This case arises out of a tariff filing by El Paso Natural Gas Company in which it sought to recover from its customers 75 percent of the value of certain oil and gas bearing properties it had transferred to a gas producer as part of a settlement agreement to resolve a "take-or-pay" contract dispute. [321 U.S.App.D.C. 14] Relying solely on a discounted cash flow methodology, the Federal Energy Regulatory Commission ("FERC" or "Commission") determined the fair market value of the properties at the time of transfer to be $98.3 million. El Paso contends that the Commission erred in rejecting "comparable sales" and other market evidence in determining the value of the properties and thus underestimated their value. After careful consideration of the record, we conclude that the Commission acted within its discretion in rejecting El Paso's market evidence and in relying solely on a discounted cash flow methodology.

In the 1980's, natural gas pipelines incurred massive liabilities under long-term gas purchase contracts that obligated them "either to take or to pay" for gas that they had contracted to buy at prices substantially higher than those now prevailing in the marketplace. See Public Utils. Comm'n of Cal. v. FERC, 988 F.2d 154, 157 (D.C.Cir.1993). One way in which the Commission has attempted to deal with the problem of "take or pay" liability is through the "equitable sharing" mechanism established in Order No. 500, 52 Fed.Reg. 30,334 (Aug. 14, 1987), and its progeny, which permits an "open access pipeline" (one that is obligated, like a common carrier, to provide transportation service on a non-discriminatory basis) to recover from its customers up to 75 percent of the costs it has incurred in settling take-or-pay contract liabilities. See generally United Distribution Companies v. FERC, 88 F.3d 1105, 1123-25 & accompanying footnotes (D.C.Cir.1996). In ANR Pipeline Co., 48 FERC p 61,140 at 61,553 (1989), the Commission ruled that costs eligible for recovery include not only cash but also the fair market value of any non-cash consideration used to resolve the take-or-pay liability. This appeal is a predictable consequence of that ruling, namely, a dispute between a pipeline and its customers over the fair market value of non-cash consideration for which the pipeline seeks recovery.

In December 1989, El Paso entered into a settlement agreement with one of its gas suppliers, TransAmerican Natural Gas Corporation, to resolve an ongoing take-or-pay contract dispute. Pursuant to the terms of the settlement, El Paso paid TransAmerican $302,374,350 in cash and conveyed to TransAmerican interests in certain oil- and gas-producing properties located in the La Perla Ranch field, Zapata County, Texas ("La Perla properties" or "La Perla"). The settlement agreement did not place a value on the properties; they were subsequently appraised, however, by Tom G. Calhoun II, of Calhoun Engineering, Inc., on behalf of El Paso. Calhoun utilized a discounted cash flow ("DCF") methodology to determine that their value at the time of transfer to TransAmerican was $135 million. Generally speaking, a DCF methodology determines the net present value of a property by predicting future revenues (derived by estimating the amount of oil and/or gas reserves present, the level of risk attributed to those reserves, the anticipated rates of production, future prices, and development and operating costs) and discounting the future revenues back to a "net present value" by applying one of several discounting methodologies.

In February 1990, El Paso filed a tariff with FERC seeking to recover from its customers 75 percent of the costs incurred in the TransAmerican settlement, including 75 percent of the $135 million it alleged to be the fair market value of the properties. In response to this filing, the Public Utilities Commission of the State of California, Pacific Gas and Electric Company, and Southern California Gas Company (collectively, "California Parties") retained Harry J. Gaston, of Ryder Scott Company Petroleum Engineers, to review the Calhoun appraisal and to make an independent appraisal of the La Perla properties. Gaston's appraisal, like Calhoun's, was based on a DCF methodology; but because he employed different assumptions as to estimated reserves, future gas prices, and operating costs, and because he used a different discount method, Gaston arrived at a fair market value of only $78.5 million. In order to resolve the issue of the properties' fair market value, the Commission ordered a hearing before an administrative law judge ("ALJ"). El Paso Natural Gas Co., 61 FERC p 61,107 at 61,425 (1992).

[321 U.S.App.D.C. 15] Shortly before the hearing, which was held on January 19, 21, and 22, 1993, Calhoun became aware of certain computer errors in his DCF analysis that required him to reduce his previous valuation to $120 million. Nonetheless, at the hearing, El Paso continued to claim $135 million as the fair market value of the properties on the ground that the market for gas-producing properties in South Texas in the late 1980's was so competitive that the properties would have sold for a premium above the value established by his DCF analysis. In support of this position, El Paso offered the testimony of Calhoun and Thomas E. Hassen, a market analyst with Morgan Stanley & Co. Both experts relied heavily on three sales that had occurred at about the time of the La Perla properties' transfer: (1) Pacific Gas and Electric Company's purchase of properties from Wessely Energy (TCPL) in September 1988 ("Wessely"); (2) Texaco, Inc.'s purchase of properties from Tana Production Corporation in October 1989 ("Tana"); and (3) DeKalb Energy, Inc.'s purchase of properties from Royal Producing Corporation in May 1990 ("Royal").

Calhoun emphasized that the actual sales prices of the Tana and Royal transactions were significantly above their respective DCF estimates, which he had himself performed. He testified that, although the fair market value of the La Perla properties was approximately $120 million based solely on his DCF analysis, "if placed on the open market in the business climate existing at the end of 1989, [it] would have brought a price approximately 25 percent higher ..., or approximately $150 million." He thus concluded that "$135 million is a reasonable, probably even conservative, estimate of [the properties'] market value...." See Direct Testimony of Tom G. Calhoun II at 6, reprinted at Joint Appendix ("J.A.") 77. Hassen, in turn, testified that, when viewed on the basis of price paid per thousand cubic feet ("Mcf") equivalent of gas reserves, these three properties sold for significantly more per Mcf than the $0.85 placed on the La Perla properties' reserves by a $135 million valuation. He thus concluded that $135 million was a "very reasonable" estimate of the properties' fair market value. Direct Testimony of Thomas E. Hassen at 16-17, reprinted at J.A. 63-64.

Although the California Parties' witness, Gaston, commented on the Calhoun and Hassen analyses, the California Parties did not submit any market evidence of their own. Rather, they relied on Gaston's DCF analysis to support their assertion that the fair market value of the La Perla properties at the time of that transfer was only $78.5 million.

On March 8, 1993, the ALJ issued an initial decision, in which he held that no weight would be given to El Paso's market evidence in determining the properties' fair market value because

[t]he evidence of purportedly comparable sales at the time of the transfer of the La Perla Properties ... has not been shown on [the] record to be comparable and does not warrant an augmentation of the [fair market value] of the La Perla Properties as initially submitted to the Commission by El Paso.

El Paso Natural Gas Co., 62 FERC p 63,024 at 65,101 (1993) ("Initial Decision"). Instead, the ALJ determined that their fair market value would be ascertained solely by reference to the competing DCF analyses submitted by El Paso and the California Parties. Id. The ALJ proceeded to make findings as to the appropriate reserve estimates, the risk factors to be applied to proved undeveloped and probable undeveloped reserves, future gas prices, operating costs, and the appropriate discount method to be used, and ordered El Paso to recompute its analysis using those specific figures. The parties have stipulated that, calculated according to this method, the fair market value of the La Perla properties was $98,328,420.

In an order issued June 16, 1994, the Commission affirmed the ALJ's decision to reject El Paso's evidence of comparable sales on the ground that El Paso had failed to carry its burden of showing comparability. El Paso Natural Gas Co., 67 FERC p 61,327 at 62,154 (1994) ("Order Affirming Initial Decision"). Similarly, the Commission rejected a claim by El Paso that the fair market value of the properties should have been established by utilizing Hassen's Mcf-equivalent [321 U.S.App.D.C. 16] price evidence. The Commission subsequently denied El Paso's petition for rehearing. El Paso Natural Gas Co., 69 FERC p 61,155 (1994) ("Order Denying Rehearing").

El Paso now seeks review of these orders, arguing that the Commission erred in rejecting its evidence of comparative sales and in refusing to consider the Mcf-equivalent approach to valuation.

It is incumbent on FERC, in resolving a dispute over the fair market value of a property, to utilize the valuation method that will yield the most accurate results in a particular case. Cf. United States v. 103.38 Acres of Land, 660 F.2d 208, 212 (6th Cir.1981) ("It is self-evident that the trier of fact's goal in estimating 'fair market value' is maximum accuracy."). The choice of methodologies will depend on a number of factors, such as the characteristics of the property to be valued and the quantity and quality of the information available to the decisionmaker. See United States v. 429.59 Acres of Land, 612 F.2d 459, 462 (9th Cir.1980) (determination of fair market value must take into consideration "all the facts and circumstances that would reasonably go into the making of a bargain of purchase and sale"). Because this inquiry is fact intensive, it is appropriate to give significant deference to the Commission's choice of a valuation methodology. Cf. Seravalli v. United States, 845 F.2d 1571, 1575 (Fed.Cir.1988) (decisionmaker "must have considerable discretion to select the method of valuation that is most appropriate in the light of the facts of the particular case"). We must, of course, defer to the Commission's findings of fact so long as they are supported by substantial evidence in the record. See 15 U.S.C. § 717r(b) (1994).

We agree with El Paso that evidence of contemporaneous sales of comparable properties is generally the preferred method of valuation. See United States v. 320.0 Acres of Land, 605 F.2d 762, 798 (5th Cir.1979) ("Courts have consistently recognized that, in general, comparable sales constitute the best evidence of market value."). The validity of this method, however, depends on both the timing of the sales and the "comparability" of the properties involved. Where the properties are not shown to be sufficiently similar or the sales sufficiently contemporaneous, the "trier of fact ... must resort to other means of determining fair market value." 103.38 Acres of Land, 660 F.2d at 211.

In finding that El Paso had failed to establish the comparability of the properties that provided the basis for its comparable sales valuation, the Commission stated that it was applying the standard set forth in 103.38 Acres of Land and Fairfield Gardens, Inc. v. United States, 306 F.2d 167, 172 (9th Cir.1962); namely, that although the properties need not be identical, they must be "similar in many respects." See Order Denying Rehearing, 69 FERC at 61,591 & n. 10. This formulation strikes us as essentially meaningless: gold and fool's gold are similar in many respects, but the value of one hardly serves as proxy for the value of the other. The similarities, of course, must be sufficiently relevant to enable the trier of fact to determine, with reasonable confidence, the price that a buyer would be willing to pay a willing seller for the property being valued, both being informed of the relevant facts and neither being under any compulsion to buy or sell. See Reservation Eleven Assocs. v. District of Columbia, 420 F.2d 153, 155 (D.C.Cir.1969). This is the standard that FERC in fact applied when it rejected El Paso's comparable sales evidence.

The Commission found that El Paso's expert, Calhoun, had failed to establish that the properties were comparable because, "other than pointing out that the properties were located in South Texas, were 'primarily' of gas reserves, were near in time to the La Perla transfer, and were sold in excess of the 'engineering estimate' of the properties, he [had] analyzed nothing more specific." Order Denying Rehearing, 69 FERC at 61,592. It expressed particular concern over El Paso's failure to compare "significant factors, such as differences in the amount of additional drilling costs that must be incurred on each respective property," id. at 61,591, or [321 U.S.App.D.C. 17] "the cost to produce and deliver reserves." Id. at 61,592. And it noted that in the case of Tana and Royal, the transactions involved the sale of the capital stock of the companies owning those properties; that the assets of those companies "include[d] many different oil and gas leases having different production characteristics and operational considerations"; and that their sales prices could reflect "intangibles such as goodwill and other things of value apart from just proved reserves." Id.

El Paso dismisses the distinctions drawn by the Commission. It contends, for example, that, "[i]n the absence of evidence" that the purchase prices reflected values other than those of the underlying reserves, the Commission was unjustified in relying on the fact that the Tana and Royal transactions were accomplished through the sale of the corporate entities rather than of individual properties. The burden, however, was on El Paso to establish comparability; and when FERC identified factors that cast any serious question on the matter, it was incumbent on El Paso to demonstrate that the differences either were not material or that they had been accounted for. While we agree with El Paso that not every distinction identified by FERC is relevant to a determination of comparability (e.g., we are unpersuaded that the fact that Tana was bought with Texaco stock rather than cash was a material distinction), on the whole, we find that the problems identified by the Commission were both material and supported by substantial evidence in the record.

El Paso also argues that FERC erroneously rejected the Royal sale by arbitrarily excluding sales occurring after the La Perla transfer. See Order Denying Rehearing, 69 FERC at 61,593 ("one of the properties, the Royal properties, was sold after the date of the La Perla properties and cannot, therefore, be relied on"). While we agree with El Paso that there is no justification for the application of a per se rule excluding consideration of post-transaction sales, see United States v. 63.04 Acres of Land, 245 F.2d 140, 144 (2d Cir.1957) ("There is no absolute rule which precludes consideration of subsequent sales."), this was merely an alternative basis for rejecting the Royal sale. The Commission had already found that the Royal sale was not comparable because the development of its reserves would involve lower drilling costs than would the development of La Perla's and because there were greater liquid petroleum reserves at Royal than at La Perla. Order Denying Rehearing, 69 FERC at 61,592.

B. "Mcf Equivalent Prices" as Evidence of Comparability

El Paso's expert, Hassen, submitted testimony and exhibits that were designed to show that a valuation of $135 million placed a price on the La Perla properties' gas reserves that was significantly lower than the price paid for comparable reserves in other South Texas transactions. Specifically, he compared the "Mcf equivalent price" attributable to each property. The "Mcf equivalent price" reflects the price paid by the buyer for reserves in the ground. It is determined by dividing the price paid for a property by that property's proved developed and undeveloped gas equivalent reserves.

By dividing $135 million (El Paso's asserted fair market value of La Perla) by Calhoun's estimate of its reserves, Hassen arrived at an Mcf equivalent price for La Perla's reserves of $0.85, which he then compared to the Mcf equivalent prices he determined to have been paid for some 60 other gas and oil properties in South Texas. These comparisons showed that $0.85 per Mcf was significantly lower than all but one other sale in 1989; that it was significantly lower than the prices paid for Wessely ($1.45 per Mcf), Tana ($1.39 per Mcf), and Royal ($1.58 per Mcf); and that it was lower than the median price paid for South Texas gas reserves in 1988 ($1.10 per Mcf), 1989 ($1.16 per Mcf), and 1990 ($1.20 per Mcf). On the basis of these comparisons, Hassen stated that "$135 million [was] a very reasonable fair market value for [La Perla]." Direct Testimony of Thomas E. Hassen at 16-17, reprinted at J.A. 63-64.

The Commission found, however, that El Paso had failed to support its "asserted quantification of an exact 'Mcf equivalent' [321 U.S.App.D.C. 18] 'premium' price," Order Denying Rehearing, 69 FERC at 61,593, and thus declined to afford any weight to Hassen's testimony. The Commission noted, for example, that some of the properties compared by Hassen involved the transfer of assets other than gas reserves; accordingly, in order to establish an Mcf equivalent price for those properties, Hassen had to extract from the total sales price a specific portion to be assigned solely to the purchase of the gas reserves. The Commission found this "extraction" to be a "wholly speculative proposition." Id.

Similarly, the Commission noted that Hassen's study failed to account for the fact that other properties may have had "varying quantities of additional 'probable' or other higher-risk categories of estimated gas and oil reserves which were sold along with the proved reserves but which were not reflected in Hassen's calculations." Id. at 61,592. If Hassen had taken into consideration the price the purchasers might be willing to pay for these higher risk categories, the Mcf equivalent prices attributed to the Tana, Wessely, and Royal proved reserves would have been lower. Conversely, Hassen underestimated the Mcf equivalent price for La Perla by relying on Calhoun's estimates of proved reserves, which were subsequently found by the ALJ to be exaggerated.

El Paso now concedes this last point but argues that if one bases the calculation on the net proved reserves found by the Commission (approximately 97,200,000 Mcf), the resultant Mcf equivalent price for La Perla is only $1.01 per Mcf, "which is still below the ... average yearly Mcf equivalent prices and far below the Mcf equivalent prices paid for Wessely, Tana and Royal." Brief for Petitioner at 9. Indeed, those yearly averages were $1.10 for 1988, $1.16 for 1989, and $1.20 for 1990. See Exhibits to Hassen testimony, reprinted at J.A. 70-71. But the Commission found that El Paso's Mcf-equivalent analysis was unreliable because it failed to take into account important drawbacks to the La Perla properties, which "required additional, costly drilling and potential workover costs that no prudent purchaser would ignore." Order Denying Rehearing, 69 FERC at 61,591. Under these circumstances, we find no reason to question the Commission's preference for a DCF methodology. The likelihood that the variations in extraction cost invoked by the Commission might have a drastic impact is, indeed, suggested by the range of Mcf equivalent prices in the data offered by El Paso. In 1989, for example, the prices ranged from $0.55 to $1.88 per Mcf.

El Paso having failed to prove the existence of comparable sales, it was entirely proper for FERC to rely on the DCF methodology in order to determine the value of the La Perla properties. See United States v. 22.80 Acres of Land, 839 F.2d 1362, 1364 (9th Cir.1988). It is, in fact, the methodology El Paso itself relied on to support its fair market value estimate until the discovery of computer errors forced it to reduce its DCF estimate by $15 million.

For the foregoing reasons, we conclude that the Commission did not abuse its discretion by relying on a DCF analysis to ascertain the fair market value of the La Perla properties and in rejecting El Paso's alternative valuation methodologies. We acknowledge that, under the proper circumstances and properly applied, those methodologies may establish a more accurate estimate of a property's current market value. Nonetheless, the Commission presented legitimate reasons supported by the record evidence for rejecting El Paso's comparable sales evidence; and having done so, it was within its discretion to choose the next best method of valuation. Accordingly, the petition for review is

Denied.