City of El Paso v. Public Utility Commission of Texas

ON MOTION FOR REHEARING JONES, Justice. The opinion and judgment issued by this Court on August 14, 1991, are withdrawn, and this opinion is filed in place of the earlier one. The district court affirmed an order of the Public Utility Commission (“Commission”) setting rates to be charged by El Paso Electric Company (“EPEC”). The Commission issued the order, after hearing, pursuant to the Public Utility Regulatory Act (PURA), Tex.Rev.Civ.Stat.Ann. art. 1446c (Supp.1992). The City of El Paso (“City”), the State of Texas (on behalf of various state agencies located in western Texas) (“TSA”), and the Office of Public Utility Counsel (“OPC”), appellants, seek reversal of the trial court’s judgment. We will affirm in part and reverse in part. Disapproving of EPEC’s decision to invest in the Arizona Nuclear Power Project, appellants complain of the Commission’s order permitting EPEC to charge rates that allow it to recover most of the project’s costs. The Commission ordered the rate increase after conducting a lengthy evidentiary hearing, providing many opportunities for all parties to plead and present their cases; the Commission considered two sets of motions for rehearing. The Travis County district court, to which the complaining parties brought their appeal, affirmed the Commission’s order. On appeal, appellants challenge twelve different aspects of the Commission’s decision: (1) adoption of a non-unanimous stipulation; (2) failure to disallow more for “decisional” imprudence; (3) approval of deferred accounting for regulatory-lag expenses; (4) failure to disallow more for constructional imprudence; (5) exclusion of witness Hubbard’s testimony; (6) refusal to find excess capacity in EPEC’s system; (7) inclusion of common facilities in rate base; (8) inclusion of income tax expense in cost of service; (9) inclusion of a portion of Unit 2 lease payments in cost of service; (10) calculation and inclusion of other cost-of-service allowances; (11) temporary exclusion of TSA from the proceedings; (12) determination of rates and rate class for TSA. We will reverse the trial court’s judgment to the extent it affirmed the Commission’s approval and use of “deferred accounting” for the carrying costs EPEC incurred between the date the Palo Verde plant became commercially operational and the date the new rates were effective; we will affirm the remainder of the trial court’s judgment. SETTING At the heart of this dispute is EPEC’s decision to expand its power generation system by obtaining an ownership interest in the Arizona Nuclear Power Project, also known as the Palo Verde Nuclear Generating Station. EPEC and four other utility companies agreed to partially fund and otherwise assist in building one or more nuclear steam electric generating units, with attendant common facilities. In return, EPEC was entitled to 15.8% of the resulting net energy generation and the same percentage of available generating capability. Construction is complete on the common facilities and two of the five units originally planned. At the time this proceeding was heard by the Commission, a third unit was still under construction. Factors arising after construction began have induced EPEC to alter its ownership interest in the units. Originally, EPEC owned an undivided interest in each of the units as a tenant in common with the other four project participants. Although EPEC retains its undivided interest in Unit 1,'the company has sold its interest in Unit 2 and made arrangements to lease the unit back for the duration of EPEC’s involvement in the project. Because of the complexity of appellants’ points of error, we will supply additional facts from the record throughout this opinion as necessary to clarify the discussion. PROCEDURAL BACKGROUND On October 31, 1986, EPEC filed an application requesting the Commission to de *902 termine whether EPEC’s arrangement to sell and lease back its ownership interest in Unit 2 was in the public interest. On April 6, 1987, EPEC filed applications with various affected cities and the Commission to raise its rates based on the inclusion of the new nuclear plant in its generation system. The City of El Paso, one of the affected cities, approved rates within the El Paso city limits that would not have allowed EPEC to recover any of its nuclear plant investment; EPEC appealed the City’s decision to the Commission. The Commission consolidated EPEC’s appeal with its appeals from the decisions of other municipalities and with the environs case (affecting areas outside the cities) that was already before it. 1 Because of the voluminous evidence to be presented, the Commission initially divided the proceedings in the rate case into three phases. The Commission added a fourth phase after several parties tendered a non-unanimous “stipulation”- to the Commission and requested that the Commission base its decision on the stipulation. In addition, pursuant to an agreed motion, the Commission consolidated the rate case with the proceeding to approve the sale/leaseback arrangement. An examiner convened hearings on each of the four phases during August, September, October, and November of 1987. On February 1, 1988, the hearings examiner filed a report with the Commission recommending against adoption of the stipulation. On March 31, 1988, the Commission signed an order adopting and incorporating the terms of an amended and restated stipulation and increasing rates to permit EPEC to earn a return on part of its investment. The order withheld approval of the sale/leaseback arrangement until a later proceeding, effectively undoing the earlier consolidation. On May 10, 1988, the Commission made minor changes to its order in response to motions for rehearing filed by the City, OPC, TSA, and others. The City, OPC, and TSA filed second motions for rehearing which the Commission overruled on June 16, 1988. The City, OPC, and TSA each appealed the Commission’s order to a Travis County district court, and the Commission obtained a consolidation of those appeals. All the appellants participated in the ensuing district-court review, which resulted in an af-firmance of the Commission’s order. The City, OPC, and TSA each raise several challenges to the trial court’s judgment and, through it, to the Commission’s order. THE NON-UNANIMOUS STIPULATION The threshold complaint of the City and OPC is that the Commission erred in basing its final order, in part, on a non-unanimous stipulation. Both appellants argue that there is no substantial evidence to support the stipulated matters and that the Commission violated its own procedural rules by considering the stipulation. In addition, OPC asserts that some necessary findings or statements of underlying facts are lacking and that, by considering and basing its final order on the stipulation, the Commission “improperly attempt[ed] to negate statutorily created rights belonging solely to OPC.” An unarticulated assumption underlies the majority of appellants’ challenges to the Commission’s decision in these and other points of error; although they do not say so explicitly, appellants impliedly urge us to presume that, by basing its final order partially on stipulated matters, the Commission completely abdicated its responsibility to determine disputed issues. We may not so presume; indeed, the law compels a contrary presumption. In reviewing a challenged administrative order, we must presume its validity. The challenger bears the burden of showing error. Texas Health Facilities Comm’n v. Charter Medical-Dallas, Inc., 665 S.W.2d 446, 453 (Tex.1984); Continental Cars, Inc. v. Texas Motor Vehicle Comm’n, 697 S.W.2d 438, 441 (Tex.App.1985, writ ref’d n.r.e.). *903 We may not substitute our discretion or our judgment for that of the agency; we may reverse an agency’s decision only if it is unsupported by substantial evidence, is arbitrary, or results from an abuse of discretion. Railroad Comm’n v. Continental Bus System, Inc., 616 S.W.2d 179, 181 (Tex.1981). An agency’s decision is arbitrary or results from an abuse of discretion if the agency: (1) failed to consider a factor the legislature directs it to consider; (2) considers an irrelevant factor; or (3) weighs only relevant factors that the legislature directs it to consider but still reaches a completely unreasonable result. Gerst v. Nixon, 411 S.W.2d 350, 360 n. 8 (Tex.1966); Statewide Convoy Trans., Inc. v. Railroad Comm’n, 753 S.W.2d 800, 804 (Tex.App.1988, no writ). Appellants analogize the present case to a civil cause in which the court has rendered an agreed judgment without the consent of all parties. The analogy is not apt. In the present case, the decision-maker did not impose the terms of a settlement on non-settling parties. Although the parties signing the stipulation believed its terms fairly resolved disputed issues, 2 tender of the stipulation to the examiner did not bind the Commission to “adopt” it. Furthermore, the Commission did not, as appellants suggest we presume, adopt the stipulation as its final order without scrutinyThe non-signing parties had ample opportunity to argue their positions both before and after the Commission rendered its decision. A fourth phase was added to the hearings to provide a forum in which parties could object to the Commission’s use of the stipulation as a partial basis for its final order. In addition to presenting evidence, the parties submitted briefs concerning use of the stipulation, filed exceptions to the proposed final order incorporating stipulated matters, and moved for rehearing after the Commission rendered its decision. Having urged their objections to the Commission’s use of the stipulation at each of these stages, appellants yet failed to show that basing a final order on a non-unanimous stipulation would be improper. In a similar case, coincidentally involving the Palo Verde Nuclear Generating Station, the New Mexico Supreme Court approved the state Public Service Commission’s adoption of a non-unanimous stipulation: [The Commission] can adopt a contested stipulation by, first, affording any non-stipulating party an opportunity to be heard on the merits of the stipulation (i.e., whether it is a fair and reasonable resolution of the controversy before the Commission) and second, making an independent finding, supported by substantial evidence in the record, that the stipulation does indeed resolve the matters in dispute in a way that is fair, just and reasonable and in the public interest. Attorney General of New Mexico v. New Mexico Public Service Comm’n, 111 N.M. 636, 640, 808 P.2d 606, 610 (1991). The New Mexico court relied on Mobil Oil Corp. v. Federal Power Commission, 417 U.S. 283, 94 S.Ct. 2328, 41 L.Ed.2d 72 (1974), where the United States Supreme Court noted the distinction between considering a proposal “as a settlement” and considering it “on its merits”: “[E]ven if there is a lack of unanimity [in the stipulation], it may be adopted as a resolution on the merits_” 417 U.S. at 414, 94 S.Ct. at 2282 (quoting Placid Oil Co. v. FPC, 483 F.2d 880, 893 (5th Cir.1973)) (emphasis in original). In the present case, the requirements mentioned in the foregoing cases for the adoption of a non-unanimous stipulation were satisfied. First, the non-stipulating parties were given an opportunity to be heard on the merits of the stipulation. Indeed, as stated above, the Commission added a fourth phase to the proceedings devoted exclusively to receiving evidence and argument on the propriety of using the stipulation as a basis for resolving the contested issues. Second, the Commission made the requisite independent findings. The initial part of the Commission’s Order recited: *904 4. Even where some parties to a proceeding do not agree to a stipulated result, it is reasonable to adopt such a stipulation if: (a) The parties opposing the stipulation have notice that the stipulation may be considered by the Commission and an opportunity to be heard on their reasons for opposing the stipulation; (b) The matters contained in the stipulation are supported by a preponderance of the credible evidence in the case; (c) The stipulation is in accordance with applicable law; (d) The stipulation results in just and reasonable rates;- (e) The results of the stipulation are in the public interest, including the interest of those customers represented by parties opposing the stipulation. 5. Pursuant to the Findings of Fact and Conclusions of Law set forth below, the Commission finds the Amended and Restated Stipulation, as modified, is a reasonable basis for resolution of the issues in this case and that adoption of the Amended and Restated Stipulation, as modified, as the basis of the Commission’s Order in this proceeding is in the public interest. In addition, Finding of Fact No. 237 stated: “The provisions of the Amended and Restated Stipulation are reasonable and supported by a preponderance of the credible evidence in this record and should be adopted.” Conclusion of Law No. 28 stated: “The Amended and Restated Stipulation, as modified per Finding of Fact No. 6, represents a reasonable resolution of the contested issues in this docket, is supported in the record, is in the public interest, and should therefore be adopted, as the basis for the Commission’s Order in this case.” Appellants have not shown that use of the stipulation as a partial basis for the final order is arbitrary, unreasonable, an abuse of discretion, or involves consideration of factors other than those the legislature has directed the Commission to consider. Under such circumstances, we conclude that the Commission may generally set just and reasonable rates in an order based, in part, on a non-unanimous stipulation. On a procedural level, OPC asserts that the Commission’s rules, specifically Public Utility Commission Rules of Practice & Procedure § 21.151, 16 Tex.Admin. Code § 21.151 (1990), prohibit it from basing its order on a non-unanimous stipulation. Section 21.151 provides: After the expiration of the time for filing exceptions and replies thereto, the examiner’s report and proposal for decision will be considered by the commission and either adopted, modified and adopted, or remanded to the examiner.... OPC admits the Commission has the power to reject an examiner’s recommendation; however, it claims the Commission may not modify and then adopt a stipulation. OPC argues that the Commission violated section 21.151 by basing its final order on a modified stipulation over the examiner’s contrary recommendation. The argument is meritless. Paragraph six of the Commission’s final order expressly adopts findings of fact and conclusions of law proposed by the parties who signed the stipulation. The Commission accepted the proposed findings and conclusions rather than those recommended by the examiner only when there was a conflict between the two. The Commission expressly adopted the “examiner’s report” to the degree it was consistent with the proposed findings and conclusions; it expressly repudiated the section of the examiner’s report concerning EPEC’s prudence in investing and remaining involved in the project. We hold that the Commission was not required to accept or reject the examiner’s report in its entirety. The Commission’s authority undoubtedly extends to repudiating a part of the examiner’s report and modifying it by deletion. Within its challenge to the Commission’s use of the stipulation, OPC claims that paragraph 4(e) of the Commission’s final order “supplants” OPC’s authority to represent the interest of residential and small business consumers in ratemaking cases of this type. Paragraph 4(e) provides: “The results of the stipulation are in the public *905 interest, including the interest of those customers represented by parties opposing the stipulation.” OPC’s interpretation of the order and of its own enabling legislation is incorrect. The legislature created OPC to “advocate” the interests of residential and small commercial consumers. PURA § 15A(a). However, the Commission must set just and reasonable rates in ratemaking cases. PURA § 38. In addition, only the Commission has the authority to determine whether a sale of utility assets is in the public interest. See PURA § 63. Authority to advocate a position on behalf of small businesses and residential consumers is not equivalent to authority to decide what is in the public’s best interest. The only authority OPC possesses is the former. OPC’s contention is overruled. One challenge to the Commission’s use of the non-unanimous stipulation remains. Appellants argue that some findings of fact phrased in statutory language lack the required accompanying concise statements or findings of underlying facts. This argument, addressed to all the findings accompanying the final order, is too general to preserve error. To the extent that appellants assert generally that necessary findings of underlying fact are missing, they have waived their complaint by failing to demonstrate any error prejudicing their substantial rights. See Administrative Procedure and Texas Register Act (hereinafter “APTRA”), Tex.Rev.Civ.Stat.Ann. art. 6252-13a, § 19(e) (Supp.1992). The order addresses numerous issues, and appellants have made several specific substantial-evidence and finding-of-fact challenges. We will discuss appellants’ specific (and consequently preserved) challenges while disposing of their remaining points of error. “DECISIONAL” IMPRUDENCE The Commission concluded that due to imprudent decisions, $32 million of EPEC’s costs should not be included in rate base. The City’s third point of error and OPC’s second point contend that the disallowance is unsupported by substantial record evidence, arguing that the amount disallowed should have been greater. 3 OPC also charges that the Commission acted arbitrarily and abused its discretion in selecting the $32 million figure and that it made insufficient findings of underlying facts. Based on its anticipated load demand, EPEC first decided to participate in the nuclear power project at a 15.8% level. Confident that its decision was a prudent one, EPEC has continued to participate at the same level. When appellants challenged the prudence of EPEC's decisions, the City, EPEC, and the Commission staff each offered expert testimony on whether EPEC had acted prudently in deciding to participate in the project and in continuing to participate at the 15.8% level. The Commission concluded EPEC had not been entirely prudent in making decisions about its level of participation in the project. The issues we must resolve are: (1) whether substantial evidence supports the findings underlying the Commission’s disal-lowance of the precise $32 million figure, and (2) whether the Commission made the necessary findings of fact to allow this Court to conduct a meaningful review of imprudently incurred costs. A. Substantial Evidence and Abuse of Discretion. In conducting a substantial-evidence review, we must determine whether the evidence as a whole is such that reasonable minds could have reached the conclu *906 sion the agency must have reached in order to take the disputed action. Texas State Bd. of Dental Examiners v. Sizemore, 759 S.W.2d 114, 116 (Tex.1988), cert. denied, 490 U.S. 1080, 109 S.Ct. 2100, 104 L.Ed.2d 662 (1989); Texas Health Facilities Comm’n v. Charter Medical-Dallas, Inc., 665 S.W.2d 446, 453 (Tex.1984). We may not substitute our judgment for that of the agency and may consider only the record on which the agency based its decision. Sizemore, 759 S.W.2d at 116. The appealing party bears the burden of showing a lack of substantial evidence. Charter Medical, 665 S.W.2d at 453. It cannot meet this burden merely by showing that the evidence preponderates against the agency decision. Id. at 452. If substantial evidence would support either affirmative or negative findings, we must uphold the agency’s order, resolving any conflict in favor of the agency’s decision. Auto Convoy Co. v. Railroad Comm’n, 507 S.W.2d 718, 722 (Tex.1974); Warner v. City of Lufkin, 582 S.W.2d 165, 167 (Tex.Civ.App.1979, writ ref’d n.r.e.). Appellants’ position is that only the City’s witness, Ben Johnson, provided a method by which the Commission could quantify the amount of imprudently incurred costs. Johnson offered the opinion that EPEC had made several imprudent decisions and that, as a result, the Commission should disallow 50% of its costs. The City contends that because no other witness suggested a quantification method, the Commission, upon a finding of some decisional imprudence, should have adopted Johnson’s quantification method and, necessarily, his result. We do not agree. The Commission is empowered to hold hearings, receive evidence, make decisions, issue orders, and find facts. PURA § 16. In addition, the Commission impliedly possesses those powers necessary and convenient to making findings and decisions. PURA § 16(a). PURA does not expressly confer on the Commission power to judge witnesses’ credibility; however, the requirement that the Commission make decisions, findings of fact, and conclusions of law implies the necessary corollary power to judge credibility and to accept or reject a witness’s testimony in whole or in part. Gerst v. Guardian Sav. & Loan Ass’n, 434 S.W.2d 113, 116 (Tex.1968); Texas State Bd. of Dental Examiners v. Silagi, 766 S.W.2d 280, 283 (Tex.App.1989, writ denied). EPEC maintains that, based on information available at the time it made its decisions, its continued participation in the project would enable it to supply ratepayers with needed electricity at the lowest possible cost. Consequently, EPEC contends that all its costs should be included in rate base. The City, on the other hand, argues that available data would have informed a prudent utility manager that involvement in a nuclear project would be unduly burdensome to ratepayers. Consequently, Johnson recommended that the Commission disallow half of EPEC’s costs. However, the evidence encompassed more than Johnson’s recommendations. All the witnesses who offered their opinions about EPEC’s decisional prudence recognized the complexity of a decision to construct or purchase new generating capacity. A utility must weigh many competing concerns before undertaking an expansion project. EPEC, as a part of its decision to participate in the Arizona Nuclear Power Project, considered the following factors, among others: (1) the feasibility of obtaining financing; (2) the effect of long-term financing on EPEC’s financial integrity; (3) the potential impact on ratepayers of increasing system capacity by a significant percentage; (4) predicted expenses, revenues, and load demands for the relevant time periods; (5) EPEC’s degree of financial flexibility; and (6) the availability of alternative sources for the additional capacity that forecasts had shown would be necessary. The effects these factors have on total project costs are not susceptible of ready quantification. Requiring the Commission to adopt or reject witnesses' testimony in toto, especially when the testimony concerns a multifaceted issue such as this one, would hobble the Commission’s ability to assess each witness and render its decision based solely *907 on the testimony it found credible. Having deduced that the Commission may properly accept less than all of a witness’s testimony, we conclude the Commission committed no error in disallowing a lesser percentage of costs than Johnson recommended. The Commission could properly identify the factors which credible evidence showed EPEC should have considered when making its decisions. Likewise, the Commission could also decide that prudence would not have required EPEC to consider other factors because the evidence to the contrary was not credible. The record contains substantial evidence to support a disallowance figure of zero for decisional imprudence; the Commission would, therefore, have been acting within its discretion had it agreed that EPEC was entirely prudent in its management and planning. The substantial evidence would also have supported a Commission finding that 50% of EPEC’s costs should have been disallowed. 4 Appellants assert that the stipulation is the only possible evidence of the exact $32 million disallowance. Because the appellants do not admit that the stipulation has any evidentiary weight, they contend there is no evidence to support the $32 million figure. We disagree. Because it is a statement contrary to EPEC’s pecuniary interest, the concession has some evidentiary weight. A declaration contrary to a party’s position on a disputed issue is akin to a quasi-admission. While not binding on the declarant, as a judicial admission would be, such a concession constitutes some evidence. Mendoza v. Fidelity & Guar. Ins. Underwriters, Inc., 606 S.W.2d 692, 694 (Tex.1980); Texas Distillers, Inc. v. Howell, 409 S.W.2d 888, 890 (Tex.Civ.App.1966, writ ref’d n.r.e.). It is for the trier of fact to determine the weight to be assigned to a quasi-admission. Mendoza, 606 S.W.2d at 694. EPEC’s position has always been that it acted prudently in deciding to participate in the project at the 15.8% level. Nevertheless, EPEC conceded through the stipulation that, if its decision had been imprudent, the resulting costs that should be disallowed totaled $32 million. Such a statement is clearly contrary to EPEC’s position. Therefore, the Commission could properly consider and weigh the stipulation in quantifying the imprudently incurred costs. The range of figures supported by the testimony of expert witnesses, the complexity of the issues the Commission had to review to determine whether EPEC made prudent decisions, the difficulty of assigning a value to the effects of any component on project costs, and EPEC’s admission against interest all combine to compel our conclusion that substantial record evidence supports the $32 million disallowance. We overrule the substantial-evidence challenge to the disallowance for decisional imprudence. In addition, because OPC’s contention that the Commission abused its discretion rests on the premise that the stipulation has no evidentiary weight, and because we have concluded to the contrary, we overrule this contention as well. B. Findings of Fact. As its final challenge within this point, OPC asserts that “[t]he Commission’s findings are insufficient to comply with AP-TRA.” However, instead of arguing that specific findings of underlying fact are insufficient, OPC complains that the Commission failed to provide “explicit statements of underlying facts to support its findings as required by Tex.Rev.Civ.Stat.Ann. art. 6252-13a, § 16(e) (APTRA).” 5 Section 16(b) of APTRA provides, in part, that “[fjindings of fact, if set forth in statutory language, must be accompa *908 nied by a concise and explicit statement of the underlying facts supporting the findings.” (Emphasis added.) The Texas Supreme Court has concluded that an agency’s findings of fact need the additional support of findings of underlying facts only when the ultimate findings are in terms taken directly from the enabling legislation or when they “represent the criteria that the legislature has directed the agency to consider in performing its function.” Charter Medical, 665 S.W.2d at 451. The “statutory language” to which AP-TRA § 16(b) refers is the language in the statute that confers authority on the agency to take the complained-of action. Id. In PURA, the legislature authorized the Commission to make orders setting rates. PURA § 37. A number of PURA’s sections also detail the criteria the Commission is to consider in setting rates. See PURA §§ 38, 39, 41, and 43. Therefore, only when the Commission’s findings are stated in PURA’s express terms, or when they represent criteria the legislature has directed the Commission to consider, must the Commission also make findings of underlying fact. OPC does not direct us to a particular finding that requires, but is not accompanied by, findings of underlying fact. However, we conclude from its argument that Findings 101 through 103 are the subjects of its complaint, because those findings address the prudence and disallowance issues. Although PURA does not expressly require the Commission to make a finding of prudence before including costs in rate base, once the Commission finds a major project to have been imprudently planned or managed, it should generally disallow project costs to the extent of the imprudence. See PURA § 41(a); supra note 3. Finding of Fact 101 is phrased in statutory terms; the Commission therein decided that EPEC was “not entirely prudent in its planning and management” of the project. The phrasing of that finding in statutory terms required the Commission to make findings of underlying fact showing the basis for the Commission’s determination of imprudence and supplying the amount of the disallowance. Findings 102 and 103 accomplish these goals. Neither Finding 102 or 103 is phrased in statutory terms. Finding 103 states that “[qjuantification of the effects of imprudence requires the exercise of judgment based upon the evidence. In light of the evidence relating to prudence and the difficulties in quantification, the quantification of decisional imprudence at $32 million for Units 1 and 2 is reasonable and appropriate.” Finding 102 indicates the Commission concluded that imprudence existed only with respect to “the Company’s continuing evaluation of the level of its participation in the Palo Verde Project.” (Emphasis added.) Taken together, these findings adequately supply the required concise statement of underlying facts supporting Finding 101. OPC contends that, in order to be sufficient, the Commission’s findings of underlying fact must identify “the processes and acts found to be imprudent, the nexus between those acts and the disallowance amount, [and] the evidentiary support for the disallowance figure.” OPC fails to point either to statutory provisions or case law mandating that the Commission make such findings. Not having discovered any such authority ourselves, we conclude that such findings are not required. We overrule the City’s third point of error and OPC’s second point of error. DEFERRALS EPEC requested that its rate base be increased by the amount of carrying costs and operating and maintenance costs it incurred during “regulatory lag.” 6 The utility had “deferred” these types of costs for Units 1 and 2, aggregating each type of cost for each unit into a separate capital *909 account. 7 EPEC obtained the Commission’s prior permission to defer Unit 1 costs. The Commission reserved the right, however, to refuse subsequently to include the deferred costs in rate base to the extent they were unreasonable, related to plant not used and useful, or were spent or incurred imprudently. Although EPEC did not obtain prior permission to defer its post-in-service costs for Unit 2, it nevertheless deferred them, apparently assuming that obtaining prior approval a second time was unnecessary. After the rate-increase proceeding was completed, the Commission included the deferred costs for both units in rate base. In this appeal, the City and OPC contend the Commission erred, first, in entering an order permitting EPEC to defer Unit 1 costs and, second, in subsequently including the “deferred-costs assets” for both units in rate base. 8 OPC and the City lodge several arguments against the deferred accounting procedure used here, including that the practice constitutes impermissible retroactive ratemaking and that it violates the “original-cost” standard contained in PURA § 41(a). 9 Thus, they assert that the Commission exceeded its authority by including the deferred post-in-service costs in rate base. A reviewing court's role in construing a statute is to “seek out the legislative intent from a general view of the enactment as a whole, and, once the intent has been ascertained, to construe the statute so as to give effect to the purpose of the Legislature.” Hightower v. State Comm’r of Educ., 778 S.W.2d 595, 597 (Tex.App.1989, no writ); see also Medeiros v. Insurance Co. of N. Am., 781 S.W.2d 404, 406 (Tex.App.1989, no writ); Sexton v. Mount Olivet Cemetery Ass’n, 720 S.W.2d 129, 137 (Tex.App.1986, writ ref'd n.r.e.). As a general rule, an administrative agency is a creation of the legislature and, as such, has only those powers expressly conferred and those necessary to the accomplishment of its duties. State v. Jackson, 376 S.W.2d 341, 344 (Tex.1964); Sexton, 720 S.W.2d at 137; Railroad Comm’n v. Atchison, T. & S.F. R.R., 609 S.W.2d 641, 643 (Tex.Civ.App.1980, writ ref'd n.r.e.). The present case is governed by PURA, which expressly grants to the Commission “the general power to regulate and supervise the business of every public utility within its jurisdiction and to do all things, whether specifically designated in this Act or implied herein, necessary and convenient to the exercise of this power and jurisdiction.” PURA § 16(a). A reviewing court may determine, as a matter of law, the scope of an agency’s statutory authority. See Gage v. Railroad Comm’n, 582 S.W.2d 410, 412 (Tex.1979). The power of an agency to take such actions as may be “necessary” to perform an express duty is not without limits. This Court has previously held that [t]he agency may not, however, on a theory of necessary implication from a specific power, function, or duty expressly delegated, erect and exercise what really amounts to a new and additional power or one that contradicts the statute, no matter that the new power is *910 viewed as being expedient for administrative purposes. Sexton, 720 S.W.2d at 137-38 (emphasis added). Thus, if there is no specific express authority for a challenged action, and if the action is inconsistent with a statutory provision or ascertainable legislative intent, we must conclude that, by performing the act, the agency has exceeded its grant of statutory authority. PURA requires the Commission to set rates at a level that will permit each utility “a reasonable opportunity to earn a reasonable return on its invested capital used and useful in rendering service to the public over and above its reasonable operating expenses.” PURA § 39(a). This provision imposes many complex tasks on the Commission: What are the utility’s reasonable operating expenses? What portion of the utility’s expenditures constitute capital investment? What portion of the utility’s invested capital is used and useful in rendering service? How should the value of the utility’s used-and-useful invested capital be calculated? What is a reasonable return on the utility’s used-and-useful invested capital? Historically, one of the most vexing questions for regulatory authorities has been how to calculate the value of a utility’s invested capital. See Charles F. Phillips, The Regulation of Public Utilities 305-29 (2nd ed. 1988) (hereinafter cited as “Phillips”). In the landmark case of Smyth v. Ames, 169 U.S. 466, 18 S.Ct. 418, 42 L.Ed. 819 (1898), the United States Supreme Court established the legal basis of the so-called “fair value” doctrine: [T]he basis of all calculations as to the reasonableness of rates to be charged by a corporation ... must be the fair value of the property being used by it for the convenience of the public. And in order to ascertain that value, the original cost of construction, the amount expended in permanent improvements, the amount and market value of its bonds and stock, the present as compared with the original cost of construction, the probable earning capacity of the property under particular rates prescribed by statute, and the sum required to meet operating expenses, are all matters for consideration, and are to be given such weight as may be just and right in each case. We do not say that there may not be other matters to be regarded in estimating the value of the property. What the company is entitled to ask is a fair return upon the value of that which it employs for the public convenience. Id. at 546-47, 18 S.Ct. at 434. Smyth v. Ames commanded that both the original cost of construction, on the one hand, and the current reproduction or replacement cost, on the other hand, must be “considered” in setting rates. As debate raged as to which of the two cost measures should receive the greater weight or emphasis, the Smyth fair-return doctrine received increasing criticism over the years. Finally, Smyth was abandoned by the Supreme Court in Federal Power Comm’n v. Hope Natural Gas Co., 320 U.S. 591, 64 S.Ct. 281, 88 L.Ed. 333 (1944). The Court in Hope held that regulatory commissions are not bound by any particular formula in determining rates, as long as the rates established “enable the company to operate successfully, to maintain its financial integrity, to attract capital, and to compensate its investors for the risks assumed.” 320 U.S. at 605, 64 S.Ct. at 289. The Texas experience roughly paralleled that of the federal system. In Railroad Commission v. Houston Natural Gas Corporation, 289 S.W.2d 559 (Tex.1956), after a thorough historical review, the Texas Supreme Court held that pre-PURA statutes mandated a fair-value method of valuation, which the court defined as “a reasonable balance between original cost less depreciation and replacement cost new less an adjustment for present age and condition.” Id. at 572. As originally adopted in 1975, PURA incorporated this fair-value definition. 10 In 1983, however, the legislature amended PURA to make Texas a pure “original-cost” state. Section 41(a) of PURA now provides: *911 Sec. 41. The components of invested capital ... shall be determined according to the following rules: (a) Invested Capital. Utility rates shall be based upon the original cost of property used by and useful to the public utility in providing service including construction work in progress at cost as recorded on the books of the utility. The inclusion of construction work in progress is an exceptional form of rate relief to be granted only upon the demonstration by the utility that such inclusion is necessary to the financial integrity of the utility. Construction work in progress shall not be included in the rate base for major projects under construction to the extent that such projects have been inefficiently or imprudently planned or managed. Original cost shall be the actual money cost, or the actual money value of any consideration paid other than money, of the property at the time it shall have been dedicated to public use, whether by the utility which is the present owner or by a predecessor, less depreciation. PURA § 41(a) (emphasis added). In addressing the arguments made by the City and OPC in the present case, we deem it convenient to discuss separately the two different types of costs for which the Commission allowed deferred-accounting treatment: (1) carrying costs, and (2) operating and maintenance costs. A. Carrying Costs. 11 As a general rule, the only assets that may be included in a utility’s rate base (so that the utility earns a return on the value of such assets) are those found to be “used and useful” in providing service to the utility’s customers. As quoted above, for example, section 41(a) of PURA specifically states that rates must be based on “the original cost of property used by and useful to the public utility in providing service.” When a new plant is built, the utility must invest large amounts of capital during construction. Until the plant is completed, however, it is usually not considered a used and useful asset. Accordingly, a rigid application of the used-and-useful rule could prohibit the utility from earning a return on this invested capital until the new plant is completed and its cost is included in rate base by the regulatory authority. Thus, under such a rigid application, equity capital that had been or *912 could have been earning a return for the utility would, when devoted to construction of the new plant, be unable to earn a return until the new plant was completed and its cost included in rate base; further, any interest actually paid on borrowed funds would not earn a return, even though the payment of such interest might have required the investment of additional capital. It has been widely if not universally conceded that utilities should, in fairness and occasionally out of economic necessity, be compensated for these carrying costs, especially for major capital construction projects. Two methods have been developed to compensate utilities for such costs. The first method capitalizes the carrying charges incurred during the construction period as allowance for funds used during construction (AFUDC). AFUDC is recorded part as current income, part as an offset to interest expenses, but no cash payments are made by ratepayers during construction. The payments from ratepayers to recover the carrying charges begin when the completed plant goes on stream. The entire cost of the plant (including AFUDC) is added to rate base, and it earns a rate of return on investment and is depreciated over the life of that plant. James Bonbright, et al., Principles of Public Utility Rates 246 (2nd ed. 1988) (hereinafter cited as “Bonbright”). The second method, as the Bonbright treatise explains it, is to include construction work in progress (CWIP) in the rate base. (CWIP includes accrued AFUDC on investment not in rate base.) The regulatee recovers its carrying charges currently from ratepayers through the return component of its rates, rather than adding them to the cost of construction for recovery when the plant is in service. The return on CWIP is recorded as income on a current basis (like AFUDC), and actual cash payments are made by the ratepayers currently (unlike AFUDC). Id. Section 41(a) of PURA expressly permits the inclusion of CWIP in rate base where the utility demonstrates that “such inclusion is necessary to the financial integrity of the utility,” and CWIP may be included only to the extent that the project has not been “inefficiently or imprudently planned or managed.” PURA § 41(a). In the present case, CWIP was not requested; rather, EPEC accrued AFUDC in a capital account while the plant was under construction. When the new plant began commercial operation, FERC accounting rules required EPEC to cease accruing AFUDC in a capital account; any such costs that continue after commercial operation begins must thereafter be recorded as expenses. Except for deductions for imprudence, the “original cost” of the plant, including AFUDC, was included by the Commission in EPEC’s rate base. In addition, however, the Commission allowed EPEC to defer, and later included in rate base, the carrying costs that EPEC incurred between the date of commercial operation and the effective date of the new rates that included in rate base the “original cost” of the Palo Verde plant. These carrying costs appear to be simply a continuation of AFUDC under a different name. OPC contends that the Commission’s action violated the provision in section 41(a) permitting only the cost of the new plant “at the time it shall have been dedicated to public use” to be included in rate base. We agree. The legislature has made it clear in section 41(a) that the value of new plant is, for rate-base purposes, to be measured by its original cost at the time the plant is dedicated to public use. As stated above, it has been generally recognized that carrying costs associated with the construction of a new plant are essentially part of the “original cost” of constructing the plant, and the utility should be compensated for them by including at least part of those costs in rate base. Nonetheless, the legislature apparently chose to simplify the calculation of a plant’s original cost by placing a “cut-off date” on construction and acquisition costs: such costs must be calculated as of the *913 time the physical asset being constructed or acquired is placed in public service. 12 As stated earlier, the post-in-service carrying costs that were allowed to be “deferred” and which were included in rate base in the present case are indistinguishable from the AFUDC that was properly accrued and capitalized before commercial operation began. Accordingly, any procedure that permits such costs to be included in rate base would effectively allow the inclusion in rate base of a construction cost of the plant that was incurred after the plant’s dedication to public use, thereby violating the mandate of section 41(a) that the original cost of new plant be calculated as of the date the plant is placed in public service. In effect, such a procedure would, by an accounting device, permit the Commission to let in through the back door what the legislature has expressly prohibited coming in the front door. Even without the unique wording of section 41(a), the Washington Utilities and Transportation Commission reached the same conclusion when faced with a request to extend the period of capitalization of AFUDC from the in-service date of a new plant to the date when new rates went into effect: [AJccrual of AFUDC after the in-service date of a utility plant would result in a utility plant with a value exceeding its “original” cost. The original cost concept requires that the value of utility plant be determined at the time it is first placed in service to the public. To grant this petition would establish a dangerous and unwarranted precedent leading to further requests to disregard the original cost concept. In re Puget Sound Power & Light Co., 62 PUR4th 436, 440 (Wash. Util. & Transp. Comm’n 1984). Whether or not one agrees that the original-cost method of valuation requires, as a general proposition, that the value of a utility plant must be determined at the time it is first placed in service to the public, the language of section 41(a) clearly mandates that approach. Accordingly, the Commission contravened section 41(a) when it allowed post-in-service carrying costs to be included in EPEC’s rate base. EPEC and the Commission present several arguments against such a construction of section 41(a). First, they argue that the phrase “at the time it shall have been dedicated to public use” does not mean the time the plant itself is placed in service. They assert, instead, that the phrase refers to the money spent to construct the plant, and that such money is dedicated to public use at the time it is spent. We cannot agree with this interpretation of section 41(a). For example, section 41(a) states that “original cost” is “the actual money cost... of the property at the time it shall have been dedicated to public use.” Thus, the reference to “property” in section 41(a) is obviously to property being acquired or constructed in exchange for the payment of money, not to the funds themselves used to pay for its acquisition or construction. Just as clearly, the term “it” in the phrase “at the time it shall have been dedicated to public use” refers back to the “property” being acquired or constructed. Thus, in the case of new plant, section 41(a) requires that the plant’s original cost be determined as of the time the new plant is placed in service. EPEC and the Commission next argue that “dedicated to public use” does not refer to the time a plant begins commercial operation. We disagree. Having determined that the “cut-off date” contained in section 41(a) refers to the property being acquired or constructed, and not to the money used to pay for its acquisition or construction, the question becomes: When is a new plant dedicated to public use? We conclude that new plant is dedicated to public use when it is first placed in public service. First, the plain meaning of the statutory provision supports the proposition that a plant has not been “dedicated to *914 public use” until it has been placed in public service, and a plant is placed in public service when it begins operating commercially. Second, under FERC rules and Commission practice, a utility must cease accruing AFUDC when a new plant begins commercial operation. Simple logic dictates that the most appropriate time to determine the original cost of a capital asset is when existing accounting rules require that a significant, ongoing cost of that asset cease being capitalized and start being expensed. In this connection, EPEC and the Commission also argue that section 41(a) does not contain any temporal limitation (i.e., “cut-off date”) on the determination of the original cost of capital assets. They stress that section 41(a) provides that original cost is the actual money cost of property “at the time it shall have been dedicated to public use, whether by the utility which is the present owner or by a predecessor.” They contend’ that the emphasized clause above shows that the purpose of section 41(a) is simply to prevent utilities from selling or transferring a plant to another utility and having the purchasing utility use its purchase price as the original cost of the plant. Thus, they argue, the last sentence of section 41(a) merely requires that original cost be the cost to whichever utility first placed the plant in public service, not that original cost must necessarily be determined at the specific time that the plant was placed in service. We disagree. The language of section 41(a) could hardly be clearer in this regard: “Original cost shall be the actual money cost ... of the property at the time it shall have been dedicated to public use_” The clause that follows, “whether by the utility which is the present owner or by a predecessor,” is simply one of clarification, emphasizing that the time of dedication to public use is the critical date, irrespective of whether that dedication was made by the current owner or a predecessor. Ignoring the statute’s plain language, EPEC cites Office of Consumers’ Counsel v. Public Utilities Commission, 18 Ohio St.3d 264, 480 N.E.2d 1105 (1985), in which the Ohio Supreme Court construed the relevant Ohio statute to mean that original cost would be the cost “to the person that first dedicated the property to the public use”; the court went on to hold that the statute “establishes which entities’ costs are to be utilized in establishing a rate base. [It] does not affect the timing of property valuation.” Id. 480 N.E.2d at 1107. A comparison of the Ohio statute with the Texas statute, however, shows why the Texas statute cannot rationally be given the same construction. The Ohio statute provided: “Such original cost of property ... shall be the cost, as determined to be reasonable by the commission, to the person that first dedicated the property to the public me_” Ohio Rev.Code Ann. § 4909.-05(E) (emphasis added). The Texas statute provides: “Original cost shall be the actual money cost ... of the property at the time it shall have been dedicated to public use, whether by the utility which is the present owner or by a predecessor, less depreciation.” PURA § 41(a). The two statutes could not be more different in their focus and meaning. The Ohio statute focuses on “who”; the Texas statute focuses on “when.” Accordingly, the Office of Consumers’ Counsel case is inapposite. EPEC and the Commission next argue that our construction of section 41(a) will prevent the inclusion in rate base of recognized elements of invested capital, such as working capital, “accumulated deferred federal income tax,” and rate case expenses, none of which have a “commercial operation” date or an “in-service” date. We disagree that our holding will have such an effect. As stated above, the purpose of section 41(a) was to establish a method of valuing tangible property acquired or constructed by the utility. Although the term “property” can, in an appropriate context, certainly have a meaning broader than just tangibles, the history of the original-cost/replacement-cost debate as to the proper method of valuing a utility’s invested capital indicates that the crux of the dispute has related primarily, if not exclusively, to plant-in-service. Indeed, the supreme court in the Alvin Case held that “the Texas statutes require a physical *915 property valuation rate base.” Houston Natural Gas, 289 S.W.2d at 564 (emphasis added). We recognize that a utility’s invested capital — and therefore its rate base — can include more than plant-in-service. For example, one noted commentator identifies the following four elements of a utility’s rate base: (1) “tangibles, which includes ‘used and useful’ land, buildings, and equipment (plant)”; (2) “other elements of value, [which] includes working capital, property held for future use, and intangibles”; (3) “customer contributions and tax deferrals, [which are] frequently deducted from the rate base, since those components do not represent investor-supplied capital”; and (4) “construction work in progress.” Phillips, supra at 302. The Bonbright treatise also identifies four elements of rate base, although it arranges the categories somewhat differently: “(1) net plant in service; (2) property held for future use; (3) working capital; and (4) construction work in progress (CWIP) — no AFUDC.” Bon-bright, supra at 237. Although “invested capital” can include more than tangible assets, it is simply not feasible to apply the original-cost formula contained in section 41(a) to certain types of assets, e.g., intangibles and working capital. Such assets have no money “cost” by which they are acquired or constructed; indeed, in some instances they more closely resemble the payment of money than a tangible asset for which money is paid. Nonetheless, we do not believe the legislature intended for section 41(a) to limit a utility’s rate base to the original cost of tangible assets, and we do not so hold. We hold only that the original-cost formula (“the actual money cost ... of the property at the time it shall have been dedicated to public use”) states a mandatory method for the valuation of tangible assets, i.e., plant-in-service. This holding neither addresses nor affects the issue of whether — and to what extent — other types of assets may be included in rate base. EPEC and the Commission next argue that sections 2, 16, 27, and 39 of PURA grant broad enough powers to the Commission to allow it to use “deferred accounting” procedures. Without discussing those statutory provisions in detail, we note our general agreement that they grant broad power and discretion to the Commission. However, they do not expressly authorize inclusion of post-in-service carrying costs in rate base, and we cannot construe them to impliedly permit an action that is contrary to or inconsistent with another section of PURA. Sexton, 720 S.W.2d at 137-38. And as we have held, allowing post-in-service carrying costs to be included in rate base is inconsistent with section 41(a). 13 We conclude, therefore, that the Commission exceeded its authority when it included in EPEC’s rate base the carrying costs incurred by EPEC after the Palo Verde plant began commercial operation. B. Operating and Maintenance Costs. 1. PURA § 41(a) The foregoing discussion makes it clear that we consider the purpose of PURA § 41(a) to be the establishment of a method of determining the value of tangible capital assets that a utility has acquired or constructed. Section 41(a) prohibits post-in-service carrying costs from being included in rate base because carrying costs constitute part of the actual money cost of acquiring or constructing new plant. Operating and maintenance (0 & M) costs, on the other hand, are not part of the cost of acquiring or constructing new plant; rather, they are expenses associated with maintaining the plant after it is already in service. To illustrate the distinction, if the Palo Verde plant had been completely shut down and abandoned the day after it became operational, EPEC’s carrying costs *916 would have continued unabated until all funds borrowed for its construction were repaid; the 0 & M costs, however, would have ceased. Because post-in-service 0 & M costs are not part of the “actual money cost” of acquiring or constructing the plant, the original-cost formula contained in section 41(a) simply has no application to such expenditures. Accordingly, whatever other objections may be made to the inclusion in rate base of post-in-service 0 & M costs, such inclusion is not inconsistent with PURA § 41(a). 2. Retroactive Ratemaking Initially, we note that EPEC applied for and received from the Commission permission to defer post-in-service 0 & M costs on Palo Verde Unit 1 before that unit became commercially operational. Accordingly, we question whether the inclusion of Unit 1 0 & M costs has any retrospective effect at all. Indeed, the City does not even lodge a retroactive-ratemaking complaint about the inclusion in rate base of 0 & M costs as to Unit 1. However, because we must address whether the inclusion in rate base of 0 & M costs for Unit 2 constitutes improper retroactive ratemaking, we will assume without deciding that the Commission’s inclusion of Unit 1 0 & M costs in rate base did have a retrospective effect. As stated previously, sections 2, 16, 27, and 39 of PURA expressly grant broad powers to the Commission. We believe those provisions give the Commission discretionary authority to allow deferral and capitalization of post-in-service 0 & M costs, and to permit the Commission to include such costs in rate base, unless such a procedure is inconsistent with other state law. Thus, to determine the validity of the Commission’s action in the present case, we must determine whether the deferral, capitalization, and inclusion in rate base of such costs is inconsistent with a statutory or constitutional prohibition of retroactive ratemaking. See Texas Ass’n of Long Distance Tel. Cos. (TEXALTEL) v. PUC, 798 S.W.2d 875, 881-82 (Tex.App.1990, writ denied); Southwestern Bell Tel. Co. v. PUC, 615 S.W.2d 947, 953 (Tex.Civ.App.), writ ref'd n.r.e., 622 S.W.2d 82 (Tex.1981). a. Statutory prohibitions. In order to satisfy the first prong of the retroactivity test, the action allegedly having retrospective effect must not contravene any statutory prohibition. As stated above, we have concluded that the deferral, capitalization, and later inclusion in rate base of post-in-service 0 & M costs is not contrary to PURA § 41(a). Appellants also contend, however, that inclusion of such 0 & M costs is inconsistent with PURA § 43(f). Section 43(f) provides that if, after hearing, the Commission finds the existing rates to be unreasonable or in violation of law, it shall fix new rates “by order,” which rates are “thereafter to be observed until changed.” Although many jurisdictions have construed the term “thereafter” to give the regulatory authority power to prescribe rates prospectively only, the Texas Supreme Court stated in one case that the term in section 43(f) gives Texas agencies “discretion” in setting the effective date of new rates. See Railroad Comm’n v. Lone Star Gas Co., 656 S.W.2d 421, 425-26 (Tex.1983). 14 Nonetheless, this *917 Court has held that PURA § 43(f) prohibits the Commission from making new rates effective at a date earlier than the date of the order fixing those rates. See PUC v. GTE-SW, 833 S.W.2d 153 (Tex.App.—Austin, 1992, writ denied); PUC v. General Tel. Co., 777 S.W.2d 827 (Tex.App.1989, writ dism’d); cf. TEXALTEL, 798 S.W.2d at 882-84 (rates may be made effective after order fixing the level of revenues but before final approval of tariffs). In the present case, the effective date of the new rates was not prior to the date of the order fixing the rates. Therefore, the Commission’s action here was not inconsistent with our holdings in GTE-SW and General Telephone. Appellants argue, however, that the inclusion in rate base of deferred 0 & M costs had the effect of implementing the new rates as of the date the Palo Verde plant became commercially operational, i.e., retroactively. We decline to construe the term “thereafter” in section 43(f) to have such sweeping effect. We are not willing to say that the term “thereafter” in PURA § 43(f) constitutes a blanket prohibition of any consideration by the Commission of a utility’s past gains or losses in fixing future rates. Thus, even if the “thereafter” language in section 43(f) precludes the Commission from making new rates effective at a date earlier than the date of the order fixing those rates, the Commission’s action in the present case — permitting deferral, capitalization, and inclusion in rate base of EPEC’s post-in-service 0 & M costs — is not inconsistent with such a prohibition. We conclude, therefore, that the Commission’s inclusion in rate base of EPEC’s deferred post-in-service 0 & M costs is not inconsistent with Texas statutory law. b. Constitutional prohibition. To withstand the second prong of appellants’ retroactive-ratemaking challenge, the Commission’s order must not violate Article I, § 16 of the Texas Constitution, which provides: “No bill of attainder, ex post facto law, retroactive law, or any law impairing the obligation of contracts shall be made.” Courts often recite the rule that ratemaking is a legislative activity, even when delegated to an administrative body. See, e.g., Houston Natural Gas Corp., 289 S.W.2d at 563. For that reason, it has often been stated that rates set after an agency hearing generally must have a prospective effect, just as would laws enacted by the legislature. 15 Id.; see also Tex. Const. Ann. art. I, § 16 (1984). Appellants complain that allowing EPEC to recover post-in-service 0 & M costs through deferred accounting would permit EPEC to charge ratepayers an additional amount for services that they have already received and paid for. In this regard, the following general observations about retroactive laws are instructive: [One,] a law is retroactive if it assumes to give effect to a past event, in order to create a present right or duty. [Two,] ... a law is retroactive when it assumes to give to a past event the effect of creating rights and duties ab initio, or as of some time prior to the retroactive law.... ... Under limitations [one and two], it is true, a law gives to an event which has already transpired a juristic significance it did not have at the time it occurred; but it does this only in order that it may thereby regulate future conduct. Bryant Smith, Retroactive Laws and Vested Rights, 5 Tex.L.Rev. 231-33 (1927). In addition, the supreme court has recognized that statutes permitting agencies to consider prior conduct have retrospective effect. Texas Water Rights Commission v. Wright, 464 S.W.2d 642, 648-49 (Tex.1971). In the present case, the past event given significance by inclusion in rate base of the deferred post-in-service 0 & M costs is the *918 start-up of commercial operation of Units 1 and 2. Appellants assert that allowing EPEC to earn a return on such deferred-cost assets gives the start-up of those units the effect, ab initio, of creating a new duty for ratepayers: they are thereafter required, without Commission approval, to pay higher rates for services received. Thus, recovery of such 0 & M costs would, in appellants’ view, effect a change in charges after ratepayers have consumed the service. See, e.g., Lone Star Gas, 656 S.W.2d 421. For purposes of the ensuing discussion, we will assume without deciding that inclusion in rate base of the post-in-service 0 & M costs has a retrospective effect. A retrospective effect alone, however, will not invalidate an agency action. Wright, 464 S.W.2d at 648. Notwithstanding a retrospective effect, a rate order may avoid constitutional infirmity if it does not substantially impair or destroy vested rights, McCain v. Yost, 155 Tex. 174, 284 S.W.2d 898, 900 (1955); TEXALTEL, 798 S.W.2d at 882, and if it does not change the substantial rights and obligations of the implied contract between a utility and its ratepayers, Southwestern Bell, 615 S.W.2d at 956; Amarillo Gas Co. v. City of Amarillo, 208 S.W. 239, 240 (Tex.Civ.App.1919, no writ). Although courts have often failed to explicate or apply the latter consideration, we conclude that it is significant in the present case. Several courts have concluded that an implied contract exists between a utility and its ratepayers, creating both the utility’s duty to provide a defined service and the ratepayers’ duty to pay a defined rate. See, e.g., Amarillo Gas Co., 208 S.W. at 240; Southwestern Bell, 615 S.W.2d at 956. Under this implied contract, ratepayers have a right to pay a constant rate for service until, by legislatively approved procedures, the old rate is formally challenged. See TEXALTEL, 798 S.W.2d at 882. The setting of new rates permissibly adjusts the respective rights and obligations of the ratepayers and the utility. Only the rights and obligations existing between rate settings are constitutionally protected against alteration by retroactive ratemaking. 208 S.W. at 240. Therefore, only if new rates alter the ratepayers’ right to pay a set rate for a specified service will they violate the prohibition of article I, § 16. The Amarillo Gas case, involving city ordinances which set consumer gas rates, typifies true retroactive ratemaking. Among other things, the first ordinance included a provision which allowed consumers a 10% discount if they paid their bills within ten days. The subsequent ordinance eliminated the discount and imposed a 10% surcharge on payments made after ten days. A dispute arose when the gas company attempted to collect surcharges on bills for gas the company had supplied before the effective date of the second ordinance. The question on appeal was whether imposing the surcharge on bills for gas supplied before the effective date “ehange[d] the substantial rights and obligations of this contract as to transactions already had under it.” 208 S.W. at 240. The court concluded that the practical and necessary result of the amendment was to require the consumer to pay considerably more for his gas than he would have been required to pay under the old rates_ The rates established by the new ordinance did inevitably make a substantial change in the rights and obligations of the consumer, and we conclude [that they] cannot be applied to the gas consumed prior to the time the ordinance took effect. Id. Unlike Amarillo Gas, the present case involves no attempt to charge ratepayers an additional sum for service already purchased. Before the effective date of the new rates, EPEC’s ratepayers paid Commission-authorized rates for the service they were obtaining. EPEC invested in the Palo Verde plant to equip itself to provide additional service to its ratepayers. When the new plant became commercially operational, the ratepayers began receiving the benefit of the new service before being charged for it. Therefore, because the Commission had not defined the substantial *919 rights and obligations of a new implied contract between EPEC and its ratepayers, the ratepayers had no substantial right to pay a certain rate for service being provided by the Palo Verde plant. Significantly, the ratepayers also had not paid rates which would allow EPEC a return on its investment in the plant. Therefore, including in new rates costs incident to the interim benefit provided by the operation of Palo Verde plant does not change a substantial right belonging to the ratepayers. Cf. Business & Prof. People for the Pub. Interest v. Illinois Commerce Comm’n, 205 Ill.App.3d 891, 150 Ill.Dec. 750, 563 N.E.2d 877 (1990) (deferral of regulatory-lag costs allowed because costs of new plant had not been taken into account in setting existing rates), rev’d on other grounds, 146 Ill.2d 175,166 Ill.Dec. 10, 585 N.E.2d 1032 (1991). Just as allowing inclusion in rate base of the Palo Verde plant’s post-in-service O & M costs does not change the ratepayers’ substantial rights, likewise it does not impair or destroy vested rights. Whether legislation substantially impairs or destroys vested rights necessitates consideration of whether the retrospective effect: (1) advances or retards the public interest; (2) effectuates or defeats the bona fide intentions or reasonable expectations of affected persons; and (3) surprises persons who have long relied on a contrary state of the law. Southwestern Bell, 615 S.W.2d at 956-57; see also Wright, 464 S.W.2d at 648. The public has an interest in obtaining a reasonable quantity and quality of service. The utility should generate the service safely, under the guidance of efficient management, and make the service obtainable at reasonable rates. See Phillips, supra, at 164. In the present case, the Commission could reasonably have concluded that including post-in-service O & M costs in rate base would advance these interests. Further, a ratepayer could not reasonably expect a utility to spend millions of dollars building a nuclear facility, use the facility to generate electricity, and then not seek a return on its investment therein. In addition, the fact that the Commission had previously granted EPEC a certificate of convenience and necessity to participate in the project thereafter precluded any interested persons from reasonably claiming surprise at finding themselves obligated to pay the costs of building and operating the new plant. All of these important considerations support the conclusion that the Commission has not substantially impaired or destroyed vested rights by including in rate base EPEC’s post-in-service O & M costs. Therefore, because the new rates neither impair vested rights nor change substantial rights or obligations of implied contract, we conclude that deferral, capitalization, and inclusion in rate base of such costs does not violate the constitutional prohibition against retroactive ratemaking. C. Disallowance for Imprudence. OPC also argues in point of error five that the Commission erred by not reducing the deferred O & M costs in proportion to the imprudence disallowance before including the assets in rate base. OPC cites no authority that would require the Commission to reduce the “deferred-costs asset” because of imprudence. Instead, it points to the Commission's reservation of the right to exclude the capitalized costs from rate base contained in the order allowing deferral of Unit 1 costs. Apparently, OPC has construed this reservation as a representation that the costs would not be included; we do not find any such representation in the order. D. Substantial Evidence. As its final complaint within this point of error, OPC contends that the total figure assigned by the Commission to the deferred-costs asset is not supported by substantial evidence. We do not agree. Our review of the record shows that EPEC provided ample documentation of the costs it had incurred and deferred after the Palo Verde plant became commercially operational. We sustain OPC’s and the City’s fifth points of error to the extent they complain *920 of the Commission’s deferral, capitalization, and inclusion in rate base of EPEC’s post-in-service carrying costs. We overrule OPC’s and the City’s fifth points of error to the extent they complain of the Commission’s deferral, capitalization, and inclusion in rate base of EPEC’s post-in-service 0 & M costs, and in all other respects. CONSTRUCTIONAL IMPRUDENCE By its third point of error, OPC raises three distinct complaints about the disal-lowance of $28 million in construction costs as a result of EPEC’s imprudent planning or management of the project’s construction. First, OPC asserts that substantial evidence does not support the Commission’s method of quantifying the imprudently incurred costs. Second, OPC argues that the quantification method used by the Commission results in retroactive ratemak-ing. Finally, OPC complains that “[t]he Commission’s finding regarding construction cost impacts is conclusory and does not indicate the underlying facts relied upon by the PUC.” We construe this contention to be a challenge to the sufficiency of Finding of Fact 100 and an assertion that the Commission should have made additional findings of underlying facts to support Finding 100. EPEC maintains that OPC has waived all of its contentions except that concerning retroactive ratemaking because it failed to present legal bases for its complaints in its second motion for rehearing. We do not agree. Although OPC’s second motion for rehearing contained no citations to legal authority for its complaints, the motion did contain statements of OPC’s legal bases for its argument. While motions for rehearing must point out the specific finding challenged and the legal basis for the challenge, they need not contain citations of authority. Burke v. Central Educ. Agency, 725 S.W.2d 393, 397 (Tex.App.1987, writ ref’d n.r.e.). A. Retroactive Ratemaking. OPC asserts that by disallowing a smaller portion of construction costs than OPC recommended, “the Commission has retroactively increased the value of the Company’s revenue requirement fixed in prior rate proceedings, and shifted these additional hypothetical costs to future ratepayers.” Essentially, OPC argues that the Commission may not legally adopt a method recognizing any savings that ratepayers may have realized because of delays in completing construction. The Commission staff’s expert, Morris Jacobs, and the City’s expert, Richard B. Hubbard, agreed that financing costs had increased because of delays. Jacobs also testified, however, that ratepayers had received an unexpected benefit because of the delay: they had had the present use of money they would otherwise have had to pay in rates if the construction had been completed and the costs included in rate base as scheduled. Consequently, according to Jacobs, the true amount of imprudently incurred costs can be determined only by reducing the increased financing costs by the amount of the benefit to ratepayers. OPC objects to offsetting the ratepayers’ benefit against the increased financing costs, apparently because the former accrued before the Commission could set rates including the plant in rate base. Accordingly, OPC argues that the savings cannot be taken into account without having a retrospective effect on rates. We fail to see how recognizing an actual benefit ratepayers derive from delays would impermissibly increase revenue requirements “fixed in prior rate proceedings.” We conclude that by offsetting such a benefit against increased financing costs in determining the constructional-imprudence disallowance, the Commission did not change the substantial rights of the ratepayers or impair or destroy any of their vested rights. See McCain v. Yost, 155 Tex. 174, 284 S.W.2d 898, 900 (1955); Amarillo Gas Co. v. City of Amarillo, 208 S.W. 239, 240 (Tex.Civ.App.1919, no writ). Therefore, we reject OPC’s retroactive-ratemaking argument. B. Substantial Evidence. Finding of Fact 100 provides: “Staff witness Jacobs presented a credible quantifica *921 tion of construction management imprudence related to costs of delay in the amount of $28 million.” The gist of OPC’s purported substantial-evidence argument is that the Commission erred in adopting Jacobs’s method of quantifying imprudently incurred costs. This is not, however, a substantial evidence challenge. OPC has not attempted to show that application of Jacobs’s method results in a figure unsupported by substantial evidence; nor does OPC argue that substantial evidence supports another, but different, figure. We must, therefore, make two distinct determinations: (1) whether the finding underlying the Commission’s $28 million disallowance of imprudently incurred construction costs is supported by substantial evidence, and (2) whether Jacobs’s method considered all imprudently incurred costs. Turning to the latter first, we recognize that we must not disturb an agency’s exercise of discretion unless it is arbitrary or unreasonable, Murphy v. Rowland, 609 S.W.2d 292, 297 (Tex.Civ.App.1980) and that we must allow the agency some leeway to select the method by which it carries out its own legislative mandate, Railroad Comm’n v. Humble Oil & Refining Co., 193 S.W.2d 824, 833 (Tex.Civ.App.1946, writ ref’d n.r.e.), aff'd, 331 U.S. 791, 67 S.Ct. 1523, 91 L.Ed. 1820 (1947). Therefore, we will not reverse the Commission’s decision to use Jacobs’s method unless OPC can show that the Commission made its decision arbitrarily and unreasonably. Jacobs did not relate any particular imprudent construction management actions to any specific delays. Jacobs’s testimony supports the amount of the Commission’s disallowance. In addition, other witnesses testified that some construction delays were unavoidable and not the result of management imprudence. Although the City’s witness, Hubbard, supplied testimony linking specific construction decisions with resulting delays, we conclude that substantial evidence supports Jacobs’s quantification method. C. Finding of Fact 100. OPC challenges the sufficiency of Finding of Fact 100 and alleges that it lacks necessary findings of underlying facts. Finding 100 is not a finding “set forth in statutory language” such that it must have a “concise and explicit statement of the underlying facts.” See AP-TRA § 16(b). Instead, it is itself a finding of underlying fact which supports Finding 99. Finding 99 is phrased in statutory language and states that EPEC was imprudent to some degree. Finding 100 concisely sets out the underlying facts that (1) based on Jacobs’s credible quantification, (2) $28 million in construction costs would be disallowed. OPC contends the Commission was obligated to explain its adoption of Jacobs’s method rather than Hubbard’s. In addition, OPC claims the Commission had a duty to identify individual instances of construction imprudence. Appellants have not cited any authority that would require the Commission to explain why it found a particular witness’s testimony credible or determined a particular figure to have resulted from imprudent construction management. We consider these arguments to be challenges to the sufficiency of the finding. APTRA § 16(b) does not delineate a standard for sufficient findings of underlying fact. The supreme court articulated the established principles in Charter Medical: The characteristics of proper findings of fact, as well as their purposes, are well established. Valid findings of fact must be clear and specific. A mere conclusion or a recital of evidence is inadequate. The required underlying facts may not be presumed from findings of a conclu-sional nature. In general, underlying findings of fact must be such that the reviewing court can fairly and reasonably say that the underlying findings support the statutorily required criteria. * * * * * * Proper underlying (basic) findings of fact should follow the guidelines we previously have noted: they should be clear, specific, non-conclusory, and supportive of the ultimate statutory finding. Mere recitals of testimony or references to or summations of the evidence are improp *922 er. Such findings should be stated as the agency’s findings. The findings should relate to material basic facts and should relate to the ultimate statutory finding that they accompany. Charter Medical, 665 S.W.2d at 451-52 (citations omitted); see also State Banking Bd. v. Allied Bank Marble Falls, 748 S.W.2d 447 (Tex.1988). The supreme court has recently reconsidered the question of sufficiency of underlying fact findings. See Goeke v. Houston Lighting & Power Co., 797 S.W.2d 12 (Tex.1990). The court expressed the opinion that, although there is no precise form for an agency’s articulation of underlying facts, certain “guidelines” exist to prevent the types of abuses courts have found. 16 Those specific guidelines are that the findings: (1) must be more than mere recitals of testimony; (2) should be stated as the agency’s findings; and (3) should relate to the ultimate statutory findings. Id. at 15. We will review appellants’ sufficiency complaints under the Goeke guidelines. Finding 100 is not conclusory; it indicates not only the evidence on which the Commission relied in making Finding 99, but also the Commission’s conclusion that the figure found using the method was credible. The finding obviously supports Finding 99, an ultimate statutory finding. Therefore, Finding 100 is a sufficient finding of underlying fact. For all of the foregoing reasons, we overrule OPC’s third point of error in its entirety. THE EXCLUSION OF HUBBARD’S TESTIMONY In support of its request that plant construction costs be included in rate base, EPEC offered evidence of prudent construction management. In response, the City called Hubbard, who testified that EPEC had managed the construction imprudently and that, as a result, construction costs had been unreasonably high. EPEC moved to strike sections of Hubbard’s testimony, arguing that his conclusions were inadmissible because they were speculative and not based on concrete information. The hearing examiner excluded the challenged testimony. The City complains of the exclusion. EPEC answers that the City waived its complaint by failing to state a legal basis to support it in the City’s second motion for rehearing. An administrative litigant may preserve a complaint only by giving the agency an opportunity to review the legal ground on which the complaint is based. Sears v. State Bd. of Dental Examiners, 759 S.W.2d 748, 750 (Tex.App.1988, no writ); Burke, 725 S.W.2d at 397. Our review of the City’s motions for rehearing convinces us that EPEC is correct. The City failed to provide the Commission any basis for finding the testimony admissible. For this reason, we overrule the City’s second point of error. If we had resolved the complaint on its merits, however, we would have found no reversible error. Hubbard testified at length about the duration of design-problem delays and the necessity of reworking the designs. We must assume the Commission considered this testimony and gave it due weight. The examiner excised only minute sections from the thick attachments to Hubbard’s direct testimony. Ample evidence existed from which the Commission could have found that EPEC had managed the construction imprudently. The City has not shown that the exclusion prejudiced its substantial rights, and has therefore failed to carry its burden of showing harmful error. EXCESS CAPACITY The City and OPC, each in its fourth point of error, complain that the Commission erred in finding that EPEC had no system excess capacity. Both appellants contend that Finding of Fact 107, in which *923 the Commission concludes that no present excess capacity exists, is unsupported by substantial evidence in the record. In its argument, the City specifically challenges the action of the Commission in: (1) including in the load determination the amount of power EPEC has contracted to sell to the Texas-New Mexico Power Company (TNP); (2) allowing EPEC to reduce estimated system capacity by retiring three gas-fired units earlier than it had originally planned; (3) rescheduling maintenance because of the alteration in system components; and (4) calculating the reserve requirement. OPC joins in the last of these four specific complaints. In addition, OPC complains that Findings of Fact 111 through 113 do not explicitly state the facts the Commission relied on and the reasoning it used in disregarding the Examiner’s recommendations. OPC also asserts the Commission erred by not making a conclusion of law clarifying the relation, if any, between the “used and useful” standard of PURA § 39(a) and the excess capacity concept. Finally, OPC complains that the Commission “abrogated OPC’s right to represent the interests of residential and small commercial customers on excess capacity issues in future rate cases of El Paso Electric Company.” EPEC’s application to increase rates sought only the inclusion in rate base of costs related to Units 1 and 2. At the time the Commission heard evidence on the application, Unit 3 was not complete and had not begun commercial operation. It would have been improper for the Commission to have determined, at that time, whether excess capacity would exist on EPEC’s system once Unit 3 became operational. The Unit 3 issues, including excess capacity, are not yet ripe for determination. Consequently, to the extent the City seeks resolution of Unit 3’s used-and-useful status, we overrule its point of error for lack of ripeness. A. Substantial Evidence. Appellants complain that the Commission erred in disregarding the examiner’s recommendations for treatment of four disputed issues. The most heated debate arose when the Commission selected the “largest single hazard plus 5%” method for determining reserve requirements. The examiner recommended that the Commission use the “20% of peak load” method.. Use of the examiner’s recommended method would have resulted in excess capacity of approximately 50% of the units EPEC was requesting be included in rate base. Appellants claim the Commission’s use of an improper method was the principal cause of the “no excess capacity” finding. Two of EPEC’s expert witnesses explained EPEC’s method of determining reserve requirements. The company uses one of the methods outlined by the Western Systems Coordinating Council, an organization responsible for promoting reliable operations among the interconnected bulk power system to which EPEC belongs. The Council recommends three different methods of determining reserve requirements, of which the “largest single hazard plus 5%” is one. There was expert testimony that EPEC selected this method because, of the other two possibilities, one would set an unnecessarily high reserve and the other would produce an insufficient reserve to protect against a blackout in the event of a significant loss in system generation capacity. Considering these expert opinions, we cannot say that reasonable minds could not have reached the conclusion the Commission must have reached in order to make the finding it did. We conclude that substantial evidence supports the Commission’s adoption of the “largest single hazard plus 5%” method of determining reserve requirements. The City also complains that the Commission determined the overall demand on the system to be much higher than it should have been, thereby inflating the system capacity found necessary. The City contends the Commission erred by: (1) including in its calculation the power EPEC has contracted to supply TNP; (2) omitting three gas-fired units from generation capability because of plans to retire them early; and (3) adopting a maintenance schedule that requires removal of some units from *924 the line during the summer peak period. Two experts offered their opinions that the Commission acted reasonably in approving the maintenance schedule and including the TNP obligation in peak load. In addition, one witness recognized that EPEC’s objective in planning generation capacity was to “meet forecasted load demands with adequate system reliability while minimizing total system cost.’’ The Commissioners could reasonably have found that adding two nuclear power units to the system and retiring three gas-fired units would achieve this objective. We conclude that substantial evidence in the record supports the Commission’s finding that there is no appreciable excess capacity in the EPEC system. B. Other Contentions. OPC complains that the Commission has abrogated OPC’s right to represent its clients — residential and small commercial users — in future rate cases. OPC bases this contention on Finding 107. The relevant portion of that finding states that the excess capacity findings in the present case will not be considered “precedents in any manner in cases involving the addition of future generating capacity to the system, including Palo Verde Unit 3, or in any reconsideration proceeding conducted pursuant to paragraph 11 of the Amended and Restated Stipulation.” The finding merely recites the Commission’s refusal to address issues not yet ripe for determination. OPC has had a full and fair opportunity to litigate the excess capacity issue with respect to Units 1 and 2. Even without Finding 107, general principles of issue preclusion would bar OPC from relitigating the excess capacity issue with respect to those units. Finding 107 does no more than that. OPC next complains of Findings of Fact 111-113; it asserts that each finding requires additional findings of underlying fact. The challenged findings are as follows: 111.The largest single hazard plus five percent (“LSH + 5”) criterion for determining a reasonable reserve margin is used by EPEC and recommended by the Western Systems Coordinating Council, of which EPEC is a member. 112. Based on the evidence presented, use of the LSH + 5 criterion is reasonable for application to the EPEC system in this case. 113. Using the LSH + 5 criterion, EPEC should carry 258 MW of reserve capacity in 1988. These findings are not ultimate findings; therefore, the Commission had no duty to make additional findings of underlying fact. OPC also argues that the Commission was bound to explain its reasons for rejecting the examiner’s recommendation and adopting a different method of determining reserve requirements. Again, OPC has not shown any authority requiring the Commission to do so. Therefore, we also find this argument to be meritless. In its motions for rehearing to the Commission and at the trial court level, OPC objected to the Commission’s failure to explicate the relationship between the concept of excess capacity and PURA’s “used and useful” standard. We need not resolve this issue in order to dispose of the present case; therefore, we express no opinion on the matter. We overrule the City’s and OPC’s fourth points of error. COMMON FACILITIES As a part of its application to increase rates, EPEC requested that costs incurred in constructing the facilities to be used in common by all of the generating units, including those not yet completed, be included in rate base. The Commission found it reasonable to include such costs and made two findings of fact about which OPC now complains. In its sixth point of error, OPC asserts generally that the Commission erred by deciding to consider the common facilities as “plant-in-service.” OPC argues that the Commission, by refusing to apportion the costs of the common facilities to each unit, has changed its position on apportionment questions and that such a shift in position *925 is improper. OPC also asserts that the findings of fact relating to treatment of common facilities costs are insufficient and that additional findings of underlying fact are necessary to show the evidence on which the Commission relied in determining whether to include the common facilities costs in rate base. OPC has failed to brief adequately its complaint on this issue. Each of its legal contentions comprises but a single sentence, and only in conjunction with its change-of-position complaint has OPC provided any supporting authority at all. OPC also fails to point out evidence in the record demonstrating either that the Commission has erred or that any error has substantially prejudiced OPC’s rights. Finally, OPC cites no authority for the propositions that the Commission was obligated to make more findings of fact than it did, that the Commission had to allocate costs rather than following generally accepted accounting principles, or that OPC was unable to present its appeal adequately because of the Commission’s alleged failure to explain why it declined to apportion the common facilities’ capital costs. Such a complete failure to develop and support a complaint waives the complaint. Helle v. Hightower, 735 S.W.2d 650, 654 (Tex.App.1987, writ denied). We conclude OPC has waived its sixth point of error, and we overrule it. INCOME TAXES In its final point of error, OPC generally challenges the cost-of-service determination on grounds that the amount found includes a “hypothetical” federal income tax expense. In the most cursory of fashions, OPC alleges the Commission erred in including the federal income tax expense in EPEC’s cost of service because: (1) no evidence supports the inclusion; (2) the Commission made no findings of fact or conclusions of law regarding federal income tax expense; and (3) the Commission failed to inquire whether EPEC had actually incurred all of the tax expense. As a matter of interest, we note that OPC has overlooked Finding of Fact 186, which expressly addresses federal income tax expense. Consequently, OPC’s complaint that the Commission made no findings with respect to federal income taxes is merit-less. OPC’s briefing of this point of error is wholly inadequate. The only authority to which OPC has drawn this Court’s attention is the opinion in PUC v. Houston Lighting & Power Company, 748 S.W.2d 439 (Tex.1987). Even more than in its sixth point of error, this multifarious seventh point contains conclusory statements unsupported by authority. In its motion for rehearing, OPC complains that this Court “misunderstood” OPC’s position on this issue; there follows a lengthy exposition of the position OPC intended to argue in its initial brief. The detailed discussion in OPC’s motion for rehearing further underscores the inadequacy of the argument on this point in its original brief. OPC has waived its seventh point by failing adequately to support or to argue the offending issue in its original brief. LEASE PAYMENTS ON UNIT 2 As required by PURA § 63, EPEC notified the Commission of the sale/leaseback arrangement for Unit 2. The sole issue to be determined as to that transaction was whether the sale/leaseback was consistent with the public interest. In January 1987 the hearings examiner stayed proceedings in the sale/leaseback matter so that the Commission could consider the public interest issue along with EPEC’s rate case. In response to a motion filed by EPEC, the Commission consolidated the two matters under the docket number for the rate case. When the Commission rendered a final order, however, it did not decide whether the sale/leaseback transaction was consistent with the public interest; instead, it specifically reserved that issue for later determination. Nonetheless, the Commission made fact findings that allowed Unit 2 lease payments to be included in EPEC’s cost of service to the extent they did not exceed the amount the Commission would have included in rate base for Unit 2 capital costs if EPEC had retained an ownership *926 interest. The partial inclusion of lease payments in cost of service prompts the City’s substantial evidence challenge in its final point of error. PURA § 63 requires a utility to report a contemplated or consummated transaction within a reasonable time if the total transaction consideration exceeds $100,000. Further, the section provides that [o]n the filing of a report with the commission, the commission shall investigate the same with or without public hearing, to determine whether the action is consistent with the public interest. In reaching. its determination, the commission shall take into consideration the reasonable value of the property, facilities, or securities to be acquired, disposed of, merged or consolidated. If the commission finds that such transactions are not in the public interest, the commission shall take the effect of the transaction into consideration in the rate-making proceedings and disallow the effect of such transaction if it will unreasonably affect rates or service. The provisions of this section shall not be construed as being applicable to the purchase of units of property for replacement or to the addition to the facilities of the public utility by construction. PURA § 63 (emphasis added). Pursuant to PURA, the Commission must, at some point, decide whether the sale/leaseback transaction is consistent with the public interest. The statute designates no time within which the Commission must make that determination after the utility files its report. The section expressly allows the Commission to decide the issue without holding a public hearing; consequently, we do not construe PURA to require the Commission to resolve that question in the context of a formal ratemaking proceeding. The Commission has discretion to consolidate proceedings with common issues when consolidation would serve judicial or administrative economy. See Alamo Express, Inc. v. Union City Transfer, 309 S.W.2d 815, 821 (Tex.1958). The City does not deny this, but asserts that once the Commission had consolidated the proceedings, it was powerless to sever them. The City contends that at that point the Commission became bound to settle the public interest question in its order setting rates. This contention fails to recognize the Commission’s discretion to regulate its docket so that only issues which can reasonably and fairly be tried within the framework of a single proceeding are tried together. We conclude the Commission has the power to sever. Any other result would defeat the legislative intent in delegating duties to the Commission for more efficient administration. Therefore, in the interest of accomplishing the legislative purpose underlying the Commission’s creation, we deem it essential that the Commission’s power to consolidate be balanced by a corresponding power to sever. During the hearing, the Commission apparently concluded that “the effect” of the sale/leaseback transaction did not include the entire amount of lease payments made. EPEC would have incurred certain costs even if it had retained its ownership interest in the unit instead of arranging the sale/leaseback. Therefore, even if the Commission were ultimately to find the transaction inconsistent with the public interest, the cost that EPEC would have incurred had it retained ownership would be includable in rates because it was not an “effect” contemplated by the disallowance provision of PURA § 63. Consequently, the Commission did not err in including this amount in cost of service. By post-submission brief, the City argues that deferral of the public-interest determination implies a finding that EPEC failed to carry its burden of proof on that issue. In support of this argument, the City directs our attention to the supreme court’s recent decision in Coalition of Cities for Affordable Utility Rates v. PUC, 798 S.W.2d 560 (Tex.1990). The City did not make this argument below; therefore, it is not properly before this Court. See City of San Antonio v. Texas Water Comm’n, 407 S.W.2d 752 (Tex.1966). Even if we were to agree that the Commission impliedly found EPEC had failed to carry its burden of showing that the transaction was consistent with the public inter *927 est, we would find no reversible error in the inclusion of part of the payments in cost of service. Finding of Fact 122 provides: “EPEC’s proposed ‘book break-even’ calculation of the portion of the lease payment may be included in cost of service in this instance, as it is not in excess of the amount that would result if calculated using the traditional ratemaking plant in service/rate base methodology.” The City fails to recognize that the Commission is responsible not only for determining whether the transaction in question is consistent with the public interest, but also for disallowing the effect of the transaction “if it will unreasonably affect rates or service.” PURA § 63. Only if inclusion of the effect will unreasonably affect rates will it be disallowed. Therefore, Finding 122 implies, consistent with the Coalition analysis, that the Commission found that EPEC carried its burden of showing that inclusion of the relevant portion of the lease payment would not unreasonably affect rates. The City has not challenged this implied finding; therefore, even if the City’s untimely argument were correct, it would not show reversible harm. We overrule the City’s seventh point of error. COST-OF-SERVICE ALLOWANCES The City makes several general complaints about the Commission’s revenue-requirements determination and, in addition, specifically challenges five separate components of the cost-of-service allowance. The general complaints are that: (1) substantial evidence does not support the Commission’s revenue requirements finding; (2) the Commission applied no statutory standard in determining revenue requirements; and (3) the Commission failed to make all required findings of underlying facts. We conclude that, except for the specific challenges to five component amounts, the City has waived its complaints by failing to show that particular cost-of-service component amounts are unsupported by substantial evidence. Nor has the City identified a statutory standard requiring the Commission to supply findings of underlying facts in addition to those already made. Although the City contends that the Commission decided the matter without referring to PUC Substantive Rule § 23.21(b), the findings obviously refer to that rule. Therefore, we overrule the City’s general complaints and proceed to address the challenges to specific component amounts. A. Fuel and Purchased Power Expense. The City contends that the Commission overstated expenses for fuel and “purchased power” because it included in that amount the price of 25 megawatts of electricity not actually purchased by EPEC. Staff witness Stan Kaplan calculated EPEC’s reasonably predictable fuel costs based on the assumption that the new rates would become effective January 1, 1988. In estimating the amount of purchased power costs, Kaplan anticipated that EPEC would purchase 75 megawatts in January 1988 and 50 megawatts per month for the balance of the calendar year. However, the prolonged hearings delayed the new rates’ effective date to a point approximately three months beyond the January 1, 1988, date Kaplan had assumed. Nonetheless, the Commission’s final order was based on Kaplan’s prediction, which included an anticipated first-month purchase of 75 megawatts. The City argues that the cost of 25 megawatts, which EPEC did not purchase after the new rates became effective, should be excluded from cost of service. A utility’s allowable expenses are calculated by adjusting its historical test year expenses for known and measurable changes. 16 Tex.Admin.Code § 23.21(b) (1991). Based on the evidence filed before and the testimony adduced during the hearing, the Commission determines the amount of the utility’s reasonably predictable purchased power costs for the “rate year,” i.e., the first twelve months after the rates will become effective. 16 Tex.Admin.Code § 23.23(b)(2)(B) (1991). This determination inherently involves estimation and the making of a number of assumptions. One necessary assumption is that rates will become effective on some specific *928 date. In the present case, that assumption turned out to be incorrect by three months. On that basis, the City argues that the Commission’s determination of EPEC’s reasonable and necessary operating expenses was invalid. We do not agree. If we were to hold the relevant determination in this case invalid, we would be imposing an onerous burden on the Commission; it would have to recalculate each element of every component of revenue requirements whenever a witness’s assumption that new rates would become effective on a certain date later proved to be incorrect. The recalculation time alone could conceivably delay rendition of a new order long enough once again to alter the effective date. Such a process might never end. We conclude that, under the circumstances of this case, the Commission did not abuse its discretion and did not act arbitrarily or capriciously by refusing to recalculate purchased-power cost once it became apparent that the actual effective date would not coincide with the assumed effective date. We overrule the City’s contention regarding this component. B. Operating and Maintenance Expenses. The City next asserts that the Commission found an improper amount of “operating and maintenance expenses” because it: (1) listed the expense as a single-line item; (2) provided no findings of underlying fact to explain its reasoning in adopting the figure; (3) found a figure unsupported by any evidence; and (4) included rate-case expenses in operating and maintenance expense after having severed them out of the docket. The City points to no duty compelling the Commission to find, as underlying facts, the amounts comprising a sum which is itself a component of a statutorily mandated criterion. PURA directs the Commission to find the amount of “reasonable and necessary operating expenses,” not operating and maintenance expense; likewise, PURA does not expressly mandate consideration of operating and maintenance expense when the Commission determines net income. See PURA § 41(c). In the present case, the Commission found specific amounts for “operating expenses” and for that category’s components, one of which was labelled “operating and maintenance expenses.” Because the Commission had no duty to itemize the subcomponents of “operating and maintenance expenses,” its failure to find them as underlying facts cannot be considered error. See, e.g., Frost v. PUC, 672 S.W.2d 883, 885 (Tex. App.1984, writ ref’d n.r.e.). We conclude the Commission did not act arbitrarily and capriciously in declining to further subdivide the components of operating expense. During the proceeding, EPEC offered evidence of the amount of operating and maintenance expense, to which the City did not object. Although this figure was undisputed, the Commission reduced it before including the amount in revenue requirements as cost of service. We will not reverse the Commission’s order absent a showing that the City’s substantial rights were prejudiced by the inclusion of the reduced expense figures in cost of service. Since the City has not shown harm, we overrule its challenge. Finally, the City contends the Commission improperly included the rate-case cost in operating and maintenance expense. The City bases its claim of error on the alleged prior severance of the rate-case expense issue from the proceeding. In paragraph 16 of its final order, the Commission held that “[t]he issue as to the reasonableness of the Company’s and the Cities’ rate case expense incurred in the prosecution of this case is severed from this docket.” The City argues that, because of paragraph 16, no regulatory commission expense should be included in the revenue requirements. We do not agree. The Commission’s staff provided testimony supporting the findings of that portion of the regulatory commission expense that was undisputed, and the Commission included only these undisputed amounts in cost of service. We conclude that the City’s contentions regarding operating and maintenance expense are meritless. *929 C.Employee Benefits. The City next challenges the Commission’s findings on employee benefits. It asserts that the Commission erred by concluding that the evidence supported staff witness Young’s adjustments to the 401-k plan expenses and the Tax Reduction Act Stock Option Plan (TRASOP) expenses. The City argues that the Commission abused its discretion by finding a 401-k plan expense in excess of the amount requested by EPEC was reasonable and necessary when the evidence does not support the finding. According to the City, by doing so, the Commission has violated its own rules. The City has not indicated which of its substantive rules the Commission violated, and it identifies no evidence that would tend to show that the Commission acted without reference to any guiding legal principles. In addition, the staff offered evidence that the greater sum was necessary to provide the benefit to all existing employees and reasonably anticipated additional employees. We conclude that substantial evidence supports the Commission’s findings that the amount recommended by the staff was reasonable and necessary. The City next contends that the Commission erred in including any TRA-SOP costs in employee benefits expense. The basis for this argument appears to be that the Commission had rejected inclusion of these expenses in two prior dockets; in the City’s view, apparently, the Commission’s prior holdings estop it from including the expense in later dockets. The City cites no authority for this proposition. The legislature has given the Commission discretion to determine which of a utility’s expenses are reasonable and necessary and, hence, may be recovered. PURA § 41(c). Because the reasonableness determination is one committed to agency discretion, it may be overturned only by a showing that the agency either based its decision on legally irrelevant factors, failed to consider legally relevant factors, or reached a completely unreasonable result after weighing only legally relevant factors. Gerst, 411 S.W.2d at 360; Statewide Convoy, 753 S.W.2d at 804. The City has not shown that the Commission erred in any of these respects. Consequently, we conclude that the City’s attacks on the Commission’s findings regarding employee benefits are without merit. D. Taxes Other Than Federal Income Taxes. The City contends that no evidence supports the Commission’s finding of a specific amount for expenses incurred for taxes other than federal income taxes. This item of expense increased by $992,773 after the stipulation phase of the hearing, and the City claims that neither EPEC nor the staff can identify evidence supporting the upward adjustment. Contrary to the City’s assertion, EPEC offered testimony supporting the tax expense alterations. EPEC’s witness May-hew testified that taxes are uniquely tied to other elements of revenue requirements. Therefore, when other components of the revenue requirement were adjusted, the tax expense necessarily was changed to accurately reflect the expense EPEC would incur. Mayhew explained that one way of isolating the tax effect of alterations in revenue requirements would be to compare the two reconciliation statements, line by line, identifying changes and recalculating taxes based on them. The City has not attempted to show that the recalculation was done incorrectly, and we conclude that its contention on this issue is meritless. E. Depreciation Add-Back. In its final challenge to the Commission’s cost-of-service findings, the City attacks Finding of Fact 187. By that finding, the Commission included in cost of service an element referred to as “depreciation add-back,” the purpose of which was to account for EPEC’s transition from a “flow-through” system of tax accounting to a “normalization” system. The City challenges this finding on the grounds that: (1) inclusion of “depreciation add-back” amounts to retroactive ratemaking, (2) the inclusion is not supported by substantial *930 evidence, and (3) certain necessary findings of underlying fact have been omitted. The City complains that it “cannot know how the Commission reached its determination because there are no underlying findings of fact.” Without providing legal authority for its position, the City argues that the Commission failed in its duty to make additional findings of underlying fact that would show “a logical nexus between the conclusion of the underlying fact[s] and the evidence.” For the same reasons that we held such findings of “nexus” unnecessary in connection with the Commission’s “deci-sional-imprudence” findings, we conclude they are also unnecessary here. The City next argues that inclusion of depreciation add-back constitutes retroactive ratemaking because [t]he evidence does not identify any shortfall in depreciation reserves to cover test year tax timing reversals. The evidence does show that some of the alleged deficiency may be attributed to years before the Company’s first rate proceeding, which would mean that the burden imposed on ratepayers by the Commission as a result of Finding of Fact No. 187, inured to the benefit of the Company and its shareholders in the past. The City misconstrues the nature of the transition from a flow-through accounting system to a normalization system. As is common for utilities, EPEC depreciated its assets at an accelerated rate for tax purposes while depreciating them using the straight-line method on its ratemaking books. Under a flow-through system, any tax benefit that resulted from the practice of keeping one set of books for tax purposes and another for ratemaking purposes was passed on to the ratepayers as it accrued. As time passed, the utility would incur, first, a very low, then a medium, and, finally, a relatively high level of income tax liability. The portion of rates attributable to income taxes that ratepayers paid over the asset’s life would also rise, corresponding to the utility’s actual tax liability. Under normalization, on the other hand, while actual tax liability follows the same increasing path, rates reflect that the ratepayers’ contribution to the payment of the utility’s income taxes remains constant throughout the asset’s useful life. To the extent that this creates an “overpayment” of taxes during the early years of an asset’s useful life, the utility accumulates the excess in a deferred-income-tax account. The accumulated funds are later used during the stage of the asset’s useful life when ratepayer payments are not enough to satisfy the utility’s actual tax liability. At the end of the asset’s useful life, the total overpayments and underpayments will match, such that ratepayers will have paid an amount equal to the actual tax liability incurred by the utility. See generally GTE-SW, 833 S.W.2d at 164-65. Federal law now requires that all public utilities which accelerate depreciation for federal income tax purposes use the normalization system. 26 U.S.C. § 168(f)(2), (i)(9) (Supp.1992). In most cases, a utility that had been using the flow-through system had to switch to normalization in the middle years of the useful lives of its assets, rather than at the beginning or end of those lives. As a result, the utility had not built up a deferred-tax account with which to pay its taxes during the later years in which its actual tax liability will exceed the tax payments ratepayers will be making under normalization. The utility therefore faces an increasing tax liability without a means of recovering the increased expense; yet, in spite of the deficiency in funds available to satisfy the tax liability, the utility must pay its taxes as they become due. Consequently, in this case, the Commission has included in cost of service a one-time adjustment to put EPEC in the position it would have occupied had it used normalization all along. The City complains bitterly about the allegedly retroactive effect of the new rates because the adjustment the Commission made is called “depreciation add-back.” This label does indeed make it sound as if the Commission has obliged present and prospective ratepayers to pay the utility a second time for assets already *931 depreciated. However, we do not decide the propriety of Commission action based on the name the Commission has elected to apply to it. The true effect of the “depreciation add-back” adjustment is to allow the utility to obtain from present and prospective ratepayers its actual current and future tax expenses. Consequently, this adjustment to the deferred-tax account does not, in any way, constitute retroactive rate-making. As a final matter, we note that the City’s assertion of inadequate evidence to reveal a shortfall in the reserves needed to pay taxes in the test year or ensuing years is incorrect. Moisés Rodriguez, the supervisor of the EPEC’s tax accounting section, testified that, while EPEC had adjusted reserves to compensate for the shift to normalization as it affected the timing differences related to depreciation, EPEC had not done so for differences related to the tax “bases” of all its assets. In Mr. Rodriguez’s words, “[t]he net result is that the current accumulated deferred Federal income tax balance does not fully reflect the timing difference that occurred prior to 1979.” Rodriguez concluded that an adjustment was necessary to bring EPEC into compliance with federal law. We therefore conclude that substantial evidence supports the Commission’s adjustment to the deferred-tax element of EPEC’s cost of service. Having concluded that all of the City’s challenges to the Commission’s cost-of-service allowances are without merit, we overrule the City’s sixth point of error. EXCLUSION OF TSA DURING PROCEEDINGS On October 22, 1987, the examiner orally granted EPEC’s motion to strike TSA as a party, excluding TSA from the proceedings. TSA appealed the decision to the Commission, but the Commission extended its time for making a decision such that it rendered the order setting new rates before ruling on TSA’s appeal. In the meantime, on November 6, 1987, EPEC withdrew its motion to remove TSA from the proceedings, and the examiner readmitted TSA and reinstated its party status. During the fifteen days that TSA did not participate in the hearing, more than thirty witnesses offered testimony on various issues. TSA contends that the October 22 order violated its due process rights by preventing it from cross-examining these witnesses. In addition, TSA alleges that the Commission intentionally postponed considering its appeal solely to prevent TSA from obtaining a stay of the proceedings from the Texas Supreme Court until such time as that court could determine TSA’s entitlement to party status. Finally, TSA asserts that the Commission erred in refusing to make findings of fact and conclusions of law concerning the due process claim. The supreme court’s holding in State v. Thomas, 766 S.W.2d 217 (Tex.1989), is dis-positive of TSA’s complaint. TSA had the right, under the Texas Constitution, to intervene in the proceedings. Id. at 219. However, the wrongful exclusion of TSA will necessitate reversal of the Commission’s order only if the error prejudiced substantial rights of TSA. APTRA § 19(e). After TSA appealed the October 22 order to the Commission, that body granted itself five extensions of time to consider the complaint. These extensions allowed the Commission to avoid deciding the issue; the final extension postponed consideration of the appeal until after the Commission had signed a final order in the docket. Nevertheless, TSA participated as a party in all proceedings after its reinstatement on November 6. TSA suffered no harm, therefore, from the Commission’s failure to rule on its appeal. The harm, if any, stems from TSA’s inability to cross-examine the thirty-plus witnesses who testified while it was absent from the proceedings. TSA does not complain of its inability to cross-examine approximately two dozen of the witnesses who testified during its absence. TSA claims to have suffered harm only by losing the opportunity to cross-examine: (1) three prudence and deferral witnesses who had testified during the first two days TSA was excluded; and (2) four *932 rate-design witnesses. As to the latter, TSA has not preserved any error; the examiner specifically stated in his oral ruling that he would permit TSA to recall and cross-examine any rate-design witnesses. As to the former, the examiner made it clear he would entertain a motion from TSA to recall them. TSA recalled only EPEC witness May-hew, a rate-design expert. It made no motion to recall and cross-examine the three prudence and deferral witnesses. Further, while TSA was still present during the original rate-base phase of the hearing— the only phase to which deferral and prudence issues would have been pertinent — it did not seek to cross-examine witnesses or present evidence of its own. Under these circumstances, and because TSA failed to request available relief that would have made it whole, we cannot say that the error in excluding TSA for fifteen days prejudiced substantial rights of TSA. We therefore overrule TSA’s first point of error. In addition, because Thomas dictates the conclusion that TSA should not have been excluded, we need not rule on TSA’s complaint regarding the Commission’s failure to make the requested findings of fact and conclusions of law on that issue. TSA’S RATE CLASSIFICATION The third phase of the proceedings, the rate-design segment, afforded the parties an opportunity to offer evidence in support of or in opposition to the proposed method of apportioning the anticipated rate increase among the various rate classes. In addition, because TSA sought to be reassigned to the city/county governmental-consumer class (rate class 41) from the general services class (rate class 24), the parties also offered evidence on the classification issue. The examiners recommended that TSA not be reassigned and that the proposed apportionment method be approved. The Commission adopted both recommendations and the examiners’ underlying reasoning. In its second and third points of error, TSA argues that the Commission erred by approving a rate for TSA that is not cost-based and by refusing to move TSA to the city/county rate class. A. Rate Class 41 TSA urges us to conclude that the Commission erred in refusing to include TSA in the city/county rate class, which arguably pays lower rates than the general services class. TSA asserts that there is no reasonable basis for differentiating TSA from the city and county governmental consumers; that EPEC offered no proof of a factor justifying different treatment for TSA than for the city and county consumers; and consequently, that PURA § 38 required the Commission to reassign TSA to the city/county class to prevent EPEC from charging unreasonably discriminatory rates. The Commission has broad discretion to determine whether a particular rate design would result in just, reasonable, and non-discriminatory rates. In making the determination, the Commission may consider factors in addition to the cost of providing service, keeping in mind the overriding considerations of consistency and the utility’s burden of proving that its proposed rates are just and reasonable. See PURA § 40; Texas Alarm & Signal Assoc. v. PUC, 603 S.W.2d 766, 773 (Tex.1980). Absent unreasonably discriminatory rates, we will not overturn the Commission’s approval of a rate design. PUC v. AT & T Communications of the Southwest, 777 S.W.2d 363 (Tex.1989). A customer seeking reassignment to a different class must show that its conditions of service are similar to those of the members of the class to which it seeks reassignment. The issue is one of fact, to be resolved by reference to the particular circumstances of each case. Ford v. Rio Grande Valley Gas Co., 174 S.W.2d 479, 480 (Tex.1943); Amtel Communications v. PUC, 687 S.W.2d 95, 102 (Tex.App.1985, no writ). Existing classification schemes previously approved by the Commission are, prima facie, not unreasonably discriminatory, and the complaining party has the burden of proving that the classification produces unreasonably dis *933 criminatory rates. Ashley v. City of Gilmer, 271 S.W.2d 100, 102 (Tex.Civ.App.1954, writ ref'd); see also Ford, 174 S.W.2d at 480; Amtel Communications, 687 S.W.2d at 102. Resolution of the burden-of-proof issue disposes of the dispute here. TSA failed to offer proof that its load characteristics were similar to those of the city and county governmental customers. In addition, TSA offered no proof that its constituent agencies are similar to the city/county consumers in other respects which the Commission considers when classifying a customer. Having offered no proof of these similarities, TSA has failed to carry its burden of showing that its rate is unreasonably discriminatory. Another consideration which persuades us that the Commission did not abuse its discretion in refusing to reassign TSA is that the Commission’s final order, following a suggestion in the Examiners’ Report, directed EPEC to perform the appropriate studies, so that during [EPEC’s] next general rate case, the load and usage characteristics, at the state agencies, as a group, including any state universities and colleges, can be compared to the load and usage characteristics of both Rate Classes 24 and 41. Considering that the current classification scheme has apparently existed unchallenged for some fifty years, that the Commission has expressed its intention to investigate TSA’s assignment to the general services class in the next rate case, and that requiring EPEC to produce the needed information in this proceeding could have resulted in a significant delay, we conclude that the Commission acted reasonably by refusing to reassign TSA to the city/county rate class. We overrule TSA’s third point of error. B. Cost-Based Rates TSA next contends that it is entitled to rates based on the utility’s actual cost of serving only the agencies constituting TSA. To support this argument, TSA cites four Texas constitutional provisions it claims require EPEC to charge TSA a rate based on the cost of serving only TSA. See Tex. Const, art. Ill, §§ 44, 51, and 53 (1984), and art. XVI, § 6 (Supp.1992). The cited constitutional provisions prevent the State from depleting its treasury by disbursing State funds without obtaining a corresponding benefit for the public. See, e.g., University of Texas System v. Robert E. McKee, Inc., 521 S.W.2d 944, 948 (Tex.Civ.App.1975, writ ref’d n.r.e.); State v. City of Austin, 160 Tex. 348, 331 S.W.2d 737, 742 (1960); Olshan Demolishing Co. v. Angleton Indep. School Dist., 684 S.W.2d 179, 185 (Tex.App.1984, writ ref’d n.r.e.); and State v. City of Dallas, 319 S.W.2d 767, 775-76 (Tex.Civ.App.1959), aff'd, 160 Tex. 348, 331 S.W.2d 737, 742 (1960). We need not decide, however, whether rates that are not cost-based violate these provisions. TSA’s challenge to the proposed rate, like its challenge to its classification, is resolved by examining the burden of proof on the issue. Even assuming for the purposes of this discussion that TSA correctly identified a constitutional entitlement to cost-based rates, we conclude that it had the burden of proving that its new rates were not cost-based. Because it did not carry this burden, we will overrule its second point. TSA insists that EPEC had the burden of proving that the rate it proposed for TSA was based on its cost of serving TSA. This argument is premised on: (1) the overall burden of proof a utility bears in ratemak-ing proceedings imposed by PURA § 40; and (2) EPEC’s exclusive control of cost-of-service information. We conclude that these considerations are insufficient to impose on EPEC the burden of proving that its proposed rates are based on its costs to serve this select group of customers. All individual state agencies were originally assigned to the general services rate class fifty years ago; until now, they have not complained of that assignment. The agencies comprising TSA intervened in this ratemaking proceeding as a newly formed group seeking to be reassigned to the city/county class. The crux of the group’s *934 cost-based-rates argument is that EPEC was obligated to prove that the rate EPEC anticipated charging the group was based on the utility’s cost of serving only the members of the group. We conclude that the Commission acted reasonably both in refusing to impose such a burden on EPEC and in ordering EPEC to produce information necessary to evaluate the issue in the next ratemaking case. The Texas Supreme Court has held that a utility is not required to compare profits and rates of return between services. Texas Alarm & Signal, 603 S.W.2d at 772. The considerations supporting the decision not to require such a comparison also support the decision not to require a utility to determine and offer proof of the costs of serving individual customers or subclasses of customers who share one or more characteristics. Cf . City of Corpus Christi v. PUC, 572 S.W.2d 290, 294-96 (Tex.1978). Comparing rates of return between services requires the utility to determine the expenses it incurred and the adjusted value of property it used in producing each service. Determining cost-based rates for subclasses of consumers would require the utility to determine the expenses it incurred and the adjusted value of property it used in producing service for each individual customer. This would be even more onerous a burden than that rejected by the supreme court in Texas Alarm & Signal. In addition, requiring EPEC to determine costs of service for individual customers would generate more costs, which would then be passed on to consumers. The increase in rates that could result from added costs of the ratemaking proceeding is a factor the Commission can and should consider. Id. at 772 n. 7. Allocation of the burden of proof to a complaining party is reasonable in circumstances in which individual customers have combined to form a subclass, which then asserts an entitlement to rates based not on the costs of serving all customers, but of serving only the members of the subclass. If the complaining subclass were not assigned the burden of proof in such circumstances, the utility would arguably be obligated, in every rate-making proceeding, to present evidence of the cost of serving every subclass that customers could define based on shared characteristics. Even assuming that it would be possible for the utility to satisfy such an obligation, it is questionable whether the vastly increased costs which such a presentation would entail would be in the public’s interest. Further supporting our conclusion that the Commission did not abuse its discretion in approving the new rates is the fact that TSA did not request the information before the hearings, even though the Commission could have compelled EPEC to produce the information necessary to prove TSA’s claim of being overcharged. Instead, TSA asserted, during the rate-design phase, that it needed the information but had no access to it. Because TSA made little attempt to acquire the information it needed to carry its burden of proof, its policy argument is unsympathetic. Therefore, even assuming for the sake of argument that TSA’s constitutional theory is correct, the record it has brought this Court is insufficient to show harm. We find no reversible error on this record. TSA’s second point of error is overruled. CONCLUSION In the context of a myriad of complex issues and often-contentious parties, the Commission must be allowed to weigh all competing interests in setting rates that will be fair to all consumers. With the exception of the use of deferred accounting as to EPEC’s carrying costs incurred during the regulatory-lag period, we conclude that the Commission acted within its discretion in setting new rates for EPEC. We reverse that portion of the trial court’s judgment which affirmed the Commission’s approval and use of deferred accounting as to the carrying costs incurred by EPEC between the date Palo Verde Units 1 and 2 became commercially operational and the effective date of the new rates. We affirm the remainder of the trial court’s judgment. We remand the cause to the Commission for such further *935 proceedings as may be necessary or appropriate to implement this Court’s judgment. Footnotes [1] All of the affected cities set rates within their respective city limits prohibiting EPEC from recovering its nuclear plant investment. Only El Paso sought review of the Commission’s order. [2] EPEC, the Commission staff, and four corporate intervenors which purchased significant amounts of electricity from EPEC all signed the stipulation. [3] All parties appear to agree that, to some degree, EPEC made imprudent decisions. On its face, PURA § 41(a) contemplates an imprudence disallowance only within the framework of a request that “construction work in progress” be included in rate base. The parties, however, have not briefed or argued whether a “prudence” standard governs the inclusion of new nuclear plant-in-service in rate base. We will therefore discuss appellants’ points in the terms they have chosen. We do not express an opinion on the possible existence of a distinction between the reasonableness standard guiding the PUC in setting rates and the prudence standard it uses pursuant to PURA § 41(a) in deciding whether to include the value of construction work in progress in rate base. [4] We express no opinion on the proposition that, because the Commission’s prudence decision and resulting disallowance rested on a number of factors, the Commission could reasonably have concluded that the substantial evidence would support a figure anywhere within the range from zero to 50% of the total project costs. [5] We construe OPC's allegation to refer to the requirements of section 16 (⅛), which deals with findings of fact. APTRA § 16(e) addresses agency motions for rehearing. [6] "Regulatory lag” is the period between the date a new plant begins commercial operation (the “in-service” date) and the effective date of the new rates that result from including the new plant's costs in rate base. [7] "Deferral” is an accounting procedure a utility can use to record all costs of a certain type in two accounts, one an expense account and the other a capital asset account. As the expense account balance increases, so does the capital asset account balance, so that the two balances remain identical. Recording costs this way preserves evidence the utility may later use to seek inclusion of the capital account balance in rate base and a corresponding increase in rates. [8] TSA also purports to complain of the deferral procedure. However, TSA failed to address the alleged error in its motions for rehearing to the agency. Therefore, it has waived any error. APTRA § 16(e); Burke v. Central Educ. Agency, 725 S.W.2d 393, 396 (Tex.App.1987, writ ref'd n.r.e.). [9] The City did not preserve a section-41(a) argument in its motion for rehearing to the Commission. The City also argues in its brief to this Court that the Commission failed to make necessary findings of underlying fact to support its conclusion that Unit 2 costs should be included in rate base, but the City likewise did not assign such failure as error in its motion for rehearing. Both complaints have been waived. APTRA § 16(e); Burke v. Central Educ. Agency, 725 S.W.2d 393, 396 (Tex.App.1987, writ ref'd n.r.e.). [10] The relevant sections of PURA originally provided: Sec. 39. In fixing the rates of a public utility the regulatory authority shall fix its overall revenues at a level which will permit *911 such utility to recover its operating expenses together with a reasonable return on its invested capital. Sec. 40. (a) The regulatory authority shall not prescribe any rate which will yield more than a fair return upon the adjusted value of the invested capital used and useful in rendering service to the public. ****** Sec. 41. The components of adjusted value of invested capital and net income shall be determined according to the following rules: (a) Adjusted Value of Invested Capital. Utility rates shall be based upon the adjusted value of property used by and useful to the public utility in providing service including where necessary to the financial integrity of the utility construction work in progress at cost as recorded on the books of the utility. The adjusted value of such property shall be a reasonable balance between original cost less depreciation and current cost less an adjustment for both present age and condition. The regulatory authority shall have the discretion to determine a reasonable balance that reflects not less than 60% nor more than 75% original cost, that is, the actual money cost, or the actual money value of any consideration paid other than money, of the property at the time it shall have been dedicated to public use, whether by the utility which is the present owner or by a predecessor, less depreciation, and not less than 25% nor more than 40% current cost less an adjustment for both present age and condition. The regulatory authority may consider inflation, deflation, quality of service being provided, the growth rate of the service area, and the need for the public utility to attract new capital in determining a reasonable balance. 1975 Tex.Gen.Laws, ch. 721, §§ 39-41, at 2341-42 (PURA §§ 39-41, since amended); see also Southwestern Bell Tel. Co. v. PUC, 571 S.W.2d 503, 512-16 (Tex.1978). [11] There are actually two types of such “costs” associated with capital construction projects: (1) interest paid on debt capital (i.e., borrowed funds); and (2) inability to earn a fair return on equity capital. While the difference between the two is relevant for some purposes, it does not appear to be so for purposes of our decision in the present case. Accordingly, we refer to both collectively as "carrying costs.” [12] "Since no program of rate regulation is self-executing, one of the most important virtues of an original cost valuation method is that of relative ease of administration.” H. Louis Nichoís & Randall Hagan Fields, Rate Base Under PURA: How Firm is the Foundation?, 28 Baylor L.Rev. 861, 866 (1976). [13] Our holding does not prevent post-in-service carrying costs from being amortized and recovered by a utility; it merely prevents them from being included in rate base. Moreover, even if the practical effect of this holding were to prevent recovery of such costs incurred during regulatory lag, "[a]ny change in protection for the utility against undue regulatory lag should come from the legislature.” Railroad Comm’n v. Lone Star Gas Co., 656 S.W.2d 421, 427 (Tex.1983). [14] One commentator has recently argued against giving undue importance to the "thereafter” language contained in many state utility regulatory acts: It is difficult to understand why legislatures would have delegated such broad powers to commissions, but would have simultaneously limited that authority with off-hand wording in a provision merely describing the process for the entry of a rate order. The most probable reason, then, for the inclusion of the “thereafter" language in the enabling statutes is the simple fact that after the commission enters a rate order, a utility cannot mechanically collect rates in the past. [[Image here]] ... If commissions in their rate orders are allowed to consider only losses or gains fore-casted to occur “thereafter,” then it is difficult to discern how commissions have the authority to correct mistakes in past rate orders, to allow recoveries for past extraordinary gains or losses, to change accounting treatment for past gains or losses, or to grant refunds or surcharges after reversal of a rate order. Stefan Krieger, The Ghost of Regulation Past: Current Applications of the Rule Against Retro *917 active Ratemaking in Public Utility Proceedings, 1991 U.Ill.L.Rev. 983, 1034-35 (1991). [15] A mechanical recitation of this rule to support application of the retroactive-ratemaking prohibition may not be justified. See Krieger, supra at 1035-37. Professor Krieger points out that many states, including Texas, expressly require the use of adjudicative rather than rule-making procedures for rate hearings. Id. at 1037; see APTRA § 3(2). [16] Within the context of this opinion, we need not decide whether the language the supreme court used in Goeke reduces the requirements set forth in Charter Medical and Allied Bank in order for underlying findings of fact to be considered sufficient to comply with APTRA § 16(b).