George K. Cracraft, Judge.
This case was commenced by General Telephone Company of the Southwest in May of 1985 before the Arkansas Public Service Commission seeking $4,600,000.00 to $6,410,615.00 in additional revenues, which would increase basic local rates by 39 % to 56 %. The Company requested that the appellee authorize rates sufficient to achieve a return on equity of 16.25 % and an overall rate of return on rate base of 11.37%. In March of 1986, following hearings on the merits, the appellee by its Order No. 10 authorized appellant to collect $809,001.00 of the requested additional revenues and allowed a return on equity of 12.13 % with a return on rate base of 8.78 %. One month later, the appellee’s staff moved for rehearing on certain issues and, after rehearing, the appellee entered Order No. 18 on October 28,1986, and revised Order No. 10 downward to reflect a revenue deficiency of $159,165.00.
The original award of $809,001.00 had been appealed to this court and, after the modification in October, 1986, the issues involved in rehearing were appealed. Because the record was voluminous and related to both the original orders and those orders which resulted from rehearing, the second appeal was consolidated with the first, and the entire case is now before this court.
The issues raised by appellant on appeal are as follows:
1. Without substantial evidence in the record, and by Orders No. 10 and 11, the Arkansas Public Service Commission arbitrarily and capriciously and without notice used the modified balance sheet approach to find the cash working capital component of rate base and a cost free component of capital structure, resulting in a confiscation of property and return on investment.
2. The overall rate of return of 8.78 % and the return on equity of 12.13% constitute the unlawful taking and confiscation of property, are not supported by substantial evidence, and double leverage as applied fails to take into account financial risk.
3. The finding of Order No. 18 that the third quarter 1985 Arkansas intralata toll pool rate of return of 11.3889% was appropriate for General Telephone Company of the Southwest’s toll revenue requirement is not supported by substantial evidence and the method by which the 11.3889% was calculated is contrary to established ratemaking principles; and has the effect of overstating toll revenues for the test period.
4. The Arkansas Public Service Commission by Order No. 9 arbitrarily and capriciously overruled General Telephone Company of the Southwest’s motion to extend the final order deadline, and denied General Telephone Company of the Southwest the opportunity to present the latest and best toll pool results occurring within the adjusted test year.
5. The accounting adjustment finding contained in Order No. 18 as appropriate for cost savings projects arbitrarily and capriciously included adjustments taking place, if at all, after the close of the adjusted test year; and has the effect of understating reasonable and necessary expenses for the test period.
6. The Arkansas Public Service Commission arbitrarily and capriciously granted a rehearing to the staff of the Arkansas Public Service Commission by Order No. 13 because the staff is not a statutory or procedural party who can petition for rehearing; and arbitrarily and capriciously overruled General Telephone Company of the Southwest’s motion to dismiss the rehearing by Order No. 15.
We find no error and affirm the orders of the appellee.
Our standard of review of appeals from the Public Service Commission is limited by the provisions of Ark. Stat. Ann. Section 73-229.1 (Supp. 1985), which defines our scope of review as a determination of whether (1) the Commission’s findings of fact are supported by substantial evidence; (2) the Commission has regularly pursued its authority; and, (3) the order under review violated any right of appellant under the laws or Constitution of the State of Arkansas or the United States. In reviewing such cases, we must give due regard to the expertise of the Commission, which derives its ratemaking authority from the general assembly.
Although we do not judge the wisdom of the decisions of the Commission, our review is not a mere formality but seeks to determine whether the findings are properly supported and that there has been no abuse of discretion. We do not advise the Commission how to discharge its function of arriving at findings of fact or the exercise of its discretion. The question of reasonableness of the action of the Commission relates only to its findings of fact and whether it has acted arbitrarily. Southwestern Bell Telephone Co. v. Arkansas Public Service Commission, 18 Ark. App. 260, 715 S.W.2d 45 (1986).
The Commission has broad discretion in choosing its approach to rate regulation and is free within its statutory authority to make the pragmatic adjustments which may be called for under particular circumstances. The appellate court is generally not concerned with the methodology used by the Commission in reaching its result, so long as its findings are based on substantial evidence and are not arbitrary. In these cases, it is the result reached, not the method employed, which controls; and judicial inquiry is concluded if the decision is supported by substantial evidence and the total effect of the order is not unjust, unreasonable, unlawful or discriminatory. Federal Power Commission v. Hope Natural Gas Co., 320 U.S. 591 (1944); General Telephone Co. v. Arkansas Public Service Commission, 272 Ark. 440, 616 S.W.2d 1 (1981); Southwestern Bell Telephone, supra. It is not the theory, but the impact of the rate order that counts in determining whether rates are just, reasonable and nondiscriminatory, and if the total effect of the rate order cannot be said to be unjust, unreasonable, or discriminatory, judicial inquiry is concluded and infirmities in the method employed are deemed unimportant. Southwestern Bell Telephone Co., supra; Walnut Hill Telephone Co. v. Arkansas Public Service Commission, 17 Ark. App. 259, 709 S.W.2d 96 (1986). Guided by these observations as to our standard of review, we address the issues presented by appellant on appeal.
Appellant first questions the validity of a methodology known as the “modified balance sheet approach” to determine appellant’s working capital requirements. Appellee’s staff witness Rodney Merritt described this methodology as follows:
The basic theory behind [the Modified Balance Sheet] approach is simply that assets which are necessary to provide utility service, which are not considered elsewhere in rate base, and which are not interest bearing, should be included in rate base, thereby allowing the Company to earn a return on these assets. Additionally, current, accrued, and other liabilities are considered to be a source of funds used to finance the assets of the Company and are placed in the capital structure at their respective costs.
(R. 497). Working capital is the cash and other non-plant investment in assets a utility must maintain in order to meet its current financial obligations and provide utility service to its customers in an economical and efficient manner. Since at least part of working capital represents a contribution from investors, an amount is generally included in the calculation of a utility’s rate base upon which a return is allowed.
The appellee has in previous rate cases utilized what is known as a “lead/lag study” for determination of the working capital requirement of a utility. Appellant described a lead/lag study in its brief at page 5 as follows:
A lead/lag study measures the cash working capital requirement as net lag days between the time utility service is rendered to a customer and the time customers pay for that service reduced by the offsetting leads between the time goods and services are rendered to the utility and the time the utility pays for goods and services. The investor in the utility supplies the capital for the net lag from the time costs are incurred until payment is received.
A lead/lag study generally will result in a ratio of lead or lag days to the total number of days in a year, and that ratio is then applied to various components on the company’s income statement to yield the working capital requirement. As is pointed out in the briefs, the appellee’s own filing requirements require a utility to file a rate request to calculate leads and lags. The modified balance sheet approach (MBSA) was adopted as a check on lead/ lag studies in a previous Commission docket,
wherein the appellee directed appellee’s staff to check the results of any lead/ lag studies by comparison with a calculation accomplished through the modified balance sheet approach. According to staff witness Merritt, a lead/lag study is the generally accepted measure for determining working capital. He testified:
Staff accepts a properly prepared lead/lag study and proves the results of that study by comparing it to the results of the modified balance sheet approach. In the absence of a properly prepared lead/lag study, Staff will calculate the working capital requirement using the modified balance sheet approach.
(R. 496-97).
Merritt testified that the appellee’s staff rejected appellant’s lead/lag study here because it contained deficiencies which resulted in calculations which were unverifiable or errors which were uncorrectable. While Merritt testified that, although some alleged deficiencies in the lead/lag study could be corrected by appellee’s staff, certain problems could not be corrected and cited as an example the inclusion of expenses in the study which were not those of the test year utilized in the rate case. Merritt testified that this deficiency was serious because:
The expense inputs to the study must be functionalized and properly measured to arrive at individual lead days. Those individual lead days and expenses are combined to determine composite lead days and dollar days for functional categories.
The $100,000,000 discrepancy [an error found by Staff] impacts at least four (4) functional categories and the lead days for those categories. Correcting errors of this magnitude would involve a great deal of analytical work on individual expense inputs and recalculation of lead days and dollar days.
(R. 495-96).
Because of the deficiencies Merritt testified he found with appellant’s lead/lag study, he proceeded to determine appellant’s working capital requirement by the use of the modified balance sheet approach. Merritt testified in some detail as to which asset accounts were included or excluded in rate base and articulated in detail his reasons for those actions. He also explained why current, accrued, and other liabilities were placed in appellant’s capital structure. Merritt explained his treatment of liabilities as follows:
The rationale for placing liabilities in the capital structure is simply that all liabilities are a source of funds used to finance the assets of a company. Moreover, the distinction cannot be made as to which asset each liability is funding, i.e., liabilities are fungible. This has long been recognized for the largest liability on the balance sheet, long-term debt. If this were not the case, one could identify specific debt funding non-utility investments and simply exclude the interest expense associated with that debt and accordingly, include in the cost of service the interest expense associated with debt financing utility property only. However, since that distinction cannot be made, long-term debt is placed in the capital structure and the weighted cost applied to the rate base.
If we are to determine the total cost of funds to a company, current, accrued, and other liabilities cannot be ignored, nor can they be netted against working capital assets and placed in the rate base. To do so would violate the concept of fungibility for these liabilities while accepting it for others. This would be both inconsistent and theoretically incorrect.
(R. 501-02). (Emphasis in original). Essentially, then, the modified balance sheet approach proceeds on the notion that all liabilities are representative of funding sources and that these funds cannot be traced to any particular rate base asset; therefore, all funding sources are accounted for on the liability side of the balance sheet at their respective costs.
In our view, near the heart of appellant’s and the appellee’s disagreement is the contention that liabilities are “fungible.” The parties do not quarrel that at least one funding source is “fungible”: long-term debt. Appellant claims, however, that the other funding sources can be traced to particular assets, whereas the appellee contends that it is inconsistent to treat one source of funding as fungible and other sources as not fungible.
Arkansas Statutes Annotated Section 73-237 (Repl. 1979) puts the burden of justifying any proposed change in rates upon appellant. There was evidence that appellant’s lead/lag study contained errors and problems which could cause reasonable persons to doubt its reliability. For example, an auditor found a $100,000,000 expense error in the original lead/lag study. Although corrections were apparently attempted, the appellee’s staff was unable to verify the corrections or reconcile the results to the modified balance sheet approach. While appellant argues that an error of $100,000,000 was relatively insignificant because the net revenue effect was under $30,000, we are not so much concerned with the relative magnitude of any error as we are with the appellee’s finding that the lead/lag study was not reliable. Whether the lead/lag study should be accepted is a matter for the trier of fact, and, on the record before us, we cannot say the appellee’s rejection thereof is not based on substantial evidence.
In this case, the appellee did not accept appellant’s evidence as to working capital based on a lead/lag study but, instead, accepted the appellee’s staffs evidence as to the working capital requirement resulting from calculations utilizing the modified balance sheet approach. A working capital allowance represents only a portion of the utility’s rate base upon which it is allowed to earn a return. It is apparent that no particular methodology is precise and that a determination of working capital is in many respects an exercise of discretion as to what particular method yields the most fair and equitable result in each case. Without question, the particular amount of working capital allowance, along with the particular methodology used to derive that amount, is a matter of educated opinion, expertise, and informed judgment of the Commission and not one of mathematically demonstrable fact.
As the trier of fact in rate cases, it is within the province of the appellee to decide on the credibility of the witnesses, the reliability of their opinions, and the weight to be given their evidence. The appellee is never compelled to accept the opinion of any witness on any issue before it. The appellee is not bound to accept one or the other of any conflicting views, opinions, or methodologies. Arkansas Public Service Commission v. Continental Telephone Co., 262 Ark. 821, 561 S.W.2d 645 (1978).
The evidence also sharply conflicted as to whether “return on return” is allowable in calculating rates. The parties seem to agree that, during the time the utility renders service to the customer until the time the company receives payment from the customer for that service, the utility theoretically advances to the customer some amount of money, which is includable as working capital in the rate base. Whether the company advances the “cost” as opposed to the “price” of the service to the customer is the point at which the witnesses diverge. Appellant contends that the utility advances both the expenses incurred in connection with rendition of utility service along with the return (or profit) the utility will earn as a result of rendering service. Therefore, according to appellant, appellant advances the “price” of the service to the customer. The opposing view proceeds from the posture that the ratepayer pays the return or profit and that the only advance by appellant is the cost it actually incurs in providing the service. The appellee found and concluded that the utility only advances the “cost” of service and not the “price” (i.e., cost plus profit or return) to the customer and that allowing the application of net lag days to revenues is improper because it allows appellant to earn a return on top of a return. The appellee argues that acceptance of appellant’s position on this issue results in a portion of profit being placed in rate base and another profit allowed thereon, which it contends overcompensates the utility at the expense of the ratepayer. From our review of the record, we cannot conclude that the finding of the Commission is not supported by substantial evidence.
The appellant’s argument that use of the modified balance sheet approach to determine cash working capital was contrary to the appellee’s Rules of Practice and Procedure is without merit. As stated in Rule 9.01 of the Rules, the purpose of the appellee’s filing requirements is to define the information desired by the appellee from a utility when a rate application is filed and not to establish particular ratemaking principles. The appellee’s rules as to filing requirements merely provide a starting point from which the appellee may proceed to discharge its obligation to investigate the reasonableness of any rate application. We cannot conclude that the use of certain forms and the requirement of particular data with a rate application establishes intractably the methodology or means by which the appellee is to pursue the exercise of the authority delegated it by the legislature.
The appellant’s argument that the modified balance sheet approach was utilized in its rate case without sufficient notice is likewise without merit. The record reflects that the appellee’s staff used the modified balance sheet approach as a check on appellant’s lead/lag study and, when irreconcilable discrepancies resulted, sought to recalculate appellant’s working capital allowance. The record reflects that the appellee’s staff’s use of the modified balance sheet approach and the staff’s problems with appellant’s lead/lag study were manifested more than a month prior to hearing when the appellee’s staff filed its testimony. Appellant had ample opportunity to dispute the appellee’s staffs position and in fact cross-examined the staff extensively on the modified balance sheet approach.
The appellee adopted the appellee’s staffs recommendation as to appellant’s rate of return. Appellant sought a return on equity of 16.25 % and an overall rate of return of 11.37 %. The staff recommended a return on equity of 12.13 % and an overall rate of return on rate base of 8.78%, using what is known as “double leverage.” Appellant claims that these figures are not supported by substantial evidence, primarily because they do not take into account appellant’s financial risk.
Appellant is a wholly-owned subsidiary of General Telephone and Electric (GTE), a fact which makes calculation of the required return on equity for the subsidiary more difficult because there exists no public market for its stock. A commonly-used and widely recognized principle known as “double leverage” is used in situations such as this to determine the cost of a subsidiary corporation’s equity. This theory holds that the cost of capital for a parent corporation equals the cost of equity for its subsidiary. Appellee’s staff witness Kilburn and Attorney General witness Wilson used the double leverage approach, and the appellee argues that, without recognizing double leverage, appellant’s equity return would be 19.8%.
The first step in using double leverage is to calculate the parent corporation’s cost of capital. Staff witness Kilburn used the discounted cash flow method (DCF)
to obtain GTE’s cost of equity. Kilburn followed tried and true methods of applying the DCF method, utilizing market-specific information as to GTE within the DCF formula to determine GTE’s cost of equity. Kilburn also used a comparative sample of nine major independent telephone companies and again estimated the proper return on equity according.to the DCF model. Kilburn found the cost of equity for GTE (the parent corporation) to fall within the range of 13.10% to 14.14%, with a midpoint of 13.62%. To this was added a flotation cost of .03 %
from which she arrived at a cost of equity for GTE of 13.65 %. In determining the weighted cost of capital for GTE, Kilburn arrived at 12.13%, which cost was imputed to appellant as its cost of capital under the double leverage theory.
The concept of double leverage has been recognized by the Arkansas Supreme Court on several occasions. General Telephone Co. v. Arkansas Public Service Commission, 272 Ark. 440, 616 S.W.2d 1 (1981); Arkansas Public Service Commission v. Lincoln-Desha Telephone Co., 271 Ark. 346, 609 S.W.2d 20 (1980); Southwestern Bell Telephone Co. v. Arkansas Public Service Commission, 267 Ark. 550, 593 S.W.2d 434 (1980). In Arkansas Public Service Commission v. Lincoln-Desha Telephone Co., supra, the Arkansas Supreme Court said:
Corporations are usually financed partly with debt capital and partly with equity capital. “Leverage” is a financial term used to describe the situation in which a corporation is funded by debt in addition to the equity supplied by the stockholders. A corporation is said to be “leveraged” to the extent that debt is included in its capital structure. The leverage arises from the advantage gained by equity holders through the rental of capital at a lower rate than the return they receive on their equity. Thus, we see that by use of leverage the equity owners are able to earn an overall rate of return in excess of the cost of capital. The added earnings above the cost inure to the benefit of the stockholders as they then receive a higher rate of return than if the institution had been financed entirely by equity.
271 Ark. at 348-49, 609 S.W.2d at 22. In that same case, our supreme court stated that “double leverage is merely an extension of the concept of leverage to a parent-subsidiary corporate relationship.” Id. at 348, 609 S.W.2d at 22, citing New England Telephone and Telegraph Co. v. Public Utilities Commission, 390 A.2d 8 (Me. 1978).
As earlier noted, appellant claims that the use of double leverage does not properly account for the financial risk inherent in appellant. Risk is not particularly quantifiable, but rather is a matter of discretion on the part of the appellee in the application of its expertise. The evaluation of financial risk and what constitutes a “fair return” is a judgment call involving a good bit of educated speculation at best, in light of the evidence from expert witnesses. While appellant makes some appealing arguments relative to the DCF calculations made by staff witness Kilburn, we cannot say that the appellee’s finding that the market place accounts for all risk at the parent corporation level is not supported by substantial evidence. Simply put, the appellee chose the testimony of one witness over that of another as to the rate of return in this case.
Before proceeding with the remaining issues, we address the last issue raised by appellant in its supplemental brief, which pertains to whether the appellee’s staff may properly seek rehearing before the appellee. Arkansas Statutes Annotated Section 73-229.1 (Supp. 1985) provides that “[a]ny party to a proceeding before the Commission aggrieved by an order issued by the Commission may apply for a rehearing within thirty (30) days after the service of such order.” Appellant argues that the appellee erred in granting a rehearing to the appellee’s staff because, under Section 73-229.1, the appellee’s staff is not and cannot be an “aggrieved” party. The applicable statutes do not spell out whether the appellee’s staff may be treated as a party for purposes of such rehearings. The appellee, however, through its Rules of Practice and Procedure, has required that the staff be bound by and conform to the appellee’s rules as a party in proceedings before the appellee.
An agency’s or department’s interpretation of its own rules and regulations is not binding upon the courts but is nevertheless highly persuasive; the agency’s interpretation of its own rules is controlling unless plainly erroneous or inconsistent. Boone County v. Apex of Arkansas, Inc., 288 Ark. 152, 155, 702 S.W.2d 795 (1986); Clinton v. Rehab Hospital Services Corp., 285 Ark. 393, 688 S.W.2d 272 (1985). In addition, appellant did not question the status of the appellee’s staff as a party to the proceedings until after an order granting the staff’s petition for rehearing had been entered.
Appellant is a member of what is known as the Arkansas IntraLata Toll Pool (AITP). The toll pool serves as a clearing house for revenues generated in the form of charges for toll telephone service, which is primarily long distance. The pool divides up revenues it collects periodically among the various telephone companies which are its members, including an amount of money representative of a rate of return earned by the pool. Depending upon the rate of return the toll pool earns, the member companies receive a share of the pool’s profits which come into play “above the line” in the revenue requirement calculation. The intralata toll pool rate of return is important because it results in the inclusion of a particular dollar figure in appellant’s revenue requirement which can serve to offset the revenue requirement which must be recovered from ratepayers in their basic rates. The lower the toll pool rate of return, the more revenues that must be recouped from basic rates; the higher the toll pool rate of return, the lower the amounts that must be recovered in basic rates.
In its original order, the appellee adopted an intralata toll pool rate of return (10.08%) it had established in a previous docket as the proper target toll pool rate of return for appellant, which results in revenues of $11,503,710.00 being imputed into appellant’s revenue requirement calculation. Appellant had originally requested a 5.21 % toll pool rate of return and subsequently amended that figure upward to 9.5%. Upon rehearing, the realized toll pool rate of return for the third quarter of 1985 was introduced, which showed that the actual rate of return for the third three-month period of 1985 was 11.3889%. After rehearing, that figure was adopted, and the original order was modified to reflect the 11.3889% figure instead of 10.08% as the estimated intralata toll pool revenue requirement to be applied in calculating rates for appellant as a result of their rate application.
Appellant does not argue here about 10.08 % being an inappropriate target rate of return. Appellant complains, however, that the appellee’s adoption of a higher target rate of return upon rehearing, based upon the actual performance of the toll pool during the third quarter of 1985 is not supported by substantial evidence. The effect of the appellee’s revision of the target toll pool rate of return from 10.08% to 11.3889% results in more money being imputed from the toll pool to appellant’s revenue requirement, thereby lowering the revenue requirement which must be recovered from ratepayers in the form of basic rates. Since witnesses for both sides below testified that actual results for the third quarter of 1985 would be the only rate of return data which would not be estimates and would be desirable from an accuracy standpoint, we cannot say the appellee’s adoption of third quarter results is arbitrary or not supported by substantial evidence.
In this case, appellant sought an extension of the statutory deadline in order to allow it to gather fourth quarter 1985 intralata toll pool rate of return information. There is a ten-month statutory deadline established by the General Assembly within which rate cases must be decided. Both parties argue that resolution of the issue as to whether the appellee should have extended the statutory deadline to allow introduction of fourth quarter 1985 toll pool results would turn on an inquiry as to whether the appellee abused its discretion in extending the statutory deadline or in refusing to do so.
Arkansas Statutes Annotated Section 73-217 (Supp. 1985) provides that the appellee must act within ten months of the filing of the rate application. The legislature enacted this limitation to reduce regulatory delay and to induce some certainty into the ratemaking process. We do not agree, however, that the statutory deadlines provided in Ark. Stat. Ann. Section 73-217 may be waived. In Southwestern Bell Telephone Co. v. Arkansas Public Service Commission, 267 Ark. 550, 559, 593 S.W.2d 434, 440 (1980), our supreme court said:
The time limit was imposed by the General Assembly. It must be remembered that the PSC is a creature of the legislature and that, in rate-making, it is performing a legislative function, which has been delegated to it. City of Ft. Smith v. Department of Public Utilities, 195 Ark. 513, 113 S.W.2d 100; Arkansas Power & Light Co. v. Arkansas Public Service Com’n., 226 Ark. 225, 289 S.W.2d 668. The commission was created to act for the General Assembly and it has the same power that body would have when acting within the powers conferred upon it by legislative act. Department of Public Utilities v. Arkansas Louisiana Gas Co., 200 Ark. 983, 142 S.W.2d 213. The General Assembly certainly has not surrendered the power to fix time limitations on the actions of its own agency.
We hold this language applicable in this case. The Arkansas General Assembly has not surrendered to the appellee the power to fix time limitations in rate cases, and, consequently, the time limitation provisions of Section 73-217 may not be waived or disregarded by any party or by the appellee itself. We cannot conclude that the action of the Commission was arbitrary.
Finally, appellant contends that giving effect to certain cost-saving projects
was improper in light of Section 73-217.5 because no savings would be realized from those projects during the pro forma year, i.e., the twelve months following the close of the rate case test year. Arkansas Statutes Annotated Section 73-217.5 (Supp. 1985) reads in full as follows:
For the purpose of justifying the reasonableness of the proposed new rate schedule or rate schedules, a utility may utilize either a historical test period of 12 consecutive calendar months, or a forward looking test period of 12 consecutive calendar months consisting of 6 months of projected data which together shall be the period or test year upon which fair and reasonable rates shall be determined by the Commission. Provided, however, that the Commission shall also permit adjustments to any test year so utilized to reflect the effects on an annualized basis of any and all changes in circumstances which may occur within 12 months after the end of such test year where such changes are both reasonably known and measurable.
The test year ended December 31, 1984, and the record shows (and the appellee found) that several of appellant’s witnesses testified that the savings projects would be implemented during the pro forma year. The appellee’s staff, using data supplied by appellant, annualized the savings to account for those which could be realized from the projects during the pro forma year. Appellant contends that, even though the projects may have been implemented during the pro forma year, savings realized therefrom are not reasonably known and measurable. We cannot quarrel with the appellee’s finding of fact based upon evidence presented by appellee’s staff which was based upon data supplied by appellant. We cannot conclude that it was arbitrary or not supported by substantial evidence.
The decision of the appellee is affirmed in all respects.