UNITED STATES OF AMERICA 71 FERC 61,291 FEDERAL ENERGY REGULATORY COMMISSION
Williams Pipe Line Company ) Docket Nos. IS90-21-003, ) IS90-31-003, IS90-32-003, ) IS90-40-003, IS91-1-003, ) SP91-3-003, SP91-5-003, ) IS91-28-003, IS91-33-003, ) OR93-1-001 Enron Liquids Pipeline Company ) IS90-39-003, IS91-3-001, ) and IS91-32-001 ) (Phase I) and Williams Pipe Line Company ) IS92-26-000, IS95-2-000, ) and IS95-7-000
OPINION NO. 391-A
OPINION AND ORDER GRANTING IN PART AND DENYING IN PART REHEARING
Issued: June 6, 1995
UNITED STATES OF AMERICA FEDERAL ENERGY REGULATORY COMMISSION
Before Commissioners: Elizabeth Anne Moler, Chair; Vicky A. Bailey, James J. Hoecker, William L. Massey, and Donald F. Santa, Jr.
Williams Pipe Line Company ) Docket Nos. IS90-21-003, ) IS90-31-003, IS90-32-003, ) IS90-40-003, IS91-1-003, ) SP91-3-003, SP91-5-003, ) IS91-28-003, IS91-33-003, ) OR93-1-001 Enron Liquids Pipeline Company ) IS90-39-003, IS91-3-001, ) and IS91-32-001 ) (Phase I) and Williams Pipe Line Company ) IS92-26-000, IS95-2-000, ) and IS95-7-000
OPINION NO. 391-A
OPINION AND ORDER GRANTING IN PART AND DENYING IN PART REHEARING
(Issued June 6, 1995)
On August 29, 1994, Williams Pipe Line Company (Williams) and the Association of Oil Pipe Lines (AOPL) filed requests for rehearing and clarification of the "Opinion and Order on Initial Decision, on Motion Proposing Rate Standards, and on Complaint and Protest" that was issued by the Commission on July 28, 1994 (Opinion No. 391). 1/ Texaco Refining and Marketing Inc. (Texaco) also filed a petition for rehearing of Opinion No. 391. That opinion affirmed and modified an initial decision of the presiding administrative law judge (ALJ) relating to tariffs filed by Williams proposing changes in its rates for the transportation of petroleum products. As discussed below, the Commission grants in part the requests for rehearing and clarification and denies the remainder of the requests for rehearing.
On February 6, 1995, Murphy Oil USA, Inc. (Murphy) filed a motion in Docket Nos. IS92-26-000, IS95-2-000, and IS95-7-000
1/ Williams Pipe Line Co., 68 FERC 61,136 (1994).
Docket No. IS90-21-003, et al. - 2 -
asking the Commission to sever from those consolidated proceedings a limited issue involving certain shorthaul rates filed by Williams. As discussed below, the Commission denies Murphy's motion.
Background
The background of this proceeding is described in Opinion No. 391 and will not be repeated here. 2/ Opinion No. 391 addressed exceptions to the ALJ's initial decision in Phase I 3/ of the proceeding. 4/
In Opinion No. 391, the Commission affirmed the ALJ's decision to limit the scope of Phase I to a determination of Williams' market power in the relevant markets. 5/ In defining market power for these proceedings, the Commission adopted the ALJ's findings that (1) no specific percentage of price increase is required as a test for market power; (2) the relevant product market is delivered pipelineable petroleum products; (3) the use of destination markets is appropriate for determining the geographic scope of the markets; and (4) Williams' destinations are the relevant BEAs. 6/
In analyzing Williams' markets, the Commission accepted the ALJ's use of the Herfindahl-Hirschman Index (HHI) to develop an initial screen; however, in its examination of the markets, the Commission used the HHI less stringently than did the ALJ, employing it as an indicator to be evaluated along with other factors. In calculating the HHIs, the Commission affirmed (1) the ALJ's use of capacity rather than delivery data; (2) his methodology for calculating and measuring effective capacity; and
2/ Id. at 61,654-55.
3/ Williams elected to bifurcate the proceedings in accordance with the procedures adopted by the Commission in Buckeye Pipe Line Co., 44 FERC 61,066 (1988), order on reh'g, 45 FERC 61,046 (1988), Opinion and Order on Initial Decision, 53 FERC 61,473 (1990) (Opinion No. 360), order on reh'g, 55 FERC 61,084 (1991) (Opinion No. 360-A) (collectively (Buckeye).
4/ Williams Pipe Line Co., 58 FERC 63,004 (1992).
5/ 68 FERC 61,136 at 61,656.
6/ Id. at 61,658-61. BEAs are areas of the contiguous United States that have been established by the Bureau of Economic Analysis of the U.S. Department of Commerce and are intended to represent actual areas of economic activity. A major city is at the hub of each BEA.
Docket No. IS90-21-003, et al. - 3 -
(3) where appropriate, the inclusion of private pipelines, barges, refineries, and potential competition. The Commission declined to apply a mechanical analysis in assessing the effect of truck transportation of products from sources outside of a BEA. 7/
The Commission also evaluated a variety of other factors bearing on competition, finding that (1) the ALJ had not applied an excessively high standard in assessing market share; (2) exchanges should be entitled to little weight in the post- screening review of the markets; (3) there is no precise formula by which to determine whether sufficient weight has been given to excess capacity in the analysis; (4) the evidence does not support a conclusion that the presence of vertically integrated companies in Williams' markets justifies less regulation of Williams; (5) where buyer power is shown, it should be entitled to some weight; and (6) profitability is a neutral factor in this case. 8/
The Commission then turned to an examination of the individual BEA markets. It found that Williams had established that it lacks market power in the Chicago, St. Louis, Oklahoma City, Tulsa, Wichita, Springfield/Decatur, Peoria, Rockford, Wausau, Dubuque, Davenport, Columbia, and Minneapolis/St. Paul markets. 9/ The Commission also found that Williams had failed to establish that it lacks market power in the Springfield (MO), Eau Claire, Des Moines, Kansas City, Lincoln, Fargo, Grand Forks, Duluth, Rochester, Sioux City, Topeka, Omaha, Grand Island, Sioux Falls, Aberdeen, Quincy, Cedar Rapids, Waterloo, and Ft. Dodge markets. 10/ The Commission directed the ALJ to proceed with Phase II of the proceedings for the purpose of establishing base rates for these noncompetitive markets.
The Commission addressed exceptions to the ALJ's rulings on certain claims of discrimination raised by the intervenors. In each case, the ALJ had found that the alleged forms of discrimination were intended to meet competitive conditions and did not constitute an abuse of market power. The Commission affirmed the ALJ's reservation for Phase II of the issues of cross-subsidization and justifiable differences in costs. 11/ Further, the Commission denied Williams' motion proposing rate
7/ Id. at 61,663-70.
8/ Id. at 61,672-75.
9/ Id. at 61,675-78, 61,682.
10/ Id. at 61,679-86.
11/ Id. at 61,688.
Docket No. IS90-21-003, et al. - 4 -
standards to apply to Phase II of this proceeding, indicating that the method to be used for establishing base rates would be at issue in the second phase. 12/
Requests for Rehearing and Clarification
Williams argues that the Commission erred in (1) finding that Williams had failed to show that it lacked market power in several of its markets; (2) improperly recalculating the HHIs for Williams' markets by excluding certain external sources; (3) relying unduly on delivery-based market shares rather than effective capacity-based market shares; (4) failing to accord the HHI primary significance over other statistics, specifically market share statistics; (5) failing to use the HHI as a screen to identify markets in which Williams should be presumed to lack significant market power; (6) declining to give sufficient weight to exchanges; and (7) ignoring evidence of Williams' inability to sustain profitably a 15 percent price increase.
Williams challenges the Commission's determinations that Williams failed to demonstrate its lack of market power in the following nine markets: (1) Springfield, MO; (2) Kansas City; (3) Lincoln; (4) Quincy; (5) Des Moines; (6) Omaha; (7) Eau Claire; (8) Fargo; and (9) Grand Forks. Williams also asserts that the Commission erred in reserving for Phase II further analysis of the discrimination claims on a cost basis in light of the Commission's affirmation of the ALJ's decision that Williams proved that the challenged rate disparities are justified by competition. Finally, Williams asks the Commission to clarify that, in Opinion No. 391, the Commission has not ruled on the merits of Williams' proposed rate standards in any respect and that the parties have the right in Phase II to rely on the Phase I record regarding costs and cost standards, subject to the right of any party to supplement that evidence.
Texaco maintains that the Commission erred in rejecting the consideration of corridors or origin-destination pairs in this case. In the alternative, if the Commission believes that origins are not relevant, Texaco argues that Williams must post a single rate to each destination from all origins, a rate no higher than that of Williams' principal competitors. Finally, Texaco asserts that the Commission erred in finding that Williams cannot exercise significant market power in those markets in which the pipeline is responsible for over 45 percent of the deliveries.
AOPL faults Opinion No. 391 for failing to adhere to the standards for evaluating market power articulated by the Commission. Further, AOPL contends that the Commission attached
12/ Id. at 61,695.
Docket No. IS90-21-003, et al. - 5 -
undue importance to market share by calculating that factor based on delivery data and by ignoring the concept of relative ability to sustain a price increase over a significant period of time as a measure of market power. AOPL also asserts that the Commission erred by requiring rate differentials found in Phase I to be justified by competitive conditions to be evaluated in Phase II for possible cross-subsidization, as well as requiring the rate differentials to be justified based on costs. Finally, AOPL objects to the Commission's rejection of Williams' proposed use of the stand alone cost method for Phase II.
Discussion
I. Analysis of Market Power
Williams, AOPL, and Texaco challenge the Commission's holdings respecting the general concepts employed in the market power assessment.
A. Exclusion of External Sources
In Opinion No. 391, the Commission declined to apply a mechanical analysis with a specific mileage limit to exclude external sources. While the Commission recognized that a great deal of product arrives in Williams' markets via trucks, many of which travel from origins more than 50 miles from a BEA, the Commission examined the 65 to 70 mile limit established by the ALJ and found that the distance of a source from a BEA might have little bearing on the economic ability of the source to compete effectively in the BEA. 13/
Williams acknowledges that the Commission judged the effects of external competition based on the facts of the individual BEAs; however, Williams argues that the record does not support the Commission's unduly conservative conclusions concerning the effect of truck transportation of petroleum products into Williams' markets. According to Williams, the intervenors' own evidence confirmed that trucks compete with pipelines at distances of 70 and 100 miles, particularly in rural areas. 14/ Williams also cites the surveys conducted by its Witness Jones, which Williams claims support its position that trips in excess of 100 miles are common and that trips in excess
13/ 68 FERC 61,136 at 61,669-70.
14/ Citing the testimony of Conoco's Witness Stockebrand (Ex. 749, Ex. 753, Tr. 46/7921-26); Witness Swerczek (Tr. 47/8228-32); staff's Witness Alger (Tr. 42/7068-128); Ex. 801; Ex. 735; Tr. 45/7803; Tr. 48/8479-81.
Docket No. IS90-21-003, et al. - 6 -
of even 200 miles are not infrequent. 15/ Williams points to its Witness Bollen's survey of gasoline station operators 16/ and two statistical tests performed by its Witness Schink, all of which Williams argues confirm the results of Witness Jones' surveys. 17/
Williams states that the only conflicting testimony considered by the ALJ concerning the economic limits on truck transportation was submitted by Kerr-McGee. However, Williams argues that the ALJ incorrectly interpreted Kerr-McGee's information in reaching his conclusion that the limit beyond which trucks generally cannot travel cost-effectively is 70 miles; when corrected, Williams states that the actual effective range is increased to 90 to 95 miles.
Moreover, continues Williams, in recalculating the HHIs to reflect what the Commission considered to be reasonable economic limits on truck movements from external sources, Opinion No. 391 compounds the unduly conservative approach of the ALJ. Specifically, Williams states that the Commission improperly accepted the intervenors' position regarding the length of haul for external sources, which in all cases was claimed to be to the center of the particular BEA. However, Williams contends that trucks do not have to penetrate all parts of a BEA to make a price increase unprofitable. Williams concludes that, because the effective capacity-based HHI already discounts the competitive strength of external sources, discarding those sources entirely, as the Commission did, constitutes an unwarranted double discounting of external sources, creating the false impression that there is no competitive impact from sources serving only part of a BEA.
The Commission grants partial rehearing on this issue. As will be discussed in greater detail in the context of the individual BEAs, in reassessing the viability of external sources, we have carefully examined the record evidence highlighted by Williams in its rehearing request concerning truck deliveries to the individual BEAs and the relative costs of truck, barge, and alternative pipeline deliveries into these markets. Williams sponsored Exhibits 303-337, which in effect aggregate and summarize the other evidence it cites in its request for rehearing, to validate its claim that specified external sources competitively serve the individual BEA markets and, thus, should be included in the assessments of those markets. Williams' exhibits include refinery, truck, and gas
15/ Citing Ex. 295 at 26-27; Ex. 308; Exs. 311-337.
16/ Citing Ex. 501 at 7, 23-24.
17/ Citing Ex. 260 at 126-175.
Docket No. IS90-21-003, et al. - 7 -
station surveys that document truck shipments by origin and destination, as well as extensive tables for each BEA that compare transportation costs for Williams with those of a number of internal and external sources. These cost tables provide pipeline tariffs and a consistent reference guide to trucking costs for distances ranging from eight to 425 miles. On rehearing, the Commission is now persuaded that these comparative cost tables provide a sound basis on which to evaluate the ability of external sources to serve a market competitively, as the comparative costs are highly relevant in determining whether a source can represent effective competition. We are now inclined to give less weight to the Kerr-McGee evidence relating to the economic limits of truck transportation, as it is less comprehensive. Our revised findings are addressed in the discussion of the individual BEAs below.
B. Use of Delivery-Based Market Shares Versus Capacity-Based Market Shares
In Opinion No. 391, the Commission affirmed the ALJ's use of capacity data in calculating the HHIs for Williams' markets, but held that use of delivery data as an additional and secondary indication of market concentration in determining the market share does not produce a distortion and, in fact, permits each methodology to offset the inherent deficiencies of the other. 18/
On rehearing, Williams and AOPL challenge the Commission's reliance on delivery data. Both emphasize that delivery data focus on historical events and provide no indication of a future response to a price increase. AOPL argues that the correct indicator of market power is market behavior -- the potential competitive response by the market to any attempt to exercise market power. According to AOPL, excessive reliance on market share, particularly when calculated on the basis of actual deliveries rather than capacity, ignores the potential for competitive response.
AOPL asserts that a determination of market behavior requires an evaluation of a host of factors, only one of which is market share. AOPL also emphasizes the potential ability to sustain a small but significant nontransitory increase in price as a key factor in assessing market behavior.
According to AOPL, the Commission's undue reliance on delivery-based market share information distorts the ultimate determinations of market power. For regulatory purposes, as opposed to merger policy, AOPL claims that the focus on market share should be more on market dominance. Typically, states
18/ 68 FERC 61,136 at 61,665.
Docket No. IS90-21-003, et al. - 8 -
AOPL, federal courts have employed market share thresholds of 60 percent and higher to indicate dominance. 19/
Williams and AOPL assert that the record in this proceeding demonstrates that delivery data -- especially where competitors are involved -- are not readily available and are of questionable accuracy. 20/ In contrast, states AOPL, capacity data tend to be public information and are more readily verified. AOPL concludes that, if the Commission continues to use market share both for HHI purposes and as a separate and discrete screen, then at least the potential competition afforded by the capacity of the various competitors in the market should be used for both purposes.
Although the Commission found delivered pipelineable petroleum products to be the product in this case, Williams argues that the product actually is the capacity to supply
19/ Citing U.S. v. Aluminum Company of America, 148 F.2d 416 (2d Cir. 1945); U.S. v. Grinell Corp., 384 U.S. 563 (1966); Holleb & Company v. Produce Terminal Cold Storage Co.; 532 F.2d 29 (7th Cir. 1976); Hiland Dairy v. Kroger Co., 402 F.2d 968 (8th Cir. 1968), cert. denied, 395 U.S. 961 (1969).
20/ Citing Tr. 27/3755, 3758, 3760-61; Tr. 32/4836-37. Williams points out that its delivery data are unavoidably flawed and incomplete, rendering market shares calculated using these data highly suspect. Specifically, Williams explains that delivery data are available only from Williams and not from any of its competitors. Further, asserts Williams, even the delivery data it provided cannot be assigned accurately to BEA markets because of serious reporting errors and omissions on the bill of lading records that are filled out by truckers who deliver the product to its final destination. Williams states that these truckers often enter home office or billing address locations in the delivery destination field or report vague destinations, such as "various." Williams states that all of its pipeline deliveries to non-Williams off-line terminals are recorded in Williams' delivery data as being consumed at the site of these off-line terminals. In fact, maintains Williams, trucks take this product to locations both inside and outside the BEA market. Finally, Williams points out that, in at least two BEAs, it shares terminals with other companies, making it impossible to determine truck deliveries by pipeline because bill of lading records do not provide that information.
Docket No. IS90-21-003, et al. - 9 -
petroleum products in a BEA. 21/ Finally, Williams asserts that, even if delivery-based market share data have some probative value, that would not justify simply disregarding capacity-based market share data which, according to Williams, demonstrate clearly that Williams lacks market power.
The Commission denies rehearing on this issue. We acknowledge Williams' concerns regarding reliance on the delivery data that were available in this case. As Williams has noted, there are serious deficiencies in the market delivery data available in this proceeding. First, such data were only available for Williams' deliveries. Second, delivery data are not kept in a uniform or consistent fashion that would permit accurate or uniform analysis. However, those concerns were taken into account in our choice of capacity as the basis for our market assessment. Our primary measure of market concentration was capacity-based, and we used Williams' delivery figures only as a secondary consideration. We did not, as Williams alleges, disregard capacity-based market shares, nor did we base any market finding solely on delivery-based market share. As discussed below, we have rejected Texaco's argument that we do so. 22/
We emphasize again that we find it redundant to use a capacity-based HHI and a capacity-based percentage of a given market, as though these were two separate measures. They are not. While expressed differently, both measure Williams' share of available capacity in the market. Deliveries into a market by individual pipelines may be somewhat less than their available capacity. Therefore, it is more informative to compare relative capacity and relative deliveries when accurate figures are available. It is even more useful when actual delivered product costs are also available for comparison. Any serious discrepancy between relative capacity, relative delivered costs, and relative delivery-based market shares would cause us to examine a particular market more closely.
We also believe that our use of capacity-based HHIs addresses sufficiently the issues raised by AOPL concerning market behavior and market dominance. AOPL claims that our consideration of delivery data was misplaced and led the Commission to ignore the more important market power measure of relative ability to sustain a price increase over a significant
21/ Williams cites the testimony of Witness Schink, claiming that because marketers and refiners determine where they will market product and how capacity is deployed, available capacity best depicts the likely response to an exercise of market power. Tr. 28/3881-83.
22/ See infra part I.C.
Docket No. IS90-21-003, et al. - 10 -
period of time. Our use of capacity based HHI's does, contrary to AOPL's assertion, permit consideration of market responses to a price increase. Our capacity-based HHI's take into account both excess and potential capacity, the existence of which in a market provides the potential for rapid response to a significant price increase.
AOPL is correct that we did not use as the basis for our analysis the measurement of specific responses to specific price increases. The record in this case does not contain data that would enable us to make the price analysis that AOPL seeks. The parties have been inconsistent and sporadic in their exploration of the potential for new entry under sustained price increases. We are unable to use price increases/potential capacity as a tool for market definition in this case because Williams has not presented sufficient evidence on alternatives in that context. Additionally, there is not sufficient information on actual costs in the record to permit us to perform the analysis independently.
In Opinion No. 391, we found the relevant product to be delivered pipelineable petroleum products, recognizing that a large volume of product arrives in Williams' markets via other modes of transportation and through exchanges. 23/ We find no merit in Williams' argument that capacity is the true product in this case, which is essentially the same point that Texaco raises in its contention that the Commission should use corridors in analyzing the markets in this proceeding. We have consistently rejected the use of corridors in this proceeding, and find nothing new in the claims of Williams and Texaco that persuade us to reverse that position.
C. Weight Accorded Effective Capacity-Based HHI Statistic
In Opinion No. 391, the Commission affirmed the ALJ's consideration of Williams' delivery-based market share in markets identified for further examination by the HHI screening. The Commission concluded that the ALJ did not establish an excessively high standard in assessing delivery-based market share, nor did he establish an absolute threshold of 70 percent as indicative of market power. For example, the ALJ did not exclude automatically from examination those markets in which Williams' share is below 50 percent, although the Commission pointed out that the ALJ found those markets to be "less troublesome." 24/
23/ 68 FERC 61,136 at 61,659.
24/ 68 FERC 61,136 at 61,672.
Docket No. IS90-21-003, et al. - 11 -
Williams states that the HHI statistic takes into account the market shares of all market participants. Therefore, if any market participant has a large share, the resulting high HHI will indicate that the market is highly concentrated and that one or more of the market participants may have market power. A low HHI, on the contrary, indicates that no participant can have market power. Thus, reasons Williams, market shares for individual market participants are relevant only when the HHI statistic is high.
Williams further states that the effective capacity-based HHI statistic used by staff also takes into account the number of companies supplying the market and the capacity of these companies to do so. Williams explains that, for external supply sources, the capacity to serve the market is measured not only in terms of the capacity of the sources but also in terms of the reach of trucks from these sources into the BEA. According to Williams, an external source typically serves only a small fraction of the market, and the effective capacity-based HHI credits the source only to that extent. Therefore, asserts Williams, the effective capacity-based HHI appropriately discounts the competitive influence of external supply sources relative to internal suppliers. Williams concludes that the Commission should consider the effective capacity-based HHI as the primary statistic and should use market share statistics only when the HHI is high.
Rehearing is denied on this issue. We find that the question of the weight accorded capacity has been addressed adequately in our calculations, as reflected in Opinion No. 391. Although in Buckeye, we relied on delivery data in calculating the HHIs, we recognized that circumstances might warrant the use of other appropriate data in other cases. 25/ In the instant case, we used capacity data because it was more complete than the delivery data. In fact, delivery data was only available for Williams and not for other market participants. We also emphasize again that we have based no market decision solely on delivery-based market share, but have employed that factor only as a secondary indicator of market power.
In its request for rehearing, Texaco contends that the Commission erred in finding that Williams cannot exercise significant market power when Williams has over 45 percent of the deliveries in a particular market, such as in the Columbia BEA. In the alternative, Texaco asks the Commission to reconsider its arguments relating to the competitiveness of the challenged BEAs where Williams was allowed to set rates on its own. Texaco offered no specific examples other than Columbia, which will be
25/ Opinion No. 360, 53 FERC 61,473 at 62,667; Opinion No. 360-A, 55 FERC 61,084 at 61,261.
Docket No. IS90-21-003, et al. - 12 -
addressed below in the context of the individual BEAs. It thus appears that Texaco's real argument relates to the magnitude of a price increase in a particular market, rather than to the general concept of market share. Therefore, the Commission denies Texaco's request for rehearing on the question of whether a delivery-based market share of 45 percent or greater can permit Williams to exercise market power in a market.
D. Use of the HHI as a Screen
Williams asserts that the Commission correctly affirmed the ALJ's use of a 2500 HHI as an initial screen to identify markets warranting further scrutiny, but erred in holding that the HHI is an analytical tool rather than an irrebuttable presumption of lack of market power. 26/ Although, in its rehearing request, Williams maintains that an HHI of 2500 should constitute an irrebuttable presumption that a market is workably competitive, Williams declines to seek rehearing on this point, but does ask the Commission to hold that markets with HHI's less than 1800 are irrebuttably deemed workably competitive. AOPL contends that such screens do not prejudice the parties' rights, but rather focus the debate and streamline the proceeding.
Williams asserts that the record in this case supports its position. Williams cites the testimony of the economic witnesses, 27/ the Commission's decision in Buckeye, 28/ the Department of Justice study addressing oil pipeline deregulation, 29/ and the ALJ's decision in this proceeding 30/ as being in agreement that the HHI alone should be the initial screen for identifying markets requiring further analysis. Williams states that the superiority of the HHI as an index of market power stems from the fact that the HHI, unlike delivery-based market share, reflects both the number and size of all firms in the market. 31/ Williams disputes the
26/ See 68 FERC 61,136 at 61,661-63.
27/ Citing Ex. 208 at 3; Ex. 619 at 47; Ex. 932 at 6.
28/ Citing 53 FERC 61,473 at 62,663.
29/ Citing Ex. 18 at 29-31. U.S. Dep't of Justice, Oil Pipeline Deregulation (1986) (Oil Pipeline Study).
30/ Citing 58 FERC 63,004 at 65,023.
31/ Citing FTC v. University Health, Inc., 938 F.2d 1206, 1211 n.12 (11th Cir. 1991); FTC v. PPG Industries, Inc., 798 F.2d 1500, 1503 (D.C. Cir. 1986); First and First, Inc. v. Dunkin' Donuts, Inc., 1990-1 Trade Cases, 68,989 at 63,368 (continued...)
Docket No. IS90-21-003, et al. - 13 -
Commission's statement that its approach in this case is consistent with Buckeye, claiming that the dispute in Buckeye was between an HHI threshold of 1800 and 2500, rather than whether a threshold, if adopted, would in fact serve as an irrebuttable presumption. Further, contends Williams, in Buckeye, the Commission found all markets with HHIs below 1800 to be workably competitive.
Rehearing is denied on this issue. An HHI of 1800 merely indicates that the shippers in a market have at least six good choices, suggesting that the market is not captive. However, the Commission continues to believe that, for its purposes in assessing market power, the HHI should be used as an analytical tool, along with other criteria, rather than establishing an irrebuttable presumption of lack of market power when it falls below a specific number. This practice gives the Commission the greatest flexibility in discharging its legal responsibilities under the Interstate Commerce Act (ICA). We are not persuaded that the parties' claimed need for certainty mandates a different decision. The parties still must offer evidence sufficient for the Commission to assess market power in each market, and our unwillingness to establish a firm HHI threshold in a particular case does not add significantly to that obligation.
The testimony cited by Williams is not as definitive as Williams suggests. The passages cited explore the preferences of each witness for a particular HHI value (1800 vs. 2500), as being indicative of a lack of market power. However, the witnesses do not espouse a particular number as the basis for an irrebuttable presumption of competition. Rather, each witness addresses the need to consider a number of factors in making a market power determination.
Further, we do not read the Oil Pipeline Study to require that a particular HHI can establish an irrebuttable presumption of competitiveness. Although the study does state that, with a BEA of less than 2500, competitive concerns are presumed to be small relative to the costs of pipeline regulation, 32/ we find no language that would bar additional analysis, particularly if other factors present in a case suggest that further examination may be of value.
We also disagree that Buckeye supports the notion that a particular HHI can serve as an irrebuttable presumption of lack of market power. In that proceeding, the parties argued for an
31/(...continued) (E.D. Pa. 1990); United States v. LTV Corp., 1984-2 Trade Cases 66,133 at 66,336 n.5 (D.D.C. 1984).
32/ Oil Pipeline Study at 30.
Docket No. IS90-21-003, et al. - 14 -
HHI threshold, but the Commission concluded that, while the use of HHIs is the appropriate first step in evaluating the likelihood of market power, knowing the degree of concentration in a market merely "provides useful information about where on the competitive spectrum that market likely lies and what other factors will have to be weighed to enable a finding as to the existence or absence of significant market power." 33/ Although the Commission stated that, where a market had a "particularly low" HHI, it did not undertake further review, the Commission did not recognize any specific number as raising an irrebuttable presumption of competitiveness. 34/
Neither do we agree that the ALJ's decision supports the notion of a particular HHI number raising an irrebuttable presumption of lack of market power. While the ALJ chose to use 2500 as an initial screen for his analysis, he acknowledged that in Buckeye, the Commission did not adopt a particular number, 35/ and he recognized that caution suggested a look at BEAs with HHIs below 2500 where Williams has a high market share. 36/ His choice of 1800 as a lower limit appears to be based in large part on the fact that several parties had urged adoption of that number as the screen. Accordingly, based on all of these considerations, we deny rehearing on this issue. 37/
E. Weight Accorded Exchanges
The Commission, in Opinion No. 391, affirmed the ALJ's conclusion that exchanges should be entitled to little weight in the post-screening review of the markets in this case. Further, the Commission recognized that the potential for double counting
33/ 53 FERC 61,473 at 62,667.
34/ Id. See also 55 FERC 61,084 at 61,254.
35/ 58 FERC 63,004 at 65,013.
36/ Id. at 65,023.
37/ In both the Buckeye proceeding and this proceeding we have used an initial screen of 2500. The Commission did so to follow the recommendations and the precedent of the Justice Department in its work on oil pipelines. More recently, the Commission has requested comments on a Staff Paper on Market-Based Rates for Natural Gas Companies. In that Paper, Staff proposed an initial screen of 1800 for analyzing natural gas pipeline transportation markets, proposing a relatively strict standard when the Commission considers market-based rates for natural gas pipelines. Request for Comments on Alternative Pricing Methods, Docket No. RM95-6-000, 70 FERC 61,139 at 61,403, 61,408.
Docket No. IS90-21-003, et al. - 15 -
exists where capacity is included in the HHI and the exchange utilizing the capacity is added into the HHI as well or considered a mitigating factor. The Commission pointed out that exchanges do not create new barrels of product and do not always involve the owner's taking physical possession of the barrels. Finally, the Commission concluded that exchanges tend to be negotiated with reference to Williams' rates rather than disciplining the rates. 38/
Williams disputes the conclusion that assigning significant weight to exchanges would result in possible double counting. According to Williams, it never asked the Commission to take exchanges into account as some sort of adjustment to HHIs, but merely asked the Commission to consider the extensive volume and pricing information contained in the record on exchanges. Williams claims that this record shows a consistent pattern of intense price rivalry among alternative sources, even in the BEAs found not workably competitive.
The Commission's use of supply capacity as a measure of market concentration already takes exchanges into account, as explained in Opinion No. 391. 39/ While exchanges may obviate the use of specific pipeline corridors between two markets, they do not obviate the need for ultimate delivery into the BEA. This must be accomplished with existing delivery capacity, including trucks, pipelines, barges, and direct refinery sales. All of these alternatives, where viable, have been included in our HHI calculations for individual markets.
We do acknowledge the considerable traffic in exchanges documented by Williams and the fact that this permits shippers to bypass all or part of the Williams system. However, this does not change the fact that delivery of exchanged product into a BEA must take place through the facilities of some existing supplier. The ability of suppliers to deliver exchanges into the BEA is already accounted for in our HHI analysis.
Williams has offered specific examples of what it claims to be the effects of exchanges in the Omaha, Duluth, Grand Island, and Sioux City BEAs. Although Williams alleges that product is received on exchange in the Omaha BEA, thereby eliminating the need for transportation on its system, as we have stated above, our use of supply capacity as a measure of market concentration already considers exchanges. Although Williams has not sought rehearing of our market power determinations relating to the Duluth, Grand Island, and Sioux City BEAs, the same principle applies in those markets.
38/ 68 FERC 61,136 at 61,673.
39/ Id. at 61,672-73.
Docket No. IS90-21-003, et al. - 16 -
F. Ability to Sustain a 15 Percent Price Increase
The Commission accepted the ALJ's definition of market power as "a firm's ability to sustain a price increase over a significant period of time, or to exclude competition." 40/ The Commission also affirmed the ALJ's rejection of any specific rate increase as a litmus test for market power. 41/
Williams and AOPL challenge the Commission's failure to establish a specific price increase threshold for determining the existence of market power. First, they contend, use of such a threshold flows from the definition of market power adopted by the ALJ. Williams and AOPL also object to the Commission's reliance on the Merger Guidelines 42/ for the proposition that a small but significant nontransitory price increase is only a methodological tool and does not establish a tolerance level for inferences concerning market power. According to Williams and AOPL, the point of the statement in the Merger Guidelines is that an small but significant nontransitory price increase is not intended to license a particular quantum of future price, and in any event, that is not what Williams seeks to do. Rather, state Williams and AOPL, the evidence of Williams' inability to sustain a 15 percent price increase demonstrates that it lacks the ability to increase its rates by that amount today.
Williams contends that the Commission ignored a great deal of evidence when it accepted the ALJ's finding that Williams studied the impact of a hypothetical rate increase in only three of its markets where market concentration suggested further scrutiny. According to Williams, the record demonstrates that its rates have not increased significantly in real terms over its pre-existing rates and actually have decreased in real terms, particularly its average rates to each of the BEAs where it was found not to have shown that it lacks market power. Further, emphasizes Williams, its average systemwide rates also decreased 6.6 percent in real terms, although its systemwide average nominal rates increased 13 percent.
Williams maintains that the record contains evidence that would have allowed the ALJ and the Commission to test the effect of a 15 percent increase in all of the markets that were subjected to further analysis. Williams states that it presented BEA-by-BEA evidence of the number of additional truckloads of product per day from non-Williams sources that would make a 15
40/ Id. at 61,657.
41/ Id. at 61,658.
42/ Dep't of Justice and Federal Trade Commission Horizontal Merger Guidelines (April 2, 1992).
Docket No. IS90-21-003, et al. - 17 -
percent real price increase by Williams unprofitable. 43/ Williams claims that this evidence is extremely conservative because it relies on the delivery-based market share data presented by the staff, which includes 100 percent of Williams' deliveries to non-Williams terminals even if they were trucked outside of the destination BEA. Despite that flaw, Williams seeks to demonstrate that, in most of the BEAs where the ALJ found that Williams had failed to demonstrate that it lacks market power, fewer than ten additional truckloads per day would be sufficient to make a 15 percent increase unprofitable. Finally, Williams states that the trucking cost table presented by Kerr-McGee and relied on by the ALJ also showed the extent to which external sources could cost-effectively serve each BEA as a result of a 15 percent real price increase by Williams. Williams concludes that the analysis, which it depicted in the maps contained in the individual BEA discussions in its Brief on Exceptions 44/ and the Appendix to its Brief Opposing Exceptions, 45/ confirmed that Williams would lose sufficient business to render such a price increase unprofitable.
The Commission denies rehearing on this issue. Opinion No. 391 thoroughly explains the Commission's reasons for rejecting the use of a specific level of rate increase as a litmus test for market power. Williams and AOPL raise no new arguments that were not addressed by the Commission in its previous order. For example, the Commission rejected Williams' argument that rate decreases in certain of its markets requires the Commission to establish a specific rate increase threshold for determining the presence or absence of market power. The Commission also pointed out that the Merger Guidelines reject mechanical application of the concept of a small but significant nontransitory price increase.
We also find a lack of agreement on the use and definition of a small but significant nontransitory price increase in this proceeding. When Williams alludes to the importance of such a price increase in measuring its competition, it looks at the
43/ Williams explains that it calculated its daily deliveries to each BEA by multiplying its net percentage share of deliveries by total daily consumption in the BEA. Citing Ex. 306 at 2; Ex. 354 at 3; Ex. 217 at 8. Then Williams determined the number of tank truckloads that would be required to displace 15 percent of those deliveries and presented the results in the individual BEA discussions in its Post-Hearing BEA Appendix.
44/ Brief on Exceptions of Williams Pipe Line Co. at 72-113.
45/ Brief Opposing Exceptions of Williams Pipe Line Co., BEA Addendum.
Docket No. IS90-21-003, et al. - 18 -
effect of a 15 percent price increase on competition in its markets, citing the reference to 15 percent in Buckeye, but adding 15 percent to its 1990 proposed rate increase. 46/ By contrast, Witness Alger, in his assessment of transit pipelines that could build terminals profitably in particular BEAs, used as his small but significant nontransitory price increase a $0.10 increase over pre-1990 rates. 47/ There is no need for the Commission to resolve this conflict.
G. Use of Corridors
Opinion No. 391 affirmed the ALJ's use of destination markets, rejecting Texaco's contention that corridors should be used in determining the relevant geographic markets in this proceeding. The Commission found that the real economic concern of shippers is the delivered product and its price rather than whether the product travels between specific locations via pipeline. Additionally, the Commission recognized that, if geographic markets were limited to specific origin/destination pairs, this would fail to recognize the economic concern of the shippers as well as eliminating from consideration competitive suppliers who bring product into the markets without utilizing the specific corridors. The Commission also explained that the availability of exchange options mitigates situations in which a shipper might otherwise be captive and that the use of destinations as the relevant market allows for those exchanges to be recognized and considered as the viable options that they are for shippers. Finally the Commission found that the use of destinations as the relevant markets is consistent with Buckeye. 48/
Texaco argues that, just as Williams may have captive and non-captive destination markets, with relief tailored accordingly, there should be recognition that both captive and non-captive corridors can exist. 49/ Texaco's request for rehearing on this issue is denied. In Opinion No. 391, the Commission thoroughly explained its reasons for rejecting the use of corridors, principally that defining geographic markets by specific corridors would eliminate from consideration those suppliers who bring product into a market without using the corridors.
46/ See, e.g., Ex. 208 at 38.
47/ Tr. 41/6997, 7006; Ex. 627.
48/ 68 FERC 61,136 at 61,660-61.
49/ Texaco states that the Commission recognized this in the NOPR for market-based ratemaking for oil pipelines. Citing 68 FERC 61,137, slip op. at 10, n.18 (July 28, 1994).
Docket No. IS90-21-003, et al. - 19 -
Texaco argues in the alternative that, if the only consideration is for destination markets, and corridors are immaterial, then there should be only a single competitive rate to each market where Williams lacks significant market power. Texaco maintains that, if Williams posts more than one rate to the market destination, it demonstrates either that (1) there is a segmented market in the destination, with one segment insulated from another, or (2) only some customers are receiving the benefit of any competition, contradicting the finding that competition will hold the rates within just and reasonable levels. Texaco asserts that the single rate it advocates should be set at the lowest of the rates posted, subject to verification that the lowest rate is no higher than the rate of the principal competitors.
Texaco has cited the Dubuque BEA as an illustration, claiming that, while Williams claims to face competition in this market, the range of rates in this BEA makes it appear that only some customers benefit from this competition. However, Texaco is incorrect when it assumes that competition requires that there be only one pipeline rate into a BEA. Competition will tend to effect a single market price of delivered product, but that does not require all suppliers to have identical costs at every level of the delivery chain. Competition, in fact, may require differential pricing, as will be discussed in greater detail later in this order. We also emphasize that competition does not preclude price increases. In any of its markets, if Williams' cost for product transportation is above market levels, it will lose customers. In the case of Dubuque, shippers not only have pipeline alternatives to use instead of Williams, but may receive barge deliveries as well. 50/
Texaco mistakenly assumes here that Phase I of this proceeding is for the purpose of establishing just and reasonable rates in individual markets, rather than for the purpose of performing a market analysis to determine market power. Market power that controls access to a destination is relevant to that market, and market power that controls egress from an origin market is relevant to that market; however, the options open to shippers in an origin market do not prove or disprove competition in the destination market. The fact that a pipeline may be able to post more than one rate to a market could also reflect differences in distances, costs, product prices, density of traffic, and trucking or delivery costs for competitors in various sectors of a market.
Texaco appears to be seeking an inappropriate result -- that Williams should post a single rate from all destinations into
50/ See Ex. 321.
Docket No. IS90-21-003, et al. - 20 -
each of its competitive markets. The Commission rejects that argument as well. To require a single transportation rate, which would be set at the lowest of the rates posted, subject to the verification sought by Texaco, would be to impose by regulation the desired outcome of a single party without regard to market realities. It would also be inconsistent with the Commission's philosophy of light-handed regulation where market forces are effective.
II. Challenged BEA Findings
With respect to the nine BEA markets where it challenges the Commission's findings, Williams argues that Opinion No. 391 (1) failed to take into account the recent entry of the Heartland and Cenex pipelines into Williams' markets; (2) ignored the surveys and interviews demonstrating that external sources are serving Williams' markets on a regular and ongoing basis; (3) failed to consider the impact of barge competition; (4) failed to consider the competition created by the Department of Defense (DOD) annual bidding process; and (5) misconstrued staff analyses concerning the potential for profitable construction of terminals on pipelines traversing several of Williams' markets. Texaco asks the Commission to grant rehearing and find the Columbia BEA to be noncompetitive, claiming that Williams has been able to raise its rates significantly in that market.
A. Springfield, MO
Despite the ALJ's 1317 HHI for this market and Williams' 38 percent delivery-based market share, the Commission found that Williams had failed to demonstrate that it lacks market power in this BEA. The Commission explained that this is a large BEA and that five of the external sources utilized in the ALJ's calculation are at too great a distance from the BEA border to provide economic competition. The Commission's recalculated HHI was over 3000. 51/
On rehearing, Williams states that the Commission excluded all external sources as well as three pipelines that do not have operating terminals within the BEA. As a result, contends Williams, the recalculated HHI is high because only three suppliers with capacity sufficient to supply the entire BEA were considered. 52/ Williams argues that the Commission is incorrect about the potential for terminals on the three excluded pipelines because the staff found that two terminals could be
51/ 68 FERC 61,136 at 61,679.
52/ The three pipelines are Williams, CRA-Farmland, and Cherokee.
Docket No. IS90-21-003, et al. - 21 -
constructed profitably within this BEA and should be included in the HHI calculation.
Further, maintains Williams, the exclusion of external supply sources was based on inaccurate information, which cited distances from external sources to the City of Springfield, rather than distances to the border of this large BEA. According to Williams, Phillips supplies this BEA market from both Jefferson City and Kansas City. Williams contends that the Commission eliminated both sources, although Kansas City is twice as far as Jefferson City from the market. Williams argues that inclusion or exclusion of external sources should be based on observed trucking activity, such as the survey and interview evidence in the record.
Williams explains that the HHI used by staff discounts the external sources by recognizing that these sources are capable of serving only part of a BEA market. However, Williams argues that even these external supply sources are capable of competing with an internal source. Williams notes that Phillips at Kansas City is capable of competing in slightly more than one-half of the market, while Shell at Wood River, Illinois, competes in less than four percent of the market.
Williams also argues that the staff's recalculated HHI is conservative because it excludes sources that logically would supply the BEA but were not documented by the trucker interviews or the gas station manager surveys. Williams lists six external sources that logically could serve the Springfield BEA but were excluded because there was no documentation on the record. 53/ Williams concludes that these additional supply sources require an HHI even lower than the staff's calculation.
We will grant rehearing with respect to the Springfield BEA. Our further examination of the record indicates that the Springfield BEA could support the profitable construction of terminals for the ARCO and Explorer transit pipelines, although not for the KCPS line. 54/ We will, therefore, include these two internal sources in our recalculation of the HHI for this market.
We reach a different conclusion regarding the viability of external sources. We have examined carefully the documentation of delivery costs provided by Williams for this BEA, which
53/ Williams lists Chase at Valley Center, Kansas; Texaco at El Dorado, Kansas; Derby at Wichita, Kansas; Total at Arkansas City, Kansas; ARCO at Kansas City, Kansas; and Sinclair at Tulsa, Oklahoma.
54/ See Ex. 627.
Docket No. IS90-21-003, et al. - 22 -
demonstrates that delivery costs for Phillips and Shell far exceed those of Williams. 55/ Phillips' delivery costs into this BEA through Jefferson City are nearly double those of Williams. Shell's terminal which, according to Williams, is 50 miles further from the BEA, would necessarily be even more costly, although Williams provides no costs for this source. Therefore, on the basis of this evidence, we are unable to conclude that deliveries by Phillips and Shell are economically viable competitive sources for this BEA.
Truck deliveries from the Sun refinery in Tulsa are questionable; it is not clear that such deliveries would occur because both Conoco and Williams offer pipeline service from Tulsa to a number of terminals in the Springfield BEA, at delivered costs well below the cost of trucking. 56/ If Williams were to raise its rates, the Sun alternative could become more attractive. However, because Williams assigns only a five percent market share to Sun, including or excluding it barely changes the market concentration.
We will not include the other six potential suppliers cited by Williams. As Williams itself admits, these sources are not documented in the record and are included by Williams for the first time on rehearing.
With the inclusion of the ARCO and Explorer pipelines as internal sources, we can find that Williams lacks significant market power in this market. Even with the continued exclusion of Phillips and Shell as external sources, customers in this BEA now have a choice of taking deliveries from five alternative suppliers in addition to Williams. This is reflected in the recalculated HHI of 1800 for the Springfield BEA. Therefore, we grant rehearing for this market.
B. Kansas City
For the Kansas City BEA, the Commission recalculated the HHI to eliminate sources more than 100 miles from the BEA, finding that the record did not show them to be competitive with Williams. The recalculated HHI of more than 2500, in addition to Williams' delivery-based market share of 63 percent and the lack of offsetting circumstances, caused the Commission to find that Williams had failed to establish that it lacks market power in the Kansas City BEA. 57/
55/ See Ex. 311.
56/ Id.
57/ 68 FERC 61,136 at 61,680.
Docket No. IS90-21-003, et al. - 23 -
Williams asserts that the Commission erred by relying on an inaccurate high delivery-based market share and by improperly excluding external supply sources. According to Williams, the market share cited by the Commission actually represents the combination of Williams' and Phillips' pipeline deliveries to this market because the two companies share breakout facilities. Williams also states that the shippers have exaggerated the distances from the external supply sources to the BEA border. Williams urges the Commission to adopt the original staff HHI of 2340 because this calculation appropriately recognizes the effect on this market, albeit small, of the external sources located more than 100 miles from the BEA border.
Williams maintains that it included in the HHI calculations only external sources that had been identified and documented as serving the BEA by the gas station manager surveys or trucker interviews. However, in the case of the Kansas City BEA, Williams contends on rehearing that there are four other external supply sources that could be included in the HHI calculation, thereby lowering the HHI. 58/
On review of the record, we find that some of these external sources may indeed deliver into this BEA at competitive prices. Truck deliveries from two external sources -- the Farmland refinery in Coffeyville, Kansas, and the Conoco terminal at Mt. Vernon, Missouri -- were improperly excluded from our calculation for this market. Although both sources face some competition from pipeline deliveries, they are still economically feasible. Product from both Coffeyville and Mt. Vernon can serve the border counties of the BEA in viable competition with product delivered at Kansas City and trucked back to the border counties. On this basis, we will include trucking from the Farmland refinery and the Mt. Vernon terminal as separate and competitive alternatives to Williams' pipeline in this BEA. 59/
However, the record does not contain capacity figures for the alleged external supply sources at Salina, Phillipsburg, El Dorado, and Council Bluffs that would allow us to examine their potential as competitors.
For this BEA, we also agree that the delivery data are somewhat misleading. Because Williams shares breakout facilities with Phillips in this BEA, the data on market share include deliveries for both companies. With this correction, Williams'
58/ These sources are Kaneb at Salina, Kansas; NCRA at Phillipsburg, Kansas; Texaco at El Dorado, Kansas; and NCRA at Council Bluffs, Iowa.
59/ See Ex. 315.
Docket No. IS90-21-003, et al. - 24 -
market share falls almost by half, causing us considerably less concern about Williams' market power.
On review, we find that six other suppliers in addition to Williams can serve this BEA competitively. Therefore, our HHI calculation, which now includes Amoco, ARCO, KCPS/Texaco, Phillips, Williams, Farmland, and Conoco (at Mt. Vernon only), is 2400. This figure, along with a corrected market share for Williams of 36 percent, causes us to find that Williams has successfully demonstrated that it lacks market power in this BEA. Accordingly, rehearing is granted.
C. Lincoln
The Commission concluded that Williams had failed to prove that it lacks market power in the Lincoln BEA. In this market, the Commission found Williams' delivery-based market share to be 65 percent. Eliminating the NCRA pipeline, which does not have a terminal in the BEA, produced a recalculated HHI of over 3000. The Commission also found no evidence of any factors offsetting the high market share and HHI numbers. 60/
Williams argues that the Commission erred in excluding the NCRA and Heartland pipelines in its HHI recalculation and in relying too heavily on Williams' pre-Heartland delivery-based market share. Williams cites the evidence presented by staff's Witness Alger that NCRA would find it profitable to build a terminal because the net present value of the incremental profits that would be generated would exceed the cost of constructing the terminal. Williams further states that the Heartland pipeline, with a rated capacity of 50,000 barrels per day, began operation in September 1990 and was fully operational throughout 1992.
On review, we find that the evidence demonstrates that, even though NCRA presently has no terminal in the Lincoln BEA, it does have a terminal at Council Bluffs, some 60 miles from the City of Lincoln, that can deliver product competitively into the Lincoln area. 61/ Given this fact, the question of a potential internal NCRA terminal is moot. We will include the Council Bluffs terminal in the HHI calculation. However, our reading of Witness Alger's testimony concerning the Heartland terminal at Lincoln differs from that of Williams. Witness Alger merely agreed that if Williams' assumptions about the economics of the potential NCRA terminal were accepted, then an NCRA terminal would be profitable; however, he did not accept Williams'
60/ 68 FERC 61,136 at 61,680.
61/ See Ex. 327.
Docket No. IS90-21-003, et al. - 25 -
assumptions. 62/ We believe that his assumed price increase of $0.10 over pre-1990 rates is more realistic than Williams' assumed price increase of 15 percent over pre-1990 rates.
Although Williams presented evidence that Heartland was under construction, we excluded it from our analysis of the market as an uncertainty. However, the Heartland terminal at Lincoln became fully operational in 1992, in which year Williams' deliveries fell by a significant percentage. While the record does not contain any evidence of costs for this new pipeline, it is in fact an internal source in this BEA, with an effective capacity equal to that of Williams or Kaneb. With the inclusion of Heartland and the NCRA Council Bluffs terminal, our recalculated HHI is 1542. Therefore, based on our determination that Williams lacks significant market power in this market, we grant rehearing.
D. Quincy
The Commission examined the Quincy BEA and reversed the ALJ's decision, based on Williams' 74 percent delivery-based market share and the recalculated HHI of 6559. 63/ The Commission found that Williams has the only pipeline terminal in the BEA and that evidence demonstrated that it would not be profitable for Amoco, which has a proprietary pipeline traversing the BEA, to build a terminal there. 64/
Williams argues that, contrary to the Commission's conclusion, staff Witness Alger determined that it would be profitable for Amoco to build a terminal in the Quincy BEA. Williams also states that, in evaluating external supply sources in this BEA, it included only external sources that had been identified and documented as serving the BEA by the gas station manager surveys or trucker interviews. However, Williams contends that there are three other external supply sources that could have been included. 65/
Next, Williams contends that, contrary to its findings in Buckeye, the Commission erred in not recognizing the strong competitive effects of the barge facilities that are located virtually in the center of the Quincy BEA market at Canton and 62/ Tr. 42/7185.
63/ The ALJ found the HHI to be 2026. 58 FERC 63,004 at 65,023.
64/ 68 FERC 61,136 at 61,685.
65/ Williams lists Conoco at St. Charles, Missouri; ARCO at Ft. Madison, Iowa; and Shell at St. Louis, Missouri.
Docket No. IS90-21-003, et al. - 26 -
LaGrange, Missouri. 66/ Williams also maintains that its effective capacity-based market share in Quincy is 25 percent, demonstrating that it cannot exercise market power in this market.
On rehearing, we find that Williams is correct in its contention regarding the possible construction of an Amoco terminal. Using his own conservative assumptions, staff Witness Alger found that it would be profitable to construct this terminal, and that its inclusion in the HHI calculation might alter his finding on competition in this BEA. 67/ We have accepted Witness Alger's assessment of the potential profitability of such terminals in other markets and will do so here as well by including the potential Amoco terminal in the market and the HHI calculation.
With respect to external sources that might serve as alternatives in this BEA, the trucking and delivery costs presented in the record lead us to reaffirm our exclusion of Phillips as an economically viable alternative in the BEA. The trucking costs alone from Phillips' terminals are almost equal to Williams' total delivered costs in the Quincy market. By contrast, these cost data indicate that ARCO, while not a least- cost supplier, could serve this BEA in competition with Williams. 68/
The external barge sources cited by Williams, which are St. Louis and Gulf Coast suppliers, also must be eliminated. If they serve the BEA through the internal barge terminals at Canton and LaGrange, they should not be counted twice as sources. If supplies are offloaded from barges at St. Louis and trucked to Quincy, the 79 mile trip to the border of the BEA would be uncompetitive with more direct barge deliveries because barge transport costs are approximately one to two mils per mile as opposed to approximately one cent per mile for trucks.
We also recognize that there are year-round barge facilities within this BEA at LaGrange and Canton, Missouri, which together are capable of providing more than 10 times the BEA's consumption of product. 69/ If these barge facilities were included at an effective capacity to serve the entire BEA, the HHI would be
66/ Citing 50 FERC 63,011 at 65,055-56; 53 FERC 61,473 at 62,669 (199O).
67/ See Ex. 627.
68/ See Ex. 318.
69/ BEA Appendix to the Opening Post-Hearing Brief of Williams (BEA Appendix), n.30.
Docket No. IS90-21-003, et al. - 27 -
2700. According to the evidence provided by Williams, barge deliveries into this BEA account for some 28 percent of total deliveries. 70/ However, the HHI calculation relied on by the ALJ in making his findings credits these facilities with a very low capacity. Like the ALJ, we merely relied on Williams' numbers in reaching our conclusion in Opinion No. 391.
We will recalculate the HHI to include Amoco on the basis that a terminal could be constructed profitably in this BEA. We will exclude Phillips because we find that it is not economically viable, and we will exclude the external barge sources to avoid double counting, as explained above. The recalculated HHI is 3100, almost half of our previous HHI, but not enough by itself to change our finding about the concentration of this market. However, it is commonly viewed that the existence of waterborne traffic, coupled with expandable capacity for waterborne deliveries, makes an oil market competitive. 71/ The staff in the past has suggested a more conservative approach, holding that expandable waterborne capacity, coupled with waterborne deliveries that account for at least 10 percent of total deliveries into a market, create a presumption of competition in that market. 72/ We will adopt this more conservative approach. Accordingly, because the conditions in the Quincy market satisfy this presumption, we find that Williams does not have significant market power in this market. Despite the seemingly high HHI for this BEA, we will grant rehearing.
E. Des Moines
The Commission found that Williams has a 78 percent delivery-based market share in the Des Moines BEA. Eliminating two external sources, barge terminals at Dubuque and Bettendorf, Iowa, the Commission's recalculated HHI was over 2500. In concluding that Williams had failed to establish that it lacks market power in this BEA, the Commission explained that Williams had not offered other considerations to offset the two high numbers. 73/ 70/ See Ex. 211 at 3.
71/ See, e.g., John A. Hansen, U.S. Oil Markets 40, 49, 64 (1983); Department of Justice, Oil Pipeline Study at 36, 64 n.75.
72/ Commission Staff Proposal for Revisions to Oil Pipeline Regulations Pursuant to the Energy Policy Act of 1992 at 46- 47, 54 (March 1993). See also Oil Pipeline Study at xii, 17, 36 (indicating the importance of port facilities in determining the competitiveness of a market).
73/ 68 FERC 61,136 at 61,680.
Docket No. IS90-21-003, et al. - 28 -
Williams emphasizes that Heartland began operating in this BEA in September of 1990, causing Williams to cut its rates from Central Oklahoma origins to Des Moines in November. Despite this action, Williams states that it lost substantial volumes to Heartland. 74/ Williams argues that these losses verify the strong pro-competitive effect of Heartland and also Williams' lack of market power.
Williams maintains that the Bettendorf and Dubuque barge facilities were documented to be serving the Des Moines BEA through interviews with truckers and surveys of gas station managers, as validated statistically by Witnesses Bollen and Schink. Although the Commission found that the barge facilities at Bettendorf and Dubuque to be approximately 160 miles distant, Williams states that these are the approximate distances to the central city of the BEA, not to the BEA border. Williams also points out that neither of these external barge sources serves the entire BEA market.
Williams claims that it did not include external sources unless they were specifically identified by the truck driver interviews or through the telephone surveys. Thus, Williams urges consideration of two other documented sources for this BEA: NCRA's pipeline terminal in Council Bluffs and the Mississippi River barge terminals in Canton, Missouri. Williams points out that trucks from Amoco's terminal in Council Bluffs and from Williams' terminal in Omaha serve a large part of the Des Moines BEA; therefore, reasons Williams, it is logical that trucks from NCRA's terminal in Council Bluffs should have the same reach within the Des Moines BEA as do trucks from Amoco's Council Bluffs and Williams' Omaha terminals. Williams calculates that NCRA competes in 41 percent of this market. Noting that trucks from the barge terminal in LaGrange, Missouri, serve the Des Moines BEA, Williams assumes that the effective supply capacity for the barge docks at Canton is at least as great as the effective supply capacity of the barge docks at LaGrange, given the fact that the Canton docks are approximately seven miles closer to the Des Moines BEA border than the LaGrange docks.
We reject Williams' argument that we failed to give adequate consideration to the effects of Heartland's entry into the market. Our calculations did take full account of Heartland, viewing it as equal to Williams in its ability to serve the market. Because our HHI in this case is a simple capacity-based measure, the price and volume changes experienced by Williams are irrelevant to, but implied by our calculations.
74/ Williams claims that Heartland's FERC Form No. 6 submitted December 31, 1992, confirms that these barrels, along with volumes lost by Williams at Lincoln, were diverted to Heartland.
Docket No. IS90-21-003, et al. - 29 -
As to external sources, Williams correctly notes that NCRA can serve this BEA competitively for a small share of the market. 75/ By contrast, Williams' cost evidence shows that ARCO deliveries from Ft. Madison exceed Williams' own delivery costs by almost $1.00 per barrel. 76/ We will, therefore,, include NCRA but not ARCO in our recalculated HHI for this market. We will also include the barges at LaGrange and Canton as minor sources.
We disagree with Williams' contention that the barge terminals at Dubuque and Bettendorf can serve 72 percent and 81 percent of the BEA market respectively in competition with Williams. We conclude that we properly excluded trucked deliveries from these barge terminals as regularly viable sources. The trucking and delivery costs presented in the record for this BEA show a disparity of $1.00 to $2.00 per barrel between Williams' pipeline rates and the parallel trucking costs, 77/ making it unlikely that anyone would choose trucking from these terminals.
In contrast, the record shows that pipeline deliveries from these barge terminals into the BEA are cost-effective. However, our reexamination of the record reveals that three of the four terminals at Bettendorf and one of the two at Dubuque are connected to Williams. 78/ Access to the BEA from these terminals is thus primarily through Williams unless product is trucked. If we include deliveries made into the Des Moines BEA by pipeline, then we must recalculate the HHI to consolidate some of this capacity into Williams' share of the market. We can ascribe 3/4 of the capacity at Bettendorf and one-half the capacity at Dubuque to Williams.
With this addition to Williams' share of capacity in the Des Moines market, and reflecting the other changes discussed in this section we have a recalculated HHI of 2897. Shippers in this market have a choice of only two other pipelines in the BEA, with limited service alternatives from external sources. Accordingly, because Williams exercises significant market power in this market, rehearing is denied with respect to the Des Moines BEA.
75/ See Ex. 326.
76/ Id.
77/ Id.
78/ Tr. 23/3017-19, 41/6971-72.
Docket No. IS90-21-003, et al. - 30 -
F. Omaha
On review of the ALJ's determination, the Commission affirmed that Williams exercised significant market power in the Omaha BEA. The Commission based its decision on Williams' delivery-based market share of 46 percent and an HHI of 2786. The Commission rejected Williams' arguments concerning competition from the Heartland pipeline. Finally, the Commission rejected Williams' contention that the reduction in its rates to Omaha was conclusive evidence of competition, emphasizing that a rate increase or decrease per se does not prove or disprove market power. 79/
Williams states that both the ALJ and the Commission failed to recognize the presence of a Heartland terminal in Lincoln, Nebraska, which also serves this BEA. Williams points out that the Heartland Lincoln terminal is located just 25 miles from the Omaha BEA border and less than 50 miles from the City of Omaha. Based on documented truck movements from Williams' terminals in Lincoln, Williams argues that trucks from the Heartland terminal at Lincoln logically could serve a large portion of the Omaha BEA.
Williams also points out that this terminal, owned and operated by Conoco, was served exclusively by Williams prior to the construction of the Heartland pipeline. Subsequently, asserts Williams, shipments to Conoco's Lincoln terminal were largely rerouted through the new pipeline, which sharply reduced Williams' shipments to the Lincoln BEA.
Williams also argues that the Omaha market can be served from the Heartland terminal at Des Moines, and that a strong potential exists for construction of a new terminal or modification of an existing facility served by the Heartland pipeline within the Omaha BEA. 80/ Finally, Williams states that it delivers substantial volumes of product in Omaha for ultimate use at Offutt Air Force Base. Williams argues that this business is constantly at risk because the DOD puts the fuel contract out for competitive bid every year.
On rehearing, the Commission finds that Williams is partly correct. The trucking cost information provided by Williams for this BEA confirms that the Heartland terminal in Lincoln, located 25 miles from the BEA border and 58 miles from Omaha, could be 79/ 68 FERC 61,136 at 61,684.
80/ Williams points out that the Heartland Pipeline, which is jointly owned by Conoco and Enron, runs through the southern portion of the BEA and currently makes LPG deliveries to an Enron terminal in Plattsmouth, Nebraska.
Docket No. IS90-21-003, et al. - 31 -
considered a viable supply source for this BEA. 81/ Although the record contains no evidence of costs for this terminal, Williams notes that its share of deliveries in this BEA has dropped substantially since Heartland began operations, which reduces Williams' delivery-based market share.
By contrast, we do not find that Heartland's terminal in Des Moines, some 130 miles from Omaha and 68 miles from the BEA border, should be included as an economically viable source for this BEA. According to evidence presented by Williams, the additional trucking charges from Des Moines would make deliveries from this terminal uncompetitive in this BEA. Trucking costs to the border would be approximately 70 cents and approximately $1.20 to Omaha, exclusive of pipeline and terminal fees. 82/
There is insufficient evidence to permit us to rule on the inclusion of the potential Heartland terminal near Omaha; in fact, Williams does not include any capacity figure for it. We can acknowledge the possibility of its construction, but will not include it in our HHI calculation.
Although Williams argues that its market share in this BEA is subject to serious competition because a large volume is subject to annual bids for DOD contracts, we conclude that Williams' ability to compete for DOD sales is no different than its ability to compete for other business. Such decisions are made on the basis of Williams' transportation costs, as those affect the delivered price.
Under our revised market analysis, we find that shippers in this BEA have a choice of five internal sources, including barges, as well as the Heartland terminal at Lincoln. The recalculated HHI is 2300. Accordingly, we find that Williams lacks significant market power in this market, and grant rehearing.
G. Eau Claire
In finding that Williams failed to prove that it lacks market power in the Eau Claire BEA, the Commission relied on Williams' 59 percent delivery-based market share and recalculated the HHI, eliminating the Koch pipeline, which lacks a terminal, and the Badger terminal, which is at too great a distance to serve as effective competition for Williams. The recalculated
81/ See Ex. 316.
82/ Id.
Docket No. IS90-21-003, et al. - 32 -
HHI was over 3000, which the Commission found was not offset by other factors. 83/
Williams claims that the Commission improperly excluded Koch as a supplier for this market because Koch does not have an operating terminal on its transit pipeline. However, contends Williams, Koch serves the entire Eau Claire BEA from its Minneapolis-area refinery. Williams also contends that the Commission erroneously excluded the Badger pipeline terminal in McFarland, Wisconsin, believing the truck haul to be too lengthy. However, according to Williams, the uncontroverted gas station manager surveys and trucker interviews substantiated that this was a regular and ongoing movement.
Further, Williams states that only external sources from which product movements could be documented were included in the HHI. However, in the case of the Eau Claire BEA, Williams states that there is one external source that logically would be capable of serving the entire market, namely Amoco's pipeline terminal located at Minneapolis, only 67 miles from the border of the Eau Claire BEA. According to Williams, it has been documented that trucks serve the entire Eau Claire BEA from truck racks at Ashland's and Koch's refineries and Williams' terminal, all of which are located in Minneapolis.
In this market, Williams argues that delivery-based data overstate Williams' market share and are unreliable. Specifically, Williams points out that it does not operate a terminal within the Eau Claire BEA, but instead delivers to three third-party terminals operated by shippers. Given this fact, Williams states that it has no information as to where the product from these terminals is delivered; therefore, all of the product shipped to these terminals is counted as being delivered to the county where the terminals are located. However, Williams reasons, trucks from these terminals clearly deliver product to locations throughout the Eau Claire BEA and into adjacent BEAs.
On rehearing, the Commission acknowledges the weakness of the delivery-based data in this market. The Commission also agrees that its HHI calculation was in error in this BEA. The record demonstrates that Koch could serve the BEA competitively from its St. Paul refinery, 74 miles from Eau Claire itself and 57 miles from the edge of the BEA. In the initial calculation, following the method established by staff's Witness Alger, this external capacity was combined with Koch's internal capacity. 84/ When Koch was excluded as an internal source, we inadvertently also dropped the external refinery capacity
83/ 68 FERC 61,136 at 61,679.
84/ See Ex. 619 at 60-61.
Docket No. IS90-21-003, et al. - 33 -
properly attributable to Koch. Therefore, the total exclusion of Koch from the HHI calculation was an error.
Williams also urges the inclusion of several Koch pipeline terminals as sources for this BEA. A review of the trucking and delivery costs in the record shows that, with delivery costs nearly twice that of Williams, none of these is a viable alternative source. We only include the refinery. Further, on the basis of these trucking and delivery costs, we continue to find that Badger would not be a competitive alternative to Williams in this BEA. 85/
Adding Koch's St. Paul refinery to our previous analysis, but not the Badger terminal, reduces Williams' market share and results in an HHI of 2500. Williams is the only internal pipeline source for this BEA, but the market is served by at least three refineries at prices that undercut Williams. 86/ Given the significant potential for price discipline from these sources, we find that Williams lacks significant market power in this market, and we grant rehearing.
H. Fargo
The Commission recognized that the Fargo BEA covers a large area and found that the external sources cited by Williams are too distant to serve as viable alternative sources. In this BEA, Williams has a delivery-based market share of 51 percent, and the Commission's recalculated HHI is in excess of 3000. Given these facts, the Commission found that Williams had failed to prove the market to be workably competitive. 87/
Williams states that Cenex was constructing a pipeline into Fargo during the evidentiary hearing. According to Williams, this eight-inch pipeline, which transports product to Fargo from the Cenex refinery in Montana, has a design capacity exceeding the consumption of petroleum products in the entire Fargo BEA. Williams points out that the Cenex pipeline is connected to Williams' terminal at Fargo in the same way that Phillips pipeline is connected to Williams' terminal in Kansas City, allowing Cenex to deliver product to the Fargo BEA through Williams' tank truck loading racks. Further, states Williams, Cenex has chosen to use Williams' truck racks instead of constructing its own only because Williams offered Cenex an attractive rate for providing these racking services.
85/ See Ex. 333.
86/ The refineries are owned by Ashland, Koch, and Murphy. See BEA Appendix at 4-5.
87/ 68 FERC 61,136 at 61,681.
Docket No. IS90-21-003, et al. - 34 -
Williams argues that the Commission improperly excluded all external supply sources, which, Williams maintains, have been documented as occurring on a regular and ongoing basis by the gasoline station manager surveys and the truck driver interviews and statistically validated by Witnesses Bollen and Schink. According to Williams, the shipper-intervenors' own trucking distribution data are consistent with the trucking distances determined in the driver interviews. Further, states Williams, trucking distances can be expected to be longer in remote rural areas such as Fargo. Finally, Williams argues that its delivery data do not provide a reliable indication of the location of its deliveries in this market.
We find the evidence demonstrates that Cenex can deliver product directly into Fargo; therefore, it should be included in the HHI calculation as an internal source. 88/ Adding Cenex to this market will reduce Williams' delivery-based market share.
We also have reviewed the evidence relating to the three additional external sources that Williams claims should be included in the HHI calculation for this market. These sources include the Interprovincial facility at Winnipeg (168 miles from the BEA), and the Koch and Ashland refineries at Minneapolis (175 miles from the BEA). According to evidence on trucking and delivery costs submitted by Williams for this BEA, we find that the cost of deliveries to Fargo from these sources exceed the cost of deliveries on Williams' system by 40 to 112 percent, and thus, would not offer serious price discipline for Williams' rate increases. By contrast, product from Amoco's refinery in Mandan can be delivered to Fargo by pipeline or by truck (a distance of more than 100 miles) at a cost that far undercuts Williams. 89/
The HHI recalculated to include Cenex, but not Interprovincial, Koch, or Ashland, is 2500. The following factors suggest that this market is competitive: (1) shippers in this BEA have a choice of service from four pipelines, even without external sources; 90/ (2) Williams offers proportional discounts in the western half of the Fargo BEA in order to meet the competition it faces from Kaneb and Amoco; 91/ and (3)
88/ Tr. 26/ 3519-21.
89/ See Ex. 330.
90/ The Fargo BEA is served by Amoco and Kaneb, in addition to Williams and Cenex. Request of Williams for Rehearing and Clarification of Opinion No. 391 at 73. Citing Staff BEA Appendix at 31; Ex. 217 at 31.
91/ See Ex. 15.
Docket No. IS90-21-003, et al. - 35 -
according to the cost evidence supplied by Williams, its rates into this BEA are considerably undercut by Amoco and by trucking deliveries from Amoco's refinery in Mandan. 92/ Accordingly, we find that Williams lacks significant market power in this market, and grant rehearing with respect to this BEA.
I. Grand Forks
The Grand Forks BEA is also large, and Williams' delivery- based market share is 56 percent. Because external sources were not shown to offer effective competition, the Commission's recalculated HHI was greater than 3000, and the Commission found that the Grand Forks BEA is not a workably competitive market. 93/
According to Williams, the Cenex pipeline into Fargo is located only 64 miles from the BEA's southern border and 77 miles south of the City of Grand Forks. Williams argues that the record contains ample evidence of truck deliveries from Amoco's Fargo terminal and the Cenex terminal at Minot into the Grand Forks BEA. Further, Williams claims that truck driver interviews and gas station manager telephone surveys confirmed specific towns within the BEA being served and the extent of the service.
Williams lists seven other external sources, located at distances ranging from 68 to 189 miles to the BEA border. 94/ Williams claims that these sources can serve varying percentages of the market -- from four percent to 22 percent. 95/ According to Williams, this demonstrates that external sources need not reach the center city of a BEA to be competitive with Williams.
Williams also claims that it is under constant pressure to charge competitive rates to Grand Forks, largely because of the highly competitive nature of the DOD business in the BEA. Williams states that, in Grand Forks, off-line deliveries to the DOD represented nearly one-quarter of total shipments to the BEA
92/ See Ex. 330.
93/ 68 FERC 61,136 at 61,681.
94/ Request of Williams for Rehearing and Clarification of Opinion No. 391 at 77. They are Winnipeg at Winnipeg, Manitoba; Kaneb at Jamestown, North Dakota; Amoco at Fargo, North Dakota; Mandan, North Dakota; and Sauk Center, Minnesota; Ashland and Koch at Minneapolis, Minnesota; Murphy at Superior, Wisconsin; and Cenex at Minot, North Dakota.
95/ Id. at 78.
Docket No. IS90-21-003, et al. - 36 -
during 1990, but that it has lost a substantial portion of this business.
We agree with Williams that the now-completed Cenex terminal in Fargo should be included in our HHI calculation. However, as for external sources, the trucking and delivery costs in the record for this BEA indicate that only Amoco, in addition to Cenex, can make competitively viable deliveries into this BEA. According to the delivery cost data provided by Williams, all other options are 60 to 120 percent more costly than Williams and so cannot be expected to offer sufficient pricing discipline. 96/
Again Williams has argued that it faces serious competition because a large portion of its sales is subject to annual contract awards by the DOD. As stated above in the discussion of the Omaha BEA, we consider Williams' ability to compete for the DOD business to be no different than its ability to compete for other business in its markets. Therefore, as with the Omaha BEA, we will not consider the DOD business as a separate factor in our market power analysis of the Grand Forks BEA.
Our recalculated HHI for this BEA is 3500, reflecting market shares for Williams, Amoco, and Cenex. This figure remains too high to permit a finding that Williams lacks significant market power in this market. Accordingly, rehearing is denied with respect to the Grand Forks BEA.
J. Columbia
The Commission determined the HHI for this BEA to be 1738 and acknowledged that Williams' market share is 49 percent. The Commission accepted the testimony of Witness Alger that it would be economical for Amoco to construct a terminal in the Columbia BEA, and the Commission also recognized that external sources might economically serve particular areas of the BEA. Therefore, the Commission affirmed the ALJ's finding that the Columbia BEA is a workably competitive market. 97/
Emphasizing Williams' 49 percent market share, Texaco states that Williams has demonstrated its ability to raise rates by 44 percent. Further, argues Texaco, if the Commission continues to hold that BEAs such as Columbia are competitive, then Williams should be estopped from charging rates above the 15 percent standard for which Williams argued. In the Columbia BEA, Texaco states that the rate first went from 87 cents to 110.40 cents, an increase of 26 percent. Texaco also claims that subsequent
96/ See Ex. 335.
97/ 68 FERC 61,136 at 61,678.
Docket No. IS90-21-003, et al. - 37 -
increases to 119.10 cents and then to 124.90 cents are near the 15 percent mark.
On review of the record, we find that the Columbia BEA is served by three pipelines having terminals within the BEA. 98/ According to Witness Alger, a fourth pipeline, Amoco, could build a terminal profitably on its transit pipeline in this BEA. 99/ Moreover, the BEA is served competitively by the Conoco terminal just outside the BEA boundary and by two barge terminals within 30 miles of the BEA. 100/ We continue to find that Williams lacks significant market power in the Columbia BEA, with an HHI of 1800.
Texaco errs in suggesting that the Williams' 49 percent share of deliveries in this market should be dispositive of a market power finding. As discussed at the outset, market delivery data in this proceeding are incomplete and inconsistent, and cannot be used as a primary indicator of market power. We have used delivery data primarily as a matter for further scrutiny when the more reliable capacity-based HHI was sufficiently high to warrant further examination of the market. That is not the case with Columbia.
Texaco also errs in its argument that an increase in Williams' rates -- even a significant one -- is indicative of market power. The existence of competition does not automatically imply an inability to raise rates or even that low rates should prevail. The existence of competition means that price increases above efficient, market-driven equilibrium prices will not be sustainable for any length of time. 101/
Texaco has cited evidence of price increases in this BEA, but has offered no evidence that these price increases are undue, incongruent with efficient competitive pricing, or indicative of the exercise of market power. Texaco has not discussed the effect of Williams' 1985 price freeze on real rates, has not compared Williams' rates with those of competing sources in this market, nor has it compared the rates paid by Texaco with those paid by other shippers on Williams to this BEA. Absent any such showings, and given the wide range of alternatives to Williams in this BEA, we find no basis for granting rehearing with respect to the Columbia BEA.
98/ See Ex. 619 at 19.
99/ See Ex. 627.
100/ See Ex. 314.
101/ Merger Guidelines at 4.
Docket No. IS90-21-003, et al. - 38 -
III. Discrimination Claims
The intervenors raised a variety of discrimination claims. 102/ The ALJ found that each alleged form of discrimination was intended to meet market conditions and was not an abuse of Williams' market power; however, he reserved for evaluation in Phase II the questions of whether one service cross-subsidizes another and whether, in certain instances, differences in costs also could be used to justify the alleged discrimination. 103/ On review, the Commission affirmed all of the ALJ's determinations, but stated that the ALJ's findings did not establish that the rate differentials are lawful. The Commission determined that it would review Williams' rate differentials in Phase II using the cost information required in that phase to determine whether the proposed rates are just and reasonable or unjustly discriminatory. 104/
On rehearing, Williams and AOPL argue that all of the discrimination claims have been resolved in Phase I and should not be relitigated in Phase II. Williams emphasizes that rate disparities can be sustained solely on the ground that they are reasonably related to differing competitive conditions, 105/ a defense that the ALJ accepted. However, Williams concedes that the issue of whether one service cross-subsidizes another is a legitimate question for Phase II.
102/ The discrimination issues addressed by the ALJ are as follows: (1) the disequalization of certain "Group 3" rates, resulting in relatively higher rates to certain northern destinations from relatively more distant origins in Oklahoma than from Kansas; (2) the approximate equalization on a per-mile basis of rates from northern and southern origins on Williams' system; (3) alleged discrimination against rural destinations in relation to urban destinations; (4) discounts for relatively higher volume shippers (or shipper groups); (5) "proportional rate" discounts -- i.e., discounts for movements to certain destinations on Williams' system which then move by truck into certain designated counties; and (6) Williams' operation on an "open stock" (or "fungible") basis whereby Williams is able to meet shipper demand at a particular destination without moving that barrel from a particular origin.
103/ 58 FERC 63,004 at 65,025-28.
104/ 68 FERC 61,136 at 61,688.
105/ Citing Associated Gas Distributors v. FERC, 825 F.2d 981, 1011 (D.C. Cir. 1987) (AGD), cert. denied, 485 U.S. 1006 (1988).
Docket No. IS90-21-003, et al. - 39 -
The ALJ's decision is replete with citations to evidence indicating that Williams relied on its extensive calculations of competitors' and shippers' delivered prices into various Williams markets to justify its newly-instituted rate differentials. According to the ALJ, Williams then based its competitive discounts on these determinations of what the market could bear. Williams did not rely primarily on its own cost-of-service/cost allocation data, and only briefly mentioned cost differentials between routes where rate differentials were established or where discounts are provided. 106/
The ALJ noted that, absent full cost evidence, any reliance on costs to support these discounts and rate differentials would have to await Phase II. 107/ However, he also concluded that this would not be necessary because Williams had justified its rate differentials on the basis of competitive exigencies. He correctly pointed out that any talk of cost justification is no longer of significance once the competitive need for rate differentials has been established. 108/ The Commission agrees.
Opinion No. 391 responded to claims that these discounts constitute undue discrimination by requiring the allegations to be examined in Phase II. The Commission is concerned that discounts to some shippers not entail cross-subsidies by others. Simply put, customers in noncompetitive markets should not be required to pay excessive rates in order to make up for any losses suffered in competitive markets. We recognize, however, that rate differentials -- even if not based on corresponding cost differences -- are not necessarily tantamount to cross- subsidies.
In fact, discounts to customers with other shipping alternatives are recognized as benefitting captive customers, so long as the non-captive customers contribute something to common costs. 109/ Therefore, competitive discounts do not constitute or require subsidies from captives, and can be found to be not unduly discriminatory in this Phase I rehearing. On the basis of the competitive justifications provided by Williams, the Commission finds that for Williams to charge such rate differentials is not unduly discriminatory. Accordingly, the
106/ See 58 FERC 63,004 at 65,024-28.
107/ Id. at 65,025, 65026, and 65,027.
108/ Id. at 65,025 and 65,026.
109/ Policy Statement Providing Guidance with Respect to the Designing of Rates, 47 FERC 61,295 at 62,053 (1989), citing AGD, 824 F.2d 981, 1011-11.
Docket No. IS90-21-003, et al. - 40 -
Commission clarifies its previous order on this point, and will not require re-examination of the issues of alleged discrimination on a cost basis in Phase II of this proceeding.
Nevertheless, questions of rate design, that is, how rates and rate differentials should be set in noncompetitive markets is properly the subject of Phase II. We also fully expect Phase II to explore the role of efficient and not-undue price discrimination in both competitive and captive markets. However, to insist that the issue of cross subsidies can only be fully explored in the context of cost-of-service or point-to-point cost allocations would be to misread our intent. These issues can also be considered, for example, by examining the cost and revenue contributions of relevant services or markets. The rights of the participants in Phase II to examine these important questions of cross subsidies, cost contributions, and discounts do not depend on the use of a single sanctioned method for determining justness and reasonableness.
IV. Whether the Commission Has Ruled on the Merits of Williams' Proposed Rate Standards
Following issuance of the ALJ's decision, Williams filed a motion with the Commission proposing a standard for adjudicating in Phase II the maximum reasonable level of rates in any market not found workably competitive in Phase I. Williams proposed that it make two showings in any rate filing: (1) that the overall earnings generated by the proposed rates do not exceed the revenue requirement permitted by Opinion No. 154-B and its progeny, 110/ and (2) that the total of its proposed rates do not exceed the stand-alone costs of its services in total. 111/ Williams also proposed a standard for adjudicating in Phase II the minimum reasonable level of rates in any market not found workably competitive in Phase I. Under Williams' proposal, the minimum rate standard for any given service would be short-run marginal 112/ or incremental costs. In proposing a one-year time horizon to measure short- term marginal costs, Williams asserted that such short-term costs
110/ Williams Pipe Line Co., 31 FERC 61,377 (Opinion No. 154- B), modified, 33 FERC 61,327 (Opinion No. 154-C) (1985); ARCO Pipe Line Co., 52 FERC 61,055 (Opinion No. 351), reh'g denied, 53 FERC 61,398 (Opinion No. 351-A) (1990).
111/ Williams claimed that the total stand-alone costs would equal a rate of return ceiling based on a current cost rate base for a hypothetical, optimally efficient pipeline designed to supplant the present Williams pipeline as a whole.
112/ Marginal cost is the cost of producing one more unit.
Docket No. IS90-21-003, et al. - 41 -
consisted only of variable fuel and power costs. The rate maximum for each service would be the stand-alone costs.
In Opinion No. 391, the Commission pointed out that the rate standards proposed by Williams would govern not only the establishment of base rates in this case, but also would establish how Williams' rates would be judged in future cases. The Commission explained that it had adopted a final rule instituting a simplified and generally applicable ratemaking methodology for oil pipelines. The Commission noted that, in the rulemaking, it had considered requests by Williams and AOPL to establish a stand-alone maximum rate standard for changing rates. However, the final rule rejected the stand-alone methodology as a primary vehicle for rate changes in favor of an indexing methodology. Therefore, because the indexing methodology adopted by the Commission is inconsistent with Williams' proposed stand- alone methodology, the Commission rejected Williams' proposed use of the stand-alone methodology to justify future rate changes. The Commission stated that the only question remaining is what base rates will be allowed for Williams in this case and will serve as the basis for future indexing. The Commission set for hearing in Phase II the method to be used for establishing base rates, but directed the ALJ and the parties, in developing such a method, to give particular attention to the allocation of costs between the competitive and noncompetitive markets to ensure that customers in the noncompetitive markets do not subsidize those in the competitive markets. 113/
Williams and AOPL seek clarification of Opinion No. 391 insofar as its denial of Williams' motion could be read to have ruled on the merits of any cost standards proposed by Williams in Phase I or to preclude Williams from continuing to advocate such standards in Phase II. Williams and AOPL ask the Commission to clarify that, in Phase II, Williams may continue to advocate the cost-based standards that it advocated in Phase I.
Both Williams and AOPL rely on Order No. 561, et al. 114/ They emphasize that Williams did not intend for the rate standards proposed in its motion or the similar standards proposed in the rulemaking by AOPL to apply in lieu of the indexed rate cap. Rather, states Williams, these standards were proposed in the present case to enable Williams to establish base rates to the extent necessary in Phase II. Indeed, states Williams, in Order No. 561-A, the Commission expressly (1)
113/ 68 FERC 61,136 at 61,695.
114/ Citing Revisions to Oil Pipeline Regulations Pursuant to the Energy Policy Act of 1992, order on reh'g, Order No. 561-A, 59 Fed. Reg. 40,243, (Aug. 8, 1994), III FERC Stats. & Regs. Preambles 31,000 (July 28, 1994).
Docket No. IS90-21-003, et al. - 42 -
recognized that cost-based standards are necessary as a supplement to the index in certain circumstances; (2) declined to decide the issue of whether fully-allocated costs should be employed for that purpose; and (3) promised that proponents of the stand-alone cost methodology or other costing methodologies will not be precluded from advocating such methodologies in individual cases. 115/
Despite this promise, continues Williams, Opinion No. 391 is confusing. While the Commission set for hearing in Phase II the method to be used for establishing base rates, the opinion admonishes the participants and the ALJ to give particular attention to the allocation of costs between competitive and noncompetitive markets to ensure that customers in the noncompetitive markets do not subsidize customers in the competitive markets. This admonition, asserts Williams, could be read to suggest that the cost standards proposed by Williams, which do not rely on cost allocations in order to prevent cross- subsidies, have been prejudged.
Thus, concludes Williams, rehearing should be granted for the purpose of clarifying that Williams may advocate any cost- based standards of its choosing in Phase II and that the Commission has made no rulings on the merits of any cost issue for Phase II. Further, Williams states that the evidence regarding cost standards submitted in Phase I is massive and was prepared at considerable cost. Although Williams believes that all participants should have the right to supplement that evidence in Phase II, Williams maintains that it would be wasteful if the participants were required to construct a Phase II record from scratch. Accordingly, Williams also asks the Commission to clarify that the participants may rely on any portion of the Phase I record in Phase II, subject to supplementation by any participant.
The Commission clarifies that its ruling in Opinion No. 391 was not intended to pre-judge the validity of Williams' proposed rate standards as a means of establishing base rates in Phase II of this proceeding. As both Williams and AOPL acknowledge, future rate increases will be subject primarily to the indexing process established in Order No. 561, et al. However, in Phase II of this proceeding, Williams is free to present any method it chooses for arriving at just and reasonable rates for the markets we have determined to be noncompetitive. As we stated above, the rights of the parties to address important questions relating to cross subsidies do not depend on the use of a single sanctioned method for determining justness and reasonableness of rates. Finally, we agree with Williams that it is efficient in terms of both time and costs to permit the parties to rely on data already
115/ Citing Order No. 561-A, slip op. at 47.
Docket No. IS90-21-003, et al. - 43 -
submitted in Phase I, supplemented as determined to be necessary by the ALJ and the parties.
V. Motion to Sever Limited Issue for Investigation and Decision
On October 31 and November 10, 1994, Williams filed proposed tariffs in Docket Nos. IS92-26-000, IS95-2-000, and IS95-7-000. In orders issued November 30 and December 9, 1994, the Commission's Oil Pipeline Board accepted and suspended the proposed rates, subject to refund, but stayed these consolidated proceedings pending rehearing of Opinion No. 391 and completion of the Phase II investigation in that earlier proceeding. 116/
Murphy Oil USA, Inc. (Murphy) owns and operates an oil refinery in Superior, Wisconsin. Murphy states that it is captive to Williams for shorthaul shipments from its refinery across the state line to the Duluth and Wrenshall, Minnesota terminals. Murphy asks the Commission to sever from the proceedings in Docket No. IS92-26-000, et al., the limited issue of whether Williams' shorthaul rates from Superior to Duluth and Wrenshall create unlawful discrimination against shorthaul intrastate shipments from Minnesota refineries to Williams' own Minneapolis terminal. Murphy states that, because the Phase II investigation in Docket No. IS90-21-000, et al., is now only in its preliminary stages, there is likely to be a lengthy delay in the resolution of this issue.
Williams filed an answer opposing Murphy's motion to sever. Williams states that Murphy has failed to make even a prima facie case that the rates identified are discriminatory or that it has in any way been injured as a result of the alleged discrimination. Williams also contends that the revenues involved do not warrant a separate investigation into the challenged rates. Finally, Williams states that Murphy failed timely and properly to protest the rate increases it now challenges.
The Commission denies Murphy's motion to sever. Since the Commission commenced the investigation in Docket No. IS90-21-000, it has routinely suspended subsequent Williams rate filings and consolidated them with Docket No. IS90-21-000, as occurred in this instance. As discussed in greater detail above, differences in rates do not necessarily establish discrimination. Williams correctly points out that a variety of non-mileage factors can influence a particular rate and justify a differential. The issue of just and reasonable rates for the Duluth BEA remains to
116/ Williams Pipe Line Co., 69 FERC 62,175; 69 FERC 62,206 (1994).
Docket No. IS90-21-003, et al. - 44 -
be resolved in Phase II of the proceedings in Docket No. IS90-21- 000, et al.
The Commission is satisfied that staying the proceedings in Docket No. IS92-26-000, et al., pending resolution of the earlier proceeding will promote administrative efficiency and that Murphy will suffer no injury. It appears that the volumes and revenues associated with Murphy's shipments to Duluth and Wrenshall are rather minimal and do not outweigh the potential cost to the parties and to the Commission that would result from separately investigating the rates Murphy challenges. In any event, both of the Oil Pipeline Board orders clearly require Williams to maintain accurate records of the revenues it receives under these tariff sheets so that refunds with interest can be made in the event the proposed rates are found to be unjustified.
The Commission orders:
(A) Rehearing and clarification are granted and denied as discussed in the body of this order.
(B) No further rate review is required for the following markets where Williams has established that it lacks market power: Springfield (MO), Kansas City, Lincoln, Quincy, Omaha, Eau Claire, Fargo, and Columbia.
(C) The ALJ is directed to continue with Phase II of this proceeding for the purpose of establishing base rates for the following markets where Williams has failed to establish that it lacks market power: Des Moines and Grand Forks. These markets are in addition to those listed in ordering paragraph (C) of Opinion No. 391 that were unchallenged on rehearing.
(D) Murphy's motion to sever is denied.
By the Commission.
( S E A L )
Lois D. Cashell, Secretary.