On 27 March 1986, Duke Power Company (“Duke”) filed an application with the North Carolina Utilities Commission (“Commission”) seeking authority to increase annual electric revenues for its North Carolina retail customers by 14.7%, or $289,316,000. On 14 April 1986, the Commission declared Duke’s application to be a general rate case, established the test period, and scheduled public hearings. Thereafter, the Attorney General, North Carolina Industrial Energy Consumers, the Commission’s Public Staff, Carolina Utility Customers Association and Wells Eddleman intervened. On 31 October 1986, the Commission, Chairman Wells and Commissioner Cook dissenting in part, issued its Order Granting Partial Rate Increase, from which these appeals are taken, and which authorized Duke to increase its revenues for North Carolina retail customers by $133,080,000.
On intervenors’ appeals the questions presented are whether the Commission erred in: (1) reaffirming its decision to allow Duke *691to recover, in rates charged to its customers, costs expended on its abandoned Perkins and Cherokee nuclear power stations; (2) deciding that 13.4% is a fair rate of return on Duke’s common equity; (3) adopting Duke’s actual capital structure as it existed on 31 July 1986 as a proper capital structure for rate-making purposes; and (4) failing to reduce Duke’s equity capital for rate-making purposes by deducting from it Duke’s investment in certain of its wholly owned, nonregulated subsidiaries. We hold that the Commission’s conclusion regarding the appropriate rate of return on common equity is not supported by adequate factual findings, and we remand for further proceedings on this question. We affirm the Commission’s decision on all other questions presented.
I.
The Attorney General, the City of Durham, and Wells Eddleman contend the Commission erred in reaffirming its earlier decisions to allow Duke to recover from its ratepayers costs incurred in its now cancelled Cherokee and Perkins nuclear power station projects. Duke responds that our earlier decision, State ex rel Utilities Comm. v. Eddleman, 320 N.C. 344, 358 S.E. 2d 339 (1987), means this matter is res judicata and appellants are therefore barred from pressing this contention in the instant case. We agree with Duke and decline to address the merits of appellants’ arguments.
In Eddleman we considered whether the Commission improperly allowed Duke to recover costs incurred in the construction of its abandoned Cherokee and Perkins nuclear power stations. Duke began constructing these stations in the mid-1970’s in order to accommodate then predicted increases in electricity consumption. When consumption increases were not commensurate with these predictions, Duke’s Board of Directors terminated construction of both plants. Duke sought, nevertheless, to recover in its rates costs, amortized over periods of years, incurred in the construction of these abandoned plants. The Commission, as it had done in the past without challenge on appeal, decided to permit this procedure. The Attorney General, the City of Durham, and Wells Eddleman appealed to this Court, contending for various reasons that the Commission’s decision was legally impermissible. The Court, one Justice not participating, affirmed the Commission’s *692decision by an evenly divided vote. Concerning the effect of our decision we declared, “following the uniform practice of this Court, the decision of the Utilities Commission is affirmed, not as precedent but as the decision in this case.” Id. at 380, 358 S.E. 2d at 362.
The doctrine of res judicata treats a final judgment as the full measure of relief to be accorded between the same parties on the same “claim” or “cause of action.” C. Wright, Federal Practice and Procedure, § 4402 (1969). “The essential elements of res judicata are: (1) a final judgment on the merits in an earlier suit, (2) an identity of the cause of action in both the earlier and the later suit, and (3) an identity of parties or their privies in the two suits.” Hogan v. Cone Mills Corporation, 315 N.C. 127, 135, 337 S.E. 2d 477, 482 (1986).
Applying the law enunciated in Hogan to the facts of the present case, we conclude, as a result of our Eddleman decision, that appellants are barred by the doctrine of res judicata from reasserting their claim that the Commission’s treatment of the costs of the abandoned Perkins and Cherokee power stations is legally impermissible. Here and in Eddleman the parties and the claims are identical: the Attorney General, the City of Durham, and Wells Eddleman contend that the Commission erred in allowing Duke to include in its rates, on an amortized basis, costs incurred in the construction of its abandoned power plants. A final judgment on the merits of this claim was reached in Eddleman. Although our evenly divided decision had no precedential value, it operated, nevertheless, as the final decision in the case on this claim as to these parties.
Appellants contend that because the Court in Eddleman was evenly divided there was no final judgment on the merits of their claim. This argument stands at odds with our decision in Seay v. Insurance Company, 213 N.C. 660, 197 S.E. 151 (1935). In Seay we held that when a judgment is affirmed on appeal because of an evenly divided Court, the lower decision becomes the law of the case and is determinative of the rights of the parties with regard to the fully litigated claim. Id. at 661, 197 S.E. at 152. Seay involved a claim by an insurance agent to recover commissions he contended the defendant company owed him. In the initial action the superior court reversed a judgment of nonsuit entered *693against the agent in municipal court. Defendant appealed, and this Court affirmed the superior court by an evenly divided vote. The agent brought a subsequent suit on the same claim against the insurance company’s successor in interest, and again a judgment of nonsuit was entered against him. This Court reversed, holding that our earlier decision to reverse the judgment of nonsuit was determinative of the rights of the parties in any subsequent action on the same contract. Id. at 661, 197 S.E. at 152.
Seay controls the present case on the res judicata point. While our decision in Eddleman to affirm the Commission has no precedential value, it does finally determine the rights of the parties in that litigation on the abandoned plant cost issue. Since those parties and that issue are the same as in the instant case, those parties may not here relitigate that issue.
II.
The Public Staff argues the Commission’s conclusion that 13.4% is a fair rate of return on Duke’s common equity is not supported by adequate findings of material facts. We agree and remand for further proceedings on this issue consistent with this opinion.
We note at the outset that the Commission has labeled its determination that 13.4% is a fair rate of return on common equity a “finding.” This, of course, does not make it so. What constitutes a fair rate of return on equity, as the Commission in other parts of its order recognizes, is ultimately a matter of judgment. Matters of judgment are not factual; they are conclusory and based ultimately on various factual considerations. Facts are things in space and time that can be objectively ascertained by one or more of the five senses or by mathematical calculation. Facts, in turn, provide the bases for conclusions. State ex rel Utilities Comm. v. Eddleman, 320 N.C. at 351, 358 S.E. 2d at 346. What constitutes a fair rate of return on common equity is a conclusion of law which must, in turn, be predicated on adequate factual findings.
All parties relied principally on what is known as “discount flow methodology” (“DCF”) to determine the appropriate common equity rate of return. According to this method the proper rate of return is determined by adding to the common stock’s current *694yield a rate of increase which investors will expect to occur over time. C. F. Phillips, Jr., The Regulation of Public Utilities 356-57 (1984). While the use of DCF methodology presents some difficulties, especially in determining investor expectations, the parties in the present case agreed that a Duke-specific DCF is the best method for determining Duke’s rate of return on common equity.
Duke witness Dr. Charles E. Olson performed a Duke-specific DCF study which, standing alone, suggested a return requirement of 11.9% to 12.4%. He “checked” the results of his study by performing a DCF study on a group of electric utilities comparable in risk to Duke. This study suggested a return requirement of 12.5% to 13.0%. Dr. Olson performed another check — a “risk premium study” — which suggested a return requirement of 13.75%. He acknowledged that the risk premium method is not as accurate as the DCF method. Dr. Olson’s ultimate recommendation was for a 13.5% to 14.0% return on common equity.
Dr. Olson testified that in arriving at his recommendation he made two upward adjustments to the return requirement suggested by his Duke-specific DCF. The first adjustment was for the purpose of reimbursing Duke for the costs of issuing common stock in the future. Such costs are known as “flotation,” or “financing,” costs. The second adjustment was made to protect Duke’s investors against dilution of their investment should Duke be required, during unfavorable market conditions, to issue stock below book value. Dr. Olson’s financing cost adjustment would add one-half of one percent (.5%) and his “down market” adjustment, an additional .5% to the rate of return on Duke’s common equity otherwise suggested by his Duke-specific DCF. Both adjustments, when added to the rate of return otherwise suggested by Dr. Olson’s Duke-specific DCF, result in a rate of return on common equity of 13.4%. Each adjustment translates into a cost of $21.2 million annually to Duke’s North Carolina ratepayers. Together they add $42.4 million annually to these ratepayers’ electric bills.
Public Staff witness George T. Sessoms testified that he performed DCF studies which suggested that a 12.3% return on Duke’s common equity would be proper. He arrived at this figure by performing a Duke-specific DCF and a DCF analysis of a group *695of electric utility companies with risks similar to Duke’s. The Duke-specific study indicated an investor return of 11.5°/o to 12.3%; and the study of the comparable group, 12.0% to 12.9%. From these ranges, Sessoms concluded that Duke’s return requirement on its common equity should be 12.2%, only .2% less than suggested by Dr. Olson’s similar, unadjusted, Duke-specific DCF. Sessoms then added .1% to this figure to compensate for those financing costs which, from Duke’s past history of common stock issues, might be reasonably anticipated to occur in the future.
Attorney General witness Dr. John W. Wilson recommended an 11% rate of return on Duke’s common equity. He based his conclusion on a DCF study which employs a regression and correlation analysis of the historical growth rate of Duke and 79 other electric companies. Dr. Wilson made no adjustment for financing costs or down markets.
The Commission concluded that Duke should have the opportunity to earn 13.4% on its common equity. This is precisely the return suggested by Dr. Olson’s upwardly adjusted Duke-specific DCF study. To support this conclusion the Commission recited the testimonies of witnesses Olson, Sessoms and Wilson. It highlighted the problems inherent in the DCF method of determining the required rate of return on common equity, emphasizing that the volatility characterizing then current stock market conditions may have skewed the witnesses’ ultimate recommendations. The Commission noted that in the last general rate case Duke was allowed a 14.9% common equity return. It went on to declare that
the rate of return on common equity of 14.0% requested by the company is excessive, while the rates of return on common equity of 12.3% and 11% recommended by the Public Staff and the Attorney General, respectively, are too conservative and stringent and would severely handicap the Company in continuing to provide adequate and reasonably priced electric service to its customers.
The Commission did not state whether the 13.4% return included any adjustment for down markets. It did note that return included some adjustment for financing costs, but it did not quantify this adjustment.
*696Commission Chairman Robert O. Wells, joined by Commissioner Ruth E. Cook, dissented from the Commission’s decision to allow Duke a 13.4% return on its common equity. Chairman Wells expressed the view that the majority improperly included within the 13.4% return a .5% adjustment for financing costs and a .5% adjustment for down markets. Concerning the financing cost adjustment he noted:
Duke issued new shares of common equity five times over the entire 10-year period of 1975-1985. The total cost of issuance was $16.1 million for an average cost per issue of $3.2 million. To permit Duke to collect $21.2 million annually to cover Dr. Olson’s flotation cost fiction is totally unwarranted. However, . . . such a result is . . . implicit in the Majority having allowed Duke a 13.4% return on common equity. Mr. Sessoms added only one-tenth of one percent (.1%) for flotation costs. A .1% flotation cost adjustment will provide annual revenues of $4.2 million on a North Carolina retail basis. Such a sum will more than compensate investors for the costs of issuance of new common stock. Furthermore, the evidence is that Duke will not issue new stock in the foreseeable future. Therefore, whatever allowance is made for flotation cost will, in the foreseeable future, compensate investors for a cost Duke will not incur.
Regarding an adjustment for down markets Chairman Wells wrote:
[I]t is not the responsibility of this Commission, or of Duke ratepayers, to protect investors from swings in market price. The cost of Dr. Olson’s down market adjustment to Duke’s North Carolina retail ratepayers is another $21.2 million annually. The Majority, by allowing a 13.4% return on common equity, has in effect adopted a major portion of this down market adjustment. Again I note that Duke does not expect to issue any new common stock for the next three or four years. Indeed, the record reveals that Duke presently has surplus cash in excess of $400 million which could be used for capital expansion if needed.
The guidelines for determining a utility’s rate of return in a general rate case are set forth in N.C.G.S. § 62433(b)(4). This statute provides that the Commission shall fix an overall rate of *697return on the cost of a utility’s property as determined by subsection (b)(1) that will (1) enable a well-managed utility to produce a fair return for its shareholders, (2) allow the utility to maintain its facilities and services at a reasonable level, and (3) enable the utility to compete in the market for capital funds on terms that are reasonable and fair to its customers as well as its existing investors. See N.C.G.S. § 62-133(b)(4) (Cum. Supp. 1987). We have interpreted this statute to mean that
[T]he Legislature intended for the Commission to fix rates as low as may be reasonably consistent with the requirements of the Due Process Clause of the Fourteenth Amendment to the Constitution of the United States, those of the State Constitution, Art. I, Sec. 19, being the same in this respect.
Utilities Comm. v. Power Co., 285 N.C. 377, 388, 206 S.E. 2d 269, 276 (1974).
The overall rate of return is figured by combining the rates of return permitted on each form of capital accumulation, here long-term debt, preferred equity and common equity. In combining the rates of return each is weighted according to the ratio among the capital components of the company’s capital structure, here 42.9% long-term debt, 10.8% preferred equity and 46.3% common equity. Here the proper rates of return were determined to be 8.91% for Duke’s long-term debt and 8.27% for its preferred equity. When these are combined with the 13.4% return on common equity allowed by the Commission and all are weighted according to the ratios among all three forms of capital, the overall rate of return becomes 10.92%. It is this overall rate of return when applied to the rate base as determined by N.C.G.S. § 62-133 (b)(1) that produces the utility’s required revenues.
Proper rates of return on debt and preferred equity are relatively easy to determine because they represent returns which, in effect, have been guaranteed to those who have furnished these forms of capital. These rates of return are sometimes called “imbedded costs.” The proper rate of return on common equity is always the most difficult to determine and becomes, as we have noted, essentially a matter of judgment based on a number of factual considerations which vary from case to case.
The proper rate of return on common equity is here an extremely important determination because, as can be seen, it is the *698most expensive form of capital accumulation, which expense is ultimately borne by the ratepayer, and it is the most heavily weighted in arriving at the overall return. It is important that a reviewing court be able to determine the factual underpinnings upon which the Commission’s conclusion on this rate of return rests.
N.C.G.S. § 62-94 prescribes the scope of appellate review of a decision by the Utilities Commission. According to this standard, the reviewing court
(b) . . . may reverse or modify the decision if the substantial rights of the appellants have been prejudiced because the Commission’s findings, inferences, conclusions or decisions are:
(1) In violation of constitutional provisions, or
(2) In excess of statutory authority or jurisdiction of the Commission, or
(3) Made upon unlawful proceedings, or
(4) Affected by other errors of law, or
(5) Unsupported by competent, material and substantial evidence in view of the entire record as submitted, or
(6) Arbitrary or capricious.
(c) In making the foregoing determinations, the court shall review the whole record or such portions thereof as may be cited by any party and due account shall be taken of the rule of prejudicial error.
N.C.G.S. § 62-94 (1982).
In order to facilitate appellate review, the Commission must comply with N.C.G.S. § 62-79(a), which provides that
(a) All final orders and decisions of the Commission shall be sufficient in detail to enable the court on appeal to determine the controverted questions presented in the proceedings and shall include:
(1) Findings and conclusions and the reasons or bases therefor upon all the material issues of fact, law, or discretion presented in the record, and
*699(2) The appropriate rule, order, sanction, relief or statement of denial thereof.
N.C.G.S. § 62-79(a) (1982). “The failure to include all the necessary findings of fact is an error of law and a basis for remand upon N.C.G.S. § 62-94(b)(4) because it frustrates appellate review.” State ex rel Utilities Comm. v. The Public Staff, 317 N.C. 26, 34, 343 S.E. 2d 898, 904 (1986).
With these principles in mind, we hold the Commission did not comply with N.C.G.S. § 62-79(a) because it failed to include material factual findings sufficient in detail to permit meaningful appellate review of its conclusion that 13.4% is a fair return on common equity.
First, we note that the Commission’s approved rate of return coincides precisely with Dr. Olson’s testimony as to the proper return suggested by his Duke-specific DCF study as he adjusted it to protect Duke’s investors against down markets and to compensate for flotation costs. But the Commission made no findings as to whether it considered protecting Duke’s investors against down markets in setting a proper rate of return on common equity, and if so, the extent to which this factor was employed.
Whether and the extent to which the Commission utilized such a factor is material to its rate of return conclusion. There is no evidence in this record that Duke proposes or can be reasonably expected to issue common stock under market conditions which would cause the value of its outstanding stock to fall. At least in the absence of this kind of evidence a rate of return based, in part, on such a projected phenomenon would be improper. We agree with Chairman Wells that ordinarily “it is not the responsibility of the ratepayers to protect investors from swings in the marketplace.” Investors understand, and take the risk, that Duke might possibly at some time because of market conditions be required to issue shares at less than book value. But there is nothing in this record to show that such an event is contemplated or is a probability. Without such evidence we see no reason, nor has the Commission suggested any, for shifting the risk of possible stock issues in down markets from shareholders to ratepayers.
To attract capital, a utility does not need to charge, and is not entitled to charge, for its services rates which will make *700its shares . . . attractive to investors who are willing to risk substantial loss of principal in return for the possibility of abnormally high earnings. The reason is the utility, having a legal monopoly in an essential service, offers its investors a minimal risk of loss of principal.
Utilities Comm. v. Telephone Co., 281 N.C. 318, 337-38, 189 S.E. 2d 705, 718 (1972).
Second, the Commission acknowledges that the rate of return is designed to compensate for financing costs. It states in its order, “[t]he rate of return on common equity . . . includes an adjustment to allow for reasonable stock or issuance financing costs for the reasons generally stated by witnesses Olson and Sessoms in this case.” But the Commission failed to quantify this factor, or to specify the extent to which this factor affected the ultimate rate of return approved. This again is a missing material factual finding. Because of its absence we are unable to say whether the Commission erred in its rate of return decision.
Since no evidence was introduced that Duke intends to issue new stock for the next three or four years, and because there was no evidence regarding the probable cost of a prospective issuance, we question whether the record supports any financing cost adjustment. On the other hand, the record reveals that both Dr. Olson, testifying for Duke, and Mr. Sessoms, for the Public Staff, adjusted the rate of return derived from their DCF studies to account for future financing costs. The adjustments they suggested, however, .5% and .1% respectively, differ widely and amount to a significant difference to ratepayers. A .5% financing cost adjustment will increase rates by $21.2 million annually, while a .1% increase will cost ratepayers only $4.2 million annually.
We find nothing in the record which supports a $21.2 million annual adjustment for financing costs. The record shows that over the entire period of 1975-1985 the total cost of Duke’s new stock issues was only $16.1 million. The average cost per issue was approximately $3.2 million. The only reference in the record to Duke’s plans to issue stock in the future was the statement of its Chairman, Mr. Lee, that the company’s “present expectation is that we will be back into the capital markets for new funds in about three to four years.”
*701On the basis of this evidence we agree with Chairman Wells that since the .1% financing costs adjustment suggested by Mr. Sessoms will provide annual revenues of $4.2 million, it will “more than compensate investors for the cost of issuance of new common stock” presently contemplated by Duke. On the other hand, the .5% financing costs adjustment recommended by Dr. Olson would be, on this record, grossly extravagant and not justified.
Thus, even if the record supports some adjustment for financing costs, the extent of that adjustment is of considerable importance to ratepayers. It is a material fact which should have been specifically found by the Commission in order to permit meaningful appellate review of the Commission’s ultimate rate of return decision.
The Commission, then, was required to make specific findings showing what effect, if any, it gave to financing costs or down market protection, or both, in arriving at its common equity rate of return decision. See Utilities Comm. v. Telephone Co., 281 N.C. 318, 361, 189 S.E. 2d 705, 732-33 (1972). Failure to do so constitutes an error of law requiring a remand for further proceedings. Id. at 365, 198 S.E. 2d at 735.
On remand the Commission is directed to reconsider the proper rate of return on Duke’s common equity in light of this opinion. The Commission is directed further to support its conclusion on this issue with specific findings as to its treatment of financing costs and down market protection. The Commission may make such other findings of material facts in support of its conclusion on this issue as it deems appropriate.
III.
The Public Staff contends the Commission’s decision that Duke’s capital structure should include a common equity ratio of 46.3% is improper because it is not supported by substantial evidence. We disagree.
The ratios among common equity, preferred stock, and long-term debt used for rate-making purposes are important because of the relative expense to the utility of each form of capital accumulation. Common equity investors demand the greatest return; so, for the utility, this form of capital accumulation is, in terms of current expense, the most expensive of the three. The *702higher the common equity ratio, the greater the rates which must be charged to cover the expense of providing the higher return demanded by common equity investors.
Evidence presented by Duke tended to show that its capital structure should be set so that the company could maintain between 45.0% and 50.0% common equity. Mr. William S. Lee and Mr. William R. Stimart testified that Standard and Poor’s had recommended such a common equity ratio for AA rated utilities. Duke’s application was for a capital structure based on its actual, per book capital structure as of 31 December 1985, consisting of 46.18% common equity, 10.97% preferred stock, and 42.85% long-term debt. In rebuttal, Mr. Stimart recommended that the Commission utilize Duke’s actual capital structure as of 30 June 1986. This structure consisted of 46.49% common equity, 10.75% preferred stock, and 42.76% long-term debt. On cross-examination Mr. Stimart testified that Duke’s capital structure as of 31 July 1986 included a common equity ratio of 46.3%.
Public Staff witness Sessoms testified that the capital structure requested by Duke was too conservative for rate-making purposes. According to Sessoms, Duke’s capital structure was conservative in relation to the average capital structure of other publicly traded AA rated electric utility companies. He recommended a hypothetical capital structure of 45.0% common equity, 11.0% preferred stock, and 44.0% long-term debt.
Attorney General witness Dr. Wilson recommended ratios of 42.17% for common equity, 11.17% for preferred stock and 46.66% for long-term debt. Dr. Wilson arrived at these figures by adjusting Duke’s capital structure to omit what he considered to be equity invested in Duke’s nonregulated subsidiaries, Mill Power Supply Company, Church Street Capital Corporation, Cresent Land and Timber, and Eastover Mining Company. Dr. Wilson testified that since these nonregulated enterprises produce an equity return for Duke, Duke’s electric utility ratepayers should not be forced to subsidize them by including in Duke’s capital structure for rate-making purposes Duke’s equity investments in them.
The Commission concluded that Duke’s actual capital structure as of 31 July 1986 represented an appropriate structure for this rate-making proceeding. This capital structure is composed of *70342.9% long-term debt, 10.8% preferred stock, and 46.3% common equity. The Commission supported this conclusion by finding that the recommendations of the witnesses for the Attorney General and the Public Staff
would reduce Duke’s common equity ratio below that which was approved in Duke’s last general rate case; i.e., 45.52%. The evidence in this case does not support such a reduction. Mr. Lee, Dr. Olson, and Dr. Stimart all testified that S & P is increasing rather than decreasing the requirements for AA utilities and Dr. Olson testified that the tax reform act will have the effect of reducing Duke’s fixed charge coverage ratio. The evidence in this case supports an increase rather than a decrease in Duke’s common ratio.
Based on considerations such as these the Commission decided that the capital structure as of 31 July 1986 was “within the zone of reasonableness” according to the evidence presented. The Commission stated finally that while it considered the capital structure of 31 July 1986 appropriate in this case it was concerned about Duke’s increasing equity percentage. The Commission concluded that it “believes that it is appropriate to place Duke on notice that the Company’s actual capital structure will be closely scrutinized and examined for rate-making purposes in future general rate cases.”
Chairman Wells dissented from the majority’s decision concerning capital structure. He thought the majority failed to comply with its responsibility to fix rates as low as reasonably consistent with due process.
Turning again to the standard of review established in N.C. G.S. § 62-94, we must determine “whether there is substantial evidence, in view of the entire record, to support” the Commission’s findings which, in turn, support its conclusion on the appropriate capital ratio for rate-making purposes. State ex rel. Utilities Comm. v. Eddleman, 320 N.C. at 355, 358 S.E. 2d at 339. Our function is not to conduct a review de novo. “A reviewing court may not modify or reverse [the Commission’s] determination merely because the court would have reached a different result based on the evidence.” State v. The Public Staff, 317 N.C. at 34, 343 S.E. 2d at 903-04. On the contrary, a Commission’s order
*704will not be disturbed if upon consideration of the entire record we find the decision is not affected by error of law and the facts found by the Commission are supported by competent, material and substantial evidence, taking into account any contradictory evidence or evidence from which conflicting inferences could be drawn.
Utilities Commission v. Carolina Utilities Customers Assn., 314 N.C. 171, 179-80, 333 S.E. 2d 259, 265 (1985).
We find no error in the Commission’s decision concerning Duke’s capital structure. The testimony of Duke’s witnesses Lee and Stimart supports the Commission’s finding on changed economic conditions which, in turn, supports its conclusion that an increase, rather than a decrease, in the percentage of common equity from Duke’s last rate proceeding is justified.
IV.
The Attorney General argues that the Commission’s decision regarding capital structure is improper because it includes equity capital which Duke has invested in certain of its wholly owned, nonregulated subsidiaries. He argues that to permit inclusion of equity capital Duke has invested in its nonregulated subsidiaries in determining the equity capital ratio for rate-making purposes is unlawful because (1) it permits the utility to earn a return on property not actually used in producing electricity in violation of N.C.G.S. § 62-133(b)(1) and (4); and (2) it is arbitrary and capricious in violation of N.C.G.S. § 62-94(b)(6) in light of the Commission’s decision in an earlier proceeding to exclude from the capital structure the common equity invested in two of Duke’s subsidiaries, Cresent Land & Timber Corporation and Mill Power Supply Company. See State ex rel. Utilities Comm. v. Eddleman, 320 N.C. 344, 381, 358 S.E. 2d 339, 362, n. 11. We disagree.
The flaw in the Attorney General’s argument, as we see it, is that it assumes that when Duke invests in a subsidiary company the invested proceeds are derived wholly from capital accumulated by the sale of common equity. This, of course, is not the case. As we have noted earlier, capital is derived from the sale not only of common equity but from the sale of preferred stock and bonds. When proceeds from capital accumulated from all three sources is invested elsewhere, the assumption must be that these proceeds *705are derived from each source of capital in the same ratio as each source bears to the other on Duke’s books. Thus, if any reduction in Duke’s capital structure is to be made for rate-making purposes because of Duke’s investment of some of its capital in nonregulated companies, the reduction must be made in each source of capital according to the ratio each source bears to the other. Such a reduction would effect no change in the ratios among the capital sources nor would it effect any change in the rate of return allowed for each component of capital.
To arrive at the level of income Duke is permitted to receive from its customers, the Commission first determines the utility’s rate base, which is the original cost of all the utility’s property used in producing electricity. It then allocates this rate base according to the ratio of its capital structure. Next it applies the rates of return allowed on each component of capital to that portion of the utility’s rate base allocated to that component. See generally, N.C.G.S. § 62-133 (Cum. Supp. 1987). Whatever capital accumulated from whatever source Duke might have invested in unregulated subsidiaries, these investments have no effect on Duke’s rate base upon which its permitted level of income is figured.
The result is that deducting from Duke’s capital structure its investments in unregulated subsidiaries would have no ultimate effect on the determination of the level of income Duke is entitled to receive from its customers.
In Eddleman the Commission did adjust the equity component of Duke’s capital structure by excluding Duke’s investment in certain of Duke’s nonregulated subsidiaries. Concerning this decision the Commission declared:
Duke’s capital structure should be adjusted to exclude the company’s equity investment of $24,076,000 in two of its nonregulated subsidiaries (Cresent Land and Timber Corp. and Mill Power Supply Co.), particularly in view of the fact that the company has itself removed $21 million of long-term debt supporting such nonregulated subsidiaries .... It would clearly be inconsistent to exclude only the long-term debt portion of Duke’s nonregulated investment in deriving the company’s appropriate capital structure for rate-making purposes.
*70669 PUR 4th 375, 452 (1985). In Eddleman it was necessary to adjust Duke’s equity component in order to compensate for Duke’s adjustment to its debt component. One adjustment necessitated the other in order to maintain an appropriate ratio between the two components for rate-making purposes.
In the present case there has been no adjustment to any capital component associated with Duke’s investment in unregulated subsidiaries. Thus, the ratio among the components of Duke’s capital structure remains constant. The rates of Duke’s electric customers are not affected by inclusion or exclusion of capital invested in subsidiaries so long as the ratio among the components of the capital structure remains constant.
This is not to say that the Commission cannot, or should not, take into consideration, among other things, Duke’s investment in unregulated subsidiaries in determining either an appropriate capital structure for rate-making purposes or an appropriate rate of return to be earned on Duke’s rate base. We hold only that the Commission is not required, as a matter of law, to reduce the common equity component of Duke’s capital structure by an amount equal to Duke’s investment in its nonregulated subsidiaries in determining the appropriate capital structure for rate-making purposes.
Summarizing, we hold the Commission erred only in failing to make sufficient material factual findings necessary to support its conclusion that 13.4% is a fair rate of return on common equity. This portion of the Commission’s decision is reversed and the matter is remanded to the Commission for further proceedings consistent with this opinion. As to the other issues brought forward in these appeals, the Commission’s decision is affirmed.
Affirmed in part; reversed in part and remanded.