255 F3d 816 Cf v. Lp

255 F.3d 816 (D.C. Cir. 2001)

CF Industries, Inc., Petitioner
v.
Surface Transportation Board and United States of America, Respondents
Farmland Industries, Inc. and Koch Pipeline Company, L.P., Intervenors

No. 00-1209, 00-1213 and 00-1248

United States Court of Appeals FOR THE DISTRICT OF COLUMBIA CIRCUIT

Argued May 7, 2001
Decided July 27, 2001

On Petitions for Review of an Order of the Surface Transportation Board

Mitchell F. Hertz argued the cause for petitioners CF Industries, Inc. and Farmland Industries. With him on the briefs were Frederic L. Wood, Jeffrey O. Moreno, Scott A. Harvey, Jeffrey A. Rosen, Daryl Joseffer and Daniel T. Donovan. James D. Senger entered an appearance.

John G. Roberts, Jr. argued the cause for petitioner Koch Pipeline Company, L.P. With him on the briefs were Samuel M. Sipe, Jr., F. Michael Kail and Jonathan S. Franklin.

Theodore K. Kalick, Attorney, Surface Transportation Board, argued the cause for respondents. With him on the brief were Ellen D. Hanson, General Counsel, and John M. Nannes, Acting Assistant Attorney General, Robert P. Nicholson, Attorney, and John P. Fonte, Attorney, U.S. Department of Justice.

Frederic L .Wood, Jeffrey O. Moreno, Scott A. Harvey, Jeffrey A. Rosen, Mitchell F. Hertz, Daryl Joseffer and Daniel T. Donovan were on the brief for intervenors CF Industries, Inc. and Farmland Industries, Inc. James D. Senger entered an appearance.

Samuel M. Sipe, Jr., F. Michael Kail, John G. Roberts, Jr. and Jonathan S. Franklin were on the brief for intervenor Koch Pipeline Company L.P.

Before: Edwards, Randolph, and Garland, Circuit Judges.

Opinion for the Court filed by Circuit Judge Garland.

Garland, Circuit Judge:

1

In 1996, Koch Pipeline Company, L.P. raised shipping rates on its anhydrous ammonia pipeline. Pipeline customers CF Industries, Inc. and Farmland Industries, Inc. challenged the rate increase before the Surface Transportation Board (STB). The Board found the new rates unreasonable. In these consolidated petitions for review, Koch disputes the STB's decision to lower the rates, while CF and Farmland attack the Board's decision not to lower them still further. We affirm both decisions and deny the petitions for review.

2

* Anhydrous ammonia is a hazardous compound of nitrogen and hydrogen that is used both to manufacture fertilizers and as a direct fertilizer application. A significant amount of the compound is produced in Louisiana, Oklahoma, and Texas, and shipped to users in the Midwest. Demand is seasonal, increasing sharply during the spring planting season and to a lesser extent in the fall. During the year, shippers fill large storage terminals throughout the Midwest to ensure availability when needed. When the spring comes, the terminals are rapidly emptied through deliveries to local retailers, who in turn distribute the anhydrous ammonia to farmers for immediate application. CF Indus., Inc., No. 41685 at 2-3 (S.T.B. May 9, 2000) (Final Order).

3

The STB inherited the Interstate Commerce Commission's (ICC's) jurisdiction over interstate "transportation by pipeline ... when transporting a commodity other than water, gas, or oil." 49 U.S.C. 15301; ICC Termination Act of 1995, Pub. L. No. 104-88, 106(a), 109 Stat. 803, 922. This jurisdiction includes anhydrous ammonia pipelines. See CF Indus., Inc. v. FERC, 925 F.2d 476, 478 (D.C. Cir. 1991) (affirming ICC jurisdiction over anhydrous ammonia pipelines). A pipeline carrier's rates must be "reasonable" and non-discriminatory, 49 U.S.C. 15501, and if the Board determines that they are not, it "may prescribe the rate ... to be followed," id. 15503(a), and direct the repayment of overcharges, id. 15904(b)(1), (c)(2). The Board must consider, "among other factors[,] ... the need for revenues that are sufficient, under honest, economical, and efficient management, to let the carrier provide that transportation," as well as "the availability of other economic transportation alternatives." Id. 15503(b)(2), (3).

4

Koch owns one of two pipelines that transport anhydrous ammonia to the Midwest in pressurized, liquid form. Koch's pipeline originates in Louisiana and connects that state's producers to numerous Midwestern terminals. The other pipeline, owned by the Mid-American Pipeline Company (MAPCO), originates in Texas and Oklahoma. Koch purchased its pipeline from Gulf Central Pipeline Company in 1988 as part of a $200 million package that also included the Gulf Central Storage and Terminal Company and a natural gas company. Koch continued to charge Gulf Central's shipping rates until 1996, when Koch raised its rates. For the locations relevant here, the increases averaged almost 20%.

5

CF Industries and Farmland Industries are farmer-owned cooperatives that produce anhydrous ammonia in Louisiana and ship it to the Midwest via Koch's pipeline. Both producers also occasionally ship by rail, and CF ships a significant amount of ammonia by barge. Farmland's production facilities are not located near a river and thus have no barge access. On March 27, 1996, CF Industries filed a complaint with the Board, alleging that Koch's rate increases were unreasonable.1 Four months later, the STB granted Farmland's petition to intervene as a complainant.

6

In May 1997, the STB issued an initial order to govern the proceedings. See CF Indus., Inc., No. 41685 (S.T.B. May 14, 1997) (Initial Order). Two parts of that order are important here. First, the Board decided that it would only prescribe rates at locations where it determined Koch to be "market dominant," finding no justification for the agency "to inject itself into the pricing of services" where competitive alternatives act "as an effective constraint on a pipeline's rates." Id. at 5. In assessing the existence of effective competitive alternatives, the Board said it would be guided by the railroad market dominance guidelines issued by its predecessor agency, the ICC, and by the precedent developed under those guidelines. Id. at 5 (citing Market Dominance Determinations & Consideration of Prod. Competition, 365 I.C.C. 118, 129 (1981), aff'd sub nom. Western Coal Traffic League v. United States, 719 F.2d 772 (5th Cir. 1983) (en banc), modi fied, Product & Geographic Competition, 2 I.C.C.2d 1 (1985)).2

7

Second, the Board stated that for locations where it found Koch to be market dominant, it would evaluate Koch's rates using the Constrained Market Pricing (CMP) principles articulated in the ICC's Coal Rate Guidelines. Id. at 6 (citing Coal Rate Guidelines, Nationwide, 1 I.C.C.2d 520 (1985), aff'd sub nom. Consolidated Rail Corp. v. United States, 812 F.2d 1444 (3d Cir. 1987)).3 Under CMP, the Board said, a complainant may choose among several rate constraints, including the "stand-alone cost" and "revenue adequacy" constraints. Id. at 6.

8

In January 1998, CF moved to amend its complaint to add a challenge to the rates Koch had charged prior to its 1996 increase. CF argued that its amendment merely clarified the relief requested in its initial complaint, where it had asked for refunds of the rate increases as well as "such other relief as the Board deems just and proper." Compl. p 53. Koch opposed the amendment.

9

On May 9, 2000, the STB issued its final decision. At the outset, the STB denied CF's motion to amend its complaint as untimely--"having been filed almost 2 years after the initial complaint and 4 months after the close of discovery." Final Order at 2 n.4. It also held that CF was estopped from challenging pre-increase rates based on a settlement agreement CF had signed with Koch's predecessor, Gulf Central. Id.

10

The Board then turned to Koch's 1996 rate increases, addressing first the question of Koch's market dominance. In analyzing this issue, the STB considered several possible competitive alternatives to Koch's pipeline, only one of which is at issue here: "intermodal" competition from river barges.4 The STB concluded--and Koch does not dispute in this proceeding--that barge shipping does not compete with the pipeline for Farmland's business because Farmland lacks access to barge transportation. Id. at 11. However, the Board also concluded--and this Koch does dispute--that barges do not effectively compete with the pipeline for CF's transportation to numerous pipeline destination points. Id. at 13-17.5

11

Having found Koch to be market dominant at a number of terminals, the Board went on to ask whether Koch's rate increases for service to those terminals were reasonable. To make that determination, the Board applied the revenue adequacy test of CMP, which asks whether rates generate revenues sufficient to "cover all costs and provide a rate of return on investment equal to the current cost of capital." Id. at 21 (citing Coal Rate Guidelines, 1 I.C.C.2d at 535). It found that, even without the rate increases, Koch would more than recover its total investment in the pipeline by the end of 2000, and that, with the exception of its first year of owner ship, Koch's return-on-investment (ROI) "has exceeded its cost of capital in all years and by increasingly larger margins so that by 1996 its ROI (21.52%) was almost twice the cost of capital (11.80%)." Id. at 26. The Board concluded that, "based on all reliable measures, it is clear that the pipeline is earning adequate revenues and that Koch's 1996 rate increases are not warranted." Id.

12

Koch challenges the STB's determinations regarding both market dominance and rate reasonableness. We consider those challenges in Parts II and III below. In a separate petition for review, which we briefly address in Part IV, CF and Farmland challenge the Board's denial of CF's motion to amend its complaint to include an attack on Koch's preincrease rates.

II

13

The STB's market dominance guidelines define "market dominance" as "an absence of effective competition, from other carriers or modes of transportation, for the traffic or movement to which a rate applies." See Market Dominance, 365 I.C.C. at 128 (applying the statutory standard for the regulation of railroad rates, 49 U.S.C. 10707(a)); see also 49 U.S.C. § 15503(b)(3) (requiring STB, when prescribing pipeline rates, to consider, inter alia, "the availability of other economic transportation alternatives"). " '[E]ffective competition['] ... means that, if a carrier raises the rate for such traffic, then some or all of that traffic will be lost to other carriers or modes." Market Dominance, 365 I.C.C. at 129. Such competition, the guidelines state, "serves as a constraint on the ability of [the] carrier to raise rates." Id. To evaluate market dominance, the guidelines employ a "flexible" approach that rests on "case-by-case" analysis of competitive alternatives to a carrier's facility. Id. at 119.

14

In evaluating the effectiveness of barge transport as a competitive alternative, the Board first reviewed CF's claim that barge transport has qualitative disadvantages compared to pipeline transport. The Board noted that barge companies lack sufficient hauling and storage capacity to handle a significant shift of anhydrous ammonia traffic from pipeline to barge; that barge transport involves higher costs than pipeline transport, which could make a shift prohibitively expensive; that barges, unlike pipelines, are hindered by floods, low water, and icing; and that barge trips take from days to weeks, while pipeline injection and withdrawal is essentially instantaneous. See Final Order at 11. The Board also noted CF's evidence that pipelines are a safer mode for transporting a hazardous product like anhydrous ammonia. Id. at 8.

15

Next, the Board considered an "Alternative Inbound Study" developed by CF, which indicated that, because of insufficient storage capacity at CF's barge destination points, it would have to make prohibitively large expenditures or investments to shift from pipeline to barge.6 Although Koch criticized CF's claim of inadequate storage capacity, the Board accepted Koch's criticism only in part, finding that Koch's own restatement of CF's study ignored storage costs. The Board also considered a "matching" study prepared by Koch to demonstrate the effectiveness of barge competition, but found Koch's study unreliable. Id. at 13-14. The Board then consolidated data from both parties and recalculated where necessary to account for the problems it had identified. Id. at 15.

16

Finally, "[a]s a measure of the effectiveness of barge competition," the Board "compared Koch's revenues ... under the old rate structure to its revenues under the new rate structure, assuming all cost-competitive traffic would be diverted." Id. at 15. That comparison showed that "Koch's revenues for the CF traffic that it would retain--based on its increased rates--would exceed pre-increase revenues" at a number of terminal points. Id. As a result, the Board concluded that Koch does not face effective barge competition at those locations. Id. at 15 & n.43 (citing Market Dominance, 365 I.C.C. at 128-29, 131; Aluminum Ass'n Inc., 367 I.C.C. 475, 489, aff'd sub nom. Aluminum Co. of Am. v. ICC, 761 F.2d 746 (D.C. Cir. 1985); Salt River Project Agric. Improvement, 1 I.C.C.2d 684, 691 (1985), aff'd, Salt River Project Agric. Improvement & Power Dist. v. United States, 762 F.2d 1053 (D.C. Cir. 1985)).7

17

Koch attacks the STB's market dominance determination on two grounds. First, it contends that the Board's comparison of its pre-and post-increase revenues was an inappropriate test for market dominance. Second, it contends that the Board disregarded record evidence of effective competition, particularly Koch's own "matching" study.

18

We consider these two challenges below. We do so with the understanding that "since decisions concerning market dominance are peculiarly within the expertise of the [Board], our review of such decisions must be particularly deferential." Aluminum Co., 761 F.2d at 750 (Scalia, J.). We may vacate the Board's market dominance determination only if it is "arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law ... [, or if it is] unsupported by substantial evidence." Arizona Pub. Serv. Co. v. United States, 742 F.2d 644, 649 (D.C. Cir. 1984) (quoting 5 U.S.C. 706(2)(A), (E)).

19

* Koch contends that the Board resolved the issue of market dominance based on a single, unprecedented, and irrational standard: whether Koch could raise its rates and still increase its net revenues. We conclude that none of those three adjectives appropriately characterizes the standard employed by the Board.

20

First, although the STB's comparison of revenues before and after the rate increase was important to its analysis, it was not the only factor the STB examined. Rather, that comparison came last in the STB's analysis. It was employed only after the Board had reviewed a host of qualitative considerations that it found to limit the effectiveness of barge competition--including capacity, reliability, speed, and safety--and after it had reviewed the studies proffered by both Koch and the complainants.8 While the Board's market dominance guidelines contemplate the use of such qualitative considerations, they do not exclude the application of quantitative analysis as well. See Market Dominance, 365 I.C.C. at 119 n.5.9

21

Second, the revenue comparison methodology is hardly unprecedented. To the contrary, it is well in accord with both the market dominance guidelines and the agency precedent cited by the Board. See 365 I.C.C. at 131 ("If the loss in future revenues exceeds the gains from exercising market power in the short term, then a rail carrier will be deterred from charging excessive rates."); id. at 129 (stating that "effective competition ... means that, if a carrier raises the rate for such traffic, then some or all of that traffic will be lost to other carriers or modes"); Aluminum Ass'n Inc., 367 I.C.C. at 489 ("A potential loss in future revenues which merely exceeds any gain from exercising market power in the short term will deter a rail carrier from charging excessive rates."); Salt River, 1 I.C.C.2d at 691 ("[T]he potential loss of future revenues, provided that the loss exceeds any gains from the exercise of market power, will deter rail carriers from charging excessive rates."), aff'd, Salt River, 762 F.2d at 1053; see also Burlington N. Inc., Finance Docket No. 32549, 1995 WL 528184, at *44 n.72 (I.C.C. Aug. 23, 1995) (noting that market power is defined as "the ability profitably to sustain higher prices or lower service quality" (internal quotation omitted)). It is also in accord with judicial precedent. See Burlington N. R.R. v. STB, 114 F.3d 206, 212 (D.C. Cir. 1997) (upholding STB market dominance determination where carrier "could recoup profits on lost incremental coal traffic by charging higher rates" on shipments that complainant could not avoid); Arizona Pub. Serv. Co., 742 F.2d at 654 (noting that effectiveness of competition "would depend (at least) on ... the extent of the monopolist's profit that the railroads could reap by raising their prices, and ... the amount of traffic [they] would lose by raising prices").10

22

Finally, Koch contends that the Board's methodology-testing whether a firm is market dominant by asking whether it can increase its net revenues by raising its prices--is irrational. Given that this methodology is consistent with the market dominance guidelines, which are not themselves questioned by Koch, a challenge to its rationality is misplaced.11 In any event, we do not find the methodology irrational. To the contrary, it represents an accepted method of measuring market power, based on the recognition that although a firm in a competitive market cannot raise its prices without a net loss of revenue, a firm with market power can.12 See IIA Phillip E. Areeda & Herbert Hovenkamp, Antitrust Law: An Analysis of Antitrust Principles and Their Application p 501, at 85 (1995) ("A defendant firm has market power if it can raise price without a total loss of sales. Market power ... is large when a firm can profit by raising prices substantially without losing too many sales."); id. p 503a, at 93; see also Landes & Posner, 94 Harv. L. Rev. at 937; Richard A. Posner & Frank H. Easterbrook, Antitrust: Cases, Economic Notes & Other Materials 348-49 (2d ed. 1981).13 Of course, pre cisely how high a firm can raise its prices without reducing net revenues is an important consideration. But the amount by which Koch increased its rates here--an average of 20%-is well above the standard usually employed to signal a substantial degree of market power. See IIA Areeda p 552a, at 223 (suggesting use of 5% or 10% differential); cf. Merger Guidelines 1.11 (utilizing 5% price-increase test).

23

Koch correctly points out that the ability to increase revenues by raising prices is not always indicative of market power. As Koch notes, "a firm in a fully competitive market that is pricing below market levels would expect to earn greater revenues by raising its prices to meet its competitors." Br. for Pet'r Koch ("Koch Br.") at 16. But the normal assumption in examining assertions of market power is that the current price is at least the competitive price. See IIA Areeda p 537b, at 200. Koch questions the validity of that assumption in this case, pointing to the fact that it had not raised its prices in eight years prior to 1996. But Koch is a for-profit institution, not an eleemosynary one, and it has provided no reason to believe that it priced below market for eight years, rather than calculated that those prices were the most the market would bear. See IIA Areeda p 501, at 85 (noting that rational profit-maximizing firm has no reason to sell for less than market price).14

B

24

Koch also contends that the STB, in making its market dominance finding, improperly rejected certain studies Koch submitted. Koch first takes issue with the Board's rejection of its "matching" study. In that study, Koch identified retailers that had received anhydrous ammonia shipped by both barge and pipeline during the same year. Koch argued that those retailers could meet all of their ammonia needs from barges, and thus that the tonnage needed to serve those retailers would be diverted from pipeline to barge. Such a diversion, Koch argued, would constrain the pipeline's rates. The Board rejected Koch's matching study as unreliable because Koch included retailers whose costs for shipping by barge were greater than for pipeline when the total costs of barge shipping--including the cost of trucking the anhydrous ammonia from the barge river terminals to the areas served by inland pipeline terminals--were included in the calculation. Final Order at 13-14 & n.36. The cost differential indicated to the Board that barges were not serving as a competitive alternative to pipeline transport for those retailers, but instead were meeting different needs--such as providing additional anhydrous ammonia when sufficient supplies were unavailable by pipeline. Id.; see IIA Areeda p 534b at 180 ("The absence of close price relationships among products presumptively indicates that they are in separate markets.").

25

Koch claims that the STB improperly rejected its matching study because the Board arbitrarily viewed as noncompetitive any barge shipping that was more costly than Koch's pipeline. It is, of course, well-accepted that a significant cost differential may render one product an ineffective restraint on the pricing of another, even if the two could otherwise serve as substitutes for one another. Market Dominance, 365 I.C.C. at 134 (stating that evidence of product competition should show that shipper can obtain feasible substitutes "without substantially greater cost, transportation or otherwise").15 As Koch itself notes, "the Board's market dominance analysis ask[s] whether complaining shippers ha[ve] feasible competitive alternatives such that they could switch a sufficient amount of traffic to protect themselves from an unreasonable price increase if they wished." Koch Br. at 24 (emphasis added). If barge costs were significantly above those of pipeline, switching to barge would not protect the shippers from Koch's "unreasonable price increase." Accordingly, Koch concedes that "at some point an alternative will be so costly or impracticable that it is simply not feasible." Id. at 26.

26

The trouble with the STB's analysis, Koch contends, is that the Board regarded barge shipping as noncompetitive if it cost even one cent more than pipeline. Id. at 10, 24. Koch bases this contention on a single sentence in the STB's opinion, in which the Board stated that Koch's matching study had improperly indicated matches "where total barge costs were higher than total pipeline costs." Final Order at 13. By "higher," Koch contends, the Board must have meant higher by as little as one cent.

27

But the footnote to the cited sentence makes clear that Koch has misread the STB's test. See Final Order at 13 n.36. In that footnote, the Board provides an example to explain what it regarded as the flaw in Koch's study: a case in which Koch claimed a "match" notwithstanding that the total bargedelivered costs were $43.41 per ton, while pipeline-delivered costs were only $36.80--a difference of approximately 18%. Although a small difference in the price charged by an alternative carrier would not prevent it from constraining Koch's prices, a difference of 18% is well-recognized as having that effect. Cf. IIA Areeda p 537b, at 200, p 552a, at 223 (suggesting 5% or 10% difference as significant).

28

Koch also attacks the STB for rejecting its "restatement" of CF's Alternative Inbound Study. In that restatement, Koch identified 137,000 tons of anhydrous ammonia that it said could be diverted from pipeline to barge. See Final Order at 12. The Board rejected Koch's figures, in part because they failed to account for storage costs CF would have to pay at terminals it did not own. Id. at 14 & n.38. Koch argues that in so doing, the Board "ignored unrefuted evidence" that shifting the anhydrous ammonia to the nextclosest CF storage terminals would still be less costly than pipeline. Koch Br. at 27. But Koch's evidence was not unrefuted; rather, it was vigorously disputed by CF's witnesses. Compare Verified Reply and Rebuttal Statements of C. Phillip Baumel on Behalf of Koch Pipeline Co., J.A. at 23842, 370-74, with Verified Opening and Rebuttal Statements of Fred Mugica on Behalf of CF Industries, Inc., J.A. at 216-18, 494-95. Although Koch may regard CF's responses as constituting ineffective refutation, the Board acts within its authority when it chooses between contending accounts of the evidence, see Burlington N. R.R., 114 F.3d at 213, and this court has no power "to substitute its judgment for that of the agency," id. at 210-11 (quoting Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 416 (1971)).

III

29

We next turn to the STB's conclusion that Koch's 1996 rate increases were unreasonable. Although the Board announced in its initial order that it would apply Constrained Market Pricing (CMP) principles, Koch contends that it failed to do so. While the components of Koch's argument are interrelated, it is most convenient to consider them in three parts: (1) that the Board should have determined rate reasonableness using a standard consistent with the stand-alone cost (SAC) constraint of CMP; (2) that instead, the Board used original cost ratemaking (OCR), a discredited methodology; and (3) that in so doing, the Board failed to account for the expenditures Koch would eventually have to make to replace the pipeline.

30

In reviewing the STB's determination of rate reasonableness issues, we again apply the deferential standards of the Administrative Procedure Act, 5 U.S.C. 706(2)(A), (E), and uphold Board decisions as long as they are supported by substantial evidence and are not arbitrary or capricious. See McCarty Farms, Inc. v. STB, 158 F.3d 1294, 1300 (D.C. Cir. 1998). "Because Congress has expressly delegated to the Board responsibility for determining whether a [carrier's] ... rate is reasonable, the Board is at the zenith of its powers when it exercises that authority, and [is] therefore entitled to particular deference." Burlington N. R.R., 114 F.3d at 210 (internal quotations omitted). Moreover, as with the market dominance guidelines, we note Koch's declaration that it does not challenge the rate guidelines themselves, but only their asserted misapplication in this case.

31

* The purpose of CMP is "to ensure that a carrier does not use its market dominance to charge its captive ... shippers more than they should have to pay for efficient ... service." Coal Rate Guidelines, 1 I.C.C.2d at 524; see Consolidated Rail, 812 F.2d at 1457 (affirming Coal Rate Guidelines). To accomplish this, CMP imposes four separate constraints on rates, two of which are relevant here: "revenue adequacy" and SAC. Coal Rate Guidelines, 1 I.C.C.2d at 521.16 The first constraint, revenue adequacy, examines the existing carrier on a system-wide basis to determine the revenues it needs to "provide a rate of return on net investment equal to the current cost of capital (i.e., the level of return available on alternative investments)." Id. at 535. The SAC constraint, by contrast, limits a carrier's rates to those necessary to generate "the revenue that a hypothetical new, optimally efficient carrier would need to meet in order to serve the complaining shippers" alone. Final Order at 7; see Coal Rate Guidelines, 1 I.C.C.2d at 542; McCarty Farms, 158 F.3d at 1301. SAC is intended to ensure that a shipper "not bear the costs of any facilities or services from which it derives no[ ] benefit." Coal Rate Guidelines, 1 I.C.C.2d at 528.17

32

In this case, CF and Farmland elected to rely on the revenue adequacy constraint. Holding that revenue adequacy and SAC provide "alternative methodologies for examining the reasonableness of a carrier's rates," and that "complainants may use any methodology that is consistent with CMP," Final Order at 7, the Board employed the revenue adequacy approach and found Koch's 1996 rate increases unnecessary to ensure adequate revenues, id. at 27. In so doing, the STB rejected the relevance of Koch's SAC evidence, which purportedly would have justified the company's rate increases. Id. at 22.

33

Koch contends that the STB should have accepted its SAC model, or at least refused to employ a revenue adequacy methodology that yielded different results. The Board's conclusion to the contrary, however, is consistent with the agency's rate guidelines. Those guidelines state that SAC and revenue adequacy are "separate constraining factors" on the maximum rates a carrier may charge, Coal Rate Guidelines, 1 I.C.C.2d at 521, and that "[c]arriers do not need ... and ... are not entitled to any higher revenues" than the revenue adequacy standard permits, id. at 535. Moreover, the guidelines expressly contemplate that "the rate to an individual shipper may vary depending upon which of the two CMP approaches is used," and that it is the complaining shipper who may "decid[e] which approach to pursue." Id. at 534 n.35. See Consolidated Rail, 812 F.2d at 1451 (noting that the guidelines provide that the constraints "may be used individually or in combination to analyze whether the rate [increase] is unreasonably high" (quoting Coal Rate Guidelines, 1 I.C.C.2d at 548)).18

34

In sum, the Board's determination that Koch could not charge rates higher than those permitted by the revenue adequacy constraint, and therefore that Koch's SAC evidence was not relevant even if it would have yielded a different result, was a reasonable reading of the agency's rate guidelines and is not subject to reversal by this court. See Auer v. Robbins, 519 U.S. 452, 461 (1997).

B

35

Koch also attacks the Board's rate-reasonableness analysis on the ground that it assertedly employed OCR principles, which limit the carrier to a return based on the depreciated cost paid by the original owner of the capital assets. See generally Ashley Creek Phosphate Co., No. 40131, 1992 WL 52672, at *1 n.4, *8 (I.C.C. Mar. 12, 1992) ("OCR uses original cost as of the actual installation date of the assets as the basis for computing the capital costs.").19 Koch devotes a considerable part of its argument to pointing out the flaws in the OCR approach, but we need not consider those flaws because the STB did not use original cost in the analysis it undertook in this case. The Board relied, instead, on "[a]quisition-cost valuation--the amount [Koch] paid in an arm's-length transaction" to acquire the pipeline. Final Order at 24.

36

Although it may appear that Koch and the STB are talking past one another, a footnote to Koch's brief discloses the source of the disconnect. Koch Br. at 36 n.9. There, Koch asserts that the $77.2 million figure the STB used as Koch's initial investment in the pipeline was not Koch's true acquisition cost, but rather the previous owner's depreciated original cost as of the date Koch bought the pipeline. Koch argues that by using that figure to determine revenue adequacy, the STB effectively employed an OCR analysis. Id. at 36.

37

This argument cannot prevail. As the STB explains, the $77.2 million figure was the figure Koch itself assigned to its pipeline assets on a form Koch filed with the Federal Energy Regulatory Commission (FERC) shortly after the acquisition. See Final Order at 23 n.61; 1988 FERC Form 6 (reproduced in Br. for CF Before STB, Ex. 1 at 111). The form's instructions required the use of acquisition cost.20 Moreover, Koch submitted no other figure for its investment base to the STB, notwithstanding the Board's direction in its initial order that the parties' evidence "should include ... pipeline investment." Initial Order at 7.21 Nor has Koch suggested an alternative figure on this appeal. Accordingly, the STB reasonably concluded that it could "properly use Koch's own $77.2 million valuation as a reliable estimate of its cost of acquiring--and the value of its initial investment in--the pipeline." Final Order at 25.

C

38

Finally, Koch argues that the Board should have used replacement cost, rather than either acquisition or original cost, as the baseline for its revenue adequacy constraint. Koch defines replacement cost as "the current cost of replacing assets (or constructing them anew)." Koch Br. at 30.

39

Much of Koch's argument in favor of replacement cost relies on the fact that (a variant of) replacement cost is utilized in SAC analysis. SAC uses replacement cost, however, because "the theory of SAC ... assumes that a new entrant can potentially enter the market today[; thus,] asset value must be based on the cost of acquiring assets today (at their current value)." Coal Rate Guidelines, 1 I.C.C.2d at 544-45 (emphasis added). Revenue adequacy analysis, by contrast, is intended to determine whether the revenues of an existing company are sufficient to "provide a rate of return on [that company's] net investment equal to the current cost of capital." Id. at 535. Hence, the considerations that compel the use of replacement cost in the SAC constraint do not apply to the independent constraint of revenue adequacy.

40

Koch also contends that the STB should have used replacement cost because it best simulates competitive pricing. If rates of return are not calculated based on the cost of replacing the pipeline, Koch asserts, it will not be able to compete equally with other firms for available financing in order to replace its current facilities. The Board, however, believes that as long as a carrier is permitted revenues adequate to "cover all costs and provide a rate of return on investment equal to the current cost of capital," it will be able to "compete equally with other firms for available financing in order to maintain, replace and, if necessary, expand its facilities and services." Final Order at 21. In the Board's view, the use of acquisition cost as the investment base ensures such revenue, and it is therefore unnecessary also to "requir[e] captive pipeline shippers to provide in advance a revenue stream to pay for investments not yet made and assets that are not in place." Id. at 27-28.22

41

Our role is not to choose the optimal method of pipeline ratemaking, but only to ensure that the STB's choice is a rational one. We cannot say that the Board has acted irrationally here.23 Moreover, Koch's attack on the economic implications of eschewing reliance on replacement cost methodology is largely beside the point. As Koch has stressed, it does not challenge the regulations that govern the Board's approach to ratemaking, but instead contends that they were not followed. This circumscribes our role even further, as we must uphold the Board's interpretation of its regulations as long as that interpretation is reasonable. See Auer, 519 U.S. at 461; Thomas Jefferson Univ. v. Shalala, 512 U.S. 504, 512 (1994).

42

Koch correctly notes that during the 1980s, the ICC actively "consider[ed] using a replacement cost methodology in valuing assets in future revenue adequacy determinations." Standards for Railroad Revenue Adequacy, 364 I.C.C. 803, 818 (1981). But while the Commission regarded replacement cost as "conceptually the best method available," id. at 820, in the end it concluded that replacement cost methodology "cannot be practically implemented in a manner that we can be confident would produce accurate and reliable results." Standards for Railroad Revenue Adequacy, 3 I.C.C.2d 261, 277 (1986).

43

In promulgating the 1990 guidelines cited by the STB in this case, the ICC again considered the appropriate investment base for analysis of revenue adequacy. See Railroad Revenue Adequacy, 6 I.C.C.2d at 940. After notice and comment, the Commission rejected the railroads' argument that "the soundest valuation of assets from an economic perspective is current replacement cost," id. at 937, and their further argument that original "predecessor costs are closer to that measure than are acquisition costs," id. Responding to the same concerns Koch raises here, the ICC concluded "that railroad assets would be replaced as long as competitive returns are allowed on new and existing investments, [and that].... if investors can reasonably expect competitive returns, capital can be attracted when it is needed, making advance funds accumulation unnecessary." Id. at 939 (adopting view of Railroad Accounting Principles Board). Accordingly, the Commission announced that it would "use acquisition costs in valuing railroad investment bases in this and subsequent revenue adequacy determinations." Id. at 939.24 This court affirmed that decision in Association of American Railroads v. ICC, 978 F.2d 737, 741 (D.C. Cir. 1992). It is therefore clear that the Board reasonably interpreted its rate guidelines as using acquisition rather than replacement cost as the investment base upon which to determine revenue adequacy. See Final Order at 24 (citing Railroad Revenue Adequacy, 6 I.C.C.2d at 940).

44

We conclude that, in assessing the legality of Koch's rate increase, the STB employed a reasonable ratemaking methodology consistent with the Board's guidelines and precedents. That being the case, Koch's challenge to the reversal of its rate increase must fail.

IV

45

We now briefly address CF and Farmland's charge that the STB improperly denied CF's motion to amend its complaint to include a challenge to Koch's pre-increase rates. "Leave to amend any document is a matter of the Board's discretion," and we find no abuse of that discretion here. 49 C.F.R. 1104.11.

46

The STB denied CF's motion on the ground that it was "untimely--having been filed almost 2 years after the initial complaint and 4 months after the close of discovery." Final Order at 2 n.4.25 The Board had warned in its initial order that, "[b]ecause 49 U.S.C. 15901(c) requires that this investigation be concluded within three years after its initiation," it was "critical that th[e] investigation be conducted in an orderly and timely fashion." Initial Order at 4. Under those circumstances, it was reasonable for the Board to reject an amendment that would have substantially expanded the scope of the proceedings.

47

CF contends that no such expansion was required here-indeed, that no amendment was formally required--because its original complaint encompassed a request for relief from pre-increase rates. That complaint, however, stated that CF had filed "in order to seek a directive from the Board requiring Koch Pipeline to maintain its rates at current levels" and to recover damages "in an amount equal to any rate increase." Compl. p 6. Although CF notes that the complaint also "requested all 'just and proper' relief," Br. for Pet'rs CF & Farmland at 22 (quoting Compl. at 21), that kind of boilerplate request was insufficient to put the Board or the defendant on notice that a pre-increase reduction was also on the table.26

48

CF further contends that the denial of leave to intervene was inconsistent with STB precedent. But the Board has frequently expressed concern with the timing and extent of amendments, see, e.g., Grain Land Coop, No. 41687, 1999 WL 1117130, at *4 (S.T.B. Dec. 8, 1999) (permitting second amendment because it was "simply a clarification" of first); Seapac Servs. Inc., No. 40534, 1992 WL 88109, at *1 (I.C.C. May 4, 1992) (noting that amendment would not "unduly broaden" or "unduly delay" proceeding), and has denied amendment in circumstances similar to those here, see SouthWest R.R. Car Parts Co., No. 40073, 1988 WL 225131 (I.C.C. Dec. 1, 1988) (denying amended complaint's untimely claims). Accordingly, we conclude that the Board did not abuse its discretion in denying CF's motion to amend its complaint.

V

49

Koch's challenge to the final decision of the Board is essentially a narrow one, questioning not the STB's regula tions but rather their application in a single ratemaking. We conclude that the STB's finding that Koch was market dominant was a reasonable application of the Board's market dominance guidelines, and that the Board's rollback of Koch's 1996 rate increase was a reasonable application of its ratemaking guidelines. At the same time, we hold that the Board did not abuse its discretion in denying CF's request to amend its complaint to seek still further rate reductions.

The petitions for review are therefore

50

Denied.

Notes:

1

The complaint also charged that Koch's rates discriminated against CF in favor of a Koch affiliate, a charge the STB found moot for most of the terminals and rejected on the merits for two. Final Order at 28. CF does not appeal the Board's disposition of the discrimination claim.

2

See ICC Termination Act 204(a), 109 Stat. at 941 (providing that all ICC orders and regulations shall continue in effect until modified or revoked by the STB), reprinted in 49 U.S.C. 701 note.

3

"Notwithstanding the title..., those guidelines are not limited to any one commodity. Coal cases, which typify captive, highvolume, repetitive rail traffic, were the springboard for our analysis because of the prevalence of coal rate challenges." Rate Guidelines--Non-Coal Proceedings, Ex Parte No. 347 (Sub.-No. 2), 1995 WL 705171 (I.C.C. Dec. 1, 1995).

4

In parts of its opinion not challenged in this court, the STB found that Koch does not face effective: (1) "intramodal" competition from the MAPCO pipeline, because that pipeline originates in Texas and Oklahoma rather than Louisiana where complainants are located, Final Order at 8; (2) "geographic" competition from anhydrous ammonia produced at other locations, except at the one point at which the MAPCO and Koch pipelines cross and share a common storage terminal, id. at 17-19; (3) "product" competition, i.e., the ability of farmers to substitute other fertilizer products for anhydrous ammonia, id. at 20-21; or (4) "intermodal" competition from trucks or railroads, id. at 8-11.

5

The STB found that barge shipping does constrain Koch's prices to CF at one pipeline destination, which is located on the Missouri River at Palmyra, Missouri. Final Order at 12. CF does not appeal that finding.

6

According to CF's study, Koch could raise its rates an additional 20-50% without facing effective barge competition. Final Order at 11 (citing CF exhibits and verified statements).

7

There were four points at which Koch would not increase net revenue if all cost-competitive tonnage were diverted to barge. Final Order at 16. As to these points, however, the Board found that the tonnage that could actually be diverted, given capacity constraints at the storage terminals, was "too small an amount ... to constrain Koch's rates," and that Koch would "still earn greater revenues on its retained traffic (with the rate increases) at each of these pipeline points than what it earned for all of the traffic at each of those points under the prior rates." Id. at 17.

8

Koch points out that, in a footnote, the STB stated: "While we do not rely on" quantitative measures like the revenue comparison methodology "as a substitute for a thorough qualitative examination of all possible competitive alternatives ... , we are not restricted from using any valid tool ... where, as here, the other evidence leaves the question [of effective competitive alternatives] unresolved." Final Order at 15-16 n.43. But rather than read this footnote, as Koch does, as indicating that the Board's ultimate conclusion rested solely on the outcome of the revenue comparison, we read it, in accord with the text of the STB opinion, as confirming that the Board undertook a "thorough qualitative examination of all possible competitive alternatives" before reaching its conclusion.

9

Koch correctly notes that the market dominance guidelines did reject the use of certain quantitative presumptions: e.g., high price/cost ratios, observed market share percentages, and substantial shipper-related investments. 365 I.C.C. at 120-26. None of those presumptions was employed in this case.

10

Our opinion in Salt River is not to the contrary. Although we affirmed an ICC finding of lack of market dominance even though the alternatives "may not [have] exert[ed] effective market pressure" on the defendant railroad's rates, we did so because the complainant shipper used the railroad "only under exceptional and unpredictable circumstances." 762 F.2d at 1064 & n.14. We concluded that in enacting the market dominance inquiry of 49 U.S.C. 10707, Congress did not intend to include a situation in which a carrier had only "transitory market power" over a shipper. Id. at 1062. Here, CF is not merely an occasional user of Koch's pipeline, and the latter's exertion of market power cannot be characterized as "transitory."

11

See Aluminum Co., 761 F.2d at 751 ("It is not possible for petitioners in this proceeding to challenge the validity of [the ICC's market dominance guidelines]. The time for direct review of the Commission's guidelines has long passed." (citing 28 U.S.C. 2344)).

12

Although techniques exist for measuring market power more directly, they involve data not typically available to courts or regulators, and data which the parties agree are not part of the record in this case. See William M. Landes & Richard A. Posner, Market Power in Antitrust Cases, 94 Harv. L. Rev. 937, 939-43 (1981) (noting that market power can be measured directly with knowledge of firm marginal cost or elasticity of demand).

13

In the closely related area of merger analysis, the Department of Justice defines the scope of the relevant product market by asking whether a hypothetical profit-maximizing firm that is the only seller of the product could profitably impose a "small but significant and nontransitory increase in price." See U.S. Dep't of Justice, Horizontal Merger Guidelines 1.11 (1992); see also IIA Areeda p 536, at 195-96 (noting that a product is a market of its own if a firm controlling all of its output could profit from a significant price increase).

Koch cites antitrust cases that it regards as inconsistent with the proposition discussed in the text, but those cases are inapposite. Blue Cross holds that high prices or high profits alone do not necessarily evidence monopoly power, see Blue Cross & Blue Shield United of Wisc. v. Marshfield Clinic, 65 F.3d 1406, 1411-12 (7th Cir. 1995), but the STB used neither as an index of market dominance here. Brooke Group held that industry-wide price increases do not necessarily evidence conscious parallelism in an oligopolistic market, Brooke Group Ltd. v. Brown & Williamson Tobacco Corp., 509 U.S. 209, 237 (1993), a point not at issue in this case.

14

Although it is true that during 1988-95 Koch's prices were potentially subject to rate regulation by the ICC, the Commission never imposed regulation and Koch never tried to raise its rates. Koch asserts that its pre-increase rates were substantially below those of "its competitor pipeline [MAPCO] and barge competitors." Koch Br. at 16-17. But the Board rejected that contention because "[s]uch comparisons, especially to carriers that largely serve different markets or bear different transportation characteristics and operating costs, are not particularly instructive." Final Order at 16 n.43. Indeed, Koch's contention assumes the answer to the question at issue: Are the MAPCO pipeline and barge carriers really Koch's competitors? The Board found, and Koch has not challenged here, that the MAPCO pipeline is not an intramodal competitor because it does not serve complainants' Louisiana production facilities, and that it does not pose geographic competition except at one terminal. As for the barges, the STB concluded that the fact that their rates were substantially above those of Koch showed not that Koch's rates were sub-market, but rather that barge transport is an ineffective competitive alternative to pipeline. As explained in Part II.B infra, we find that conclusion to be a reasonable one.

15

See Arizona Pub. Serv. Co., 742 F.2d at 650-51 (holding that in light of significant cost difference, fact that complainant received oil by both trucks and railroad did not make trucks an effective competitive alternative); Atchison, Topeka & Santa Fe Ry. v. ICC, 580 F.2d 623, 636 (D.C. Cir. 1978) ("Shippers must be able to make the choice to use an alternative service without absorbing substantial economic loss."); see also Salt River, 762 F.2d at 1059 (noting that market dominance guidelines require findings on the relative costs of potential alternatives, and that without evidence of relative costs, the "mere possibility" that shipper "could" use alternative carrier does not render alternative an effective competitor); IIA Areeda p 534b, at 180 (noting that "substantial differences in production, transportation, or other costs among ... products prevent one from operating as an effective competitive check on the prices of the other").

16

The other two are market efficiency and phasing. 1 I.C.C.2d at 521; see Consolidated Rail, 812 F.2d at 1450-51.

17

The governing statute directs that, when prescribing a rate, "the Board shall consider, among other factors

(1) the effect of the prescribed rate ... on the movement of traffic by that carrier;

(2) the need for revenues that are sufficient ... to let the carrier provide that transportation or service; and

(3) the availability of other economic transportation alternatives.

49 U.S.C. 15503(b). It also states that "[i]n prescribing [a] rate ... the Board may utilize rate reasonableness procedures that provide an effective simulation of a market-based price for a stand alone pipeline." Id. 15503(a) (emphasis added).

18

The Board's position is not inconsistent with our decisions in Burlington and PEPCO. In Burlington, we remanded a ratemaking because the ICC had returned to a pre-CMP standard without explanation. See Burlington N. R.R. v. ICC, 985 F.2d 589, 599 (D.C. Cir. 1993). In PEPCO, we upheld the ICC's use of SAC in determining the reasonableness of the rates of a railroad with inadequate revenues, but noted that where the other CMP constraints are applicable (they were not applicable in PEPCO), "stand-alone cost is not the only ceiling on rates." Potomac Electric Power Co. v. ICC, 744 F.2d 185, 193-94 (D.C. Cir. 1984).

19

See also Railroad Revenue Adequacy--1988 Determination, 6 I.C.C.2d 933, 935 n.3 (1990) (stating that "original" or "predecessor" cost "represents the cost of the asset when it was first dedicated to public service, plus any subsequent improvements, less depreciation and retirements").

20

The instructions define the entry for "carrier property" as "[t]he cost of property owned that is devoted to transportation service," and define cost as "the amount of money actually paid for property or services." See 18 C.F.R. Pt. 352, def. 11, instr. 2-3. The instructions further state that, "[i]n accounting for a 'purchase,' " i.e. an acquisition of a "distinct operating system" involving a price over $250,000, "the assets shall be recorded on the books of the acquiring carrier at cost as of the date of acquisition." Id., instr. 3-11 (emphasis added).

21

Indeed, the same order advised Koch that CF anticipated that revenue adequacy would be "an important and relevant reasonableness cap in this case," Initial Order at 6, and as we discuss in the next section, the revenue adequacy constraint uses acquisition cost as the measure of a carrier's pipeline investment.

22

Koch quotes the Board's decision in CSX Corp. for the proposition that "carriers cannot attract and retain capital unless they are given the opportunity to be compensated for the real value of the property, not just the book value." Koch Br. at 30 (quoting CSX Corp., No. 33388, 1998 WL 456510 at *40 (S.T.B. July 23, 1998)). But the "book value" rejected by the STB in CSX was the "predecessor book value," 1998 WL 456510 at *40, not the railroad's own acquisition cost--which, in fact, the Board ultimately relied upon in that case. Id. at *38.

23

Our decision in City of Los Angeles v. United States Dep't of Transp., 103 F.3d 1027, 1032-33 (D.C. Cir. 1997), is not to the contrary. In that case, we vacated a DOT decision to set fees in reliance on original cost rather than fair market value, because the Department mistakenly thought it was statutorily required to use original cost. The STB had no such misconception in this case.

24

Koch contends that the Board's Railroad Revenue Adequacy standards were intended to apply only to the annual revenue adequacy calculations the STB performs for railroads, and not to "the ratemaking context." Koch Br. at 32 n.7. But the Board's interpretation to the contrary is a reasonable reading of the regulations. See Coal Rate Guidelines, 1 I.C.C.2d at 535 (using Railroad Revenue Adequacy standards to define the revenue adequacy constraint of the CMP methodology). Nor, as the STB explained, does Arkansas & Missouri Railroad, 6 I.C.C.2d 619, 627 (1990), aff'd sub nom. Missouri Pac. R.R. v. ICC, 23 F.3d 531 (D.C. Cir. 1994), compel the use of replacement cost in revenue adequacy determinations. Final Order at 25 n.64. In that case, the ICC used replacement cost because "an arm's length purchase price" cost could not be determined as a consequence of the way in which the purchase agreement had been structured. 6 I.C.C.2d at 626; see 23 F.3d at 534.

25

The Board also denied permission to amend on the ground that CF was estopped from challenging the rates in effect prior to 1996 by a settlement agreement with Koch's predecessor. In light of our affirmance of the Board's determination with respect to untimeliness, we do not address the STB's alternative rationale.

26

Farmland suggests that its petition to intervene should be treated differently, but that petition likewise sought to bar Koch from "charging any increase in rates established on or about April 1, 1996," Pet. p 34, and sought as damages the difference between a reasonable rate "and the rate paid by Farmland ... during the period beginning on April 1, 1996," Id. p 29. Moreover, in granting Farmland's petition, the Board relied on the fact that "Koch did not oppose Farmland's intervention because the subject matter of Farmland's complaint is substantially similar to that of the complaint brought by CF." CF Indus., Inc., No. 41685, slip. op. at 1 (S.T.B. July 25, 1996).