Public Utility Com'n of Texas v. Gte-Sw

ON MOTION FOR REHEARING POWERS, Justice. We withdraw our previous opinion of June 19,1991, and substitute this opinion in order to respond more conveniently to matters raised in the parties’ motions for rehearing, which we hereby overrule. In a suit for judicial review authorized by the Public Utility Regulatory Act (PURA), Tex.Rev.Civ.Stat.Ann. art. 1446c, § 69 (Supp.1992), the district court affirmed in part and reversed in part a final order issued by the Public Utility Commission in a rate proceeding, remanding the cause to the Commission with instructions that it “take such action and enter such orders as are consistent with” the district-court judgment. We affirm that part of the district-court judgment reversing the Commission order. We reverse that judgment to the extent it affirms the Commission’s order. We remand the cause to the district court with instructions that the cause be remanded to the Commission for further proceedings not inconsistent with our opinion. THE CONTROVERSY In 1984 General Telephone Company of the Southwest applied to the Commission for an order increasing the rates it is permitted to charge its customers for intrastate telecommunication services. See PURA § 43. Several parties opposed the application. After an intervening lawsuit in district court, and a subsequent appeal involving certain aspects of the controversy, the Commission concluded the contested case by a final order dated April 7, 1989. The Commission order required the Company to reduce its rates to the extent necessary to diminish its annual revenues by about $59 million. The Commission purported to make the new rates retroactive by assigning them an effective date of January 1, 1987. To effectuate that element of the order, the Commission required the Company to refund about $140 million to its customers through credits on future invoices. After the Commission overruled various motions for rehearing, numerous parties in the contested case sued for judicial review of the order in the statutory cause of action authorized by PURA § 69. The district court consolidated the several causes. After final hearing, the court reversed that *156 portion of the agency order which required a refund but affirmed the remainder of the order. Several litigants appeal to this Court: the Commission, the Company, the State Purchasing and General Services Commission, the Office of Public Utility Counsel, and eighty-six municipalities. 1 See Texas Administrative Procedure and Texas Register Act (APTRA), Tex.Rev.Civ. Stat.Ann. art. 6252-13a, §§ 19(e), 20 (Supp. 1992). The appellants assail, by various points of error, those parts of the Commission order adverse to their respective interests. We will discuss below the several points of error. To assist in understanding our discussion, however, we should set out first the general statutory context in which the Commission arrived at its final order in the contested case. RATEMAKING IN THE COMMISSION In PURA, the Legislature directed the Commission to accomplish two objectives: (1) protect the public interest in state-wide availability of an adequate, efficient, safe, and reasonable telecommunications service; and (2) assure that the rates charged and paid for such service are “just and reasonable.” See PURA §§ 18(a), 35(a), 38. In PURA § 37, the Legislature empowered the Commission to “fix and regulate” the rates a utility charges for intrastate telecommunication services. The Legislature placed in the Commission a generous discretion in rate matters, if one looks only to certain broadly worded provisions of PURA. For example, a telecommunications utility must prove that any rates it proposes are “just and reasonable,” which is to say that they permit “a reasonable opportunity to earn a reasonable return on ... invested capital ... over and above ... reasonable and necessary operating expenses.” PURA §§ 18(a), 39, 40. Notwithstanding such broad criteria, however, the Legislature circumscribed Commission discretion by defining in PURA itself such essential terms as “invested capital,” “net income,” and “expenses disallowed.” PURA § 41. See generally Texas Alarm & Signal Ass’n v. Public Util. Comm’n, 603 S.W.2d 766 (Tex.1980); Southwestern Bell Tel. Co. v. Public Util. Comm’n, 571 S.W.2d 503 (Tex.1978). In addition, the Legislature directed the Commission to consider specified factors in fixing the rates that a public utility may charge. In PURA § 39(b), for example, the Legislature required the Commission to “consider, in addition to other applicable factors,” several aspects of a utility’s operations, namely: “[the utility’s] efforts to comply with the statewide energy plan, the efforts and achievements of such utility in the conservation of resources, the quality of the utility’s services, the efficiency of the utility's operations, and the quality of the utility’s management.” Still other factors may become applicable from sources of law outside PURA. The Constitution of the United States precludes, of course, any rate that is “confiscatory” of the utility’s property. Public Serv. Comm’n v. Great N. Util. Co., 289 U.S. 130, 135, 53 S.Ct. 546, 548, 77 L.Ed. 1080 (1933). And the Commission has provided by its own regulations for the consideration of such additional factors as inflation, deflation, service-area growth rate, and a utility’s need to attract new capital. Tex. Sec’y of State, 16 Tex.Admin.Code § 23.21(c)(1)(C) (Supp.1992). We believe our observations in reference to a similar statutory scheme apply to the rate-setting provisions of PURA. The applicable law and facts may require “the Commission to ascertain the existence, absence, and interaction of any number of factors. These factors may vary from case to case and time to time, requiring perhaps a different orchestration in each instance.” Morgan Express v. Railroad Comm’n, 749 S.W.2d 134, 137 (Tex.App.1987, writ denied). In view of the complexity inherent in the subject-matter, “[w]asteful and fruitless attempts at perfection are neither ex *157 pected nor required” of the Commission. Id. RATE OF RETURN The terms of PURA § 39(a) require the Commission to calculate and fix the Company’s charges to its customers according to the Commission’s determination of the level of “overall revenues” necessary to permit the Company “a reasonable opportunity to earn a reasonable return on” its rate base, or the “invested capital used and useful in rendering service to the public over and above its reasonable and necessary operating expenses.” 2 See generally Don R. Butler, The Rate of Return in Texas — The Neglected Issue, 28 Baylor L.Rev. 937, 938 (1976). The Commission concluded that a rate of 11.05% would yield a “reasonable return” within the meaning of PURA § 39(a). The Company challenges on appeal the Commission findings on which the conclusion rests. These relate to only one criterion of what constitutes a “reasonable return”: whether the rate is sufficient to yield a level of earnings that is high enough to attract new capital from external sources. The Company’s complaint pertains even more narrowly to a single technique or method by which the Commission might estimate a rate that is sufficient to attract new capital: the “double leverage” method sometimes employed when the capital stock of a utility is owned wholly by another corporation. See generally J. Rhoads Foster, Fair Return Criteria and Estimation, 28 Baylor L.Rev. 883, 886-90 (1976). The evidence may be summarized by the following table displaying the Company’s capital structure and the 11.05% “cost of capital” upon which the Commission’s findings rest. The table, which we have altered somewhat, is taken from the examiner’s proposal for decision, incorporated in the Commission’s final order: COST OF CAPITAL — GTE SOUTHWEST CORPORATION (After the Subsidiary Risk Adjustment) Component Amount 000’s Ratio Cost Weighted Cost 10.01% 4.16% Long-Term Debt $ 779,910 ⅞© to t — H 7.54 .06 Short-Term Debt 15,500 CO OO * 7.13 .12 Preferred Stock 31,530 OO ÍD T-i 11.99 6.71 Common Equity 1,049,517 CO 05 1C Total 11.05% While all other figures in the table appear to be actual costs, the 11.99% “cost” for “common equity” 3 or common stock is artificial, rendering artificial as well the 6.71% “weighted cost” of that element and the resulting total of 11.05%, or the rate of *158 return arrived at by the Commission, in its final order, as a “reasonable return." The artificial figure of 11.99% was derived from the testimony of witnesses who explained their various expert opinions of what a “reasonable return” would be. According to their opinions, a “reasonable return” ranged from 10.54% to 15.0%. Three witnesses buttressed or explained their opinions by referring to the “double leverage” method of estimating the “cost of capital” for a utility corporation when, as here, its common stock is owned entirely by another corporation. That method simply imputes to the utility a cost of “common equity” equal to the parent corporation’s total weighted cost of capital, based on an assumption that the parent supplies that component of the utility’s capital from the aggregate of the parent’s own capital structure, and not any particular component thereof. See Foster, supra, at 889. In the present case, for example, the capital structure of the Company’s parent, consisting of debt, preferred stock, and “common equity” or common stock, was shown to have a total weighted cost of 12.49%. The “double leverage” method of estimating the Company’s cost would have required that the 12.49% be inserted in the table above as the Company’s cost of “common equity.” The 11.05% rate of return selected by the Commission does not result from this manner of using the "double leverage” method of estimating cost of capital. The Commission employed instead the theory of a witness who varied the “double leverage” method in arriving at her opinion of the Company’s cost of capital — she reduced the parent corporation’s total weighted cost of capital (12.49%) to account for the fact that the Company was a public utility with a diminished economic risk. While the Commission adhered to the theory of a reduced economic risk, it did not merely adopt the exact reduction or rate of return arrived at by the witness. The Commission determined rather to reduce the 12.49% to 11.-99% to account for the diminished economic risk (the latter figure being shown in the table above). The ultimate 11.05% rate of return follows from simple arithmetical calculations. The Company argues that the diminished-risk factor was already reflected in the capital structure of its parent, in the minds of investors, and thus in the 12.49% total weighted cost of capital. Accordingly, any further reduction to account for the diminished-risk factor amounted to a double reduction for the same factor, an abuse of discretion requiring reversal of the Commission’s final order. APTRA § 19(e)(6). We believe, however, that the argument misses the essential point that both the “double leverage” method itself and the theory of a reduction to account for the Company’s diminished economic risk were only instruments employed in estimating the Company’s hypothetical cost of capital in a situation in which the Company’s actual cost of capital would not serve the purpose of determining the “reasonable return” required by PURA § 39(a). The method and the reduction were neither required nor prohibited by statute or regulation. No one was bound in a statutory sense by the techniques employed by any expert witness or a resulting opinion. For example, the several expert witnesses gave their respective opinions that a proper hypothetical rate of return on “common equity capital” for the Company was 15%, 12.1%, 10.85%, 4 and 10.54%, the last three being based, at least in part, on the witness’s use of the “double leverage” method. The figure chosen by the Commission, 11.99%, was no more than the Commission’s own estimate converted into a finding, an estimate within the range made by the testimony of the various expert witnesses. We hold the Commission did not abuse its discretion. The Company argues next that the Commission acted contrary to its established policy by making the diminished-risk adjustment mentioned above. It is not sug *159 gested that the policy was one of general applicability, laid down in a formal rule previously promulgated by the Commission to govern in all cases of a particular class. Rather, the policy is found, according to the Company, in the Commission’s own precedents in similar contested cases. The Commission’s unexplained departure from established policy, the Company contends, amounted to an abuse of discretion. AP-TRA § 19(e)(6). We might accurately compare the circumstances in the Company’s case with those in the earlier contested cases referred to only if we had before us the agency record compiled in those other cases. Those records are not before us, and were not before the district court. Instead, the Company refers to so much of the record in those cases as is reflected in a publication entitled Public Utility Commission Bulletin. We have read the report of those cases and other cases in the bulletin which indicate that the Commission did make the adjustment mentioned. Given the resulting doubt about whether such an administrative policy exists at all, and the absence of any means of determining the similarity of the other contested cases to the present case, we reject the Company’s argument. In a separate point of error, the Company contends the Commission’s choice of 11.05%, as a rate sufficient to yield a reasonable return, was “not reasonably supported by substantial evidence of record.” APTRA § 19(e)(5). The Company argues the point jointly with its abuse-of-discretion complaint discussed above. We believe the Company directs its separate point of error at the absence of evidence explicitly supporting the Commission’s choice to reduce the 12.49% figure to exactly 11.99%, that is to say, a want of evidence to support fixing the difference at precisely .50%. The .50% difference is an inference from the body of evidence on the point in question; none of the evidence suggested that figure explicitly. It is, however, within the range of the evidence received, as discussed above. The “substantial evidence” test or rule refers simply to “such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.” Pierce v. Underwood, 487 U.S. 552, 563, 108 S.Ct. 2541, 2549, 101 L.Ed.2d 490 (1988); Consolidated Edison Co. v. NLRB, 305 U.S. 197, 229, 59 S.Ct. 206, 216, 83 L.Ed. 126 (1938); see generally Kenneth C. Davis, Administrative Law Text § 29.02, at 527-30 (3d ed. 1972); Bernard Schwartz, Administrative Law § 10.7, at 596-600 (2d ed. 1984). The Company suggests no reason why the .50% difference is an unreasonable conclusion even though it falls squarely within the range of conclusory inferences that a reasoning mind might have drawn from the relevant evidence, which included testimony that a reasonable rate of return to “common equity” might range from 10.54% to 15%. No such reason appears to us. For the reasons given, we overrule the Company’s two points of error complaining of the rate of return. PAYMENTS TO AFFILIATES As mentioned previously, PURA § 39(a) requires a rate of return that will allow the Company to recoup “its reasonable and necessary operating expenses” while simultaneously affording the Company “a reasonable opportunity to earn a reasonable return on its invested capital.” The Commission included within the Company’s operating expenses the estimated amounts the Company will pay two affiliate companies, GTE Directories and GTE Service Corporation. The Cities complain on appeal that the Commission erroneously included these sums in calculating the Company’s operating expenses and, by extension, its rate of return. Because operating expenses are limited to those that are “reasonable and necessary,” and because the sums paid between affiliates might be fixed at something less than an arms-length transaction, rational administration required the Commission to scrutinize closely any sums the Company expected to pay its two affiliates. See Douglas W. Hawes, Utility Holding Companies, at 10-1 (1984). The same prin *160 ciple is found in the terms of PURA § 41(c)(1). That statute explicitly requires the Commission to exclude such expenses “except to the extent that the [Commission] shall find such payment to be reasonable and necessary for each item or class of items as determined by the commission.” Moreover, the statute spells out the character of finding that the Commission must make before such expenses may be included: Any such finding shall include specific findings of the reasonableness and necessity of each item or class of items allowed and a finding that the price to the utility is no higher than prices charged by the supplying affiliate to its other affiliates or divisions for the same item or class of items, or to unaffiliated persons or corporations. PURA § 41(c)(1). In other words, each finding that a payment to an affiliate is “reasonable and necessary” must include subsidiary findings: (1) that each item or class of items is “reasonable and necessary”; and (2) that the price paid by the utility is no higher than the prices charged to other specified purchasers. GTE Directories The evidence showed that the Company made payments to GTE Directories for the printing and publishing of the Company’s telephone directories. GTE Directories is an “affiliate” within the meaning of PURA § 3(i)(4). 5 The Cities' complain on appeal, in several respects, of the Commission's inclusion of the Company’s payments to GTE Directories as “reasonable and necessary operating expenses.” Specifically, they argue that the Commission’s final order did not contain the specific findings required by PURA § 41(c)(1). The sole relevant finding contained in the Commission’s final order is that numbered 16; no other finding pertains to the Company’s payments to GTE Directories. Finding of Fact 16 states as follows: The testimony of GTE witness Keys refutes the conclusion that GTE Directories’ agreement with GTE Southwest is unfavorable in comparison with its other customers. It would not be appropriate to adjust GTE Directories’ prices for return and tax components. Because the matching principle would require ignoring the revenues from the arrangement with GTE Directories if the expenses are not allowed, the customers benefit from recognizing the arrangement for rate-making purposes. Accordingly, expenses of $19,400,000 and revenues of $47,416,000 should be included in the cost of service. While the foregoing purports to be a “finding,” it hardly determines anything at all except perhaps in the last sentence, which concludes starkly and opaquely that the stated sums “should” be included in the Company’s cost of service. The balance of the finding declares merely that: (1) a witness’s testimony refutes any conclusion that the directory agreement “is unfavorable in comparison with the [Company’s] other customers”; (2) any adjustment of the prices “would not be appropriate”; and (3) the Company’s “customers benefit from recognizing the arrangement for rate-making purposes.” We are instructed not to be “hypertechnical” in our evaluation of whether an agency’s findings comply with a statutory mandate that the agency make findings of fact; and we are even told that the “evidence and testimony” in a contested case amounts in some manner to a sufficient finding of fact. State Banking Bd. v. Allied Bank, 748 S.W.2d 447, 449 (Tex.1988). We need not explore these mysteries, however, for we cannot imagine a more flagrant failure by an administrative agency *161 to comply with the terms of its governing statute. Absolutely nothing in the Commission’s order suggests the Commission attempted to include in its order what PURA § 41(c)(1) explicitly demands in the clearest and most imperative language: (1) “specific findings of the reasonableness and necessity ” of the sums to be paid for directories; and (2) “a finding that the price to the [Company] is no higher than prices charged by [GTE Directories] to its other affiliates or divisions for the same item or class of items, or to unaffiliated persons or corporations.” If a statute means anything at all, PURA § 41(c)(1) required those specific findings before the affiliate payments might be included, and the Commission did not make them in its final order in this case. We therefore sustain the Cities’ point of error. 6 See APTRA § 19(e)(1). GTE Service Corporation GTE Service Corporation is another subsidiary of GTE Corporation. Three divisions of GTE Service Corporation charge its sister subsidiaries for various services it provides them; they are: (1) “Telephone Operations Headquarters,” or “Telops,” which files interstate-access tariffs and supplies certain planning and development services to those subsidiaries that are telephone-operating companies, such as the Company here; (2) “GTE Corporate Departments,” which supplies to all affiliates certain services such as consolidated accounting, treasury services, and the preparation and filing of the parent’s consolidated tax return; and (3) “Diversified Products Headquarters,” which supplies services only to subsidiaries that are not telephone-operating companies, such as GTE Communications Systems and GTE Government Systems. The parties dispute exactly how GTE Service Corporation charges its sister affiliates for its services. In general, however, the charges are assessed only for services supplied: for example, the Company receives services only from “Telops” and “GTE Corporate Departments,” but not from “Diversified Products Headquarters,” and pays accordingly. With this qualification, we may refer hereafter to the amounts the Company pays its affiliate, GTE Service Corporation. The Cities assail the Commission’s decision to include as “reasonable and necessary operating expenses” the estimated sums the Company will pay GTE Service Corporation, contending the payments were unreasonable as a matter of law because the allocation system employed by GTE Service Corporation results in a disproportionate share of those expenses being borne by the Company, and thus by its customers through the rates charged them. The Company disputes any contention that it pays more than a “fair share” under the allocation system. We are unable to decide that issue; it is a question for the Commission and not this court. But the Cities raise an additional contention, directed again at the Commission’s failure to comply with the statutory requirements of PURA § 41(c)(1). The Commission’s Finding of Fact 13 is the sole finding, in the final order, that refers to the Company’s payments to GTE Service Corporation. It states as follows: GTE Service Corp provides to GTE Southwest the classes of services described in section III.D.3.a of the Report. The testimony of the GTE witnesses summarized in section III.D.3 of the Report establishes by a preponderance of the evidence that (1) the allocation for *162 mula properly reflects differences between the purchasing affiliates; (2) the prices charged for each class of service are reasonable relative to the cost of obtaining them from alternative sources; and (3) the services are reasonable and necessary for the provision of utility service. The testimony of Mr. Gillespie establishes that $258,000 should be deducted as a disallowance of legislative advocacy expenses and $268,000 should be deducted for Signaling System 7, which is related to services that are not being provided. Finding of Fact 13 incorporates by reference the following paragraph in the examiner’s report: Introduction — Four GTE witnesses discussed the operation of GTE Service Corp and the allocation and direct billing of its expenses to GTE Southwest. About $9,406,000 of GTE Service Corp expenses were allocated to GTE Southwest. [The report describes here the division of GTE Service Corp. into three groups and summarizes the functions of each group.] In general, the expenses of corporate departments are allocable to all GTE subsidiaries, the expenses of telops are allocable only to telephone operating companies, and the expenses of products and systems are allocable only to non-telephone subsidiaries. As mentioned previously, PURA § 41(c)(1) requires the exclusion of payments to an affiliate, from a utility’s “reasonable and necessary operating expenses,” unless the Commission makes these findings in its final order: a general finding that the payments are “reasonable and necessary for each item or class of items”; and specific findings that (1) each item or class of items supplied by the affiliate is reasonable and necessary and (2) “the price to the utility is no higher than prices charged by [the] affiliate to its other affiliates or divisions for the same item or class of items, or to unaffiliated persons or corporations.” The Cities complain that the Commission did not make the second specific finding — a finding that the prices charged by GTE Service Corporation to the Company were “no higher than prices charged” other affiliates or unaffiliated persons or corporations for the same item or class of items. As a result, the Cities contend, the premise of PURA § 41(c)(1) required that the Commission exclude from the rate calculation the Company’s estimated payments to its affiliate GTE Service Corporation. We agree. If we assume that Finding of Fact 13 amounts to “specific findings” that each item or class of items supplied by GTE Service Corporation is reasonable and necessary (a matter we do not decide), the fact remains that nothing therein purports to determine that the prices charged the Company by GTE Service Corporation are “no higher than” those charged by it to its other affiliates or divisions, or unaffiliated persons or corporations. Instead, Finding of Fact 13 determines only that the prices charged the Company by GTE Service Corporation are “reasonable relative to the cost of obtaining them from alternative sources." This is not the same thing. See APTRA § 19(e)(1). In its motion for rehearing, the Commission invites our attention to section III.D.3.e. of the report, where the examiner writes: “Accordingly, Mr. Bredeweg’s testimony establishes that the prices charged to GTE-Southwest were no higher than prices charged to other subsidiaries.” The Commission apparently argues that, by mentioning a section of the examiner’s report in a finding of fact, it thereby adopts everything in that section of the report. We disagree. PURA § 41(c)(1) requires specific findings that the price charged to the utility is no higher than prices charged by the supplying affiliate to its other affiliates or unrelated entities. The Commission made no such finding here. A sentence in the examiner’s report, summarizing the opinion of a witness, will not substitute for a Commission decision in the form of a finding, particularly when there is no specific incorporation of the examiner’s report, when there is an actual finding dealing with the subject, and when the *163 actual finding does not satisfy the statutory requirement. Nothing in the order explains the departure from what the statute required; and, of course, the Commission was not free to disobey the terms of its governing statute. We therefore sustain the Cities’ point of error. See APTRA § 19(e)(1). HYPOTHETICAL FEDERAL INCOME TAX ALLOWANCE Federal income taxes are includable as part of a utility’s operating expenses for ratemaking purposes. Public Util. Comm’n v. Houston Lighting & Power Co., 748 S.W.2d 439, 441 (Tex.1987), appeal dism’d, 488 U.S. 805, 109 S.Ct. 36, 102 L.Ed.2d 16 (1988); Suburban Util. Corp. v. Public Util. Comm’n, 652 S.W.2d 358, 363 (Tex.1983). Public Counsel and the Cities contend the Commission erroneously calculated the Company’s federal income-tax liability in estimating its operating expenses. They contend the calculation was erroneous in two respects: first, it failed to account for a pro-rata share of the tax savings realized by the Company by reason of its parent’s filing a consolidated tax return; second, it did not include the income-tax deductions actually taken by the Company for expenses disallowed by PURA § 41(c)(3). Both tax-accounting practices, the Cities and Public Counsel assert, violate the “actual taxes incurred” formulation of Houston Lighting, 748 S.W.2d at 442. We will sustain the point of error. The Consolidated Income-Tax Return The Company’s parent files a consolidated federal income-tax return on behalf of itself and its subsidiary companies. Not all the subsidiaries are regulated utilities. Each subsidiary furnishes the parent a return reflecting its particular federal income-tax liability or absence thereof, which the parent uses to compute the combined tax liability of all the participating subsidiaries. A. The Statutory Requirements PURA § 41(c)(2) requires for rate-calculation purposes that a utility which is a subsidiary of a holding company compute its federal income-tax expense as follows: If the public utility is a member of an affiliated group that is eligible to file a consolidated income tax return, and if it is advantageous to the public utility to do so, income taxes shall be computed as though a consolidated return had been so filed and the utility had realized its fair share of the savings resulting from the consolidated return, unless it is shown to the satisfaction of the regulatory authority that it was reasonable to choose not to consolidate returns. The amounts of income taxes saved by a consolidated group of which a public utility is a member by reason of the elimination in the consolidated return of the inter-company profit on purchases by the public utility from an affiliate shall be applied to reduce the cost of the property or services so purchased. The investment tax credit allowed against federal income taxes, to the extent retained by the utility, shall be applied as a reduction in the rate based contribution of the assets to which such credit applies, to the extent and at such rate as allowed by the Internal Revenue Code. We have not found, nor has any party pointed out to us, a Texas decision construing PURA § 41(c)(2). We believe, however, that § 41(c)(2) contains three separate requirements. Firstly, upon showings that the utility is a member of an affiliated group eligible to file a consolidated income-tax return and that it would be advantageous to the utility to do so, the Commission must (1) compute the utility’s income taxes as though it had filed a consolidated return, and (2) impute to the utility its fair share of tax savings. Secondly, any taxes saved by the consolidated group because of the elimination of the intercompany profit on purchases by the utility from an affiliate must be applied to reduce the costs of the services or property purchased. Thirdly, any investment-tax credit retained by the utility must be applied as a reduction in the rate-base contribution of the asset to which the credit applies. Each part of the statute must be given effect. See Perkins *164 v. State, 367 S.W.2d 140, 147 (Tex.1963). Therefore, the statute imposes three distinct duties on the Commission. B. The Commission’s Findings The Company argues that the Commission satisfied the requirements of PURA § 41(c)(2). The Cities and Public Counsel contend, however, that the Commission calculated the Company’s income-tax expense on the basis of a hypothetical individual tax return when a consolidated return would have yielded the Company certain tax savings associated with the parent’s investment in non-regulated businesses. Thus, the Company’s customers were denied the benefit of such savings in violation of PURA § 41(c)(2) and Houston Lighting. The relevant finding by the Commission declared: FF 12. The evidence establishes that GTE Southwest properly accounts for the benefits from filing a consolidated tax return with the GTE affiliates: (1) its purchases from GTE Communications Systems are subject to intercompany profit elimination, (2) it amortizes investment tax credits over the life of the related plant, and (3) GTE Corp’s consolidation method properly assigns tax losses and tax benefits to the sources of income or losses that gave rise to the tax or benefit. The Commission’s order also contains the following “Conclusion of Law,” as well as several other “Conclusions of Law” that we have set out in a footnote: 7 CL 11. Because the benefits of the consolidated tax return are properly accounted for in GTE Southwest’s invested-capital and cost-of-service accounts, the staff’s model calculates the proper allowance for the company’s federal income tax expense in accordance with section 41(c)(2) of PURA without a separate adjustment for consolidated tax savings. Appended to the final order is “Schedule V.” It appears to represent federal-income-tax expenses used in calculating the Company’s rates. 8 The table indicates that *165 the Commission made no adjustment to reflect the difference between the tax liability shown on the Company’s individual return and that of the consolidated return. The various conclusions of law and the one conclusion labeled a finding of fact, augmented by “Schedule V,” amount to only one essential determination by the Commission: an opaque generality that the Company “properly accounts for the benefits from filing a consolidated tax return with the GTE affiliates.” This does not conform to PURA § 41(c)(2), which explicitly requires a determination of whether any tax savings would result from a consolidated return; and if so, an inclusion of a fair share of that savings in computing the Company’s operating expenses, in order that the savings might be passed on to customers in the form of a lower rate. The Company calls our attention to Finding of Fact 72, which says, “For the reasons stated in section III.B. 9 of the Examiner’s Report, GTE-Southwest’s federal income tax should not be adjusted for consolidated tax savings.” We do not believe that the Examiner’s Report satisfies the requirements of PURA § 41(c)(2). Although the report states that “the GTE consolidated return provides GTE-Southwest its ‘fair share of savings,’ ” the Examiner limited his understanding of consolidation to the elimination of intercompany profit and amortization of investment tax credits. He refused to reduce the Company’s tax expense by the losses of unregulated affiliates because he perceived a violation of the normalization rules. We do not agree that the normalization rules preclude consideration of the losses of unregulated affiliates, as we will discuss below. Because the Examiner’s Report does not demonstrate compliance with the first sentence of PURA § 41(c)(2), the Examiner’s Report will not satisfy the statutory requirements. 10 C. Normalization Issues Once the Commission computes a utility’s income taxes as though a consolidated return had been filed, the statute directs the Commission to impute to the utility its “fair share of the savings resulting from the consolidated return.” PURA § 41(c)(2). The statute does not define the expression “fair share,” but the supreme court has held that the tax savings generated by a utility’s imprudent expenses *166 “should inure to the benefit of its ratepayers.” Houston Lighting, 748 S.W.2d at 442. To effectuate that principle, in a case in which the utility had actually obtained a tax savings, Houston Lighting required application of the corollary that “(t)he utility’s rates must reflect the tax liability actually incurred” 11 in lieu of any greater, hypothetical tax liability that might plausibly be constructed in the circumstances. 12 Id.; see also Federal Power Comm’n v. United Gas Pipeline Co., 386 U.S. 237, 244, 87 S.Ct. 1003, 1007, 18 L.Ed.2d 18 (1967); Suburban Util., 652 S.W.2d at 362-64. *165 [[Image here]] *166 The Company argues on appeal that the “actual taxes paid” corollary of Houston Lighting violates the “normalization” requirement of the Internal Revenue Code and the Commission’s substantive rules regarding that subject. See 26 U.S.C. § 167(Z)(3)(G) (1988), repealed by Omnibus Budget Reconciliation Act of 1990, Pub.L. No. 101-508, § 11812(a)(1), 104 Stat. 1388-534 (1990) (repeal applies to property placed in service after November 5, 1990); 26 U.S.C. § 168(i)(9) (Supp.1991); Tex. Sec’y of State, 16 Tex.Admin.Code § 23.-21(b)(1)(D) (Supp.1991-1992). On that basis, the Company would avoid the effect of the corollary in the present case. We should first determine whether the “actual taxes paid” corollary of Houston Lighting conflicts with the “normalization” requirement of the statute and rule. The “normalization” requirement is one element of rate calculation. The Commission’s rule 23.21(b) requires that a utility’s “cost of service” include, for rate-calculation purposes, “[fjederal income taxes on a normalized basis” and that such taxes “be computed according to the provisions of” PURA. Nothing in PURA or the Commission’s rules elaborates on “normalization.” It appears, however, that “normalization” refers to the technique by which utilities are positively encouraged to seek income-tax savings that may ultimately inure to the benefit of ratepayers, even though the benefit does not inure immediately. Simply stated, a utility is authorized to depreciate its capital assets at an accelerated rate and calculate its actual income-tax liability accordingly; for rate-calculation purposes, however, the utility is authorized to depreciate the same assets on a straight-line basis. The resulting difference is accumulated in a deferred-tax account that the Commission may ultimately divide, on an equitable basis, between the utility and its customers. See PURA § 27(e); Elizabeth Warren, Tax Accounting in Regulated Industries: Limitations on Rate Base Exclusions, 31 Rutgers L.Rev. 187, 187-194 (1978). “Normalization” under the Internal Revenue Code and the Commission’s rules purports to authorize the use of a fictitious depreciation expense for rate-calculation purposes: the expense is determined on a straight-line basis for rate-calculation purposes even though the utility determines the expense on an accelerated basis in calculating its actual income-tax liability. But this does not mean that the holding in Houston Lighting contradicts the “normalization” requirement; rather, it means that the corollary of that decision does so. In Houston Lighting, the “actual taxes incurred” corollary was necessary to effectuate the principle laid down in the holding in *167 that decision: the tax savings actually enjoyed by a utility “should inure to the benefit of the ratepayers.” See Houston Lighting, 748 S.W.2d at 442. The “normalization” requirement effectuates that principle in a different way: by encouraging utilities to seek tax savings for the ultimate benefit of its ratepayers when the Commission eventually divides the savings equitably between the utility and its ratepayers in calculating rates. We hold, for this reason, that there exists no contradiction between the “normalization” requirement and Houston Lighting. 13 The Company contends that using the tax losses of unregulated affiliates to reduce the income tax expense of a regulated affiliate violates normalization rules. We reject this argument for two reasons. Firstly, although no Texas case has addressed this question directly, the parties in Houston Lighting apparently raised the issue of normalization violations in their motion for rehearing. 14 In his dissent to the court’s overruling of the motion for rehearing, Justice Wallace asserted that taxable revenues and deductible expenses must be segregated on the basis of whether they are generated by a utility or non-utility. Id. He also warned that permitting a deduction to go to the entity that did not bear the loss could cause the I.R.S. to deny the utility the right to depreciate its assets on an accelerated basis, an argument the Company vigorously presses in this case. No other member of the supreme court joined Justice Wallace’s dissent; we therefore conclude that the supreme court rejected the dissenter’s contentions. 15 Secondly, we believe that the very text of PURA § 41(c)(2) refutes the Company’s contention that reducing a utility’s tax expense because of losses by non-regulated affiliates violates normalization rules. If the first sentence of PURA § 41(c)(2) requires a separate computation, as we think it does, consolidation must mean something more than elimination of intercompany profit and amortization of investment tax credits. We believe that it requires the Commission to include in the Company’s tax expense any losses that may reduce the utility’s tax liability, even if unregulated affiliates suffered the losses. 16 As discussed above, “normalization” is an accounting method for depreciation deduc *168 tions; it does not speak to the issue of reducing a utility’s tax expense because of its affiliates’ losses. Finally, the Company argues that it receives no benefit from a consolidation of tax returns. It explains that it determines its separate income-tax liability and pays that amount to its parent corporation. The parent, GTE Corporation, then pays the consolidated-tax liability and distributes the amount left over to those subsidiaries whose investments, expenses or losses gave rise to the tax benefit. We do not pass on the question whether it was advantageous for the Company to consolidate returns; this is a question for the Commission. 17 We do believe, however, that the Commission was required by statute to find either (1) that it was not advantageous for the Company to consolidate returns or (2) that the Commission had computed taxes as though a consolidated return were filed and the utility had received its fair share of savings from the consolidated return. The Commission cannot satisfy the explicit requirements of the statute merely by a general finding that the Company’s income-tax expense should not be adjusted for consolidated tax savings. D. Basis for Reversal In their motions for rehearing, both the Company and the Commission argue that extensive underlying findings of fact support the Commission’s ultimate findings. They mistake the basis for our holding. The issue is not whether the Commission’s final order contains the underlying facts required by APTRA § 16(b), but whether the Commission complied with its governing statute, PURA § 41(c)(2). It did not, and its failure in that regard requires reversal by a reviewing court. APTRA § 19(e)(1). Disallowed-Expense Deductions The Company incurs certain operating expenses which PURA § 41(c)(1) and (3) prohibit the Company to include as operating expenses for rate-making purposes. These include lobbying expenses, certain advertising expenses, and payments to various affiliates. The Cities and Public Counsel contend the Commission erred by not requiring the Company to pass on to its customers, through the rate calculation, the tax deductions taken for these “disallowed expenses.” We agree. The Texas Supreme Court held, in Houston Lighting, that any tax savings actually enjoyed by a utility should inure to the benefit of ratepayers. Houston Lighting, 748 S.W.2d at 442; see also Suburban Util., 652 S.W.2d at 362-64. In Houston Lighting, the utility actually took a tax deduction for expenses incurred in the management of its Allen’s Creek Nuclear Plant, which expenses the Commission had disallowed in computing the utility’s costs of service for ratemaking purposes. Even though the utility could not pass on to ratepayers the expenses it had incurred in connection with the nuclear plant, the court held the utility must include, in its rate calculation, the tax savings obtained when deductions were taken for these expenses in calculating the utility’s income-tax liabilityIn regard to the tax deductions taken for “disallowed expenses,” the Commission’s order includes two “Conclusions of Law”: CL 14. Section 43(c)(3) [sic] prohibits the Commission from considering for rate-making purposes any expenditure found to be unreasonable. To include disallowed expenses in the computation of income taxes is to consider them for rate-making purposes in violation of section 43(c)(3) [sic]. CL 16. Including disallowed expenses in the computation of federal income tax expense violates the consistency requirement in the normalization rules. Accordingly, such a treatment would violate the requirement in PURA and the Commis *169 sion s substantive rules requiring utilities to compute federal income taxes on a normalized basis. The first conclusion conflicts with the supreme court’s holding in Houston Lighting that any tax savings taken by a utility must inure to the benefit of the ratepayers, regardless of whether the expense bringing about the deduction is includable in the utility’s cost of service under PURA § 41(c)(3). The Company argues (as the Commission stated in its Conclusion of Law 14) that including deductions taken for disallowed expenses in a utility’s rate is equivalent to taking into account a “disallowed expense” in ratemaking, in violation of PURA § 41(c)(3). 18 We believe the matter settled, however, by the holding in Houston Lighting. Furthermore, the issue does not concern a violation of § 41(c)(3), but rather compliance with § 39 and Houston Lighting. The terms of PURA § 41(c)(3) forbid passing along to ratepayers certain expenses, but say nothing of savings. For the reasons stated, we sustain the point of error raised by Public Counsel and the Cities regarding disallowed-expense deductions. 19 APTRA § 19(e)(1). EFFECTIVE DATE OF THE COMMISSION’S FINAL ORDER The provisions of PURA § 43 govern when a utility wishes to change its existing rates. The utility must file in the Commission a statement of intent to change its rates, including therein a schedule of the proposed new rates and the date they are to become effective. PURA § 43(a). If the schedule portends a “major change” in rates, as defined in PURA § 43(b), the Commission must “enter on a hearing to determine the propriety of such change.” PURA § 43(c). “Pending the hearing and decision,” however, the Commission may suspend the effective date of the utility’s proposed new rates for as many as “150 days beyond the date on which the schedule would otherwise go into effect.” PURA § 43(d). If the Commission suspends the proposed rate schedule, the utility’s existing rates continue in force unless the Commission, in its discretion, establishes temporary rates in lieu of the existing rates. Id. If the Commission suspends the utility’s existing rates but fails to determine within 150 days the utility’s rate-change application, the utility may unilaterally implement a system of unofficial rates of its own choosing, not to exceed the proposed rates, provided the utility files and the Commission approves a bond to secure the utility’s obligation to refund or credit any excess produced by the unofficial rates over and above the rates finally determined by the Commission. PURA § 43(e). After hearing, the Commission must determine and fix the utility’s level of rates if the agency finds the proposed rates “unreasonable or in any way in violation of any provision of law.” PURA § 43(f). The Commission must include the rates fixed by it in an order “served upon the utility,” and “these rates are thereafter to be observed until changed, as provided by” PURA. Id. Not all rate changes are initiated by a utility. It is the Commission’s duty “to insure that every rate made, demanded, or received by any public utility ... shall be just and reasonable,” and not “unreasonably preferential, prejudicial, or discriminatory, but ... sufficient, equitable, and consistent in application to each class of con *170 sumers.” PURA § 38. Accordingly, the Commission itself is authorized to initiate a rate-change proceeding “on its own motion or on complaint by any affected person.” PURA § 42; see Public Util. Comm’n v. Pedernales Elec. Coop., 678 S.W.2d 214 (Tex.App.1984, writ ref’d n.r.e.). If the Commission, after reasonable notice and hearing, finds in such a case “that the existing rates of any public utility for any service are unreasonable or in any way in violation of any provision of law,” the Commission “shall determine the just and reasonable rates ... to be thereafter observed and in force, and shall fix the same by order to be served on the public utility.” Id. These “rates shall constitute the legal rates of the public utility until changed as provided in” PURA. Id. The record of agency proceedings and the briefs indicate that the Company first initiated, in February 1984, a rate-change proceeding under PURA § 43. Its proposed rates were suspended for 150 days, and continued to be suspended by agreement for an extended period thereafter. The Commission did not establish temporary rates, and the Company did not implement a system of unofficial “bonded” rates. The Company thus continued to charge, pendente lite, its existing rates. By an amendment of the tax code, effective October 1, 1985, the legislature excluded from the gross-receipts tax any sums received by a utility from its sale of telecommunication services. Simultaneously, however, the amendment included the sale of such services within the scope of the sales tax. See 1985 Tex.Gen.Laws, ch. 206, § 10, at 793 [since repealed]; 1959 Tex. Gen.Laws, 3d C.S., ch. 1, art. 11.06, at 304 [since repealed]; 1985 Tex.Gen.Laws, ch. 206, § 3, at 792 [since amended, now codified as Tex.Tax Code Ann. § 151.0101(a)(6) (Supp.1992)]; Tex.Tax Code Ann. § 151.010 (Supp.1992). The Company, after October 1, 1985, began including in its customer invoices a “surcharge” to collect the sales tax while continuing to charge for its telecommunication services according to its existing rates. The Company’s collection of the sales tax was, of course, neutral in its effect on the Company, which simply remitted those collections to the State. But the calculation of charges according to the Company’s existing rates resulted in a “windfall” in a certain sense: because the Company was no longer subject to the gross-receipts tax, the existing rates yielded the Company an equivalent increase in net income. In 1986, the Cities complained to the Commission regarding “the double collection of state taxes,” and the Commission apparently initiated a temporary-rate proceeding under PURA § 43(d). Before this was completed, however, the Company sued in district court to enjoin the Commission proceeding. The court did not enjoin the proceeding, but the temporary-rate issue was apparently abandoned by all parties and the Company continued to charge its existing rates. 20 On March 27, 1987, the Company withdrew its original application for a rate increase according to the “package” which accompanied its statement of intent. On June 1, 1988, however, the Company filed a new “package,” proposing rates that would increase its annual intrastate revenues by about $81.4 million. Although the record indicates that the Company’s proposed rates were suspended, it does not indicate that the Commission established temporary rates or that the Company implemented a system of unofficial “bonded” rates subject to refund. After final hearing, the Commission required the Company to reduce its rates as necessary to diminish its annual revenue by about $59 million, and to refund to its customers, by credits on future invoices, *171 some $140 million. The Commission’s final order also determined: (1) the Commission possessed authority under PURA §§ 42 and 43 to make new rates retroactive to the date the General Counsel filed an answer to the Company’s first application or the date the Company filed its second application; (2) PURA § 43(i) (dealing with rates proposed by a “local exchange company”) “does not make the Commission’s retroactive rate-making authority contingent on the date of its final determination in this case”; and (3) the rates required by the Commission’s final order would be effective January 1, 1987, “in order to do equity in light of [the Company’s] dilatory tactics ... in this case” and because “[t]he gross receipts taxes and federal income taxes embedded in the company’s rates were at higher levels than the company actually paid since that date.” The district court reversed the Commission’s order because it purported to fix the effective date of the Company’s rate retroactively. We agree with this holding and the resulting reversal of the Commission order. On appeal to this Court, the Cities, State Purchasing, Public Counsel, and the Commission complain that the district court erred in reversing the retroactive effective date of the order. The Cities complain that the Commission did not go far enough back in time when it gave retroactive effect to its new rates — that the effective date of January 1, 1987, still permitted the Company a “windfall” by reason of its collecting the sales tax while being exempt from the gross-receipts tax during the calendar year 1986. Because we hold PURA § 42 and § 43(f) preclude the Commission from making its official rates effective at a date earlier than that of its final order under either statute, we will overrule the parties’ points of error regarding retroactivity. We need not, in consequence, discuss the Cities’ contention that the Commission should have set an effective date for the rate decrease which would have allowed ratepayers to recover the “over-collection” received by the Company in 1986. We have held previously that PURA does not authorize the Commission to make its new official rates effective at a date earlier than the date of the order fixing those rates. See Texas Ass’n of Long Distance Tel. Co. (TEXALTEL) v. Public Util. Comm’n, 798 S.W.2d 875, 882 (Tex.App.1990, writ denied); Southwestern Bell Tel. Co. v. Public Util. Comm’n, 615 S.W.2d 947, 955 (Tex.Civ.App.1981, writ ref’d n.r.e., 622 S.W.2d 82); see also Public Util. Comm’n v. United Fuel Gas Co., 317 U.S. 456, 464, 63 S.Ct. 369, 374, 87 L.Ed. 396 (1943). We shall, however, expand on the reasoning by which we reached that view. The purpose of making rates effective at a date earlier than the order fixing them is to compensate for the effects of “regulatory lag,” or unusual delay in the proceeding in which the Commission fixes the rates. Railroad Comm’n v. Lone Star Gas Co., 656 S.W.2d 421, 427 (Tex.1983). Whether any regulatory body possesses the power to make its new rates effective earlier than the date it fixes the rates is a question of statutory construction — did the legislature delegate that power to the regulatory authority? The issue thus comes within the general rules: a commission has only those specific powers conferred upon the agency by law, in clear and express language, as well as any power necessary to make effective any specific power that is granted in clear and express language. One may not, however, on a theory of necessary implication, impute to the commission a power, function, or duty that really amounts to a new and additional power or one that contradicts a relevant statute, no matter how expedient the new power might be for purposes of administration. See Sexton v. Mount Olivet Cemetery Ass’n, 720 S.W.2d 129, 137-38 (Tex.App.1986, writ ref’d n.r.e.). Moreover, the possible implication of a power does not extend so far as to vest in the agency an implied power to supplant a method or procedure that the legislature itself has designated for the circumstances. The legislature’s method or procedure prevails over that of the agency; “the prescribed method excludes all others, and must be followed.” Cobra Oil & Gas Corp. v. Sadler, 447 S.W.2d 887, 892 (Tex.1968); Foster v. City of Waco, 113 *172 Tex. 352, 255 S.W. 1104, 1105 (1923); Balios v. Texas Dep’t of Public Safety, 733 S.W.2d 308, 311 (Tex.App.1987, writ ref'd). The only relevant statutory provisions requiring construction are those contained in PURA. Nothing in PURA purports to delegate to the Commission, in clear and express language, a power to assign new rates an effective date earlier than the order in which they are fixed. Indeed, PURA establishes a uniform principle that such rates shall be effective no earlier than the date of the order fixing them. This is declared most explicitly in PURA § 26(g), a statute not applicable here, but it is also declared explicitly in those that are applicable — PURA §§42 and 43(e). Both direct the Commission to determine and fix just and reasonable rates in an order served on the utility, and conclude with the mandate that such rates shall “thereafter” be “observed” until changed in a manner authorized by PURA. The word thereafter “in terms thus gives the Commission power to prescribe such rates prospectively only.... There is no basis in the statute for concluding the Commission’s orders can be retroactive to the date when the Commission’s inquiry into the rates was begun; on the contrary, the explicit language of the statute precludes such a construction.” United Fuel Gas Co., 317 U.S. at 464, 63 S.Ct. at 374 (emphasis added). Far from conferring “in clear and express language” the power to make rates effective at a date earlier than the order fixing them, the language in PURA §§ 42 and 43(e) provides precisely to the contrary. We shall, nevertheless, continue in our analysis of the Commission’s claim to have the power by implication. The Commission determined that the early effective date of the new rates was required “in order to do equity in light of [the Company’s] dilatory tactics.” This falls squarely within the purpose of a legislative grant of the power to make rates effective at a date earlier than the date they are fixed — the purpose of providing a remedy for the effects of “regulatory lag.” But the legislature provided in PURA its own statutory remedy for “regulatory lag.” We refer, of course, to the system of rate suspension, temporary rates fixed by the Commission, and “bonded-in” rates fixed by the utility when unusual delay occurs, as this system is established in PURA § 43(d) and (e) to provide for rates effective pendente lite. There is no similar provision for temporary and “bonded-in” rates in Commission-initiated rate proceedings under PURA § 42, the legislature being content evidently to rely upon the Commission’s diligence in ferreting out when and to what extent a utility’s rates have become unjust, unreasonable, or in violation of law, using its investigative powers for the purpose, as it did in Pedernales Elec. Coop., 678 S.W.2d at 214. As indicated in Cobra Oil & Gas Corp. and the other decisions listed above, we are not free to impute to the Commission, on a theory of necessary implication, a power to provide its own remedy for “regulatory lag” to be utilized in lieu of the remedy prescribed by the legislature, especially when the Commission’s remedy contradicts the clear and express language that utility rates shall be effective after the Commission’s order fixing them. In the present case, the terms of PURA §§ 43(d) empowered the Commission to fix temporary rates to serve the purpose of compensating for the Company’s “dilatory tactics.” No additional power was required by way of necessary implication. In its motion for rehearing, Public Counsel argues that if the Commission lacks the power to make rates effective at a date earlier than the order that fixes them, then the Commission should have the power under PURA § 42 to suspend a utility’s existing rates and to set temporary rates fixed by the Commission. (The Commission initiated a proceeding under PURA § 42 after the Company initiated its own rate-change application under PURA § 43.) As proof of the Commission's need, Public Counsel cites the length of the proceeding in the present cause, which began in 1984. We reject the argument. The power to suspend the Company’s rates and to establish temporary rates existed in the present case by virtue of PURA § 43(d), for the Company itself initiated a proceeding to change *173 its rates under PURA § 43. This is, therefore, not an apt case to raise the contention. Moreover, it is not our place to supply omissions in legislation concerning the problem of “regulatory lag”; the change should come from the legislature. Lone Star Gas Co., 656 S.W.2d at 427. More importantly, perhaps, it appears to us that such power might be inconsistent with the general tenor and purpose of PURA § 42. In those proceedings, the legislature elected apparently to rely upon a diligent Commission in initiating rate inquiries on its own initiative. In any case, we are not free in the present case to consider whether such a power exists by necessary implication; and, of course, we do not purport to make a decision in the matter. In its motion for rehearing, the Commission contends we err in finding that PURA does not authorize the Commission to make its new rates effective earlier than the date of its order and service of the order on the utility. The Commission argues first from PURA § 2, which requires the Commission to balance the interests of utilities and consumers in setting rates. 21 We reject the argument. PURA §§ 42 and 43 provide expressly and clearly that rates shall be effective after they are found to be just and reasonable by the Commission, in an order served upon the utility, as discussed above at length. If the Commission possessed the power to make its new rates effective at an earlier time, by necessary implication from PURA § 2, then the more specific language of PURA §§ 42 and 43 would be unnecessary and indeed meaningless. It is a familiar rule of statutory construction, of course, that a specific provision governs and qualifies a general provision. Ayre v. Brown & Root, 678 S.W.2d 564, 566 (Tex.App.1984, writ ref’d n.r.e.). The Commission argues next that our reasoning and interpretation of PURA §§ 42 and 43 are contrary to the precedents mentioned below. We do not believe the supreme court rejected our views by its decision in Railroad Comm’n v. Moran Util. Co., 728 S.W.2d 764 (Tex.1987). There, the court expressly based its decision solely on an explicit provision for the refund of illegally collected sums, as authorized by the terms of Tex.Rev.Civ.Stat.Ann. art. 6055 (1962). Moran Util., 728 S.W.2d at 767 n. 2. The court noted, moreover, that reasonable statutory procedures affecting a utility’s rates (such as those setting the effective date of rate orders) do not amount to an unconstitutional confiscation of the utility’s property. Id. at 768. Nor do we believe the supreme court rejected our views by its decision in Lone Star Gas Co. The court in Lone Star expressly limited the effect of its decision to cases arising before September 1, 1983, the effective date of “the new Gas [Utility] Regulatory Act.” Lone Star Gas Co., 656 S.W.2d at 427. It is not within our power to override this express limitation concerning the effect of the supreme court’s decision. The significance of September 1, 1983, was that it was the effective date of the new Gas Utility Regulatory Act, an enactment that made express provisions to compensate for regulatory lag — provisions allowing for “bonded-in” temporary rates in terms almost identical to those found in PURA § 43(e). 1983 Tex.Gen.Laws, ch. 263, § 20, at 1209. This provision of the new enactment cured the problem raised in Lone Star Gas Co. — there was no provision for a utility to “bond-in” rates in rate eases initiated before the governing body of a municipality (the City of Kaufman in that case) and reviewed de novo by the Commission: “The rates within the city of Kaufman over which the Commission exercised appellate jurisdiction and which represent most of Lone Star’s revenues in controversy are not subject to being ‘bonded in.’ ” Lone Star Gas Co., 656 S.W.2d at 425; see also Arkansas Louisiana Gas Co. v. Railroad Comm’n, 586 S.W.2d 643 (Tex.Civ.App.1979, writ ref’d n.r.e.). Moreover, *174 the new enactment provided expressly that “[a]ny rates, whether temporary or permanent, set by the railroad commission shall be prospective and observed from and after the applicable order of the commission, except interim rate orders necessary to effect uniform systemwide rates.” 1983 Tex.Gen Laws, supra, at 1196. In summary, we believe it clear that the supreme court held as it did in Lone Star Gas Co. because there was in the relevant regulatory system no provision prescribed by the legislature to compensate for the effects of unusual regulatory lag; hence it was permissible to rely upon an implied power. Such is not the present case, of course, where the legislature has provided for “bonded-in” rates for that purpose and those statutory provisions in PURA § 43(d) were available for the purpose. In Lone Star Gas Co. the implied power did not supplant a legislative remedy; in the present cause, it would. We therefore hold that the district court did not err in reversing the Commission’s order insofar as it purported to implement a retroactive effective date. 22 The points of error regarding that issue are overruled. 23 THE DISTRICT-COURT JUDGMENT The district-court judgment reverses the Commission’s final order on the Company’s single complaint that the Commission erred as a matter of law when the agency assigned the new rates an effective date of January 1, 1987. Concerning the Company’s other complaints and those of the other parties in the cause, the district-court judgment declares them moot or decided adversely to the party making them, without identifying any particular complaint or party. We affirm that portion of the district-court judgment that reverses the Commission’s final order on the ground that that order purports to set an effective date for the rate change earlier than the date of the order fixing the rate. After reversing the Commission’s final order, however, the district-court judgment goes further. It directs the Commission to implement a surcharge that would allow the Company to recover any sums previously refunded by the Company, together with “interest and other costs associated with such refund.” The judgment expressly leaves to the Commission the issue of whether the surcharge should include an amount allowing the Company to recover “attorney’s fees.” We disagree with this aspect of the district court judgment, and remand the cause to that court with instruction that the remand to the agency be general. The Revised Model State Administrative Procedure Act of 1961 (from which APTRA was taken) allowed reviewing courts to modify agency orders in appropriate circumstances. This authorization to modify orders was deleted in APTRA “in order to retain the Texas rule that courts may not write certain orders for the agency or direct it to do so in specific terms.” Dudley McCalla, The Administrative Procedure and Texas Register Act, 28 Baylor L.Rev. 445, 492 (1976). The Texas Supreme Court had stated earlier that a court “has the right to review the action of the Railroad Commission, and to strike its orders down if they are illegal; but it has no authority to write a proration order for the Commission, nor to prescribe the terms of the subsequent order to be entered by the Commission.” Marrs v. Railroad Comm’n, 142 Tex. 293, 177 S.W.2d 941, 950 (1944). The court added that the discretion to devise a proration scheme had been committed to the Railroad Commission, and it would be a usurpation *175 of the Commission’s authority for the court to prescribe the terms of the order. Id. That same reasoning applies in this case; a reviewing court may not make substantive decisions committed by statute to the Public Utility Commission. In its motion for rehearing, the Company cites opinions by this Court for the correct proposition that a reviewing court may control the scope of a remand to the agency. See First Sav. & Loan Ass’n v. Lewis, 512 S.W.2d 62, 64 (Tex.Civ.App.1974, writ ref’d n.r.e.). This Court has also stated in a utility rate case that if a reviewing court sustains a utility’s allegations of error, “the proper judgment will be one of remand to the agency with instructions to re-determine the utility’s annual revenue increase in light of the final decisions made upon the utility’s various contentions of error.” Southwestern Bell Tel. Co., 615 S.W.2d at 955. It does not follow, however, that the power to control the scope of a remand necessarily authorizes a district court to modify the agency’s final order. As the supreme court declared in Marrs, the terms of the final order are matters committed to agency discretion as long as the agency corrects the assignment of errors sustained on appeal. The terms of APTRA § 19(e) authorize expressly only a general remand when a party’s substantial rights are prejudiced by an agency error of the kind described in subsections (1) through (6). It is sufficient that the cause must be remanded generally to the Commission in order to effectuate our decision and judgment, which rests on the several grounds given above. 24 We therefore affirm that part of trial-court judgment reversing the Commission’s order. We reverse that judgment insofar as it affirms the Commission’s order. We remand the cause to the district court with instructions that the cause be remanded to the Commission for proceedings not inconsistent with our opinion. Footnotes [1] The Cities’ briefs on appeal list eighty-eight cities, but only eighty-six are shown on the cities’ notice of appeal. [2] Throughout our opinion we have added the emphasis where it appears in quotations. [3] Throughout our discussion, we refer to the Company’s common stock as its "common equity" merely because we do not wish to deviate from the terminology used by the Commission in the proceeding below. We note, however, that the term "equity” is broadly defined as "the spirit and the habit of fairness, justness, and right dealing which would regulate the intercourse of men with men, — the rule of doing to all others as we desire them to do to us.” Black’s Law Dictionary 634 (rev. 4th ed. 1968). The term has also come to mean "the remaining interest belonging to one who has pledged or mortgaged his property, or the surplus of value which may remain after the property has been disposed of for the satisfaction of liens.” Id. Although some use the term "common equity” to mean "common stock," see Utilities Comm’n v. Public Staff, 323 N.C. 481, 374 S.E.2d 361, 364 (1988); Douglas W. Hawes, Utility Holding Companies, § 5.07 (1984), we do not approve such misuse of the term "equity.” We employ that usage only for the reason stated above. [4] The 10.85% figure is the cost of equity that Public Counsel’s witness Carol Szerszen recommended during her cross examination. We note that the Examiner’s Report attributes to Szerszen a recommended cost of equity of 10.-73%. [5] GTE-Southwest argues that because it and GTE Directories are involved in a joint venture, it does not pay an expense to GTE Directories, and the arrangement therefore does not come within the scope of § 41(c)(1). The administrative law judge explicitly rejected this argument, and the Commission made no findings that would support such a position. We may not find facts which the Commission declined to find. See Gulf Land Co. v. Atlantic Ref. Co., 134 Tex. 59, 131 S.W.2d 73, 84 (1939) (‘‘[T]he Commission determines primarily and finally all fact issues — that is, all issues that are not estáb-lished as a matter of law."). [6] We need not address the Cities' other grounds of attack on Finding of Fact 16, as we sustain the Cities' point of error for the reasons stated in the text of our opinion. Furthermore, the Company contends and the Commission concluded in its “Finding of Fact 16,” the Commission could not equitably disallow expenses incurred by the Company in its transactions with GTE Directories and, simultaneously, include in the rate calculation revenues received from the directory sales. We disagree. As Commissioner Campbell pointed out in her dissent, the payments to GTE Directories comprise only part of the expense the Company incurs in connection with the publication of the directories. The Commission did not "disallow” other expenses incurred in the publication of the directories which are not payments to affiliates. [7] Other "Conclusions of Law” relevant to the consolidated-tax-return issue include the following: CL 13. Section 27(e) [sic] requires utilities to keep separate accounts to show profits and losses from the sale of equipment not integral to the provision of utility service, and it prohibits the Commission from considering such profits in determining a utility’s rates. CL 15. Because GTE Corp’s consolidated return assigns to each subsidiary its proper income tax liability, reducing GTE Southwest’s income-tax expense would reduce its deferred federal income-tax liability in violation of the normalization provisions of the Internal Revenue Code and related regulations. Such a treatment would also violate the consistency requirement in the normalization rules. As a result, it would violate the requirement in PURA and the Commission’s substantive rules requiring utilities to compute federal income taxes on a normalized basis. [8] [[Image here]] [9] We assume the reference to the Examiner’s Report contained a typographical error. The discussion of the consolidated-tax-return issue is in section III.C. Section III.B. discusses a completely different issue. [10] The utility can avoid the computation required by PURA § 41(c)(2) by convincing the regulatory authority that it is reasonable to choose not to consolidate returns. Perhaps the Commission’s conclusory determinations can be said to amount to a determination that the exception in PURA § 41(c)(2) applies: that the Company had "shown to the satisfaction of the [Commission] that it was reasonable to choose not to consolidate returns.” In either case, however, the Commission’s conclusory determinations refer necessarily to the statutory factors that PURA § 41(c)(2) required the Commission to consider in calculating the Company’s income-tax liability for inclusion in its operating expenses for rate-making purposes. Thus, if the Commission believed that the exception applied in this case, it was bound to include in its final order an ultimate finding to that effect, as well as findings of basic or underlying fact that supported its conclusory determinations. APTRA § 16(b); Texas Health Facilities Comm’n v. Charter Medical-Dallas, Inc., 665 S.W.2d 446, 451 (Tex.1984). The absence of an ultimate finding and supporting findings implies that the Commission did not act upon the exception carved from the general rule. See APTRA § 19(e)(1). [11] In an amicus curiae brief, Texas Utilities Electric Company argues that this broad statement is dicta and therefore does not control on income-tax issues. In the absence of any other opinion enlarging the boundaries of the Commission's discretion with regard to tax savings, however, we take this to be the supreme court’s view of the matter. It is up to that court to narrow its language if it wishes to do so. [12] It has been said that "the imprecision of the ‘actual taxes paid’ formulation is exceeded only by the name of the Holy Roman Empire: two out of the three words are wrong. Taxes, yes. But not necessarily actual taxes, since inexact estimations are often allowed.” City of Charlottesville, Va. v. F.E.R.C., 774 F.2d 1205, 1215 (D.C.Cir.1985) (emphasis in original), cert. denied, 475 U.S. 1108, 106 S.Ct. 1515, 89 L.Ed.2d 914 (1986). Furthermore, as Judge (now Justice) Scalia points out in the majority opinion in Charlottesville, taxes included in the rate calculation are often not "paid,” but instead "incurred,” as "normalization" rules permit utilities to pass along to ratepayers, on a straight-line schedule, taxes deferred by the use of accelerated depreciation. Charlottesville, 774 F.2d at 1215. [13] We note that the Internal Revenue Service has withdrawn the notice of a proposed rule entitled Normalization: Inconsistent Procedures and Adjustments, 55 Fed.Reg. 49294 (1990), which addressed the extent to which "certain utility ratemaking procedures and adjustments that are based on tax savings attributable to the filing of a consolidated tax return” violate normalization requirements. See 56 Fed.Reg. 19825 (1991). [14] In Houston Lighting the Commission disallowed $166 million as imprudent expenses for ratemaking purposes, but required the utility to pass on to its ratepayers the tax write-off for these expenses. The supreme court required the shareholders of Houston Lighting & Power’s parent corporation, Houston Industries, Inc., to bear the losses. Houston Lighting, 748 S.W.2d at 443 (Wallace, J., dissenting). In Houston Lighting the unregulated entity had to bear the losses of the regulated affiliate; here, the regulated affiliate had to bear the losses of unregulated affiliates. We do not believe this distinction warrants a different result. [15] Texas Utilities argues that ”[t]o suggest that the [Houston Lighting] majority rejected the possibility that their decision could result in normalization violations or were unpersuaded by such arguments ... is to incorrectly stretch silence into affirmative disapproval.” We disagree. The normalization argument was obviously made to the supreme court in the motion for rehearing. From their silence on the issue of normalization violations and their statement that tax savings "should inure to the benefit of the ratepayers," we infer that the court saw no violation of normalization rules. [16] Texas Utilities contends that requiring the consolidation of tax returns of regulated and nonregulated affiliates is inequitable and confiscatory. It argues that the utility's customers benefit when the nonregulated affiliates’ losses reduce the tax liability of the consolidated group, but the utility’s customers are not required to pay higher rates when the nonregulat-ed affiliates’ gains create higher taxes. According to Texas Utilities, this confiscates the assets of the nonregulated entity, simply because it is affiliated with the utility. If this result is truly inequitable, that is a matter for the legislature to remedy by amending PURA § 41(c)(2), or for the supreme court to remedy by narrowing the holding of Houston Lighting. One notes, however, that the parent company controls the way in which it structures the ownership of its affiliates. [17] While GTE-Southwest may receive no tax savings from this method, the parent corporation surely does. It benefits to the extent of the time value of money that it saves from deferred taxes. Moreover, GTE-Southwest finds itself in the position of arguing that it should be allowed to subsidize its affiliates with the revenue it receives from ratepayers. We do not believe this argument is consistent with the holding in Houston Lighting. [18] The following are relevant portions of PURA § 41(c)(3): Expenses Disallowed. The regulatory authority shall not consider for ratemaking purposes the following expenses: (A) legislative advocacy expenses, whether made directly or indirectly, including but not limited to legislative advocacy expenses included in trade association dues; [[Image here]] (D) any expenditure found by the regulatory authority to be unreasonable, unnecessary, or not in the public interest, including but not limited to executive salaries, advertising expenses, legal expenses, and civil penalties or fines. [19] We limit our holding to disallowed non-capital expenses. We do not decide the proper treatment of disallowed capital expenses. [20] State Purchasing and the Commission both assert that the Commission did not abandon the temporary-rate issue, but instead was "judicially prevented" from setting such rates. We disagree with their contention. The record reflects that the district judge never signed an order enjoining the Commission from setting temporary rates. The Commission apparently pursued the matter no further after receiving a letter from the district judge stating he was prepared to issue a temporary injunction. The Commission cannot now complain of the district judge's action when it failed to secure a ruling on the temporary-rate issue. [21] PURA § 2 reads in part, "The purpose of this Act is to establish a comprehensive regulatory system which is adequate to the task of regulating public utilities as defined by this Act, to assure rates, operations, and services which are just and reasonable to the consumers and to the utilities.” [22] Because we base our holding on the PURA provisions against retroactive ratemaking, we need not address Public Counsel’s fourth point of error, which contends the district court erred in its conclusion that setting a retroactive effective date for a rate change is unconstitutional. [23] Because we hold that PURA prohibits the Commission from setting an effective date for a rate change which precedes the final order, we need not address an argument posed by the Cities and State Purchasing that the Commission's use of a retroactive effective date constituted proper punishment for the Company’s use of unspecified "dilatory tactics.” [24] Because we reverse the district-court judgment on the grounds stated above, we need not address the Commission's point of error alleging that the district court erred in not sustaining its pleas to the jurisdiction and motions to strike the intervention of several parties. In its motion for rehearing, the Commission contends that this Court should have decided whether Public Counsel was entitled to intervene. The Commission asserts that Public Counsel failed to invoke the jurisdiction of the district court. According to the Commission, a party's right to appear before the court is a fundamental issue; therefore, this Court’s reversal of other legal rulings by the district court does not settle the question whether the district court had jurisdiction over Public Counsel and its claims. We reject this argument. This Court is to address every issue "necessary to final disposition of the appeal." Tex.R.App.P.Ann. 90(a) (Pamph.1992). Deciding the issue of Public Counsel’s right to intervene was not necessary to the Court’s disposition because the Cities raised the same points of error that Public Counsel raised.