Respondent New England Electric System (NEES), a holding company registered under § 5 of the Public Utility Holding Company Act of 1935,1 controls both an integrated electric utility system and an integrated gas utility system.2 Section 11(b) of the Act requires the Securities and Exchange Commission to limit the operations of a holding company system to a single integrated public utility system, except the Commission may permit the holding company to continue control of any additional integrated utility system that the Commission determines, among other things, "cannot be operated as an independent system without the loss of substantial economies which can be secured by the retention of control by such holding company of such system. * * *"3 In 1957 the Securities and Exchange Commission instituted proceedings to determine whether NEES should be permitted to retain control of both the electric and gas systems. The Commission initially found that the electric companies constituted a single integrated electric utility system, 38 S.E.C. 193 (1958), and NEES elected to retain those companies as its principal system. NEES urged, however, that it should also be permitted to retain the gas system. After extensive hearings, the Commission refused respondent permission to do so, and ordered the gas system divested. 41 S.E.C. 888 (1964).

In reaching its conclusion the Commission construed the statutory phrase "loss of substantial economies" in Clause A of § 11(b)(1) to require a showing that the "additional system cannot be operated under separate ownership without the loss of economies so important as to cause a serious impairment of that system." In its first review of the Commission's order, the Court of Appeals for the First Circuit held that the Commission had erroneously construed the statute; in the court's view, "loss of substantial economies" merely "called for a business judgment of what would be a significant loss. * * *" The court therefore set aside the Commission's order and remanded for reconsideration in light of that test. 346 F.2d 399, 406. We reversed, approving the Commission's construction, and remanded to the Court of Appeals for review of the challenged order in light of the proper meaning of the statutory term. SEC v. New England Electric System, 384 U.S. 176, 86 S.Ct. 1397, 16 L.Ed.2d 456 (NEES I). On remand, the Court of Appeals again set aside the Commission's order. 376 F.2d 107.4 That court, "after a fresh review of all the evidence," concluded "that the Commission's opinion does not reveal that application of both reason and experience to facts which merits endorsement as the responsible exercise of expertise." Id., at 111. We granted certiorari. 389 U.S. 816, 88 S.Ct. 81, 19 L.Ed.2d 65. We reverse and remand to the Court of Appeals with direction to enter a judgment affirming the Commission's order.

The question for our decision is whether the Court of Appeals properly held that, on the record, the Commission erred in finding that NEES failed to prove a case for retention of the integrated gas utility system. We address that question against the background of a congressional objective to protect consumer interests through the "elimination of 'restraint of free and independent competition.' * * * One of the evils that had resulted from control of utilities by holding companies was the retention in one system of both gas and electric properties and the favoring of one of these competing forms of energy over the other." NEES I, 384 U.S., at 183, 86 S.Ct. at 1401.5 Congress therefore ordained separate ownership—and divestiture where necessary to reduce holdings to one system—as the " 'very heart' of the Act." Id., at 180, 86 S.Ct. at 1399. Although Congress was aware that some economic loss might be suffered by the parent holding company or the separated integrated utility, Congress relented only to the extent of authorizing the Commission to permit retention of an additional integrated utility if that permission might be granted under the narrow exception provided by § 11(b)(1). But "retention of an 'additional' integrated system is decidedly the exception," and the burden is on the holding company to satisfy the "stringent test" set by the statute. Id., at 180, 182, 86 S.Ct. at 1399; cf. United States v. First City Nat. Bank, 386 U.s. 361, 366, 87 S.Ct. 1088, 1092, 18 L.Ed.2d 151.

Congress committed to the Commission the task of determining whether a holding company has met the burden of showing that its situation falls within the narrow exception under § 11(b)(1). The Clause A determination whether separation entails a loss of economies likely to cause a serious impairment of the system involves an element of prediction which necessarily calls for difficult and expert judgment. That judgment requires the assessment of many subtle and often intangible factors not easily expressed in precise or quantifiable terms. This is the very nature of economic forecasting. The task calls for expertise and is not simply "an exercise in counting commonplaces." United States v. Drum, 368 U.S. 370, 384, 82 S.Ct. 408, 415, 7 L.Ed.2d 360; see NEES I, 384 U.S., at 184-185, 86 S.Ct. at 1402-1403. Judicial review of that expert judgment is necessarily a limited one. See Gray v. Powell, 314 U.S. 402, 412-413, 62 S.Ct. 326, 332-333, 86 L.Ed. 301; NLRB v. Hearst Publications, 322 U.S. 111, 131, 64 S.Ct. 851, 860, 88 L.Ed. 1170; Atlantic Ref. Co. v. FTC, 381 U.S. 357, 367-368, 85 S.Ct. 1498, 1505-1506, 14 L.Ed.2d 443; United States v. Drum, supra, 368 U.S. at 375-376, 82 S.Ct. at 410-411. Congress expressly provided that "(t)he findings of the Commission as to the facts, if supported by substantial evidence, shall be conclusive." 15 U.S.C. § 79x(a); see Universal Camera Corp. v. NLRB, 340 U.S. 474, 71 S.Ct. 456, 95 L.Ed. 456; cf. NLRB v. Erie Resistor Corp., 373 U.S. 221, 236, 83 S.Ct. 1139, 1149, 10 L.Ed.2d 308. In our view, the Court of Appeals in this case indulged in an unwarranted incursion into the administrative domain.6 The Commission's order has adequate support in the record and should have been affirmed.

As of 1958, the test year selected for purposes of these proceedings,7 NEES' eight gas subsidiaries provided retail service to some 237,000 customers in a relatively compact 660-square-mile franchise area in Massachusetts. NEES' electric companies also served 75% of this area and about 78% of the gas customers were also electric customers. NEES' gross investment in gas plant and equipment was about $56,300,000 and gross gas revenues for 1958 were about $22,700,000. The eight gas companies were organized administratively as a Gas Division with centralized management, marketing and supply operations, and merchandising departments.8 The chief executive of the Gas Division was also president of each gas company and ultimately responsible to NEES' vice president in charge of management; in short, top management rested with executives having joint control over both electric and gas operations.

The Commission had before it a "severance study," a cost analysis and projection prepared for NEES by a professional public utilities management consulting firm, Ebasco Services, Inc. This study projected a loss of economies of approximately $1,100,000 annually for the gas system as the result of its separation from NEES. The Commission dealt with this study in alternative ways. It analyzed the study and concluded that "(t)he Ebasco estimate is inadequately supported in a number of important aspects and leaves considerable doubts which (NEES has) not satisfactorily overcome in the record." Then it went on to find that even if the estimated $1,100,000 in loss of economies were accepted as accurate "it would not lead us to conclude that such a loss is so substantial, when compared with the loss of economies involved in prior divestment cases and viewed in light of the objectives of the Act, as to warrant retention of the gas properties. * * *" 41 S.E.C., at 895, 897. Because we conclude that the record supports the Commission's decision on the latter ground, we have no occasion to consider whether the Commission's strictures on the reliability of the Ebasco study are well founded.

The Commission's ultimate finding that the projected $1,100,000 loss of economies annually did not constitute a loss of "substantial" economies within Clause A of § 11(b)(1) was reached primarily upon the basis of its subsidiary findings upon three matters: (1) That NEES' estimated losses were not significantly out of line with those found insubstantial in previous cases; (2) that other nonaffiliated Massachusetts gas companies,9 all but one of them smaller than the NEES gas system, are apparently able to operate successfully without electric utility affiliations; (3) that NEES did not establish that independent management devoted solely to promoting gas sales would not result in benefits to offset some of the projected losses. The Court of Appeals held that none of the three subsidiary findings was supported by substantial evidence. We disagree.

The Commission, consistent with its practice in prior cases,10 weighed NEES' estimated $1,100,000 losses in relative rather than absolute terms, calculating the losses as a percentage of NEES' 1958 revenues, expenses, and income.11 It found these loss ratios to be "lower or not significantly higher than corresponding ratios of gas systems whose divestment we have required on the ground that the estimated loss of economies was not substantial within the meaning of clause A." 41 S.E.C., at 898. The cases with which these particular comparisons were made involved companies outside Massachusetts.12 The Court of Appeals held that the comparisons with the loss ratios of companies involved in prior cases were "largely irrelevant" because "* * * these ratios are significant only as they affect the investment structure of the companies in the particular case, and different companies may be compared only on the assumption that both operate at the same level." 376 F.2d at 113, 115. The court's ultimate conclusion was that only close analysis of NEES' own "particular circumstances" was relevant to the Commission's inquiry.

It is significant, however, that the Court of Appeals' criticism of the Commission's use of ratios relied heavily on the court's reading of the statistical data in evidence as showing that the projected loss of economies "would decrease (NEES') rate of return from 6.4 per cent in 1959 to 4.1 per cent on the projected basis," or some 30% below, "an average rate of 5.9 per cent for the nonaffiliated Massachusetts gas companies * * *." 376 F.2d, at 114. But, as the Commission has noted, the court's computation that the separated companies would realize a return of only 4.1% contained a serious error, for it overlooked the allowance to be made for income tax deductions generated by the projected losses. The actual rate of return taking such deductions into account would be a significantly higher 5.2%.13

In any event, we may agree that the ratios of losses of revenues, expenses, and income are necessarily affected by differences in capital structure, management, market position, and other factors. But it by no means follows that the Commission's comparisons are for that reason irrelevant to the determination whether a projected loss of economies is so important as to cause a serious impairment of the separated system. It was well within the range of the Commission's administrative discretion to use the loss ratios, as it did, "as a guide in adjudicating the pending case." Philadelphia Co., 28 S.E.C. 35, 50, n. 24. The Commission in its expert judgment may so employ evaluative factors it considers relevant.14

Indeed, NEES apparently recognized that its burden to establish that its situation comes with Clause A included the burden of showing that the projected loss of economies would be more serious for its separated system than the comparable level of losses in the other cases already decided by the Commission. Respondent attempted to prove that the gas system's distance from sources of supply gives it only a very narrow competitive advantage over oil as a fuel, and, further, that the system's growth potential is more limited by a lack of new housing expansion in the area serviced by the gas companies. As we shall see below, the Commission found that NEES had not made a case in either respect insofar as those matters bore on whether the projected loss of economies threatened serious impairment of the separated system.

The Commission's resort to data concerning the operations of the nonaffiliated Massachusetts gas companies was a response to NEES' argument, supported by the Massachusetts Department of Public Utilities, that the projected loss of economies from separation of the gas system would require the gas companies to seek rate increases which might seriously impair or destroy any hope of a successful operation. Natural gas in 1959 enjoyed in New England the smallest price advantage over oil of any section of the country. The annual differential was $7 over oil for a typical New England house compared with $27 to $118 in favor of gas in the rest of the country.15 NEES contended that the predicted rate increase would substantially or entirely eliminate the gas system's already narrow price advantage over oil competitors. The Commission's answer was to inquire about the economic health of the already nonaffiliated Massachusetts gas companies. The Commission found that these companies were apparently able to earn a fair return although not enjoying the supposed advantages of affiliation with electric utilities; and it could find no evidence that they did not face the same competitive conditions as NEES.16 The Commission found further that, despite NEES' insistence that its market conditions differed from the nonaffiliated companies because of relatively stagnant franchise areas offering less sales growth,17 there was no evidence that this would prevent the separated gas system—which would emerge as the second largest independent in the State—from competing as effectively as the smaller independents who had long held their own. Finally, the Commission noted that after severance the gas system's operating ratio would be more favorable or only slightly higher than the ratios of nine independents and therefore concluded that it "would be entering the realm of speculation at this time to assume that rate increases would ensue from severance." 41 S.E.C., at 899.18

The Court of Appeals rejected the comparison of these operating ratios, again on the ground that such ratios fail to take account of special characteristics of individual companies. The court observed that since all New England gas companies operated on a "small cushion * * * (t)he significance of this is not negated by observing that non-NEES companies in Massachusetts seem to be surviving, for the focus must be on the specific characteristics of the NEES companies, the only ones affected by the Commission's order." 376 F.2d, at 113. The court further held "irrelevant the comparison of operating ratios, since a business may operate relatively efficiently, yet at a level too low to attract investors." 376 F.2d, at 114, n. 6. For the reasons already stated for our disagreement with the Court of Appeals' view of the Commission's use of other ratios, we disagree that this comparison was either irrelevant or outside the limits of the Commission's administrative discretion. The dissection and evaluation of an economic projection is a function Congress committed to the Commission, not the courts. A court may believe it would have done the job differently and better; but judicial inquiry must be addressed to whether what the Commission did is fatal to its ultimate conclusion that the holding company failed to carry its burden of showing a loss of "substantial" economies within the meaning of Clause A. In assessing NEES' forecast of the need for rate increases because of the projected loss of economies, it was proper for the Commission to consider the performance of other Massachusetts gas companies which were already operating independently. NEES was afforded every opportunity to sustain its burden of showing that the separated gas system would wither into critical health despite the contrary inferences suggested by the comparison made by the Commission. It cannot be a basis for finding error that the Commission found the attempt unpersuasive, given the gas system's size,19 and the prognosis of efficiencies comparable to those achieved by the independents.20

The Commission conceived that the projected loss of economies would in some measure be offset by advantages realized by the separated system under the direction of "a management solely interested in and devoted to the gas operations. * * *" 41 S.E.C., at 901. NEES, again supported by the Massachusetts Department of Public Utilities, took the position that its operation of the companies had already achieved all possible benefits of interservice competition. The Commission found the argument unpersuasive, relying again on a comparison with the nonaffiliated Massachusetts gas companies. This was a comparison of the sales performance of the gas companies under NEES management with the sales performances of the independents. All seven of the comparable independents showed substantially higher gas sales and revenues per customer and lower costs to customers.21 The Commission found unpersuasive NEES' explanation that this was accounted for by the greater residential growth potential of the areas serviced by the independents.22

The Court of Appeals held that the test of "serious impairment" under Clause A already took account of offsetting benefits to be realized from separation and therefore "that done, the general judgment has no independent significance in an individual case." 376 F.2d, at 115-116. Whatever the merit of the general premise, see NEES I, 384 U.S., at 184-185, 86 S.Ct., at 1402-1403, we understand the Commission's finding to have been simply that the projected $1,100,000 loss of economies did not in fact take into account any offsetting benefits on the assumption that joint operation had already achieved the advantages of independence. See 41 S.E.C., at 900-901. The Commission's conclusion that NEES' assumption was not proved has support in the record and the Court of Appeals was not justified in rejecting it.

The judgment of the Court of Appeals is reversed and the case is remanded to that court with direction to enter a judgment affirming the Commission's order. It is so ordered.

Judgment of Court of Appeals reversed and case remanded with direction.

Mr. Justice DOUGLAS and Mr. Justice MARSHALL took no part in the consideration or decision of this case.

Mr. Justice HARLAN, whom Mr. Justice STEWART joins, concurring.

Given the earlier decision of the Court in this case, SEC v. New England Electric System, 384 U.S. 176, 86 S.Ct. 1397, 16 L.Ed.2d 456, which I continue to believe wrongly construed the statute but by which I consider myself bound, I join today's opinion of the Court.