Central Power & Light Co./Cities of Alice v. Public Utility Commission

36 S.W.3d 547, 2000 WL 1028257

• Court: Court of Appeals of Texas
• Decided: February 28, 2001
• Docket: 03-99-00111-CV
• Status: Published

Opinion

J. WOODFIN JONES, Justice.

This is an administrative appeal from an order of the Public Utility Commission (the Commission) reducing rates charged for electric utility service by the Central Power and Light Company (CPL). The City of Alice and 85 other cities (Cities), as well as the Office of Public Utility Counsel (OPC), intervened in the administrative proceedings and appealed portions of the Commission’s order to the district court. The district court reversed two aspects of the Commission’s order, only one of which has been appealed, and affirmed the order in all other respects. The Commission, Cities, and OPC appeal that portion of the trial court’s judgment determining that the Commission’s calculation of consolidated tax savings constituted retroactive rate-making. Cities also bring two independent issues on appeal, complaining that the Commission (1) violated the affiliate-transaction statute by including factoring costs paid by CPL in cost-of-service calculations, and (2) erroneously included expenses associated with the promotion of increased consumption of electricity in its rates. CPL brings four issues on appeal, complaining that (1) the Commission set an unreasonably low rate of return on invested capital that is not supported by substantial evidence, (2) certain findings and conclusions in the Commission’s order should be reversed and stricken, (3) the Commission should have included actual attendant impacts from two new facilities in the test-year calculations, and (4) CPL should have been allowed to recover $12.6 million in disallowed affiliate expenses. 1 We will affirm the trial court’s judgment in part and reverse and remand in part.

FACTUAL AND PROCEDURAL BACKGROUND

Overview of Electric Utility Ratemaking

Utility ratemaking is governed by the Public Utility Regulatory Act (PURA). See Tex. UtiLCode Ann. §§ 36.001-.353 (West 1998 & Supp.2000). 2 Utility rates are set prospectively, based on the utility’s actual costs in the most recent historical test year, adjusted for known and measur able changes. See 16 Tex. Admin. Code § 23.21(b) (1999). The Commission is required to set rates that “will permit the utility a reasonable opportunity to earn a reasonable return on the utility’s invested capital used and useful in providing service to the public in excess of the utility’s reasonable and necessary operating expenses.” PURA § 36.051. Thus, PURA directs the Commission to examine financial information taken from a historical test year and calculate rates based on three factors: (1) the amount of invested capital that the utility uses to provide service to its customers, (2) a reasonable return on that invested capital, and (3) the amount of the utility’s reasonable and necessary operating expenses.

In determining a utility’s invested capital, also known as its “rate base,” the Commission determines the “original cost, less depreciation, of property used by and useful to the utility in providing service.” PURA § 36.053(a). The rate of return on that invested capital is figured by determining the cost of capital. The elements of cost of capital include interest paid on debts, dividends paid on preferred stock, and earnings on common stock. See Southern Union Gas Co. v. Railroad Comm’n, 692 S.W.2d 137, 141 (Tex.App.— Austin 1985, writ ref d n.r.e.). The cost of debt capital is easily calculated; it is simply the interest paid to creditors. See id. The cost of preferred stock is also easily quantified; it is the required dividend that must be paid to preferred shareholders. See id. The cost of common stock is not so readily ascertainable; experts testify regarding their assessment of the return on equity necessary to enable the utility to attract necessary capital from investors. See id. Once the utility’s cost of capital is determined, that figure is applied to the rate base to arrive at a proper rate of return. See id.; see also PURA § 36.052.

The final element of the Commission’s ratemaking formula is the determination of the utility’s “reasonable and necessary” operating expenses. The Commission starts with the utility’s expenses incurred during the test year and then adjusts those expenses for known and measurable changes. See 16 Tex Admin. Code § 23.21(b) (1999). Operating expenses include depreciation expenses, federal income taxes and other taxes, and other “reasonable and necessary” costs of doing business. See id. § 23.21(c). Some payments to affiliates for services may also be included in operating expenses, although there is a statutory presumption against allowing such expenses. See PURA § 36.058 (West Supp.2000).

Proceedings in this Case

CPL initiated this ratemaking proceeding in November 1995 when it filed a request with the Commission seeking a rate increase. See PURA § 36.151. The rate-fifing package accompanying the request for an increase included data based on the 1995 test year. See PURA § 35.153. Cities and some twenty-six other entities intervened to oppose the request.

The Commission referred the case to the State Office of Administrative Hearings (SOAH), where hearings were held between February and October 1996. In January 1997, the SOAH Administrative Law Judges (ALJs) who presided over the proceedings issued a proposed decision that would have ordered an increase in CPL’s revenues. The Commission rejected this proposal and instead issued an initial order on March 31, 1997, ordering a decrease in CPL’s rates. In response to motions for rehearing, the Commission remanded the cause to SOAH for further hearings. The Commission issued a Second Order on Rehearing on October 15, 1997, ordering an initial rate reduction of $26.8 million in 1997, and further reductions of $13 million in May 1998 and May 1999. It is that order that is the subject of this appeal.

CPL, Cities, and Texas Industrial Energy Consumers (TIEC) 3 filed suit in dis- triet court seeking judicial review of the Commission’s order. See PURA § 15.001. The district court reversed the portion of the Commission’s order calculating CPL’s consolidated tax savings, holding that the method used constituted retroactive rate-making. One other portion of the order, not challenged here, was also reversed. The trial court affirmed the remainder of the Commission’s order. With no party entirely satisfied by the trial court’s judgment, CPL, the Commission, Cities, and OPC have all appealed the portions of that judgment adverse to them.

DISCUSSION

Issue raised by the Commission

The Commission, Cities, and OPC all complain that the trial court erred in ruling that the Commission’s calculation of CPL’s consolidated tax savings constituted retroactive ratemaking. CPL asks us to affirm the trial court’s reversal of that portion of the Commission’s order, arguing that the Commission’s attempt to recover consolidated tax savings is an impermissible form of retroactive ratemaking and that the method employed by the Commission in this calculation is arbitrary and capricious and not supported by substantial evidence.

Ratemaking has been likened to a legislative activity, even though it is carried out by an administrative agency. See City of Alvin v. Public Util. Comm’n, 876 S.W.2d 346, 362 (Tex.App.— Austin 1993), judgm’t vacated w.r.m. sub nom. Public Util. Comm’n v. Texas-New Mex. Elec. Co., 893 S.W.2d 450 (Tex.1994). Therefore, the constitutional prohibition on ex post facto or retroactive laws applies. See id.; Tex. Const, art. I, § 16. Utility rates generally may have only prospective effect, and the Commission may not set rates that allow a utility to recoup past losses or refund excess utility profits to consumers. See State v. Public Util. Comm’n, 883 S.W.2d 190, 199 (Tex.1994); City of Alvin, 876 S.W.2d at 362. This principle is reflected in PURA § 36.111, which states that “[t]he rates established in the order shall be observed thereafter until changed as provided by this title.” PURA § 36.111(b) (emphasis added).

The rates a utility charges its customers must be set so as to produce revenues equal to the sum of two amounts: (1) the utility’s “reasonable and necessary operating expenses,” and (2) “a reasonable return on the utility’s invested capital used and useful in providing service to the public.” PURA § 36.051; see Cities for Fair Util. Rates v. Public Util. Comm’n, 924 S.W.2d 933, 935 (Tex.1996). Federal income tax expense is one of the operating expenses a utility may recover in its rates. See Cities for Fair Util. Rates v. Public Util. Comm’n, 884 S.W.2d 540, 544 (Tex. App.— Austin 1994), rev’d in part on other grounds, 924 S.W.2d 933 (Tex.1996).

When considering federal income taxes as part of a utility’s cost of service, PURA requires the Commission to determine a utility’s “fair share” of the tax benefits that result when its parent company flies a consolidated tax return:

(a) Unless it is shown to the satisfaction of the regulatory authority that it was reasonable to choose not to consolidate returns, an electric utility’s income taxes shall be computed as though a consolidated tax return had been filed and the utility had realized its fair share of the savings resulting from that return, if:

(1) the utility is a member of an affiliated group eligible to file a consolidated income tax return; and

(2) it is advantageous to the utility to do so.

PURA § 36.060(a). The supreme court has noted that the legislature has granted the Commission broad discretion in determining a utility’s “fair share of the savings” arising from the filing of a consolidated tax return by the parent company. See Public Util. Comm’n v. GTE-SW, Inc., 901 S.W.2d 401, 410 (Tex.1995).

In addition to owning CPL, parent company Central and South West Corporation (CSW) also owns numerous other entities. Some of those other entities are also regulated utilities, others are non-regulated entities; some provide services to the regulated entities, others are businesses with activities totally unrelated to CPL’s provision of electric service. Many of the affiliates are unprofitable, so CSW benefits from filing a consolidated tax return for all of its subsidiaries. The most obvious saving that arises from filing a consolidated tax return is the reduction of overall annual tax expense by using losses of unprofitable affiliates to offset gains of profitable entities; this allows CSW to reduce overall taxable income and thereby reduce taxes owed for that year. As one of CSW’s currently profitable entities, CPL has paid less in taxes in recent years than it would have if there had been no affiliate losses to offset CPL’s gains.

Less obvious and less easily quantifiable are the secondary benefits conveyed to the unprofitable affiliates by profitable companies like CPL. When these proceedings began, federal income tax law allowed a business to carry forward its losses for no more than fifteen years. See 26 U.S.C.A. § 172(b)(1)(B) (West 1988), amended by Pub.L. 105-34, § 1082(a)(2) (1997). 4 Hence, if an unprofitable CSW subsidiary failed to earn profits of its own within fifteen years of sustaining a net loss, CSW would lose the tax benefit of that loss but for the existence of profits earned by affiliates such as CPL. Additionally, by filing a consolidated tax return, CSW is able to enjoy the tax benefits of the loss in the year in which the loss is incurred, rather than having to wait until a later year when a loss affiliate might earn a profit against which the loss can be applied. The presence of inflation generally means a tax loss has less value in later years. 5 Moreover, the unprofitable, non-regulated affiliates benefit from their association with profitable entities such as CPL. Essentially, CPL finances the losses of those unprofitable companies, thereby giving them an economic edge over competitors.

Recognizing that CSW’s non-regulated unprofitable affiliates benefit from filing a consolidated return with profitable utilities, the Commission ordered a “consolidated tax savings” adjustment to recognize these secondary benefits. In its findings of fact related to this issue, the Commission laid out the following formula for calculating the consolidated tax savings:

112A. In view of the advantage CSW competitive affiliates gain over competitors and to compensate CPL for the benefits it conveys to unprofitable CSW affiliates, CPL should be compensated for the value of the tax shield it provides to CSW affiliates.

112B. The value of the tax shield CPL provides to CSW competitive affiliates is equal to the amount of consolidated tax savings over the last fifteen years that would not have been realized by CSW affiliates as of the test year, but for their affiliation with CPL, multiplied by the time-value of money.

112C. CPL’s long-term cost of borrowing, 7.96%, is appropriate to measure the time-value of money for purposes of consolidated tax savings because that percentage reflects CPL’s cost of borrowing.

112D. The consolidated tax savings, calculated by multiplying the FIT [federal income tax] savings by the time-value of money, equals $6,056,178.

CPL complained, and the trial court agreed, that the Commission’s calculation of that approximately $6 million consolidated tax savings represents an attempt to retroactively recover past tax savings. To do so would violate the prohibition against retroactive rate-making. See Public Util. Comm’n, 883 S.W.2d at 199; City of Alvin, 876 S.W.2d at 362.

We disagree that the Commission’s calculation constitutes retroactive ratemak-ing. The formula initially calculates fifteen years’ worth of savings enjoyed by CSW as a result of using its profitable affiliates’ gains to utilize the tax benefits of its unprofitable affiliates’ losses. The Commission calculated the losses and gains of each subsidiary for the years 1980 to 1994, the fifteen-year period concluding with the test year. Then, looking at the gains of each subsidiary standing alone, the Commission calculated which of those losses would have been offset by the subsidiary’s own gains during that fifteen-year period. The remaining losses — losses that would have had no value to CSW during the fifteen-year period if entities like CPL had not earned profits against which those losses could be applied — were multiplied by the tax rate for the year each loss was sustained to determine each year’s tax savings to CSW. Then, all fifteen years’ worth of savings were combined, producing total tax savings to CSW of about $373 million. In order to determine CPL’s “fair share” of these savings, the total figure was multiplied by the percentage that represents CPL’s share of CSW’s overall profits during those years (20.42%). The resulting figure — approximately $76 million in tax savings — was multiplied by CPL’s long-term debt rate, 7.96%, resulting in the $6 million consolidated tax savings to CPL reflected in the Commission’s order.

What the Commission apparently concluded is that CPL’s fair share of CSW’s tax savings — $76 million — is tantamount to free capital provided to CPL over the last fifteen years. Because of those savings, CPL was able to invest or spend $76 million it would not otherwise have had. In effect, the Commission asked how much CPL would have had to pay each year in interest charges if it had borrowed that $76 million instead of being “given” that amount through CSW’s tax savings. Likening the $76 million in savings to loan principal and multiplying that figure by CPL’s long-term debt rate, the Commission has essentially determined that CPL would have had to pay $6 million per year to obtain $76 million in capital. Thus, $6 million represents the annual time-value of the $76 million that CPL has been given by virtue of the tax shield its profits provide CSW for the losses of other affiliates.

CPL argues that because the Commission recalculated fifteen years worth of past tax savings to come up with the $6 million figure, its method necessarily constitutes retroactive ratemaking. CPL believes that because the rates charged during the fifteen-year period were governed by previous Commission orders, the Commission cannot revisit those years now to recalculate what it claims are previously unaccounted-for savings. We disagree. It is true that the method employed here is novel, or at least has not been tested in any published cases. 6 Nonetheless, we see nothing in the Commission’s method that violates the prohibition against retroactive ratemaking. The savings figure calculated by the Commission has only prospective application and so does not violate the literal language of PURA section 36.111. See PURA § 36.111(b). The Commission has not attempted to recoup past losses or refund excess utility profits to consumers with this consolidated tax savings adjustment. See Public Util. Comm’n, 883 S.W.2d at 199; City of Alvin, 876 S.W.2d at 362. Moreover, the Commission has not attempted to recoup the actual $76 million “free capital” allocated to CPL in an attempt to recoup past savings. Instead, the Commission multiplied that “principal” by the long-term debt rate in order to recognize today’s value of obtaining that capital. As long as the Commission is not trying to recoup past savings, but only trying to recover today’s benefit from those past savings, its calculation of consolidated tax savings does not constitute retroactive ratemaking. We therefore hold that the trial court erred in concluding that the Commission’s computation of CPL’s “fair share of the savings resulting from [fifing a consolidated tax return]” violated the prohibition on retroactive ratemaking. PURA § 36.060.

Nonetheless, we uphold the trial court’s reversal of this portion of the Commission’s order because we conclude that the Commission’s method for calculating consolidated tax savings is unsupported by the record. We must reverse and remand for further administrative proceedings “if substantial rights of the appellant have been prejudiced because the administrative findings, inferences, conclusions, or decisions are ... not reasonably supported by substantial evidence considering the reliable and probative evidence in the record as a whole.... ” Tex. Gov’t Code Ann. § 2001.174 (West 2000). In searching for substantial evidence, we review the record to determine whether some reasonable basis exists for the agency’s action. See Mireles v. Texas Dep’t of Pub. Safety, 9 S.W.3d 128, 131 (Tex.1999) (citing City of El Paso v. Public Util. Comm’n, 883 S.W.2d 179,185 (Tex.1994)).

No witness testified in support of a method of calculating consolidated tax savings by determining today’s value of past tax savings. The witnesses cited by the parties discussed whether this kind of savings should be quantified at all since the savings attributed to CPL were provided at no cost to the utility. The witnesses also discussed whether tax savings were adequately accounted for in CSW’s own tax allocation plan, which the Commission ultimately decided to disregard. The Commission’s having decided that the savings at issue here were to be taken into consideration in calculating consolidated tax savings, the debate centered on how the savings should be reflected in the utility’s rates. Some witnesses advocated deducting the savings from CPL’s rate base, while others advocated deducting them from CPL’s cost of service.

The method used by the Commission must be supported by substantial evidence in the record. See Railroad Comm’n v. Moran Util. Co., 728 S.W.2d 764, 766 (Tex.1987); Railroad Comm’n v. Lone Star Gas Co., 618 S.W.2d 121, 124 (Tex.Civ.App.— Austin 1981, no writ). In crafting its order, the Commission is not bound to accept the testimony of any witness. See Lone Star Gas, 618 S.W.2d at 125. The agency may accept or reject in whole or in part the testimony of the various witnesses who testify. See id. It follows, then, that substantial evidence will support a method of calculating consolidated tax savings that is an amalgam of methods proposed by different witnesses. However, the method ultimately employed must have some support in the record. While agency expertise may be utilized in evaluating the evidence in the record, that expertise cannot be a substitute for proof. See id. at 124-25.

The record as cited by the parties here does not support the Commission’s decision to attempt to account for the time-value of past tax savings. The Commission looked at fifteen years of tax savings and treated that total as free capi tal that CPL would have had to pay interest on each year if it had borrowed that amount. The Commission then treated the utility’s consolidatéd tax savings as if they represented a loan on which CPL did not have to pay interest and deducted that “unpaid interest” from CPL’s cost of service. The specific choices to use fifteen years’ worth of data and to reflect the savings in CPL’s cost of service are each supported by the record. 7 However, the witnesses who testified regarding the cost-of-service approach calculated consolidated tax savings by averaging previous years’ savings to project the annual tax savings that CPL could anticipate in the future. None of those witnesses proposed using today’s value of past savings to account for the earlier savings. In generating this calculation, the Commission embarked on its own method that goes beyond anything in the record. While the Commission does have broad discretion to determine a utility’s “fair share” of a reduction in its federal income tax liability that resulted from the filing of a consolidated tax return, its determination still must be supported by evidence in the record. See GTE-SW, 901 S.W.2d at 410; Moran Util., 728 S.W.2d at 766; Lone Star Gas, 618 S.W.2d at 124. The Commission’s issue on appeal is overruled.

Issues raised by CPL

CPL’s Rate of Return on Invested Capital

In its first issue, CPL complains that the Commission set an unreasonably low rate of return on invested capital that is not supported by substantial evidence. The rates a utility charges its customers must be set so as to produce revenues equal to the sum of two amounts: (1) the utility’s “reasonable and necessary operating expenses,” and (2) “a reasonable return on the utility’s invested capital used and useful in providing service to the public.” PURA § 36.051; see Cities for Fair Util. Rates, 924 S.W.2d at 935. Thus, a utility’s rates should be set at a level “sufficient to repay expenses, without a return or profit on those expenses, and to provide a return on the invested capital included in the rate base, without repaying that investment.” Cities for Fair Util. Rates, 924 S.W.2d at 935.

In this rate-making proceeding, the Commission initially set the rate of return on CPL’s invested capital at 10.9%. However, the Commission ordered a lower rate of return on the equity portion of certain of CPL’s assets that were deemed “excess cost over market” (ECOM). ECOM is the amount by which the depreciated original cost of a utility’s generation assets exceeds the potential competitive market values of those assets; these are also known as “stranded,” unrecoverable costs. The Commission determined that CPL had $800 million of ECOM. Based on CPL’s capital structure, the equity portion of these assets amounted to $363 million. The Commission allowed only a 7.96% rate of return on that portion of CPL’s invested capital.

Most of CPL’s ECOM is apparently attributable to its investment in the South Texas Project (STP), a nuclear power plant in which CPL owns a 25.2% interest. The Commission allowed CPL to accelerate depreciation of that investment from a 32-year life to a 20-year remaining depre-ciable life. CPL requested this accelera tion because it anticipated deregulation of the electric industry and feared that, if it were unable to recover now some of the costs invested in constructing STP, those costs would become stranded when CPL was unable to raise rates in a competitive market. Depreciation is an expense that CPL is allowed to recover in its rates, and by accelerating the depreciable life of STP from 32 years to 20 years, the Commission effectively allowed CPL to recover an additional $15 million per year in rates and at the same time protected that amount of its investment from possibly becoming stranded. Because the Commission concluded that allowing accelerated depreciation for STP reduced CPL’s risk of under-recovery of that investment, it decided to apply a lower rate of return to that portion of CPL’s ECOM. 8 As an additional reason for reducing the rate of return on ECOM, the Commission also explained that ECOM is “economically less useful than CPL’s other invested capital.”

Taken together, the 10.9% general rate of return and the 7.96% rate on the equity portion of CPL’s ECOM resulted in an overall rate of return on invested capital of 10.09%. CPL complains that the Commission reduced the rate of return on ECOM based on irrelevant and erroneous assumptions about the usefulness of CPL’s STP investment and the reduction of risk in a hypothetical, future marketplace. We need not consider these complaints, however, if the overall rate of return has sufficient support in the record. It is well settled that a reviewing court is not bound by the reasons given by an agency in its order, provided there is a valid basis for the action taken by the agency. See, e.g., Texas Health Facilities Comm’n v. Charter Medicalr-Dallas, Inc., 665 S.W.2d 446, 452 (Tex.1984); Railroad Comm’n v. City of Austin, 524 S.W.2d 262, 279 (Tex.1975); Public Util. Comm’n v. Southwestern Bell Tel. Co., 960 S.W.2d 116, 121 (Tex.App.— Austin 1997, no pet.). Thus, the relevant question here is whether the overall 10.09% return on capital allowed by the Commission is supported by substantial evidence. This Court has held that when the Commission’s figure for cost of capital falls within the range of relevant evidence in the record, the finding is supported by substantial evidence. See Texas Util. Elec. Co. v. Public Util. Comm’n, 881 S.W.2d 387, 417 (Tex.App.— Austin 1994), rev’d in part on other grounds, 935 S.W.2d 109 (Tex.1996); City of Alvin, 876 S.W.2d at 366; Public Util. Comm’n v. GTE-SW, Inc., 833 S.W.2d 153, 158 (Tex.App. — Austin 1992), affd in part and rev’d in part, 901 S.W.2d 401 (Tex.1995) (cost-of-equity issue not addressed by supreme court).

CPL asserts that the overall 10.09% rate is arbitrary and capricious and not supported by substantial evidence. A court applying the substantial evidence standard may not substitute its judgment for that of the agency. See Mireles, 9 S.W.3d at 131 (citing Charter Medical, 665 S.W.2d at 452); Tex. Gov’t Code Ann. § 2001.174. The issue for the reviewing court is not whether the agency’s decision was correct, but only whether the record demonstrates some reasonable basis for the agency’s action. See id.

CPL claims that no evidence supports the 10.09% rate of return, i.e., that this number falls outside the range of evidence recommended by any witness. We disagree. Four witnesses testified to their recommendations of an appropriate rate of return on invested capital. A close reading of the evidence reveals that the rate adopted by the Commission is within the range of evidence offered by these witnesses. CPL offered the testimony of Samuel C. Hadaway with Financial Analysis Consultants, who estimated an appropriate return on capital for the utility to be 12.25%. In reaching his conclusion, Hada-way discussed and compared several dif ferent models for estimating an appropriate cost of capital and the corresponding rate of return for each model. Using a constant-growth, “discounted cash flow” (DCF) method applied to a comparable group of electric utilities, Hadaway calculated a range of 9.3% to 9.6% return on capital. 9 Using a non-constant-growth (multi-stage) DCF method, he calculated a range of 10.4% to 13.3%. 10 Using a different method — the “risk premium” method — Hadaway came up with a range of 11.7% to 12.3%. 11 Hadaway rejected the constant-growth DCF method as flatly “not valid” because he believes it gives unreliable results where rates are expected to fluctuate or when future growth rates are highly uncertain. Hadaway recommended that “no weight be given to the constant growth DCF model.” He found most rehable a method that combined elements of the non-constant-growth DCF and risk premium methods, and settled on a recommended return on equity of 12.25%.

Texas Industrial Energy Consumers (TIEC), a group of industrial customers of CPL, offered the testimony of economic consultant Michael Gorman. He used the same three basic models introduced by Hadaway to calculate CPL’s cost of equity and to estimate its needed return on equity, but his figures were lower than Hada-way’s. Like Hadaway, Gorman analyzed a comparable group of publicly traded utilities to come up with figures appropriate for CPL. Using the constant-growth DCF model, he calculated a mean rate of return at 9.57%. Using a non-constant-growth DCF model, he calculated a rate of 10.25%. Using a risk-premium model, he calculated a mean rate of 10.9%. Like Hadaway, Gor-man rejected the constant-growth DCF model and recommended a rate of return within the range of 10.25% to 10.9%, pinpointing his recommendation at 10.6%.

Cities enlisted the testimony of Eugene Rasmussen, a consulting analyst of economic and financial problems faced by public utilities. Unlike CPL and TIEC’s witnesses, Rasmussen specifically suggested using a constant-growth DCF method. He testified that “[t]he constant growth DCF method I use has been relied upon by many regulatory bodies. It is the same method Dr. Hadaway and this Commission used for many years.... The conceptual basis for the constant growth DCF is not in question.” He went on to dispute Had-away’s rationale for rejecting the constant-growth DCF method, explaining why it is not necessary to use some form of non-constant-growth DCF method in order to recognize the changes that will be brought by competition. Using the constant-growth DCF method, Rasmussen recommended a 10.625% rate of return.

Finally, OPC offered the testimony of economist Dr. Carol A. Szerszen. She also testified that she preferred the DCF method. Using a single-company estimate with the constant-growth DCF method, she came up with a range of 9.6% to 10.6%.

However, Szerszen did not feel that a company’s cost of equity should be estimated based on a single company. Instead, she said that the financial characteristics of other companies in the same industry should be compared. Using the same group of publicly traded electric utilities Hadaway had used with the constant-growth DCF model, Szerszen said that a comparable-company DCF rate is 10%. She ultimately recommended a 10.5% DCF rate for CPL.

Because most of the witnesses who produced a range of numbers that dipped as low as or lower than the overall rate of 10.09% settled on by the Commission also criticized the constant-growth DCF model that yielded those numbers, CPL claims there is no evidence to support a rate of return that low. CPL cites us to Railroad Commission v. Lone Star Gas Co., 618 S.W.2d 121, 125 (Tex.Civ.App.— Austin 1981, no writ), for the proposition that a witness’s testimony criticizing the DCF method is not evidence that would support the agency’s ultimate decision to adopt a rate of return based on that method. Lone Star Gas is distinguishable, however, because in that case no other witnesses testified in support of the DCF method. See id. at 123-24. Thus, the only evidence the Railroad Commission had before it regarding the DCF method was critical of that method; under those circumstances, this Court held that there was no evidence in the administrative record to support the Railroad Commission’s decision to use that method. See id. at 125. That is not the case here. Although CPL witness Hadaway and TIEC witness Gorman both criticized the reliability and usefulness of the constant-growth DCF formula, Cities’ witness Rasmussen and OPC witness Szerszen each testified in support of that method as it is applied to CPL. Hence, there is substantial evidence in the record to support the Commission’s decision to consider figures generated using the constant-growth DCF method.

Having determined that the constant-growth DCF method was valid, the Commission was free to use the figures generated by any witness using that method. The agency is the sole judge of the weight to be accorded the testimony of each witness. See Public Util. Comm’n v. Houston Lighting & Power Co., 715 S.W.2d 98, 109 (TexApp. — Austin 1986), rev’d in part on other grounds, 748 S.W.2d 439 (Tex.1987). In weighing evidence, the agency may accept or reject the testimony of witnesses or may accept part of a witness’s testimony and disregard the remainder. See id. (citing Southern Union Gas Co., 692 S.W.2d at 141-42). Therefore, even while rejecting the criticism of the constant-growth DCF method by two witnesses, the agency was free to accept other portions of the testimony of those same witnesses — Hadaway, who calculated a rate of return range going as low as 9.3%, and Gorman, who calculated a range as low as 9.57%. The figures generated by these witnesses, as well as by Szerszen, who calculated a return range as low as 9.6%, constitute substantial evidence supporting the overall rate of 10.09% ultimately adopted by the Commission. CPL’s first issue is overruled.

Challenged Findings of Fact and Conclusions of Law

In its second issue on appeal, CPL asks us to reverse certain findings of fact and conclusions of law contained in the Commission’s decision. In those findings, the Commission stated that if the legislature decided to deregulate the retail sale of electricity and did not provide for recovery of stranded costs, then utilities such as CPL would have no constitutional right to recover such costs. 12 The stranded costs at issue are the difference between the actual market value of generation plants in a deregulated environment and the remaining balance of original costs after depreciation. That difference is what is referred to as “excess cost over market” or ECOM. CPL complains that whether and how those costs can be recovered are issues that will be addressed by the legislature in the event of deregulation and are not properly questions before the Commission. The Commission’s findings, according to CPL, are merely advisory and have “no relevance to any issue in this case.” Appellees Cities and the Commission agree that the challenged conclusions and findings are immaterial and have no binding legal effect.

Even assuming that CPL is correct that the challenged findings are wrong, the Administrative Procedure Act prevents us from reversing or remanding an agency decision unless “substantial rights of the appellant have been prejudiced.” Tex. Gov’t Code Ann.2001.174 (West 2000). Where, as here, all parties agree that the findings are superfluous to the Commission’s decision and order, we fail to see how those findings could have prejudiced CPL’s rights. In an analogous context, courts generally may disregard immaterial jury findings. See Spencer v. Eagle Star Ins. Co., 876 S.W.2d 154, 157 (Tex.1994) (trial court may disregard jury finding if issue is immaterial); cf. Charter Medical, 665 S.W.2d at 453 (improper and irrelevant agency findings were not considered by court). CPL claims that it lost $970 million in stock equity and that its bonds were downgraded as a result of the Commission’s initial order. It does not cite us to anything in the administrative record to support the existence of such a loss, nor to record evidence that the challenged findings and conclusions were the cause of any such loss. 13 We decline to address such a speculative claim of harm.

To the extent CPL may argue that the challenged findings and conclusions could have precedential impact on future proceedings, we note that “any complaint [CPL] has regarding standards to be applied in future cases is not yet ripe for consideration.” El Paso Elec. Co. v. Public Util. Comm’n, 917 S.W.2d 846, 857 n. 6 (Tex.App.— Austin 1995), judgment withdrawn by agr., 917 S.W.2d 872 (Tex. App. — Austin 1996) (withdrawing judgment but not majority or dissenting opinions). Additionally, it is doubtful that these findings could have any impact on other cases. A challenge to superfluous findings is tantamount to appealing dicta in a court opinion. It is well settled that, as a general rule, dicta has no precedential value. See Boswell v. Pannell, 107 Tex. 433, 180 S.W. 593, 596-97 (1915); Lester v. First Am. Bank, 866 S.W.2d 361, 363 (Tex. App.— Waco 1993, writ denied). It follows that the findings and conclusions at issue here would likewise have no res judicata or collateral estoppel effect in future proceedings. Because no decision is required on this point, CPL’s second issue is overruled.

Stepped-in Rate Decrease

In its third issue on appeal, CPL complains that the Commission exceeded its authority by ordering rate reductions in three consecutive years. The Commission’s order adopted what the Commission termed a “rate-reduction ‘glide-path,’ ” which set rates in three steps. The first step of the rate reduction was effective from the date of the order until May 1998. Then, in May 1998 and again in May 1999, rates were further reduced “to recognize that the accumulation of additional depreciation prior to those dates will reduce the level of [CPL’s] invested capital, thereby reducing its return and income tax expense.” CPL has filed a motion to dismiss this issue from its appeal pursuant to a settlement reached with the Commission. We grant the motion and therefore will not address this issue.

Additions to Test-Year Calculations

In its fourth issue on appeal, CPL argues that the Commission should have included the attendant costs of two new facilities that were placed into service shortly after the end of the test year. A utility seeking a rate increase must present the Commission with revenue and expense data from a historical test year. See City of El Paso, 883 S.W.2d at 188. As noted above, however, “[c]hanges occurring after the test period, if known, may be taken into consideration by the regulatory agency ... in order to make the test-year data as representative as possible of the cost situation that is apt to prevail in the future.” Id. (quoting Suburban Util Corp., 652 S.W.2d at 366).

Nonetheless, the Commission’s authority to allow post-test-year adjustments for “known and measurable changes to historical test-year data” is discretionary, and its own substantive rules permit such changes only where “the attendant impacts on all aspects of a utility’s operations can be with reasonable certainty identified, quantified, and matched.” 16 Tex. Admin. Code § 23.21(b) (1999). The Commission disallowed two requested post-test-year adjustments sought by CPL because it found that the utility failed to account for the attendant impact of those investments on its revenues.

The test year in this proceeding ended June 30, 1995. Shortly thereafter, on August 6, 1995, the East Tie project was put into service. The East Tie project is a high-voltage, direct-current connection between the Electric Reliability Council of Texas and the Southwest Power Pool. CPL sought a post-test-year adjustment of approximately $20 million to include the construction costs for East Tie in the test year. In their proposal for decision, the ALJs noted that East Tie construction costs were prudent and reasonable, yet still recommended that the costs not be added to CPL’s rate base. The Commission agreed, disallowing the adjustment because it found that East Tie enables CPL to serve more customers, and “the impact on CPL’s revenues of that load growth 14 is not taken into account” in the adjustment sought by CPL for its invested capital in East Tie.

CPL asserts that load growth is not an attendant impact of CPL’s investment in East Tie; it claims that the actual effect of East Tie is to provide increased reliability and access to potential fuel savings, and that these purposes have no direct impact on CPL’s operating costs. However, CPL has cited us to nothing in the record to support its contention that there is no causal link between CPL’s load growth and the completion of the East Tie. It offers only the testimony of CPL witness James Bruggeman, who testified that “the East Tie was not built to meet load growth.” CPL’s reason for constructing the asset does not necessarily speak to what impact its construction may have on revenues. The burden was on CPL to account for the attendant impact of its proposed adjustments to test-year numbers, and having failed to quantify the revenue impact of the East Tie (even if the evidence would have shown only that the East Tie did not affect revenues), it has not shown that the Commission’s decision not to allow the adjustment was arbitrary or capricious.

CPL also sought a post-test-year adjustment of approximately $14 million for a new coal-blending facility, completed in October 1995, at Coleto Creek, an existing CPL power plant. The coal-blending facility enables CPL to run Coleto Creek ■with a cheaper coal mix. The Commission disallowed the requested post-test-year adjustment for Coleto Creek because, again, CPL failed to account for the attendant impact of load growth, despite testimony from witnesses for OPC and Cities that showed known load growth had increased since the end of the test year. CPL claims that the Coleto Creek facility was constructed as a cost-savings measure and does not change Coleto Creek’s generation capacity, and therefore the facility cannot have load growth as an effect. However, CPL has cited us to nothing in the record to show what did cause the load growth, and therefore has not conclusively demonstrated that there is no causal link between the load growth and completion of the Coleto Creek facility. Because it had the burden of accounting for the attendant impact of expenses it sought to include in the test year, CPL is not entitled to a reversal of this portion of the Commission’s order. CPL’s fourth issue is overruled.

Affiliate Expenses

In its fifth issue on appeal, CPL argues that the Commission arbitrarily refused to allow CPL to recover $12.6 million in reasonable and necessary affiliate expenses. In setting utility rates, the Commission is required to allow the utility to receive revenues providing a “reasonable return” on the utility’s invested capital and in excess of the utility’s reasonable and necessary operating expenses. See PURA § 36.051. Because of the possibility for self-dealing between affiliated companies, however, expenses paid to an affiliated entity are presumptively not included in the rate base. See id. § 36.058(a); Railroad

Comm’n v. Rio Grande Valley Gas Co., 683 S.W.2d 783, 786 (Tex.App.— Austin 1984, no writ). A utility can overcome this presumption against affiliate expenses only if it demonstrates that its payments are “reasonable and necessary for each item or class of items as determined by the commission.” PURA § 36.058(b). In allowing the expenses, the Commission must make two findings:

(1) a specific finding of the reasonableness and necessity of each item or class of items allowed; and

(2) a finding that the price to the electric utility is not higher than the prices charged by the supplying affiliate to its other affiliates or divisions or to a nonaffiliated person for the same item or class of items.

Id. § 36.058(c).

During the administrative proceedings, Cities and OPC challenged affiliate payments made by CPL to Central and South West Services (CSWS). CPL and CSWS are subsidiaries of CSW. CSW has centralized many of its affiliates’ financial, administrative, engineering, and other services within CSWS, which results in lower overall costs for it and all other CSW subsidiaries. 15 The Commission disallowed sixty-five work orders totaling some $12.6 million in expenses paid to CSWS on the ground that CPL failed to satisfy section 36.058(c)(2); in its findings of fact the Commission found that, for the various categories of charges, CPL failed to prove that the price charged by the affiliate was no higher than the price the supplying affiliate charges other affiliates, divisions, or unaffiliated companies.

The Commission found fault with the allocation factors employed by CSWS in dividing up the costs of its services among the affiliates who benefitted. The disallowed work orders were divided among six different kinds of allocation factors, each of which the Commission found CPL had not proven to fairly distribute costs among CSW affiliates. CPL complains that the Securities and Exchange Commission (SEC) requires that it use the questioned allocation factors, and that the Commission’s rejection of the factors on the basis of section 36.058(c)(2) conflicts with federal law. Services provided by CSWS must be “performed economically and efficiently for the benefit of associated companies at cost, fairly and equitably allocated among” affiliated companies. 15 U.S.C.A. § 79m(b) (West 1997) (Public Utility Holding Company Act, “Contracts by Subsidiary or Mutual Service Companies”). To ensure that costs are “fairly and equitably allocated,” the SEC periodically audits and approves the allocation formulas to be used by the utility to allocate costs among affiliates. During the “test year” for this ratemaking proceeding, the SEC reaffirmed its prior approval of CSWS’s allocation formulas. CPL claims that because it used the SEC-approved allocation formulas, it has done all that is required to satisfy the requirement in PURA section 36.058(c)(2) that “the price to the electric utility is not higher than the prices charged by the supplying affiliate to its other affiliates.” PURA § 36.058(c)(2).

SEC approval of the allocation factors does not preempt Commission review or disallowance. The Public Utility Holding Company Act (PUHCA) is meant to supplement state regulation, not to supplant it. See Alabama Elec. Co-op., Inc. v. Securities & Exch. Comm’n, 353 F.2d 905, 907 (D.C.Cir.1965); see also 15 U.S.C.A. § 79 (West 1997). The PUHCA specifically provides that “[njothing in this chapter shall affect ... the jurisdiction of any other commission, board, agency, or officer ... of any State or political subdivision of any State.” 15 U.S.C.A. § 79u. Further, “in the case of accounts of any company whose methods of accounting are prescribed under the provisions of any law of ... any State, the rules and regulations or orders of the [SEC] ... shall not be inconsistent with the requirements imposed by such law....” Id. § 79t(b). Applicable federal statutes clearly rebut any notion that SEC approval of its allocation factors precludes disallowance by the Commission of charges assigned according to those factors.

The SEC itself reached the same conclusion when, in 1995, CSW sought permission from the SEC to reorganize. The reorganization request also addressed the services provided by CSWS. The Commission and other state utility commissions presented concerns to the SEC, asking the SEC to adopt principles for application of allocation factors it approved. The SEC declined to take on the task of reviewing the application of allocation factors, explaining in its Order Authorizing Restructuring that SEC approval of the restructuring

in no way precludes [any state’s] regulatory authority from scrutinizing and disallowing the pass-through of costs in rates for services rendered to customers of CSWS. It is the Commission’s intention that each state commission having jurisdiction over an Operating Company will retain the right to review and to disallow costs of CSWS services that may be subject to recovery in rates.

Thus, while the SEC requires CPL to use allocation factors it has approved, the Commission retains the right to review the application of those allocation factors to charges submitted by the utility for inclusion in its rate base. The SEC gives only general approval to the allocation factors; how they are employed is squarely within the purview of the Commission.

CPL next urges that even if we do not agree that the SEC’s approval of various allocation factors preempts the Commission’s rejection of the questioned expenses, CPL proved it was entitled to recover those expenses under PURA section 36.058(c). CPL cites us to the proposed decision issued by the ALJs and relies heavily on the ALJs’ recommendation that most of the affiliate expenses paid to CSWS be allowed. CPL seemingly argues that because the ALJs approved the work-order system and concluded that CPL had proved the disputed expenses, the expenses were supported by substantial evidence and the PUC’s holding to the contrary is therefore arbitrary and capricious. However, the ALJs’ recommendations are not binding on the Commission. See Tex. Gov’t Code Ann. § 2003.049(g) (West 2000). Reviewing the same evidence the ALJs had before them, the Commission came to a contrary conclusion on the allowance of $12.6 million in affiliate expenses. We are not allowed to substitute our judgment for that of the Commission. See Tex. Gov’t Code Ann. § 2001.174; Mireles, 9 S.W.3d at 131. We review Commission orders under the substantial evidence rule. See Util.Code § 15.001. The issue for the reviewing court is not whether the agency’s decision was correct, but only whether the record demonstrates some reasonable basis for the agency’s action. See Míreles, 9 S.W.3d at 131 (citing City of El Paso, 883 S.W.2d at 185).

The Commission disallowed sixty-five work orders for CSWS services paid by CPL totaling $12.6 million. Twenty-eight of those work orders (totaling $7.7 million in claimed expenses) were allocated according to formulas based on the number of employees at the CSW affiliates. Cities’ witness Michael L. Arndt, a public utility rate consultant, testified before the Commission that using a subsidiary’s number of employees to determine cost allocations is suspect because companies with no employees whose services are performed by CSWS are allocated no costs whatsoever using these factors, even when they benefit from the services. 16 He further testified that rapid expansion in some of CSW’s non-utility companies is not accounted for in employee-based allocations using the test-year’s numbers, resulting in the effective “subsidization of these non-utility operations until costs are re-examined in future rate proceedings.” He testified that using those allocators results in the under-allocation of costs to several subsidiaries. OPC witness Carol Szerszen, an economist, likewise testified that allocation formulas based on employee numbers are unreliable methods of assuring that each subsidiary pays for services in proportion to the benefits received. The testimony of these witnesses satisfies us that, as to these twenty-eight work orders, the Commission had a reasonable basis for its finding that “CPL failed to meet its burden of showing that its method of allocating these indirect costs among CSW and its affiliates results in a charge for the services to CPL that is no higher than the' price the supplying affiliate charges to other affiliates, divisions, or unaffiliated companies.”

Another twenty-three work orders totaling $3.2 million in expenses employed an allocation factor that automatically assigned 20% of the costs of service to CSWS and divided the remaining 80% among affiliates on the basis of the total assets of the CSW affiliates or the common stock equity of the affiliates. Arndt testified that the method of allocating a flat 20% to CSWS is “an arbitrary allocation methodology which has no demonstrated relation to cost causation.” The 20% factor, he testified, has been used without revision since 1978. Szerszen echoed these assertions. Their testimony provides a reasonable basis for the Commis sion’s finding that the 20% allocation factor is “arbitrary and dated.”

Arndt further testified that basing allocation on the remaining 80% of the total assets of the affiliates results in assessing no costs against some of the affiliates, even though they benefit from the services reflected in the work order charges (e.g., public relations, financial reporting, government monitoring, and legal analysis). Szerszen stated that asset size is not necessarily indicative of the time and resources expended by CSWS in administering subsidiary activities. Szerszen also criticized basing allocation on common equity amounts, testifying that such methods do not reliably measure the benefits received from allocated services. The testimony of these witnesses provides a reasonable basis for the Commission’s decision to disallow the $3.2 million in expenses that were allocated according to the total-asset or common-stock-equity methods.

The PUC disallowed another fourteen work orders totaling $1.7 million in expenses charged to CPL by CSWS. Those work orders allocated costs using four different allocation methods. First was a method that allocated costs to CSW subsidiaries based on the total assets of the subsidiaries. 17 Arndt noted that this method “overallocates charges to utility operations which are capital intensive and allocates no charges to non-utility operations .... ” The Commission disallowed charges in five work orders because it found that “CPL used this factor to allocate costs that are not reasonably related to the assets of the companies to which costs are allocated.” Arndt criticized another allocation factor, based on the average of the utilities’ peak load, energy sales, and number of customers, because it failed to allocate any costs to non-utility subsidiaries, even though they benefitted from the services reflected in the challenged work orders (i.e., integrated long-range plan support, production services, and other costs). The Commission disallowed the expenses in seven work orders using this allocation factor because it did not believe the allocation was reasonably related to the level of service provided.

An allocation method based on peak average load was likewise criticized by Arndt for failing to assess costs against CSW’s non-utility subsidiaries. The Commission accordingly disallowed $345,466 in expenses because the allocation was not reasonably related to the level of services provided. Finally, Arndt challenged an allocation method based on CSWS’s total billing for use in allocating charges for investment-related costs, including annual report and meeting, planning and analysis, and dividend reimbursement. Arndt asserted that this method undercharged non-utility subsidiaries while overstating costs to utility operations. The Commission agreed that CPL failed to show that this allocation factor results in an allocation reasonably related to the level of service required. Regarding all of the above-mentioned allocation factors, Szerszen generally criticized such size-based allocation factors — those based on factors such as subsidiary asset size, common equity amounts, employee numbers, and CSWS’s total billings — for underalloeating costs to smaller, non-utility subsidiaries. We conclude that the testimony of witnesses Arndt and Szerszen provides a reasonable basis for the Commission’s decision to disallow the disputed charges for failing to satisfy the requirement of PURA section 36.058(c)(2); there is substantial evidence to support the Commission’s finding that CPL had failed to prove that the price charged to CPL by CSWS “is not higher than the prices charged by the supplying affiliate to its other affiliates or divisions or to a nonaffiliated person for the same item or class of items.” PURA § 36.058(c)(2). 18

Finally, CPL claims that even if we accept that there are flaws in the allocation factors used as they are applied to the disputed charges, the Commission’s decision to allow none of the charges is clearly an arbitrary and unreasonable result. Under substantial evidence review, we must reverse the Commission’s order if it is “arbitrary or capricious or characterized by abuse of discretion or clearly unwarranted exercise of discretion.” Tex. Gov’t Code Ann. § 2001.174(2)(F). CPL seems to argue that because it showed that it incurred at least some costs, as evidenced by the work orders it paid to CSWS, allowing it to recover none of those costs is arbitrary and unreasonable. In so arguing, CPL misconstrues its burden in presenting those charges to the Commission. The statutory presumption is that payments made to affiliates are not allowed. See PURA § 36.058(a). Because the Commission found that CPL did not overcome this presumption, CPL was simply not entitled to include those costs in its rate base. In rejecting the allocation factors used by CSW in assessing costs against CPL, the Commission was not required to substitute its own allocation method to enable CPL to satisfy its statutory burden and include some of the costs. We overrule CPL’s fifth and final issue.

Issues Raised by Cities

Factoring Expenses

In addition to their complaint that the trial court erroneously found the Commission’s Order constituted retroactive rate-making, Cities also raise two independent issues on appeal against the Commission and CPL. They complain in issue two that CPL should not have been allowed to include in its cost of service certain “factoring expenses” paid by CPL to CSW Credit, Inc. and that, by allowing these expenses, the Commission violated the affiliate transaction statute. They ask us to reverse the trial court and remand to the Commission with instructions to modify the Order.

“Factoring” is the selling of accounts receivable to a third party at a discount. The factoring expense is the cost of the discount. CPL factors its accounts receivable to CSW Credit, Inc., an affiliated company, and then receives cash from CSW Credit, Inc. within a few days. This practice allows CPL to generate working capital without having to wait to receive cash from customers several weeks after billing. CPL asked the Commission to allow $10,204,374 as a cost of service for factoring. The Commission adopted the ALJs’ suggestion to disallow approximately $900,000 of the cost, resulting in an allowed factoring expense of $9,035,440. Cities complain on appeal that none of the factoring expenses should have been included as a cost of service because the statutorily required findings were not made.

As noted above, because of the possibility for self-dealing between affiliated companies, PURA regards affiliate transactions with some skepticism and requires the utility to meet a high burden of proof before costs paid to affiliated companies may be allowed as an expense. See PURA § 36.058; Rio Grande Valley Gas Co., 683 S.W.2d at 786. Section 36.058 presumes that such costs are not allowed, but provides in Subsection (b) that the regulatory authority may allow the costs “only to the extent that the regulatory authority finds the payment is reasonable and necessary for each item or class of items as determined by the commission.” PURA § 36.058(b). The statute goes on to specify that:

a finding under Subsection (b) must include:

(1) a specific finding of the reasonableness and necessity of each item or class of items allowed; and

(2) a finding that the price to the electric utility is not higher than the prices charged by the supplying affiliate to its other affiliates or divisions or to a nonaffiliated person for the same item or class of items.

PURA § 36.058(c) (emphasis added). Cities complain that the Commission’s Order does not contain these required findings and that CPL was therefore not entitled to include these expenses in its cost of service.

CPL argues first that Cities failed to raise this complaint in their third motion for rehearing with sufficient specificity to preserve the error. In its motion for rehearing, a complaining party must direct the regulatory body to (1) the particular finding of fact, conclusion of law, ruling, or other action by. the agency that the complaining party asserts was error; and (2) the legal basis upon which the claim of error rests. See Burke v. Central Educ. Agency, 725 S.W.2d 393, 397 (Tex. App.— Austin 1987, writ refd n.r.e.). In their third motion 'for rehearing, Cities stated, “Cities request rehearing on the Commission’s decision to include in cost of service factoring expense ... because the Commission failed to apply the affiliate transaction standard and failed to make the findings of fact required by PURA § 2.208(b).” 19 A few paragraphs later, the motion reads: ’

The Commission erred in failing to apply § 2.208(b) of PURA 95. Under the affiliate transaction statute, payment to CSW Credit, Inc., “... may not be allowed either as capital cost or as expense except to the extent that the regulatory authority shall find such payment to be reasonable and necessary ...” The Commission failed to make the “specific finding” required by § 2.208(b).

CPL seems to argue that Cities’ motion for rehearing may have preserved the right to complain that the Commission failed to make a “specific finding of the reasonableness and necessity of each item,” as required by section 36.058(c)(1), 20 but that the motion is silent as to the required finding of fact under (c)(2). CPL therefore contends Cities cannot complain on appeal that the Commission failed to find that the factoring costs were not “higher than the prices charged by the supplying affiliate to its other affiliates or divisions or to a nonaffiliated person for the same item.” PURA § 35.058(c)(1), (2). We find this assertion without merit.

Cities’ motion for rehearing tracks the language used in the statute, which allows the agency to include an affiliate expense “only to the extent that the regulatory authority finds the payment is reasonable and necessary.” Id. § 36.058(b). Subsection (c) states that a determination of reasonableness and necessity under subsection (b) must include specific findings both that (1) each item is reasonable and necessary, and (2) the prices charged by the affiliate are in line with prices charged to the affiliate’s other customers. We hold that Cities’ motion for rehearing, which complained that the Order did not satisfy section 36.058(b), when considered in conjunction with the structure of the statute, was sufficient to notify the Commission that it needed to make both specific findings required under section 36.058(c). That is, Cities sufficiently notified the Commission of both the nature of and legal basis for the error. 21

CPL and the Commission next claim that the Commission’s order did contain the findings necessary to satisfy the requirements of section 36.058(c). The Commission’s order included the following two findings of fact:

52. CPL’s necessary and reasonable factoring related cost of service is shown in Schedule II adopted by the Commission as a part of this Order.

[[Image here]]

71. The costs of CPL’s transactions with CSW Corporation, CSWS, Transok, WTU, and SWEPCO during the test year that were reasonable, necessary, not above approximate cost, and not higher than what the affiliate charged others for the same service are shown in Schedule II adopted by the Commission as a part of this Order.

Schedule II includes $9,035,440 in factoring costs. CPL and the Commission assert that, taken together, these two findings of fact show that the Commission specifically approved the factoring costs and made the required findings with regard to all of the expenses listed in Schedule II. We disagree that the findings of fact can be read so broadly.

Section 36.058(c) unequivocally states that the Commission’s approval of payment to an affiliate must include two specific findings. Finding of fact 52 satisfies the section 36.058(c)(1) requirement of “a specific finding of the reasonableness and necessity of each item or class of items allowed.” Finding of fact 71 does not, however, satisfy the section 36.058(c)(2) requirement of “a finding that the price to the electric utility is not higher than the prices charged by the supplying affiliate to its other affiliates or divisions or to a nonaffiliated person for same item or class of items.” Finding of fact 71 pertains to transactions with several enumerated affiliate entities, but CSW Credit, Inc. is not included in the list. The Commission and CPL suggest that the finding was implicitly made with regard to the factoring costs as well by the incorporation of the costs listed in schedule II; however, any such implicit finding does not satisfy the statutory requirement of a specific finding, especially in light of the fact that the finding expressly listed certain entities, but did not include CSW Credit, Inc. We agree with Cities that the Commission failed to make both findings required by the affiliate transaction statute, and that the findings in the order do not support the Commission’s decision to allow the factoring costs. We therefore sustain Cities’ second issue.

Expenses Related to Promotion of Electricity Consumption

In their-final issue on appeal, Cities complain that the Commission erroneously included in CPL’s rates certain ex penses associated with the promotion of increased consumption of electricity, thereby violating the Commission’s own rules. 22 The Commission and CPL counter that the expenses stemmed from Demand-Side Management (DSM) programs designed to promote the efficient use of electricity, and that the expenses were properly allowed under Tex. Util.Code Ann. §§ 34.171, .172 (West 1998). 23

After Cities challenged the expenses, CPL witnesses Sam Schertz, Director of Marketing for CPL, and Neil Felber, CPL’s Director of Regulatory and Accounting Support, testified in rebuttal. They stated that, due to a bookkeeping error in CPL’s test-year accounts, approximately $3 million in DSM expenses were erroneously attributed to an account for expenses related to the promotion of electricity consumption. The error was discovered, and in its rate filing package CPL included an adjustment to its test-year accounts transferring those expenses to a separate account related to encouraging customers to use electricity safely and efficiently. Notwithstanding the initial allocation to the wrong account, the actual nature of the charges supports their allowance as a DSM expense. After considering the evidence, the Commission ultimately approved $2.3 million demand-side management expenses and a $6.4 million rate-base amount, amortized over ten years. We hold that the testimony of witnesses Felber and Schertz constituted substantial evidence that the expenses at issue were properly allowed as DSM expenses, and we overrule Cities’ third issue.

CONCLUSION

The Commission, Cities, and OPC joined to appeal the trial court’s judgment reversing and remanding the portion of the judgment involving consolidated tax savings. While we disagree with the trial court’s reasoning insofar as it concluded that the Commission’s determination of consolidated tax savings constituted retroactive rate-making, we will affirm the trial court’s judgment reversing and remanding to the agency that portion of the Commission’s order because we hold that the method used by the Commission in those calculations is unsupported by substantial evidence in the record.

As to the issues raised by CPL, we hold that (1) substantial evidence supports the 10.09% overall rate of return on invested capital established by the Commission’s order; (2) the harmless error rule prevents us from overturning the challenged statements in the Commission’s order questioning CPL’s right to recover certain costs in the event of deregulation; (3) CPL failed to quantify the revenue impact of the East Tie or Coleto Creek facilities and so has not shown that the Commission’s decision not to allow inclusion of those costs in the test year was arbitrary or capricious; and (4) CPL did not overcome the statutory presumption that payments made to affiliates are not allowed and so was not entitled to include those costs in its rate base. We therefore affirm the portions of the Commission’s order challenged by CPL. We dismiss the portion of CPL’s appeal challenging the ordered “stepped-in” rate decrease, pursuant to CPL’s motion.

In addition to their appeal of the consolidated tax savings issue, Cities also brought two independent issues on appeal. Because we hold that the Commission failed to make the statutorily required findings to support its inclusion of CPL’s factoring costs in its cost of service calculations, we sustain Cities’ complaint on that issue, re verse the portion of the order pertaining to CPL’s factoring costs, and remand to the agency that portion of the cause. Regarding Cities’ final issue, we hold that substantial evidence supports the Commission’s allowance of the challenged expenses and affirm that portion of the trial court’s judgment.

1 . CPL also brought a fifth issue, complaining that the Commission’s order unlawfully orders a "stepped-in” rate decrease. The parties have since reached agreement on that issue, and CPS has filed a "Notice of [CPL's] Withdrawal, Pursuant to Agreement, of One Portion of Its Appeal.” We will treat this "notice” as a motion and dismiss that issue from this appeal.

2 . The provisions of PURA as amended in 1995 were relied on and cited during the Commission proceedings. In 1997, PURA was recodified and moved from the Revised Civil Statutes to the new Utilities Code. See Act of May 8, 1997, 75th Leg., R.S., ch. 166, § l, Title 2, 1997 Tex. Gen. Laws 713, 715 (codified at Tex. UtiLCode Ann. §§ 36.001-.353). The codification did not make substantive changes in PURA. For the sake of convenience, therefore, we will cite to the current act, referring to specific provisions as "PURA §_”

3 . TIEC is a group of CPL’s industrial customers.

4 . The tax code currently allows net operating losses to be carried forward to "each of the 20 taxable years following the taxable year of the loss.” 26 U.S.C.A. § 172(b)(l)(A)(ii) (West Supp.1999).

5 . It is possible that, in spite of inflation’s tendency to devalue losses that are carried forward, an increase in the tax rate could make a loss more valuable in later years than in the year the loss is incurred. This happens when tax-rate increases outpace inflation.

6 . For purposes of analogy, the parties cite us to three other retroactive ratemaking cases. See State v. Public Util. Comm’n, 883 S.W.2d at 199 (allowing deferred accounting treatment of generation plant operating costs incurred after prior rate case does not constitute retroactive ratemaking); City of Alvin, 876 S.W.2d at 362 (adjusting utility expenses by amount of corporate franchise tax refund received during test year is not retroactive ratemaking); Southern Union Gas v. Railroad Comm’n, 701 S.W.2d 277, 279-80 (Tex.App.— Austin 1985, writ ref'd n.r.e.) (spreading investment tax credits received by utility over life of property and applying ratable portion of tax credit to test year is not retroactive ratemaking).

7 . OPC witness Russell Needier recommended using fifteen years of data in calculating consolidated tax savings because that was the maximum allowable time that a company could forward losses to offset taxable income in a later year. Although Needier ultimately recommended using a rate base method, rather than an expense (cost of service) method, to calculate consolidated tax savings, the rationale for deciding to use fifteen years’ worth of data seems to apply to either method and so is supported by substantial evidence.

Cities' witness Michael Arndt advocated using the expense method to reflect consolidated tax savings, and he estimated CPL’s income tax expense in the future by averaging together five years’ worth of tax savings. Though their figures differed, this method is the same as that offered as an alternative proposal by Needier.

8 . A utility’s required return on capital is directly related to the risk perceived by potential investors. The higher the perceived risk of investing in the utility, the higher the rate of return has to be in order to attract the necessary capital.

9 . The DCF méthod, as described by the ALJs, "assumes that stock prices are based on the present, or discounted, value of all future dividends that investors expect to receive." DCF analysis may be done under either a constant-growth or a non-constant-growth method. The constant-growth DCF model assumes that utility earnings will grow at essentially the same rate over time.

10 . The non-constant-growth DCF method assumes different rates of dividend growth at different times. Advocates of this method believe that deregulation will cause utility earnings to grow erratically in the future in a way that cannot be adequately accounted for under the constant-growth method. With either the constant-growth or non-constant-growth method, other variables and assumptions— such as whether to consider figures from only one company or a group of comparable companies, or how many years’ data should be considered — impact the calculations and conclusions.

11 .The risk-premium method is based on the assumption that equity securities are more risky than debt and that in exchange for assuming greater risk, equity investors will require a higher rate of return. Calculations vary according to the length of the term of study.

12 . Specifically, CPL asks us to strike eleven findings of fact and twenty-six conclusions of law, all of which are related to the general conclusion that CPL has no absolute constitutional right to recover ECOM expenses.

13 . To support its claim of substantial harm, CPL cites us only to an appendix to its brief that contains the affidavits of Samuel C. Had-away. The appellate record consists of the clerk’s record and reporter's record, and the parties’ argument in their briefs must be supported by appropriate citation to the record.

See Tex.R.App. P. 34.1, 38.1(g). The parties' briefs are not part of the record on appeal. Cities assert that the Hadaway affidavits were submitted after the close of evidence and have not been subject to cross-examination, but they, too, have failed to offer record cites on this issue.

14 . “Load growth” is an increase in the level of electricity consumption by a utility's customers.

15 . OPC witness Dr. Carol Szerszen testified that CSW has four utility subsidiaries and eight non-utility subsidiaries. Among the non-utility subsidiaries are CSW Credit, Inc., which provides accounts-receivable factoring to other affiliates and non-affiliates; Transok, Inc., which gathers, stores, markets, and transports natural gas to CSW affiliates and non-affiliates; CSW Energy, which was formed to develop co-generation and other non-utility generating projects; CSW Communications, which was formed to create a fiber optics system for internal CSW communications, with reserve capacity to be eventually marketed to third parties; and CSW International, established to make international business investments.

16 . Testimony was received that several CSW subsidiaries had no employees as of October 1995, including CSW Credit, CSW Leasing, Enershop, CSW Communications, and CSW International. Those companies’ services are apparently provided by CSWS and/or CSW Energy.

17 . This is a separate allocation factor from the one discussed above that allocates costs according to the subsidiaries' assets after allo-eating a flat 20% of costs to the parent company.

18 . CPL relies on the direct and rebuttal testimony of its witness Sandra Bennett, assistant controller at CSWS, to defend the application of the challenged allocation factors to the disputed charges. CPL characterizes the testimony of OPC witness Szerszen and Cities' witness Arndt as “conclusory” and "erroneous” and claims that the testimony of Bennett disproves many of their assertions. Even assuming for the sake of argument that we agreed that Bennett's testimony is more convincing, we would still not be allowed to substitute our judgment for that of the Commission. See Tex. Gov’t Code Ann. § 2001.174. In fact, the evidence in the record may preponderate against the decision of the agency and nonetheless amount to substantial evidence. See Charter Medical, 665 S.W.2d at 452 (citing Lewis v. Metropolitan Savings & Loan Ass’n, 550 S.W.2d 11, 13 (Tex. 1977)). Having already determined that substantial evidence supports the Commission’s decision, we need not consider CPL’s assertion that the evidence weighed more heavily in its favor.

19 . Current PURA § 36.058 was previously codified in Vernon’s Annotated Civil Statutes article 1446c-0, section 2.208(b). See Public Utility Regulatory Act of 1995, 74th Leg., R.S., ch. 9, sec. 1, § 2.208(b), 1995 Tex. Gen. Laws 31, 55 (now codified at Tex. Util.Code Ann. § 36.058).

20 . It is undisputed that the Commission made this finding of fact, as will be discussed infra.

21 . We further note that former PURA section 2.208, in effect when the Commission issued its order and Cities filed their motion for rehearing, combined into one paragraph what the current section 36.058 breaks down into several subsections. See Public Utility Regulatory Act of 1995, 74th Leg., R.S., ch. 9, sec. 1, § 2.208(b), 1995 Tex. Gen. Laws 31, 55 (now codified at Tex. Util.Code Ann. § 36.058). We held above that the complaint made — using the language in current section 36.058(b) — is sufficient to reference the specific findings required in the following subsection (c). The error asserted in the motion for rehearing that the Commission had not made the necessary findings pursuant to former section 2.208 was certainly sufficient to point the Commission to the required findings listed in the same paragraph.

22 . The Commission’s Substantive Rules provide: "The following expenses shall never be allowed as a component of cost of service: ... (F) funds promoting increased consumption of electricity or water.” 16 Tex. Admin. Code § 23.21(c)(2)(f) (1999).

23 . Though in effect during the ratemaking proceedings, these sections have since been repealed. See Act of May 21, 1999, 76th Leg., R.S., ch. 405, sec. 58, § 61(2), 1999 Tex. Gen. Laws 2543, 2624.