concurring and dissenting.
The Public Utility Commission issued a final order in the true-up proceeding to finalize “stranded costs” and other true-up balances for the AEP Texas Central Company and CPL Retail Energy, L.P. (collectively “TCC”). The district court affirmed the Commission’s final order in most respects, but reversed on three issues. For the reasons discussed below, I would affirm the district court’s judgment in part, reverse in part, and remand this cause to the Commission for further proceedings.
I. Factual and Procedural Background
In 1999, the legislature determined it was in the public interest to restructure and partially deregulate the Texas retail *277electric power industry. See generally Tex. UtiLCode Ann. § 39.001 (West 2007). To accomplish this mandate, the legislature enacted Senate Bill 7 (“SB 7”), which amended the Public Utility Regulatory Act (“PURA”).1 See Act of May 27, 1999, 76th Leg., R.S., ch. 405, 1999 Tex. Gen. Laws 2543 (now codified in Chapter 39 of the PURA, Tex. UtiLCode Ann. §§ 39.001-.910 (West 2007)); see also CenterPoint Energy Houston Electric, LLC v. Gulf Coast Coalition of Cities, 252 S.W.3d 1, 7-12 (Tex.App.-Austin 2008, no pet. h.) (op. on reh’g) (describing statutory framework for transition to competitive retail electric market) (hereafter “CenterPoint ”). SB 7 required each integrated electric utility to separate its business activities into three separate units — a power generation company, a transmission and distribution utility, and a retail electric provider. See Tex. Util. Code Ann. § 39.051 (West 2007).
As part of the transition from regulation to retail competition, the legislature authorized each electric utility to recover “all of its net, verifiable, nonmitigable stranded costs incurred in purchasing power and providing electric generation service.” Tex. UtiLCode Ann. § 39.252(a) (West 2007). The term “stranded costs” is defined in section 39.251 of the PURA,2 but generally speaking, stranded costs represent prudently incurred expenditures made by the utilities in a regulated environment — previously recoverable over time through regulated rates paid by consumers — that have become unrecoverable in a competitive market. See Reliant Energy, Inc. v. Public Util. Comm’n, 101 S.W.3d 129, 132 (TexApp.-Austin 2003) (hereafter “Reliant I ”), rev’d in paH sub nom. CenterPoint Energy, Inc. v. Public Util. Comm’n, 143 S.W.3d 81 (Tex.2004) (op. on reh’g). Recovery of stranded costs is one of the final steps in the transition from traditional cost-of-service regulation to retail competition.
In addition to the recovery of stranded costs, the legislature’s deregulation plan required the Commission to determine each electric utility’s final fuel balance and capacity auction true-up award. See Tex. UtiLCode Ann. §§ 39.201, .202(c), .262(d) (West 2007). Once determined by the Commission, the net sum of the final fuel balance and the capacity auction true-up award would result in a credit or bill from the affiliated power generation company to the transmission and distribution utility. See id. § 39.262(d).
To recover its stranded costs and finalize its other true-up balances, TCC filed an application with the Commission seeking a total true-up balance of $2,406,271,176, including interest through September 2005. This amount included a requested capacity auction true-up award of $482,664,890, less TCC’s final fuel balance of $176,698,379. Several consumer groups intervened in the proceedings before the Commission to *278challenge TCC’s requested recovery. Among the intervenors were the State of Texas, the Office of Public Utility Counsel (OPC), the Texas Industrial Energy Consumers (TIEC), the Cities served by TCC (Cities),3 the Alliance for Valley Healthcare (AVH), the Alliance for Retail Markets, the Brownsville Public Utility Board, the Commercial Customers Group (CCG), Occidental Power Marketing, Reliant Energy, Inc., and the Texas Cotton Ginners’ Association. On review of TCC’s application, the Commission made several adjustments to the amounts requested by TCC. These adjustments related to TCC’s failure to use commercially reasonable means to mitigate potential stranded costs in relation to the sale of TCC’s share of the South Texas Nuclear Project (STP), the bundling of certain TCC gas plants as part of the sale of the Coleto Creek Coal Plant, disallowances to TCC’s capacity auction true-up award, and other items. In sum, the Commission awarded TCC a total recovery of $1,475,933,779.
TCC and six intervenors sought judicial review of the Commission’s final order in district court. See Tex. UtiLCode Ann. §§ 15.001, 39.262(j) (West 2007); Tex. Gov’t Code Ann. §§ 2001.171, .176 (West 2000). The district court reversed the Commission’s order on three issues. The district court held that the Commission erred by making adjustments to the net book value of TCC’s generation assets, by excluding the testimony and report offered by an expert witness, and by applying an interest rate specified in a rule that the supreme court had previously invalidated. TCC and six of the intervenors4 have filed separate appeals challenging the district court’s judgment, and the Commission has filed a cross-appeal.
II. Discussion
On appeal, the parties urge this Court to reverse the district court’s judgment on various grounds. The Commission urges us to reverse those portions of the district court’s judgment which reversed the Commission’s final order. Specifically, the Commission urges that the district court erred in its findings that the Commission was prohibited from making adjustments to the net book value of TCC’s generation assets, that the interest rate used by the Commission was invalid, and that the Commission erred in excluding the testimony and report of an expert witness. TCC urges us to reverse portions of the district court’s judgment on the grounds that the district court erred in upholding the Commission’s disallowance and reduction to TCC’s capacity auction true-up award, and in affirming the Commission’s treatment of Accumulated Deferred Investment Tax Credits, Excess Deferred Federal Income Taxes, and excess mitigation credits. In addition, like the Commis*279sion, TCC urges that the district court erred in reversing the Commission’s use of the interest rate specified in the Commission’s true-up rule.
The six intervenors likewise assert various challenges to the district court’s judgment. The Joint Intervenors argue that the district court erred in reversing the Commission’s adjustments to the net book value of TCC’s generation assets and in reversing the Commission’s exclusion of the testimony and report of an expert witness. The Joint Intervenors further argue that the district court erred in affirming the Commission’s adjustments to the net book value of the STP and the Coleto Creek Coal Plant, as well as the Commission’s inclusion of construction work in progress (CWIP) in its calculation of net book value. The Joint Intervenors also argue that the district court erred in affirming the Commission’s failure to use the ECOM, or excess-cost-over-market, method to value TCC’s nuclear generation assets and in affirming the Commission’s determination that TCC’s auction of its share of the STP met statutory requirements. Finally, the Joint Intervenors argue that the Commission erred in allowing TCC to recover excess mitigation credits paid to retail electric providers, allowing TCC to recover interest on its capacity auction true-up award, and by refusing to reduce stranded costs to account for certain profits achieved by TCC from the sale of its generation assets.
In addition to the claims raised by the Joint Intervenors, OPC/CCG also raise three challenges to the district court’s judgment affirming the Commission’s final order. First, OPC/CCG argue that the Commission violated the PURA by allowing TCC to recover stranded costs for generation assets that were not “uneconomic.” Like the Joint Intervenors, OPC/ CCG also allege that the Commission erred in its failure to use the ECOM method to value TCC’s nuclear generation assets. Finally, OPC argues that the Commission erred in failing to reduce TCC’s stranded cost recovery by the total amount of TCC’s excess earnings for 1999-2001.
A. Standard of Review
This Court reviews the Commission’s final order under the substantial evidence rule. See Tex. UtiLCode Ann. § 15.001; Tex. Gov’t Code Ann. § 2001.174 (West 2000). Under the substantial evidence rule, we give significant deference to the agency in its field of expertise. Railroad Comm’n v. Torch Operating Co., 912 S.W.2d 790, 792 (Tex.1995); Texas Health Facilities Comm’n v. Charter Medical-Dallas, Inc., 665 S.W.2d 446, 452 (Tex. 1984). We presume that the agency’s order is valid and that its findings, inferences, conclusions, and decisions are supported by substantial evidence. City of El Paso v. Public Util. Comm’n, 883 S.W.2d 179, 185 (Tex.1994); Charter Medical, 665 S.W.2d at 452. The complaining party has the burden to overcome this presumption. City of El Paso, 883 S.W.2d at 185; Ham-mack v. Public Util. Comm’n, 131 S.W.3d 713, 725 (Tex.App.-Austin 2004, pet. denied). In conducting a substantial evidence review, we evaluate the entire record to determine whether the evidence as a whole is such that reasonable minds could have reached the conclusion the agency must have reached in order to take the disputed action. Texas State Bd. of Dental Exam’rs v. Sizemore, 759 S.W.2d 114, 116 (Tex.1988); Suburban Util. Corp. v. Public Util. Comm’n, 652 S.W.2d 358, 364 (Tex.1983). We may not substitute our judgment for that of the agency on the weight of the evidence on questions committed to the agency’s discretion. Tex. Gov’t Code Ann. § 2001.174; Charter Medical, 665 S.W.2d at 452; H.G. Sledge, Inc. v. Prospective Inv. & Trading Co., *280Ltd., 36 S.W.3d 597, 602 (Tex.App.-Austin 2000, pet. denied).
Under a substantial evidence review, the issue for the reviewing court is not whether we believe the agency’s decision was correct, but whether the record demonstrates some reasonable basis for the agency’s action. Charter Medical, 665 S.W.2d at 452; Central Power & Light Co. v. Public Util. Comm’n, 36 S.W.3d 547, 561 (Tex.App.-Austin 2000, pet. denied). The evidence in the record may preponderate against the decision of the agency and nevertheless amount to substantial evidence. Charter Medical, 665 S.W.2d at 452; Meier Infiniti v. Motor Vehicle Bd., 918 S.W.2d 95, 98 (TexApp.-Austin 1996, writ denied). We will sustain the agency’s order if the evidence is such that reasonable minds could have reached the conclusion that the agency must have reached in order to justify its action. Charter Medical, 665 S.W.2d at 453; Suburban Util. Corp., 652 S.W.2d at 364. We must uphold the agency’s order unless the agency’s decision is not reasonably supported by substantial evidence, in violation of a constitutional or statutory provision, in excess of the agency’s statutory authority, made through unlawful procedure, affected by other error of law, arbitrary or capricious, or characterized by an abuse of discretion. See Tex. Gov’t Code Ann. § 2001.174(2)(A)-(F).
Several issues raised by the parties involve questions of statutory construction, which we review de novo. See, e.g., City of San Antonio v. City ofBoeme, 111 S.W.3d 22, 25 (Tex.2003) (appellate courts review matters of statutory construction de novo); In re Humphreys, 880 S.W.2d 402, 404 (Tex.1994) (questions of law are always subject to de novo review). When construing a statute, our primary goal is to determine and give effect to the legislature’s intent. City of San Antonio, 111 S.W.3d at 25. To determine legislative intent, we look to the statute as a whole, as opposed to isolated provisions. State v. Gonzalez, 82 S.W.3d 322, 327 (Tex.2002). We begin with the plain language of the statute at issue and apply its common meaning. City of San Antonio, 111 S.W.3d at 25. Where the statutory text is unambiguous, we adopt a construction supported by the statute’s plain language, unless that construction would lead to an absurd result. Fleming Foods of Tex., Inc. v. Rylander, 6 S.W.3d 278, 284 (Tex.1999). We give serious consideration to an agency’s interpretation of the statutes it is charged with enforcing, so long as that interpretation is reasonable and consistent with the statutory language. Tarrant Appraisal Dist. v. Moore, 845 S.W.2d 820, 823 (Tex.1993); Steering Comms. for the Cities Served by TXU Elec. v. PUC, 42 S.W.3d 296, 300 (Tex.App.-Austin 2001, no pet.). This is particularly true when the statute involves a complex subject matter. Steering Comms. for Cities, 42 S.W.3d at 300. Courts, however, “do not defer to administrative interpretation in regard to questions which do not he within administrative expertise, or deal with a nontechnical question of law.” Rylander v. Fisher Controls Int’l, Inc., 45 S.W.3d 291, 302 (TexApp.-Austin 2001, no pet.).
Additionally, several issues raised by the parties challenge the Commission’s authority. The Public Utility Commission “is a creature of the Legislature and has no inherent authority.” Public Util. Comm’n v. GTE-SW, Inc., 901 S.W.2d 401, 407 (Tex.1995). Like other state administrative agencies, the Commission “has only those powers that the Legislature expressly confers upon it” and “any implied powers that are necessary to carry out the express responsibilities given to it by the Legislature.” Public Util. Comm’n v. City Pub. Serv. Bd., 53 S.W.3d 310, 316 (Tex. 2001). It is not enough that the power *281claimed by the Commission be reasonably useful to the Commission in discharging its duties; the power must be either expressly conferred or necessarily implied by statute. The 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.
B. The Statutory Scheme
Before addressing the issues raised by the parties, it is helpful to review the statutory scheme allowing for the recovery of stranded costs and other true-up balances.
1. Recovery of stranded costs
When the legislature mandated the transition from traditional cost-of-service regulation to retail competition, it recognized that many utility companies had made very large investments to build power generation plants, and that while the costs of these power plants might be recoverable from ratepayers in a regulated environment, these same costs “might well become uneconomic and thus unrecoverable in a competitive, deregulated power market.” CenterPoint Energy, 143 S.W.3d at 82; see also City of Corpus Christi v. Public Util. Comm’n, 51 S.W.3d 231, 237-38 (Tex.2001). The legislature called these uneconomic assets “stranded costs.” Cen-terPoint Energy, 143 S.W.3d at 82; see also Tex. Util.Code Ann. §§ 39.001(b)(2), .251(7) (West 2007). As part of the transition to retail competition, the legislature made an express finding that it was in the public interest to allow utility companies to recover their stranded costs. See Tex. Util.Code Ann. § 39.001(b)(2) (West 2007). The legislature thus “set forth a comprehensive scheme for estimating, finalizing, and recovering those costs.” CenterPoint Energy, 143 S.W.3d at 83; see Tex. Util. Code Ann. §§ 39.201, .251-.254, ,256-.265, .301-313 (West 2007).
The legislature provided a mechanism in section 39.262 of the PURA for each utility to recover its stranded costs. See Tex. Util-Code Ann. § 39.262. This mechanism requires each transmission and distribution utility, its affiliated retail electric provider, and its affiliated power generation company to jointly file an application with the Commission to finalize stranded costs. Id. § 39.262(c). The legislature provided a formula to calculate a utility’s stranded costs. Id. § 39.251(7). Under this formula, a utility’s stranded costs equal the excess amount of the net book value of generation assets minus the market value of those assets — i.e., SC = NBV - MV". Id.
Each utility has the burden of quantifying its stranded costs using one of four statutory methods. Id. § 39.262(h). These methods include sale-of-assets, stock valuation, partial stock valuation, or exchange of assets. Id. § 39.262(h)(l)-(4). TCC chose to establish its stranded costs using the sale-of-assets method. Under this method, TCC was required to demonstrate that it sold its generating assets in a “bona fide third-party transaction under a competitive offering.” Id. § 39.262(h)(1). Once this showing has been made, the total net value realized from the sale will establish the market value of the generation assets sold. Id.
The legislature recognized, however, that during the transition from regulation to retail competition, utilities would not necessarily have a business incentive to reduce their potential stranded costs. Accordingly, the legislature provided a statutory incentive — the legislature required each utility to “pursue commercially reasonable means to reduce its potential stranded costs, including good faith attempts to renegotiate above-cost fuel and purchased power contracts or the exercise of normal business practices to protect the *282value of its assets.” Id. § 39.252(d). The legislature also required the Commission to consider each utility’s efforts to mitigate potential stranded costs when determining the utility’s final stranded cost balance. Id. In addition, the legislature required the Commission to ensure that each utility does not overrecover stranded costs. Id. § 39.262(a); see also CenterPoint Energy, 143 S.W.3d at 98-99.
2. Determination of non-stranded-cost true-ups
As part of the transition to deregulation, the legislature also required the Commission to conduct additional true-up proceedings to calculate each utility’s final fuel balance and capacity auction true-up award. Tex. Util.Code Ann. § 39.262(d). The parties do not dispute the Commission’s determination of TCC’s final fuel balance, but TCC and the Joint Interve-nors both challenge the Commission’s determination of TCC’s capacity auction true-up award. Following is a brief description of the purpose of the capacity auction and its statutory requirements.
To ensure the availability of power in the emerging competitive market, the legislature required power generation companies to auction entitlements to at least 15% of their installed generation capacity beginning 60 days prior to the start of customer choice on January 1, 2002. See Tex. UtiLCode Ann. §§ 39.153, .156 (West 2007); Reliant I, 101 S.W.3d at 137. The utilities’ obligation to auction generation capacity continued until the earlier of 60 months (five years) after January 1, 2002, or the date on which the Commission determines that 40 percent or more of the electric power consumed by residential and small commercial customers within the affiliated transmission and distribution company’s service area before the onset of customer choice is provided by nonaffiliat-ed retail electric providers. See Tex. Util. Code Ann. § 39.153(b). The legislature directed the Commission to adopt rules defining the scope of the capacity entitlements to be auctioned and the procedure for the auction of those entitlements. Id. § 39.153(e)-(g); see also 16 Tex. Admin. Code § 25.381 (West 2007) (PUC Capacity Auction Rule).5
The legislature was concerned that consumers and generation companies could be harmed by “distortions and fluctuations in the market price of power during the first two years of deregulation.” See Center-Point Energy, 143 S.W.3d at 96. To alleviate this concern, the legislature designed the capacity auction true-up proceeding. Id. The purpose of the capacity auction true-up proceeding was to provide those companies forced to participate in the required capacity auctions with a guaranteed return on the sales of power, while at the same time ensuring that consumers would not be harmed by fluctuating prices and market instability. Id. The “guaranteed return,” or capacity auction true-up award, represents the difference between the price of power projected by the Commission in the 2001 ECOM model6 and the *283price of power obtained through the auctions. See Tex. UtiLCode Ann. § 39.262(d)(2). The Commission was charged with calculating this return in the capacity auction true-up proceeding. Id. § 39.262(d). As a result of the capacity auction true-up proceeding, consumers and generation companies were guaranteed that generation companies “will receive no more and no less” than a set margin for power sales predetermined by the Commission in 2001 when the ECOM model was run in compliance with section 39.201. CenterPoint Energy, 143 S.W.3d at 96.
With this statutory scheme in mind, I examine the parties’ complaints regarding the Commission’s determination of TCC’s stranded costs and other true-up balances.
C. Stranded Cost True-up
The legislature defined stranded costs as the positive excess of the net book value of a utility’s generation assets over the market value of those assets. Tex. Util. Code Ann. § 39.251(7). The parties raise several challenges to the Commission’s determinations regarding both the market value and net book value of TCC’s generation assets.
1. Determination of market value
The Joint Intervenors and OPC/CCG challenge the Commission’s determinations regarding the market value of TCC’s interest in the STP. As a preliminary matter, OPC argues that the Commission erred in allowing TCC to recover stranded costs for the STP because there was no evidence that the STP was an “uneconomic asset.”7 Assuming the Commission properly allowed TCC to recover stranded costs for the STP, the Joint Intervenors and OPC/ CCG argue that the Commission erred in its market valuation of the STP because it failed to use the ECOM method prescribed in section 39.262(i) for valuing nuclear assets. The Joint Intervenors also argue in the alternative that the Commission erred in its conclusion that the STP auction was a bona fide third-party transaction within the meaning of section 39.262(h).
(A) Was the STP an “uneconomic asset”?
OPC argues that TCC cannot recover stranded costs for its share of the STP because the STP was not an “uneconomic asset.” In support of this argument, OPC cites to section 39.001(b)(2) of the PURA. In that section, the legislature found that it was in the public interest to “allow utilities with uneconomic generation related assets and purchased power contracts to recover the reasonable excess costs over market [value] of those assets and purchased power contracts.” Tex. UtiLCode Ann. § 39.001(b)(2) (emphasis added). Based on this language, OPC argues that TCC was only permitted to recover stranded costs for its demonstrated “uneconomic generation related assets.” See id. As a result, OPC argues that the Commission erred in allowing TCC to recover stranded costs for the STP because “the undisputed evidence in the record shows that net margins over the life of the STP far exceed the net book value of the plant.”
OPC’s argument involves a matter of statutory construction. When construing a statute, our primary goal is to determine and give effect to the legislature’s intent. City of San Antonio, 111 S.W.3d at 25. To determine legislative intent, we look to the statute as a whole, as opposed to isolated provisions. Gonzalez, 82 S.W.3d at *284327. In addition to the language relied on by OPC in section 39.001(b)(2), section 39.251(7) of the PURA defines the term “stranded costs” as “the positive excess of the net book value of generation assets over the market value of the assets, taking into account all of the electric utility’s generation assets.” Tex. UtiLCode Ann. § 39.251(7) (emphasis added). Section 39.251(3) defines the term “generation assets” as “all assets associated with the production of electricity, including generation plants, electrical interconnections of the generation plant to the transmission system, fuel contracts, fuel transportation contracts, water contracts, lands, surface or subsurface water rights, emissions related allowances, and gas pipeline interconnections.” Id. § 39.251(3) (emphasis added). Section 39.252(a) provides that “[a]n electric utility is allowed to recover all of its net, verifiable, nonmitigable stranded costs incurred in purchasing power and providing electric generation service.” Id. § 39.252(a) (emphasis added).
When read together, these provisions allow a utility to recover all of its stranded costs for generation related assets when the net book value exceeds the market value of those assets. See id. §§ 39.251(3), (7), .252(a). None of these provisions includes the phrase “uneconomic assets,” nor does the legislature otherwise limit the recovery of stranded costs to only “uneconomic assets.” In section 39.252(a), the legislature expressly provided that a utility may recover “all of its net, verifiable, non-mitigable stranded costs.” See id. § 39.252(a) (emphasis added). There is no indication in other provisions of the PURA that the legislature intended to limit a utility’s recovery of stranded costs by its use of the phrase “uneconomic assets” in section 39.001(b)(2). It does not follow then that the legislature intended to limit recovery of stranded costs in the manner urged by OPC. Because the STP is a generation asset and all generation assets must be valued to determine TCC’s stranded costs, I conclude that the Commission properly determined that the STP was eligible for inclusion in the calculation of TCC’s stranded costs. I would overrule OPC’s issue one.
(B) Was the Commission required to use the ECOM method to determine the market value of the STP ?
OPC/CCG and the Joint Intervenors argue that the Commission was required to use the ECOM method in section 39.262(f) when determining the market value of the STP. OPC/CCG argue that TCC must use the ECOM method to value its nuclear assets absent a stock valuation, and that the sale-of-assets method can never be used for that purpose. In contrast, the Joint Intervenors argue that because a utility has a duty to mitigate its stranded costs, the ECOM method in section 39.262® must be used to value nuclear assets unless another method is shown to produce lower stranded costs.8
These arguments also involve a matter of statutory construction. As previously stated, the primary objective when construing a statute is to ascertain and give effect to the legislature’s intent. City of San Antonio, 111 S.W.3d at 25. To determine legislative intent, courts look to the statute as a whole, as opposed to isolated provisions. Gonzalez, 82 S.W.3d at 327. A reviewing court gives deference to the Commission’s interpretation of a statute it is charged with enforcing as long as that *285interpretation is reasonable and consistent with the plain language of the statute. Tarrant Appraisal Dist, 845 S.W.2d at 823; Steering Comms. for Cities, 42 S.W.3d at 300. Accordingly, I begin with the plain language of the PURA.
In relevant part, section 39.262(h) provides, “Except as provided in Subsection (i) ... the affiliated power generation company shall quantify its stranded costs using one or more of the following methods.Tex. Util.Code Ann. § 39.262(h). Section 39.262(h) goes on to specify four methods a utility may use to quantify its stranded costs: sale-of-assets, stock valuation, partial stock valuation, and exchange-of-assets. Id. § 39.262(h)(l)-(4). In addition, section 39.262(i) provides, “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 Subsections (h)(2) and (h)(3), the electric utility shall quantify its stranded costs for nuclear assets using the ECOM method....” Id. § 39.262(i).
The Commission determined that, when read together, these statutes allow TCC to quantify its stranded costs for nuclear assets using any of the methods in section 39.262(h), including the sale-of-assets method. The Commission adopted PUC Substantive Rule 25.264, which confirms its interpretation allowing TCC to quantify its stranded costs for nuclear assets using the methods established in section 39.262(h). Rule 25.264 states:
The market value of an affiliated power generation company’s nuclear assets may be established by compliance with any of the four methods of quantification specified in Public Utility Regulatory Act (PURA) § 39.262(h) and related requirements specified in § 25.263 of this title (relating to True-up Proceeding). If the electric utility or its affiliated power generation company values some of its assets using the sale of assets or an exchange of assets, any remaining assets shall be combined in one or more transferee corporations as described in PURA § 39.262(h)(2) and (3) for purposes of determining their market value, or the electric utility or its affiliated power generation company shall quantify its stranded costs for remaining nuclear assets using the “excess costs over market” or ECOM method.
16 Tex. Admin. Code § 25.264 (2006) (Pub. Util.Comm’n). In its order adopting Rule 25.264, the Commission explained:
This rule is necessary to firmly establish the methods that may be employed to determine the stranded cost of nuclear power generation assets. In addition, the rule is needed to serve the public interest and legislative policy stating that utilities with uneconomic generation-related assets should be allowed to recover the reasonable excess costs over market value of those assets. In order to assure that the market value of nuclear generation assets is properly quantified in a manner that reduces, to the extent possible, the amount of excess costs over market value for those assets, the rule clarifies that a public utility and its affiliated companies may use any of the valuation methods specified in PURA § 39.262(h) and (i) to quantify the market value of nuclear generation assets.
Tex. Pub. Util. Comm’n, Rulemaking Proceeding Concerning Quantification of Stranded Costs of Nuclear Generation Assets, Substantive Rule § 25.264, Docket No. 27464, Order at 2 (May 23, 2003).
The plain language of section 39.262(h) allows a utility or its affiliates to quantify stranded costs “using one or more of the following methods,” including the sale-of-assets method. Tex. Util.Code Ann. *286§ 39.262(h). That section does not preclude a utility from using the sale-of-assets method to quantify its stranded costs for nuclear generation assets. Id. Nor does the plain language of section 39.262(i) require a utility to use the ECOM method to quantify its stranded costs for nuclear generation assets. Id. § 39.262(i). By its use of the word “remaining,” the language of section 39.262© presumes that it is a fallback provision and that a utility will use one of the four methods in section 39.262(h) to quantify stranded costs for nuclear assets. Id. Only if a company has “remaining” assets will section 39.262© come into play.
The definition of “market value” in section 39.251(4) likewise confirms this interpretation. Section 39.251(4) defines market value “for nonnuclear assets and certain nuclear assets, [as] the value the assets would have if bought and sold in a bona fide third-party transaction or transactions on the open market under Section 39.262(h) or, for certain nuclear assets, as described by Section 39.262©, the value determined under the method provided by that subsection.” Id. § 39.251(4). The plain language of section 39.251(4) contemplates that nuclear assets might be “bought and sold in a bona fide third-party transaction.” Id.
The Commission’s interpretation allowing TCC to quantify its stranded costs for nuclear assets using the sale-of-assets method was reasonable and consistent with the plain language of the statute. In addition, PUC Rule 25.264 comports with the statute.9 For these reasons, I would overrule OPC/CCG’s issue two.
To the extent the Joint Intervenors argue that TCC was required to use the ECOM method to quantify its stranded costs for nuclear assets because that method would have produced lower stranded costs, I would likewise reject that claim. To adopt the Joint Intervenors’ argument would require the utility to know in advance which method would produce the least amount of stranded costs. It is only in hindsight that a utility will know whether the method it has chosen to quantify its stranded costs will produce lower stranded costs than the ECOM method. The plain language of section 39.262(h) does not require a utility to engage in forecasting the future. Although the legislature required utilities to mitigate their potential stranded costs and provided that a utility may recover only its “net, verifiable, nonmitiga-ble stranded costs,” the legislature also gave the utility the option of choosing which method it would use to quantify stranded costs. See Tex. UtiLCode Ann. § 39.262(h). That the utility had this choice is confirmed by the mandatory language of section 39.262(h), which states that “the affiliated power generation company shall quantify its stranded costs using one of the following methods.” Id. § 39.262(h). I would overrule Joint Inter-venors’ issue three.
(C) Was the sale of TCC’s share of the STP a bona fide third-party transaction under a competitive offering?
In the alternative, the Joint Intervenors argue that the Commission erred in its conclusion that TCC’s sale of its share of the STP was a bona fide third-party transaction under a competitive offering as required under section 39.262(h). This argument is without merit.
*287The primary complaint of the Joint In-tervenors is that the Commission failed to follow its precedent in the Texas-New Mexico Power true-up, where the Commission determined that TNMP’s sale of generation assets was not a bona fide third-party transaction under a competitive offering. But the record reflects that the facts of TCC’s sale of the STP differed from TNMP’s sale of its generation assets. The Commission determined that one of the most important aspects for determining whether a transaction meets the statutory requirements is whether the utility acquires competent and independent advice before and during the transaction. The record reflects that TCC chose several financial advisors to assist with the sale of its share of the STP, not just one, as TNMP did. The record also reflects that TCC’s choice of financial advisors was not motivated by an inside connection as was TNMP’s.10 The Commission also noted that two additional strengths of TCC’s sales process not present in that of TNMP’s was the marketing of the offering and the aggressive negotiations with bidders. The Commission determined that the basic structure of TCC’s process promoted competitive and good-faith participation by bidders. Because the facts of TCC’s sales process differed substantially from that of TNMP’s, the Commission was not bound to follow its precedent in that case or, in any event, it did not constitute precedent as to this transaction. On this record, there was a reasonable basis to support the Commission’s determination that TCC’s sale of its share of the STP was a bona fide third-party transaction under a competitive offering in compliance with section 39.262(h). I would overrule Joint Intervenors’ issue four.
2. Adjustments to net book value
(A) Commercial unreasonableness
In its final order, the Commission made two adjustments to the net book value of TCC’s generation assets to address what it determined to be commercial unreasonableness and the failure to mitigate stranded costs by TCC. Specifically, the Commission reduced the net book value of TCC’s Coleto Creek Coal Plant by $8 million and reduced the net book value of TCC’s share in the STP by $68.7 million. The district court found that these two adjustments were barred by section 39.252(d) of the PURA. The Commission and the Joint Intervenors appeal the reversal of these two adjustments. In addition, the Joint Intervenors contend that the Commission’s calculation of these two adjustments was in error. In a related issue, the Joint Intervenors also argue that the Commission erred in failing to adjust net book value for TCC’s commercial unreasonableness in failing to use “bridge power sales agreements.”
For the reasons stated in Justice Pem-berton’s opinion, a majority of this Court affirms the district court’s judgment that section 39.252(d) of the PURA precludes the Commission from making adjustments to the net book value of TCC’s generation assets for “commercial unreasonableness.”11 I disagree with the majority’s decision.
*288As an initial matter, these claims require the Court to consider the Commission’s authority under section 39.252(d) to make adjustments to the net book value of TCC’s generation assets — a question this Court already decided in Reliant I, 101 S.W.3d at 149. Section 39.252(d) provides:
An electric utility shall pursue commercially reasonable means to reduce its potential stranded costs, including good faith attempts to renegotiate above-cost fuel and purchased power contracts or the exercise of normal business practices to protect the value of its assets. The commission shall consider the utility’s efforts under this subsection when determining the amount of the utility’s stranded costs; provided, however, that nothing in this section authorizes the commission to substitute its judgment for a market valuation of generation assets determined under Sections 39.262(h) and (i).
Tex. UtiLCode Ann. § 39.252(d). In Reliant I, this Court held that section 39.252(d) authorized the Commission to make adjustments to net book value when a utility does not pursue commercially reasonable means to reduce potential stranded costs. See 101 S.W.3d at 149. We further observed that the relevant statutory goal was calculating not just an accurate stranded cost amount, but “calculating an accurate ‘verifiable, non-mitigable stranded cost[]’ amount.” Id. (quoting Tex. UtiLCode Ann. § 39.252(d)). And we recognized that “[cjompliance with the duty to pursue commercially reasonable means to mitigate its potential stranded costs is what makes stranded costs non-mitigable.” See id. Given the plain language of section 39.252(d), we held in Reliant I that nothing in section 39.252(d) prohibits the Commission from reducing net book value if it concludes that a utility fails to comply with the duty to mitigate potential stranded costs. Id.
I disagree with TCC’s assertion that Reliant I “does not address the issue of when the Commission may properly adjust net book value.” This Court’s holding in Reliant I made clear that the Commission may properly adjust net book value when it determines that a utility has not pursued commercially reasonable means to reduce its potential stranded costs. Id. I see no reason to re-examine that holding here. Accordingly, I would reaffirm this Court’s prior holding in Reliant I that section 39.252(d) allows the Commission to adjust the net book value of a utility’s generation assets when the utility fails to “pursue commercially reasonably means to reduce its potential stranded costs.” Id.; see also Tex. UtiLCode Ann. § 39.252(d).
This Court also considered the Commission’s authority to adjust net book value for commercial unreasonableness in the CenterPoint true-up appeal. See Center-Point, 252 S.W.3d at 33-34. In Center-Point, the Commission’s final order set forth a primary and alternative holding. Id. at 12-13. Under its primary holding, the Commission adopted its own market valuation of CenterPoint’s generation assets based on its finding that CenterPoint had failed to establish the market value of those assets as required under section 39.262(h). Id. at 13, 16-18. Under its alternative holding, the Commission adopted the market value proposed by CenterPoint, but made reductions to net book value based on commercial unreasonableness. Id. at 13, 32. This Court considered whether the Commission’s reductions to net book value for commercial unreasonableness should also be applied to its primary holding. Id. at 32-34. This *289Court concluded that reductions to net book value for commercially unreasonable behavior were not meant to be punitive in nature. Id. at 33. The Court stated that where a utility’s commercially unreasonable behavior benefits the utility financially and lessens the impact of stranded costs by increasing the utility’s stranded cost recovery, the amount of stranded costs recovered by the utility should be modified to capture the utility’s unreasonable behavior. Id. The Court further explained that, if the commercially unreasonable behavior has no financial impact or, if the financial impact is irrelevant to or has already been accounted for in the valuation of the utility’s generation assets, any adjustment to the net book value of the utility’s generation assets would be contrary to the legislative directive. Id.
This Court ultimately concluded that the Commission’s decision in the CenterPoint true-up proceeding to limit the adjustments to net book value to its alternative holding was reasonable because the negative effects of any commercially unreasonable behavior on the part of CenterPoint had no effect on the valuation of Center-Point’s generation assets. Id. at 34. But this Court’s statements in CenterPoint suggest that an adjustment for commercial unreasonableness is appropriate if the valuation method chosen does not account for the commercially unreasonable behavior and the utility’s commercially unreasonable behavior benefits the utility unfairly by increasing its stranded cost recovery. See id. at 33.
TCC argues that section 39.252(d) only imposes a duty to mitigate potential stranded costs prior to a utility’s divestiture of its generation assets and that a utility has no duty to mitigate potential stranded costs during or after divestiture. Thus, because the Commission determined that TCC’s sales of the Coleto Creek Coal
Plant and its share of the STP were bona fide third-party transactions under a competitive offering within the meaning of section 39.262(h)(1), TCC maintains that the Commission could not thereafter adjust the net book value of these assets for commercial unreasonableness. TCC’s argument is essentially one of timing — namely that a utility is required to pursue commercially reasonable means to mitigate stranded costs only up to the point that it decides to sell its generation assets, but not thereafter.
This is not what the legislature provided. See Tex. UtiLCode Ann. § 39.252(d). Nowhere in section 39.252(d) does the legislature limit a utility’s duty to mitigate potential stranded costs in the manner urged by TCC. The determination of whether a sale is a bona fide third-party transaction under a competitive offering is distinct from the determination of whether TCC used commercially reasonable means to mitigate stranded costs. As the Commission explained in its final order, the analysis of whether a sale is a bona fide third-party transaction under a competitive offering is directed towards the requirements of section 39.262(h)(1) in determining the market value of generation assets. See id. § 39.262(h)(1). In contrast, a determination of commercial reasonableness of a company’s mitigation efforts is made under section 39.252(d), and applies not only to the sales process, but also to other actions that the seller may take to mitigate stranded costs. See id. § 39.252(d). The Commission’s interpretation of the statute is further supported by the plain language of section 39.262(a), which provides that a utility “may not be permitted to overrecover stranded costs through the procedures established by this section or through the application of the measures provided by the other sections of this chapter.” Tex. UtiLCode Ann. § 39.262(a). The legislative intent is *290clear — a utility may not be permitted to overrecover stranded costs. See id. In light of the statute’s plain language and deferring to the Commission’s reasonable construction of the statute it is charged with enforcing, I would hold that the Commission’s finding that TCC’s sales of its generation assets were bona fide third-party transactions under a competitive offering does not preclude it from determining that TCC’s conduct undertaken in pursuit of those sales was commercially unreasonable. Under a plain reading of the statute, the former simply does not foreclose the latter. See id.
To the extent it holds otherwise, the majority restricts a utility’s duty to mitigate stranded costs in a manner not contemplated by the legislature. There is nothing temporal in the statute that precludes the Commission from adjusting the net book value of a utility’s generation assets based on commercially unreasonable conduct occurring before, during, or after an asset sale — particularly where the market valuation method employed by TCC does not take into account its commercially unreasonable behavior. Accordingly, I would sustain the Commission’s issue one and Joint Intervenor’s issue one, and I would reverse the district court’s judgment that section 39.252(d) prohibits the Commission’s adjustments to net book value. Because the majority does not, I respectfully dissent.
As a result of my conclusion that section 39.252(d) does not prohibit the Commission’s adjustments to net book value, it is necessary to consider the parties’ challenges to the Commission’s calculation of individual adjustments to net book value and whether the valuation method chosen by TCC accounts for the alleged commercially unreasonable behavior found by the Commission. See CenterPoint, 252 S.W.3d at 33-34. Unlike its order in CenterPoint, the Commission adopted only one holding in its final order here. In that order, the Commission adopted two adjustments for commercial unreasonableness: (1) an $8 million reduction to net book value based on TCC’s decision to bundle and sell the Coleto Creek Coal Plant with other less desirable generation assets; and (2) a $68.7 million reduction to net book value based on TCC’s failure to develop an estimate of the value of its share of the STP, the corresponding lack of knowledge of the STP’s intrinsic value, and the lack of a walk-away price for its share of the STP.
(i) TCC’s choice to bundle the gas plants with Coleto Creek
The record shows and TCC admits that its decision to bundle the gas plants with the Coleto Creek Coal Plant for purposes of divestiture decreased the market value of the Coleto Creek Coal Plant by $8 million — the amount of the Commission’s adjustment for commercial unreasonableness. The Commission accepted TCC’s proposal to reduce net book value for Cole-to Creek by $8 million and rejected the intervenors’ request for a much greater reduction. The $8 million reduction to net book value was based on the difference between Sempra’s bid for the Coleto Creek Coal Plant alone and its bid for the bundle of plants, which was $8 million lower than the stand-alone bid for Coleto Creek.
TCC’s decision to bundle the gas plants with Coleto Creek financially benefitted the utility and lessened the impact of stranded costs. Moreover, the valuation method chosen — the total net value realized from the asset sale — did not account for the decreased market value caused by TCC’s decision to bundle the plants. Because the effect of TCC’s commercially unreasonable decision to bundle the Coleto Creek Coal Plant with other less desirable assets was not captured in the valuation *291method chosen by TCC, I would conclude that the Commission’s $8 million reduction to TCC’s net book value for the Coleto Creek Coal Plant was appropriate. I would therefore consider the Joint Inter-venors’ challenge to this adjustment.
The Joint Intervenors challenge the Commission’s $8 million adjustment to the net book value of the Coleto Creek Coal Plant on the grounds that it is not supported by substantial evidence. They argue that the Commission’s adjustment to the net book value of Coleto Creek should have been higher than $8 million.
When conducting a substantial evidence review, a reviewing court must determine whether the evidence as a whole is such that reasonable minds could have reached the conclusion reached by the Commission. City of El Paso, 888 S.W.2d at 186; Size-more, 759 S.W.2d at 116. The true test is not whether we believe the agency reached the correct conclusion, but whether some reasonable basis exists in the record for the action taken by the agency. Charter Medical, 665 S.W.2d at 452. The Commission is the sole judge of the weight to be accorded the testimony of each witness. Central Power & Light Co., 36 S.W.3d at 561. We may not substitute our judgment for that of the Commission on the weight of the evidence on questions committed to the Commission’s discretion. Tex. Gov’t Code Ann. § 2001.174; see Charter Medical, 665 S.W.2d at 452; H.G. Sledge, 36 S.W.3d at 602. The Commission was entitled to accept or reject in whole or in part the testimony of the various witnesses who testified. See City of Corpus Christi v. Public Util. Comm’n, 188 S.W.3d 681, 695 (TexApp.-Austin 2005, pet. denied); Central Power & Light Co., 36 S.W.3d at 561.
The Commission concluded that TCC’s action of bundling Coleto Creek with the old gas-fired and hydroelectric plants adversely affected the amount bid for Coleto Creek, thereby increasing TCC’s stranded costs. The prevailing bidder, Sempra/Riv-erstone, submitted a bid for Coleto Creek alone and a separate bid for the bundled plants. The bid for Coleto Creek alone was $8 million higher than the bid for the bundled plants together. Because the bids were virtually contemporaneous and were made by the same bidder for the same facilities, the Commission concluded that TCC’s decision to bundle its gas-fired and hydroelectric plants with Coleto Creek increased TCC’s stranded costs by $8 million. Accordingly, the Commission reduced TCC’s net book value by $8 million. The Joint Intervenors complain that this adjustment was not high enough because other evidence in the record suggested that TCC could have received an additional $68.2 to $87.2 million for Coleto Creek.12 Because the Commission was entitled to accept or reject all or part of the testimony presented, see City of Corpus Christi, 188 S.W.3d at 695; Central Power & Light Co., 36 S.W.3d at 561, I would conclude that the Commission’s decision to reduce the net book value of the Coleto Creek Coal Plant was supported by substantial evidence, and I would overrule Joint Interve-nors’ issue five.
(ii) Failure to develop estimated value of the STP share
The record shows that TCC never developed an estimate of the value of its share *292of the STP that was supported by economic or financial analysis. Thus, prior to the auction of its share of the STP, TCC had not formulated how much its share was worth.13 Without this information, TCC could not make an informed decision as to whether the negotiated sales price was reasonable. The valuation method chosen could not have accounted for TCC’s lack of knowledge.
The Commission determined that TCC’s failure to develop an estimate of the value of its share of the STP, the corresponding lack of knowledge of the STP’s intrinsic value, and the lack of a walk-away price for the STP was commercially unreasonable and amounted to a failure to mitigate stranded costs. To account for TCC’s commercial unreasonableness, the Commission reduced the net book value of TCC’s share in the STP by $68.7 million. Under the standard set forth in Center-Point, I would conclude that the Commission’s reduction to TCC’s net book value for its share of the STP was appropriate. I would therefore consider the Joint Inter-venors’ challenge to this adjustment.
The Joint Intervenors argue that the Commission’s adjustment should have been higher. We review the Commission’s adjustment under the substantial evidence rule, and we presume that the Commission’s determination is supported by substantial evidence. City of El Paso, 883 S.W.2d at 187; Office of Pub. Util. Counsel v. Public Util. Comm’n, 185 S.W.3d 555, 567 (Tex.App.-Austin 2006, pet. denied).
The Joint Intervenors have the burden of proving otherwise. Office of Pub. Util. Counsel, 185 S.W.3d at 567.
The record reflects that the Commission calculated the adjustment to net book value for TCC’s share of the STP based on the SEC filing of GC Power,14 the “Project Crayola” documents,15 and a comparison of the pricing differential between Texas Genco’s16 sale of its share in the STP and TCC’s sale of its share in the STP. GC Power’s SEC filing states, “The purchase price for the nuclear acquisition ... was $707.6 million, which includes $700 million in cash paid to Texas Genco Holdings, Inc. and approximately $7.6 million in acquisition costs.” The “Project Crayola” documents likewise demonstrated that the price of Texas Genco’s sale of its share in the STP was $700 million for 1100 MW, or approximately $636/kW. This was $109/kW higher than the price of TCC’s sale to Cameco for $527/kW. The Commission calculated its adjustment to net book value by multiplying the $109/kW difference times TCC’s 630 MW share of the STP, which produced an adjustment of $68.7 million.
The Joint Intervenors argue that this adjustment was too low because Texas Genco actually received more than $636/kW for the sale of its share in the STP. The Joint Intervenors argue that the SEC filing shows that Texas Genco received $700 million for its original 770 MW share of the STP, not including the 330 MW it received when exercising its *293right of first refusal to purchase a portion of TCC’s share. But, as explained in the rebuttal testimony of TCC witness Mujeeb Qazi, the Joint Intervenors misconstrue the SEC filing relied upon by the Commission to determine the price that Texas Genco received for its share of the STP. As Qazi testified, the sale of Texas Genco’s share of the STP included the 330 MW that it received from TCC under the right of first refusal; thus, Texas Genco sold the entire 1100 MW share of the STP to GC power for $700 million. The Commission is the sole judge of the weight to be accorded the testimony of each witness. Central Power & Light Co., 36 S.W.3d at 561. The Commission was entitled to accept or reject in whole or in part the testimony of the various witnesses who testified. See City of Corpus Christi, 188 S.W.3d at 695; Central Power & Light Co., 36 S.W.3d at 561. We may not substitute our judgment for that of the Commission on the weight of the evidence on questions committed to the Commission’s discretion. Tex. Gov’t Code Ann. § 2001.174; see Charter Medical, 665 S.W.2d at 452; H.G. Sledge, 36 S.W.3d at 602. Based upon Qazi’s testimony, I would conclude that there is a reasonable basis in the record to support the Commission’s $68.7 million adjustment to the net book value of TCC’s share of the STP, and I would overrule Joint Intervenors’ issue two.
(in) Bridge Power Sales Agreements
In their seventh issue on appeal, the Joint Intervenors assert that the Commission erred in failing to adjust net book value for TCC’s commercial unreasonableness in failing to use bridge power sales agreements.17 According to the Joint In-tervenors, the Commission’s failure to require TCC to use bridge power sales agreements resulted in TCC’s failure to mitigate its potential stranded costs and allowed TCC to overrecover stranded costs in violation of sections 39.252(d) and 39.262(a) of the PURA. Because TCC retained the profits from the interim power sales, the Joint Intervenors argue that, as a result of TCC’s failure to use bridge power sales agreements, the Commission should have reduced TCC’s stranded cost recovery by $206 million.
Applying the standard set forth in Cen-terPoint, however, I would conclude that TCC’s failure to use bridge power sales agreements does not warrant an adjustment for commercial unreasonableness. Because TCC’s failure to use such agreements allowed TCC to retain the profits from power sales prior to closing the asset sale, the failure to use such agreements was accounted for in the valuation method chosen in the form of a lower asset sales price. I would overrule Joint Intervenors’ issue seven.
(B) Inclusion of construction work in progress
The Commission included in net book value the amount that TCC had spent on generation related construction work in progress as of December 31, 2001.18 In their sixth issue, the Joint Intervenors argue that the Commission erred by includ*294ing CWIP in the net book value of TCC’s generation assets. Specifically, the Joint Intervenors argue that the Commission failed to show that the inclusion of CWIP satisfied the requirements of section 36.05419 of the PURA and that these requirements were made applicable to the calculation of TCC’s stranded costs through section 39.260.20
This Court recently rejected the same claim in our consideration of the appeal of the true-up proceeding of CenterPoint Energy. See CenterPoint, 252 S.W.3d at 45-48. In CenterPoint, we upheld the Commission’s inclusion of CWIP in net book value and explained that the Commission’s decision was proper for several reasons. Id. at 46, 47-48. We rejected the construction urged by the Joint Intervenors in CenterPoint on the ground that it ignored the “unique role” of section 36.054 in the ratemaking context. Id. at 46-47. We explained that were we to adopt the construction urged by the Joint Intervenors, we would forever deny a utility’s right to recover its CWIP expenses — a result not contemplated by the legislature when it allowed utilities to recover “all” of their stranded costs. See id. at 47 (citing Tex. UtiLCode Ann. § 39.252(a)). We also concluded that the plain language of section 39.260 did not incorporate the requirements of section 36.054 into the Commission’s determination of stranded costs because those requirements do not relate to any type of accounting rule or principle. Id. Finally, we concluded that the Commission’s rule regarding stranded cost recovery includes CWIP in net book value but makes no mention of the requirements in section 36.054. Id. at 47. Because the Joint Intervenors make the same arguments this Court previously rejected in CenterPoint, I would overrule Joint Inter-venors’ issue six.
D. Excess Earnings/EMCs and Interest
I next turn to the parties’ claims regarding excess earnings and excess mitigation credits. In their third issue, OPC/CCG argue that the Commission erred in failing to reduce TCC’s stranded cost recovery by TCC’s excess earnings in 1999-2001 as required by sections 39.254 and 39.262(a) of the PURA. Additionally, the Joint In-tervenors argue that the Commission erred in its failure to reduce net book value by the amount of excess mitigation credits paid to TCC’s affiliated retail electric provider — CPL Retail Energy. In a related claim, TCC argues that the district court erred in affirming the Commission’s *295order because the Commission required TCC to pay interest twice on the same excess earnings.
The record reflects that TCC had $42,209,382 of excess earnings during 1999-2001. The Commission applied $860,765 as a reduction in the net book value of TCC’s generation assets as required under section 39.254 of the PURA. The remaining $41,348,617 of excess earnings was paid by TCC to its affiliated retail electric provider and other retail electric providers in the form of excess mitigation credits. The record reflects that TCC paid $21,351,322 to its affiliated REP CPL Retail Energy.21 OPC/CCG contend that the entire amount of excess mitigation credits should be applied to reduce the net book value of TCC’s generation assets, thereby reducing TCC’s stranded cost recovery. The Joint Intervenors agree, but they also argue in the alternative that, at a minimum, the Commission should have reduced TCC’s stranded cost recovery by the amount of excess mitigation credits paid to CPL Retail Energy.
This Court considered claims similar to those raised by OPC/CCG and the Joint Intervenors in CenterPoint. See 252 S.W.3d at 34-39. We held that allowing the utility to recover as stranded costs the amounts paid to its affiliated REP would allow the utility to overrecover stranded costs in violation of section 39.262(a). Id. at 38-39. We explained that section 39.262 of the PURA requires the Commission to treat the electric utility and its affiliated REP as one entity for the purpose of stranded cost recovery. Id. at 38-39. Therefore, to the extent a utility’s affiliated REP was allowed to retain excess mitigation credits, the utility could not recover that amount as stranded costs because allowing such a recovery would result in an overrecovery of stranded costs in violation of section 39.262(a). Id. at 39.
Adhering to this Court’s decision in Cen-terPoint, I would conclude that the Commission erred in allowing TCC to recover as stranded costs those amounts paid to CPL Retail Energy as excess mitigation credits, but consideration of these claims must also take into account this Court’s prior decision in City of Corpus Christi v. Public Utility Commission, 188 S.W.3d 681 (Tex.App.-Austin 2005, pet. denied). In City of Corpus Christi, this Court held that the Commission exceeded its statutory authority when it ordered TCC to pay excess mitigation credits in the first instance. 188 S.W.3d at 689. The supreme court denied the Commission’s petition for review on April 29, 2007 — more than one year after the Commission’s final order in the instant true-up proceedings. Accordingly, we are called upon to reconcile our earlier decision in City of Corpus Christi with our more recent decision in Center-Point, and the Commission’s final order in TCC’s true-up proceedings.
This analysis begins with the Commission’s final order. Finding that the appellate process in City of Corpus Christi was “not yet complete,” the Commission “declined] to provide for [an] adjustment [to TCC’s stranded costs] in this Order.” Stated differently, the Commission did not consider the effects of our decision in City of Corpus Christi, much less our recent decision in CenterPoint, when it finalized TCC’s stranded costs in the instant true-up proceeding. Because the Commission should be given the opportunity to consider the effects of our decisions in City of Corpus Christi and CenterPoint in the first instance,221 would reverse the district *296court’s judgment to the extent it affirmed the Commission’s order regarding the treatment of excess mitigation credits, and remand this issue, including the claims raised in OPC/CCG’s issue three and Joint Intervenors’ issue ten, to the Commission for further proceedings.
For the same reasons, I would reverse the district court judgment to the extent it affirmed the Commission’s order regarding the treatment of interest as it relates to the payment of excess mitigation credits, and remand this issue, including the claim raised in TCC’s issue three, to the Commission for further proceedings.
E. Interest on Stranded Costs
In its final judgment, the district court concluded that the Commission “erred in applying Rule [16 Tex. Admin. Code § ] 25.263 to determine the interest rate on stranded costs, because the Supreme Court invalidated the rule in [Cen-terPoint Energy].” Both the Commission and TCC appeal this portion of the district court’s judgment. Because I conclude that the supreme court’s decision in Center-Point Energy did not invalidate that portion of the rule regarding the interest rate, I would reverse the judgment of the district court. See 143 S.W.3d at 84.
The Commission’s true-up rule in effect at the time, PUC Substantive Rule 25.263, provided in relevant part:
The TDU shall be allowed to recover, or shall be liable for, carrying costs on the true-up balance. Carrying costs shall be calculated using the cost of capital established in the utility’s UCOS proceeding, and shall be calculated for the period of time from the date of the true-up final order until fully recovered.
26 Tex. Reg. 10498, 10525 (2001) (Pub.Util. Comm’n) (codified as amended at 16 Tex. Admin. Code § 25.263(0(3». The rule states that the proper interest rate to be applied to a utility’s stranded costs is the cost of capital established in the utility’s UCOS proceeding.23 Id. This interest rate was not at issue before the supreme court in CenterPoint Energy.
In CenterPoint Energy, the supreme court declared “that Rule 25.263(0(3) is invalid.” 143 S.W.3d at 99. At first glance, this statement would appear to invalidate the entire rule, including that portion of the rule regarding the interest rate to be applied to a utility’s stranded costs. Upon closer examination of the supreme court’s opinion, however, I am not persuaded that the supreme court spoke so broadly.
It is clear from the supreme court’s opinion that the only issue before the court in CenterPoint Energy was the date on which a utility begins to accrue interest on its stranded cost balance. As originally adopted, the Commission’s rule provided that a utility did not begin to accrue interest on its stranded costs until the Commission issued a final order in the utility’s stranded cost true-up proceeding. 26 Tex. Reg. at 10525; see also CenterPoint Ener*297gy, 143 S.W.3d at 83. The supreme court determined that the Commission’s rule was invalid because it was inconsistent with the legislative intent that a utility recover all of its net, verifiable, non-mitigable stranded costs, “that ‘exist on the last day of the freeze period [December 31, 2001],’ ” which would include carrying costs from the first date of retail competition, or January 1, 2002. CenterPoint Energy, 143 S.W.3d at 84 (quoting Tex. Util. Code Ann. § 39.201(g)). Thus, despite its broad statement that rule 25.263(Z )(3) was invalid, it is clear from the text of the supreme court’s opinion that the court only invalidated that portion of the rule relating to the date on which carrying costs, or interest, began to accrue. See id.
Furthermore, the Commission’s rules have a severability clause. See 16 Tex. Admin. Code § 25.3 (2007) (Pub.Util. Comm’n). The severability clause states,
If any provision of this chapter is held invalid, such invalidity shall not affect other provisions or applications of this chapter which can be given effect without the invalid provision or application, and to this end, the provisions of this chapter are declared to be severable.
Id. § 25.3(a). To determine whether it is proper to sever that portion of rule 25.263 invalidated by the supreme court in Cen-terPoint Energy from the remainder of the rule we apply a two-prong test:
(1) will the function of the regulatory statute as a whole be impaired without the invalid part of the rule; and
(2) is there any indication that the agency would not have adopted the rule but for the invalid part?
See Texas Dep’t of Banking v. Restland Funeral Home, Inc., 847 S.W.2d 680, 683 (Tex.App.-Austin 1993, no writ) (citing K Mart Corp. v. Cartier, Inc., 486 U.S. 281, 294-95, 108 S.Ct. 1811, 100 L.Ed.2d 313 (1988)). If, in the court’s view, the answer to either inquiry is “yes,” then severance is not justified and the entire rule must fall. Id.
Applying this test in the context of rule 25.263(Z )(3), I would conclude that severance is justified. The function of the rule as a whole will not be impaired without the invalid part of the rule, and there is no indication that the Commission would not have adopted the rule but for the portion invalidated by the supreme court. See id. I would sustain the Commission’s issue two and TCC’s issue four, reverse the district court’s judgment, and render judgment affirming the Commission’s final order on this issue.
F. Evidentiary Ruling
In its third issue on appeal, the Commission seeks reversal of that portion of the district court’s judgment finding that the Commission erred in striking the expert report of Ross A. Sollosy, a consultant hired to advise the Commission and paid for by the Commission. The Joint Inter-venors also appeal the district court’s judgment on this issue. For the reasons that follow, I agree with the Commission and the Joint Intervenors that the district court’s judgment was in error.
This Court reviews the Commission’s decision to admit or exclude evidence for abuse of discretion — the same standard we apply to trial courts. City of Amarillo v. Railroad Comm’n, 894 S.W.2d 491, 495 (Tex.App.-Austin 1995, writ denied). An agency has broad discretion when deciding whether to admit expert testimony in an administrative hearing, and its decision will not be disturbed absent an abuse of discretion. Fay-Ray Corp. v. Texas Alcoholic Bev. Comm’n, 959 S.W.2d 362, 367 (Tex.App.-Austin 1998, no pet.).
The record reflects that Sollosy and his firm Navigant were hired to advise the Commission and its staff regarding TCC’s *298sale of its share of the STP and other generation assets. During the hearing before the Commissioners, Commission staff introduced Sollosy’s expert report into evidence. In addition, TCC introduced excerpts from Sollosy’s deposition testimony, a deposition exhibit, and Sollo-sy’s expert report. TCC’s introduction of Sollosy’s testimony, exhibit, and report was cumulative of other evidence and testimony provided by other TCC witnesses. After Sollosy’s testimony and report were introduced into evidence, it was discovered that Navigant, in its responses to the parties’ requests for information (RFIs), had failed to produce documents responsive to the parties’ RFIs. Initially, the Commissioners asked Sollosy to review his RFI responses and supplement them as necessary. Several days later, Navi-gant’s general counsel appeared unexpectedly at the hearing and disclosed more extensive problems. He admitted that Navigant’s responses to the RFIs were misleading and incomplete because several boxes of material related to Navigant’s prior involvement with electric generation asset sales, as well as information relied upon by Sollosy in forming his opinions regarding TCC’s asset sales and preparing his expert report, had not been produced. Navigant’s general counsel indicated that there were potentially sixty to one hundred boxes of responsive materials that had not been produced. He stated that these documents were, at the time, being pulled from a warehouse and reviewed and redacted and that they would be available in Austin in a week.24
In response to this development, various intervenors moved to strike Sollosy’s testimony and expert report because they had not had the opportunity to review these documents in preparation of their case before cross-examining Sollosy. The Commission granted the motion to strike.25 As a result of this decision, the Commission struck, among other exhibits, TCC Exhibit 28, including excerpts from Sollosy’s deposition testimony, a deposition exhibit, and Sollosy’s expert report.
TCC argues on appeal that we should affirm the district court’s judgment reversing the Commission’s decision to exclude TCC Exhibit 28 because the Commission’s decision was contrary to its own sanctions rule and the rules of civil procedure because it was a sanction “against TCC” even though TCC was not the offending party. I disagree.
The Commission’s rule regarding sanctions for discovery abuse expressly allows the Commission to strike “pleadings or testimony, or both, in whole or in part” as a remedy for discovery abuse. 16 Tex. Admin. Code § 22.161(c)(8), (d) (2007) (Pub.UtiLComm’n). The Commission’s decision to exclude Sollosy’s testimony and expert report was not a sanction “against TCC.” Sollosy was a witness for Commission staff, not TCC. Thus, if anything, excluding Sollosy’s testimony was a sanction against Commission staff, not TCC. Furthermore, the record reflects that TCC provided additional testimony from various witnesses addressing the same issues addressed by Sollosy — namely, whether the sale of TCC’s share of the STP was a bona *299fide third-party sale under a competitive offering in compliance with section 39.262(h) of the PURA, whether the price TCC received for its share of the STP was reasonable, and whether the bundling of TCC’s Coleto Creek plant with other facilities was commercially reasonable. This evidence demonstrates that Sollosy’s testimony and expert report were merely cumulative. In light of the fact that Sollosy’s testimony and expert report were cumulative of other evidence introduced by TCC, I cannot conclude that the Commission’s decision violated its own sanctions rule, much less that the Commission abused its discretion in striking Sollosy’s testimony and report.
For the same reasons, even if the rules of civil procedure apply to discovery in Commission proceedings, I cannot conclude that the Commission’s decision to strike Sollosy’s testimony and report was contrary to those rules. Like the Commission’s rule regarding sanctions for discovery abuse, the rules of civil procedure give courts the discretion to exclude evidence as a sanction for discovery abuse. See, e.g., Tex.R. Civ. P. 193.6(a), 215.2(b)(4). Because Sollosy was a witness for Commission staff, not TCC, and his testimony and report were cumulative of other evidence offered by TCC, I would conclude the Commission’s decision striking Sollo-sy’s testimony and report comports with the rules of civil procedure and was not an abuse of discretion.26 Therefore, I would sustain the Commission’s third issue and Joint Intervenors’ eighth issue, reverse that portion of the district court’s judgment finding that the Commission erred in striking Sollosy’s testimony and expert report, and render judgment affirming the Commission’s decision excluding Sollosy’s testimony and expert report.
G. PUC Treatment of TCC’s Tax Accounts
In its second issue on appeal, TCC argues that the district court erred in upholding the Commission’s treatment of TCC’s federal tax accounts for Accumulated Deferred Investment Tax Credits (AD-ITC) and Excess Deferred Federal Income Tax (EDFIT) because of the threat of “normalization violations.”27 TCC requests this Court to reverse the district court’s judgment and remand this issue to the Commission for further consideration. The Commission joins in this request asserting that federal tax law regarding normalization violations is in flux. The Joint Intervenors dispute whether a remand to the Commission is appropriate and argue that the district court properly rejected the request for a remand to the Commission.
This Court recently addressed a similar issue in CenterPoint. See CenterPoint, 252 S.W.3d at 41-45. In that case, we affirmed the Commission’s treatment of CenterPoint’s ADITC and EDFIT accounts and the Commission’s related reductions to CenterPoint’s stranded cost recovery, see id. at 44; however, we remanded the issue for the Commission to consider whether it was appropriate to provide a remedy in the event that the *300IRS later determined that a normalization violation had occurred. Id. at 44-45. For the same reasons expressed in Center-Point, I would affirm the Commission’s treatment of TCC’s ADITC and EDFIT accounts, but I would also conclude that a remand is appropriate for the Commission to consider the effects of the IRS’s private letter ruling issued to TCC after the close of evidence before the Commission. Therefore, I would reverse that portion of the district court’s judgment affirming the Commission’s decision not to consider the effects of the private letter ruling issued to TCC, and remand this cause back to the Commission for further proceedings.
H. Capacity Auction True-up
In addition to determining TCC’s stranded costs, the legislature charged the Commission with reconciling TCC’s capacity auction true-up award. See Tex. Util. Code Ann. § 39.262(d). TCC sought a capacity auction true-up award of $482,664,890, which the Commission reduced by $420,695,372. The district court affirmed the Commission’s reduction to TCC’s proposed capacity auction true-up award. On appeal, TCC argues that the district court’s judgment should be reversed because the Commission’s reduction was “unsupported by substantial evidence, arbitrary and capricious, grossly excessive, disproportionate to any downward bias from alleged deficiencies in TCC’s capacity auction process, and reaches a completely unreasonable result.” I disagree.28
To ensure the availability of power to all retail competitors in the newly deregulated market, the legislature required each electric utility, including power generation companies, to auction off entitlements to “at least 15 percent” of the electric utility’s installed generation capacity. See Tex. UtiLCode Ann. § 39.153(a) (emphasis added). The Commission concluded, and no one disputes, that TCC failed to auction entitlements to at least 15% of its installed generation capacity during 2002 and 2003. The dispute centers upon whether TCC satisfied the requirements of the relevant “safe-harbor” provisions and, therefore, could be deemed to have met the 15% statutory requirement or, if not, to what award, if any, TCC was entitled.
As part of the transition to a competitive retail electric market, the legislature required each formerly bundled utility to auction off 15% of its installed generation capacity before and after the date on which customer choice began — January 1, 2002. See Tex. UtiLCode Ann. § 39.153(a), (b). These auctions were designed to encourage retail competition by providing a source of power to new competitors in the market. See CenterPoint, 252 S.W.3d at 47-48; Reliant I, 101 S.W.3d at 137. The legislature also directed the Commission to adopt rules defining the scope of the capacity entitlements to be auctioned and the procedures to be followed.29 See Tex. Util. Code Ann. § 39.153(e)-(g); 16 Tex. Admin. Code § 25.381 (2007) (Pub.Util.Comm’n). Pursuant to the Commission’s rule, the auctions were to be held four times a year, 16 Tex. Admin. Code § 25.381(h)(1)(A)®, and the utilities were obligated to sell entitlements to four types of capacity products: baseload, gas-intermediate, gas cyclic, and gas-peaking. Id. § 25.381(c)(5), (f)(g). The amount of each type of product to be sold varied depending on the utility’s
*301generation assets, but the total amount of entitlements sold had to equal at least 15% of the utility’s total generation capacity. Id. § 25.381(e)(1).
If a utility failed to auction off at least 15% of its generation capacity, it could still be deemed to have met this requirement if it satisfied the safe-harbor provisions established by the Commission. The Commission established two safe-harbor provisions that are relevant to this appeal. First, under the Commission’s capacity auction rule, a utility will be deemed to have met the 15% requirement if it offered products in a product category and successfully sold all of the entitlements offered in at least one month in that category. Id. § 25.381(h)(l)(B)(iv); see also id. § 25.381(h)(7)(C).30 If there is no month in which all of the products in a given category were sold, the company must make a proposal to the Commission to meet the 15% requirement. 16 Tex. Admin. Code § 25.381(h)(7)(C). This safe-harbor provision became part of the Commission’s capacity auction rule in June 2002, and therefore only relates to capacity auctions in 2003.
Second, the Commission approved a settlement in two proceedings — PUC Docket Nos. 23774 and 24888 — adopting a capacity-auction-mechanics document that included a safe-harbor provision applicable for certain capacity auctions in 2002. The Commission determined that this safe-harbor provision was “significantly different” than the safe-harbor provision established in the Commission’s rule. To satisfy this safe-harbor provision, a utility must sell all of its offered entitlements in a given month. If, in a given month, all entitlements are not sold, the utility must propose to auction additional entitlements to the products that did sell in order to meet the 15% requirement. This safe-harbor provision allows a utility to remedy its failure to sell sufficient products only by offering additional entitlements for the products that did sell. In contrast, the safe-harbor provision in the Commission’s capacity auction rule is product-specific and requires the utility to sell all of the entitlements offered in at least one month for each product category. To remedy a deficiency under the rule’s safe-harbor provision, the utility may propose to modify the product’s offering price or the product category, or it may offer alternative capacity products.
In addition, the safe harbors established in Docket Nos. 23774 and 24888 have only limited applicability. The capacity-auction-mechanics document in Docket No. 23774 applies only to the initial auction— i.e., the September 2001 auction. The capacity-auetion-mechanies document adopted in Docket No. 24888 contains the same limiting language. Because Docket No. 24888 addressed only the mechanics of two auctions held in March 2002 and July 2002, the safe-harbor provision from Docket No. 24888 only applies to the initial auction in 2002 — i.e., the March 2002 auction.
1. TCC’s compliance with the 15% requirement
In its first argument on appeal, TCC contends that it should not “be required to do the impossible — successfully sell products that the market does not want.” TCC complains that the Commission required the auction of four specific products that were defined in minute detail at minimum prices established by the Commission and that to construe section 39.153 of the *302PURA in a manner that requires TCC to successfully sell at auction 15% of these Commission-designed, Commission-priced products is unreasonable and leads to absurd results. Stated differently, TCC argues that it should not be held to the 15% requirement in section 39.153 because the market did not want the products that were to be sold at auction. In support of this argument, TCC claims that none of the other utilities required to conduct capacity auctions were successful at selling “unmarketable” products.
I am unpersuaded by this argument. The language in section 39.153(a) is clear and unambiguous: it requires each utility to sell “at least 15 percent” of its generation capacity. Tex. UtiLCode Ann. § 39.153(a). Moreover, section 39.153(e) charges the Commission with authority to define the scope of capacity entitlements to be auctioned. Id. § 39.153(e). Although this section gives some guidance to the Commission,31 the legislature left it to the Commission’s discretion to determine the types of capacity entitlements to be auctioned. Id. And TCC does not argue that the Commission-designed products did not meet the requirements of section 39.153(e). Rather, TCC argues that the products were “unmarketable.” TCC, perhaps recognizing the fallacy of its own argument, dismisses it with the statement that “TCC was not required to meet this impossible requirement because TCC satisfied the applicable ‘safe harbor provisions.’ ” I
would reject the notion that TCC was not required to satisfy the 15% requirement in section 39.153(a).
2. TCC’s compliance with safe-harbor provisions
In its second argument on appeal, TCC argues that safe harbors existed during both 2002 and 2003 and that TCC met the applicable safe-harbor provisions and, therefore, the Commission should have found that TCC satisfied the 15% requirement. As part of its argument, TCC contends that safe harbors existed for all of 2002 under the settlements adopted by the Commission in Docket Nos. 23774 and 24888. However, as previously discussed, the settlements adopted by the Commission in each of these dockets only applied to the initial auction addressed in each proceeding — i.e., the September 2001 and March 2002 auctions. See discussion supra. The Commission concluded that TCC could only be deemed to have met its capacity auction obligation for 2002 if TCC met the requirements of the safe harbor for each of these two auctions. Accordingly, the Commission considered only whether TCC complied with the safe-harbor provisions for the September 2001 and March 2002 auctions.
Examining TCC’s sales in the September 2001 auction, the Commission found that TCC failed to sell all of its offering in any one month covered by the September 2001 auction.32 In light of this failure, the *303safe-harbor provision in Docket No. 23774 required TCC to propose to the Commission a plan to sell all of its offering in any one month covered by the September 2001 auction. Because TCC made no such proposal and continued to offer the same number of entitlements in its subsequent auctions, the Commission concluded that TCC failed to achieve the safe harbor for the September 2001 auction. And the Commission likewise concluded that TCC failed to achieve the safe harbor for capacity sold in 2002.
As for 2003, the Commission examined whether TCC met the safe harbor stated in rule 25.381(h)(l)(B)(iv) and (h)(7)(C). The Commission determined that there was no single month in which TCC sold all of its offered gas-intermediate or gas-peaking products. Given this failure, the rule required TCC to propose to the Commission a plan to offer modified or different capacity products to meet its 15% obligation. TCC did so, and proposed to offer modified versions of these products, but even the modified products failed to sell in sufficient quantities. According to the testimony of TCC witness Michael Isenberg, TCC sold no additional gas-intermediate entitlements and only three of its four offered gas-peaking entitlements. TCC made no additional proposals. Thus, the Commission concluded that TCC failed to satisfy the safe harbor established in the Commission’s rule for capacity sold in 2003.33
Having reviewed the record, I would conclude that the Commission’s determinations that TCC failed to satisfy the applicable safe harbor in either 2002 or 2003 was supported by substantial evidence and was neither arbitrary nor capricious.
3. TCC’s obligation to sell one- and two-year strips
In addition to its failure to sell the amount of capacity required by the PURA and the Commission’s rules, the Commission also determined that TCC failed to sell products over the terms required by the Commission’s capacity auction rule. The Commission’s rule required TCC to offer entitlements to capacity products for various time periods. Of the total number of entitlements offered by the utility, approximately 20% must be auctioned as two one-year strips for 2002 and 2003, with the strips offered jointly. See 16 Tex. Admin. Code § 25.381(h)(3)(A)(i). An additional 30% of the entitlements must be in the form of one-year strips for 2002. Id. § 25.381(h)(3)(A)(ii). The remainder of the utility’s offered capacity must be in the form of discrete months for 2002. Id. § 25.381(h)(3)(A)(iv) and (v). No one disputes that TCC failed to offer one-year or two-year strips during its initial capacity auction. TCC witness Isenberg testified that TCC offered the equivalent amount of capacity in terms of discrete months because, as part of the settlement resolving the AEP-CSW merger proceeding, TCC had agreed to sell several plants constituting 1354 MW of capacity. TCC contended that its capacity auction notices stated that it was not offering strips and that neither the Commission nor any other party ob-*304jeeted. TCC argued that the lack of an objection excused any failure to offer the strips. The Commission concluded that TCC’s failure to offer the strips violated the capacity auction rule and was not excused by TCC’s statements in the auction notices that it would not be offering the strips because the capacity auction rule plainly states that what is deemed approved in the event that no objection is made to the notice is the notice itself, not the substantive details of the auction. See id. § 25.381(h)(2)(B)®.
TCC makes this same argument on appeal and argues that the very purpose of the notice is to provide the substantive details, not simply to provide notice without content. Even if one accepts TCC’s argument, that would not excuse TCC’s failure to meet the 15% requirement, much less its failure to satisfy the applicable safe harbor for 2002 or 2003. Therefore, it is irrelevant for purposes of this appeal whether the lack of objection to TCC’s notices excused its failure to offer the yearly strips or not.
Jp. Calculation of TCC’s capacity auction true-up award
The Commission concluded that TCC capacity auctions fell substantially short of TCC’s obligations under the PURA and the Commission’s capacity auction rule. Specifically, TCC failed to sell 15% of its generation capacity as required in section 39.153; TCC failed to achieve safe harbor in either 2002 or 2003; and TCC failed to offer approximately half of its capacity entitlements in the form of yearly strips as required by rale 25.381. These defects in TCC’s capacity auctions required the Commission to determine an alternative proxy for the capacity auction price to be used in the capacity auction true-up formula stated in the Commission’s rule.
(A) Commission authority to develop an alternative calculation method
This Court considered a utility’s failure to comply with the statutory capacity auction requirements in CenterPoint. See CenterPoint, 252 S.W.3d at 48-59. In that case, we reversed the district court’s judgment and affirmed the Commission’s final order disallowing $440 million of Center-Point’s requested capacity auction true-up award. Id. at 14, 58-59, 70. Like TCC, CenterPoint had failed to comply with the statutory requirements to auction 15% of its generation capacity and likewise failed to satisfy the applicable safe-harbor provisions. See id. at 48-59. This Court held that, in light of CenterPoint’s failures, the Commission acted within its authority when it used an alternative method for establishing the capacity auction price to calculate CenterPoint’s capacity auction true-up award. Id. at 48-59, 70.
TCC presents the same arguments this Court rejected in CenterPoint. First, as previously discussed, TCC argues that it should not be held to the 15% obligation because there was simply no market for the products designed and priced by the Commission to be sold at auction. In other words, TCC argues that the failure to satisfy the 15% requirement was due to factors outside of its control. This Court rejected the same argument in Center-Point, finding that by specifying what products were to be sold and for what amount, the Commission was doing exactly as the legislature had instructed it to do. Id. at 55-56; see also Tex. UtiLCode Ann. § 39.153(e) (requiring the Commission to determine the scope of entitlements to be offered and the minimum amount of capacity to be auctioned), (f) (requiring the Commission to establish procedures for the auctions, including designating which generation units are offered and establishing an opening bid price). Moreover, the Court recognized that, although Center-Point expressed dissatisfaction with the *305Commission’s actions on appeal, Center-Point had failed to challenge the products or the price specified in the Commission’s capacity auction rule within the time period authorized by statute. See Center-Point, 252 S.W.3d at 55-56 (citing Tex. Util.Code Ann. § 39.001(f)). Like Center-Point, TCC has failed to show that it objected to the products or the price specified in the Commission’s rule. For the same reasons expressed by this Court in CenterPoint, I would reject TCC’s argument that its failure to satisfy the 15% requirement was due to factors beyond its control.
Second, TCC argues that any reduction to its proposed capacity auction true-up award was improper because TCC substantially complied with the applicable safe-harbor provisions. Again, the Court rejected this argument in CenterPoint. See id. at 53-55. The Court explained that the legislative goal of the capacity auction was to have each utility auction at least 15% of its generation capacity. Id. at 54. Moreover, the legislature’s use of the phrase “at least” was some indication that the legislature thought 15% was the absolute minimum amount of capacity that needed to be sold in order to determine an accurate estimate of the capacity auction price. Id. at 54. Thus, the Court concluded that substantial compliance must ultimately be measured by comparison to this legislative goal, and not the Commission’s safe-harbor rule. Id. Indeed, by establishing a safe-harbor provision, the Commission essentially determined what is necessary to establish substantial compliance with the legislative goal. Id. at 54-55. For the reasons expressed in Center-Point, I would conclude that TCC failed to substantially comply with the relevant requirements. See id.
Finally, TCC argues that the Commission’s reduction to TCC’s proposed capacity auction award was excessive when compared to what TCC considers to be minor deficiencies in its capacity auctions. TCC essentially argues that the PUC could have used simple mathematical calculations to determine “what would have happened” in TCC’s auctions in the absence of any deficiencies and that it was improper for the Commission to develop an alternative method for determining the “capacity auction price” to be used in the true-up formula when calculating TCC’s capacity auction true-up award. Although, in Cen-terPoint, we found the framing of this argument compelling, this Court rejected CenterPoint’s characterization of the reduction. Id. at 57. The Court explained that the situation presented in this issue was similar to the one presented in the market valuation of CenterPoint’s generation assets. Id. Because the relevant statutory provisions and the Commission’s rule assume the utility will comply with either the 15% requirement or with the safe harbor, neither one addresses the possibility of noncompliance. Id. The Court concluded that the Commission was caught between a statutory mandate requiring utilities be allowed to recover for their capacity auction costs and the utility’s noncompliance with the requirements necessary to determine an accurate award. Id. Accordingly, in the absence of compliance, the Court found the Commission had the implied authority to develop an alternative method for determining the capacity auction price to be used in calculating Center-Point’s true-up award. Id. at 57. For the same reasons, I would conclude that the Commission had the implied authority to develop an alternative method for determining the capacity auction price to be used in calculating TCC’s capacity auction true-up award. It is therefore necessary to consider whether the Commission’s decision was supported by substantial evidence.
*306 (B) Was the Commission’s calculation supported by substantial evidence?
In its final order, the Commission considered various proxies proposed by the parties for use as the capacity auction price. TCC argued that the capacity price obtained in its auction should be used but, if the Commission determined a disallowance was appropriate, it should only make narrow adjustments to address the specific deficiencies of TCC’s capacity auctions. Alternatively, TCC proposed that the Commission use the Megawatt Daily ER-COT South Zone index to establish a proxy for TCC’s capacity auction price. According to TCC, use of this index would produce prices of $25.76/MWh for 2002 and $39.60/MWh for 2003, and would reduce TCC’s proposed capacity auction true-up award by approximately $90 million. In contrast, the intervenors argued that the proper adjustment was derived by use of TCC’s actual capacity and energy sales revenues for 2002 and 2003. Under this approach, TIEC determined that the appropriate prices would be $42.50/MWh for 2002 and $48.01/MWh for 2003. The Commission adopted the approach and the proposed prices based on TCC’s actual revenues for 2002 and 2003 as urged by the intervenors.
TCC argues that the prices adopted by the Commission are not supported by substantial evidence because there is no evidence that these prices represent competitive prices. TCC contends that these prices were not competitive because the evidence showed that they were based primarily on non-market revenues such as resource-must-run (RMR) contracts and sales to TCC’s affiliated retad electric provider. I disagree.
The purpose of the capacity auctions as stated in the statute was to compare the actual prices received at auction with the prices in the 2001 ECOM model and award the difference to the utility. See Tex. Util. Code Ann. § 39.262(d)(2). Because the Commission determined and the record supports that TCC’s capacity auction was deficient in many respects, the Commission was unable to make the statutory comparison because it did not have actual prices received by TCC in the capacity auctions. Accordingly, the Commission was left to develop an alternative method for determining the prices that TCC would have received if it had performed a successful capacity auction.
This Court addressed a similar situation in CenterPoint. See CenterPoint, 252 S.W.3d at 51. Like TCC, CenterPoint also failed to comply with the statutory requirement to auction 15% of its generation capacity. Id. Having performed a deficient capacity auction, the Commission could not use the prices obtained by Cen-terPoint in the capacity auctions to determine the capacity auction true-up. Id. As a result, the Commission developed a proxy for the capacity auction price by using the average price of all capacity products sold by CenterPoint. Id. This Court concluded there was a reasonable basis for the Commission’s decision to use a proxy for CenterPoint’s capacity auction price and that the Commission’s decision was consistent with the relevant statutory provisions. Id. at 57-58. This Court affirmed the Commission’s calculation of CenterPoint’s capacity auction true-up award. Id. at 58.
Here, as in CenterPoint, the Commission used the average price of all TCC capacity sales during the relevant time period to determine the proxy for TCC’s capacity auction price. This was the approach urged by TIEC, the Cities, and OPC. TIEC witness Jeffry Pollock testified that using the average price of all of TCC’s capacity sales during the relevant time period would produce prices of *307$42.50/MWh for 2002 and $48.01/MWh for 2003. Cities witness Scott Norwood34 and OPC witness Randall Falkenberg also testified to this approach and calculated price figures that were very close to or identical to these prices.
The Commission adopted the approach urged by the intervenors and agreed that using the average price of all capacity sales was consistent with the purpose of the capacity auction true-up to reconcile the utility’s actual results in the competitive market with the power price projections from the 2001 ECOM model. The Commission recognized that the problems with TCC’s capacity auction were substantial and were more significant than those identified by the Commission in Center-Point’s capacity auction. In particular, the Commission found that, by failing to sell sufficient quantities of products other than baseload, the results of TCC’s capacity auctions over-represented the lower base-load pricing and were, therefore, distorted downwards. The Commission further concluded that the approach urged by the intervenors was consistent with the approach the Commission took in the Center-Point true-up and rejected TCC’s proposal to correct the deficiencies in TCC’s capacity auction on a “flaw-by-flaw” or “piecemeal” basis.
Consistent with this Court’s opinion in CenterPoint, I would conclude that there is a reasonable basis in the record to support the Commission’s development of an alternative method for determining TCC’s capacity auction price. The simple fact that the Commission chose to use the method urged by the intervenors instead of that urged by TCC does not make the Commission’s decision unreasonable. The method chosen by the Commission was reasonable and consistent with the relevant statutory provisions, and I would conclude that the Commission’s decision and calculation of TCC’s capacity auction true-up award was supported by substantial evidence. I would overrule TCC’s first issue.
In a related issue, the Joint Intervenors assert that the district court erred by affirming the Commission’s decision to include interest on TCC’s capacity auction true-up award. The Commission determined that including interest on TCC’s capacity auction true-up was necessary to make the utility whole and was also consistent with the supreme court’s decision in CenterPoint Energy, 143 S.W.3d at 85. As recognized by the supreme court in CenterPoint Energy, the purpose of the capacity auction true-up was to ensure that the utility earned the same margins that were projected in the 2001 ECOM model. TCC was entitled to a capacity auction award because it did not recover all of the predicted margins. Moreover, since those amounts will not be recovered until the true-up is complete, it necessarily follows that a time value for the delay must be added to make the utility whole. See id. This is likewise consistent with this Court’s decision in CenterPoint. See 252 S.W.3d at 59-62. Accordingly, for the reasons expressed in this Court’s CenterPoint opinion, I would overrule Joint Interve-nors’ issue nine.
*308III. Conclusion
Having considered all of the parties’ issues raised on appeal, I would affirm the district court’s judgment in part, reverse in part, and remand this cause to the Commission for further proceedings. Because the plain language of section 39.252(d) does not preclude the Commission’s adjustments to the net book value of TCC’s generation assets for commercial unreasonableness related to the sales of TCC’s share in the STP and the decision to bundle the Coleto Creek Coal Plant with other generation assets, I dissent from the majority’s conclusion that it does. I would reverse the district court’s judgment on this point and affirm the Commission’s reductions to net book value.
Joined in part by Justices PURYEAR • and PEMBERTON.
. Tex. UtiLCode Ann. §§ 11.001-64.158 (West 2007).
. Section 39.251(7) defines the term "stranded costs” as:
the positive excess of the net book value of generation assets over the market value of the assets, taking into account all of the electric utility’s generation assets, any above market purchased power costs, and any deferred debit related to a utility’s discontinuance of the application of Statement of Financial Accounting Standards No. 71 ("Accounting for the Effects of Certain Types of Regulation”) for generation-related assets if required by the provisions of this chapter. For purposes of Section 39.262, book value shall be established as of December 31, 2001, or the date a market value is established through a market valuation method under Section 39.262(h), whichever is earlier, and shall include stranded costs incurred under Section 39.263.
Tex. UtiLCode Ann. § 39.251(7) (West 2007).
. The 63 participating cities included: Aran-sas Pass, Beeville, Charlotte, Devine, Dilley, Eagle Pass, Edinburg, Edna, Ganado, Indian Lake, Ingleside, Ingleside on the Bay, Jourdanton, Kingsville, Los Fresnos, Luling, Lyford, Lytle, McAllen, Mathis, Mercedes, Nordheim, Odem, Palmview, Portland, Port Lavaca, Poteet, Primera, Raymondville, Rio Hondo, Sabinal, Smiley, South Padre Island, Victoria, Alamo, Bishop, Goliad, Gregory, Harlingen, La Feria, Pharr, Pleasanton, Port Aransas, Rancho Viejo, Weslaco, Cariz-zo Springs, Del Rio, Donna, Laredo, Crystal City, Corpus Christi, San Benito, Uvalde, Bay City, Bayview, Camp Wood, George West, Palm Valley, Refugio, Rio Grande City, San Juan, Sinton, and Taft.
. Those intervenors participating in this appeal include the State of Texas, Cities, AVH, CCG, OPC, and TIEC. The State, Cities, AVH, and TIEC have filed joint briefs, and I refer to them collectively as the "Joint Inter-venors.” OPC and CCG have also filed joint briefs, and I refer to them collectively as "OPC/CCG.”
. The current version of the capacity auction rule differs from that version originally promulgated by the Commission. Compare 25 Tex. Reg. 9139 (2000) (original rule), adopted 25 Tex. Reg. 12961 (effective Jan. 4, 2001), with 16 Tex. Admin. Code § 25.381 (2007) (current rule). The primary difference relevant to this appeal concerns the adoption of "safe harbors” for the purpose of compliance with the rule. As originally adopted, the capacity auction rule did not include a safe harbor — or alternate means of compliance— but the rule was amended in 2002 to add a safe harbor. See 27 Tex. Reg. 5982 (2002) (adding subsection 25.381(h)(l)(B)(iv) among other provisions). I cite the current rule unless noted otherwise.
. The ECOM model is a computer model used by the Commission to predict whether a utility will incur stranded costs. See CenterPoint Energy Houston Electric, LLC v. Gulf Coast *283Coalition of Cities, 252 S.W.3d 1, 9 (Tex.App.Austin 2008, no pet. h.) (op. on reh’g).
. I agree with TCC that CCG has failed to preserve this issue for review. Therefore, I only address this issue to the extent it is raised by OPC.
. Contrary to the suggestions of TCC and the Commission, the record demonstrates that AVH raised this claim in its motion for rehearing before the Commission and in its petition for judicial review before the district court. Therefore, I would conclude that the Joint Intervenors have preserved this claim for our review.
. Because I conclude that the Commission’s interpretation is reasonable and consistent with the statute and that the rule comports with the statute, I would not reach the issue of whether OPC/CCG's claim is an impermissible collateral attack on the validity of the rule.
. In TNMP’s case, TNMP chose Laurel Hill as its only financial advisor based on the merger agreement between TNMP’s parent company TNP Enterprises and ST Acquisition Corporation. Because the managing director of Laurel Hill was also the chief financial officer of TNP Enterprises, the Commission concluded that Laurel Hill was chosen because of its inherent relationship with TNMP’s parent company and not because it offered significant expertise in the sale of a utility’s generation assets.
. For this reason, Justices Puryear and Pem-berton do not join parts II.C.2(A)(i) through (iii) of this opinion, except to the extent we *288overrule Joint Intervenors’ issue seven and affirm the district court’s judgment and the Commission’s final order regarding the use of bridge power sales agreements.
. Cities witness Scott Norwood recommended a $68.2 million reduction to net book value based on his comparison of what TCC received for the Oklaunion Coal Plant and what it received for Coleto Creek; CCG witness Paul Wielgus testified that TCC could have received an additional $70 million for Coleto Creek; and consistent with Norwood’s testimony regarding the comparison between Oklaunion and Coleto Creek, OPC witness David Rode recommended a reduction to net book value of $87.2 million.
. It is illogical to assume that a reasonable seller would not want to know the value of what he is selling.
. GC Power purchased Texas Genco’s share of the STP.
. The "Project Crayola” documents were admitted without objection in the proceedings before the Commission.
.Texas Genco was a co-owner of the STP and exercised its right of first refusal to purchase approximately 330 MW of TCC's share in the STP. After this purchase, Texas Genco owned approximately 1100 MW of the STP’s total capacity, which it then sold to a third party.
. A bridge power sales agreement "would have enabled the buyers of TCC's generating units to sell power from those units until the sales were finalized, with TCC being compensated with a higher asset-sale price.” Stated differently, the buyers, not TCC, would have received the benefit of the interim power sales and would have paid a higher purchase price to TCC for its generation assets.
. This amount was $53,588,063, plus interest, and included only those CWIP amounts that TCC had spent as of December 31, 2001. It did not include projected costs for unfinished projects.
. Section 36.054 states:
(a) Construction work in progress, at cost as recorded on the electric utility’s books, may be included in the utility’s rate base. The inclusion of construction work in progress is an exceptional form of rate relief that the regulatory authority may grant only if the utility demonstrates that inclusion is necessary to the utility’s financial integrity.
(b) Construction work in progress may not be included in the rate base for a major project under construction to the extent that the project has been inefficiently or imprudently planned or managed.
Tex. Util.Code Ann. § 36.054.
. Section 39.260 states:
(a) The definition and identification of invested capital and other terms used in this subchapter and Subchapter G that affect the net book value of generation assets and the treatment of transactions performed under Section 35.035 and other transactions authorized by this title or approved by the regulatory authority that affect the net book value of generation assets during the freeze period shall be treated in accordance with generally accepted accounting principles as modified by regulatory accounting rules generally applicable to utilities.
Tex. Util.Code Ann. § 39.260.
. CPL Retail Energy received a total of $29,616,249, including interest of $8,264,927.
. The legislature has charged the Commission with the task of finalizing each utility’s *296stranded costs. See Tex. Util.Code Ann. § 39.262(c).
. The Commission established the proper interest rate to be applied to TCC’s stranded cost balance in TCC's unbundled cost of service, or UCOS, proceeding. See Tex. Pub. Util. Comm’n, Application of Central Power & Light Company for Approval of Unbundled Cost of Service Rate Pursuant to PURA § 39.201 and Public Utility Commission Substantive Rule § 25.344, Docket No. 22352 (Oct. 5, 2001) (order), available at http:// interchange.puc.state.tx.us (accessed June 13, 2008); see also Tex. Util.Code Ann. § 39.201 (West 2007); 16 Tex. Admin. Code § 25.344 (2007). TCC witness Randall W. Hamlett testified that the cost of capital established in TCC’s UCOS proceeding was 11.7951%. This was the rate adopted by the Commission to calculate interest on TCC’s stranded costs.
. The record reflected that some of the documents relating to Navigant's prior experience in the sale of generation assets had been returned to its clients and that copies were not kept. Navigant’s general counsel stated that Navigant could "probably get them back.”
. I disagree with TCC's assertion on appeal that the Commission did not strike the Cities' exhibits of Sollosy's testimony because the hearing transcript reflects that Sollosy’s testimony and report were struck for all purposes except for TCC’s offer of proof.
. I would also reject TCC’s argument that the exclusion of TCC Exhibit 28 was harmful error because TCC fails to address the cumulative nature of Sollosy’s testimony and expert report. To establish harm as to the exclusion of evidence, TCC must demonstrate that the excluded evidence was both controlling on a material issue and not cumulative of other evidence. See Williams Distrib. Co. v. Franklin, 898 S.W.2d 816, 817 (Tex. 1995).
. A normalization violation is essentially a violation of the tax accounting regulations of the Internal Revenue Service. See Center-Point, 252 S.W.3d at 42-43 (discussing normalization violations).
. TCC and the Commission had a fundamental disagreement about how to interpret the relevant provisions in the PURA and the Commission’s rules regarding TCC’s statutory obligation to auction capacity.
. The Commission adopted PUC Substantive Rule 25.381. See 16 Tex. Admin. Code § 25.381 (2007) (Pub.Util.Comm’n).
. The same safe-harbor provision appears in both of these portions of the Commission’s capacity auction rule.
. Sections 39.153(e)(l)-(5) provide that the entitlements:
(1) may be sold and purchased in periods of not less than one month nor more than four years;
(2) may be resold to any lawful purchaser, except for a retail electric provider affiliated with the electric utility that originally auctioned the entitlement;
(3) include no possessory interest in the unit from which the power is produced;
(4) include no obligations of a possessory owner of an interest in the unit from which the power is produced; and
(5) give the purchaser the right to designate the dispatch of the entitlement, subject to planned outages, outages beyond the control of the utility operating the unit, and other considerations subject to the oversight of the applicable independent organization.
Tex. Util.Code Ann. § 39.153(e)(l)-(5).
. The Commission’s finding was based on *303the direct testimony of TIEC witness Jeffry Pollock.
. With respect to 2003, TCC argues that the Commission's determination was flawed because the Commission improperly included TCC's mothballed plants as part of TCC’s total generation capacity. This argument is not relevant because the record demonstrates that TCC failed to meet its 15% obligation in 2003 regardless of whether TCC’s mothballed capacity was included or excluded. Indeed, the record shows, and the Commission concluded, that TCC sold only 5% of its capacity in 2003, if the mothballed capacity is included, and only 8%, if the mothballed capacity is excluded.
. To the extent TCC complains that Nor-wood admitted that TCC's revenues were "dominated” by what TCC alleges are non-competitive sales, TCC takes one statement from Norwood’s entire testimony out of context. The full record of Norwood’s testimony supports the approach urged by the intervenors. Moreover, the Commission was entitled to accept or reject any part or all of each expert’s testimony. City of Corpus Christi v. Public Util. Comm'n, 188 S.W.3d 681, 695 (Tex.App.-Austin 2005, pet. denied); Central Power & Light Co. v. Public Util. Comm’n, 36 S.W.3d 547, 561 (Tex.App.-Austin 2000, pet. denied).