Full opinion text
OPINION DAVID PURYEAR, Justice. Our opinion and judgment issued on December 20, 2007, are withdrawn, and the following opinion is substituted. This appeal concerns the transition of Texas’s energy industry from a regulated market to a competitive one. When it approved the switch to a competitive market, the legislature contemplated the possibility that the switch might saddle formerly regulated utilities with costs that they would have recovered under regulation but would be unable to recover in a competitive market. As a result, the legislature enacted statutes authorizing utilities to recover these costs in proceedings called true-up proceedings held before the Public Utility Commission (the “Commission”). The utilities involved in this case estimated the costs that they would not be able to recover due to deregulation and filed an application with the Commission seeking recovery for those costs. However, the Commission determined that not all of the relevant requirements had been satisfied when the utilities made their calculations and, therefore, performed its own estimate of the utilities’ unrecovered costs. The total amount determined by the Commission was less than the amount that the utilities originally requested. In addition to producing its own estimation, the Commission also made several reductions to the utilities’ recovery. Although the Commission allowed the utilities to recover for various construction projects that they had started, it deducted the value of certain tax benefits given to the utilities. The Commission also reduced the utilities’ recovery because it concluded that the utilities had recovered some of their costs through other means. Finally, although the Commission allowed the utilities to recover the requested amount for credits that the Commission had previously ordered them to give to their customers, it denied recovery for interest on the credits. The district court affirmed the majority of the Commission’s order but reversed the order and increased the utilities’ recovery in two respects. First, the district court concluded that the utilities should recover for the interest on the credits that they were ordered to give. Second, the district court concluded that the Commission’s decision to undertake its own estimate of one of the utilities’ costs was inappropriate and further concluded that the utilities should recover the amount originally requested. We will affirm the judgment of the district court in part and reverse and remand in part. STATUTORY FRAMEWORK To give context to the merits of this case, we will describe the statutory framework governing this case. This appeal concerns the utility market’s transition from a regulated industry to a competitive, deregulated market. See Tex. Util.Code Ann. §§ 39.001-.910 (West 2007). Prior to deregulation, utilities operated as monopolies but were regulated by the Commission and were “prohibited from charging monopoly prices.” Reliant Energy, Inc. v. Public Util. Comm’n, 101 S.W.3d 129, 133 (Tex.App.-Austin 2003) (“Reliant I”), rev’d in part sub nom., CenterPoint Energy, Inc. v. Public Util. Comm’n, 143 S.W.3d 81 (Tex.2004); see Reliant Energy, Inc. v. Public Util. Comm’n, 153 S.W.3d 174, 182 (Tex.App.-Austin 2004, pet. denied) (“Reliant II”). “[Ejach region of the state was served by a single vertically integrated utility,” Cities of Corpus Christi v. Public Util. Comm’n, 188 S.W.3d 681, 684 (Tex.App.-Austin 2005, pet. filed), which meant that the utility “produced, transported, and retailed electricity” for the region, Reliant I, 101 S.W.3d at 133. In 1999, the legislature enacted statutes that initiated the transition to a competitive retail-service industry. See Act of May 27, 1999, 76th Leg., R.S., ch. 405, 1999 Tex. Gen. Laws 2543 (current version at Tex. Util.Code Ann. §§ 39.001-.910). The legislature concluded that the “production and sale of electricity” was not an undertaking necessitating the utilization of monopolies or the “regulation of rates, operations, and services” and that it was in the public interest to allow customer choice and competition to determine the prices for these services. Tex. Util.Code Ann. § 39.001(a); see also In re TXU Elec. Co., 67 S.W.3d 130, 132 (Tex.2001) (Phillips, C.J., concurring). Accordingly, the utilities code was amended to allow for retail competition starting January 1, 2002, and to protect the interests of the citizens of Texas during the transition. Tex. Util. Code Ann. § 39.001(a); see also In re TXU Elec. Co., 67 S.W.3d at 132 (Phillips, C.J., concurring). The transition to a competitive retail market involved several changes to how utilities provided electricity. Significantly, the formerly integrated utilities were required to “unbundle” and divide into three separate entities: (1) retail electric providers, (2) power-generation companies, and (3) transmission-and-distribution utilities. Tex. Util.Code Ann. § 39.051(a)-(b); see also In re TXU Elec. Co., 67 S.W.3d at 132 (Phillips, C.J., concurring); Reliant II, 153 S.W.3d at 182. Starting in 2002, the unbundled power-generation companies owned and operated “the generating plants,” In re TXU Elec. Co., 67 S.W.3d at 132 (Phillips, C.J., concurring), and provided “wholesale generation services in competition with other generators entering the market,” Cities of Corpus Christi, 188 S.W.3d at 684. The transmission-and-distribution utilities owned and maintained “the ‘wires’ used to transport electricity from the power generation companies to all [retail electric providers] and retail consumers in the utility’s geographic area.” Id. at 685. The retail electric provider sold “electricity to end-use customers” and provided “customer service.” In re TXU Elec. Co., 67 S.W.3d at 132 (Phillips, C.J., concurring). In addition, new electricity providers were allowed to begin competing with the retail electric providers associated with the former integrated utilities. See Tex. Util.Code Ann. § 39.102(a)-(b). After the deregulation process was completed, the power-generation and retail electric markets would be subject to the “normal forces of competition” and “customer choices,” but the transmission-and-distribution utilities would remain regulated by the Commission. Id. § 39.001(a); see Cities of Corpus Christi, 188 S.W.3d at 685. However, the deregulation process is lengthy, and the Commission retained partial regulatory powers over power generation and the sale of electricity after January 2002. See, e.g., Tex. Util.Code Ann. § 39.202 (allowing Commission some control over prices charged by utilities). During the transition, affiliated retail electric providers were required to charge a “price to beat” rate to their residential and small-business customers. Id. Prior to deregulation, utilities were allowed to recover from their customers the prudent costs they incurred when acquiring power-generation assets. Reliant II, 153 S.W.3d at 183 n. 5; Reliant I, 101 S.W.3d at 134. The Commission allowed the utilities to recover these costs over time by incorporating the costs into the rates that it approved. Reliant II, 153 S.W.3d at 183 n. 5; Reliant I, 101 S.W.3d at 134. As a result, utilities made significant investments in generation-related assets with the expectation of eventually recovering their costs. See Cities of Corpus Christi, 188 S.W.3d at 685. Recognizing that this type of reimbursement would not occur under deregulation, utilities expressed their concern that under deregulation they would be unable to recover the costs for their investments because competition would drive the rates too low. Reliant II, 153 S.W.3d at 183 n. 5; Reliant I, 101 S.W.3d at 134. Because new utilities entering the market would not have “embedded generation-related costs,” they could set prices below the “level at which incumbent utilities could recover their investments.” Cities of Corpus Christi, 188 S.W.3d at 685. Therefore, the incumbent utilities would either have to charge rates that were not competitive or absorb the added expense. Id. To prevent the possibility that utilities would have to absorb the costs, the legislature provided a method by which a utility could recover its “stranded costs” or those costs representing the “portion of the net book value of [the] utility’s generation assets not yet recovered through depreciation that has become unrecoverable in a deregulated market.” Reliant I, 101 S.W.3d at 134; see also Tex. Util.Code Ann. §§ 39.001(b)(2) (finding that it is in public interest to “allow utilities with uneconomic generation-related assets ... to recover these reasonable excess costs over market of those assets”), .251(3) (defining generation assets as “all assets associated with the production of electricity, including generation plants”), .251(4) (defining market value as “the value the assets would have if bought and sold in a bona fide third-party transaction or transactions on the open market”), .251(7) (defining stranded costs as “the positive excess of the net book value of generation assets over the market value of the assets”),.252 (providing that utility is entitled to recover stranded costs); 16 Tex. Admin. Code § 25.263(g) (2007) (specifying what constitutes “net book value”). Although the legislature allowed a utility to recover stranded costs, there were express limitations imposed on this right. The utility was required to mitigate the amount of stranded costs it incurs from purchasing electricity and “providing electric generation service,” Tex. UtiLCode Ann. § 39.252(a), and was required to “pursue commercially reasonable means to reduce its potential stranded costs,” id. § 39.252(d). In addition, the Commission was authorized to consider “the utility’s efforts [to reduce its potential stranded costs] when determining the amount of the utility’s stranded costs.” Id.; see also 16 Tex. Admin. Code § 25.263(e)(4) (2007) (stating that Commission may adjust net book value of affiliated power-generation company’s generation assets if utility has failed to undertake reasonable actions to reduce its potential stranded costs); Reliant I, 101 S.W.3d at 149 (noting that terms of section 39.252 impliedly contemplate allowing adjustments to book value, which is the only other component of stranded costs besides market value). Finally, the utilities code specifies that “[a]n electric utility, together with its affiliated retail electric provider and its affiliated transmission- and-distribution utility, may not be permitted to overrecover stranded costs.” Tex. Util.Code Ann. § 39.262(a). To foster the recovery of stranded costs, the Commission used a computer model called the “Excess Cost Over Market” model (“ECOM”) to predict whether utilities would actually incur stranded costs in a deregulated market. See In re TXU Elec. Co., 67 S.W.3d at 160 (Hecht, J., dissenting). The model accounted for various factors, including fuel costs, in its calculations. Cities of Corpus Christi, 188 S.W.3d at 686. Based on this model, the Commission prepared a report for the Texas Senate in 1998 that predicted the amount of stranded costs that utilities would likely incur in the deregulated market (“1998 ECOM Report”). Reliant I, 101 S.W.3d at 134 n. 3. However, in its report, the Commission did caution that the amount predicted was only an estimate and that the amount of stranded costs that would actually result, if any, might be significantly different than the estimated amount. In re TXU Elec. Co., 67 S.W.3d at 160 (Hecht, J., dissenting). To minimize the impact on consumers and utilities, the legislature devised a three-step program for the recovery of stranded costs. The first step began in September 1999 and ended December 31, 2001. During this step, the retail electric rates charged by utilities were frozen. Tex. Util.Code Ann. § 39.052. In addition, the Legislature provided various methods for utilities to “mitigate” their stranded costs in order to lessen the impact on consumers resulting from stranded-cost recovery and to minimize the delay in the benefits resulting from competition. Id. §§ 39.254, .256; see also In re TXU Elec. Co., 67 S.W.3d at 160-61 (Hecht, J., dissenting). For example, to mitigate their stranded costs, utilities could transfer depreciation away from transmission-and-distribution assets to generation assets. Tex. Util.Code Ann. § 39.256. The second step began on the first day of competition, January 1, 2002, and ended December 31, 2003. See id. §§ 39.001(b)(1), .201(a), (b)(3), (g), (h); In re TXU Elec. Co., 67 S.W.3d at 133 (Phillips, C.J., concurring). During this stage, company-specific updates were inputted into the ECOM model to ascertain the status of stranded-cost recovery. See Tex. Util.Code Ann. § 39.201(h); Cities of Corpus Christi, 188 S.W.3d at 686. If the ECOM model calculations predicted that utilities would have stranded costs even after employing the various mitigation techniques available in the first stage, the Commission was authorized to set a non-bypassable “competition transition charge” to allow the utilities to recover these costs by collecting a fee from each customer obtaining power. See Tex. Util.Code Ann. § 39.201(b)(3); In re TXU Elec. Co., 67 S.W.3d at 133 (Phillips, C.J., concurring); Cities of Corpus Christi, 188 S.W.3d at 686-87. This charge was intended to make up the difference between the book value and the market value of a power-generation plant and, therefore, allow utilities to recover the additional expected stranded costs. In re TXU Elec. Co., 67 S.W.3d at 133 (Phillips, C.J., concurring). The affiliated power-generation companies and providers would bill the charge to the transmission-and-distribution utilities, which were allowed to pass through the charge “to retail customers” by including the amount of the charge in their “ ‘wholesale’ rates.” Id. at 160 (Hecht, J., dissenting). The charge constituted one of a number of “nonbypassable delivery charges” passed through to customers. Tex. Util.Code Ann. § 39.201(b). When the stranded-cost estimates were updated, the estimates “unexpectedly reflected that the utilities would have no stranded costs.” Reliant I, 101 S.W.3d at 135. As a result, the Commission ordered utilities to cease stranded cost mitigation efforts, “to reassign the depreciation transferred from transmission and distribution assets back to those assets, and to return monthly ‘excess mitigation credits’ to retail providers.” Id.; see In re TXU Elec. Co., 67 S.W.3d at 161 (Hecht, J., dissenting). The third step began in 2004 and is the step relevant in this appeal. Tex. Util. Code Ann. §§ 39.201, .262(c). During this stage, the Commission was required to conduct a “true-up proceeding” to determine a final calculation of a utility’s stranded costs, if any. Id. §§ 39.201(l), .262(c). The purpose of the proceeding was to reconcile the actual stranded costs incurred with the previous estimates made by the Commission. See id. §§ 39.201(l), .262(c); see also 16 Tex. Admin. Code § 25.263(a) (2007) (specifying purpose of true-up proceeding). As part of the proceeding, “each transmission and distribution utility, its affiliated electric provider, and its affiliated power generation company” were required to “jointly” file finalized stranded costs and reconcile those costs with the estimated stranded costs. Tex. Util.Code Ann. § 39.262(c). One of the most important aspects of the true-up proceeding was the determination of the actual “market value of a utility’s generation assets.” Reliant I, 101 S.W.3d at 143. The code lists several alternative methods by which an affiliated power-generation company could calculate the market value of its generation assets for the purpose of calculating its stranded costs. Tex. Util.Code Ann. § 39.262(h)(1)-(4). These valuations utilize “stock prices and anticipated income streams in a competitive market.” Cities of Corpus Christi, 188 S.W.3d at 687 (citing Tex. Util.Code Ann. §§ 39.201(l), .262(h), (i)). The true-up calculation obtained was the “final, controlling calculation of each utility’s stranded costs.” Id. at 692. The utility’s actual stranded costs were determined by subtracting the actual market value of the utility’s generation assets from the book value of those assets. Tex. Util. Code Ann. §§ 39.251(7), .252(a), .262(c), (h), (i). If the number obtained in this calculation was a positive number, then the utility was entitled to recover that amount in stranded costs. Reliant I, 101 S.W.3d at 136. The stranded-cost true-up was only one of several true-up calculations that had to be performed as part of the transition to competition. See Tex. Util.Code Ann. § 39.262(d)-(g). The utilities code establishes “two parallel true-up tracks — one for stranded costs and one for the several other true-up items.” Reliant I, 101 S.W.3d at 141. These non-stranded-cost calculations also can “result in either credits or bills to the transmission and distribution utility from its affiliated power generation company or retail electric provider.” Id. at 136 (citing Tex. Util.Code Ann. § 39.262(d)-(g)). One of the non-stranded-cost true-ups relevant to this case involves the calculation of a utility’s “capacity-auction award.” As part of the transition to a competitive market, utilities were required to auction off entitlements to some of their generation assets. See Tex. Util.Code Ann. § 39.153(a). The capacity-auction award constituted the difference between the price that a utility was predicted by the ECOM model to obtain for selling its power in the wholesale market during the second step of deregulation and the price actually obtained at auction during the first years of deregulation. See 16 Tex. Admin. Code § 25.263(i), (l) (2007). After determining the capacity-auction award, the figure was netted with another true-up award called the final fuel balance. Tex. Util.Code Ann. § 39.262(d). Once the various calculations were made, they were all considered when determining whether a utility was entitled to recover for costs. See 16 Tex. Admin. Code § 25.263(l)(1) (2007). If the true-up balance was positive and greater than the projected costs, the utility was entitled to recover the amount calculated. Based on the actual stranded costs calculated, the Commission was authorized to alter the period of time during which a utility may collect the competition transition charge or alter the amount of the charge. Tex. Util. Code Ann. §§ 39.201(l), -262(c), (d)(1), (g); 16 Tex. Admin. Code § 25.263(l)(2)(A) (2007); Reliant I, 101 S.W.3d at 137; see also Tex. Util.Code Ann. § 39.201(b) (specifying nonbypassable delivery charges). BACKGROUND CenterPoint Energy Houston Electric, LLC (“CenterPoint”); Reliant Energy Retail Services, LLC (“Reliant”); and Texas Genco, LP (“Genco”) (cumulatively “Joint Applicants”) are the unbundled components of the formerly integrated Reliant Energy: CenterPoint is the transmission- and-distribution utility, Reliant is the affiliated retail electric provider, and Genco is the power-generation company. In March 2004, they filed a joint application for a final true-up proceeding to determine their recovery for stranded costs and non-stranded costs, including their capacity-auction award. See Tex. Util.Code Ann. §§ 39.252(a), .262(c), (d)(2). In addition to the Joint Applicants, several other parties also intervened in the true-up proceeding. The intervening parties were the Office of Public Utility Counsel (“Utility Counsel”), see Tex. Util.Code Ann. § 13.003 (West 2007) (describing powers and duties of Utility Counsel), and several coalitions of interested parties that either were within CenterPoint’s service area or purchased energy from Center-Point, including the City of Houston, the Coalition of Cities, the Gulf Coast Coalition of Cities, the Houston Council for Health and Education, the State of Texas, and Texas Industrial Energy Consumers. For the sake of clarity, we will refer to these coalitions as the “Customers.” Stranded Costs In their application, the Joint Applicants asserted that they were entitled to $2,454 billion in stranded costs and $539.4 million in interest on the stranded-cost award. For ease of discussion, we will only list the specific stranded costs requested that are relevant to this appeal. First, the Joint Applicants requested $470 million in recovery for credits that the Commission had previously ordered them to give to their customers and $180 million in interest on those credits. Second, the Joint Applicants sought $147 million for various construction projects that they had begun pri- or to deregulation and for various land purchases that they made to secure locations for future power plants. After conducting a hearing, the Commission issued its final true-up order in December 2004. In its order, the Commission authorized the recovery of the $470 million that had been awarded as credits and also allowed the Joint Applicants to recover the $147 million spent on pre-deregulation construction projects. However, the Commission made significant reductions to the Joint Applicants’ requested recovery. First, it disallowed recovery for the $180 million in interest that had been credited to the utilities’ customers. Second, the Commission reduced the award by $146 million to account for various tax benefits given to the Joint Applicants. Finally, because the Commission believed that the Joint Applicants recovered some of their stranded costs through the capacity-auction process, the Commission further reduced the stranded-cost true-up award by $378.4 million. In its order, the Commission also made two alternative holdings regarding the Joint Applicants’ estimate of the value of their generation assets, which they were required to calculate as part of the recovery process. Under its primary holding, the Commission concluded that the Joint Applicants’ valuation of their assets was not valid because they did not comply with all the statutory requirements. For this reason, the Commission performed its own valuation of the Joint Applicants’ assets. See Tex. Pub. Util. Comm’n, Application of CenterPoint Energy Houston LLC, Reliant Energy Retail Services LLC, and Texas Genco LP to Determine Stranded Costs and Other True-Up Balances Pursuant to PURA § 39.262, Docket No. 29526, at 18 (Dec. 17, 2004) (Order on Rehearing) (“order”). In its appraisal, the Commission concluded that the market value of the assets was approximately $509 million higher than that estimated by the Joint Applicants. Consequently, the Commission determined that the Joint Applicants’ stranded costs were less than the amount requested and reduced their recovery accordingly. After making the reductions previously discussed and after utilizing its own market valuation, the Commission concluded that the Joint Applicants were entitled to recover $1,222 billion in stranded costs and $121 million in interest under its primary holding. Under its alternative holding, the Commission assumed that the Joint Applicants satisfied the necessary statutory requirements but made an additional reduction to the Joint Applicants’ recovery that it didn’t make in its primary holding. The Commission deducted approximately $508 million from the Joint Applicants’ recovery to account for business practices that the Commission believed were commercially unreasonable and for the tax benefit resulting from this unreasonable behavior. After making all the relevant reductions, the Commission concluded that the Joint Applicants were entitled to recover $945 million in stranded costs plus $68 million in interest under its alternative holding. The chart below details the relevant stranded-cost recovery requested by the Joint Applicants and the various modifications made by the Commission in its primary and alternative holdings: Stranded Costs Calculations in Millions of Dollars Capacity Auction In their application, the Joint Applicants also requested $1,357 billion for deficits sustained from the capacity auctions. However, in its order, the Commission reduced the requested award. The Commission concluded that the capacity-auction calculation performed by the Joint Applicants was invalid because they failed to satisfy the necessary statutory requirements. See Tex. Util.Code Ann. §§ 39.153, .262(d)(2). As with the asset valuation, the Commission performed its own estimate of the capacity-auction award and deducted $440 million from the Joint Applicants’ requested recovery. Although the Commission reduced the requested award, it did allow the Joint Applicants to recover $168 million in interest on the award to account for the fact that the Joint Applicants had been deprived of the predicted capacity-auction award for a specific period of time. The chart below details the relevant capacity-auction recovery requested by the Joint Applicants and the various modifications made by the Commission in its primary and alternative holdings: Capacity Auction Calculations in Millions of Dollars Joint Applicants’ Appeal After the order was issued, the Joint Applicants appealed the decision to the district court. See Tex. Util.Code Ann. § 15.001 (West 2007) (stating that party to proceeding before Commission is entitled to judicial review). The Customers and the Utility Counsel also appealed the order, contending that the Commission erred in several respects. After reviewing the Commission’s order, the district court issued its judgment. The district court affirmed the majority of the Commission’s order, including the decision of the Commission to perform its own assessment of the value of Joint Applicants’ assets, but reversed on two grounds. The district court’s reversal increased the amount of stranded costs that the Joint Applicants were entitled to receive. Specifically, the judgment concluded that the Commission erred by (1) preventing the joint applicants from collecting $180 million in interest on the credits and (2) disallowing $440 million from the capacity-auction true-up. Accordingly, the Joint Applicants’ recovery was increased by those amounts. The Joint Applicants, the Customers, the Utility Counsel, and the Commission all appeal the judgment of the district court. See id. §§ 15.001 (stating that any party to Commission proceeding may appeal), 39.262(j) (specifying that final order by Commission is subject to judicial review); Tex. Gov’t Code Ann. § 2001.171 (West 2000) (explaining that after exhausting administrative remedies, party aggrieved by final agency decision is entitled to judicial review of decision). STANDARD OF REVIEW The proper standard of review to utilize in this case is complicated by the fact that many of the issues are multifaceted, requiring the application of various standards in achieving a final resolution. In light of this fact and for efficiency, we will attempt to summarize the various standards that will be employed in this appeal. Several of the issues raised in this appeal involve statutory construction, which is a question of law that is reviewed de novo. See Bragg v. Edwards Aquifer Auth., 71 S.W.3d 729, 734 (Tex.2002); USA Waste Servs. of Houston, Inc. v. Strayhorn, 150 S.W.3d 491, 494 (Tex.App.-Austin 2004, pet. denied). In construing a statute, we must ascertain the legislature’s intent in enacting the statute. Fleming Foods of Tex. v. Rylander, 6 S.W.3d 278, 284 (Tex.1999). In making this determination, courts should look to the plain mean ing of the words used in the statute. See Fireman’s Fund County Mut. Ins. Co. v. Hidi, 13 S.W.3d 767, 768-69 (Tex.2000). We presume that every word was deliberately chosen and that excluded words were left out on purpose. USA Waste Servs., 150 S.W.3d at 494. When determining legislative intent, the entire act, not isolated portions, must be considered. Jones v. Fowler, 969 S.W.2d 429, 432 (Tex.1998). We may also consider the “object sought to be attained” by enacting the statute, the “circumstances under which the statute was enacted,” the “consequences of a particular construction,” and the interpretations of the statute made by an agency. Tex. Gov’t Code Ann. § 311.023 (West 2005); see City of Austin v. Southwestern Bell Tel. Co., 92 S.W.3d 434, 442 (Tex.2002). Moreover, so long as the interpretation is reasonable and consistent with the statute, we give serious consideration to an agency’s interpretation of a statute. Continental Cas. Co. v. Downs, 81 S.W.3d 803, 807 (Tex.2002); see Southwestern Bell Tel. Co., 92 S.W.3d at 441-42. This is particularly true when the statute concerns a complex subject matter. Railroad Comm’n v. Coppock, 215 S.W.3d 559, 563 (Tex.App.-Austin 2007, pet. denied); see also USA Waste Servs. of Houston, Inc. v. Strayhorn, 150 S.W.3d 491, 494 (Tex.App.-Austin 2004, pet. denied) (recognizing that legislature intends to provide agencies with centralized expertise in regulatory areas with large degree of latitude in accomplishing regulatory functions). However, courts do not defer to administrative interpretations regarding questions that are not within the agency’s expertise or that deal with nontechnical questions of law. USA Waste Servs., 150 S.W.3d at 494-95. Several issues also involve determinations regarding the Commission’s authority. As an agency, the Commission is a creation of the legislature and, therefore, “has no inherent authority.” Public Util. Comm’n v. City Pub. Serv. Bd., 53 S.W.3d 310, 316 (Tex.2001). For this reason, the Commission possesses only those powers “expressly conferred upon it.” Id. However, when conferring a power upon an agency, the legislature also “impliedly intends that the agency have whatever powers are reasonably necessary to fulfill its express functions or duties.” Id. But an agency may not “exercise what is effectively a new power, or a power contradictory to the statute, on the theory that such a power is expedient for administrative purposes.” Id. Finally, several of the issues question whether many of the Commission’s actions were adequately supported by the evidence presented. We review these types of questions under a substantial-evidence standard. Tex. Util.Code Ann. § 15.001 (West 2007) (stating that judicial review of agency action is under substantial-evidence standard); Tex. Gov’t Code Ann. § 2001.174 (West 2000) (allowing court to reverse agency determination if it is not supported by substantial evidence). Under this standard, we are prohibited from substituting our judgment for the Commission’s “as to the weight of the evidence on questions committed to agency discretion.” Cities of Abilene, San Angelo, & Vernon v. Public Util. Comm’n, 146 S.W.3d 742, 748 (Tex.App.-Austin 2004, no pet.) (citing Tex. Gov’t Code Ann. § 2001.174). In making this determination, we are not asked to verify whether “the agency reached the correct conclusion, but whether some reasonable basis exists in the record for the agency’s action.” Id. In fact, the evidence may actually preponderate against the Commission’s finding and be upheld as long as there is enough evidence to suggest that the Commission’s “determination was within the bounds of reasonableness.” Id. DISCUSSION The Commission’s Primary Market Valuation Market Valuation Before addressing the various parties’ arguments regarding the Commission’s primary market valuation, we will review the various methods by which a utility may calculate its stranded costs. The utilities code lists four primary market-based valuation methods and one alternative method for utilities to calculate the market value of generation assets — a necessary step for calculating stranded costs. The language of the statute places the burden of properly calculating the market value of the assets on the utility. Section 39.262 of the utilities code mandates that “for the purpose of finalizing the stranded costs estimate,” “the affiliated power generation company shall ” calculate the market value of the generation assets by using one of four methods: (1) the sale-of-assets method; (2) the stock-valuation method; (3) the partial-stock-valuation method; or (4) the exchange-of-assets method. Tex. Util.Code Ann. § 39.262(h) (emphasis added); 16 Tex. Admin. Code § 25.263(f)(1) (2007); see also Tex. Gov’t Code Ann. § 311.016 (West 2005) (explaining that when construing statutes, courts should interpret “shall” as imposing duty). The alternative method is found in subsection 39.262(i). See Tex. Util.Code Ann. § 39.262(f); 16 Tex. Admin. Code § 25.263(f)(2) (2007). Under this method, the market value of the generation assets is ascertained by performing an additional ECOM calculation using “updated company-specific inputs.” Tex. Util.Code Ann. § 39.262(i). Under the sale-of-assets method, the market value is determined by the “total net value realized from the sale” of the assets if they have been sold in a “bona fide third-party transaction under a competitive offering.” Id. § 39.262(h)(1). The exchange-of-assets method applies when generation assets have been transferred “in a bona fide third-party exchange transaction.” Id. § 39.262(h)(4). Under this method, the market value of the assets that were transferred may be determined by an independent appraisal of the assets. Id. If some or all of the generation assets have been transferred to “one or more affiliated or nonaffiliated corporations,” the market value of those transferred assets can be determined by using either the stock-valuation method or the partial-stock-valuation method. Both methods use the average closing price of the stocks of the corporation or corporations possessing the assets to determine the market value of those assets. Id. § 39.262(h)(2), (3). The Joint Applicants chose to employ the partial-stock-valuation method. A party may use this method when a utility or its affiliated power-generation company has transferred generation assets to a corporation and “at least 19 percent, but less than 51 percent, of the common stock” of the corporation “is spun off and sold to public investors through a national stock exchange.” Id. § 39.262(h)(3). Under this method, the market value is determined by the average daily closing price of the stock “over 30 consecutive trading days.” Id. The 30-day period is chosen by the Commission, but it must occur within 120 days of the date on which the affiliated utilities file their joint application to recover stranded costs. Id.; see id. § 39.262(c) (mandating joint filing). Because the amount of stock spun off under this method can range from 19% to 51%, it is possible that less than half of the corporation’s stock will be publicly traded and, therefore, that the corporation’s majority stockholders will have complete control over the actions of the corporation. The effect of this control might increase the value of the stock privately held, rendering the average closing price of the publicly-traded stock an inaccurate measure of the true value of the stock. For this reason, the utilities code authorizes the Commission to appoint a panel of experts to determine whether this effect, called a control premium, is present. Id. § 39.262(h)(3); Reliant I, 101 S.W.3d at 144 (explaining that “control premium is the additional value that a block of shares obtains by virtue of the fact that it carries with it the power to control the corporation”)- ⅛ other words, the panel determines the difference between the actual value of the stock and the amount that it is publicly traded for. If the panel determines that a control premium exists, the Commission shall adopt the panel’s determination of the actual value of the stock but cannot “increase the market value by a control premium greater than 10 percent.” Tex. Util.Code Ann. § 39.262(h)(3). The determination of the Commission “based on the finding of the panel conclusively establishes the value of the common stock.” Id. Over a year before the final true-up proceeding, CenterPoint distributed a little over 19% of Genco’s stock to CenterPoint’s shareholders. After distributing the stock, CenterPoint determined the market value of Genco’s generation assets by using the partial-stock-valuation method. By utilizing this method, CenterPoint determined that the market value for Genco’s generation assets was $2,907 billion. Because the majority of Genco’s stocks were owned by CenterPoint and not traded publicly, the Commission appointed a panel to determine if a control premium existed. See id. The panel determined that a control premium existed and that CenterPoint’s valuation did not accurately reflect the actual value of Genco’s stock. The panel determined that the actual value of the stock was approximately 17% higher than its trade value. See id. § 39.262(h)(3) (requiring Commission to adopt determination of panel but prohibiting it from increasing value of stock by more than 10%). Ultimately, however, the Commission concluded that the partial-stock-valuation method could not be employed because less than 19% of Genco’s stock had actually sold on a national stock exchange despite the fact that 19% had been distributed to CenterPoint’s stockholders. In an attempt to find an alternative method for determining market value, the Commission reviewed other estimates for Genco’s market value, including the report by the control-premium panel. After performing its own analysis, the Commission concluded that the market value of the assets was higher than the amount originally calculated by the Joint Applicants. Because of this, the Commission reduced the Joint Applicants’ stranded-cost recovery to an amount that was less than the amount that they originally requested. The district court affirmed the Commission’s use of an alternative method for estimating the value of the generation assets and its reduction to the Joint Applicants’ recovery. The Joint Applicants Failed to Satisfy the Requirements of the Parbial-Stockr-Valuation Method In their first issue on appeal, the Joint Applicants contend that the Commission erred when it concluded that the partial-stock-valuation method could not be employed. Under this method, the market value of generation assets is determined by using the average trading price of the stock of the corporation or corporations possessing the assets if “at least 19 percent, but less than 51 percent, of the common stock of each corporation is spun off and sold to public investors through a national stock exchange.” Tex. Util.Code Ann. § 39.262(h)(3) (emphasis added); see also Black’s Law Dictionary 974 (6th abridged ed.1991) (defining “spin-off’ as something that occurs when part of corporation’s assets and stocks are transferred to new corporation). In August 2002, CenterPoint transferred all of its generation assets to Genco. Six months later, CenterPoint distributed or spun off approximately 19% of Genco’s shares to CenterPoint shareholders. After the initial distribution, the stocks were listed on the New York Stock Exchange and were sold to public investors starting in January 2003. The stocks continued to be sold to public investors through the time of the true-up application in March 2004. See 16 Tex. Admin. Code § 25.263 (2007) (time for filing true-up application). Although CenterPoint did spin off 19% of Genco’s stock, not all of that stock was subsequently traded on a national stock exchange. For example, some of the distributed stock was placed into the retirement accounts of various CenterPoint employees and was not sold on a stock exchange. During the true-up proceeding, several employees testified that they received stocks from the spin-off and did not sell the stocks by the time of the proceeding. As a result, less than 19% of the stock actually changed ownership in the stock market. For this reason, the Commission concluded that the partial-stock-valuation method could not be used. The Joint Applicants aver that subsection 39.262(h)(3) does not require that all 19% of the spun-off stock be sold on a national stock exchange. See Tex. Util. Code Ann. § 39.262(h)(3). Rather, they assert that the requirements that stock (1) be spun off and (2) sold on a national stock exchange refer to two separate events. Stated differently, while the Joint Applicants acknowledge that at least 19% of the stock had to be spun off, they do not believe that all of the spun-off stock must subsequently be sold in a stock market. Rather, they assert that the “sold” requirement is satisfied as long as some of the stock was traded in a stock exchange. Similarly, they contend that the word “sold,” when read in the context of the statute, merely means that the stock must be offered for sale, not that it also be purchased, and refer to various definitions of the word “sell” to support this assertion. See, e.g., Webster’s New Collegiate Dictionary 1051 (1st ed.1973). The Joint Applicants also insist that interpreting the partial-stock-valuation method as requiring that all 19% of the distributed stock be sold in a stock exchange is tantamount to demanding an “unworkable and impossible requirement that defeats the entire purpose of the valuation statute.” Essentially, they argue that although market value is determined through average closing prices, many stock holders choose to retain ownership of their stock rather than sell it and that this retention plays a key role in establishing the true market value of stock. In other words, they argue that the rapid sale of stocks can lead to deflated stock prices but that stock retention helps to create a higher stock price by providing a stabilizing effect and by demonstrating that the stock is a desirable investment. Further, they assert that the benefit obtained through retention would cease to exist if all of the spun-off stock has to be sold prior to the true-up. Moreover, they insist that although not all 19% was sold, enough of the shares were sold and resold to establish an accurate market value. Specifically, they note that although 15.2 million shares were originally distributed, Genco stocks were traded 37.8 million times between January 2003 and March 2004. Finally, they assert that a rigid requirement that a utility not only spin off 19% of its stock but that 19% also be publicly traded would effectively require a utility to spin off more than 19% of stock in order to guarantee that at least 19% is traded, which they urge would lead to significant tax penalties. Specifically, they argue that CenterPoint and Genco would not have been able to file a joint tax return if more of Genco’s stock had been distributed. See 26 U.S.C.A. § 1504 (West 2002) (defining “affiliated corporation” as one in which parent corporation owns 80% of corporation’s stock). When it interpreted the relevant statutory language, the Commission determined that the phrase “sold ... through a national stock exchange,” as used in the statute, means that the stock must actually be traded through a national stock exchange (i.e. offered for sale and purchased) and not just offered for sale. From this, the Commission reasoned that at least 19% of the stock must be spun off and subsequently traded in a national stock exchange in order to satisfy the requirements of the statute. We believe that the Commission’s interpretation is correct for several reasons. First, the use of the word “and” without the insertion of a new subject in the phrase “spun off and sold” indicates that both phrases apply to the language immediately preceding them: “at least 19 percent, but less than 51 percent, of the common stock is.” See Tex. Util.Code Ann. § 39.262(h)(3). Explained another way, the statute requires that (1) at least 19% of the stock be spun off and (2) at least 19% of the stock be sold. Second, there are other definitions of the word “sold” that do not mean simply to offer for sale. For example, “sell” can also mean “to give up (property) to another for money or other valuable consideration.” Webster’s New Collegiate Dictionary 1051 (1st ed.1973). Keeping in mind that the word “sold” is the past tense of “sell,” plugging this definition into the statute leads to the conclusion that to satisfy the partial-stock-valuation requirements, at least 19% of the stock must have been purchased by public investors prior to the true-up proceeding. We believe that this construction of the statute more accurately reflects the legislative intent than the Joint Applicants’ interpretation. This construction comports with the use of the word “sold” in other provisions of the utilities code. For example, under the sale-of-assets method for determining market value, a utility may establish the market value of generation assets if the assets have been “sold.” Tex. Util.Code Ann. § 39.262(h)(1). When the word “sold” is read in the context of the remainder of the sentence, it becomes clear that “sold,” as used in this subsection, does not mean to offer for sale. The relevant portion of the provision provides as follows: “the total net value realized from the sale establishes the market value of the generation assets sold.” Id. (emphases added). This interpretation is also consistent with the emphasis placed on establishing an accurate market value apparent in the entire market-valuation subsection. Jones v. Fowler, 969 S.W.2d 429, 432 (Tex.1998) (providing that when construing statutes, courts should look to entire act). Each market valuation method listed in subsection 39.262(h) requires that certain minimum conditions be met before the utility may employ the method. Tex. Util.Code Ann. § 39.262(h). For example, a utility may employ the sale-of-assets method only if its generation assets are sold “in a bona fide third-party transaction under a competitive offering.” Id. § 39.262(h)(1). Similarly, the exchange-of-assets method may be employed only if the generation assets are transferred “in a bona fide third-party exchange transaction.” Id. § 39.262(h)(4). Moreover, under this method, the market value of the assets may be determined by offering the assets for sale if the offer is made in a way guaranteeing “broad public notice of the offer and a reasonable opportunity for other parties to bid on the asset.” Id. These requirements are designed to ensure that an accurate market value for the generation assets is calculated in order to comply with the overriding mandate present throughout the statutory scheme: that a utility be allowed to recover but not ov-errecover its stranded costs. See, e.g., id. §§ 39.252, .262(a). Given the strong legislative directive that market calculations be based on real market forces, it seems logical to conclude that the legislature fully intended that a large portion of the company’s stock — at least 19% — actually trade on a public stock exchange to ensure that an accurate market value is obtained. See id. § 39.251(4) (defining “market value” as value of assets if they had been bought and sold in “bona fide third-party transaction” or “on the open market”). Moreover, the Joint Applicants’ interpretation would lead to unreasonable results. See Lowe v. Rivera, 60 S.W.3d 366, 369 (Tex.App.-Dallas 2001, no pet.) (stating that statutes should not be construed in manner that leads to absurd results). Under their interpretation, the statute would be satisfied if 19% of the stock was spun off and offered for sale on a public stock exchange but only a few stocks actually sold through the exchange. Essentially, under the Joint Applicants’ interpretation, the market value from the sale of a handful of stocks — or even one share — could be used as a valid basis for determining stranded costs. This does not comport with the utilities code’s insistence on utilizing, to the extent possible, actual competitive market forces and reasonable business practices to determine market value. We also disagree with the Joint Applicants’ assertion that it would be impossible to comply with the requirements of the partial stock valuation. Although it may be difficult to have at least 19% of the spun-off stock actually sell on a stock exchange if only 19% is spun off, utilities can attempt to assure compliance with the statute by spinning off more than the minimum amount required. In fact, under the partial-stock-valuation method, a utility may spin off between 19 and 51% of the stock. Tex. Util.Code Ann. § 39.262(h)(3). By spinning off more than 19%, the Joint Applicants could have obtained whatever benefit might arise from certain stock holders retaining their stock and still complied with the statute by selling 19% of the stock on a national stock exchange. Furthermore, spinning off more than 19% is not the only way the statute could have been satisfied. The Commission argues that the Joint Applicants could also have chosen to comply with the statute by distributing the stock through an initial public offering. See Walden v. Affiliated Computer Servs., 97 S.W.3d 303, 327 (Tex.App.-Houston [14th Dist.] 2003, pet. denied) (explaining that initial public offering “is the commonly used term for the first offering of equity securities of an issuer to the public pursuant to a registration statement”). Under this method, public investors would purchase Genco stock from an underwriter shortly after the initial offering is made. Because the sale would involve a transfer to public investors without first going through CenterPoint shareholders, the Commission contends that the partial-stock-valuation requirements would be met as long as more than 19% of the stock was purchased in the initial offer. In other words, no more than the desired amount of stock would need to be distributed because the stock is sold directly to public investors. Although the Joint Applicants acknowledge that an initial public offering would have satisfied the necessary requirements, they insist that the market conditions during 2003 would not have allowed a successful public offering. Essentially, they argue that an offering of 15.2 million newly issued stocks would have deflated the value of the stock. Even if the value of the stock would have been temporarily lowered, the Joint Applicants appear to concede that the value would have stabilized over time at a value similar to that found by spinning off the stock first and then offering it for sale on a stock exchange. This undercuts their assertion that it would have been impossible to satisfy the partial stock valuation. It also seems to indicate that they could have satisfied the partial stock valuation without having to distribute significantly more than 19% of Genco’s stock, thereby obviating their tax concerns. In addition, the fact that the utilities code allows the partial stock valuation to be used for spinoffs of amounts much larger than 19% of a utility’s stock indicates that the partial-stock valuation provision was not enacted solely to allow affiliated utilities to file joint tax returns. Moreover, we must assume that when the legislature chose the range of values that would satisfy the spinoff requirement of the partial stock valuation, it was aware that utilities might incur negative tax consequences if they were required to distribute more than 19% of the stock. See Tex. Util.Code Ann. § 39.262(h)(3). As a result, we cannot conclude that the legislature crafted the spinoff requirements so as to prevent potential negative tax consequences for the utilities who complied. From the numerous methods for calculating market value described in the utilities code, we can infer that it was the legislature’s intent to afford the utilities discretion to consider their unique circumstances and the relevant market conditions when deciding which method to use. It was within the utilities’ discretion to consider and trade off the relative benefits and costs (e.g. taxes) when selecting a valuation method. This scheme does not, however, enable utilities to partially comply with the mandatory requirements in order to avoid a potential business cost. We must also assume that when the legislature enacted this statute, it was aware of the possibility that the recipients of a stock spin-off may hold onto their stocks for an extended period of time and that stock that is sold on a stock exchange might be resold prior to the true-up proceeding. In light of this, the legislature still required a utility to spin off and sell at least 19% of the relevant stock to comply with the partial-stock valuation method. For this reason, we also disagree with the Joint Applicants’ assertion that the subsequent reselling of the Genco stock in the stock market satisfied the legislative goal of establishing an accurate market value. For all the reasons previously given, the Commission’s interpretation requiring that a minimum proportion of a utility’s total stock be sold in the stock market in order to accurately determine market value is correct and consistent with the relevant statutory language. The Joint Applicants failed to comply with this minimum requirement. Accordingly, the Commission correctly determined that the partial-stock method could not be used to calculate the market value of the generation assets. The Commission Had the Authority to Consider Other Valuation Methods The Utility Counsel and the Customers agree that the requirements of the partial-stock method were not complied with but criticize the Commission’s decision to estimate the market value of the generation assets by a method not specifically listed in the utilities code. First, the Customers assert that the Commission should not have allowed the Joint Applicants to recover any stranded costs because they failed to meet their burden of establishing a viable market value. Essentially, the Customers assert that the burden of proving stranded costs is on the utilities and insist that if a utility fails to satisfy this burden, it should not be awarded stranded costs. See Tex. Util.Code Ann. §§ 39.252 (stating that utility is allowed to recover its “verifiable” stranded costs), .262(h) (requiring utility to “calculate its stranded costs”); see also id. § 39.003 (establishing that in contested cases, burden of proof “is on the incumbent electric utility”). However, this assertion ignores the clear legislative mandate that utilities be allowed to recover their stranded costs. See, e.g., id. §§ 39.001(b)(2) (“in public interest to ... allow utilities ... to recover” stranded costs), .252 (“utility is allowed to recover all of its net, verifiable, nonmitiga-ble stranded costs”). In fact, an entire subchapter of the utilities code is dedicated to describing the process of stranded-cost recovery. See id. §§ 39.251-265 (entitled “Recovery of Stranded Costs Through Competition Transition Charge”). Although the Utility Counsel and the Customers correctly point out that the utilities code places the burden of determining market value on the utilities, id. § 39.262(h), nothing in the code indicates that the failure of a utility to satisfy one of the market-valuation requirements should result in an automatic denial of the right to recover any stranded costs. Construing the utilities code in this manner would run afoul of the statutory scheme governing the transition to a competitive energy market and ensuring that a former regulated utility not be disadvantaged through the transition. In the alternative, the Customers argue and the Utility Counsel agrees that after concluding that the partial-stock method could not be utilized, the Commission should have used one of the other permissible valuation methods to calculate market value. See id. § 39.262(h), (i). We disagree. After considering the possibility of utilizing one of the other listed methods, the Commission concluded that none of the other methods listed in the utilities code could have been employed in this case because their requirements were not met. The stock-valuation method requires that more than 51% of the common stock of a transferee corporation be “spun off and sold to public investors.” Id. § 39.262(h)(2). However, as discussed earlier, less than 19% of Genco’s stock was actually spun off and sold. The exchange-of-assets method could also not be employed because Genco did not transfer any of its generation assets “in a bona fide third-party exchange transaction.” See id. § 39.262(h)(4). Similarly, the Commission also concluded that the two methods proposed by the Customers and the Utility Counsel — the sale-of-assets method and the alternative method found in subsection 39.262(i)— could not be employed. Subsection (i) reads, in relevant part, as follows: Unless an electric utility or its affiliated power generation company combines all of its remaining generation assets into one or more transferee corporations as described in [the stock-valuation method and partial-stock-valuation method], the electric utility shall quantify its stranded costs for nuclear assets using the ECOM method .... using updated company-specific inputs.... Id. § 39.262(i) (emphases added). The transfer of assets is a necessary component of the market valuations obtained by using either the stock-valuation method or the partial-stock-valuation method. Although the Joint Applicants did not satisfy the other requirements necessary for these two methods, namely the sale of a sufficient number of stocks in a public stock exchange, they did transfer all their generation assets to Genco. In light of this, the Commission concluded that the ECOM model could not be used to estimate market value. This determination is reasonable and consistent with the relevant statutory language, and we agree that is what the legislature intended. The sale-of-assets provision reads, in relevant part, as follows: If, at any time after December 31, 1999, an electric utility ... has sold some or all of its generation assets ... in a bona fide third-party transaction under a competitive offering, the total net value realized from the sale establishes the market value of the generation assets sold. Id. § 39.262(h)(1) (emphasis added). The Customers argue that in July 2004 Center-Point entered into a binding agreement to sell its generation assets to a third party during the true-up proceeding and that the Commission should have used the amount offered to ascertain the value of the generation assets because the offered price was in the record before the Commission. Further, in light of the statutory language stating that the sale of assets “at any time after December 31, 1999,” may be used to establish market value, see id., they ask this Court to take judicial notice of the fact that Genco was actually sold for the amount offered after the Commission issued its final order or, alternatively, to remand the case in order for the Commission to take notice of the completed sale. In related contentions, the Utility Counsel argues that the failure of the Commission to use the sale price of Genco to establish market value allowed the Joint Applicants to overrecover for stranded costs in violation of the legislative prohibition. See id. § 39.262(a). Essentially, it argues that the Commission’s market value estimate was much lower than the sale price, which allowed the Joint Applicants to recover more for stranded costs than they would have been allowed to if the sale-of-assets method had been employed. The sale-of-assets method requires that the generation assets be “sold” prior to the stranded-cost reconciliation. Id. § 39.262(h)(1). Although subsection (h)(1) does refer to a sale occurring “any time after December 31, 1999,” the Commission concluded that the word “sold,” meaning a completed act, necessarily limits consideration of a sale for market-valuation purposes to sales occurring before the true-up reconciliation. See id. Although the offer was made before the Commission issued its final order, the sale was not finalized until after the true-up proceeding, and therefore, the Commission concluded that any attempt to use the subsequent sale of Genco as the sole basis for determining market value would be improper and would be contrary to the provisions of the