Business & Professional People for the Public Interest v. Illinois Commerce Commission

JUSTICE CLARK delivered the opinion of the court: This is the second time that the propriety of orders issued by the Illinois Commerce Commission (Commission) regarding the costs of Commonwealth Edison’s (Edison’s) newest nuclear electrical generating facilities and a corresponding increase in rates has been before this court. In December 1989, this court issued an opinion overturning the Commission’s Sixth Interim Order which had granted Edison a two-step rate increase. That rate increase was due primarily to the costs incurred in constructing the Byron Unit 2, Braidwood Unit 1 and Braid-wood Unit 2 nuclear plants (Business & Professional People for the Public Interest v. Illinois Commerce Comm’n (1989), 136 Ill. 2d 192) (Business & Professional People I). On April 12, 1990, Edison filed new tariffs, thereby initiating a new rate case with the Commission. On March 8, 1991, the Commission entered two orders which attempted to comply with the instructions of this court’s opinion in Business & Professional People I (Remand Order) and set rates in accord with Edison’s 1990 tariff filing (Rate Order) respectively. Several intervenors representing various ratepayer groups filed petitions with this court for direct appeal pursuant to Rule 302(b) (134 Ill. 2d R. 302(b)). We allowed the petitions. In this case, the parties are challenging the propriety of several of the Commission’s findings contained within the Remand and Rate Orders. The parties to this appeal are as follows: 12 consumer and governmental groups that we will collectively refer to as the intervenors, the People of the State of Illinois ex rel. Roland W. Burris (Attorney General), Edison, the Commission, and the Illinois Industrial Electrical Consumers. On appeal, the intervenors argue that the Commission has allowed the consumers to bear rate increases for nonprudent plant construction and various unreasonable costs associated with that construction. Further, the intervenors allege that the plants are not used and useful as defined by the Public Utilities Act (the Act) (Ill. Rev. Stat. 1987, ch. 111⅔, par. 1—101 et seq.) and that the Commission violated various ratemaking principles and this court’s mandate in Business & Professional People I by allowing Edison to record and recover deferred charges. The Attorney General argues only that the Commission erred in determining that Edison’s plants were fully used and useful. In its cross-appeal, Edison alleges that the Commission’s orders make disallowances from the rate base which are arbitrary and unsupported by the evidence. Edison contends that the dis-allowances effectively resulted in a confiscation of its property. PROCEEDINGS BEFORE THE COMMISSION On August 21, 1987, Edison filed tariffs with the Commission requesting a $1,414 billion annual increase in rates for electric service. (See Ill. Rev. Stat. 1987, ch. 111⅔, par. 1—101 et seq.) The requested increase, approximately 26.9% over the then-existing rates, was to cover costs associated with bringing Byron Unit 2, Braidwood Unit 1 and Braidwood Unit 2 nuclear electrical generating units into service. (Byron Unit 2 began commercial operation on April 11, 1987. Braidwood Unit 1 and Unit 2 began operating on November 19, 1987, and August 5, 1988, respectively.) The Commission suspended Edison’s proposed tariffs and set this rate case for hearing. In accord with the Act, the Commission ordered that audits be conducted to ascertain the reasonableness of construction costs for the three units. (Ill. Rev. Stat. 1987, ch. 111⅔, par. 9—213.) Arthur Young and Company, now known as Ernst and Young (Ernst & Young), was selected to conduct the audit of Byron Unit 2 and O’Brien-Kreitzberg and Associates (O’Brien-Kreitzberg) was selected to conduct the audit of the Braidwood units. The auditors’ reports for Byron Unit 2 and Braid-wood Unit 1 were completed in 1988. Pending completion of the construction audits, the Commission held evidentiary hearings, beginning January 6, 1988, and extending over several months, on the nonaudit portions of the case. Without the benefit of the audit report for Braidwood Unit 2, the Commission’s staff (staff) subsequently filed a “Motion for the Commission to Defer the Resolution of the Rate Case and to Pursue an Alternative Resolution.” Staff attached to this motion a settlement proposal. The proposal provided for a $235 million rate increase effective January 1, 1989, and an additional $245 million rate increase effective January 1, 1990. The proposal also provided for a moratorium on further rate increases through December 31, 1991. Based on this proposal, on December 30, 1988, the Commission entered an order memorializing the above settlement agreement (Sixth Interim Order). The Commission subsequently amended the order on January 25, 1989, and again on February 8,1989. In Business & Professional People I this court held that the Sixth Interim Order was illegal and void. Specifically, this court found that the Commission exceeded its authority by entering the Order without the agreement of the intervenors; that the Commission decided several issues outside of the context of a traditional rate case; and that the Commission did not have the authority to unilaterally impose retroactive refunds and the rate moratorium. Upon denial of rehearing this court modified the Business & Professional People I decision on May 31, 1990, and ordered Edison to comply with its differential refund offer contained in the Sixth Interim Order. (Business & Professional People I, 136 Ill. 2d at 247.) The cause was remanded to the Commission for reconsideration of the entire rate decision. (Business & Professional People I, 136 Ill. 2d at 248.) In accord with this court’s decision, the Commission entered interim orders providing for a refund to ratepayers and rolling back Edison’s rates to the revenue level last authorized by the Commission. On April 12, 1990, prior to this court’s issuance of our modified ruling in Business & Professional People I, Edison initiated a new rate case by filing revised tariffs with the Commission. The new tariff schedules proposed to increase annual rates for electrical services by 17.7%, or $1,231 billion. The Commission suspended the new tariffs and initiated investigatory proceedings. Various parties were permitted to intervene, and hearings were held intermittently between May 4, 1990, and October 22, 1990. Evidentiary hearings were held between October 22 and November 9, 1990. The Commission’s hearing examiners subsequently filed proposed orders with the Commission and oral arguments were presented to the full Commission on February 11 and February 13, 1991. On March 8, 1991, the Commission entered orders addressing the remanded proceeding and Edison’s April 1990 rate request. In the Remand Order the Commission determined what Edison’s approximate revenue requirement would have been based on that record. Those revenue requirements were not placed in effect but were instead replaced by the revenue requirements determined in the simultaneously issued Rate Order. The Remand Order provided in part that the units were prudently constructed within the meaning of section 9 — 212 of the Act (Ill. Rev. Stat. 1987, ch. 111⅔, par. 9—212) and that each of the units was fully used and useful, premised on an application of pre-1986 law. The Remand Order also provided that completion of the plants was unreasonably delayed by a total of 32.1 months and that unreasonable costs of over $700 million were incurred during the construction of the units. The Commission’s conclusions setting forth these findings and the reasonable costs of the units are as follows: Amt. Spent Unreasonable Unreasonable Reasonable Units by Edison Delay Costs Cost Byron 2 $1,884,250,796 23.4 mos. $296,605,000 $1,587,645,796 Bdwd. 1 3,268,000,000 6.0 mos. 333,883,000 2,934,117,000 Bdwd. 2 1,863,000,000 2.7 mos. 103,271,000 1,759,729,000 Furthermore, based on the reasonable values established for each of these units, in the Remand Order the Commission determined that deferred charges of $1,728,957,300 should be added to Edison’s rate base and recovered over the lives of the units. The Commission’s findings in the Remand Order regarding prudency of construction, used and usefulness of the units, reasonable costs of the units and treatment of deferred charges were subsumed in the simultaneously issued Rate Order. Additionally, the Rate Order allowed Edison to include in its rate base approximately $480 million of unaudited capital additions to electrical generating facilities. After making numerous other adjustments to the rate base and conclusions regarding- Edison’s operating income and expenses, the Commission granted Edison a rate increase of over $750 million. The Commission further determined that it would be fair and reasonable to phase in the revenue increase over three years and to allocate it evenly among all customer classes. THE COMMISSION AND THE COURT The Commission is an administrative agency whose power is derived from the legislature. (Union Electric Co. v. Illinois Commerce Comm’n (1979), 77 Ill. 2d 364.) Pursuant to the Act, the Commission has “general supervision of all public utilities.” (Ill. Rev. Stat. 1989, ch. 111⅔, par. 4—101.) When a utility files a request for a rate increase in the form of a new tariff schedule, the Commission has the authority upon complaint or its own initiative to hear evidence, hold hearings and determine the propriety of the requested increase. (Ill. Rev. Stat. 1989, ch. 111⅔, par. 9—201(b).) The Commission must determine whether the proposed rates are just and reasonable and do so within the regulatory parameters which prohibit retroactive and single issue ratemaking. In establishing the rates that a public utility is permitted to charge its customers, the Commission must first determine the utility’s revenue requirement. The components of the revenue requirement have frequently been expressed in the formula “R (revenue requirement) = C (operating costs) + Ir (invested capital or rate base times rate of return on capital).” (Citizens Utilities Co. v. Illinois Commerce Comm’n (1988), 124 Ill. 2d 195, citing City of Charlottesville v. Federal Energy Regulatory Comm’n (D.C. Cir. 1985), 774 F.2d 1205, 1217.) The Act requires that the Commission make several critical findings regarding a utility’s investments before the costs of a new plant are included in the utility’s rate base. Initially, the Commission must determine that a plant is prudent as well as used and useful in providing utility service to the utility’s customers. (Ill. Rev. Stat. 1987, ch. 111⅔, par. 9—212.) Further, the Commission must determine that the costs of the new plant, or significant additions to an existing plant, are reasonable. (Ill. Rev. Stat. 1987, ch. 111⅔, par. 9—213; Business & Professional People I, 136 Ill. 2d at 201-02.) Throughout the rate proceedings, the utility has- the burden of proving that its investments meet these requirements. It is important to reiterate that setting utility rates is a legislative rather than a judicial function. In the rate-making scheme, the Commission and not the court is the fact-finding body (People ex rel. Hartigan v. Illinois Commerce Comm’n (1987), 117 Ill. 2d 120, 142) (Hartigan I). Its findings of fact are to be accepted as prima facie true and cannot be set aside on appeal unless they are against the manifest weight of the evidence. (City of Chicago v. Illinois Commerce Comm’n (1985), 133 Ill. App. 3d 435, 439.) Accordingly, our review of the Commission’s orders is limited to determining whether the Commission: acted within the scope of its statutory authority; set out findings of fact adequate to support its decisions; issued findings which were supported by the manifest weight of the evidence; and rendered decisions which do not infringe upon a constitutional right. City of Chicago, 133 Ill. App. 3d at 439. See Ill. Rev. Stat. 1989, ch. 111⅔, par. 10—201(e)(iv). PRUDENCY OF CONSTRUCTING THE BYRON AND BRAIDWOOD UNITS In the early 1970s, Edison petitioned the Commission to approve the construction of four nuclear electrical generating units. After holding hearings to determine whether the construction of the units was necessary to meet future demand, the Commission found that additional electrical generating capacity would be needed and approved construction of the units. Certificates of public convenience and necessity for the Byron and Braidwood facilities were issued on July 11, 1973, and March 27, 1974, respectively. See ICC Docket Nos. 57941, 58339. After construction of the units had commenced, the demand for electricity fell and the costs of construction escalated. These factors caused the Commission, on its own accord, to investigate whether completion of the Byron and Braidwood units was in the public’s interest. On October 15, 1980, after conducting numerous hearings and considering briefs submitted by the interveners and staff, the Commission concluded that substantial economic benefit would accrue to the ratepayers and Edison if the Byron and Braidwood units were completed “in as timely and economic [a] manner as possible.” (ICC Docket No. 78—0646.) These findings were made in light of the fact that completion of the plants would significantly increase Edison’s excess reserve margins. Specifically, the Commission found “that a forecasted increase in reserve margins does not indicate that delay in construction is economically advantageous when new generating units are approximately half constructed and their completion and use will displace the use of higher cost fuel.” (ICC Docket No. 78—0646.) The Commission approved continued construction of the Byron and Braid-wood units pursuant to further monitoring. In its order addressing Edison’s 1980-81 rate filing (ICC Docket No. 80— 546) and again in the fllowing year as part of Edison’s 1982 rate case, the Commission reexamined economic and engineering studies for the Byron and Braidwood units. Based on these studies, the Commission again directed Edison to complete these units in a timely and economic manner. (ICC Docket No. 82—0026.) However, in the latter order the Commission invited a consumer group and other interested parties to file a proposal describing conservation measures which may have eliminated the need for the construction of additional generating capacity. This invitation gave rise to a new Commission proceeding. On December 17, 1982, several intervenors filed a petition requesting that the Commission initiate an investigation of energy conservation alternatives for Edison. (ICC Docket No. 82—0855.) Subsequently, one of the intervenors filed a motion seeking to schedule hearings to consider the cancellation of the Braidwood units. Between April 8 and September 30, 1985, approximately 27 hearings were held. The evidence presented at the hearings compared the anticipated cost of the completion of the Braidwood units with the savings due to cancellation of the units. Based on this evidence the Commission determined that the units should not be cancelled. However, in so finding, the Commission stated: “The parties have framed their evidence in terms of revenue requirements analyses. By its nature a revenue requirements analysis assesses the issue of completion [versus] cancellation in terms of assigning specific dollar values to the major issues impacting costs. * * * The Commission emphasizes again that such dollar estimates are by no means absolute and can be considered only in a relative sense. The Commission declines to ascribe a certainty to these dollar estimates which does not exist. Given the imprecision and uncertainties of the long term forecasts and projections which the parties used as the basis for their analyses, such dollar amounts, either in favor of completion or cancellation, cannot be determinative. This is especially true when the far greater total costs of building and operating these plants over their useful lives is considered. The Commission, therefore, cannot reasonably conclude from this evidence that the two Braidwood Units as a single economic entity should be cancelled, especially considering that construction of Unit 1 is more than 90% complete. On the other hand, it is evident that Petitioners have raised substantial questions concerning the value of completing both units of Braidwood. There are a number of uncertainties which impact the decision to cancel or complete the facility as a whole. All of the evidence in this case, however, has been addressed to cancelling or completing all of Braidwood, both Units 1 and 2, and the parties agree that this evidence does not permit separate evaluation of these units. There is sufficient evidence in the record, however, for the Commission to seriously question whether both Braidwood units should be completed. The evidence clearly justifies a decision by the Commission to issue a rule to show cause why Braidwood Unit 2 should not be withdrawn or altered. Specifically, the Commission has found that the benefits to ratepayers of completing Braidwood versus the benefits of cancellation are too close to make a definitive determination as to cancellation/completion on the basis of a difference in revenue requirements alone. Indeed, the relative differences between cancellation and completion benefits are so slight that an error in any of the projections by either party could lead to significant and dramatic cost variations. The Commission cautions the Company that the Order in this docket does not imply approval or disapproval of construction of Braidwood Unit 1.” (Emphasis added.) (ICC Docket No. 82—0855.) The Commission concluded that: “(1) the evidence, while extensive, is nonetheless unpersuasive on the proposition that both units of the Braidwood station should be either cancelled or completed; and (2) a comprehensive evaluation of each unit individually will result in a Commission decision which better serves the ratepayers, the shareholders, and the economy of northern Illinois.” ICC Docket No. 82—0855. The Commission then issued an order requiring Edison to show cause why Braidwood Unit 2 should not be cancelled and the certificate of public convenience and necessity be withdrawn or altered. The Commission opened a new docket, but this docket was later suspended when'the issues were subsumed into this case. ICC Docket No. 86—0249. In the Remand Order the Commission concluded that its prior orders are prima facie evidence of the prudency of constructing the Braidwood units. (Ill. Rev. Stat. 1987, ch. 111⅔, par. 9—212.) In addition to its prior orders, the Commission based its prudency findings on the testimony of Edison’s witness George Rifakes, as well as the auditors’ findings that Edison’s initial commitment to and continued investment in the units was reasonable. The Commission also adopted these prudency findings in the Rate Order. On appeal, the interveners allege that, in reaching its conclusion that construction of the Braidwood units was prudent, the Commission commingled and misapplied the statutory tests for determining the prudency and reasonableness of continued construction. Specifically, the interveners contend that the Commission’s orders do not establish the prudency of continued construction of the Braidwood units after 1980 and that the Commission failed to determine the prudency of continued construction after that date. The prudency provision of section 9—212 of the Act (Ill. Rev. Stat. 1987, ch. 111⅔, par. 9—212) addresses whether a utility’s initial decision to construct a new facility or significant addition and any subsequent reevaluations of those decisions were sound. Prudency is to be determined by the Commission based on evidence and information which was known or should have been known at the time of certification, initiation of construction and on each subsequent evaluation. Ill. Rev. Stat. 1987, ch. 111⅔, par. 9—212. Section 9 — 212 further provides: “If the Commission has issued a certificate of public convenience and necessity for the completed facility and, to the extent that the Commission approves continued construction upon reevaluation subsequent to certification, such actions shall constitute prima facie evidence of the prudency of construction. If the Commission determines as a result of reevaluation during construction that the facility should not be completed, such determination shall constitute prima facie evidence that subsequent construction expenditures were imprudent.” (Emphasis added.) Ill. Rev. Stat. 1987, ch. 111⅔, par. 9—212. Prior to its prudency determination in the Remand Order, the Commission had not specifically approved the continued construction of the Braidwood units since December of 1982. (See ICC Docket No. 82—0026.) When confronted with the issue of completion versus cancellation of the Braidwood units in Docket No. 82—0855, the Commission found the evidence in that record caused it to seriously question whether both Braidwood units should be completed. In that docket the Commission specifically held that its order does not imply approval or disapproval of the continued construction of Braidwood Unit 1. Furthermore, the Commission entered an order requiring Edison to show cause why Braidwood Unit 2 should not be cancelled. Edison argues that because the Commission’s order in Docket No. 82—0855 did not cancel Braidwood, the Commission’s prior orders which required it to complete the units remained in effect. The Commission argues that none of the orders dealing with any of the units found that construction should not be completed. Edison’s and the Commission’s arguments ignore the specific language of the latter order. The Remand Order also ignores the specific findings of the 1986 orders (ICC Docket Nos. 82—0855, 86—0249) and makes only a passing reference to them. Instead, the Commission uses its prior orders to substantiate its conclusion that “Issuance of the certificate and the final orders of the Commission ordering completion established a prima facie case.” We believe the Commission’s reliance on orders which predate the order to show cause is insufficient to establish that the Braidwood plants were prudent. Because the Remand Order fails to set forth the conclusions of the 1986 orders, the evidence Edison presented, if any, to comply with the Commission’s rule to show cause, or an analysis of the facts and findings sufficient to permit an informed judicial review, we remand for a proper determination of the prudency of the continued construction of the Braidwood units. Ill. Rev. Stat. 1987, ch. 111⅔, par. 10—201(e)(iii). We note that the standards for prudency and reasonableness set forth in the Act are separate and distinct. The prudency test set forth in section 9—212 concerns only a utility’s decision to construct and continue to build a new electric utility generating plant, gas production facility or significant additions thereto. (Ill. Rev. Stat. 1987, ch. 111⅔, par. 9—212.) Conversely, section 9—213, which defines the reasonableness standard, addresses only whether the costs of constructing plants found to be prudent are in fact reasonable. (Ill. Rev. Stat. 1987, ch. 111⅔, par. 9—213.) The interveners contend that the Commission’s prudency finding is premised largely on evidence of the reasonableness of construetion. Because of our decision above we need not rule on this issue. Nonetheless, because the issue may arise again on remand, we note that the purpose of the reasonableness requirement set forth in section 9—213 focuses on a quantitative analysis of the cost of constructing the plant, not whether the decision to construct was prudent. (See People ex rel. Hartigan v. Illinois Commerce Comm’n (1990), 202 Ill. App. 3d 917 (Hartigan II).) A quantitative analysis of specific cost figures may be helpful in making a prudency determination, but it is only one of various factors that enter into that equation. REASONABLENESS Under the Act, the amount of a utility’s investment in new electrical generating units deemed prudent by the Commission which may be included in the rate base is limited to costs which are proved reasonable. (Ill. Rev. Stat. 1989, ch. 111⅔, par. 9—213.) The interveners assert that the Commission misapprehended the statutory reasonableness standard and related audit requirements and, thus, allowed plant costs into the rate base without requiring Edison to satisfy central protections afforded consumers by the Act. Specifically, the intervenors allege that the Commission failed to: (1) properly apply the statutory reasonableness test; (2) disallow certain “time related indirect costs” as unreasonable; and (3) require that “significant additions” to electrical generating facilities be audited. Edison disagrees with these contentions and on cross-appeal contends that the Commission imposed a standard more stringent than the statutory reasonableness test, and without adequate findings or record support failed to apply the appropriate methodology for calculating delay costs. We will consider each of these claims separately. REASONABLENESS TEST The audit of Byron Unit 2, conducted by Ernst & Young, concluded that $180,578,796 of the $1,884,250,796 spent by Edison on the construction of that unit was unreasonable. Further, Ernst & Young recognized seven unreasonable delay periods associated with the switch of steam generator models, regulatory impact, inadequate manning, late acceptance and delay in power ascension. The audit attributed all of the 23.4 months of unreasonable delay to Edison’s inadequate manning of the project during construction. The Commission adopted Ernst & Young’s finding of unreasonable delay but determined that the total unreasonable cost associated with Byron Unit 2 was $296,605 million. The audit of the Braidwood units conducted by O’Brien-Kreitzberg concluded that $146,294 million of the $3,268 billion of costs associated with the construction of Braidwood Unit 1 and $46,044 million of the $1,863 billion of costs associated with construction of Braidwood Unit 2 were unreasonable. O’Brien-Kreitzberg also determined that the Braidwood Unit 1 in-service date was unreasonably delayed by 36 months, while Edison’s unreasonable actions caused the commercial operation date of Braidwood Unit 2 to be delayed by 26.2 months. The Commission determined that of the delays found by O’Brien-Kreitzberg with respect to Braidwood Unit 1, only five months due to Edison’s licensing actions and one month due to project construction and quality management were unreasonable. The Commission also found that Braidwood Unit 2 was unreasonably delayed by only 2.7 months attributable to unreasonable project construction and quality management. The Commission determined that $333,883 million and $103,271 million of the construction costs for Braidwood Units 1 and 2, respectively, were unreasonable. With the exception of the costs of unreasonable delays, the intervenors do not attack the specific reasonableness findings of the Commission. Rather, the intervenors contend that the methodology used by the Commission in determining the reasonableness of the costs for Byron Unit 2 and the Braidwood units was flawed. The Act requires that the Commission conduct a construction audit of all new electrical utility generating plants, and significant additions to existing plants, to ascertain whether the costs associated with those plants are reasonable. (Ill. Rev. Stat. 1987, ch. 111⅔, par. 9—213.) This court has determined that the audit is the primary means by which the Commission is to determine the reasonableness of new plant costs. Hartigan I, 117 Ill. 2d at 133. Reasonableness is defined in section 9—213 of the Act (Ill. Rev. Stat. 1987, ch. 111⅔, par. 9—213). Section 9—213 provides in part: “ ‘Reasonable’, as used in this Section, means that a utility’s decisions, construction, and supervision of construction, underlying the costs of new electric utility generating plants and significant additions to electric utility generating plants resulted in efficient, economical and timely construction. In determining the reasonableness of plant costs, the Commission shall consider the knowledge and circumstances prevailing at the time of each relevant utility decision or action.” (Emphasis added.) Ill. Rev. Stat. 1987, ch. 111⅔, par. 9—213. The intervenors first contend that section 9—213 establishes a two-part test for determining whether plant costs are reasonable. The first prong of the intervenors’ test requires the Commission to consider whether the utility’s actions “resulted in efficient, economical and timely construction,” and the second prong requires a consideration of the “knowledge and circumstances prevailing at the time” the utility’s decisions were made. The intervenors allege that the Commission erred in applying the reasonableness test of section 9 — 213 because the Commission and the auditors only considered the knowledge and circumstances test. The intervenors further argue that as a result of the Commission’s failure to consider the first prong of the test, millions of dollars of unreasonable costs were included in Edison’s ratebase. In interpreting section 9 — 213, the Commission stated the Act “requires only that the cost be reasonable, and that in applying the definition of reasonable, the Commission ‘shall’ consider the knowledge and circumstances prevailing at the time of each relevant decision or action.” The Commission determined that the first sentence of the section states the definition of “reasonable” and the second sentence mandates how that definition is to be applied. Specifically, the Commission found that “the second sentence modifies the first by stating how the reasonableness shall be evaluated. The second sentence is a limitation on how the Commission makes its determination of reasonableness, not a separate test.” The Commission’s interpretation of a statutory standard is entitled to deference. (Business & Professional People for the Public Interest v. Illinois Commerce Comm’n (1988), 171 Ill. App. 3d 948, 956-57 (Business & Professional People II).) Although this court is not bound by the Commission’s interpretation of a statute, we find the Commission’s interpretation of section 9 — 213 persuasive for the following reasons. The parties’ allegations are apparently supported by conflicting rules of statutory construction. The intervenors argue that in construing a statute all statutory language should be given effect. (Business & Professional People I, 132 Ill. 2d at 64.) The intervenors further allege that the Commission’s interpretation renders the “resulted in” language of section 9—213 superfluous. On the other hand, Edison and the Commission contend that the provisions of a statute are to be construed so as to produce a consistent unified whole rather than inconsistent and contradictory parts. (Miller v. Department of Registration & Education (1979), 75 Ill. 2d 76, 81.) These parties argue that the interveners’ two-prong test renders the statutory definition inconsistent by dictating that a utility’s actions must be reasonable under the prevailing circumstances at the time when made as well as in hindsight. When interpreting a statute the primary function of this court is to ascertain and give effect to the intent of the legislature. (People v. Beam (1979), 74 Ill. 2d 240; see MCI Telephone Corp. v. Illinois Commerce Comm’n (1988), 168 Ill. App. 3d 1008.) The language used in the statute is the primary source for determining legislative intent. (People ex rel. Gibson v. Cannon (1976), 65 Ill. 2d 366.) If the language is certain and unambiguous this court need not refer to the legislative history, but must enforce the statute as enacted. Gibson, 65 Ill. 2d 366. In the present case, it is evident that the literal meaning of the words used to define reasonableness in the statute are inconsistent. The first sentence mandates that a hindsight test be used while the second sentence prohibits such a test. The interveners attempt to reconcile the clearly inconsistent standards by alleging that the Act creates a presumption that costs associated with inefficient, uneconomical or untimely construction are unreasonable. Under the interveners’ interpretation, a finding that costs associated with a utility’s new facility are unreasonable may be rebutted by evidence that the costs were reasonable based on the knowledge and circumstances prevailing at the time incurred. The Act fails on its face to make such a presumption and this court will not make hypothetical indulgences into legislative intent. We noted in Hartigan I that “the legislative history of section 30.1 [now section 9—213] suggests that an affirmative showing of the reasonableness of a utility’s construction-related costs is necessary if a sense of confidence in the ratemaking process is to be instilled in those consumers who are required to pay the increased rates resulting from those costs.” (Hartigan I, 117 Ill. 2d at 133.) Section 9—213 did not strengthen a preexisting reasonableness standard as the intervenors contend, but rather changed the Commission’s methodology for determining whether costs incurred by a utility in the construction of a new electrical generating facility were in fact reasonable. Prior to the enactment of section 30.1, the predecessor to section 9—213, the construction costs incurred by a utility were presumed to be reasonable. (City of Chicago v. Illinois Commerce Comm’n (1985), 133 Ill. App. 3d 435, 442-43; see also Hartigan I, 117 Ill. 2d at 120.) This court has previously held that the “audit required by section [9—213] has replaced the presumption of reasonableness” and that “the audit now provides the primary means by which the Commission is to determine the reasonableness of the costs associated with the construction of power plants.” (Hartigan I, 117 Ill. 2d at 133.) Our courts have not, however, squarely addressed the issue presented here today and our review of the legislative history provides us with little guidance. Hartigan II, 202 Ill. App. 3d 917. Where the statutory language is ambiguous and the legislative history is not determinative, this court must attempt to resolve the conflict by reference to the entire statute. The Commission is charged by the legislature with setting rates which are “just and reasonable” not only to the ratepayers but to the utility and its stockholders. (Ill. Rev. Stat. 1987, 111⅔, pars. 1—102, 9—201.) Prior to the enactment of section 9—213 and its predecessor, section 30.1, the use of hindsight in determining the reasonableness of costs to be included in a utility’s rate base was unprecedented. (See Peoples Gas v. Illinois Commerce Comm’n (1939), 373 Ill. 31, 61-62.) Although the addition of section 30.1 shifted the burden of proof to the utility, we cannot conclude, based on a reading of the entire Act, that the legislature intended to place an additional burden on a utility to prove that its actions were reasonable using hindsight. Such a conclusion would be fundamentally unfair. The Commission disallowed over $700 million of costs associated with the construction of Byron Unit 2 and the Braidwood units from inclusion in Edison’s rate base. The Commission made its cost determinations in light of knowledge and prevailing circumstances at the time of each relevant utility decision or action. In making its determinations, the Commission applied the proper definition of reasonable as the term is used in section 9 — 213 of the Act. In its cross-appeal Edison argues that even though it sustained its burden of proving the reasonableness of its constructioii costs under section 9 — 213, the Commission applied a “better than reasonable” standard to Edison’s conduct and disallowed millions of dollars of costs. We disagree. As an example of the alleged Commission error, Edison points to the Commission’s selection of Ernst & Young’s proposed construction schedule used to determine the period of unreasonable delay for Byron Unit 2. The Commission determined that unreasonable delays occurred during the construction of Byron Unit 2 due specifically to inadequate manning of the construction site. In so doing, the Commission made a determination based on the evidence presented by the parties. Under the auditor’s proposed construction schedule, if the site was properly manned, Byron Unit 2 could have been completed in November 1984, 23.4 months prior to its actual completion date. Conversely, according to the manpower analysis performed by Edison’s witness, Byron Unit 2 could not have been completed prior to September of 1985. The parties’ conclusions were based on different analyses of the actual construction of the plant, which analyses involved numerous assumptions. Edison argues that the auditor’s witness stated that the assumptions used by Edison in its analysis of the construction schedule for Byron Unit 2 were within a reasonable range. Therefore, Edison contends the Commission could have found that Edison’s schedule position was reasonable. Further, Edison contends that the Commission’s finding that Edison had not proven the auditor’s proposed schedule was unattainable imposed a more than reasonable standard on Edison. Edison’s argument relates to the Commission’s resolution of a disputed question of fact based on conflicting evidence of proposed construction schedules. Edison argues that it met its burden of proof because it presented evidence to show its schedule position was equally reasonable to that of the auditors. We disagree. The Commission is the trier of fact and, as such, its findings and conclusions on such questions shall be held to be prima facie true. (Ill. Rev. Stat. 1987, ch. 111⅔, par. 10—201(d).) The Commission determined, based on conflicting evidence presented, that Edison’s proposed schedule “suffered from substantial defects” and that the auditor’s schedule “provide[d] the Commission with the most reliable means of determining when Edison reasonably should have completed construction of Byron 2.” This court will not substitute its judgment for that of the Commission where, as here, it has made an evidentiary determination supported by the record. Our review of the record reveals that the Commission properly applied the definition of reasonableness set forth in section 9 — 213 of the Act while considering the knowledge and circumstances that prevailed at the time of the utility’s actions or decisions. QUANTIFICATION METHODOLOGY We next consider whether the Commission employed the proper quantification methodology for determining the cost disallowance due to Edison’s unreasonable delays in completing the plants. As . previously stated, based on the construction audit, the Commission determined that due to Edison’s unreasonable actions compleof Byron Unit 2 was delayed by 23.4 months, Braid-wood Unit 1 by 6 months and Braidwood Unit 2 by 2.7 months. At issue here is the proper method to compute the dollar amount to be disallowed as a result of these unreasonable schedule delays. The Commission adopted a modified form of the method proposed by the intervenors, and found that approximately $463 million dollars of expenses were attributable to delays caused by Edison’s unreasonable actions. Edison argues that the Commission erred in rejecting the method proposed by the auditors and that had the Commission applied the proper method, the disallowance would have been about $214 million. The intervenors allege the Commission should not have modified the method proposed by the intervenors and, therefore, the disallowance should be approximately $659 million. Before addressing the merits of this issue, it will be helpful to have an understanding of the concepts involved. A schedule delay may impact costs in several ways, depending on the nature of each particular expense. A delay may require the utility to incur expenses which could have been completely avoided but for the delay. An example of such an expense is the preoperational warranty extension Edison purchased for its Nuclear Steam Supply System (NSSS) at Byron Unit 2. But for the delay in completion of the plant, this warranty extension would not have been necessary. A second category of expenses would have been incurred at the same level regardless of the delay, but the accounting treatment for such expenses varies due to the unreasonable delay. For example, although the delay does not increase the amount of overhead and property taxes the utility incurs, the delay causes an increase in the rate base because these amounts are capitalized rather than ex-pensed in the period incurred. A third category of expense would have been incurred under a reasonable schedule, but due to the delay the expense is increased. For example, a schedule delay may increase the construction costs due to an increase in the per unit costs of labor and materials, as well as an increase in the quantity of labor and materials needed to complete the project. The impact the delay has on this latter category of expense may be referred to as escalation. In addition, a project delay may also impact system-wide costs such as the costs of fuel and capital. Time-related costs are those costs which are dependent upon the time duration of the project rather than upon any specific activity or the general level of construction activity. Indirect costs are those which are not specifically associated with any particular construction activity, but which affect the overall cost of the plant. An example of indirect costs is overhead charged to a project. There is no dispute that the costs directly associated with Edison’s unreasonable actions must be disallowed along with the related financial carrying costs recorded in the “Allowance for Funds Used During Construction” (AFUDC) account. The controversy is in how to measure the impact the schedule delays, which were caused by those unreasonable actions, had on expenditures which were otherwise reasonable. Due to the speculative nature of the calculation, the parties agree that it is impossible to determine with certainty the amount by which the costs of the projects increased due to unreasonable delays. Nevertheless, the parties presented evidence and argued in support of or opposition to various methods for figuring the costs of the delays. parties presented evidence on two different methodologies for quantifying the cost increases caused by the unreasonable delays. The auditors and Edison presented evidence on the “Present Value Revenue Requirements” (PVRR) method. This method purports to place the ratepayers in the same position they would have been in if there had not been an unreasonable delay. This method defines the cost of the unreasonable delay in terms of its impact on the company’s revenue requirements. Under this method, a disallowance is made only for those unreasonable actions which increased the present value of Edison’s revenue requirement, as opposed to those which increased the nominal cost of the plant. Thus, even though an expense may have been increased due to Edison’s unreasonable action, no disallowance is made unless the present value of Edison’s revenue requirements is also increaséd by the unreasonable delay. Under this method, time value of money concepts may be used to offset increases in the nominal cost of a plant. The auditor’s PVRR method consisted of three steps. In step A, the auditors determined the increase in the nominal cost of the plant due to the unreasonable delay. This was accomplished by removing time-related indirect costs incurred during the delay period, shifting direct costs incurred after the delay back in time to correspond with the “should-have-been-built” schedule, deflating the direct costs using the Handy-Whitman Index (HWI) to account for the effects of escalation, and finally computing a new AFUDC for the recomputed direct costs up until the “should-have-been built” date. In step B of their PVRR analysis, the auditors computed the AFUDC for the reasonable plant costs determined in step A between the “should-have-been” built date and the actual in-service date. Finally, in step C, the auditors determined the consequential costs associated with the unreasonable delay. In step C, the auditors considered the impact the delays had on system-wide costs such as fuel savings, avoided nuclear operating and maintenance expense, and “end effects” which measures the cost savings to ratepayers based on the delay in retiring the plants in the future. Under the PVRR method, no disallowance is made for delays which did not “harm” ratepayers. “Harm” to ratepayers is determined by the present value of Edison's revenue requirement. The intervenors proposed to determine the cost of schedule delay using the end-of-period AFUDC method. Under this approach, all unreasonable direct and time-related indirect costs are removed as well as the AFUDC associated with these costs. Then, rather than attempt to recast the actual cash expenditures, the escalation cost of the schedule delay is measured by the amount of AFUDC accrued between the date the plant should have been completed and the date the plant was actually completed. For example, if completion of the plant was unreasonably delayed by 10 months, then the amount of AFUDC recorded during the last 10 months of construction is deemed to be the escalation cost of the unreasonable schedule delay. Admittedly, this method does not determine the actual cost of delay, but rather uses AFUDC as a proxy for the increases in cost. Based on the evidence presented, the Commission adopted a modified form of the end-of-period AFUDC method proposed by the intervenors. The Commission disallowed unreasonable direct costs and the amount of AFUDC accumulated between the should-have-been in-service date and the actual in-service date. However, the Commission did not make a separate disallowance for time-related indirect costs. Instead, the Commission found that the time-related indirect costs were subsumed by the AFUDC disallowance. On appeal, Edison contends that the Commission erred by not adopting the auditor’s methodology for determining the cost of the unreasonable delay and, therefore, the Commission disallowed too much for the delays. On the other hand, the intervenors argue that because the Commission did not make a separate disallowance for time-related indirect expenses, some unreasonable delay expenses were included in Edison’s rate base. We will address each of these arguments separately. The question of how much construction costs increased due to unreasonable delays is a question of fact. The Commission is in the best position to determine the impact delays had on the cost of the plants. Therefore, we will not disturb the Commission’s findings unless unsupported by evidence in the record. The Commission rejected the PVRR method because it found that as applied by the auditors in this case, the PVRR method is in contravention of section 9 — 213. Specifically, the Commission interpreted section 9 — 213 to require a determination of reasonableness based solely ■ on the utility’s actions, without regard to the effect of the actions on ratepayers. The Commission pointed out that the auditor’s PVRR method makes no disallowance for unreasonable costs that do not “harm” ratepayers. Thus the Commission found that as applied in this case, the PVRR method does not comport with section 9 — 213. In addition, the Commission found that the PVRR method was prone to manipulation because it uses “numerous subjective assumptions employed over an extended period of time.” Edison argues that the Commission was incorrect when it found that the auditor’s PVRR method does not conform to section 9 — 213. Edison argues that the auditors first identified each of Edison’s unreasonable actions and then determined whether each unreasonable action increased the costs of the plant. Edison argues that, consistent with section 9 — 213, under the PVRR method the entire cost increase due to an unreasonable delay was disallowed. Edison also argues that no disallowance was made for those unreasonable actions which did not increase costs because section 9 — 213 does not require such a disallowance. Edison’s argument is based on the contention that under section 9 — 213 the reasonableness of costs must be examined in light of present value of money concepts. Edison argues that while some costs increased due to unreasonable delays, these increases in costs were offset by decreases in the financing costs associated with these expenses. That is, because the expenses were incurred later in time, the corresponding AFUDC is less and, therefore, the aggregate cost to the ratepayer is decreased. According to Edison and the auditors, some of the delays which were caused by Edison’s unreasonable actions actually benefited the ratepayers. As previously stated, section 9 — 213 provides in pertinent part: “ ‘Reasonable’, as used in this Section, means that a utility’s decisions, construction, and supervision of construction, underlying the costs of new electric utility generating plants *** resulted in efficient, economical and timely construction.” (Ill. Rev. Stat. 1987, ch. 111⅔, par. 9—213.) The Commission stated that “[a] reasonableness finding, as defined in [section 9 — 213], does not include a determination of whether unreasonable actions harmed ratepayers.” We agree with the Commission’s interpretation of this statute. As the Commission stated, a reasonableness determination must be based solely on the utility’s actions. Because the PVRR method improperly considers harm to ratepayers, it does not comport with section 9— 213. Therefore, we need not consider Edison’s argument that the Commission erred in finding the PVRR method was prone to manipulation. With respect to the modifications made to the end-of-period AFUDC method, the Commission provided two reasons for not making a separate disallowance for time-related indirect costs. First, the Commission found that a separate disallowance “incorporates the very same type of assumptions and judgment calls which tarnish the PVRR method.” While this reasoning may support a finding that some time-related indirect costs are too speculative to accurately determine, this rationale does not support a finding that all time-related indirect costs are too speculative. At least some of the time-related indirect costs may be determined without making any assumptions. For example, the amount charged to overhead expenses (i.e., administrative and general office expenses) during the unreasonable delay period may be accurately determined based on the historical costs already recorded on Edison’s books. If a project is unreasonably delayed, then the overhead capitalized for that project during the delay period is unreasonable and must not be included in Edison’s rate base. To the extent possible, the Commission should determine those time-related indirect costs which are calculable. The Commission and Edison claim that all unreasonable delay expenses, including time-related indirect costs, have already been disallowed. The Commission’s second rationale for not making a separate disallowance was its finding “that the AFUDC rate is a reasonable approximation of the rate at which costs increased due to delay. Those cost increases included both escalation and time related indirect costs (as the Auditors defined those terms).” Edison and the Commission now argue that making a separate disallowance for time-related indirect costs would have the effect of disallowing those costs twice. These parties rely on the appellate court opinion in People ex rel. Hartigan v. Illinois Commerce Comm’n (1990), 202 Ill. App. 3d 917, for support for this proposition. However, because this is a determination of fact which is specific to this case, we do not believe that the appellate court decision in that case is controlling. The Commission never made a specific finding regarding the amount of escalation or time-related indirect expense. Thus it is unclear how much of the amount disallowed is attributable to each component of delay expense. The intervenors claim that approximately $196 million in time-related indirect expenses were improperly included in the rate base. This number is apparently based on the calculation of time-related indirect costs found in the audit report. Because we are presented with only a summary finding that all unreasonable delay costs are subsumed by AFUDC, we are unable to make an informed judicial review of the Commission’s finding that AFUDC subsumes all the time-related indirect costs. We note once again that Edison has the burden of proving its expenses are reasonable. While we recognize the difficulty involved in measuring these costs, we believe that the amount involved justifies a more specific finding by the Commission. Therefore, with respect to a separate disallowance for time-related indirect costs, we hold that the Commission’s findings of fact are insufficient to allow informed judicial review (see Ill. Rev. Stat. 1987, ch. 111⅔, par. 10—201(e)(iii)), and we remand the cause to the Commission for more specific findings on this issue. A UDIT OF SIGNIFICANT ADDITIONS The intervenors next contend that the Commission failed to enforce the statutory audit requirement to establish the reasonableness of the costs associated with Edison’s forecasted capital additions to its electrical generating plants. The record establishes that Edison’s capital additions forecasted for 1990 and 1991 will cost over $1.5 billion. This amount is spread over hundreds of projects. Almost $480 million of those costs represent capital additions to nuclear and fossil fuel production plants. Twenty-nine projects, each costing over $5 million, comprised over one-half of this amount. The staff and several intervenors proposed adjustments to the cost of Edison’s capital additions prior to including them in the rate base. Although the Commission did not conduct an audit of any of the forecasted capital additions, it found that Edison presented enough evidence to establish the reasonableness of the forecasted additions. The Commission adopted the staff’s proposed adjustment of $22,172 million, adjusted the amount of forecasted capital additions to reflect a decrease in Edison’s proposed escalation rate, and allowed the remainder into the rate base. The intervenors now argue that the Commission should have conducted an audit of the substantial plant additions prior to including the costs of those additions in the rate base. In the proceedings below, the Commission noted that the intervenors first raised the issue in their reply brief. Because the intervenors’ failure to raise the issue in their initial brief deprived all other parties of the opportunity to respond, the Commission struck that portion of the intervenors’ reply brief addressing this issue. Nevertheless, the Commission addressed the merits of this issue in the Rate Order. Initially, we note that the issue regarding the audit of significant capital additions was addressed by several interveners in their opening briefs before the Commission. Therefore, the Commission’s contention that several parties did not have an opportunity to address this issue is inaccurate. Accordingly, we do not consider the issue waived. In the Rate Order, the Commission held that the costs of the forecasted projects were not subject to the audit requirement of section 9—213. The Commission stated: “Of those projects which relate to generating stations, many are not ‘additions’ at all within the meaning of section 9 — 213 ***. Moreover, none of these projects can be considered a ‘significant’ addition, either in physical or in dollar terms, in comparison to the size, complexity and cost of the generating stations to which they relate.” The parties agree that, in addition to requiring an audit of the cost of new electrical generating plants, section 9 — 213 requires that the costs associated with all significant additions to those facilities be audited. The parties disagree over whether the forecasted capital additions are significant and therefore must be audited before the costs are included in the rate base. The legislature did not codify a definition of the term “significant additions” but left the determination of whether an addition is significant and whether an audit of those projects should be conducted to the sound discretion of the Commission. The Commission’s interpretation of the statutory language is to be given deference by Illinois courts. (Business & Professional People II, 171 Ill. App. 3d 948.) However in making its determination in the instant case, the Commission has failed to articulate the standard that it applied. In addition, the Commission did not address whether even the larger projects, the cost of which approach $300 million, should be considered significant additions. In the Rate Order, the Commission makes the blanket statement that some of the capital additions “are not even additions” and that based on physical size and comparative cost evaluations none of these projects can be considered “significant.” The Commission’s conclusions make it difficult for this court to determine which projects the Commission determined were not additions, whether the Commission analyzed the range and purpose of those projects it found to be additions and whether it considered if any of the larger projects were significant. The Commission’s failure to set forth findings or analysis sufficient to allow an informed judicial review thereof necessitates that we remand this cause to the Commission for further clarification of this issue. (Ill. Rev. Stat. 1987, ch. 111⅔, par. 10—201(e)(iii).) Further, reconsideration of the Commission’s decision to include the costs associated with forecasted and still uncompleted additions in Edison’s rate base is necessary. On the basis of the Commission’s findings, we cannot conclude that uncompleted projects which are not deemed to be rate-based construction work in progress within the purview of section 9—214 should be rate based. See Ill. Rev. Stat. 1987, ch. 111⅔, par. 9—214. USED AND USEFUL The intervenors and the Attorney General appeal from the Commission’s finding that Byron Unit 2 and the Braidwood units were fully used and useful for purposes of inclusion in Edison’s rate base. The Act limits a utility to a reasonable return upon the value of its property which is used and useful in rendering service to the public. In 1986 the legislature revised the Act and added sections 9 — 211, 9 — 212, and 9— 215 (Pub. Act 84 — 617, eff. Jan. 1, 1986) to govern the Commission’s determinations of whether a utility’s investment in property is used and useful and should be included in the utility’s rate base. (Ill. Rev. Stat. 1987, ch. 111⅔, pars. 9—211, 9—212, 9—215.) Under the revised Act, like the old, the Commission is to include in a utility’s rate base only the value of a new electrical generating plant which is used and useful in providing service to public utility customers. Ill. Rev. Stat. 1987, ch. 111⅔, pars. 9—211, 9—212. The amended Act requires the utility to prove that its new plant is used and useful. The relevant sections provide in part: “A generation or production facility is used and useful only if, and only to the extent that, it is necessary to meet customer demand or economically beneficial in meeting such demand.” (Ill. Rev. Stat. 1987, ch. 111⅔, par. 9—212.) “The Commission shall have power to consider, on a case by case basis, the status of a utility’s capacity and to determine whether or not such utility’s capacity is in excess of that reasonably necessary to provide adequate and reliable electric service. Excess capacity for purposes of this section shall mean capacity in excess of that reasonably necessary to provide adequate and reliable electric service. Such consideration shall be related to the utility’s historic and projected peak. The Commission is empowered to make appropriate and equitable adjustments to rates for utility service upon a finding of excess capacity. With respect to generating capacity existing or under construction on the effective date of this amendatory Act of 1985, any such determination and adjustment to rates, and any determination as to whether such capacity is used and useful for any purpose under this Act, shall be limited to the determination and adjustment, if any, appropriate under the law in effect prior to such effective date.” Ill. Rev. Stat. 1987, ch. 111⅔, par. 9—215. The parties agree that section 9—215 of the amended Act mandates that pre-1986 used and useful standards apply because the units were under construction prior to the enactment of the amendments. (Ill. Rev. Stat. 1987, ch. 111⅔, par. 9—215; Business & Professional People I, 136 Ill. 2d 192.) Accordingly, our next inquiry is whether the needed and economic benefits tests set forth in section 9—212 of the amended Act are appropriate criteria to employ in determining whether the units are used and useful. Since this is a question of law, the Commission’s findings are not binding on this court. Business & Professional People I, 136 Ill. 2d 192. The Commission staff and the interveners recommended substantial used and useful disallowances for Byron Unit 2 and the Braidwood units, premised upon application of some form of the newly codified needed and economic benefits test. On two occasions, the hearing examiners proposed remand and rate orders which adopted the needed (excess capacity) and economic benefits analysis and concluded that, the units were not fully used and useful. The hearing examiners’ final"proposal recommended finding Byron Unit 2 100%, Braidwood Unit 1 88.12% and Braidwood Unit 2 14.03% used and useful. The examiners’ proposed disallowances were supported by the Commission’s 1989 order granting Illinois Power a rate increase for bringing its Clinton unit online. (See ICC Docket Nos. 84—0055, 87—0695, 88—0256 cons.) In that order, the Commission determined that the Clinton unit was only 27.2% used and useful based on application of the needed and economic benefits tests. Illinois Power challenged the Commission’s used and useful determinations, arguing that the Commission misinterpreted section 9 — 215 and the pre-1986 law. On February 8, 1991, exactly one month before the Commission entered the Remand Order in this case, the appellate court reversed and remanded the Commission’s findings on this issue. (Illinois Power Co. v. Illinois Commerce Comm’n (1991), 208 Ill. App. 3d 779.) In Illinois Power, the court held that the Commission erred as a matter of law when it applied the newly codified needed and economic benefits tests to determine whether Illinois Power’s Clinton nuclear power plant was used and useful for purposes of inclusion in its rate base. The appellate court stated, without citation or reference to the legislative history: “By defining ‘used and useful’ in the amended Act, and then requiring that used and useful determinations for capacity existing or under construction on January 1, 1986, be limited to the determinations and adjustments, if any, appropriate under the law in effect prior to that date, the legislature was expressing its intent to change the method by which such determinations were made. If the legislature had intended the Commission to have the discretion to use the tests set forth in the amended Act, there would have been no need for the last paragraph of section 9 — 215.” (Illinois Power, 208 Ill. App. 3d at 792.) The court concluded that the Commission “should have employed the criteria and test utilized in the pre-1986 proceedings” and directed the Commission not to employ the “needed and economic benefits tests” on remand, but rather to determine the used and usefulness of the Clinton reactor on the basis of “established pre-1986 standards.” Illinois Power, 208 Ill. App. 3d at 792. In the Remand Order, the Commission found that the Illinois Power decision “precluded [the Commission] from applying the needed and economic benefits tests,” in its analysis of whether Byron Unit 2 and the Braid-wood units were used and useful. The Commission held that it was required to determine which standards applied under pre-1986 law and to apply those standards to establish whether the new units are used and useful. The Commission concluded that the evidence presented satisfied the pre-1986 definition of used and useful, including central elements of economic dispatch, and found all three units to be 100% used and useful. On appeal, the interveners contend that the Illinois Power decision misapprehended and misapplied sections 9—212 and 9—215 of the amended Act. Further, the Attorney General alleges that the Commission’s reliance on Illinois Power is misplaced because in amending the Act the legislature did not intend to change the method by which used and useful determinations are made. The Attorney General further argues that prior to 1986 the Commission had utilized needs and/or economic benefits tests as standards to determine the used and usefulness of a utility’s investment. We disagree with the appellate court’s conclusion in Illinois Power that the legislature enacted sections 9— 212 and 9 — 215 to express its intent to change the method by which used and useful determinations are made. We believe this conclusion is contrary to the plain meaning of the amended Act and its corresponding legislative history. The court in Illinois Power concluded that if the legislature had intended to allow the Commission to apply the tests set forth in the amended Act for pre-1986 plants, there would have been no need for the last paragraph of section 9 — 215. Our reading of the statute yields a different result. We believe section 9 — 215 is intended to place limits on the Commission’s discretion only for those plants not already under construction on January 1, 1986. With respect to plants already under construction before that date, section 9 — 215 places no limits on the Commission’s discretion. Rather, the statute permits the Commission to use a wide range of tests, including the needed and economics benefits tests. This interpretation is supported by a review of the legislative history of the amended Act. Specifically, the comments of Senator Netsch, one of the sponsors of the amendment to section 9 — 215, are most insightful. During a discussion of the pertinent sections she stated: “[This amendment] clearly reaffirms the Commission’s existing authority to consider excess capacity and to adjust rates in accordance with its findings on excess capacity; and by writing into the Statute an explicit provision that deals with excess capacity. I think that the General Assembly is now sending to the Commission as well as the utilities but particularly to the Illinois Commerce Commission a clear warning that it cannot continue its inaction on this the most important issue with respect to rates and the entire regulatory process.” (84th Ill. Gen. Assem., Senate Proceedings, June 25, 1985, at 13-14.) It is evident from Senator Netsch’s statements, and those of her various colleagues in both houses of the General Assembly, that the legislature was reaffirming the Commission’s existing authority to consider excess capacity and to make appropriate used and useful adjustments. See 84th Ill. Gen. Assem., Senate Proceedings, June 30, 1985, at 13-14; 84th Ill. Gen. Assem., House Proceedings, June 25, 1985, at 240-42. Prior to the enactment of the amended Act, the legislature did not define or impose any criteria for the Commission to utilize in determining the used and usefulness of new plants. Historically, the Commission has considered a variety of factors including: need (see Candlewick Lake Utilities Co. v. Illinois Commerce Comm’n (1983), 122 Ill. App. 3d 219; Citizens for a Better Environment v. Illinois Commerce Comm’n (1981), 103 Ill. App. 3d 133, 138-39; City of Belleville v. St. Clair County Gas & Electric Co. (1915), ICC Docket No. 2237; Union Electric Co. (1985), ICC Docket No. 85—0006); economic dispatch (see Iowa-Illinois (1983), ICC Docket No. 82—0829; Central Illinois Public Service (1982), ICC Docket No. 82—0039); economic reasonableness (see Central Illinois Public Service Comm’n (1982), ICC Docket No. 82—0039); enhanced system reliability (Northern Illinois Gas Co. (1979), ICC Docket No. 79—0133; Peoples Gas, Light & Coke Co. (1953), ICC Docket Nos. 38244, 40472 cons.; Peoples Gas, Light & Coke Co. (1937), ICC Docket No. 24792); capacity used to serve ratepayers (see State Public Utilities Comm’n ex rel. City of Springfield v. Springfield Gas & Electric Co. (1919), 291 Ill. 209, 232; Candlewick Lake, 122 Ill. App. 3d 219); units are used to carry out the purposes of the utility (see Public Utilities Comm’n ex rel. City of Springfield v. Springfield Gas & Electric Co. (1919), 291 Ill. 209; Citizens for a Better Environment, 103 Ill. App. 3d 133); excess reserve margin (see Northern Illinois Gas Co. (1979), ICC Docket No. 79—0133; Central Illinois Public Utilities Service Co. (1982), ICC Docket No. 82—0039); and providing low fuel costs (see Union Electric Co. (1985), ICC Docket No. 85—0006). It is obvious from the above cases that prior to 1986 the Commission did not follow any definitive standard for determining the used and usefulness of a utility’s investments. Nor do court decisions offer a conclusive standard under pre-1986 law. Therefore, our next inquiry is whether under the law prior to 1986 it was within the Commission’s discretion to employ the needed and economic benefits tests in determining whether a unit was used and useful. The Commission concedes that although the needed and economic benefits tests had been suggested by the interveners in several rate cases, it had not been adopted by the Commission until Illinois Power. (See ICC Docket Nos. 84—0055, 87—0695, 88—0256 cons.) However, prior to 1986, the legislature had never placed any limitation on the Commission’s authority to devise and utilize a used and useful standard. When applying the various standards set forth above, the Commission and reviewing courts have consistently recognized that the Commission had wide latitude to fashion a standard and determine used and useful disallowances. (Candlewick Lake, 122 Ill. App. 3d 219; Union Electric Co. (1985), ICC Docket No. 85—0006; Citizens for a Better Environment, 103 Ill. App. 3d 133.) In Citizens, the court, in discussing the used and useful standard, found “the general legislative purpose of the Act is to afford the Commission wide discretion in its determinations.” In Candlewick, the court provided: “We also credit the Commission with considerable discretion in determining what property is ‘used and useful’ in providing services, and in determining how this relates to the setting of rates which are ‘just and reasonable.’ ” (Candlewick Lake, 122 Ill. App. 3d at 223.) Thus, we believe that under pre-1986 law the Commission had the authority to impose a needed and economic benefits standard. We must now decide whether in making its used and useful findings in this docket the Commission felt that its discretion was curtailed by the appellate court’s Illinois Power decision. In the Remand Order the Commission stated: “The Commission does not dispute that the courts have long recognized that the Commission has discretion in determining the used and usefulness of utility property, in exercising its authority to determine fair rates. This discretion, however, has been limited by the Court’s holding in the recent Illinois Power decision.” The Commission’s conclusion in the Remand Order that the Illinois Power decision governs these proceedings and that it had been directed “not to employ the needed and economic benefits tests” is evidence that the Commission clearly felt its discretion was constrained. Edison and the Commission attempt to downplay the Commission’s reliance on the Illinois Power decision. These parties emphasize that the Commission analyzed specific factual findings concerning the Edison system as well as the operation and economics of the units. Specifically, Edison argues that the Commission’s used and useful determination was based on established pre-1986 standards and that the Commission’s findings should not be disturbed. Edison’s and the Commission’s arguments miss the point. We need not address whether the Commission’s used and useful determinations were premised on a valid pre-1986 standard, whether the Commission properly applied those standards or whether the Commission’s conclusions were supported by the evidence in the record. Because of our conclusion that the Commission’s discretion on this issue was unduly constrained by the appellate court’s decision in Illinois Power, we must remand to the Commission. In utilizing its full discretion on remand, the Commission may or may not choose to adopt the newly codified needed and economic benefits tests. The Commission must be permitted to at least consider these tests. Edison also argues that in the Remand Order the Commission rejects the needed and economic benefits tests, not only because of the Illinois Power decision but because the needed and economic benefits tests are equivalent to a short term reserve margin formula. Edison contends that if this formula is adopted it would give the company no incentive to select capacity alternatives which minimize revenue requirements and cost to customers over time. As set forth above, the Commission’s reliance on Illinois Power is clear. Whether in the exercise of its unconstrained discretion the Commission will adopt a reserve margin analysis is left to its sound judgment. DEFERRED CHARGES The next issue we must address is whether the company may record and recover deferred charges. For purposes of this case, deferred charges are various costs related to Byron Unit 2, Braidwood Unit 1 and Braidwood Unit 2, which costs Edison incurred between the dates the plants were placed in service, and the date of the Rate Order. The Commission granted Edison’s request for an accounting variance relative to these plants. Based on this accounting variance, in the Rate Order the Commission permitted Edison to recover $630,681 million in deferred depreciation, $2,010,188 million of post-in-service carrying charges and $29,938 million in deferred decommissioning expenses. After adjusting for deferred income taxes, the Commission’s order allows Edison to record and recover a net total of $1,728,959 million in deferred charges. Under normal accounting procedures, during construction of a power plant, Edison records the direct costs of construction in the “Construction Work In Progress” (CWIP) account. In addition, the financial carrying costs of the plant are accumulated in the “Allowance for Funds Used During Construction” (AFUDC) account. AFUDC consists of interest on borrowed funds used for the construction of the plant plus a reasonable return on equity funds used for construction. AFUDC compensates Edison for the time value of its money, and recognizes the fact that Edison does not recover the cost of construction or receive a return on its investment until the plant is included in the company’s rate base. Under normal accounting procedures, at the time a plant is placed in service, the company stops recording costs in these accounts, and begins to amortize the accounts over the life of the plant. (See 83 Ill. Adm. Code §415.10 et seq. (1985) (“Uniform System of Accounts for Electric Utilities”).) The company recovers the costs associated with CWIP and AFUDC by filing new tariffs to take effect at the time the plant is placed in service. The company will then receive a return of its investment through increased depreciation expense, and a return on its investment through the higher rate base on which its rate of return is based. In the present case, the Rate Order was not synchronized with the in-service dates of the new plants. Byron Unit 2 was placed in service April 11, 1987, Braidwood Unit 1 November 19, 1987 and Braidwood Unit 2 August 5, 1988, respectively. Under normal accounting procedures, the company must stop recording AFUDC and begin to depreciate the plants on these dates. However the company’s investment in the plants was not recognized in a rate order until March 8, 1991. Due to the timing difference between the in-service dates and the effective date of the Rate Order, under normal accounting procedures the company would not be permitted to recover that portion of its investment in the plants which is depreciated during the interim period. Further, because the plants were not included in the rate base until the Rate Order was entered and because Edison would not be permitted to record carrying charges after the plant is placed in service, the company would not have received a return on its investment during this period. To avoid the potential adverse financial impact caused by this regulatory delay, on April 8, 1987, Edison filed with the Commission a petition for an accounting variance in Docket No. 87—0169. In this petition, Edison requested permission to postpone the start of depreciation for Byron Unit 2 and Braidwood Unit 1 as well as permission to record financial carrying charges for the period between the in-service date and the rate recognition date. This petition was later consolidated with Edison’s rate case filed in Docket No. 87—0427. Due to the above described procedural history involving Docket No. 87—0427, the Commission did not rule on Edison’s petition in Docket 87—0169 until the Remand Order. In that order the Commission permitted Edison to record deferred charges for Byron Unit 2 beginning January 1, 1988, and for Braidwood Unit 1 beginning April 25,1988. Edison filed a similar petition with respect to Braid-wood Unit-2 on August 15, 1988 in Docket No. 88—0253. On June 1, 1989, the Commission permitted Edison to record deferred charges for Braidwood Unit 2 beginning August 25, 1988. This decision was affirmed by the appellate court. (See Business & Professional People for the Public Interest v. Illinois Commerce Comm’n (1990), 205 Ill. App. 3d 891.) Because the interveners did not challenge that appellate court decision directly, they may not challenge that decision here. Therefore we will assume that the decision to allow the accounting variance for Braidwood Unit' 2 for recording purposes was correct. In the present case, our review is limited to whether Edison may record deferred charges for Byron Unit 2 and Braidwood Unit 1, and to whether Edison may recover deferred charges for any of the three units. In addition, we must consider Edison’s claim that the Commission arbitrarily shortened the deferral period. RECORDING OF DEFERRED CHARGES We first consider whether the Commission erred in granting Edison’s request for the accounting variance for recording purposes with respect to Byron Unit 2 and Braidwood Unit 1. The Commission first considered this issue at the time Byron Unit 1 was placed in service. In Docket No. 85—0092, the Commission permitted Edison to postpone the commencement of depreciation for Byron Unit 1 until the cost of the plant was recognized in a rate order and to capitalize post-in-service carrying charges during the interim period. Under the test established in Docket No. 85—0092, in order to qualify for the accounting variance, Edison must show that (1) circumstances beyond its control have created a significant regulatory lag between the in-service date and the date of the Rate Order, and that (2) denial of the accounting variance could significantly and adversely affect the company’s earnings, as well as its short-term and long-term cost of capital. (ICC Docket No. 85—0092; Re Commonwealth Edison Co. (Ill. Com. Comm’n Oct. 3, 1985), 70 Pub. Util. Rep. 4th 107, 114.) The intervenors allege that Edison has not met either prong of the two-part test established in Docket 85—0092. Initially, the intervenors contend that Edison could have avoided the delay in this case. The intervenors contend that the delays were caused by Edison’s pursuit of nontraditional ratemaking, including the settlement offer contained in the Sixth Interim Order which this court invalidated. The intervenors contend that Edison could have filed a traditional rate case in 1986, and that had Edison done so the construction audits would have been fully litigated much sooner. In addition, the intervenors contend that Edison could have obtained interim rate relief under section 9—202 of the Act (Ill. Rev. Stat. 1987, ch. 111⅔, par. 9—202) or included the plants in rate base as CWIP under section 9—214(e) of the Act (Ill. Rev. Stat. 1987, ch. 111⅔, par. 9—214(e)). The question of who is to blame for the delay period is one of fact. The Commission made the following findings on this issue: “[T]he Commission concludes that, notwithstanding the reversal and remand of the Sixth Interim Order, the significant regulatory lag between the in-service dates of these three plants and the associated rate treatment are due to circumstances beyond the control of Edison and the Commission. The Commission’s decision takes into account the requirement that the audits must be fully litigated.” In addition to the audit requirement, the Commission also based its decision on its belief that settlements between the Commission and utilities have been encouraged by the courts, and upon the appellate court opinion affirming the accounting variance for Braidwood Unit 2. In that decision, which was rendered after this court’s decision in Business & Professional People I, 136 Ill. 2d 192, the appellate court stated “Edison was not at fault for the time lag between when Braidwood Unit 2 went into service and new rates will come into effect.” Business & Professional People, 205 Ill. App. 3d at 896. The Commission is the fact finder in this case and its findings of fact are prima facie correct. We believe there is ample evidence in the record to support the Commission’s conclusion that Edison is not to blame for the regulatory delays. In making this conclusion, we are mindful that the Commission, acting through its staff, was an active participant in the negotiations which the interveners now claim led to much of the delay. Given the Commission’s participation we do not believe it would be appropriate to penalize Edison for entering into these settlement negotiations. Furthermore, we believe the Commission’s role in the negotiations placed it in a better position to determine whether Edison prolonged the negotiations unnecessarily. While the negotiations did not lead to a valid rate agreement, we cannot say that the Commission’s conclusion that Edison was not to blame is against the manifest weight of the evidence. As to the second part of the test, the intervenors presented evidence to show that Edison’s earnings and cost of capital were not in fact significantly and adversely affected during the regulatory delay period. However, the Commission decided that a determination on this prong of the test must be based on the record as it stood on January 1,1988. The Commission stated: “The Commission concludes that the second part of this test should be applied on a prospective basis, based on the record as it stood in 1988. This is the time when Edison presented evidence on its petition, and the time the sought-after accounting treatment would be implemented. For purposes of this test, to review now whether Edison suffered adverse financial impact in years subsequent to 1988 defies logic. * * * For purposes of recording Deferred Charges, it stands to reason that the Commission should refrain from employing a hindsight test to determine whether recording was reasonable. Rather, the Commission must apply this criterion using the evidence available at the time Edison was seeking approval to initiate this special accounting treatment.” Thus, the Commission based its decision on financial projections made in 1987, rather than on the actual historical data available. We find this decision was arbitrary. The decision to apply the test prospectively based on the record as it stood in 1988 is a legal determination, and is therefore entitled to less deference by this court. We note that the test, as set forth in Docket 85—0092, is whether failure to grant the accounting variance could adversely affect the company’s financial condition. In that docket, the Commission decided to grant the accounting variance within a few weeks of the plant’s in-service date. Five months later, the Commission reaffirmed that decision upon rehearing. At the time the variance was initially granted, the evidence available to the Commission was limited to projections about the impact of the regulatory delay. The present case is distinguishable from Docket 85—0092 in that at the time the Commission made the determination in this case, the Commission had at its disposal Edison’s actual financial data for the deferral period. In the instant case, the Commission arbitrarily ignored the actual financial data contained in the record. Rather than determine whether subsequent events proved Edison’s 1987 projections to be accurate, the Commission decided this issue as if it had no means to test these projections. This mistake was compounded when the Commission allowed Edison to recover the full amount of recorded deferred charges, without regard to the actual financial harm suffered. (See our discussion of single-issue ratemaking infra.) It would be anomalous to allow Edison to record and recover over $1 billion dollars in deferred charges based on projections instead of the historical evidence available in the record. This is especially true if, as the interveners claim, the actual financial data refutes those earlier projections. We, therefore, must remand for a determination of whether Edison actually suffered significant adverse financial impact due to the delay between the in-service date and the Rate Order date. If the Commission finds that Edison did suffer adverse financial impact, then it may allow the company to record deferred charges for Byron Unit 2 and Braidwood Unit 1. Our decision that the Commission acted arbitrarily is supported by the fact that the Commission did not allow Edison to record deferred charges for Byron Unit 2 from April 14, 1987, to December 31, 1987, and for Braidwood Unit 1 from November 19, 1987, through April 25, 1988. The Commission stated in the Sixth Interim Order “[t]he determination of these time periods is made in recognition that, as to Byron Unit 2 and Braid-wood Unit 1, Edison’s revenues were such that it did in fact recover its depreciation expense during 1987.” The Commission adopted these findings in the Remand Order. Thus there is precedent for relying on the actual financial data when deciding the issue of adverse financial impact. We believe the Commission should have used this approach when deciding the entire issue. RECOVERY OF DEFERRED CHARGES Because our decision above does not impact the Commission’s decision to allow Edison to record deferred charges for Braidwood Unit 2, and because the issue is likely to arise again if on remand the Commission finds that Edison may record deferred charges for Byron Unit 2 and Braidwood Unit 1, we must consider whether recovery of deferred charges is permissible. In the Rate Order, the Commission authorized Edison to include the full $1.7 billion of deferred charges in the rate base and to amortize the amount over the remaining useful lives of the plants. For ratemaking purposes, this has the dual effect of increasing Edison’s annual depreciation expense and increasing the company’s rate base on which its rate of return is based. Both of these effects increase Edison’s revenue requirement. The intervenors argue that allowing Edison to recover any deferred charges violates the Commission’s test-year principles, as well as the rules against retroactive and single-issue ratemaking. We will consider each of these claims individually. Test-Year Rules The intervenors claim that recovery of deferred charges violates the test-year principles established by the Commission’s rules. (See 83 Ill. Adm. Code §285.150 (1985).) As previously stated, a utility’s rates are a function of its annual revenues and operating expenses, as well as its rate base. In order to accurately determine the utility’s revenue requirement, the Commission established filing requirements under which a utility must present its rate data in accordance with a proposed one-year test year. The purpose of the test-year rule is to prevent a utility from overstating its revenue requirement by mismatching low revenue data from one year with high expense data from a different year. Business & Professional People I, 136 Ill. 2d at 219. The interveners contend that recovery of deferred charges allows Edison to mismatch expenses from the deferral period with the revenue data from the test year. In response, both the Commission and Edison argue that the deferred charges allowed in this case are unrecovered capital costs rather than operating expenses and, therefore, the test-year principles do not apply to deferred charges. Resolution of this issue will depend upon whether the test-year principles apply to deferred charges. The deferred charges consist of deferred depreciation, deferred decommissioning expense ánd deferred financial carrying costs. Edison argues .that there is no basis for distinguishing between these categories of deferred charges. However, because the nature of these categories of deferred charges are fundamentally different we believe we must address each category separately to determine if test-year principles apply. We begin with deferred depreciation. Edison argues that because depreciation by definition consists of the recovery of capital investments incurred and recorded in past years, it is neither an expense nor subject to test-year principles. Edison states “depreciation is clearly not like other ‘expenses,’ because depreciation charges do not reflect a current cash outlay comparable to those for wages or supplies.” In support of its position, Edison correctly points out that during construction of a plant, a utility capitalizes the direct costs of construction rather than expensing the costs as incurred. Edison further contends that if test-year principles apply to depreciation, utilities will be forced to choose between filing yearly rate cases or forgoing recovery of the costs of construction incurred during that year. Edison maintains that because this result is neither practical nor desirable, we should hold that depreciation is not subject to test-year principles. This portion of Edison’s argument focuses on whether depreciation is an operating expense. In determining whether depreciation is an expense, the critical inquiry is not how much cash was paid in a given period, but rather how much did the value of the underlying asset decline during that period. Edison’s argument improperly equates an expense with a current cash outlay and focuses on the time cash is paid for construction. While Edison is correct that construction costs are initially capitalized rather than expensed, this fact is not as important as the reason construction costs are treated in this manner. Expensing the construction costs as incurred would mismatch the costs and benefits of a plant because current ratepayers would be charged for a plant which produces no current benefit. For this reason, construction costs are capitalized and amortized over the useful life of the plant. The rationale for capitalizing construction costs does not mean that depreciation is not an expense. The plants have limited useful lives, and each year that a plant is in service a portion of its useful life is expended. Depreciation recognizes the cost of that portion of the asset which is expended in a given year, regardless of the time period in which the construction costs were actually paid. Thus, even though there is no cash outlay in the current year, depreciation is treated as an operating expense for financial reporting purposes, and more importantly for purposes of determining Edison’s revenue requirement. For this reason, we hold that depreciation is an expense subject to test-year principles. See Lindheimer v. Illinois Bell Telephone Co. (1933), 292 U.S. 151, 167-68, 78 L. Ed. 1182, 1193, 54 S. Ct. 658, 665 (“Annual depreciation is the loss which takes place in a year. In determining reasonable rates for supplying public service, it is proper to include in the operating expenses, that is, in the cost of producing the service, an allowance for consumption of capital ***. *** While property remains in the plant, the estimated depreciation rate is applied to the book cost and the resulting amounts are charged currently as expenses of operation”). Edison argues that the accounting variance did not create any new categories of expense because the ratepayers “will pay the same depreciation charges on the units they would have paid had rate recognition and in-service been synchronized.” Edison is correct.that, after adjusting for the higher carrying charges, the accounting variance does not increase the aggregate amount of depreciation expense which will be recognized over the course of the plants’ useful lives. However, because the entire cost of the plants is amortized over less than the full useful lives of the plants, the variance increases the annual depreciation expense to be recognized in each of the years following the deferral period. For this reason we find that recovery of deferred depreciation violates the test-year principles. As this court stated the last time this case was before it, “[t]he Commission may alter or amend its past practice, but it must follow the procedures set forth in its rules and the Act.” (Business & Professional People I, 136 Ill. 2d at 226.) The Commission could have amended its test-year rules to deal with the particular problems created by regulatory delay. Instead the Commission attempted to circumvent the existing rules by improperly treating depreciation as if it were not an operating expense. For this reason, we find the Commission violated its own rules to the detriment of the interveners and, therefore, committed reversible error. Business & Professional People 1, 136 Ill. 2d at 227. For similar reasons, we believe the Commission erred in holding that decommissioning expenses were not test-year items. Decommissioning expenses account for the cost of retiring the plants at the end of their useful lives. Under normal accounting procedures, Edison allocates a percentage of the total estimated decommissioning costs to each year the plant is in service, and deposits that amount in special trust accounts established pursuant to section 8—508.1 of the Act (Ill. Rev. Stat. 1989, ch. 111⅔, par. 8—508.1). In this way, the costs of retiring a plant are apportioned among the years the plant is in service, even though the actual costs of decommissioning will not be incurred until the end of the plant’s useful life. In discussing the amount of decommissioning expense that Edison should allocate to each year, the Commission found “the total amount included in these test-year expenses applicable to decommissioning costs is $66.0 million, which is based on the Company’s estimated total decommissioning costs expressed in 1987 dollars.” Despite this finding that decommissioning expense was a test-year item, the Commission permitted Edison to defer decommissioning expenses for the regulatory delay period. The Commission based this decision on its belief that ratepayers are responsible for the cost of decommissioning. Because decommissioning expense is a test-year item, we again find the Commission violated its own rules to the detriment of the interveners and, therefore, committed reversible error. Business & Professional People I, 136 Ill. 2d at 227. With respect to deferred carrying charges, we find that the test-year principles have not been violated. These deferred financing charges accumulated after the plants were placed in service. In order to determine whether test-year principles apply to these post-in-service carrying charges, we must examine how these costs would have been accounted for under normal accounting procedures, i.e., absent a regulatory delay. Under normal accounting procedures, the financing costs incurred prior to placing a plant in service are capitalized as AFUDC and amortized over the life of the plant. Thus these costs make up a portion of the depreciation expense allowed each year. However, carrying costs incurred after the plant is placed in service have consistently been treated differently for ratemaking purposes than have the pre-in-service financing costs. Under normal accounting procedures, post-in-service financing costs are neither expensed nor capitalized. Instead, these costs are recovered through the rate of return authorized on the company’s investment. This is significant in that after a plant is placed in service, the related carrying costs are not treated as operating expenses in the revenue requirement formula. We do not believe the distinction between pre-in-service and post-in-service financing costs is insignificant. Depreciation expense, including the AFUDC which accumulated prior to the in-service date, measures the decline in the value of an asset caused by its use during a year. Thus, depreciation expense is related to that portion of the asset used to produce electricity during a given year. On the other hand, post-in-service financing costs compensate Edison for the time value of its money which is still invested in the asset. These capital costs are a function of the value of the unused portion of the asset. We believe this is the fundamental reason the costs are treated differently for ratemaking purposes. As previously stated, the test-year rules are intended to prevent a utility from mismatching revenue and operating expense data. Because the post-in-service carrying charges are not operating expenses, they are not test-year items. Therefore we agree with Edison and the Commission that recovery of deferred financing charges does not violate test-year principles. Retroactive Ratemaking We next consider the interveners’ claim that allowing Edison to recover any deferred charges would violate the rule against retroactive ratemaking. Once the Commission establishes rates, the Act does not permit refunds if the established rates are too high, or surcharges if the rates are too low. (Business & Professional People I, 136 Ill. 2d at 209.) This rule is consistent with the prospective nature of the Commission’s legislative function in ratemaking. In addition, this rule promotes stability in the ratemaking process. Citizens Utilities Co. v. Illinois Commerce Comm’n (1988), 124 Ill. 2d 195, 207. The interveners argue that recovery of deferred charges assesses a current surcharge on ratepayers to compensate Edison for the revenues it may have lost during the deferral period. The intervenors contend that this constitutes retroactive ratemaking and, therefore, is prohibited under the Act. Edison and the Commission respond that the rates established in this case allow only prospective recovery of capital investments incurred in the past. According to Edison this is exactly what happens in any rate case involving the addition of new plants to the rate base. Edison notes that the Commission has never included the cost of the three plants in a valid rate order. Thus deferred charges are not intended to correct an error made in prior rate orders which had included these plants in rate base. Rather, Edison argues deferred charges are merely one aspect of the reasonable and prudent costs of the plants which should be included in rate base. We agree with Edison. The fact that deferred charges represent capital costs incurred in the past does not make recovery of these costs retroactive. Single-Issue Ratemaking We now turn to the intervenors’ argument that recovery of deferred charges constitutes single-issue rate-making. Initially, we note that the rates in effect at the time the plants were placed in service, and at the time of the Remand and Rate Orders, were established in a rate order entered in 1985. References to increases and decreases in elements of the revenue requirement are relative to the data used in setting the rates in effect at the time of the Commission’s orders. The rule against single-issue ratemaking recognizes that the revenue formula is designed to determine the revenue requirement based on the aggregate costs and demand of the utility. Therefore, it would be improper to consider changes to components of the revenue requirement in isolation. Oftentimes a change in one item of the revenue formula is offset by a corresponding change in another component of the formula. For example, an increase in depreciation expense attributable to a new plant may be offset by a decrease in the cost of labor due to increased productivity, or by increased demand for electricity. (Demand for electricity affects the revenue requirement indirectly. The yearly revenue requirement is divided by the expected demand for electricity to arrive at a per kilowatt hour rate. If actual demand is more than the estimated demand used in the formula, the utility’s revenues increase.) In such a case, the revenue requirement would be overstated if rates were increased based solely on the higher depreciation expense without first considering changes to other elements of the revenue formula. Conversely the revenue requirement would be understated if rates were reduced based on the higher demand data without considering the effects of higher expenses. In the present case, the intervenors argue that allowing recovery of deferred charges constitutes single-issue ratemaking because deferred charges measure only selected elements of the revenue formula. We agree with the intervenors that, as the record now stands, recovery of the full amount of deferred charges constitutes single-issue ratemaking. However, this argument speaks to the amount of charges which Edison may recover, rather than to whether recovery of any deferred charges is permissible. The intervenors allege that much, if not all, of the increased expenses recorded as deferred charges were offset by decreases in other operating expenses combined with higher revenues derived from increased demand. In fact, the intervenors contend that Edison’s rates would have been reduced if the Rate Order had been synchronized with the in-service dates. The intervenors contend that compensating Edison for the higher depreciation expense without considering changes in other elements of the revenue requirement formula constitutes single-issue ratemaking. We agree. Against the recommendations of the intervenors and staff, the Commission refused to limit recovery of deferred charges to the net loss Edison suffered due to the failure to synchronize. Instead, the Commission permit-Edison to recover the full amount of deferred charges recorded. In refusing to limit the amount of recovery, the Commission stated: “In response to OPC’s proposal to limit recovery of these costs on the basis that Edison is ‘only entitled to be made whole’, it should be understood that such has not been the Commission’s stated purpose in Dockets 85—0092 or 88—0253.” The Commission stated that the purpose of the accounting variance is “to preserve the opportunity to recover reasonable expenditures,” and, therefore, limitations on the recovery of deferred charges are not appropriate. This statement provides an overly narrow interpretation of the purpose of the accounting variance, which if correct would violate the rule against single-issue rate-making. In response to the interveners’ arguments that recovery of deferred charges constitutes single-issue and retroactive ratemaking, the Commission and Edison rely on the appellate court’s decision in Business & Professional People, 205 Ill. App. 3d 891. In that case the appellate court reviewed the Commission’s order allowing Edison to record deferred charges for Braidwood Unit 2. In that case, the appellate court stated: “The ICC order of June 1, 1989 was an order to affect accounting procedures, not a ratemaking decision. As such, that order does not foreclose any discussion or presentation of evidence that would normally occur when ICC conducts the ratemaking hearing for Braidwood Unit 2. Nor does the order constitute a backdoor approach to single-issue or retroactive ratemaking.” (Business & Professional People, 205 Ill. App. 3d at 896.) This decision highlights the fact that there is a fundamental difference between a decision to record deferred charges and a decision to recover deferred charges. While the Commission’s position against limiting the amount of deferred charges may be correct for recording purposes, it is not appropriate for recovery of deferred charges. There is a fundamental difference between a rate case and a case involving only accounting procedures. A decision to permit recording of deferred charges has only an indirect impact on rates and, therefore, is not subject to the same scrutiny as a decision to raise rates. It is improper to assume, as the Commission did, that having met the criteria for recording deferred charges, Edison is automatically entitled to recovery of the full amount recorded. This is especially true because recovery of deferred charges is a deviation from the normal accounting procedures. The purpose of the accounting variance is to protect Edison from adverse financial impact caused by the regulatory delay period, and to afford Edison the opportunity to recover these charges. The accounting variance should not be used to place Edison in a better position than it would have been in had synchronization been achieved. Just as it would be unfair to deny Edison recovery of its reasonable and prudent investment due to regulatory delays which the company could not control, so, too, would it be unfair if Edison were allowed to reap a windfall, at ratepayer expense, due to a regulatory delay which the ratepayers could not control. This is especially true in the present case in which at least a portion of the delay period is attributable to the illegal Sixth Interim Order to which the intervenors objected. In order to protect Edison from adverse financial impact while at the same time protecting the ratepayers, it is necessary to determine the extent of the harm Edison suffered as a result of the delay. It is possible, as the intervenors argue, that the amount Edison lost due to increased depreciation charges was offset by decreases in other expense items. If this is the case, the net effect which the regulatory delay had on Edison’s earnings is less than the amount of deferred charges recorded. Under this scenario, Edison would be granted a windfall if it could recover the full amount of deferred charges. The test established in Docket No. 85—0092 considers only whether Edison suffered some significant adverse financial impact. This test is sufficient for recording purposes, but for purposes of recovery it is also necessary to determine the extent of this harm. We trust that the Commission will be able to formulate some standard for recovery which will protect both the interests of Edison and the ratepayers. We, therefore, remand to the Commission for development of such a standard under which a determination of the proper amount of deferred charges to be recovered can be made. Consistent with our holdings above, the Commission should determine the amount of deferred charges which Edison is entitled to recover in light of the test-year principles and rules against single-issue ratemaking. This amount may then be included in Edison’s rate base for the purpose of establishing prospective rates. We believe that in this way Edison will be granted the opportunity to earn a fair return on its investment, while at the same time the interests of ratepayers will be protected. PROPER DEFEÉRAL PERIOD In its cross-appeal Edison argues that the Commission arbitrarily shortened the period for which it allowed Edison to recover deferred charges related to the two Braidwood plants. Specifically, Edison argues the Commission should have permitted Edison to begin deferring charges for Braidwood Unit 1 on January 1, 1988, and for Braidwood Unit 2 on August 5, 1988. The Commission disallowed recovery of deferred charges for these periods because the plants were “not at a sufficiently reliable operating level to justify recovery of deferred charges.” The Commission found that Braidwood Unit 1 “was in a shutdown condition from the beginning of 1988 until April 14, 1988 [April 25, 1988].” (Corrected date in original.) In addition, the Commission found that “Braidwood 2 went through approximately three or four periods when it was shut down for four to five days prior to reaching commercial operation on October 17, 1988. The Commission is of the opinion that the calculation should reflect these outages *** and accordingly, the allowance for carrying charges and deferred depreciation should be reduced by twenty days for the period from Edison’s in-service date through its commercial operation date of October 17, 1988.” The Commission disallowed recovery of deferred charges for these periods. Edison argues that the plant shutdowns were caused by licensing requirements of the Nuclear Regulatory Commission and not by any imprudence on Edison’s part. Edison argues that the deferred charges during the outages in this case are analagous to purchases of replacement power during outages of other generating facilities. Because the Commission allows recovery of the costs of such purchases unless the outage is caused by the utility’s imprudence, Edison argues the Commission should allow recovery of the deferred charges in this case. Edison further argues that the Commission has provided no rationale to differentiate the treatment of deferred charges and purchased power during outage periods. The intervenors refute this analogy by pointing out that recovery of the costs of purchased power is specifically authorized in section 9 — 220 of the Act. The question of whether Edison is entitled to recover deferred charges during the outage periods will depend upon the standard for recovery of deferred charges. Because we are remanding to the Commission for establishment of these standards, we are unable to review this issue. Therefore, upon remand the Commission should consider this issue under the standards it establishes. MANDATE IN BUSINESS & PROFESSIONAL PEOPLE I The intervenors contend that recovery of any deferred charges violates this court’s mandate in Business & Professional People I, 136 Ill. 2d 196. Specifically, the intervenors argue that because this court ordered Edison to refund the excess rates collected under the Sixth Interim Order, the company may not now recover any deferred charges which accrued prior to this court’s opinion in Business & Professional People I. The interveners argue that those deferred charges which accrued prior to this court’s opinion represent the same rates which this court found to be excessive. We disagree. This court’s opinion in Business & Professional People I was based on the improper procedures the Commission employed in setting the rates. Further, the refund provision was specifically based on Edison’s pledge to refund any excess rates collected under the Sixth Interim Order if that Order was overturned on jurisdictional grounds. Neither of these grounds for the Business & Professional People I decision was intended to foreclose consideration of the proper amount to be included in Edison’s rate base. Therefore, we do not believe that recovery of deferred charges violates the decision in Business & Professional People I. Finally with respect to deferred charges, we find the Commission failed to properly consider the time value of money when it computed Edison’s post-in-service carrying charges. The Commission permitted Edison to accrue post-in-service carrying charges at a rate of 10.51% during the regulatory delay period. This element of deferred charges compensates Edison for the carrying charges incurred during the deferral period, and is based on the interest rate for funds borrowed by Edison. In computing the post-in-service carrying charges, the Commission did not take into account the excess rates collected under the Sixth Interim Order. Edison had the use of these ratepayer funds during a portion of the deferral period. Although the excess rates must be refunded to the ratepayers with interest, the proposed interest rate of 5% paid on the refunds is significantly lower than the rate allowed for deferred carrying charges. If deferred charges are not adjusted to recognize the difference in these interest rates, Edison will receive excess compensation for the time value of its money during the deferral period. ISSUES UNRELATED TO NUCLEAR PLANTS In its cross-appeal, Edison challenges the Commission’s findings with respect to the reasonableness of its charitable contributions and payroll expenses. In addition, Edison argues that the Commission improperly disallowed the company’s western coal reserves from its rate base, and that the Commission’s treatment of deferred income taxes was arbitrary. Finally Edison argues that the rate modification plan imposed by the Commission improperly deprives the company of the full rate increase to which it is entitled. In a related argument, two intervenors contend that the Commission improperly apportioned the rate increase equally among the various ratepayer groups (i.e., commercial and residential). We will consider each of these arguments separately. PAYROLL EXPENSE Edison argues that the Commission disallowed approximately $40 million dollars in reasonable payroll and related expenses. In a related argument, the company contends the Commission improperly excluded $9 million dollars from Edison’s rate base. This $9 million represents an adjustment to the amount of payroll costs Edison was expected to capitalize as part of “Utility Plant in Service.” Edison maintains that these disallowances were not supported by evidence in the record. The evidence showed that in past years Edison budgeted for hundreds of employees that were not in fact employed by the company. Based on this evidence and an internal study prepared by Edison, the Commission disallowed the above amounts. Edison argues that the Commission incorrectly focused on the number of budgeted employees rather than the budgeted dollar amount for payroll expense. Edison states that the allowable payroll expense must be based on the cost of providing service, and that this is a function of the amount of work required, not the number of employees. Edison argues that after overtime and outside contractors are considered its past budgets have accurately estimated the total cost of providing service. Therefore, Edison argues the Commission’s findings are arbitrary and unsupported by substantial evidence. Edison raised these same arguments before the Commission. Edison’s argument challenges the Commission’s findings of fact. As previously stated, the Commission’s findings and conclusions on such questions shall be held to be prima facie true. (Ill. Rev. Stat. 1987, ch. 111⅔, par. 10—201(d).) In this instance, the evidence showed Edison’s budgets for past years overstated the number of employees needed in those years. In addition, Edison’s internal study supported a reduction in Edison’s employee level. The Commission considered this evidence, and Edison’s arguments related to this evidence, and determined that a reduction in the payroll expense was appropriate. Although this is not the only conclusion the Commission could have drawn from this evidence, we cannot conclude that this finding is unsupported by the evidence. CHARITABLE CONTRIBUTIONS Edison argues that the Commission arbitrarily disallowed $1.269 million of charitable contributions from Edison’s operating expenses. In the operating budget filed with its proposed tariffs, Edison included $3.269 million of charitable contributions. Based on a mechanical formula, which made pro forma inflation adjustments to the amount of contributions allowed in prior Commission orders, staff proposed to allow only $707,000 of contributions as operating expenses. The Commission concluded that staff’s position was contrary to the Act and that, in its “judgment, recoverable charitable contributions should be set at a more reasonable level of $2 million.” Edison argues that the Commission’s disallowance of contributions in excess of $2 million violates section 9—227 of the Act (Ill. Rev. Stat. 1987, ch. 111⅔, par. 9—227). Section 9—227 provides in part: “It shall be proper for the Commission to consider as an operating expense [charitable contributions] *** provided that such donations are reasonable in amount. In determining the reasonableness of such donations, the Commission may not establish, by rule, a presumption that any particular portion of an otherwise reasonable amount may not be considered as an operating expense. The Commission shall be prohibited from disallowing by rule, as an operating expense, any portion of a reasonable donation for public welfare or charitable purposes.” Ill. Rev. Stat. 1987, ch. 111⅔, par. 9—227. Edison argues that the Commission made an impermissible disallowance by rule because it did not make a particularized finding that any individual donation was unreasonable. Edison further claims the Commission placed an arbitrary cap of $2 million on the amount of donations allowed as an expense, without regard to whether any donations above that amount were reasonable. Edison also argues the Commission’s decision was arbitrary because it deviated from the Commission’s treatment of charitable contributions in the the Remand Order. Specifically, Edison points to the following statements from the Remand Order: “Edison states that the Act was amended in 1987 to require a particularized evaluation of the reasonableness of charitable donations. * * * The Commission concurs in Edison’s statutory interpretation, finds the amount is reasonable, and has incorporated Edison’s proposed language in the body of the Order.” The proposed language referred to above is the following: “The Commission finds Edison’s proposal to be reasonable and it is adopted. Neither [staff’s nor the intervenors’ witnesses] have made any independent determination or provided any evidence that Edison’s charitable contributions are unreasonable in amount, which showing is a condition precedent for any disallowance of charitable contributions under Section 9—227 of the Act.” Edison cites these passages as support for its position that an individual contribution must be shown to be unreasonable before a disallowance is permissible. We disagree. Edison is correct that the Act prohibits the Commission from “disallowing by rule, as an operating expense, any portion of a reasonable donation for public welfare or charitable purposes.” (Ill. Rev. Stat. 1987, ch. 111⅔, par. 9 — 227.) However, section 9—227 does not provide that every donation Edison makes to a qualified organization is presumed reasonable. Edison still has the burden of showing that a donation is reasonable in amount. When this burden of proof is considered, the Act merely states that the Commission shall not disallow any portion of that amount which Edison has shown to be reasonable in amount. Contrary to Edison’s position, we believe that the Commission must determine the reasonableness of the amount of contributions based on the total contributions rather than on an individualized basis. There are numerous charitable organizations worthy of Edison’s support. If Edison were to make a reasonable donation to each of these organizations, the aggregate total of the donations could very easily exceed a reasonable amount. Our conclusion is further supported by the testimony of Edison’s witness on this issue, who testified as follows: “Whereas the Company does not forecast charitable contributions by individual recipient, there is a high probability that many of the charities receiving contributions in 1989 will also receive funds in 1991. Individually they will not necessarily receive the same funds as 1989 but the total level of contributions in 1991 is expected to be about the same.” (Emphasis added.) (ICC Docket No. 90—0169.) Thus, Edison did not present evidence of the amount that each donee organization would receive in the 1991 test year. Without this information, it would be impossible to determine if each individual donation were reasonable in amount. Moreover, we believe it is significant that Edison’s witness testified in terms of “the total level of contributions.” Because we find the Commission must determine the reasonableness of the amount of total charitable contributions, we find the Commission did not violate section 9 — 227. The Commission found that $2 million was a reasonable amount of charitable contributions. We will not substitute our judgment for that of the Commission on findings of fact such as this. DEFERRED INCOME TAXES Edison next argues that the Commission’s treatment of certain excess accumulated deferred income taxes (ADIT) was arbitrary. Deferred income taxes arise from the timing differences between the tax treatment and financial reporting treatment of various revenues and expenses. ADIT represents the total amount of “pre-paid” taxes which Edison has already collected from ratepayers but which Edison has not yet paid to the government. Edison argues that the Commission’s decision to amortize certain portions of the ADIT account over three years was arbitrary. Before discussing this issue, it will be helpful to have an understanding of deferred income taxes. Because Edison treated some expenses differently for tax purposes and financial reporting purposes, the company’s actual tax liability did not always coincide with the tax liability included as an operating expense for ratemaking purposes. For example, with respect to some of its assets, Edison used straight-line depreciation for financial reporting purposes but accelerated depreciation for tax purposes. The accelerated depreciation method used for tax purposes increased the tax deduction the company was permitted to take in the early years of the asset’s life, which, in turn, reduced the company's tax liability in those years. This tax savings is not permanent, however, because in later years the tax benefits (i.e., reduction in tax liability related to depreciation deduction) from the accelerated depreciation method will be lower than the tax benefits under the straight-line method. In order to equalize the annual tax benefits related to depreciation of the asset, the company’s revenue requirement is determined based on the data used for financial reporting purposes. The difference between this figure and the actual tax liability in a given year is accounted for in the ADIT account. To determine the amount of ADIT to record each year, the company first determines the aggregate tax benefits it expects to receive in each year based on the financial reporting data. This calculation is dependent upon the anticipated future tax rate. The difference between the annual tax benefit under this calculation and the actual annual tax benefit is recorded in the ADIT account. In later years, when the actual tax benefit related to the asset is lower than the tax benefit for ratemaking purposes (i.e., the actual tax expense is higher than that used for ratemaking), the company amortizes a portion of the ADIT account to make up the difference. Prior to 1986, Edison computed ADIT based on an anticipated future tax rate of 46%, the rate then in effect. However, the Tax Reform Act of 1986 lowered the Federal rate to 40% for 1987 and 34% for subsequent years. Due to this reduction in the corporate rate, the amount of ADIT Edison recorded in years prior to the Tax Reform Act overstates the amount of taxes the company will be required to pay in the future. That is, because the company overestimated the amount of taxes it would pay in future years, the company has charged ratepayers more for deferred taxes than the company will actually pay to the Federal government. The difference between the amount of deferred taxes collected from ratepayers and the amount which would have been collected under the new tax rates is referred to as excess ADIT. The parties agree that Edison’s excess ADIT may be separated into three categories and that these categories may impact the rate treatment of the excess ADIT. The first category is ADIT related to property for which the company used accelerated depreciation for tax purposes. Under the Tax Reform Act, this category of excess ADIT must be “flowed back” over the remaining life of the asset. The second category relates to property for which no accelerated depreciation was taken. This category is referred to as unprotected or unrestricted excess ADIT because the Tax Reform Act does not require any particular treatment for rate-making purposes. The third category is unrelated to any property and is also not subject to restrictions by the Tax Reform Act. Our review in this case is limited to the treatment of the second category above — unrestricted excess ADIT related to property. The Commission decided that Edison should amortize this portion of ADIT over a three-year period rather than over the life of the underlying asset as suggested by Edison. Edison now argues that the Commission’s decision on this matter was arbitrary and unsupported by the evidence. We disagree. Edison argues that because this category of excess ADIT is related to property, the ADIT should be flowed back over the useful life of the property in order to match the ADIT with the benefits derived from the property in each period. Edison points out that the depreciable costs of its plants, the costs of the construction audits and the deferred charges discussed earlier in this opinion will all be amortized over the useful lives of the related plants. Edison states that the Commission provided no reasoned basis for treating property related excess ADIT differently. The Commission rejected Edison’s proposal to amortize the second category of excess ADIT over the remaining useful lives of the underlying assets. Instead, the Commission found that “a three-year flowback period is reasonable because it directly reimburses those customers who have contributed to the excess ADITs.” The excess ADIT represents money Edison has already collected from ratepayers to cover future income tax expenses which the company will not incur. We agree with the Commission that it is fair and reasonable to return these excess ADITs to the ratepayers who actually paid this money to the company. Based on its review of the evidence, the Commission decided that the best way to achieve this end was to amortize the amount over three years. We believe that the Commission’s findings on this issue were fair and reasonable as well as supported by the evidence and sound reasoning. WESTERN COAL RESERVES Edison next argues that the Commission improperly excluded $383.3 million of western coal reserves from its rate base. Edison had included this amount in its budget as property held for future use. The Commission found that none of the reserves qualified as property held for future use, although some of the reserves may be included in Edison’s rate base as part of its fossil fuel inventory. The coal reserves at issue here are an outgrowth of coal purchase contracts Edison entered into in the 1970s. At that time Edison agreed to purchase certain quantities of coal over an extended period of time. Because the demand for electricity did not grow as fast as Edison had expected, in the 1980s the company did not need as much coal as it had previously agreed to purchase. In 1982 and 1983, the company negotiated with its suppliers to reduce the amount of coal to be delivered currently. These negotiations resulted in “agreements to defer actual mining of some coal in exchange for the company’s agreement to make certain prepayments and receive what is called an ‘interest’ in the coal in the ground; i.e., the coal reserves at issue here.” (See ICC Docket No. 83—0537; 61 Pub. Util. Rep. 4th 1, 9.) In essence, the company prepaid its suppliers for the coal under the terms of the base contract, less the cost of mining and delivering the coal. In return, the company obtained the right to defer delivery of the coal until some future date. When the coal is actually needed, Edison will pay only the cost of mining and delivery. Thus, under the base coal contracts and related deferral agreements, Edison has hedged against future increases in the price of the coal. Edison now wants to include the value of these coal reserves as property held for future use. “ ‘Property held for future use’ refers to a utility’s investment in property which is not currently being used to provide service to its customers but which will be used in the future to provide such service.” (City of Chicago v. Illinois Commerce Comm’n (1985), 133 Ill. App. 3d 435, 440.) Generally a “utility is entitled to earn a return on its investment in property held for future use if the property was acquired in good faith with a definite plan for its use and it is reasonably acquired and retained to serve the utility’s customers.” (Chicago v. Illinois Commerce Comm’n, 133 Ill. App. 3d at 441.) Edison argues that because the Commission previously found that the base coal contracts and the deferral contracts were prudent (See ICC Docket No. 83—0537; 61 Pub. Util. Rep. 4th at 10), the company may include the amount of coal expected to be used during the next 10 years in its rate base as property held for future use. The Commission previously considered the rate-making treatment of these coal reserves in Docket No. 83—0537. (61 Pub. Util. Rep. 4th 1.) In that docket the Commission held that the “ ‘reserve coal’ is not property which is used and useful in providing utility service, or which is highly likely to be so used and useful in the future under a plan for its use.” (61 Pub. Util. Rep. 4th at 10.) Therefore, the Commission excluded the coal reserves from Edison’s rate base and ignored the coal reserve contracts for ratemaking purposes. Instead, the Commission directed Edison “to account for the cost of the reserve coal when it is used, for rate-making purposes, as if the coal was being bought currently on the. open market; Edison should not use the then-current base coal contract price.” 61 Pub. Util. Rep. 4th at 10. In the Remand Order, the Commission “reaffirm[ed] its decision in Docket 83 — 0537 as to the treatment of these coal reserves,” and also found that the coal reserves “are essentially coal inventory.” By treating the coal reserves as inventory, the Commission permitted Edison to include in its rate base only that amount of the reserves which was deemed proper as fossil fuel inventory. Therefore, under the Remand Order none of the reserves was included as property held for future use. In the Rate Order the Commission again “reaffirme[ed] its decision in Docket 83 — 0537 as to the treatment of these coal reserves.” Although the Rate Order did not specifically adopt the Commission’s findings contained in the Remand Order, the Rate Order “noted that the Commission also rejected this same proposal in [the Remand Order].” We believe this statement from the Rate Order was intended to reaffirm the Commission’s finding that the coal reserves are coal inventory. We conclude that the Commission’s classification of the coal reserves as inventory is within the sound discretion of the Commission and is supported by the evidence. Edison states that the Commission’s prior orders relating to the treatment of the coal reserves are irrelevant to the present rate case because the prior orders were based on the amount of reserve coal which Edison expected to use under conditions prevailing at the time of the orders. Because the amount of coal Edison expects to use has changed, Edison argues the Commission must evaluate this issue de novo to determine the amount of coal reserves to be included in Edison’s rate base. We agree with Edison that changing conditions will affect the amount of coal reserves to be included in rate base. However, we believe the Commission properly considered these changed conditions when it decided the issue of fossil fuel inventory. We note that Edison does not challenge the Commission’s findings on that issue. RATE MODERATION AND ALLOCATION OF INCREASE Edison argues that the Commission erred in adopting a rate moderation plan which failed to adequately compensate the company for its reasonable and prudent investment. In a similar argument, two interveners, the Illinois Industrial Electrical Consumers and the Chicago Transit Authority, argue the Commission erred in allocating the rate increase equally among the customer user groups. Both arguments relate to the Commission’s decisions on implementation of the specific rate increase allowed in the Rate Order. Because we reverse the Commission’s Rate Order, we need not pass on these issues. On remand the Commission will establish new rates, and presumably a new moderation and allocation plan. Because we have no way of knowing what the Commission’s decisions will be on remand, we believe it would be best if the parties addressed their arguments to the Commission on remand. Reversed and remanded. JUSTICE HEIPLE took no part in the consideration or decision of this case.