Five-Year Review of Oil Pipeline Pricing Index, 15329-15338 [06-2964]

Agencies

• DEPARTMENT OF ENERGY
• Federal Energy Regulatory Commission

[Federal Register Volume 71, Number 59 (Tuesday, March 28, 2006)]
[Rules and Regulations]
[Pages 15329-15338]
From the Federal Register Online via the Government Printing Office [www.gpo.gov]
[FR Doc No: 06-2964]


=======================================================================
-----------------------------------------------------------------------

DEPARTMENT OF ENERGY

Federal Energy Regulatory Commission

18 CFR Part 342

[Docket No. RM05-22-000]


Five-Year Review of Oil Pipeline Pricing Index

Issued March 21, 2006.
AGENCY: Federal Energy Regulatory Commission, DOE.

ACTION: Order establishing index for oil price change ceiling levels.

-----------------------------------------------------------------------

SUMMARY: The Federal Energy Regulatory Commission (Commission) is 
issuing this final order concluding its second five-year review of the 
oil pricing index, established in Order No. 561, Revisions to Oil 
Pipeline Regulations Pursuant to the Energy Policy Act of 1992, FERC 
Stats. & Regs. [Regs. Preambles, 1991-1996] ] 30,985 (1993). After 
consideration of all the initial, reply and supplemental comments, the 
Commission has concluded that the PPI+1.3 index should be established 
for the five-year period commencing July 1, 2006. At the end of this 
period, in July 2011, the Commission will once again review the index 
to determine whether it continues to measure adequately the cost 
changes in the oil pipeline industry.

DATES: March 28, 2006.

ADDRESSES: Secretary of the Commission, Federal Energy Regulatory 
Commission, 888 First Street, NE., Washington, DC 20426.

FOR FURTHER INFORMATION CONTACT: 

Harris S. Wood (Legal Information), Office of the General Counsel, 888 
First Street, NE., Washington, DC 20426, (202) 502-8224.
Robert W. Fulton (Technical Information), Office of Energy Markets and 
Reliability, 888 First Street, NE., Washington, DC 20426, (202) 502-
8003.

SUPPLEMENTARY INFORMATION: 

Before Commissioners: Joseph T. Kelliher, Chairman; Nora Mead 
Brownell, and Suedeen G. Kelly.

    1. On July 6, 2005, the Commission issued a Notice of Inquiry 
(NOI),\1\ in which it proposed to continue using the Producer Price 
Index for Finished Goods (PPI or PPI-FG) for the next five-year period 
beginning July 1, 2006, to track oil pipeline industry cost changes. 
The Commission applies the index to oil pipeline transportation tariffs 
to establish rate ceiling levels for pipeline rate changes. The NOI 
invited interested persons to submit comments on the continued use of 
PPI and to propose, justify, and fully support, as an alternative, 
adjustments to PPI. Comments and reply comments were due September 13 
and October 13, 2005, respectively.
---------------------------------------------------------------------------

    \1\ Five-Year Review of Oil Pipeline Pricing Index, IV FERC 
Stats. & Regs. [Notices] ] 35,552 (2005)
---------------------------------------------------------------------------

    2. Based on our review of the comments and reply comments received, 
and for the reasons discussed below, the Commission determines that the 
PPI plus one point three percent (PPI+1.3) should be established for 
the five-year period commencing July 1, 2006, and concludes that this 
index satisfies the mandates of the Energy Policy Act of 1992 (Energy 
Policy Act).\2\
---------------------------------------------------------------------------

    \2\ 42 U.S.C.A. 7172 note (West Supp. 1993). The Energy Policy 
Act's mandate of establishing a simplified and generally applicable 
method of regulating oil transportation rates specifically excluded 
the Trans-Alaska Pipeline System (TAPS), or any pipeline delivering 
oil, directly or indirectly, into it.
---------------------------------------------------------------------------

Background

    3. Congress, in the Energy Policy Act, required the Commission to 
establish a ``simplified and generally applicable'' ratemaking 
methodology for oil pipelines, consistent with the just and reasonable 
standard of the Interstate Commerce Act (ICA).\3\ On October 22, 1993, 
the Commission issued Order No. 561,\4\ promulgating regulations 
pertaining to the Commission's jurisdiction over oil pipelines under 
the ICA, and to fulfill the requirements of the Energy Policy Act. In 
so doing, the Commission found that using an indexing methodology to 
regulate oil pipeline rate changes, accompanied with certain 
alternative rate-changing methodologies where either the pipeline or 
the shipper could justify departure from the indexing methodology, 
would satisfy both the mandate of Congress and comply with the 
requirements of the ICA. The Commission found that the indexing 
methodology adopted in the final rule would simplify, and thereby 
expedite, the process of changing rates by allowing, as a general rule, 
such changes to be made in accordance with a generally applicable 
index, and that it would ensure compliance with the just and reasonable 
standard of the ICA by subjecting the chosen index to periodic 
monitoring and, if necessary,

[[Page 15330]]

adjustment. In determining which index to use, the Commission obtained 
the views of interested parties, including industry participants, 
shippers and others on its proposal to change its ratemaking 
methodology for oil pipelines. Dr. Alfred E. Kahn (Dr. Kahn) supported 
the establishment of an index of PPI-1 on behalf of a group of 
shippers, as the index that best tracked pipeline cost changes over a 
period of time. After extensive analysis of various suggested indices, 
the Commission adopted the PPI-1 index for the purpose of allowing oil 
pipelines to change rates without making cost-of-service filings. This 
index was chosen over others because it came the closest to tracking 
the historical changes in actual costs as reported in FERC Form No. 6 
and was to be in effect for the five-year period July 1996 through June 
2001. The Commission also committed to review every five years the 
continued effectiveness of its index.
---------------------------------------------------------------------------

    \3\ 49 U.S.C. app. 1 (1988).
    \4\ Revisions to Oil Pipeline Regulations Pursuant to the Energy 
Policy Act, FERC Stats. & Regs. [Regs. Preambles, 1991-1996] ] 
30,985 (1993), 58 FR 58753 (Nov. 4, 1993); order on reh'g, Order No. 
561-A, FERC Stats. & Regs. [Regs Preambles, 1991-1996] ] 31,000 
(1994), 59 FR 40243 (Aug. 8, 1994), aff'd., Association of Oil Pipe 
Lines v. FERC, 83 F.3d 1424 (D.C. Cir. 1996).
---------------------------------------------------------------------------

    4. In the course of establishing the appropriate index for the 
first review period 2001-2006, the Commission initially deviated from 
the methodology it had used in establishing the index as PPI-1 percent, 
concluding that the index should be retained as PPI-1, based upon a 
revision to the methodology established in Order No. 561. The 
Commission's order was reviewed by the U.S. Court of Appeals and 
remanded because the Commission departed from the Order No. 561 
methodology. Specifically, the U.S. Court of Appeals found that the 
Commission neither adequately addressed parties' concerns over using a 
new methodology, nor in the alternative articulated reasons for 
changing its averaging methodology applied in Order No. 561. Further, 
the U.S. Court of Appeals found that the Commission failed to justify 
its methodological shifts from Order No. 561 regarding outliers and the 
use of net plant. Upon remand, the Commission concluded that the most 
appropriate way to measure pipeline costs and rate ceilings, and to 
assure that the nexus drawn between them continued, was to apply the 
same methodology as it initially applied in Order No. 561. The 
Commission thus returned to the method adopted in Order No. 561 in its 
further analysis on remand. Utilizing the Kahn methodology which 
resulted in an index of an unadjusted PPI, the Commission adopted PPI 
as the appropriate index for the five-year period beginning July 2001. 
This order on remand was upheld by the U.S. Court of Appeals.\5\ In the 
current five-year review, we are applying that same methodology.
---------------------------------------------------------------------------

    \5\ 102 FERC ] 61,195 (2003), aff'd., Flying J Inc. v. FERC, 363 
F.3rd 495 (D.C. Cir. 2004).
---------------------------------------------------------------------------

Initial Comments and Initial Reply Comments

    5. On September 13, 2005, the Association of Oil Pipelines (AOPL) 
submitted its comments in response to the NOI. AOPL, as supported by a 
study done by its consultant, Dr. Ramsey Shehadeh (Dr. Shehadeh), 
contends that an index of PPI+1.3 percent rather than PPI is the 
appropriate index for the next five years. AOPL avers that application 
of the Commission's established, U.S. Court of Appeals-approved 
methodology shows that pipeline costs over the past five years 
increased at a rate of PPI+1.3 percent. AOPL maintains that the 
increased pipeline costs result from imposition of new safety and 
environmental regulatory obligations, voluntary security measures in 
the wake of 9/11 and increased energy costs. AOPL states that an index 
of PPI+1.3 percent will ensure that pipeline rates are ``just and 
reasonable'' while allowing efficient pipeline carriers to recover 
their increased costs over the next five years. These carriers, AOPL 
argues, will also be able to expand capacity to eliminate existing 
capacity constraints, and continue ongoing efforts to improve pipeline 
safety, efficiency, and security.
    6. Lion Oil Company, National Cooperative Refinery Association, 
Sinclair Oil Corporation and Tesoro Refining and Marketing Company 
(collectively the Refiners) filed a joint response on October 13, 2005, 
to AOPL's September 13 initial comments. The Refiners and the Air 
Transport Association of America (ATA), whose comments are discussed 
below, are both supported by the same study prepared by their 
consultant, Peter K. Ashton (Mr. Ashton), who urges the Commission to 
keep PPI as the index. The Refiners contend that the correct analysis 
of FERC Form No. 6 data indicates that the Commission should maintain 
PPI to determine annual rate increases. The Refiners state that PPI was 
determined appropriately by applying the methodology described by the 
Commission and the U.S. Court of Appeals for the District of Columbia 
Circuit over the past 12 years. The Refiners claim that AOPL failed to 
provide any sound support for its claim that PPI+1.3 percent is the 
inflationary index that tracked oil pipeline cost increases the best 
over the past five years. The Refiners contend that Mr. Ashton 
demonstrated a sound analysis of the data whose results indicate the 
Commission's initial view in its NOI was correct: The PPI without any 
adjustment is the index that has best tracked oil pipeline cost 
increases.
    7. On October 13, 2005, the ATA also filed a response to AOPL's 
comments. ATA contends that today's economic environment requires 
careful scrutiny of any proposed pipeline rate increases. ATA states 
the ability of pipelines to recover costs not generally encompassed by 
indexing through the cost-of-service ``safety valve'' or through the 
Commission sanctioned ``security surcharge'' ensures that pipelines can 
recover normal cost changes through indexing without, at the same time, 
having unjustified across-the-board burdens placed on the airline 
industry by implementation of an unreasonably high indexing adjustment. 
As mentioned above, ATA relied upon the same Ashton study as the 
Refiners. ATA states that its position, and that of its member 
airlines, is that the Commission should adopt a price index of PPI for 
the next five-year period.
    8. In his Sworn Declaration, Mr. Ashton claims that he employed the 
same methodology in conducting his analysis of oil pipeline cost 
increases in the 1999-2004 period as that used and adopted by the 
Commission in its previous review of the pricing index, as well as by 
Dr. Kahn. Based on his detailed analysis of historical oil pipeline 
cost data from 1999-2004, employing this methodology, Mr. Ashton 
concluded that the PPI, without any adjustment, closely tracked oil 
pipeline cost increases for that period. He states that taking the 
midpoint between the two composite averages (middle 50 percent and 
middle 80 percent of the sample) yields an annual rate of increase that 
is virtually identical to the increase in the PPI for the relevant time 
period. Mr. Ashton concludes that there is no basis for modifying the 
current PPI index since it already appropriately tracks normal industry 
average costs.
    9. In addition to conducting his own analysis, Mr. Ashton reviewed 
the submission of AOPL and its expert, Dr. Shehadeh. Mr. Ashton 
concluded that the data and analysis employed by Dr. Shehadeh are 
deficient, cannot be replicated, and therefore cannot be relied on. 
Specifically, Mr. Ashton cites the fact that much of the data 
pertaining to the later years of the study were compiled and supplied 
by AOPL, instead of Dr. Shehadeh obtaining his sample data from FERC 
Form No. 6. Mr. Ashton questions the lack of information concerning the 
source of Dr. Shehadeh's data, and the apparent lack of any attempt to 
validate or verify the information. Mr. Ashton states that, more 
significantly, in an attempt to

[[Page 15331]]

increase the sample size, Dr. Shehadeh made numerous additions and 
adjustments to the Form No. 6 data for ``potential omissions and 
potential errors,'' much of which was based on information supplied by 
AOPL. Mr. Ashton claims that Dr. Shehadeh made such adjustments without 
any clear indication as to exactly what those adjustments were, or how 
such adjustments affected the results of the study. As a result, Mr. 
Ashton concludes that any results of Dr. Shehadeh's study are deficient 
and unreliable.
    10. The U.S. Department of Transportation (DOT) also filed comments 
in response to AOPL on October 13, 2005. DOT expressed no views on the 
precise index the Commission should choose. DOT submitted its comments 
to confirm certain points raised by AOPL with respect to oil pipeline 
regulatory obligations. DOT states that it has adopted safety 
regulations that impose significant obligations and costs on pipeline 
operators.\6\ DOT states that it is concerned about the capacity of the 
underlying infrastructure of the nation's transportation networks, 
including oil pipelines, to meet growing demands placed upon them. DOT 
urges the Commission to consider seriously the financial commitment 
necessary for operators to maintain and expand pipeline system 
capacity.
---------------------------------------------------------------------------

    \6\ The DOT integrity management regulations are found at 65 
Federal Register 75378, December 1, 2000 and 67 Federal Register 
2136 (January 16, 2002).
---------------------------------------------------------------------------

    11. DOT states that in the long run, its rules will prove 
beneficial to the public and pipelines as well, but in the short run, 
the ensuing costs will prove considerable, with the additional effect 
of reducing or deferring operator revenues. DOT estimates that, over 
the seven year period 2001 through 2007, initial baseline assessments 
would cost operators more than $120 million; retesting, $14.5 million 
annually; preparation of integrity plans, almost $18 million; and 
related implementation costs, almost $10 million the first year and $5 
million annually thereafter. DOT states it could not estimate the 
repair costs incurred as a result of its required testing as it is 
impossible to predict the number and kind of conditions that would be 
disclosed, but given the fact that most repairs involve excavating 
pipeline segments and replacing sections of pipe, along with the 
requisite pressure reductions required to accommodate repairs, DOT 
believes such repair costs would be considerable. DOT noted that in 
2004 pipeline operators made more than 1,500 repairs posing immediate 
threats to pipeline integrity, and noted that one operator reported a 
single repair cost $8 million to make. DOT therefore contends that it 
is imperative that the Commission factor these costs into its 
deliberations in choosing the appropriate index for the next five-year 
period. DOT believes that failure to do so could lead to various 
outcomes inconsistent with the public interest, such as operators being 
disinclined to invest in additional capacity, abandoning older or 
marginally economic pipelines as a cost-cutting measure, or operators 
being tempted to cut corners on safety as a way of bringing costs more 
in line with revenues. DOT also cites evidence demonstrating a serious 
underinvestment in petroleum pipeline infrastructure and underscoring 
that several pipeline systems of national importance lack redundancy, 
with consequences including higher prices and less competitive markets 
for petroleum products in some regions, supply disruptions and price 
spikes due to relatively minor service interruptions, and diversion of 
petroleum products to other, less efficient and desirable 
transportation modes.
    12. DOT claims that the extent of capacity restrictions in the 
nation's pipeline infrastructure is becoming more apparent, as is the 
realization that the current regulatory mechanisms may not lead to 
appropriate reinvestment in the industry. DOT suggests that the 
Commission consider convening a workshop or technical conference to 
explore regulatory mechanisms that could facilitate critical investment 
in maintaining and expanding pipeline system capacity.
    13. On October 14, 2005, the Pipeline Safety Trust (Trust), an 
organization that promotes fuel transportation safety through education 
and advocacy, filed to respond to AOPL's comments. The Trust agrees 
with AOPL that safety requirements on the industry have significantly 
increased since the last five-year review, including but not limited to 
the new integrity management regulations. In addition, the Trust states 
that it is persuaded by the 1999-2004 data contained in AOPL's draft 
comments (which were analyzed by AOPL using a U.S. Court of Appeals-
approved methodology) that the costs on the industry have increased 
enough to justify a PPI+1.3 percent as the pricing index for the next 
five years. However, the Trust requests that prior to approving PPI+1.3 
percent, the FERC perform its own technical review of the accuracy and 
completeness of AOPL's cost data, and the reasonableness and 
appropriateness of AOPL's analytical methodology.

Exchange of Supporting Data Between Parties and Filed With the 
Commission

    14. To expedite the index review process, AOPL and the ATA and 
Refiners (hereinafter referred to as Shippers) agreed to exchange 
source data, spreadsheets, and the detail of the methodology used to 
support their respective positions of PPI+1.3 percent, and PPI. On 
November 15, 2005, AOPL and the Shippers filed their supporting 
workpapers with the Commission.

Subsequent Reply Comments and Responses

    15. On January 10, 2006, AOPL filed comments in reply to the study 
presented by the Shippers, contending that their study contains flawed 
economic analysis and incomplete and erroneous sampling of pipeline 
cost data. AOPL claims that, when corrected, the data presented by 
Shippers support an adjustment of PPI+1.56 percent, which supports 
AOPL's original position that the Commission should establish the index 
at least at PPI+1.3 percent.
    16. On January 23, 2006, the Shippers filed a joint response to 
AOPL's January 10 comments. Shippers claim that AOPL's comments distort 
the position advocated by Shippers and present non-public data upon 
which AOPL based its incorrect conclusions. Shippers contend that the 
facts they are presenting for Commission consideration are supported by 
a study conducted by Dr. Paul J. Smith (Dr. Smith), a prominent 
mathematician and statistician, as well as by a supplemental study 
performed by their consultant, Mr. Ashton. Based on new data provided 
by AOPL, Mr. Ashton added some pipelines to his study, and reconciled 
much of the data supplied by AOPL with that culled from FERC Form No. 6 
data. Mr. Ashton concludes that the Commission should be very cautious 
about establishing an index higher than the present PPI.
    17. Dr. Smith reviewed the dataset consisting of 62 firms that Mr. 
Ashton originally proposed, as well as the 81firm dataset proposed by 
Dr. Shehadeh in his January 10 rebuttal declaration on behalf of AOPL. 
Dr. Smith recommends the use of the median or geometric mean to 
estimate the five-year cost index, given the Form No. 6 data. In both 
data sets analyzed, the median and geometric mean are very close 
together. Dr. Smith argued that the use of the arithmetic mean is 
clearly not appropriate for either of these data sets. Arithmetic means 
are not representative of data from skewed distributions.

[[Page 15332]]

    18. Shippers conclude that, given the validation of Mr. Ashton's 
methodology, use of the geometric mean and choice of sample set, the 
conclusions reached by Dr. Smith, and given the real possibility of 
substantial errors in the FERC Form No. 6 data, the PPI should be used 
as the inflationary index for the next five years, or, in the 
alternative, the Commission should maintain the PPI and institute a 
rulemaking to establish new criteria and reconsider the methodology 
currently being used for determining such an index. Shippers contend 
that the fact that a significant number of large oil pipelines are 
substantially over-recovering their cost of service lends additional 
support to this conclusion.
    19. On February 9, 2006, AOPL submitted its supplemental reply 
comments in response to comments made by Shippers on January 23, 2006. 
AOPL claimed that Shippers, even after admitting to substantial 
mistakes in their analysis of oil pipeline cost data, resulting in 
flawed evidence and testimony, nevertheless urged the Commission to 
adopt a new methodology for setting the price cap index or, 
alternatively, to retain the PPI index pending a new rulemaking. AOPL 
argued that the Commission should squarely reject the Shippers' new 
position because, despite correcting errors in data and sample 
selection, Shippers' position remains fundamentally flawed. AOPL argued 
that Shippers' own cost evidence supports a substantial upward 
adjustment to the current index. AOPL states that, in sum, it is clear 
that Shippers' real complaint is not with the methodology and cost data 
used by the Commission to set its price cap index, but rather with the 
index level the faithful application of such methodology produces. The 
Commission must, as required by law, apply its established, U.S. Court 
of Appeals-approved index standard and set PPI+1.3 percent as its new 
index for the next five years.
    20. AOPL argued that, while Shippers purported to apply the 
Commission-approved methodology for measuring pipeline cost changes, 
Shippers in fact departed from that standard in several key respects. 
Even after correcting the data from their original analysis, Shippers' 
data sample omitted many eligible pipelines, failed to account for 
mergers, used incorrect data fields and data not reflected on FERC Form 
No. 6, improperly included cost data from TAPS assets that are not 
governed by the index and reflect entirely different accounting 
conventions, and most damaging, performed key calculations in the wrong 
order, thereby systematically understating cost changes.
    21. AOPL's comments addressed as well the report of Dr. Smith, 
cited by Shippers as a source of validation of its calculations. AOPL 
claimed that Dr. Smith's analysis is irrelevant to this proceeding, as 
he does not purport to address Mr. Ashton's analysis and expresses no 
opinion about the reasonableness of that analysis or its use of 
composite measures of central tendency. Dr. Smith advocates use of an 
entirely different standard, the median, does not approve of 
calculating cost changes in the wrong sequence, and in fact does not 
even analyze, much less endorse, use of the unweighted geometric mean 
in combination with the weighted mean and median that Mr. Ashton used. 
Nor, argues AOPL, did Dr. Smith analyze the middle 50 percent and 
middle 80 percent data sets. As a result of all this, any reliance 
Shippers placed on the report of Dr. Smith was misplaced.
    22. AOPL addressed Shippers' claims that the Commission must 
abandon its U.S. Court of Appeals-approved standard because of 
``manifest errors'' and because its results are sensitive to ``extreme 
data points'' by pointing out that the existence and reasonable 
treatment of outlier data was extensively addressed by the Commission 
in prior proceedings, and the Commission's methodology was specifically 
designed to take such an issue into consideration, specifically by 
employing the middle 50 percent and middle 80 percent samples. As to 
Shippers' claim that the Commission's methodology is flawed and it 
should set the index at PPI because a small minority of oil pipelines 
is over-recovering their cost of service, AOPL replied that the 
Commission recognized that, in adopting a uniform index for all 
pipelines, inevitably some pipelines would over-earn while others will 
under-earn. If Shippers truly believe that individual pipelines are 
over-earning such that rates cannot satisfy the ``just and reasonable'' 
requirement, they can file a complaint against those pipelines. AOPL 
contends that the indexing methodology is not intended to drive rates 
to cost, but instead to make sure that any rate changes were based on 
expected cost changes. AOPL further states that the Commission is not 
subject to a statutory duty to examine whole rates when pipelines 
propose index rates; rather, its inquiry is limited to a comparison of 
changes in rates and costs from one year to another.
    23. In his Supplemental Rebuttal Declaration on behalf of AOPL, Dr. 
Shehadeh argues that Mr. Ashton departed from the very methodology he 
purported to support, failed to implement accurately either the 
methodology used by Dr. Kahn or by Dr. Smith, has no support in any of 
the testimonies given in this proceeding for his findings, and for 
these reasons, such findings are unreliable as a basis for selection of 
an index for index-based regulation of oil pipeline tariffs. Dr. 
Shehadeh states that the basis for the differences between his 
conclusions and those of Mr. Ashton consist principally of errors in 
Mr. Ashton's data and his flawed order of operations in implementing 
the methods of Dr. Kahn. Specifically, Dr. Shehadeh cites the fact that 
Mr. Ashton calculates his cost changes in incorrect order--he applied 
the geometric mean over time prior to his application of the arithmetic 
mean across pipelines. Dr. Kahn correctly determined annual average 
change in costs by employing the geometric mean on the average 
cumulative changes, as opposed to Mr. Ashton, who in contrast 
determined the annual average change in costs by employing the average 
of the geometric means of each pipeline's cumulative changes.
    24. Rather than addressing the validity of Dr. Shehadeh's assertion 
concerning the order of his calculations, Dr. Shehadeh claims that 
instead, Mr. Ashton introduced an entirely new methodology based on 
measures of central tendency and composite averages, purportedly based 
on Dr. Smith's report. Dr. Shehadeh further states that Mr. Ashton's 
new methodology, especially as it pertains to use of sensitive data, is 
flawed and therefore unreliable.
    25. In conclusion, Dr. Shehadeh continues his support of the use of 
the methodology the Commission employed in its previous analysis, that 
Dr. Kahn validated, and that the U.S. Court of Appeals approved. 
According to Dr. Shehadah, Mr. Ashton's new methodology is completely 
unsupported by factual evidence, lacks economic foundation, and is 
unreliable and uninformative. Employing the same methodology used by 
Dr. Kahn and the Commission and endorsed by the U.S. Court of Appeals 
demonstrates that actual cost changes experienced by oil pipelines over 
the last five years almost equaled PPI+1.5 percent, and consequently, 
the Commission should choose as the index for the next five year period 
PPI with an adjustment factor no less than 1.3 percent.
    26. On February 21, 2006, the American Trucking Association filed 
letter comments in response to the Commission's NOI. The American

[[Page 15333]]

Trucking Association adopted the positions espoused by the Shippers and 
added no new arguments. On February 28, 2006, the International Air 
Transport Association filed letter comments, similar to that of the 
American Trucking Association, in support of the Shippers and again 
adding no new arguments.
    27. On February 24, 2006, Shippers filed additional comments, 
styled ``supplemental rebuttal,'' and a ``Sworn Rebuttal Declaration'' 
of Mr. Ashton. The purpose of this filing is to rebut the Supplemental 
Reply Comments of AOPL, which had been filed on February 9, 2006. 
Shippers contend that AOPL has made two fundamental and related errors 
in its Supplemental Reply Comments: AOPL incorrectly states that the 
Shippers and Mr. Ashton have employed a new methodology; and, even if 
Shippers have employed a different methodology, the Commission is 
within its rights to rely on that methodology.

Discussion

Methodology To Calculate the Index Differential

    28. Since Order Nos. 561 and 561-A, the Commission has primarily 
relied upon Dr. Kahn's testimony \7\ to develop the methodology to set 
the index differential \8\, which was subsequently approved by the U.S. 
Court of Appeals.\9\ Within the Commission-established method, after 
each firm's unit cost changes are calculated and weighted, two trimmed 
data sets are extracted from the master data set. Both parties have 
constructed the trimmed data sets of the middle 50 percent and middle 
80 percent. Trimming is done to remove statistical outliers, or 
spurious data points that could bias the mean of the sample in either 
direction. Table 1 provides a description of the statistical values of 
central tendency used by both parties to develop the index. The 
industry-wide cost index is calculated by averaging both composites on 
Line D and then comparing that value to the PPI-FG index data over the 
same period.
---------------------------------------------------------------------------

    \7\ Declaration of Alfred E. Kahn, August 31, 2000, in Review of 
Pipeline Pricing Index, Docket No. RM00-11-000.
    \8\ To calculate the index differential, the cost index is 
compared to the PPI-FG average index for the same time period. The 
remainder of this calculation [Cost Index-PPI-FG] is the index 
differential.
    \9\ Association of Oil Pipe Lines v. FERC, 83 F.3d at 1437.

                                 Table 1
------------------------------------------------------------------------
       Line             Middle 80  percent         Middle 50  percent
------------------------------------------------------------------------
A.................  Median...................  Median.
B.................  Weighted Mean............  Weighted Mean.
C.................  Un-weighted Mean.........  Un-weighted Mean.
D.................  Composite of 80% =         Composite of 50% =
                     (A+B+C)/3.                 (A+B+C)/3.
------------------------------------------------------------------------

    29. Both AOPL and Shippers used the same sample (with the exception 
of SFPP, L.P.\10\) to describe the central tendency of the data, in 
which the cost index calculation directly follows. However, the parties 
have differed in the way in which they calculated the pipelines' cost 
increases. The result has been that both parties calculate a different 
pipeline industry cost index; AOPL arriving at an index of PPI+1.49 
percent and the Shippers arriving at an index of PPI+0.675 percent.
---------------------------------------------------------------------------

    \10\ SFPP, L.P. was excluded by the Shippers because its cost as 
reflected in its Form No. 6 were being challenged in a current rate 
proceeding. The Commission conducted a review of the pipeline 
samples submitted by both Dr. Shehadeh and Mr. Ashton and determined 
that, using the Court of Appeals-approved methodology, the exclusion 
of SFPP, L.P. causes only a 0.02 percent decrease in the average 
annual cost difference. Thus, the exclusion of SFPP, L.P. still 
supports the use of PPI+1.3 percent.
---------------------------------------------------------------------------

    30. In simple terms, AOPL and the Shippers state that they apply 
the same methodology but they arrive at different results. Each party 
calculates total industry costs \11\ for each pipeline in the time 
period between 1999 and 2004, and then estimates the central tendency 
of the results, sums these amounts, and divides the result by the 
number of pipelines to arrive at the industry average cumulative change 
in industry costs (known as the arithmetic mean of the sample). AOPL 
then derives an annual percent change in industry costs for the 1999 
through 2004 period by employing the geometric mean on this industry 
average cumulative change in costs. AOPL's methodology tracks the 
methodology previously used by the Commission and approved by the U.S. 
Court of Appeals. The Shippers, however, depart from the prior approved 
methodology, in that the Shippers derive the cumulative change in costs 
(between 1999 and 2004) for each pipeline, by calculating each year's 
cost change for each pipeline. The year to year cost changes are 
multiplied together to arrive at the cumulative cost change for that 
pipeline. The average cost change is determined by taking the geometric 
mean of that cumulative cost change.
---------------------------------------------------------------------------

    \11\ Average total cost for an individual pipeline is the 
average of the change in operating cost (weighted by the operating 
ration) added to the average of the change in net plant (weighted by 
the residual, one minus the operating ratio). References to 
individual pipeline costs beyond this point are assumed to be 
average total costs.
---------------------------------------------------------------------------

    31. We base our analysis of the calculations in this proceeding 
upon the U.S. Court of Appeals-approved model, and have found that the 
methodology used by Dr. Shehadeh for AOPL in this proceeding conforms 
to Dr. Kahn's U.S. Court of Appeals-approved methodology. Our analysis 
shows that the Shippers' methodology, as represented by Mr. Ashton, is 
fundamentally flawed.
    32. In delineating the index differential, Mr. Ashton, in his first 
declaration,\12\ claims to have accurately applied Dr. Kahn's 
methodology in calculating average annual cost changes, but our review 
found that he deviates from Dr. Kahn's methodology in certain respects. 
In this first attempt, Mr. Ashton determines the average annual change 
in unit costs for years 1999 through 2004 by calculating the arithmetic 
average of the geometric mean of each pipeline's cumulative unit cost 
change, as opposed to Dr. Kahn's method of calculating the geometric 
mean of the arithmetic average of cumulative unit cost change (Ashton's 
Decl. at p. 14). On the basis of determining the average cost change of 
each pipeline, the use of Dr. Kahn's methodology would calculate the 
cost increase between end years 1999 and 2004 by this formula: (final 
cost - initial cost)/(initial cost)-1. Mr. Ashton erred in this step of 
the calculations by taking the geometric mean of the product of the 
individual company's yearly cost increase. Furthermore, Dr. Shehadeh 
has shown that Mr. Ashton's method results in the underestimation of 
costs.\13\
---------------------------------------------------------------------------

    \12\ Sworn Declaration of Peter K. Ashton, October 13, 2005 (p. 
3).
    \13\ Rebuttal Declaration of Ramsey D. Shehadeh, PhD at 10-11, 
January 10, 2006 by use of a theorem known as Jensen's Inequality.
---------------------------------------------------------------------------

    33. Mr. Ashton responds \14\ to Dr. Shehadeh's rebuttal, and claims 
that the newly added testimony of Dr. Smith supports Mr. Ashton's new 
methodology.\15\ However, in examining Dr. Smith's analysis, it seems 
that he has followed Dr. Kahn's approach (but not Mr. Ashton's) by 
calculating percentage cost changes for individual companies, where Dr. 
Smith states that ``the five-year percent differences in costs as 
reported'' for individual companies ``were computed as (final

[[Page 15334]]

cost - initial cost)/(initial cost)-1.'' Therefore, we cannot reconcile 
Dr. Smith's evidence with Mr. Ashton's statement that ``Prof. Smith 
clearly points out that given the underlying characteristics of the 
data and its skewed distribution, the methodology that I employ relying 
on the geometric mean is the proper methodology for computing the cost 
increases of individual pipeline companies'' (Supplemental Decl. at p. 
2) Dr. Smith's testimony regarding his recommendation of the use of the 
geometric mean, was to describe with relative accuracy the central 
tendency of the data, not the calculation of the individual cost 
increases themselves. Further, Dr. Smith's testimony in regard to this 
proceeding is incomplete because he only trimmed the cost data by 5 
percent, and he never analyzed the ``middle 50 percent'' and ``middle 
80 percent'' data sets, which excluded ``outliers,'' adopted by the 
Commission and approved by the U.S. Court of Appeals. As the Court of 
Appeals stated, ``[t]he object of excluding outliers is to prevent 
extreme and spurious data from biasing an analysis, i.e., affecting its 
result adversely.'' \16\
---------------------------------------------------------------------------

    \14\ Supplemental Declaration of Peter K. Ashton, January 23, 
2006.
    \15\ Analysis of Pipeline Index Data, submitted by Paul J. 
Smith, January 23, 2006.
    \16\ Association of Oil Pipe Lines v. FERC, 281 F.3d 239, 246 
(D.C. Cir. 2002).
---------------------------------------------------------------------------

    34. Also, in the same response, Mr. Ashton presents an ``update'' 
to his methodology, and supports it again with the analysis of Dr. 
Smith (Supplemental Decl. at p. 7). Originally, Mr. Ashton measured the 
central tendency of both of the trimmed data sets (80 percent and 50 
percent) with the median, the weighted and the un-weighted arithmetic 
mean (although he still wrongly calculates the cost changes for 
individual companies by the geometric mean). In the update, Mr. Ashton 
delineates the weighted and un-weighted means by now taking the 
geometric mean \17\
[GRAPHIC] [TIFF OMITTED] TR28MR06.000

of the unit cost change. Mr. Ashton states that this update is 
justified ``as Dr. Smith points out, in measuring the central tendency 
it is also appropriate to take the geometric mean and the median--not 
the arithmetic mean.'' The supporting evidence by Dr. Smith points out 
that both data sets (untrimmed) are not normally distributed. Dr. Smith 
states that, ``the data are more accurately described by a skewed 
lognormal distribution than by a bell-shaped normal distribution, but 
that neither distribution accurately described the data.'' Dr. Smith 
then measured the arithmetic average applied to a 5 percent trimmed 
sample in which he concluded in his analysis that, based on the 
results, ``the trimmed mean is substantially less than the arithmetic 
mean, illustrating how a few extremely large indices affect the overall 
estimate'' (p.5).
---------------------------------------------------------------------------

    \17\ This geometric mean is a statistical treatment to the data 
to find central tendency, as opposed to the geometric mean used 
previously to calculate the index of costs over time. In this 
application, the general formula (see formula above) was used, where 
denotes the cost index of firm i.
---------------------------------------------------------------------------

    35. Shippers, in their Supplemental Rebuttal filing of February 24, 
2006, contend that the use of the geometric mean was but a small change 
and an improvement to the Commission's methodology to better suit the 
underlying data.\18\ The Shippers' underlying belief in this order 
seems to be that the data in the samples that are used in the 
Commission's methodology are positively skewed, and therefore, Dr. 
Smith never had to prove that our samples were skewed. However, Dr. 
Smith never applied his alternative approach to our samples to 
determine, based on his analysis, what the best measure of central 
tendency would be. He contended that he proved that the geometric mean 
approach would be more accurate on his 5 percent trimmed sample because 
his result more closely matched the median, and therefore was ``more 
robust.'' Dr. Smith, though, never proved that the geometric mean was 
``more robust'' on the 50 and 80 percent samples.
---------------------------------------------------------------------------

    \18\ Supplemental Rebuttal Comments at 2; Ashton Supplemental 
Rebuttal Decl. at 3.
---------------------------------------------------------------------------

    36. Based on the calculations presented by the Shippers through Mr. 
Ashton's declarations, it is clear that the methodology the Shippers 
use departs from the methodology presented by Dr. Kahn. Shippers have 
not proven that their methodology is superior to that of Dr. Kahn.

Reconciling the Dataset of AOPL and Mr. Ashton

    37. In Shippers' January 23, 2006 joint response to comments made 
on January 10, 2006 by AOPL, Mr. Ashton added some pipelines to his 
study, and reconciled much of the data supplied by AOPL and culled from 
FERC Form No. 6 data. Aside from a few remaining pipelines in which 
discrepancies appear between the data that Mr. Ashton used and the data 
that Dr. Shehadeh used, Mr. Ashton is prepared to accept the 
reconciliation and changes offered by Dr. Shehadeh in his Rebuttal 
Declaration. Mr. Ashton's database is comprised of 79 pipelines that 
account for 92 percent of all barrel-miles transported in 1999.
    38. To investigate the data discrepancies, the Commission has 
examined the hard copy FERC Form No. 6 data filed by individual 
pipelines to determine whether complete data for these pipelines are 
available and whether they match the data used by Mr. Ashton or Dr. 
Shehadeh. We have compared, on a pipeline-by-pipeline basis, every 
relevant data value in Mr. Ashton's sample with the corresponding 
values in FERC Form No. 6. We then have applied to the reconciled data, 
which account for the only remaining discrepancies identified by Mr. 
Ashton, the methodology described by Dr. Kahn, adopted by the 
Commission and approved by the U.S. Court of Appeals. We now discuss 
those pipeline-by-pipeline comparisons.
    39. Navajo Pipeline Co. L.P.--Holly Energy Partners--Operating 
L.P.--Dr. Shehadeh correctly points out in Exhibit A15 of his original 
declaration and Exhibit 16 of his rebuttal declaration that Navajo 
Pipeline Co. L.P. (Navajo) was renamed Holly Energy Partners--Operating 
L.P. (Holly) in 2004. As a result, both companies filed a FERC Form No. 
6 in 2004. Mr. Ashton is correct in noting that the 2004 data reported 
by Dr. Shehadeh comes from only one company, Navajo, and is only 
partial year data. The Commission agrees that the 2004 data for Navajo 
and Holly can be aggregated to provide data for the complete year. 
However, the Commission takes issue with the values Mr. Ashton reports 
for carrier property and total barrel-miles. A review of Navajo's 2004 
FERC Form No. 6 reveals carrier property totaling $10,186,371 and 
barrel-miles totaling 4,095,048,097. Holly's 2004 FERC Form No. 6 
reports carrier property totaling $22,788,803 and total barrel-miles of 
3,330,670,969. Thus, the Commission will use $32,975,174 for carrier 
property and 7,425,719,066 for total barrel-miles.
    40. Olympic Pipe Line Company --On March 31, 2003, Olympic Pipe 
Line Company (Olympic) resubmitted its 2001 FERC Form No. 6 to report 
changes to carrier property, accrued depreciation, and operating 
revenue. Mr. Ashton is correct to use the data contained in the 
resubmitted FERC Form No. 6. However, Mr. Ashton fails to reflect the 
operating expenses provided in the updated FERC Form No. 6 and 
continues to use the figure reported in Olympic's original 2001 FERC 
Form No. 6. Thus, the Commission will use the updated $59,520,702 for 
Olympic's 2001 operating expenses.

[[Page 15335]]

    41. Premcor Port Arthur Pipeline Company--Mr. Ashton states that he 
did not include Premcor Port Arthur Pipeline Company (Premcor) in the 
dataset he used to calculate oil pipeline costs changes because Premcor 
lacked complete FERC Form No. 6 data. Dr. Shehadeh agrees with Mr. 
Ashton on this point. No data is reported for accrued depreciation for 
the years 2001-2004 even though Premcor did report accrued depreciation 
in 1999-2000. Without complete data for all six years, a company cannot 
be included in the dataset. Therefore, the Commission agrees with Mr. 
Ashton's recommendation that Premcor be excluded from the analysis.
    42. Cypress Pipe Line Company, LLC--Mr. Ashton criticizes Dr. 
Shehadeh's use of barrel-mile data for Cypress Pipe Line Company, LLC 
(Cypress) in 1999 from page 700 of the FERC Form No. 6, rather than 
data reported on page 600. Mr. Ashton states that the agreed-upon 
source for barrel-mile data is page 600 of the FERC Form No. 6. Despite 
this criticism, Mr. Ashton himself elects to use data from page 700, 
not data from page 600 as he describes. Cypress, however, errs by 
reporting the number of barrels received into the system, rather than 
total barrel-miles. This is evident from comparing line 32 of page 600, 
grand total of barrels received into system, with page 600 line 33a; 
the numbers are identical. By contrast, line 4 of page 700 reports 
total throughput in barrel-miles as 78,558,341.83. Thus, the Commission 
will use 78,558,342 for Cypress's 1999 throughput in barrel-miles.
    43. Mr. Ashton raises another substantive issue with respect to the 
reconciled dataset contained in Dr. Shehadeh's rebuttal declaration. 
Mr. Ashton notes that there are differences between the operating 
revenues for five pipelines reported on the FERC Form No. 6, page 114, 
and FERC Form No. 6, page 301. Specifically, these discrepancies occur 
with respect to Mobil Pipe Line Company, Mustang Pipe Line Partners, 
Osage Pipe Line Company, LLC, San Pedro Bay Pipeline Company, and 
SouthTex 66 Pipeline Company, Ltd.
    44. In order to rectify these differences, the Commission adjusted 
its cost calculation to use the figures endorsed by Mr. Ashton. 
However, the use of Mr. Ashton's figures proved immaterial as the 
result still supports the use of PPI+1.3 percent as the new oil index. 
When the Commission adopted the page 301 data for those pipelines for 
which Mr. Ashton noted discrepancies, and applied the U.S. Court of 
Appeals-approved methodology, the results changed by less than 0.01 
percent.

Indexing Methodology

    45. In the January 23, 2006, response of Shippers to AOPL's January 
10 comments, Shippers assert for the first time that, as an alternative 
to using the current methodology for determining adjustments to PPI 
after conducting a five-year review, the Commission should continue to 
use the current unadjusted PPI for the time being and institute a 
rulemaking to establish a new methodology for determining what the oil 
pipeline rate change index should be over the next five years. Shippers 
contend this is appropriate because there are serious defects in the 
Commission's current index review methodology. AOPL responded to this 
assertion on February 9, 2006.
    46. Notwithstanding Shippers' assertions to the contrary contained 
in their Supplemental Rebuttal Comments of February 24, 2006, Shippers' 
suggestion that the Commission should embark upon a new rulemaking 
proceeding to establish a new method for calculating pipeline cost 
changes to compare to changes in PPI is beyond the scope of our five-
year review as set forth in the NOI that instituted this proceeding. In 
the NOI, the Commission asked for comments on whether and to what 
extent the PPI should be adjusted to better reflect those cost changes, 
not whether the method for determining pipeline costs should be 
changed. The NOI specifically stated:

    The Commission proposes to continue to utilize PPI for the next 
five-year period as the index to track changes to the costs of the 
oil pipeline industry and to apply to rate ceiling levels for oil 
pipeline rate changes. The Commission invites interested persons to 
submit comments on the continued use of PPI and to propose, justify, 
and fully support, as an alternative, adjustments to PPI. (NOI, ] 4)

    The parties have filed numerous comments reflecting their positions 
on what adjustments should or should not be made to the PPI upon 
review, and only as a last-minute item has anyone suggested that the 
Commission embark on a course of discarding the Commission's current 
five-year review methodology of determining pipeline cost changes to 
compare to changes in the PPI-FG over the five-year review period. The 
information provided by Shippers is insufficient to persuade us that 
our method should be discarded. Beyond these issues, no one has 
suggested that the Commission look to change to an index other than 
PPI-FG as representative of oil pipeline industry-wide costs.
    47. Shippers first contend the use of FERC Form No. 6 data make 
application of the cost standard inaccurate. They claim that the data 
contained in the FERC Form No. 6 is sporadic, incomplete, and contain 
substantial errors. Shippers point out that out of 186 FERC regulated 
pipelines, only 79 pipelines have provided sufficient Form No. 6 data 
to warrant being included in the database for analysis. They believe 
the 42 percent sample is too small to justify the continued use of Form 
No. 6 data.
    48. The FERC Form No. 6 data is the only systematic source of 
information regarding the past costs and revenues of oil pipelines. As 
previously mentioned, Mr. Ashton concedes his sample contains 79 
pipelines that account for over 92 percent of the 1999 total barrel-
miles. In defending Order No. 561-A on appeal to the D.C. Circuit on 
this very issue, the Commission stated that ``[t]here is * * * no 
reason to believe that samples representing between 10% and 33% of the 
industry, taken from the median range of the industry cost data, were 
too small to produce reliable results.'' \19\ In addition, the Shippers 
argued before the U.S. Court of Appeals that the, ``[d]ata submitted to 
FERC have become increasingly accurate, thus eliminating the need for a 
proxy.'' \20\ Further, the U.S. Court of Appeals agreed that it is 
evidently uncontested that the reported data have become more accurate.
---------------------------------------------------------------------------

    \19\ See, Brief for Respondent Federal Energy Regulatory 
Commission, Association of Oil Pipelines v. Federal Energy 
Regulatory Commission, No. 94-1538, at 36 (July 24, 1995).
    \20\ See, Flying J Inc. v. FERC, 363 F.3d 495 (D.C. Cir. 2004).
---------------------------------------------------------------------------

    49. Second, Shippers express concern that a significant number of 
oil pipelines are not complying with FERC Form No. 6 filing 
requirements. Specifically, Shippers were concerned that the Commission 
has not consistently enforced these filing requirements nor has it 
examined the Form No. 6 data and required corrections of the errors 
noted by Mr. Ashton in his supplemental declaration.
    50. The Commission disagrees with the Shippers' assertion that the 
Commission has not consistently enforced the accurate and timely filing 
of FERC Form No. 6 data. In 1994, the Commission addressed additional 
revisions to the Form No. 6 in Order Nos. 571 and 571-A,\21\ including 
adding

[[Page 15336]]

a new page 700. The information included in the Form No. 6 was 
determined at the time to be the minimum necessary for Shippers to 
assess filed rate changes under Order No. 561.
---------------------------------------------------------------------------

    \21\ Order No. 260, 47 FR 42327 (Sept. 27 1982); FERC Stats. & 
Regs. [Regulations Preambles January 1991-June 1996] ] 31,006 (Oct. 
28 1994). Order No. 571-A, 60 FR 356 (Jan. 4, 1995); FERC Stats. & 
Regs. [Regulations Preambles January 1991-June 1996] ] 31,012 (Dec. 
28, 1994).
---------------------------------------------------------------------------

    51. Prior to 2000, FERC Form No. 6 required that a pipeline include 
its annual cost of service, operating revenues, throughput in barrels, 
and throughput in barrel-miles. The Commission found that the Form No. 
6 data was inadequate to monitor the reasonableness of a pipeline's 
filed rates. Thus, the Commission proposed the addition of the 
following reporting requirements: operating and maintenance expenses, 
depreciation expense, AFUDC depreciation, amortization of deferred 
earnings, rate base, rate of return, return on rate base, and income 
tax allowance. Since the Form No. 6 is intended to be both a financial 
and ratemaking document,\22\ these additional requirements ensured that 
the Commission had the financial, operational, and ratemaking 
information needed to carry out its regulatory responsibilities to 
monitor the oil pipeline industry in a dynamically changing 
environment.
---------------------------------------------------------------------------

    \22\ Cost of Service Reporting and Filing Requirements for Oil 
Pipelines, FERC Stats. & Regs. [Regulations Preambles 1991-1996] ] 
31,006 at 31,169 and FERC Form No. 6, p.i, I.
---------------------------------------------------------------------------

    52. In Order No. 620, the Commission required pipelines to maintain 
workpapers that fully support the data reported on page 700 including 
but not limited to the total cost-of-service calculations and all of 
its associated components.\23\ In addition, Order No. 620 provides that 
the Commission or its staff may request that a pipeline make its work 
papers available for review.\24\
---------------------------------------------------------------------------

    \23\ FERC Stats. & Regs. [Regs. Preambles] ] 31,115 (Dec. 13, 
2000); 65 FR 81335 (Dec. 26, 2000).
    \24\ Staff examined each pipeline's compliance with the 
requirements for reporting additional information on page 700 of the 
FERC Form No. 6. After substantial follow-up contacts by staff, with 
the 183 jurisdictional oil pipelines, only 12 pipelines were not in 
compliance and expressed the need for additional accounting help in 
executing a complete Form No. 6. Staff referred the pipelines to 
either the Association of Oil Pipelines personnel or staff resources 
for advice.
---------------------------------------------------------------------------

    53. Finally, Shippers submit that the U.S. Court of Appeals-
approved methodology is no longer reliable because the results are 
sensitive to small changes in the sample size. Mr. Ashton argued in his 
Supplemental Declaration that the results achieved by using the 
Commission's methodology accords undue weight to extreme data points at 
the high and low end of the spectrum, with high cost oil pipelines 
exerting a disproportionately strong impact. Using Dr. Smith's 
``trimmed percent sample,'' Mr. Ashton notes that by simply removing 
the four highest and lowest pipelines from the data set, the cost index 
fell from a range of PPI+0.66 percent and PPI+0.69 percent to PPI+0.58 
percent and PPI+0.68 percent (middle 80 percent and 50 percent 
respectively).\25\
---------------------------------------------------------------------------

    \25\ Supplemental Declaration of Peter K. Ashton at 22-24.
---------------------------------------------------------------------------

    54. The sensitivity analysis Mr. Ashton includes in his 
Supplemental Declaration does not provide a reasoned basis for the 
Commission to abandon its current methodology because the existence and 
proper treatment of ``outlier'' data were extensively addressed in 
prior Commission proceedings, and the current U.S. Court of Appeals-
approved methodology was specifically designed to take this matter into 
account. To minimize the risk that extreme and/or erroneous 
observations bias the result, the Commission uses only the middle 50 
percent and 80 percent of the relevant cost data, thus ensuring that 
the index is not driven by statistical outliers.

Possible Over-Recovery of Costs-of-Service by Large Pipelines

    55. Shippers point to the fact that a significant number of large 
oil pipelines are substantially over-recovering their cost-of-service 
as further support for maintaining the PPI without adjustment as the 
appropriate index for the subsequent five-year period. Shippers 
maintain that the index methodology was designed to enable pipelines to 
recover costs by permitting them to increase rates at the same pace as 
they are predicted to experience cost increases. Shippers contend that 
the role of the index is to accommodate normal cost changes, not to 
guarantee recovery of all costs at any time and in full. Shippers state 
that a concern of the Commission was that under an indexing system, 
rates would diverge from actual costs and the resulting rates would 
fail the just and reasonable test. According to Shippers, this concern 
has proven to be well founded, as evidenced by data reported on FERC 
Form No. 6 which indicate that a number of interstate pipelines have 
been charging indexed rates that permit them to substantially over-
recover their cost-of-service of fourteen pipelines during the 2002-
2004 period that were subject to indexing regulation, the total 
interstate operating revenues of these pipelines, and the amount by 
which each of these pipelines have been over-recovering its self-
reported cost-of-service. Shippers conclude that clearly an increase in 
the current index will further increase the amount of over-recoveries 
by these pipelines.
    56. AOPL responds that this argument is flawed as a matter of 
regulatory policy and lacks any basis in evidence. The PPI index 
reflects the year-to-year changes in industry costs in general and, as 
such, reflects changes in the ``average'' oil pipeline's cost of 
service. Under the indexing system, existing rates remain subject to 
the Commission's complaint process. If Shippers believe that individual 
pipelines are over-earning such that their rates cannot satisfy the 
just and reasonable requirement for oil pipeline rates, their remedy 
would be to file complaints against those pipelines. AOPL further 
argues that another reason why over-earning pipelines do not represent 
a flaw in the Commission's index methodology is that the index governs 
rate changes based on ``grandfathered'' rates. AOPL claims that the 
rates of most oil pipelines are deemed to be just and reasonable, 
thereby establishing a ``baseline'' for future rates. Therefore, to the 
extent a carrier was over-earning in 1992, the indexing methodology was 
not intended to drive those rates to cost, but instead to make sure 
that any rate changes were based on expected cost changes. As to 
Shippers' comparison of the cost of service to revenues for their small 
sample of pipelines, AOPL points out that five of those fourteen 
pipelines have, in whole or in part, ``market-based'' rates and as such 
are not fully subject to the index. In addition, AOPL claims that 
another of the pipelines on their list has rates based on a rate 
negotiated with Shippers in 2002. For all the above reasons, AOPL 
disputes the position taken by Shippers.
    57. The Commission is not subject to a statutory duty to examine 
the whole rate when an oil pipeline proposes an indexed rate change. 
Rather, our inquiry is limited to a comparison of the changes in the 
rates and costs from year to year. We recognized in adopting a uniform 
index for all pipelines that inevitably some pipelines would over-earn 
while others will under-earn. It is a fact simply inherent in an 
industry-wide pipeline index. Shippers' use of a sample of fourteen 
pipelines culled from the entire data set of pipelines being analyzed 
only serves to emphasize this point. Further, Shippers' own 
calculations show that many of these pipelines actually experienced a 
decrease in their over-recoveries over the short time period being 
considered. In addition, even though Shippers' calculations may 
accurately measure over-recovery for a few pipelines, AOPL shows that, 
based on Page 700 information for 2003 and 2004, pipeline

[[Page 15337]]

revenues were 20 percent below booked costs of service. For the above 
reasons, the Commission finds that the existence of such over-
recoveries does not mean that PPI is the most appropriate index.

Structural Changes in the Oil Industry

    58. Shippers claim that structural changes in the oil industry 
ensure that adequate capital will be available to pipelines if they 
charge rates determined by the PPI index level. They point to the 
emergence of publicly-traded partnerships such as master limited 
partnerships (MLPs) and limited liability companies that have elected 
to be taxed as partnerships. According to Shippers, MLPs have resulted 
in increased concentration in the pipeline industry, and are the forms 
through which many pipelines subject to indexing are owned. To evaluate 
the impact of indexing, Shippers state that the environment of MLPs 
must be reviewed as well. According to Shippers, MLPs continue to enjoy 
good access to capital markets, and the number, size and total amount 
of capital raised by MLPs continues to grow. Shippers argue that MLP's 
success at capital raising is being accompanied by an active 
acquisitions market as well as by partnerships' continued investment in 
energy infrastructure (organized growth) projects, thus proving that 
raising capital is not a problem for the oil industry. Shippers also 
argue that another important consideration is the use of funds 
generated from pipeline operations. Shippers state that MLPs generally 
distribute all available cash flow to unit-holders in the form of 
quarterly distributions (similar to dividends). Thus, Shippers contend, 
one cannot assume that increases in rates resulting from an increase in 
the Commission's index will be used to offset any increased costs for 
safety, efficiency and security, or to fund capital expansion. Shippers 
conclude that no basis exists for the proposition that extraordinary 
rate increases must be approved across-the-board in order to provide 
sufficient capital for oil pipelines to expand and operate their 
systems in a safe and secure environment. On the contrary, Shippers 
contend, the available evidence suggests that more than sufficient 
capital is presently available at rates determined by the PPI to 
achieve these objectives.
    59. AOPL responds that the Shippers are attributing what is 
occurring for only a small sample of the entire pipeline industry 
(approximately 38 energy-related MLPs exist as of August 2005) to the 
entire industry. AOPL states that what happens with MLPs means 
absolutely nothing for the majority of oil pipelines that are not owned 
and operated by MLPs. According to AOPL, the very purpose of the five-
year review of the cost index is to ensure that pipeline rates keep 
pace with cost changes in the industry so that past levels of capital 
investment can be maintained. AOPL argues that, to preserve adequate 
capital investment, the Commission must adopt a price index consistent 
with its cost standard of at least PPI+1.3 percent.
    60. The fact that oil pipelines have been able to attract capital 
in the past does not establish that they would be able to do so in the 
future if the Commission fails to set an index that is adequate. We 
believe the continuation of the methodology used in Order No. 561 to 
arrive at the new index accurately captures costs in the interstate oil 
pipeline transportation sector, and will produce an index sufficient 
for pipelines to maintain their capital investment. We find adherence 
to the Order No. 561 methodology supports an oil index of PPI+1.3 for 
the next five years. Finally, we find that no party has made a 
convincing showing that the Order No. 561 methodology is no longer 
adequate for computing the oil index.

Continuation of the Current PPI and Impairment of Pipeline Expansion

    61. Shippers claim that, contrary to the brief filed by DOT, the