TEXAS MUNICIPAL POWER V. THE BNSF RY. CO. 803
7 S. T. B.
STB DOCKET NO. 42056
TEXAS MUNICIPAL POWER AGENCY
v.
THE BURLINGTON NORTHERN AND SANTA FE RAILWAY
COMPANY
Decided September 24, 2004
Upon reconsideration, the Board modifies the rate prescription and
reparations award in this proceeding.
BY THE BOARD
In this proceeding, Texas Municipal Power Agency (TMPA) challenged
the reasonableness of the rate charged by The Burlington Northern and Santa
Fe Railway Company (BNSF) for transportation of coal in unit trains from
certain mine origins in the Powder River Basin (PRB) of Wyoming to
TMPA’s Gibbons Creek Steam Electric Station at Iola, TX. In Texas Mun.
Power Agency v. The Burlington N. & S. F. Ry. Co. 6 S.T.B. 573 (2003)
(TMPA 2003), the Board found that BNSF has market dominance over that
transportation and that the challenged rate was unreasonably high. Based
upon a stand-alone cost (SAC) analysis, the Board prescribed maximum
reasonable rates through the year 2011 and awarded reparations to TMPA.
TMPA and BNSF have each filed timely petitions for reconsideration of
various aspects of that decision. They ask the Board to reconsider various
substantive determinations made in TMPA 2003 and to correct several
asserted technical errors. The issues raised, which are discussed in turn below,
involve:
< whether BNSF should have been required to formally establish a rate
for transportation in shipper-supplied railcars and to use those cars in
place of its own railcars (Section I);
< the variable costs associated with maintenance-of-way, return on
road property and depreciation, and locomotive fuel expense
(Section II);
< those portions of the SAC analysis addressing rerouting of traffic,
revenue forecasts, locomotive fueling expense, general and
administrative expenses, bridges, utility relocation, road resurfacing,
at-grade crossings, and fencing (Section III);
< the asserted technical errors (Section IV); and
SURFACE TRANSPORTATION BOARD REPORTS804
7 S. T. B.
< the scope of the rate prescription (Section V).
Three industry associations—Western Coal Traffic League (WCTL),
National Industrial Transportation League (NITL), and North America Freight
Car Association (NAFCA) (collectively, the “associations”)—have filed
petitions to intervene in support of TMPA’s request for reconsideration with
respect to the first set of issues listed above. The associations will be allowed
to participate as amici, but not intervene as parties to this private rate dispute.
As amici, they may present their views on, and support of, issues raised by
TMPA in its petition for reconsideration. They will not, however, have
access to confidential information; nor will they be allowed to broaden the
issues in the proceeding. See Arizona Pub. Serv. Co. & Pacificorp v. The
Burlington N. & S. F. Ry. Co., 7 S.T.B. 71 (2003).
As discussed below, upon reconsideration we correct two technical errors
in the SAC computation and we revise the SAC analysis in one respect (a
modest increase in the level of executive salaries) to conform with the record
in this case. We revise the rate prescription and reparations accordingly, and
we clarify that the rate prescriptions extend to all mine origins in TMPA’s
complaint. The remaining reconsideration requests are denied.
DISCUSSION AND CONCLUSIONS
A party may seek to have the Board reconsider a decision by submitting a
timely petition that presents new evidence or changed circumstances that
would materially affect the case or that demonstrates material error in the
prior decision. 49 U.S.C. 722(c); 49 CFR 1115.3 (timely administrative
appeals). Where, as here, a petition is timely filed within 20 days after the
service date of the prior decision, the prior decision does not become
administratively final for purposes of seeking judicial review until the Board
has acted on the administrative appeal. See American Farm Lines v. Black
Ball, 397 U.S. 532, 541 (1970); ICC v. Brotherhood of Locomotive Engineers,
482 U.S. 270, 284-85 (1987).
The Board encourages parties to file timely administrative appeals where
they believe the Board has erred in some respect. The courts favor such
administrative reconsideration, as it is a more expeditious and efficient means
of achieving corrections or adjustments than resort to the federal courts in the
first instance. See Commonwealth of Pennsylvania v. ICC, 590 F.2d 1187,
1194 (D.C. Cir. 1978); B. J. Alan Co. v. ICC, 897 F.2d 561, 562-63 n.1 (D.C.
Cir. 1990). And even where judicial review is later sought, the reconsideration
process can assist the court’s review process by providing a further
exploration of the issues in advance. See American Farm Lines, 397 U.S. at
541; McCarty Farms, Inc. v. STB, 158 F.3d 1294, 1301 n.1 (D.C. Cir. 1998)
(further Board decision was “helpful to the court”).
On the other hand, the Board generally does not consider new issues
raised for the first time on reconsideration where those issues could have and
should have been presented in the earlier stages of the proceeding. Moreover,
the term “new evidence” refers to evidence that was not reasonably available
to the party when the record was developed, and not simply newly raised. See
TEXAS MUNICIPAL POWER V. THE BNSF RY. CO. 805
1
TMPA Open. WP. at 4121 (TMPA letter dated October 17, 2000).
2
TMPA Open. WP. at 4123 (BNSF letter dated November 1, 2000).
3
TMPA Open. WP. at 4126-7 (TMPA letter dated January 8, 2001).
4
TMPA Open. WP. at 4131 (TMPA letter dated February 27, 2001).
5
TMPA Open. WP. at 4132 (BNSF letter dated March 12, 2001).
6
Section 10702 provides that “[a] rail carrier providing transportation or service subject to the
jurisdiction of the Board under this part shall establish reasonable (1) rates * * * and (2) rules and
practices on matters related to that transportation or service.”
7 S. T. B.
Friends of Sierra R.R., Inc. v. ICC, 881 F.2d 663, 667 (9th Cir. 1989).
Nothing in the statute or the Board’s regulations obliges the agency to rethink
its decisions whenever a party wishes to try out a new theory or finds new
information at a late stage in the process. Connecticut Trust for Historic
Preservation v. ICC, 841 F.2d 479, 484 (2d Cir. 1988). And if a party were
free to reshape its case, so long as it did so within 20 days after a decision, the
administrative process might never end. The agency is not expected to
“behave like Penelope, unraveling each day’s work to start the web again the
next day.” Western Coal Traffic League v. ICC, 735 F.2d 1408, 1411 (D.C.
Cir. 1984).
I. Rate for Transportation in Shipper-Owned Cars
Historically, BNSF has served TMPA’s coal movements from the PRB to
the Gibbons Creek station in three train sets of BNSF-furnished railcars. In
2000, TMPA asked BNSF to establish a separate rate for movements in
shipper-supplied cars. TMPA advised BNSF that it was prepared to supply
the equipment “if the rate differential justified the acquisition costs.”
1
BNSF,
however, chose to establish a rate only in railroad-furnished cars, explaining
that it had the equipment available and had chosen to use that equipment for
this service.
2
Two months later, TMPA began objecting to a perceived drop-off in
BNSF’s service level. In a series of correspondence with BNSF, TMPA
complained that it was becoming critically short on its coal requirements due
to BNSF’s inability to provide an adequate number of train sets and crews to
meet TMPA’s needs. BNSF advised TMPA that the larger volumes of coal
coming out of the PRB required more trains and crew, but that congestion on
its network made adding a fourth train set for TMPA inadvisable. Instead,
BNSF instituted some “short-term plans” that, TMPA acknowledged, “seem
to be easing our inventory shortages.”
3
Nonetheless, in light of its past
inventory concerns, TMPA renewed its request for a rate in TMPA-supplied
cars.
4
BNSF again advised TMPA that it had sufficient equipment to fulfill
its common carrier obligations and that, if it should become necessary to use
TMPA-supplied cars, BNSF would accept them, applying the same rate it
used for movements in BNSF-supplied cars.
5
In its complaint challenging the BNSF rate for movements in railroad-
supplied cars, TMPA also charged that BNSF’s refusal to use TMPA-supplied
equipment and to formally establish a rate for shipments in shipper-supplied
cars constituted an unreasonable practice, in violation of 49 U.S.C. 10702.
6
TMPA argued that at least one trainset of shipper-supplied cars is needed to
SURFACE TRANSPORTATION BOARD REPORTS806
7
TMPA Open. Narr. at 34.
8
Id.
7 S. T. B.
supplement the BNSF fleet and satisfy TMPA’s service requirements.
7
TMPA further argued that using shipper-supplied cars for one or more
trainsets of its traffic would not cause the railroad’s cars to be idled, because
they could be shifted immediately to alternative productive uses elsewhere.
8
In TMPA 2003, 6 S.T.B. at 581-82, the Board declined to require BNSF
to accept TMPA-supplied cars. The Board explained that railroads have the
right to use their own railcars so long as they can meet their common carrier
obligations with those cars. See Atchison, Topeka & S. F. Ry. Co. v. United
States, 232 U.S. 199, 214-15 (1914) (Atchison); Shippers Comm., OT-5 v. Ann
Arbor R.R., 5 I.C.C.2d 856, 865 (1989) (SCOT-5), aff’d, Shippers Comm. OT-
5 v. ICC, 968 F.2d 75 (D.C. Cir. 1992). Examining the record in this case, the
Board concluded that TMPA had not shown that BNSF cannot meet its
common carrier obligations using its own railcars.
TMPA and the three amici have sought reconsideration, raising two
distinct issues: (A) whether the carrier can lawfully decline to accept TMPA-
supplied cars; and (B) even if the carrier is not required to accept shipper-
supplied railcars at this time, whether the railroad nonetheless should be
required to establish a rate for such movements. Although the parties often
merge these two issues, they present distinct concerns and will be discussed
separately.
A. Use of TMPA-Supplied Cars
In its petition for reconsideration, TMPA reiterates its complaints about
the inconsistency of BNSF’s service and its prediction that the railroad’s cars
would not be idled if TMPA cars were used. However, TMPA has failed to
demonstrate that BNSF has not met and cannot continue to meet its common
carrier obligations in railroad-furnished cars. Rather, the evidence
demonstrates that congestion on the rail lines was the cause of TMPA’s
inventory problems and that adding more trainsets to the already congested
network would not have alleviated that situation. The argument that BNSF
might find alternative uses for its railcars, even if correct, is irrelevant to this
inquiry. For absent evidence that BNSF does not have sufficient railcars
available to serve TMPA, BNSF has no common carrier obligation to use
shipper-supplied cars. And, given the lack of evidence that BNSF-supplied
railcars were unavailable, we cannot find that BNSF engaged in an
unreasonable practice, in violation of 49 U.S.C. 10702, by favoring the use of
its own railcars over railcars to be provided by TMPA.
WCTL argues that this analysis does not hold up for unit-train service.
WCTL cites Potomac Electric Power Co. v. Penn Central Transp. Co.,
356 I.C.C. 815 (1977) (PEPCO), aff’d in relevant part, Potomac Electric
Power Co. v. United States, 584 F.2d 1058 (D.C. Cir. 1978) for the
proposition that railroads do not have a common carrier obligation to provide
unit-train service in railroad-supplied cars. And because a carrier’s right to
TEXAS MUNICIPAL POWER V. THE BNSF RY. CO. 807
9
TMPA Open. Narr. at 34; see also id. at 117; TMPA Reply Narr. at 78; TMPA Br. at 29.
7 S. T. B.
favor the use of its own railcars derives from its obligation to provide cars for
that service, WCTL concludes that railroads may not refuse to accept shipper-
supplied cars for unit-train service.
In PEPCO, the shipper had asked the defendant railroad (Penn Central
Transportation Company) to establish a rate for “unit-train” service in
railroad-supplied cars. Instead, Penn Central established one tariff rate for
“train-load” service in railroad-supplied cars and another rate for unit-train
service in shipper-supplied cars. The train-load service Penn Central offered
in its own railcars was nearly identical to the unit-train service PEPCO
sought: both called for shipments of over 7,000 tons of coal per train from
eastern mines to PEPCO’s three electric generating plants in railroad-supplied
cars. But in unit-train service railcars and locomotives are joined for
uninterrupted, round trip, shuttle-type service, and Penn Central did not want
to commit its own railcars to the repeated movement of a single unit-train.
Penn Central preferred the greater operating flexibility of train-load service,
whereby it could use those railcars to serve other shippers if needed.
PEPCO challenged Penn Central’s refusal to dedicate the railroad’s cars
to unit-train service as an unreasonable practice. The Board’s predecessor,
the Interstate Commerce Commission (ICC), found no merit to PEPCO’s
claims, as the record clearly established that Penn Central was providing
comparable train-load service to PEPCO in railroad-supplied cars. PEPCO,
356 I.C.C. at 827 (1977). The court of appeals agreed. Potomac Elec. Power
Co., 584 F.2d at 1063 (“In these circumstances, the railroad’s refusal to hold
out a complete unit-train service did not violate the Interstate Commerce
Act.”).
Thus, PEPCO does not stand for the broad proposition that a carrier is
never required to provide service in unit trains. In PEPCO, the ICC made a
case-specific determination that it was reasonable for the carrier to offer
trainload service, instead of unit-train service, where that service applied from
the same origin at the same minimum weight as the unit-train service. See
PEPCO, 365 I.C.C. at 827. To the extent that PEPCO stands for a general
proposition, it is that how a railroad satisfies its common carrier obligation is
left to the railroad to decide in the first instance. So long as the railroad offers
service that satisfies its common carrier obligations (the critical inquiry), it
need not provide the particular service that the shipper would prefer.
In this case, the issue of whose equipment is to be used appears to be
more about rate levels than service. TMPA complains that, in insisting on
using the railroad’s cars, BNSF is “leveraging its monopoly over TMPA’s
coal transportation service to collect a guaranteed, 80% mark-up on its
variable railcar costs.”
9
Under 49 U.S.C. 10707(c)(1)(A), a railroad may
charge a captive shipper up to 180% of its variable cost of serving that shipper
without regulatory intervention, and the Board cannot prescribe a rate that
would yield revenues below that level. See West Texas Util. Co. v. Burlington
N. R.R., 1 S.T.B. 638, 677 (1996) (West Texas), aff’d sub nom. Burlington N.
SURFACE TRANSPORTATION BOARD REPORTS808
10
The term “OT-5” refers to procedures published at the time in the Official Railway
Equipment Register, OT-5, Circular E, for assigning reporting marks and handling private cars.
Under those procedures, a noncarrier would file an application with the Association of American
Railroads, which would forward it to the carrier on whose lines the cars were to be placed in service.
The originating carrier had the discretion to grant or deny OT-5 approval.
7 S. T. B.
R.R. v. STB, 114 F.3d 206 (D.C. Cir. 1997). Therefore, it may indeed be more
profitable for a railroad to use its own railcars to provide service to a captive
shipper, at least when a rate prescription is set at the regulatory floor.
But when Congress established the 180% revenue-to-variable (R/VC)
floor, it was well settled that railroads may use their own railcars in
preference to shipper-supplied cars. As far back as 1914, the law has been
clear that railroads have the right to use their own cars so long as they can
meet their common carrier obligations with those cars. Atchison, 232 U.S. at
214-15. Because Congress legislates against the backdrop of then-existing
law, one must presume that, in enacting 49 U.S.C. 10707(d)(1)(A), Congress
intended to permit railroads to charge captive shippers 180% of the variable
cost of supplying the railcars. See, e.g., United States v. Baxter Int’l, Inc.,
345 F.3d 866, 900 (11th Cir. 2003) (“We presume that Congress legislates
against the backdrop of established principles of state and federal common
law, and that when it wishes to deviate from deeply rooted principles, it will
say so.”).
Although TMPA may want a lower rate for transportation using its own
cars, the Supreme Court has held that a railroad is under no obligation to
provide a lower rate for service in shipper-supplied cars. As the Court has
explained, because railroads must stand ready as common carriers to provide
service in railroad-supplied cars (see 49 U.S.C. 11101, 11121) they “cannot
be compelled to accept [railcars] tendered by the shipper on condition that a
lower freight rate be charged.” Atchison, 232 U.S. at 214-15.
B. Rate for TMPA-Supplied Cars
NITL/NAFCA argue that the railroad should nonetheless be obliged to
provide a rate for transportation in shipper-supplied cars (TMPA 2003,
6 S.T.B. at 581-82) in light of the balance struck in SCOT-5. In that case, a
shipper group filed a complaint challenging the practices of most of the
nation’s railroads regarding the use and control of privately owned covered
hopper cars in grain transportation. The complainants alleged, among other
things, that the railroads had defaulted on their common carrier obligation by
not supplying a sufficient number of railcars when requested by shippers.
They also objected to the industry practice of sometimes denying “OT-5
approval” for assigning reporting marks to privately owned railcars—which is
required before private cars can be put onto the rail system—solely for
economic reasons.
10
The ICC concluded that it was an unreasonable practice for carriers to
deny such approvals for reasons other than safety, mechanical factors, or
inadequate track storage space. SCOT-5, 5 I.C.C.2d at 864 (1989). The ICC
explained that there should be no artificial impediment to the use of private
cars when carrier cars are unavailable. But the ICC also made clear that
TEXAS MUNICIPAL POWER V. THE BNSF RY. CO. 809
11
TMPA Open. WP. at 4132 (BNSF letter dated March 12, 2001).
12
See Procedures To Expedite Resolution of Rail Rate Challenges To Be Considered Under
the Stand-Alone Cost Methodology, 6 S.T.B. 805 (2003).
7 S. T. B.
“[a]llowing shippers freely to register private cars under the OT-5 process
does not, in itself, require the railroads to use private cars when the railroads
have cars of their own available.” Id. at 870.
Here, NITL/NAFCA argue that, just as the ICC found it unlawful for
carriers to routinely refuse OT-5 approval for cars that the owners were ready
to put into the rail system, the Board should find that a carrier must establish a
common carrier rate for transportation using those cars. Otherwise, the 10-
day period that railroads have for establishing a common carrier rate (see
49 CFR 1300.3) would create another artificial barrier to the use of privately
supplied railcars.
While there may be merit to that argument, we need not and do not
resolve that issue here, because the delay concerns expressed by
NITL/NAFCA do not apply in this case. The record indicates that BNSF has
already informed TMPA of the common carrier rate it would charge, at least
initially, for shipments in TMPA-supplied railcars should movements in
shipper-supplied cars become necessary
11
—a fact that the amici would not
have known, as the evidence in this case was filed prior to the Board’s
clarification of the requirement that parties file public versions of their
evidence in rate cases,
12
and a fact that was not reported in TMPA 2003. The
rate the carrier has quoted for movements in shipper-supplied cars is the same
as the rate for movements in railroad-supplied cars, but, as previously
discussed, BNSF cannot be compelled to accept shipper-supplied railcars on
the condition that a lower freight rate be charged. Atchison, 232 U.S. at
214-15.
II. Variable Cost Issues
The Board may consider the reasonableness of a challenged rate only if
the carrier has market dominance over the traffic involved. 49 U.S.C.
10701(d)(1), 10707(b). Market dominance is “an absence of effective
competition from other carriers or modes of transportation for the
transportation to which a rate applies.” 49 U.S.C. 10707(a). The statute
precludes a finding of market dominance, however, where the carrier shows
that the revenues produced by the movements at issue are less than 180% of
the carrier’s variable costs for providing that service. 49 U.S.C.
10707(d)(1)(A).
The variable costs associated with the traffic at issue also determine the
floor for regulatory rate relief, because the Board cannot prescribe a rate that
is below the 180% R/VC jurisdictional floor. See West Texas, 1 S.T.B. at 677
(1996). Therefore, the rate prescribed in this case was the higher of the rate
level produced by the SAC analysis or the regulatory floor (the 180% R/VC
rate level). See TMPA 2003, 6 S.T.B. at 608-11.
Under 49 U.S.C. 10707(d)(1)(B), a carrier’s variable costs are to be
determined using the Board’s Uniform Railroad Costing System (URCS),
SURFACE TRANSPORTATION BOARD REPORTS810
7 S. T. B.
with adjustment where appropriate. URCS is a general purpose costing model
used to determine the variable cost of a movement. The model determines,
for each Class I railroad, what portion of each category of costs shown in its
Annual Report to the Board (STB Form R-1) represents its system-average
variable unit cost for that cost category for that year. In considering whether
to allow adjustments to the system-average variable costs produced by URCS,
the Board evaluates whether the party proposing to use a different figure has
shown that its proposed figure would better reflect the variable costs of
serving the particular traffic at issue than the URCS system-average figure.
These adjustments are known as “movement-specific” adjustments.
In this case, each party proposed various movement-specific adjustments
to the URCS system-average figures. In TMPA 2003, the Board accepted
some of the adjustments proposed by the parties but not others. The Board
then concluded that the challenged rate produces revenues that exceed 180%
of the variable costs of providing service from the two mine origins for which
traffic data had been submitted. In their petitions for reconsideration, each
party objects to certain component parts of our variable cost analysis. Their
objections are discussed individually below.
A. Maintenance-of-Way Costs
To calculate the variable costs associated with maintenance-of-way
(MOW), the Board used system-average URCS rather than the “speed
factored gross ton” (SFGT) formula accepted in prior cases. The original
SFGT formula was developed in 1973 to allocate MOW costs between
passenger and freight traffic. It was based on research on track degradation
and MOW activities from 1950 to 1970, and the precise parameters of the
formula were estimated using those data and standard regression analysis.
See generally Nat’l R.R. Passenger Corp. & Consolidated Rail Corp.,
10 I.C.C.2d 863, 872-74 (1995) (Amtrak). The Board had accepted
adjustments based on the SFGT formula in prior SAC cases. But in this case,
the Board concluded that the SFGT formula is so outdated as to no longer be
reliable. TMPA 2003, 6 S.T.B. at 633.
TMPA objects to the departure from precedent. It argues that use of the
SFGT formula was well-settled and that the Board has previously rejected
arguments that the formula was out-of-date. TMPA also contends that the
record contained evidence demonstrating that the SFGT formula accurately
calculates average MOW costs because of adjustments that have been made
to the formula to reflect changes in maintenance practices, heavier wheel
loads, and traffic densities. To justify some movement-specific adjustments
for these high-density lines, TMPA pointed to evidence that BNSF’s system-
average MOW expense has fallen over time, allegedly showing that there
remain economies of density in the maintenance of rail lines.
It is undisputed that much has changed since the original SFGT formula
was derived. As BNSF had pointed out, in 1983 the ICC changed the
accounting system that is used by railroads from a retirement, replacement,
and betterment system to a depreciation accounting system, which treats a
significantly greater portion of maintenance as capital cost (as opposed to an
TEXAS MUNICIPAL POWER V. THE BNSF RY. CO. 811
13
See TMPA Open Narr. at 73.
7 S. T. B.
operating expense) than the prior accounting system had. Track materials
have become more durable. In 1978 the average rail line had 111-lb. rail,
only 6% of which was continuous welded rail (CWR); by 1999, the average
weight of rail was 125-lb., and 62% of that was CWR. And industry
maintenance practices have changed over the years, so that more of the
maintenance that is being performed is planned maintenance, which is
capitalized under depreciation accounting. TMPA 2003, 6 S.T.B. at 633 n.92.
Furthermore, the average densities of the railroads of the 1970’s bear no
relation to the average densities of the railroad today. The SFGT formula
reflected railroad lines with an average density of 10-15 million gross tons
(MGT), and densities of 25 MGT were relatively rare when that formula was
developed. See Amtrak, 10 I.C.C.2d at 877. In contrast, in 2000 the average
density along the route BNSF uses for TMPA’s traffic was approximately 60
MGT. In short, present-day railroad maintenance expenses and practices bear
little resemblance to those of 30 to 50 years ago.
The SFGT formula, however, has not been re-benchmarked to address
these changes. And scholars have long cautioned of the dangers of using a
regression analysis to extrapolate beyond the range of existing data. See, e.g.,
Rudolf Freud & William Wilson, Regression Analysis: Statistical Modeling of
a Response Variable (1998), at 65 n.53. See also Amtrak, 10 I.C.C.2d at 877
(“Applying [the SFGT] regression results to circumstances outside the
relevant range of data upon which the regression equations are based may not
produce valid results.”).
The fact that the formula has been modified by consultants does not
correct the problem. To the contrary, such changes could render the new
formula less reliable than the original formula. For example, the formula was
changed to add an “R-Factor”—an index purportedly designed to reflect
current maintenance practices, price levels, and departmental overheads.
13
The R-Factor was derived by comparing the SFGT formula with the system-
average MOW expenses of the carrier. For example, if the original SFGT
formula would yield a system-average MOW figure of $100 million for a
carrier, when in fact its MOW expense was $150 million, the R-Factor would
be 1.5. The formula would then be indexed by this percent markup (or
markdown as the case may be). In this case, TMPA applied an R-Factor of
0.86, indicating that the original SFGT formula overstates BNSF MOW
expenses by 14%. But the extent to which the adjustment yields reliable
results or simply masks the unreliableness of the old formula is impossible to
determine. In its circular fashion, the R-Factor does ensure that the SFGT
formula produces accurate system-average MOW expenses, but there is no
evidence that this adjustment ensures that the formula accurately models the
relationship between MOW expenses and density with respect to individual
lines.
In sum, there is no evidence that the SFGT formula produces MOW
expenses that are comparable to current actual MOW costs for any rail line.
Compare San Antonio v. Burlington N. R.R., STB Docket No. 36180 (ICC
served April 11, 1986) (SFGT-calculated MOW costs were shown to be
SURFACE TRANSPORTATION BOARD REPORTS812
7 S. T. B.
within 8% of BNSF’s actual 1978 MOW expenditures). Thus, although the
SFGT formula originally fit the data well, the passage of time and profound
changes in this industry have plainly rendered unreliable the continued use of
the SFGT formula. We need not use an outdated formula forever simply
because it was accepted in prior cases. See Atchison, T. & S. F. Ry. v. Wichita
Board of Trade, 412 U.S. 800, 808 (1973); JSG Trading Corp. v. USDA,
176 F.3d 536, 544 (D.C. Cir. 1999) (an agency is not strictly bound to follow
the test applied in prior cases, so long as it articulates a principled rationale
for departing from that test).
B. Return on Road Property and Depreciation Expense
In TMPA 2003, 6 S.T.B. at 635, the Board accepted, with one correction,
the movement-specific adjustments that TMPA had proposed for calculating
variable road property return-on-investment and depreciation expense. The
Board found that TMPA’s adjustments were reasonable and supported, and
that BNSF had failed to discredit them. On reconsideration, BNSF argues
that the Board should have followed the precedent in Wisconsin Power &
Light Co. v. Union Pacific R.R., STB Docket No. 42051 (STB served May 14,
2002) (WPL Recons.) at 4-6, where, to calculate the MOW component of
variable cost, the Board rejected a similar aggregate approach in favor of a
segment-by-segment approach. Although the determination in WPL Recons.
dealt with a different variable cost expense, BNSF contends that the same
type of analysis should be applied to road property expenses.
However, BNSF never proposed that the Board use a segment-by-
segment procedure when making movement-specific adjustments to the road
property expense item. Rather, it argued (unsuccessfully) that the Board
should not make any movement-specific adjustment to this expense item. (In
contrast, in the WPL Recons. case, the Union Pacific Railroad Company’s
initial case had included movement-specific adjustments to the MOW
expense item using a segment-by-segment weighted allocation procedure.) It
is inappropriate for BNSF to raise an issue in a petition for reconsideration
that it chose not to raise initially. Furthermore, the issue in WPL Recons. is
not identical to the issue sought to be presented here, as it related to a
different expense item. The Board did not have the benefit of a full record
and debate on whether a segment-by-segment approach would be appropriate
for road property expenses. As BNSF failed to challenge the particular
manner in which TMPA performed its movement-specific adjustment below,
its request for reconsideration of the calculation of variable road property
expenses is denied.
C. Locomotive Fuel Expense
TMPA seeks reconsideration of the Board’s determination with respect to
locomotive fueling expense, in TMPA 2003, 6 S.T.B. at 635-36. Before the
filing of evidence, the Board had ordered BNSF to let TMPA conduct a
TEXAS MUNICIPAL POWER V. THE BNSF RY. CO. 813
14
Texas Mun. Power Agency v. Burlington N. & S. F. Ry. Co., STB Docket No. 42056 (STB
served February 9, 2001) at 8.
15
TMPA Open. Narr. at 96-98; TMPA Reply Narr. at 71-72; TMPA Reb. Narr. at 96, 186.
16
TMPA Reply Narr. at 72 n.66; TMPA Reply e-WP. “Fuel Summary Reply.123”; TMPA
Reb. Narr. at 193.
17
TMPA Open. Narr. at 96-97.
7 S. T. B.
special study of locomotive fuel consumption.
14
The parties had negotiated
and agreed upon a study methodology, which was tested and validated by
TMPA’s witness. The special study examined a three-locomotive consist (the
number of locomotives used to move trains transporting TMPA’s traffic)
using an event recorder. This device documented the amount of time that a
locomotive consist operated at a particular throttle setting. Based on
manufacturer-specified fuel consumption levels at different throttle settings,
an estimate of the gallons of fuel consumed to serve TMPA was developed.
TMPA, however, was not satisfied with the results of this study, which
showed fuel consumption rates higher than BNSF’s system-average rates.
TMPA argued that the study results were too high because certain operating
characteristics of the study trains (number of locomotives in consist and cycle
time) did not accurately portray the characteristics of its traffic when
compared to train movements conducted in the ordinary course of business.
15
TMPA therefore proposed adjustments to the fuel consumption data based on
a three-locomotive consist and 2001 train cycle times.
16
Specifically, TMPA
increased the running time and decreased the idle time of the study
locomotives, and decreased the number of locomotives consuming fuel, based
on a presumption that fuel consumption varies directly with the number of
locomotives and cycle time.
17
The Board was not persuaded that TMPA’s adjustments produced a more
accurate picture of the variable locomotive fuel cost than did the unadjusted
study results (with exclusion of a few anomalous study observations).
Accordingly, the Board rejected TMPA’s adjustments and instead used the
results of the jointly conducted fuel study to determine the variable
locomotive fuel expense.
On reconsideration, TMPA again argues that discrepancies between the
operating characteristics of TMPA trains moving during the fuel study period
and those moving in the ordinary course of business required adjustments to
produce accurate variable cost results. TMPA contends that its proposed
adjustments were consistent with BNSF’s service in the period from April to
November 2001, relying on a chart comparing average running and idle times
extrapolated from train movement records during this post-contract period
with the running and cycle times of the shorter, encapsulated study period
(July to September 2001).
While there may have been differences in cycle times and the number of
locomotives in the consists during the study period as compared to other time
periods, TMPA’s proposed adjustments would not properly correct for any
such discrepancies. TMPA’s adjustments were based on the assumption that
fuel consumption is directly variable with cycle time. That is not the case,
however. It is the locomotive’s throttle position that determines the amount
of fuel consumed, not simply the aggregate time spent running as opposed to
SURFACE TRANSPORTATION BOARD REPORTS814
18
TMPA Reply Narr. at 77 (acknowledging that the fuel study measured fuel consumption as
the product of the rate of fuel consumption at a given throttle setting multiplied by the amount of
time the locomotive runs at that throttle setting).
19
BNSF Reb. V.S. Fisher at 50.
7 S. T. B.
idling.
18
The throttle position, in turn, is dependent on such factors as the
weight and makeup of the train, the grade and curvature over which the train
is moving, train spacing over the line, and weather and track conditions.
Indeed, BNSF submitted a scatter diagram showing the absence of a linear
relationship between locomotive running time and fuel consumption for the
fuel study trains.
19
TMPA alleges that its proposed adjustments were necessary because the
original study suggested the anomalous result of fuel consumption at levels
above system-average using BNSF’s most efficient locomotives. But in light
of the fact that TMPA’s trains are heavier than BNSF’s average trains, thus
requiring greater tractive effort for movement, it is not unreasonable for the
joint fuel study to have shown that fuel consumption was slightly above
system-average. Accordingly, TMPA has failed to show that it was improper
for the Board to use the results of the locomotive fuel expense study.
III. Stand-Alone Cost Issues
The Board’s finding that the challenged rate was unreasonable and its
determination of the maximum reasonable rate were based on a SAC analysis.
To make a SAC presentation, a shipper designs a stand-alone railroad
(“SARR”) specifically tailored to serve an identified traffic group, using the
optimum physical plant or rail system needed for that traffic. Based on the
traffic group, services to be provided, and terrain traversed, a detailed
operating plan must be developed to define further the physical plant that
would be needed for the SARR. It is assumed that investments normally
would be made prior to the start of service and that recovery of the
investments would occur over the economic life of the assets. A computerized
discounted cash flow (DCF) model simulates how the SARR would likely
recover its capital investments, taking into account inflation, Federal and state
tax liabilities, and a reasonable rate of return. The annual capital costs are
combined with the annual operating costs to calculate the total annual revenue
requirements of the SARR.
The defendant carrier’s rates are then judged against the revenue
requirements of the SARR. Traffic and rate level trends for the traffic group
are forecast into the future to determine the future revenue contributions from
that traffic. By comparing the total revenues generated by the traffic group
against the revenue requirements of the SARR, the Board determines whether
there would be over- or under-recovery of costs. Because the analysis period
is lengthy, a present value analysis is used that takes into account the time
value of money, netting annual over-recovery and under-recovery as of a
common point in time. If the revenues from the traffic group are determined
to be less than the revenue requirements of the SARR, then the challenged
rates are considered reasonable.
TEXAS MUNICIPAL POWER V. THE BNSF RY. CO. 815
7 S. T. B.
The parties seek reconsideration of several components of the SAC
analysis conducted in this case. Their objections are discussed below.
A. Rerouting of Traffic
To increase the amount of traffic that would flow over the hypothetical
SARR in this case, referred to as the Gibbons Creek Railroad (GCRR),
TMPA proposed an operating plan for the GCRR that would transport certain
non-issue traffic using a different route than the route BNSF uses for that
traffic. The rerouted traffic consisted of PRB coal destined to the Parrish,
Nelson, Fayette, and Big Brown utility plants. For these movements, BNSF
uses its Front Range route, which proceeds southwest along the Rocky
Mountains through Denver, CO, then southeast into Texas and Louisiana.
However, to achieve maximum traffic densities for its SARR, TMPA
hypothesized that the GCRR would haul that traffic over the somewhat
longer, but higher-density, Central Corridor route, which proceeds east from
Northport through Kansas City, and then south into Oklahoma and Texas.
The map below, reproduced from TMPA 2003, 6 S.T.B. at 593, illustrates the
proposed rerouting.
SURFACE TRANSPORTATION BOARD REPORTS816
7 S. T. B.
Proposed Rerouting
This rerouting proposal presented an issue of first impression. The Board,
guided by the underlying purpose and objectives of the SAC test, see Coal
Rate Guidelines, Nationwide, 1 I.C.C.2d 520 (1985) (Guidelines), aff’d sub
nom. Consolidated Rail Corp. v. United States, 812 F.2d 1444 (3d Cir. 1987),
first set forth general principles relating to the rerouting of traffic. TMPA
2003, 6 S.T.B. at 594-98. The Board concluded that a SAC analysis may
reroute traffic that the SARR would handle entirely from origin to destination,
so long as the proposed route is reasonable and would meet the shipper’s
transportation needs. But to reroute cross-over traffic (i.e., traffic for which
the SARR would not replicate the full length of the defendant carrier’s current
move), the SAC analysis must fully account for the ramifications of assuming
that the residual carrier (i.e., the portion of the defendant carrier’s system that
would not be replaced by the SARR) would alter its handling of the traffic.
Applying these principles, the Board permitted the proposed rerouting of
the Big Brown traffic, which the GCRR would carry entirely from the origin
mines to the plant, because the evidence demonstrated that the GCRR would
provide the same or superior service as the shipper receives from BNSF.
However, because that traffic moves in shipper-supplied cars, the Board
reduced the revenues from the Big Brown movement to account for the
additional (per-mile) car maintenance costs that the utility would incur due to
the longer route.
The Board disallowed the proposed rerouting of the Fayette, Parish, and
Nelson traffic because TMPA’s evidence had failed to address and account
TEXAS MUNICIPAL POWER V. THE BNSF RY. CO. 817
7 S. T. B.
for the variety of off-SARR costs (both capital and operating costs) that could
be associated with rerouting this cross-over traffic. This deficiency was
highlighted by TMPA’s failure to address the concerns raised by BNSF
regarding the potentially serious operational problems associated with
rerouting this traffic through the Houston area. TMPA 2003, 6 S.T.B. at 596.
Moreover, TMPA had not demonstrated that the revenues from the rerouted
traffic would be sufficient to cover all costs for the entire movement over the
longer new route, including all off-SARR costs.
Although the Board did not permit this rerouting, it did not exclude the
traffic to these three utilities from the SAC analysis entirely, as BNSF had
proposed. Instead, the Board assumed that the GCRR would carry this traffic
up to the point at which the historical routing diverged from the BNSF lines
replicated by the GCRR (i.e., Northport), where it would be interchanged with
the residual BNSF. TMPA 2003, 6 S.T.B. at 597.
Both parties seek reconsideration of the Board’s treatment of rerouted
traffic. TMPA objects to the disallowance of the proposed rerouting of the
Fayette, Parish, and Nelson traffic, while BNSF objects to the acceptance of
the rerouting of the Big Brown traffic.
1. Overview
In evaluating TMPA’s arguments here on reconsideration, it is important
to consider them in light of the underlying principles of the SAC test, which is
to determine whether the carrier is being charged more for its service than a
hypothetical, efficient carrier would charge.
Here, TMPA has sought to reroute traffic onto its SARR that in actuality
travels out of the PRB south on another route on the BNSF system. And the
Board has allowed such so-called “rerouting” in circumstances where it is
demonstrated that such rerouting would not affect the service those shippers
expect or currently receive. However, in this instance TMPA has sought to
take that concept one step farther, and reroute traffic that traverses another
route onto its system for a long length, and then arbitrarily hand that traffic
back to the BNSF at a point of its choosing, without regard to whether such a
handoff is practical or feasible. In this instance, the handoff point happens to
be near Houston, TX, one of the nation’s most crowded rail terminals. TMPA
contends that BNSF could then carry unit trains of coal through Houston to
their destination at the power plants in question.
BNSF has contended that its existing lines through Houston do not have
the necessary infrastructure to handle this demanding type of traffic. For its
part, TMPA has not submitted evidence that either demonstrates that BNSF’s
Houston lines could handle such additional traffic, or how much it would cost
to upgrade those lines to handle this traffic. The Board has properly
disallowed these movements in this instance, because it found that TMPA had
neither determined nor included the additional infrastructure and operating
costs of moving this coal traffic through Houston.
Allowing re-routing of movements in the way proposed by TMPA would
distort the SAC test by permitting the complainant to artificially increase
SURFACE TRANSPORTATION BOARD REPORTS818
20
See TMPA Open. Narr. at 125-26.
21
BNSF Mot. to Dismiss at 18-19; see also BNSF Reply Narr. at I-13-17; BNSF Reply V.S.
Mueller at 30-32.
22
See Union Pac. Corp.--Control & Merger--Southern Pac. Rail Corp., 1 S.T.B. 233 (1996),
aff’d sub nom. Western Coal Traffic League v. STB, 169 F.3d 775 (D.C. Cir. 1999).
23
Joint Pet. for Service Order, 2 S.T.B. 725 (1997), modified and extended, 2 S.T.B. 744
(1997), further modified and extended, 3 S.T.B. 44 (1998), terminated, Joint Pet. for a Further
Service Order, 3 S.T.B. 612 (1998).
24
BNSF Mot. to Dismiss at 19.
7 S. T. B.
traffic on, and revenue allocated to, its SARR, all while passing on the
ramifications of such a rerouting (in this case, funneling 8,000- foot unit trains
of coal through the city of Houston) to BNSF. The Board properly disallowed
such a reroute unless the complainant shows that it has identified what these
additional infrastructure and operational costs would be and ensured that these
costs are fully accounted for.
2. Fayette, Parish, and Nelson
TMPA presents three arguments as to why the rerouting the Fayette,
Parish, and Nelson traffic should have been allowed. First, it maintains that
its rebuttal testimony incorporated all of the on-line and off-line costs that
BNSF had identified and that this should have been sufficient and accepted as
the best evidence of record. Second, TMPA argues that the Board should not
have applied new evidentiary requirements that had not been advocated by
either party without first giving TMPA an opportunity to address them.
Third, it argues that the Board should not have made findings about Houston
congestion without first affording parties an opportunity to address such
extra-record matters.
A cursory summary of the record in this case will put TMPA’s
evidentiary arguments into context. Notwithstanding the novelty of its
rerouting proposal, TMPA offered no evidence to support that proposal as
reasonable and consistent with the Guidelines; TMPA simply asserted that its
proposal should be accepted.
20
BNSF raised several objections to the proposal, including whether it was
proper to propose an operating plan that would impose additional costs on the
residual BNSF by changing the off-SARR routing of the cross-over traffic.
BNSF pointed out that TMPA’s proposal to route the three cross-over
movements through Iola, TX, would require BNSF to handle the traffic in a
new manner. To illustrate the problems such a rerouting might present, BNSF
noted that the ability to meet the cycle time commitments in the governing
contracts would be called into serious question, as the new routing of traffic
would be through Houston, a congested terminal area.
21
BNSF reminded the
Board of the traffic problems that have arisen after the “UP/SP” merger,
22
and
of the emergency service order issued to address those problems.
23
BNSF
asserted that, while those emergency conditions had abated, the effect of
rerouting coal trains through Houston would be “uncertain, at best.”
24
TEXAS MUNICIPAL POWER V. THE BNSF RY. CO. 819
25
TMPA Reply to Mot. to Dismiss at 5.
26
See TMPA Reb. Narr. at 44-45 (asserting that BNSF’s interchange preferences were
“irrelevant”).
27
See TMPA Reb. Narr. at 604.
7 S. T. B.
Despite TMPA’s statement that it would “respond with appropriate
rebuttal evidence further demonstrating the propriety of including these
movements in its stand-alone traffic group,”
25
TMPA’s rebuttal testimony
contained no such evidence. Instead, TMPA simply repeated its assertion
that, because a shipper may include cross-over traffic in its traffic group, it
can dictate where the cross-over interchange would occur—even if that would
impose off-SARR costs on the defendant carrier’s operations.
26
The only
concession TMPA made was to include some (but not all) of the off-SARR
road property investment that BNSF had claimed would need to be incurred
as a result of such a rerouting.
27
Given the record that was presented in this case, TMPA’s arguments for
reconsideration of this issue ring hollow. With respect to TMPA’s first
argument, it is true that TMPA’s rebuttal testimony accepted (for the most
part) the additional off-SARR investment costs that BNSF had quantified.
See TMPA 2003, 6 S.T.B. at 597. But BNSF had not purported to address all
of the costs associated with the off-SARR rerouting—it had not, for example,
sought to quantify the cost of rerouting these movements through the
congested Houston area—and BNSF had vigorously objected to the rerouting
even if the costs it did quantify were included in the SAC analysis. The first
issue with respect to rerouting was not whose evidence was the best evidence
of record, but whether TMPA’s rerouting assumptions were consistent with
the purpose and goal of the SAC test. Thus, TMPA’s concession that some
additional road property investment would be needed for the residual BNSF
to handle that traffic south of Iola, TX, was not by itself sufficient to justify
its rerouting proposal in light of the spectrum of concerns that BNSF had
raised.
As for TMPA’s second argument, the Board did not impose a
qualitatively different or undue burden on TMPA; the Board simply required
TMPA to support its case. See Guidelines, 1 I.C.C.2d at 544 (“A proponent
of a particular stand-alone model must identify, and be prepared to defend, the
assumptions and selections it has made.”). Here, TMPA had offered no
evidence of the off-SARR ramifications of its rerouting proposal.
Accordingly, it had failed to defend its traffic selection in the face of serious
objections raised by BNSF. As the Board stated in TMPA 2003, 6 S.T.B. at
597-98, TMPA cannot avoid the potential impact that might result from its
rerouting proposal by choosing to terminate the SARR before the point at
which those impacts would occur (e.g., Houston).
Turning to TMPA’s third argument, concerning the discussion of
potential congestion in Houston (which TMPA characterizes as extra-record),
BNSF had specifically raised the concern of rerouting heavy coal trains
through the congested Houston area as an illustrative example of the problems
SURFACE TRANSPORTATION BOARD REPORTS820
28
See BNSF Reply Narr. at I-14-17; BNSF Reply V.S. Mueller at 30-32; BNSF Mot. to
Dismiss at 4, 16-19.
7 S. T. B.
associated with the proposed rerouting.
28
It was, therefore, entirely
appropriate for the Board to address those operational considerations.
Moreover, an expert agency may draw on its general expertise and the Board
properly did so here with respect to rail service in Houston—an issue that had
been the subject of numerous Board decisions. See TMPA 2003, 6 S.T.B. at
596-97 & n.52.
TMPA’s final argument is that, even if it was proper to disallow the
particular rerouting that it had proposed, the Board should not have reverted
to using the historical routing of that traffic, under which the SARR would
need to interchange the traffic with the residual BNSF at Northport, NE.
Instead, TMPA now argues, the Board should have maximized the SARR’s
participation in the movement by assuming that this traffic would be rerouted
as far as Fort Worth, TX, where the GCRR could interchange the traffic with
BNSF, which in turn could transport the rerouted traffic to its respective
destinations without going through Houston. However, this was not how
TMPA had structured its SAC presentation and TMPA had not suggested any
alternative scenarios. Thus, no record had been presented as to whether such
an alternative rerouting would itself pose operational concerns or place
additional costs on the residual BNSF.
In sum, as the proponent of the novel rerouting proposal, it was TMPA’s
responsibility to support its proposal. It failed to do so here. Therefore the
Board properly disallowed the proposed rerouting of the Fayette, Parish, and
Nelson traffic through the Iola interchange.
3. Big Brown
BNSF argues that, because its rail transportation contract for the Big
Brown movement specifies the Front Range route, the Board was required to
give effect to that routing provision. BNSF relies on the Board’s statement in
West Texas, 1 S.T.B. at 658 n.41, that a SAC analysis assumes that a SARR
would replace the defendant railroad but does not assume that a governing
contract would be displaced or its terms changed.
The issue in West Texas was whether the SAC analysis could ignore the
contractual rights of a connecting railroad. In that case, the complainant had
proposed an operating plan that would have routed all of the traffic destined
for a particular utility plant via the SARR to Amarillo, TX, where it would
have been interchanged with a third-party carrier. This assumption
contravened a contract provision stating that 75% of the traffic would be
interchanged with a third-party carrier at Denver. The Board therefore
rejected the complainant’s effort to route all of the traffic through Amarillo.
The Board explained that the contract terms could not be ignored simply to
garner a greater proportion of the revenue from that traffic.
The ruling in West Texas reflects the broader principle that a complainant
may not assume an operating plan that would divert traffic away from other
railroads—see Arizona Elec. Power Coop., Inc. v. Burlington N. & S. F. Ry.
TEXAS MUNICIPAL POWER V. THE BNSF RY. CO. 821
7 S. T. B.
Co. et al., 6 S.T.B. 322, 327-28 (2002)—as the analysis of the reasonableness
of a defendant carrier’s rate should be based on the extent of the defendant
carrier’s participation in the movement. And the West Texas ruling reflects
the general SAC principle that the proponent of a SARR may not assume
a changed level of service to suit its proposed configuration unless it also
presents evidence showing that the affected shippers, connecting carriers, and
receivers would not object. See, e.g., McCarty Farms, Inc. v. Burlington N.,
Inc., 2 S.T.B. 460, 467 (1997).
Here, in contrast, it is BNSF itself that has the contract for the Big Brown
movement and handles that movement from origin to destination. With the
proposed rerouting, the SARR would merely step into the shoes of the BNSF.
The traffic would not be diverted from a third-party connecting carrier, as
there is no connecting carrier involved in that movement. Furthermore,
TMPA presented evidence showing why the affected shipper presumably
would not have any reason to object to the change in service. Thus, there is
no inconsistency with the objectives of the SAC test. See TMPA 2003,
6 S.T.B. at 591.
BNSF argues that the rerouting of the Big Brown traffic would not meet
the needs of the shipper because it would deny the shipper several enumerated
benefits of the routing provision that were placed in the contract at the
shipper’s request. BNSF seeks to support this claim with a newly presented
verified statement from its Vice President of Coal Marketing. But because
that evidence could have, and should have, been presented when the record
was developed on this issue, it is not timely now.
B. Revenue Forecasts
To forecast the revenues that would be generated after 2004 for traffic
not under contract, the Board used the most recent rate forecast from the
Department of Energy’s Energy Information Administration (EIA) as the best
evidence of record. This forecast—one of several forecasts advanced by
BNSF—was found to be preferable to the forecast proposed by TMPA for
two reasons. First, it was a recent forecast from an official and impartial
source. (The EIA, an independent statistical arm of the Department of
Energy, was created by Congress for the express purpose of providing policy-
neutral data and forecasts.) Second, use of EIA’s rate forecast was consistent
with the parties’ reliance on EIA’s tonnage forecast. TMPA 2003, 6 S.T.B. at
603.
TMPA argues that the Board should have accepted its rate forecasts,
which were based on specific contract information, rather than rely on a more
general, regional forecast. TMPA’s forecasts had differentiated between
traffic based on whether it would be captive or competitive traffic for the
GCRR. For captive traffic, TMPA’s forecasts escalated the rate at the end of
the contract period by the weighted average rate adjustment for captive
shippers remaining under contract. For competitive shippers, TMPA’s
forecasts estimated a new competitive market rate and, for subsequent years,
SURFACE TRANSPORTATION BOARD REPORTS822
7 S. T. B.
escalated that rate by the weighted average rate adjustment for competitive
shippers remaining under contract. Overall, TMPA forecast a 1.7% average
annual increase in rail coal transportation rates, in contrast with EIA’s
forecast of 1.4%.
Contrary to TMPA’s assertions, its forecast was not the best evidence of
record as to post-contract rate changes beyond 2004. As the Board explained,
TMPA 2003, 6 S.T.B. at 602, TMPA’s rate projections were more reflective
of past rate changes and were not the best evidence of what changes in rates
would reasonably be expected in the future. Moreover, the Board regards the
forecasts developed by EIA, a neutral governmental source, as more reliable
than forecasts developed by private parties for litigation, which are inherently
subject to manipulation. Indeed, TMPA itself acknowledged the
reasonableness of EIA forecasts when it used the EIA forecast for traffic
volumes to demonstrate the reasonableness of its own traffic volume
projection. Finally, having used the EIA forecast for tons, it was appropriate
to use EIA’s rate forecast as well. As explained in Duke Energy Corp v. CSX
Transp., Inc., 7 S.T.B. at 448-449, the forecasts of future transportation rates
cannot be divorced from the forecasts of future demand for coal transportation
(tonnages), as the two matters are interrelated. EIA’s coal demand forecasts
reflect EIA’s rate forecasts, and tonnage and rate forecasts should be
internally consistent where possible. Thus, where EIA tonnage forecasts are
used, it is preferable to use the matching EIA rate forecasts as well. This
provides a single, consistent, and independent source for both components of
the revenue forecast.
C. Locomotive Fueling
As previously discussed, the parties conducted a joint fuel study that was
used in the variable cost analysis to determine the fuel expenses associated
with handling the TMPA traffic alone. For the GCRR’s operating costs, the
Board again relied on this special study to estimate the fuel expense
associated with serving all of the traffic in the traffic group, an approach
advocated by BNSF. TMPA 2003, 6 S.T.B. at 663. TMPA seeks
reconsideration of this determination.
TMPA again asserts that adjustments to the results from the fuel study
are required to address differences between the trains studied and the traffic
group. In the variable cost analysis, it was the differences in cycle times that
TMPA sought to address. In the SAC analysis, it was differences in the route
that TMPA argued necessitated an adjustment, as the joint fuel study
examined movements that traveled over the Front Range route, while the
traffic group would travel over the Central Corridor route. TMPA contends
that coal shipments over the Central Corridor route should have a lower fuel
consumption rate because that route has less severe grades and curves than
the Front Range route.
TMPA’s proposed adjustment, however, bore no relationship to the
perceived problem, as TMPA sought to use the same adjustment it advocated
for the variable cost analysis. That adjustment addressed alleged discrepancies
TEXAS MUNICIPAL POWER V. THE BNSF RY. CO. 823
29
TMPA Open. Narr. at 96-98; TMPA Reply Narr. at 72; TMPA Reb. Narr. at 186-87.
30
TMPA Open. Narr. at 196; TMPA Reb. Narr. at 379-80.
7 S. T. B.
in cycle times and locomotive consists, not route characteristics.
29
Nowhere
in TMPA’s variable cost evidence on the proposed fuel study adjustments did
TMPA describe any adjustment based on differences in route characteristics.
And the Board could find no explanation in TMPA’s SAC evidence of any
proposed adjustment to fuel study results to account for differing route
characteristics. Rather, TMPA simply stated that it was submitting fuel costs
for the GCRR based on the consumption rate from the locomotive fuel study
results.
30
Accordingly, the fuel consumption estimates drawn from the joint
fuel study were the best evidence of record and were therefore properly used
in the SAC analysis.
D. General and Administrative Expenses
To determine the general and administrative (G&A) expenses that the
SARR would incur, the Board reviewed the record in detail to assess what a
least-cost, most efficient railroad would need to manage the company. That
inquiry ran the gambit, from the number of outside directors to the number of
purchasing agents. With a few exceptions, the Board accepted as feasible
TMPA’s evidence of the number of employees; the Board also accepted
TMPA’s proposed salary levels. Accordingly, the Board assumed that the
GCRR would be staffed by 63 G&A employees, at an annual expense of
$13.38 million. TMPA 2003, 6 S.T.B. at 675. BNSF seeks reconsideration,
arguing that the total G&A costs accepted in this case are exceedingly low
when compared with those of real-world railroads. But BNSF’s cost-
comparison data does not show what it would cost to run a specialized,
optimally efficient railroad. As explained in TMPA 2003, 6 S.T.B. at 659-60,
the structure of the GCRR would be substantially simpler than that of the
BNSF or any other large-scale, general commodity rail carrier. Under these
circumstances, the costs incurred by BNSF or other large carriers are not
necessarily a reliable indicator of the costs that would need to be incurred by
the GCRR.
BNSF argues that the Board’s analysis on this issue erroneously shifted
the burden of proof to BNSF to show that TMPA’s evidence was infeasible.
This is incorrect. The Board found that TMPA’s evidence was feasible and
supported. See TMPA 2003, 6 S.T.B. at 676-77 (finding that TMPA’s
operating department proposal would be reasonable and that BNSF’s
proposed additional layer of management would be unnecessary), 679
(agreeing with TMPA’s proposal to outsource marketing functions and
finding that TMPA’s customer service staffing were reasonable), 683 (finding
TMPA’s proposed law and administration department reasonable),
685 (finding TMPA’s arguments on legal staffing persuasive, its proposed
staffing level for safety and claims reasonable, and its human resources
staffing adequate), 686-87 (finding TMPA’s indexation of wages realistic and
verifiable), 687-88 (finding TMPA’s proposals on non-operating personnel
SURFACE TRANSPORTATION BOARD REPORTS824
31
See TMPA Reb. e-WP “Salaries” & BNSF Reply WP. 05194-051200.
32
See TMPA Open WP. at 5959.
7 S. T. B.
salaries supported). Thus, under SAC principles, it does not matter whether
BNSF’s higher-cost evidence may also have been feasible and supported, so
long as TMPA’s evidence—which represented the least cost for these
expenses—was feasible and supported. See Guidelines, 1 I.C.C.2d at 542.
Finally, BNSF asserts that TMPA’s executive salary assumptions were
unsupported. TMPA estimated the salary for the president, vice-president of
law and administration, and controller from public filings by Florida East
Coast Industries. The salary estimate for Vice President-Finance was drawn
from public filings by Wisconsin Central Transportation Corporation. TMPA
drew its remaining G&A salary estimates from BNSF’s 2000 Annual
Wage Forms A & B.
31
Upon reconsideration, BNSF is correct that TMPA’s workpapers do not
support TMPA’s salary estimates for the senior executives. For example, the
public filing by Florida East Coast Industries shows that in 2000, the
President of the Florida East Coast Railway (FEC) received total
compensation of $506,400, in the form of a salary of $300,000 and a bonus of
$206,400.
32
TMPA, however, used only the salary to develop its estimate.
Similarly, TMPA ignored non-salary compensation of $173,000, $86,000, and
$31,580 paid to the VP-Law and Administration, Controller, and VP-Finance
respectively. Accordingly, we will modify the SAC analysis with respect to
executive compensation.
BNSF maintains that the Board should use BNSF’s evidence of salaries
paid to executives of several regional and short-line railroads to calculate
salaries for the GCRR’s executives. However, with the exception of the VP-
Law position, including the non-salary compensation components paid by the
FEC produces lower total compensation figures than proposed by BNSF.
Therefore, our revised SAC analysis uses BNSF’s evidence only with respect
to the salary for VP-Law, where it is the least-cost evidence on record. For
the other executive positions, we will continue to rely on TMPA’s evidence,
but revised to include the total annual compensation paid to the FEC
executives (excluding stock options that we cannot value), and not just the
salary component. This will increase the total G&A operating expense
by $550,221 (from $13,380,139 to $13,930,360).
E. Road Property Investment
1. Barriers to Entry
The ultimate objective of the SAC constraint is to simulate a competitive
rate standard for non-competitive rail movements by determining the rate that
would be available to the shipper in a contestable market environment. As
explained in Guidelines, 1 I.C.C.2d at 529, the railroad industry is not
considered contestable due to its significant barriers to entry and exit. It is
only by excluding from the SAC analysis the costs and other limitations
TEXAS MUNICIPAL POWER V. THE BNSF RY. CO. 825
33
The statement in TMPA 2003 6 S.T.B. at 726 (2003), that TMPA only included 62 rail
bridges south of Bridger Junction was incorrect, as that was the number of bridges TMPA excluded.
7 S. T. B.
associated with entry and exit barriers that the Board can approximate the cost
structure of firms operating in a contestable market.
The SAC analysis treats as a barrier to entry any type of cost that a new
entrant would have to incur that was not actually incurred by the defendant
carrier. West Texas, 1 S.T.B. at 670 (1996). Thus, a defendant railroad is not
entitled to earn a return on investments it did not incur, but it can earn a
reasonable return on the current replacement costs of investments it made.
However, costs associated with modern construction practices are not treated
as barriers to entry, as such practices are merely a present-day substitute for
the procedures used when the original rail line was constructed. See FMC
Wyoming Corp. and FMC et al. v. Union Pac. R.R., 4 S.T.B. 699, 800-01
(2000).
It is often difficult to determine if a defendant railroad or its
predecessor(s) incurred a particular construction cost, given the passage of
time. We therefore rely on the Engineering Reports (“Engrg Rpts”), which
are a compendia of data collected by the ICC in the early part of the 20th
Century. They detail the material quantities required to build most rail lines
in place in the United States at the time. These reports included only those
items for which the ICC concluded a carrier had paid all or some part of the
cost. See Texas Midland R.R., 75 I.C.C. 1, 115-86 (1918). Therefore, if an
item appears in the Engrg Rpts, the Board includes in the SAC analysis some
cost to replicate that facility, albeit only a percentage of that cost if there is
reason to conclude that the defendant carrier (or its predecessor) did not bear
the full cost of that item.
Both parties object to how these barrier-to-entry principles were applied
in this case to four particular expenses: rail bridges, utility relocation, at-
grade crossings, and road resurfacing.
i. Bridges
South of Bridger Junction, TMPA had included the costs for only those
rail bridges needed to cross over natural barriers and other rail lines that
predated BNSF’s line. TMPA asserted that BNSF’s predecessors had not
incurred the cost to build the remaining 62 rail bridges now on the BNSF lines
that would be replicated by the GCRR.
33
BNSF responded by noting that the
excluded bridges all had construction dates within the last 20-30 years –
inferring that they had either paid to construct that bridge (after highways
became prevalent) or paid to replace bridges that had become obsolete. The
Board excluded any cost for constructing those rail bridges, explaining that, if
BNSF had replaced bridges as they became obsolete, those costs should be
considered a maintenance expense, rather than a capital investment expense.
BNSF seeks reconsideration of that decision. BNSF contends that, if it
had paid to replace those bridges, the SAC analysis should include the
replacement cost of those bridges. BNSF argues that, even if the SAC
SURFACE TRANSPORTATION BOARD REPORTS826
7 S. T. B.
analysis did not need to include the cost of initially constructing the bridges,
the SARR would need funds to replace the bridges as they age and that, by
excluding the capital for the construction, the Board wrongly omitted any
provision for the capital recovery needed for replacement of the bridges. This
is because the DCF model used by the Board automatically calculates the
amounts needed in future years to replace SARR assets, based on the amounts
included in the original construction costs, but does not calculate a
replacement cost for items excluded from the original investment base.
Finally, BNSF argues that the Board also erred by excluding all operating
costs associated with maintenance of the bridges.
Upon reconsideration, we agree that, where BNSF has paid to replace an
obsolete bridge, even if it did not pay for the original bridge, it should be
entitled to earn a return on that investment just as it would if it had paid for
the original bridge. Treating that capital outlay as a maintenance expense is
insufficient, as the result would be to include in the SAC analysis only enough
cost to cover future operating expenses to maintain that bridge. The SAC
analysis would not provide for a return on that investment.
Here, however, BNSF has failed to prove that it paid to construct or
replace the contested bridges. It simply asserted that it had, based on nothing
more than a bridge inventory list. If BNSF had paid to replace or construct
those bridges in the last 20-30 years, we would expect BNSF to have much
stronger evidence showing that financial outlay. Thus, on this record, it was
proper to exclude the cost to construct or replace the contested bridges. And,
as discussed in Section IV below, because the Board accepted BNSF’s
evidence regarding MOW expenses, which presumably included the annual
operating cost to maintain all of the bridges along the right-of-way (ROW),
no change to that expense is necessary.
ii. Utility Relocation
Where there was no evidence that BNSF or its predecessor paid to
relocate utility lines when it constructed its rail lines, the Board excluded that
cost as a barrier to entry. TMPA 2003, 6 S.T.B. at 705-06. BNSF seeks
reconsideration of that treatment. It argues that the costs of siting or
relocating utilities should be included regardless of whether the rail line
predated or postdated the utilities, as utilities are necessary to provide power
to a railroad and BNSF invested capital in the utilities’ poles and lines.
BNSF’s argument rests on the unsupported assertion that the GCRR
would benefit from the utility lines. But in this case, the inclusion of recent
technology (encoded track circuitry) would make pole lines unnecessary.
Therefore, it was appropriate to exclude those costs.
iii. Road Surfacing and At-Grade Crossings
In this case, the Board included the costs of at-grade crossings south of
Bridger Junction because the Engrg Rpts demonstrated that BNSF’s
predecessor had incurred crossing costs south of Bridger Junction. TMPA
2003, 6 S.T.B. at 741-42. But the Board excluded the cost to surface new
TEXAS MUNICIPAL POWER V. THE BNSF RY. CO. 827
7 S. T. B.
roads or resurface existing roads damaged during construction of that line,
because the lines predated the adjacent roads and there was no evidence that
BNSF had incurred the costs for surfacing (or resurfacing) such roads. Id. at
708.
Both parties seek reconsideration of the Board’s treatment of these two
costs. BNSF argues that, because the Engrg Rpts show that the railroad paid
for at-grade crossings, that means the road predated the railroad and that the
railroad therefore would have incurred costs for road resurfacing. TMPA
responds that BNSF’s position rests on the incorrect presumption that Engrg
Rpts provide evidence of who paid for an item, instead of just the existence of
an item. TMPA argues that Engrg Rpts, which were the basis for the Board’s
decision on this issue, catalog the existence of at-grade crossings, but do not
specify whether the incumbent, or some other party, paid for the crossings,
and that BNSF has provided no other evidence that it or its predecessor paid
for the crossings.
As explained above, we do not accept TMPA’s limiting characterization
of the Engrg Rpts. Rather, inclusion of an item in those reports generally
means that the railroad paid at least some portion of the expense for that item
and that expense is therefore properly included in the SAC analysis. Here,
because the Engrg Rpts included these assets, it is reasonable to assume
that BNSF paid at least part of the expense associated with at-grade crossings
south of Bridger Junction. The evidence in this record and past cases
suggests that when railroads did pay for at-grade crossings, they generally
paid 10% of those expenses. Thus, the Board properly included only 10% of
the replacement cost in the SAC analysis. TMPA 2003, 6 S.T.B. at 742.
There is no comparable evidence to indicate that BNSF paid any cost to
surface or resurface roads damaged by such construction, or to accommodate
traffic during construction. Absent such evidence, the Board’s longstanding
presumption is that other parties (such as the state) bore those expenses. Even
where the railroad paid some portion of the costs to construct at-grade
crossings and crossing protections, there is no basis for assuming, without
supporting evidence, that the railroad, rather than the state, paid for
resurfacing or relocation of public roads.
In sum, contrary to the suggestion of the parties, it is not inconsistent to
conclude, based on the evidence in this case, that BNSF or its predecessor
paid a small portion of the cost of at-grade crossings, but that the state paid
the entire cost to surface or resurface public roads damaged or diverted as a
result of that construction.
2. Fences
Relying on BNSF’s evidence that the Engrg Rpts showed fencing of 75%
of the ROW, and on BNSF’s citation to state laws and deeds from adjacent
landowners demonstrating the railroad’s obligation to fence the ROW, the
Board included fencing costs for 75% of the ROW south of Bridger Junction.
TMPA 2003, 6 S.T.B. at 740. TMPA seeks reconsideration of the Board’s
rejection of its evidence that only 20% of the ROW required fencing.
SURFACE TRANSPORTATION BOARD REPORTS828
7 S. T. B.
TMPA’s witnesses had spent 5 days traveling along the BNSF route that
would be replicated by the GCRR’s route from the PRB through Texas, and
from their visual observations they concluded that only 20% of the line was
fenced. TMPA also submitted photographs to show the lack of fencing.
TMPA notes that the Board relied on similar evidence in PPL Montana, LLC
v. Burlington N. & S. F. Ry. Co., 6 S.T.B. 286, 316-17.
Here, BNSF’s evidence was the better evidence of record. TMPA’s
evidence consisted of two pages of workpapers purporting to summarize its
witnesses’ inspection of unspecified portions of the ROW, along with
miscellaneous photographs (some of which note that there are no fences
shown) in an unlabeled section of its rebuttal workpapers. The photographs
appear to have been taken at four locations on the ROW (all on the same
subdivision within 80 miles of each other), but with no organized sampling
procedure for the remainder of the ROW. Given that both parties agree that at
least 25% of the ROW is unfenced, a handful of photographs showing
sections with no fencing was not sufficient support for TMPA’s claim that
fully 80% is unfenced. Thus, TMPA’s evidence was simply not on par with
the evidence in PPL, where the shipper supported its fencing contentions with
detailed, organized observations in 20 pages of workpapers.
IV. Technical Errors
Several technical errors in the numbers used by the Board in TMPA 2003
have been brought to our attention by the parties. First, the parties agree that
the wrong value was used for system-average locomotive depreciation. The
Board used a figure of $0.00000105, instead of the correct value of
$0.00006377. The parties should correct this mistake when they calculate the
variable costs in future years.
Second, as BNSF has noted, the Board mistakenly separated construction
costs for Rails and for Other Track Materials (OTM), when it should have
combined them for input into DCF accounts. This error resulted in the
application of the wrong asset-life assumptions to OTM costs, as well as the
costs for Ballast, Track Labor & Equipment, and Fences and Roadway Signs.
This error is corrected herein.
Finally, BNSF notes that the MOW figure reported in the decision and
used in the DCF analysis ($83.3 million) conflicted with the MOW figure
provided to the parties in the Board’s electronic workpapers ($87.8 million).
The Board inadvertently inserted the wrong table into the decision’s appendix
(one that did not reflect the final determination of the Board) and provided the
parties with some electronic workpapers showing the wrong numbers as well.
The DCF analysis then incorrectly used the $83.3 million figure to calculate
the maximum reasonable rate and reparations.
For the reasons discussed in the decision, see TMPA 2003, 6 S.T.B. at
696-697, the Board generally relied on the evidence submitted by BNSF to
develop MOW expenses. However, BNSF’s MOW cost estimates were
based, in part, on the number of track miles the GCRR would have, and the
Board had restated the total track miles from BNSF’s estimate of 2,546 down
to 2,401. The Board therefore accepted BNSF’s total MOW expense (of
TEXAS MUNICIPAL POWER V. THE BNSF RY. CO. 829
34
The prior decision erroneously stated that the complaint named only 4 mines. TMPA 2003,
6 S.T.B. at 581 n.3.
7 S. T. B.
$93,025,753), but reduced that expense by roughly 5.6% due to the smaller
SARR network. This error is corrected here. The correct figure of
$87,800,819 is shown in our new electronic workpapers at “STB Restated
MOW1” spreadsheet “STB Restatement.”
V. Scope of Rate Prescription
The challenged BNSF rate applied to coal movements to TMPA’s
Gibbons Creek plant from 16 named PRB mines: Antelope, Buckskin, Belle
Ayr, Black Thunder, Caballo, Caballo Rojo, Coal Creek, Cordero, Dry Fork,
Eagle Butte, Glovis Point, Jacobs Ranch, North Antelope, North Rochelle,
Rawhide, and Rochelle. TMPA’s complaint was not limited to specific
mines,
34
and its SAC evidence covered all 16 mines. Moreover, it is
undisputed that there are no other feasible means of transporting coal from the
PRB to the Gibbons Creek plant. Thus, the rate prescription in this case—the
higher of the SAC rate or the 180% R/VC regulatory floor—should apply to
movements from any of the 16 mines. However, the findings and ordering
paragraph in TMPA 2003, 6 S.T.B. at 608, 612-13 (2003) only mentioned the
two mines from which TMPA’s traffic had moved during the 9-month time
period covered by the evidentiary record (Caballo Rojo and Cordero).
In its petition for reconsideration, TMPA asks that the rate prescription
not be limited to the traffic from those two mines. In fact, it was the Board’s
intention that the rate prescription extend to any mines from which traffic
might move, so long as those mines were covered by both the complaint and
the SAC analysis. In TMPA 2003, 6 S.T.B. at 582-83, the Board expressly
stated that, if Gibbons Creek traffic were to move in the future from other
mines, the prescription would apply so long as the 180% R/VC jurisdictional
threshold was met, and that the parties should themselves make the
determination whether the threshold was met using the variable costing
procedures and findings contained in Appendix A of that decision. In other
words, the order was intended to be self-effectuating: only if the parties could
not agree on the computation of the variable cost of a movement would there
be any need to return to the Board.
BNSF argues that the Board has no authority to assess the reasonableness
of a challenged rate until we have determined that the revenues for the
movement exceed the R/VC threshold contained in the statute. That,
however, is not our reading of the relevant statutory provisions. Under the
statute, we must make an affirmative finding that the rail carrier has “market
dominance,” 49 U.S.C. 10707(b), which is defined as “an absence of effective
competition from other rail carriers or modes of transportation for the
transportation to which a rate applies,” 49 U.S.C. 10707(a). Once the Board
makes that qualitative finding, we may then proceed to examine the
reasonableness of the rate. 49 U.S.C. 10707(c). There is no requirement that
the Board make an independent quantitative determination of the R/VC ratio
SURFACE TRANSPORTATION BOARD REPORTS830
7 S. T. B.
for a movement; rather, we are precluded from finding market dominance
only if the defendant carrier proves that the rate charged results in an R/VC
below 180%. 49 U.S.C. 10707(d)(1)(A). Thus, absent a railroad showing
that the R/VC level for a movement is below 180%, the Board may proceed to
examine the reasonableness of a rate for a movement for which there is no
competitive alternative.
The railroad is not disadvantaged by the rate prescription extending to all
of the mines that were embraced in the complaint and the SAC analysis. The
railroad is protected by the terms of the prescription itself, which assures that
the prescribed rate cannot fall below the 180% R/VC floor for any movement
in any year. Indeed, by their nature rate prescriptions apply to future
movements, before the information necessary to calculate the variable costs of
those future movements is known. Thus, even as to the two mines from
which traffic had already moved, the parties were directed to make the
appropriate variable cost calculations for future years based on the guidance
provided in the Board’s TMPA 2003 decision regarding how those costs are to
be determined. If the parties cannot agree on the proper R/VC calculation
from another mine even with that guidance, either party may return to the
Board to have the dispute resolved.
BNSF argues that this approach represents a departure from precedent.
In Wisconsin Power & Light Co. v. Union Pac. R.R., 5 S.T.B. 955 (2001)
(WPL)—where, as here, the Board’s decision addressed only two of the
multiple mine origins covered by the complaint, as traffic had only moved
from two mines during the period covered by the evidentiary record—the
Board stated that, for shipments from the other mines, the parties should
return to the Board only if they could not agree on what the 180% R/VC rate
floor should be. See WPL at 960 (2001). In any event, to the extent that there
has been a change in Board policy, we are persuaded that the better policy is
for a rate prescription to be self-effectuating where a mine is embraced in
both the original complaint and the SAC evidence. There is no sound legal or
public policy reason why TMPA should be required to re-litigate its rate
complaint, in whole or in part, to obtain the benefit of the rate prescription
when it shifts traffic from one of the mines covered by its rate complaint to
another mine covered by that same complaint.
VI. Recalculated DCF
The DCF analysis compares the stream of revenues that would be
generated by the traffic group to the stream of costs that the GCRR would
incur, discounted to a common point in time. To do that, the DCF model
computes and distributes the total cost of the GCRR over the 20-year analysis
period, thus determining the amount of revenues that would be needed by the
GCRR to cover its operating expenses, meet its tax obligations, recover its
investment, and obtain an adequate return on that investment.
The results of our revised DCF calculations are shown in Table 1 below.
As that table shows, in the first 10 years of the 20-year SAC analysis period,
the traffic group would generate greater revenues than the GCRR’s revenue
requirements; but thereafter the annual revenues are forecast to fall below the
TEXAS MUNICIPAL POWER V. THE BNSF RY. CO. 831
7 S. T. B.
GCRR’s revenue requirements. The sum of the present values of over-
recoveries exceeds the under-recoveries, thus demonstrating that the existing
rate level is too high. The last column of Table 1 shows the percentage
amount by which the revenues from the traffic group would need to be
reduced so that over the entire 20-year SAC analysis period this traffic group
would generate just enough revenue to cover the GCRR’s revenue
requirements. The rate prescription for TMPA’s Gibbons Creek traffic
reflects that percentage reduction.
Table 1
Revised DCF Analysis
(millions of dollars)
Year
GCRR
Revenue
Requirements
BNSF
Forecast
Revenues Difference
Present
Value Cumulative
Percent
Rate
Reduction
2001 661.0 694.5 33.5 33.6 33.6 2.36%
2002 866.4 926.9 60.6 53.3 86.9 3.19%
2003 886.0 916.9 30.9 24.5 111.4 1.64%
2004 905.8 931.0 25.2 18.1 129.5 1.32%
2005 929.1 970.8 41.7 27.0 156.5 2.10%
2006 947.8 978.6 30.8 18.0 174.6 1.54%
2007 972.7 1,006.4 33.7 17.8 192.3 1.63%
2008 997.2 1,027.7 30.5 14.5 206.9 1.45%
2009 1,021.6 1,043.9 22.4 9.6 216.5 1.05%
2010 1,047.0 1,059.8 12.8 5.0 221.5 0.59%
2011 1,073.1 1,072.6 (0.5) (0.2) 221.3 0.00%
2012 1,099.8 1,084.9 (14.9) (4.7) 216.5 0.00%
2013 1,127.5 1,099.0 (28.5) (8.2) 208.4 0.00%
2014 1,155.5 1,110.2 (45.3) (11.7) 196.7 0.00%
2015 1,183.9 1,126.6 (57.3) (13.4) 183.3 0.00%
2016 1,215.8 1,149.7 (66.1) (13.9) 169.4 0.00%
2017 1,246.9 1,170.8 (76.1) (14.5) 155.0 0.00%
2018 1,278.0 1,192.0 (85.9) (14.7) 140.2 0.00%
2019 1,310.4 1,217.4 (93.1) (14.4) 125.8 0.00%
2020 1,344.3 1,246.8 (97.5) (13.6) 112.2 0.00%
2021 341.2 311.7 (29.5)
(4.0) 108.2 0.00%
* 2001 data is for the 2
nd
, 3
rd
, and 4
th
quarters of the year.
** 2021 data is for only the 1
st
quarter of the year.
NOTE: The DCF model limits the revenue reductions in
2001through 2010 to 49% of the overpayments, in order to offset the
underpayments that would occur in 2011 through 2021.
SURFACE TRANSPORTATION BOARD REPORTS832
7 S. T. B.
Under the revised SAC analysis, the prescribed rate is the higher of the
SAC rate, as shown in Table 2, or the regulatory rate floor (the 180% R/VC
rate level), which the parties should compute in a manner consistent with the
procedures and findings in TMPA 2003, 6 S.T.B. at 614-44 (2003). (As
discussed above, the Board cannot calculate the regulatory rate floor for
movements from 14 of the challenged mine origins, or for any of the mines
beyond 2001, as we do not have the necessary variable cost information.)
Table 2
Revised Rate Prescription
Year
Tariff
Rate
SAC Rate
Reduction
SAC
Rate
STB
Prescribed
Rate
2001 Q2 $19.09 2.54% $18.61 Higher of SAC rate
or
180% R/VC rate
2001 Q3 19.28 2.36% 18.83
2001 Q4 19.39 2.18% 18.97
2002 19.55 2.36% 19.09
Higher of
SAC rate
or
180% R/VC rate
2003 20.06 3.19% 19.42
2004 20.64 1.64% 20.30
2005 21.26 1.32% 20.98
2006 21.89 2.10% 21.43
2007 22.53 1.54% 22.18
2008 23.18 1.63% 22.80
2009 23.88 1.45% 23.53
2010 24.60 1.05% 24.34
2011 25.33 0.00% 25.33
2012 26.09 0.00% 26.09
2013 26.88 0.00% 26.88
2014 27.68 0.00% 27.68
2015 28.51 0.00% 28.51
2016 29.37 0.00% 29.37
2017 30.25 0.00% 30.25
2018 31.16 0.00% 31.16
2019 32.09 0.00% 32.09
2020 33.05 0.00% 33.05
2021 Q1 33.05 0.00% 33.05
Finally, based on the revised SAC analysis, we restate the reparations
awarded to TMPA for the unreasonable portion of the rate that it has paid
prior to this revised rate prescription taking effect. The amount of reparations
for movements in the 2nd, 3rd and 4th quarters of 2001 are shown in Table 3.
TEXAS MUNICIPAL POWER V. THE BNSF RY. CO. 833
7 S. T. B.
Table 3
Revised Reparations
Tons Tariff
Rate
Amount
Paid
Prescribed Rate
Reparations
2Q2001 454852 $19.09 $8,683,119 $18.61 $218,329
3Q2001 524538 19.28 7319128 18.83 $236,042
4Q2001 455980 19.39 8841450 18.97 $191,512
Totals 1,291,370 $24,843,697 $645,883
The parties should now re-calculate what reparations, if any, are due after
the 4th Quarter 2001 until this revised rate prescription takes effect. Interest
is also awarded in accordance with 49 CFR 1141.
This decision will not significantly affect either the quality of the human
environment or the conservation of energy resources.
It is ordered:
1. The rate prescription and reparations award for movements of the
issue traffic are revised as discussed above and set forth in Tables 2 and 3 of
this decision.
2. This decision is effective October 27, 2004.
By the Board, Chairman Nober, Vice Chairman Mulvey, and
Commissioner Buttrey.
