-term variation in the mix of products sold is not mistaken for actual changes in the levels of prices. App. 618 (footnote omitted).

20

DESC elsewhere takes issue with Kalt and Killen's analysis of the PMM's index number problem, claiming that Kalt and Killen wrongly assert that the PMM mixes different fuel products. However, the PMM index for kerosene-type jet fuel does encompass several different fuel products, including Jet-A, JP-5 and JP-8, a fact DESC does not and cannot dispute. App. 616-17.

12

Case 1:03-cv-00288-EJD

Document 29

Filed 05/30/2006

Page 13 of 17

26.

Beyond the PMM's index number problem, Kalt and Killen further conclude that

DESC's PMM-based pricing clause failed to reflect the marketplace because "the formulas used by the government to create the index did not track changes in the value of the military jet fuel." App. 611. As they explain, DESC simply used the wrong fuels, and thus the wrong PMM data, in its pricing clause. App. 619-21. 27. Thus, for the military jet fuel known as JP-4, DESC based price adjustments on

the PMM data for unleaded gasoline and commercial jet fuel, weighted 70 percent and 30 percent, respectively. App. 620. However, this combination of unleaded gasoline and commercial jet fuel does not reflect the physical characteristics or market value of JP-4. Kalt and Killen state: "The DESC's simplistic use of a 70:30 ratio of gasoline and kerosene-type jet fuel prices to proxy for JP-4 market prices is inappropriate for economic price adjustment purposes given the technical and economic considerations of how JP-4 and other refined products are actually produced." App. 620. 28. Kalt and Killen explain that a proper measure of the physical characteristics and

market value of JP-4 should not be based on unleaded gasoline and commercial jet fuel but rather on a combination of naphtha, commercial jet fuel, and diesel fuel. App. 620-21. Notably, both this Court and the Federal Circuit in Barrett also concluded, over DESC's objection, that naphtha and not unleaded gasoline was the proper fuel to use for measuring the value of JP-4. Barrett, 242 F.3d at 1061 ("However, the government does not dispute the Court of Federal Claims' finding that the naphtha-based index `provides data that more closely reflect the product [JP-4] Barrett was actually selling.'").21

21

DESC elsewhere claims that Kalt and Killen's analysis assumes that an index used to adjust prices must be an index for the identical product the government purchases. This is a misstatement of Kalt and Killen's analysis. As set forth above, Kalt and Killen's analysis of JP-4 prices uses an index based on the prices of naphtha, commercial jet fuel, and diesel fuel. App. 620-21.

13

Case 1:03-cv-00288-EJD

Document 29

Filed 05/30/2006

Page 14 of 17

29.

For these and other reasons, Kalt and Killen observe that commercial buyers and

sellers do not use the PMM as a price index. They conclude that the PMM's failure to pass "[t]his `market test' reinforces the conclusion that a PMM-based escalator of the type used by the DESC is simply not suited to represent actual movements of established market prices for any fuel, much less military jet fuel." App. 619.22 VI. DESC'S MINORITY PRICE PREFERENCES AND FUEL CONTRACT AUCTIONS 30. Beyond DESC's use of the PMM to set military fuel prices, DESC distorted the

market price for fuel in two other ways. First, DESC provided a ten-percent price preference to minority-owned suppliers. Under this price preference, DESC awarded fuel contracts to minority-owned suppliers if their bids were within ten percent of that of the otherwise successful bidder. Def. App. 7; Second Amended Complaint ¶ 31. In some years, more than twenty percent of the fuel offered (as a percent of that purchased) was subject to DESC's minority price preference. App. 279 (showing volumes offered by minority firms with the designation "SD" and total volume awarded in the procurement); App. 283 (same); Second Amended Complaint ¶ 31. 31. Second, DESC conducted an auction for fuel contracts, whereby DESC offered

fuel contracts to suppliers that agreed to match the price, or adjusted price, of other bidders. Under this auction, DESC awarded contracts to small businesses willing to match the price, or adjusted price, offered by the otherwise successful bidder. Def. App. 10-13; Second Amended Complaint ¶ 30. More than fifty percent of the fuel DESC procured commonly was set aside for award to small businesses. App. 286; Second Amended Complaint ¶ 30. Both DESC's minority price preference and its auction, by their very nature, placed downward pressure on bid prices, as
22

DESC elsewhere takes issue with Kalt and Killen's conclusions because they compare the PMM to Platts spot prices. DESC's assertion is nothing short of extraordinary in view of the fact that DESC currently uses Platts spot prices to set military fuel prices in its military fuel contracts.

14

Case 1:03-cv-00288-EJD

Document 29

Filed 05/30/2006

Page 15 of 17

non-minority and non-small businesses were forced to compete on an uneven playing field. Second Amended Complaint ¶ 32. VII. DESC'S WAIVER DEFENSE 32. As Mr. Taigen makes clear in his Declaration, Chevron entered into the contracts

with DESC containing DESC's pricing clause, because the pricing clause was non-negotiable and because Chevron had no other choice if it wanted to remain a military fuel supplier. App. 638-39. As Mr. Taigen confirms: DESC maintained that the way it set fuel prices was non-negotiable and that its fuel prices reflected fair market value throughout its contracts. In entering the fuel contracts, Chevron understood that it could not be a fuel supplier to DESC unless it accepted the way DESC set prices. Chevron continued to enter fuel contracts with DESC because Chevron sought to continue to be a supplier to DESC. By continuing to enter fuel contracts with DESC, Chevron did not . . . intend that it would be paid less than the fair market value of fuel or at any time relinquish the right to payment of fair market value. App. 638-39 (internal paragraph numbers omitted). See also App. 640-41 (Johnson Declaration ¶¶ 3-4). 33. Moreover, as Kalt and Killen explain in their Expert Report, while Chevron

entered a large number of contracts with DESC, Chevron "only had a maximum of 11 annual observations (1983 to 1994, the reference year for the last DESC contracts that used PMM-based escalators) from which to draw its inferences" about the PMM. App. 629. Eleven observations, Kalt and Killen explain, are not enough to determine a consistent bias in the PMM or assess how the PMM will perform during any specific contract period, particularly since the PMM's relationship to the marketplace was "neither stable nor predictable." App. 628. 34. As Kalt and Killen explain:

For example, over a shorter period of time, a bidder might observe a number of periods in which the PMM-based escalator is biased in such a way that it tends to under-represent upward movements in prices. However, this pattern does not permit the conclusion that the bias will always be in this same direction or even in the same direction in the next round of contracting. In fact, it would not be unusual to observe repeated runs of results that do not fairly approximate the result expected over the long run even in an otherwise unbiased game.

15

Case 1:03-cv-00288-EJD

Document 29

Filed 05/30/2006

Page 16 of 17

Consider, for example, a repeated coin flip. Assuming a fair coin, the expectation is that a flip would result in a heads result 50% of the time. And, in fact, over a large number of samples, this is undoubtedly true. However, as is the case here, over a much smaller set of flips, it is quite possible to find repeated runs of heads in apparent defiance of statistical expectations. In fact, as we contemplated illustrating our point here, we performed this experiment and ended up with a repeated run of 8 heads in a row and 15 heads out of 20 coin flips. Although we did not expect the process of coin flipping to be biased, and there is no basis for presuming the coin we used was biased, we certainly found ourselves in a situation in the short run where, had we predicted 50% tails, we would have lost our "bet." App. 629 35. Furthermore, given the PMM's random variation with respect to the marketplace,

a supplier might well be expected to "alternatively, over- or under-predict the bias," thereby further complicating a supplier's assessment of the PMM's true bias. App. 630. As Kalt and Killen explain: The typical response of a bidder to the uncertainty around the bias introduced by the DESC's use of a PMM-based escalator would be to try to forecast this bias by reference to some set of historical prices. Assuming that the deviation of PMMbased escalators from actual market price escalators was a random walk, bidders might, for example, assume that the bias observed in past periods would be present in future periods. In the absence of any better information, this may be a reasonable assumption; however, this forecasting approach does not necessarily lead to consistent after-the-fact results due to the unstable nature of the bias introduced by the DESC's use of a PMM-based escalator. Using the past relationship between the PMM-based escalator and a market-based escalator to forecast the future relationship will, alternatively, over- or under-predict the bias. For example, when the bias is high in one year, the forecast bias for the next year is likely to be too high given the limited availability of data as to the "true" relationship. App. 630. 36. Accordingly, Kalt and Killen reject DESC's assertion that, through entering a

series of contracts with DESC, Chevron must be deemed to have intended to relinquish payment of fair market value. They state: "Such assertions ignore that a company such as ChevronTexaco, based on the information available at the time that it bid, could not have understood that its bids would consistently receive less than fair market value." App. 627. 37. Chevron filed its Complaint soon after the Federal Circuit in Barrett put a final

end to DESC's public assertions that there were no remedies for its military fuel prices. As 16

Case 1:03-cv-00288-EJD

Document 29

Filed 05/30/2006

Page 17 of 17

explained in the Declaration of Todd Grubin, Vice President, Operations for Chevron U.S.A. Inc., Chevron submitted claims under the fuel contracts when it first learned that remedies were available to permit payment of fair market value in accordance with Chevron's intent in entering into the contracts. App. 642; Barrett, 242 F.3d 1055. 38. DESC threatened essentially the entire U.S. refining industry in writing with

claims of fraud should refiners pursue the same claims the Federal Circuit upheld in Barrett. App. 563-94. Respectfully submitted,

s/J. Keith Burt J. Keith Burt Mayer, Brown, Rowe & Maw LLP 1909 K Street, N.W. Washington, DC 20006 (202) 263-3208 (phone) (202) 263-5208 (fax) Counsel for Plaintiffs, Chevron U.S.A. Inc., Texaco Inc., and Texas Downstream LLC May 30, 2006

17