CF Industries, Inc. v. Surface Transportation Board

United States Court of Appeals

FOR THE DISTRICT OF COLUMBIA CIRCUIT

Argued May 7, 2001 Decided July 27, 2001

No. 00-1209

CF Industries, Inc.,

Petitioner

v.

Surface Transportation Board and

United States of America,

Respondents

Farmland Industries, Inc. and

Koch Pipeline Company, L.P.,

Intervenors

Consolidated with

00-1213, 00-1248

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. Ni-

cholson, Attorney, and John P. Fonte, Attorney, U.S. Depart-

ment 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: In 1996, Koch Pipeline Company,

L.P. raised shipping rates on its anhydrous ammonia pipeline.

Pipeline customers CF Industries, Inc. and Farmland Indus-

tries, 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.

I

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 availabil-

ity 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 imme-

diate application. CF Indus., Inc., No. 41685 at 2-3 (S.T.B.

May 9, 2000) (Final Order).

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. s 15301; ICC Termination Act of 1995, Pub.

L. No. 104-88, s 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. s 15501, and if the Board determines that they are

not, it "may prescribe the rate ... to be followed," id.

s 15503(a), and direct the repayment of overcharges, id.

s 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 availabili-

ty of other economic transportation alternatives." Id.

s 15503(b)(2), (3).

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 pur-

chased 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 ship-

ping rates until 1996, when Koch raised its rates. For the

locations relevant here, the increases averaged almost 20%.

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 produc-

ers also occasionally ship by rail, and CF ships a significant

amount of ammonia by barge. Farmland's production facili-

ties 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 Farm-

land's petition to intervene as a complainant.

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 alterna-

tives 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 agen-

cy, the ICC, and by the precedent developed under those

guidelines. Id. at 5 (citing Market Dominance Determina-

tions & 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-

__________

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.

fied, Product & Geographic Competition, 2 I.C.C.2d 1

(1985)).2

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 artic-

ulated 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, in-

cluding the "stand-alone cost" and "revenue adequacy" con-

straints. Id. at 6.

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.

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 agree-

ment CF had signed with Koch's predecessor, Gulf Central.

Id.

__________

2 See ICC Termination Act s 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. s 701

note.

3 "Notwithstanding the title..., those guidelines are not limited

to any one commodity. Coal cases, which typify captive, high-

volume, repetitive rail traffic, were the springboard for our analysis

because of the prevalence of coal rate challenges." Rate Guide-

lines--Non-Coal Proceedings, Ex Parte No. 347 (Sub.-No. 2), 1995

WL 705171 (I.C.C. Dec. 1, 1995).

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

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-

__________

4 In parts of its opinion not challenged in this court, the STB

found that Koch does not face effective: (1) "intramodal" competi-

tion 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 anhy-

drous 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 anhy-

drous 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.

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 increas-

es are not warranted." Id.

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 pre-

increase rates.

II

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. s 10707(a)); see also 49

U.S.C. s 15503(b)(3) (requiring STB, when prescribing pipe-

line rates, to consider, inter alia, "the availability of other

economic transportation alternatives"). " '[E]ffective compe-

tition['] ... 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" ap-

proach that rests on "case-by-case" analysis of competitive

alternatives to a carrier's facility. Id. at 119.

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 signif-

icant shift of anhydrous ammonia traffic from pipeline to

barge; that barge transport involves higher costs than pipe-

line transport, which could make a shift prohibitively expen-

sive; 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 transport-

ing a hazardous product like anhydrous ammonia. Id. at 8.

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.

Finally, "[a]s a measure of the effectiveness of barge

competition," the Board "compared Koch's revenues ... un-

der the old rate structure to its revenues under the new rate

structure, assuming all cost-competitive traffic would be di-

verted." 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 Domi-

nance, 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,

__________

6 According to CF's study, Koch could raise its rates an addi-

tional 20-50% without facing effective barge competition. Final

Order at 11 (citing CF exhibits and verified statements).

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

Koch attacks the STB's market dominance determination

on two grounds. First, it contends that the Board's compari-

son of its pre-and post-increase revenues was an inappropri-

ate test for market dominance. Second, it contends that the

Board disregarded record evidence of effective competition,

particularly Koch's own "matching" study.

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.

s 706(2)(A), (E)).

A

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 in-

crease its net revenues. We conclude that none of those

three adjectives appropriately characterizes the standard em-

ployed by the Board.

__________

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.

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 safe-

ty--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 quanti-

tative analysis as well. See Market Dominance, 365 I.C.C. at

119 n.5.9

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 prece-

dent 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

__________

8 Koch points out that, in a footnote, the STB stated: "While

we do not rely on" quantitative measures like the revenue compari-

son methodology "as a substitute for a thorough qualitative exami-

nation 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 substan-

tial shipper-related investments. 365 I.C.C. at 120-26. None of

those presumptions was employed in this case.

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 quota-

tion 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 complain-

ant 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

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

__________

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 con-

cluded that in enacting the market dominance inquiry of 49 U.S.C.

s 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."

market dominance guidelines, which are not themselves ques-

tioned 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-

__________

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.

s 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 Depart-

ment 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 s 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.

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 s 1.11 (utilizing 5% price-increase test).

Koch correctly points out that the ability to increase reve-

nues 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 competi-

tors." 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

__________

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 in-

creases 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

B

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

__________

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 differ-

ent 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 ques-

tion at issue: Are the MAPCO pipeline and barge carriers really

Koch's competitors? The Board found, and Koch has not chal-

lenged here, that the MAPCO pipeline is not an intramodal compet-

itor 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.

to the Board that barges were not serving as a competitive

alternative to pipeline transport for those retailers, but in-

stead were meeting different needs--such as providing addi-

tional anhydrous ammonia when sufficient supplies were un-

available 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.").

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 differen-

tial 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 competi-

tive 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

__________

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 substan-

tial 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 pro-

duction, transportation, or other costs among ... products prevent

one from operating as an effective competitive check on the prices

of the other").

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.

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.

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 barge-

delivered 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).

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 next-

closest 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 wit-

nesses. Compare Verified Reply and Rebuttal Statements of

C. Phillip Baumel on Behalf of Koch Pipeline Co., J.A. at 238-

42, 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 con-

stituting ineffective refutation, the Board acts within its au-

thority 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

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 interrelat-

ed, 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 expendi-

tures Koch would eventually have to make to replace the

pipeline.

In reviewing the STB's determination of rate reasonable-

ness issues, we again apply the deferential standards of the

Administrative Procedure Act, 5 U.S.C. s 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.

A

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

__________

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

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 "complain-

ants 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 pur-

portedly would have justified the company's rate increases.

Id. at 22.

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 con-

clusion to the contrary, however, is consistent with the agen-

cy'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 guide-

lines 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

__________

(3) the availability of other economic transportation alterna-

tives.

49 U.S.C. s 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. s 15503(a) (emphasis added).

[increase] is unreasonably high" (quoting Coal Rate Guide-

lines, 1 I.C.C.2d at 548)).18

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 guide-

lines and is not subject to reversal by this court. See Auer v.

Robbins, 519 U.S. 452, 461 (1997).

B

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 consider-

able 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

__________

18 The Board's position is not inconsistent with our decisions in

Burlington and PEPCO. In Burlington, we remanded a ratemak-

ing 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 con-

straints are applicable (they were not applicable in PEPCO),

"stand-alone cost is not the only ceiling on rates." Potomac Elec-

tric 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 "predeces-

sor" cost "represents the cost of the asset when it was first

dedicated to public service, plus any subsequent improvements, less

depreciation and retirements").

valuation--the amount [Koch] paid in an arm's-length trans-

action" to acquire the pipeline. Final Order at 24.

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 acquisi-

tion 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.

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 invest-

ment." 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

__________

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 'pur-

chase,' " 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 reason-

ableness 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.

$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

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 replac-

ing assets (or constructing them anew)." Koch Br. at 30.

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, howev-

er, 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 com-

pel the use of replacement cost in the SAC constraint do not

apply to the independent constraint of revenue adequacy.

Koch also contends that the STB should have used replace-

ment 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 facili-

ties 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 "re-

quir[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

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 meth-

odology 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).

Koch correctly notes that during the 1980s, the ICC active-

ly "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

__________

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.

be confident would produce accurate and reliable results."

Standards for Railroad Revenue Adequacy, 3 I.C.C.2d 261,

277 (1986).

In promulgating the 1990 guidelines cited by the STB in

this case, the ICC again considered the appropriate invest-

ment 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 (adopt-

ing view of Railroad Accounting Principles Board). Accord-

ingly, the Commission announced that it would "use acquisi-

tion 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

__________

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 regula-

tions. See Coal Rate Guidelines, 1 I.C.C.2d at 535 (using Railroad

Revenue Adequacy standards to define the revenue adequacy con-

straint 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 determina-

tions. Final Order at 25 n.64. In that case, the ICC used replace-

ment cost because "an arm's length purchase price" cost could not

be determined as a consequence of the way in which the purchase

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).

We conclude that, in assessing the legality of Koch's rate

increase, the STB employed a reasonable ratemaking method-

ology 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

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.

s 1104.11.

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. s 15901(c) requires that this inves-

tigation 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.

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

__________

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.

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

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 South-

West 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

Koch's challenge to the final decision of the Board is

essentially a narrow one, questioning not the STB's regula-

__________

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 com-

plaint brought by CF." CF Indus., Inc., No. 41685, slip. op. at 1

(S.T.B. July 25, 1996).

tions but rather their application in a single ratemaking. We

conclude that the STB's finding that Koch was market domi-

nant 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 rate-

making 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

Denied.