Full opinion text
SPOTTSWOOD W. ROBINSON, III, Circuit Judge: In D. C. Transit System, Inc. v. Washington Metropolitan Area Transit Commission (Transit I), we reviewed the Commission’s Order No. 245, promulgated after a public hearing in 1963, by which Transit was authorized to raise its token fare for passenger transportation within the District of Columbia and between the District and points in Maryland. We then resolved several subsidiary issues germinated by that order, but were unable to pass judgment as to the legal propriety of the fare increase. The order in that aspect was predicated upon the Commission’s finding that a margin of return of $1,480,746, representing a 4.87% rate of return on Transit’s operating revenues, was just and reasonable. Upon examination of the record, however, we discovered that we had “no intelligible basis for disposing of the competing claims before us that the return allowed by the Commission is, on the one hand, too high, and, on the other, too low.” Accordingly, we remanded the case to the Commission “for further proceedings * * * to determine the margin of return over and above operating expenses that Transit should be allowed.” On September 17, 1965, shortly before our remanding order was certified to the Commission, Transit filed a new tariff making numerous fare changes for 1966, including elevations in the District cash fare from 25 to 30 cents, and in the District token fare from 21.25 to 25 cents. Exerting a power conferred by the Washington Metropolitan Area Transit Regulation Compact (Compact), the Commission promptly suspended the new tariff pending a determination as to its reasonableness, and thereafter conducted a public hearing on the proposals it contained. On January 26, 1966, in purported response to our remand, and without prior notice or further hearing for the purpose, the Commission entered Order No. 563, in supplementation of Order No. 245, reaffirming the return for 1963 the latter had previously allowed. It simultaneously issued Order No. 564, in which it found that a margin of return for 1966 of about $2,000,000 above operating revenues, symbolizing a return rate of 6.03%, would be fair and reasonable, but that the- existing fare scale would generate net earnings estimated at only $648,357. The Commission nonetheless felt that a fare increase, save in a minor respect not questioned here, was unnecessary because of the availability of $2,-166,933 in a special reserve which had been created in consequence of our decision in Bebchick v. Public Utilities Commission. In lieu of an increase, the Commission permitted Transit to draw upon the reserve for approximately $1,-350,000 to accommodate the anticipated deficit in its earnings. Timely petitions for rehearing, denied by the Commission, ripened for our present review various issues stemming from these orders. No. 20,200 brings contentions that Order No. 563 is invalid for lack of notice and hearing, and is erroneous on the merits. No. 20,-201 presents importantly the claim that the margin of return awarded by Order No. 564 is unreasonable, and it, like No. 20,202, includes attacks upon other Commission findings and conclusions. We proceed now to a consideration of the problems thus engendered, in the sequence we deem best suited to exposition of the reasons for which we set each of the Commission’s orders aside. I 1963 Margin Of Return Following our remand, the Commission promulgated in Order No. 563 its supplemental findings to Order No. 245, and “affirmed” its earlier finding that a margin of return representing a rate of 4.87% on operating revenues “was fair and reasonable.” Petitioners in No. 20,200 attack Order No. 563 on the ground that it does not comply with our instructions governing the Commission’s task on remand. The disclosures made by the record, now to be delineated, lead us to agree. We pointed out in Transit I that the return of 4.87% possesses no inherent validity, and voiced our “need to know more than we have been told about why the Commission thought this was the appropriate margin.” We observed that the margin of return properly allowable over legitimate operating expenses is “the sum of money needed to attract the capital, both debt and equity, required to insure financial stability and the resulting capacity of the utility to render the service upon which the public depends.” And we stressed the necessity for “inquiries and findings — judgmental as the latter may often be because ratemakers must be prophets of the future as well as historians of the past —into such things as the capital programs in prospect, what such programs entail in terms of down-payments as well as financing, the cost of borrowing money, working capital needs, the desirable ratio of debt to equity, the incentives required by a stockholder to keep his money in the business and the dividends and growth rates requisite to supply these incentives, the opportunities in these respects provided in comparable businesses, and the related matters which must be prayerfully explored by the conscientious regulator before he can begin to say why he fixed upon 4.87 rather than 6.5 or 3.2.” Despite these explicit guidelines, we search in vain the Commission’s supplemental opinion for the vital findings, or for any manifestation that the Commission has made the inquiries referred to. The bulk of Order No. 563 is devoted to a summary of the testimony of various witnesses given during the 1963 hearing. The summary covers such matters as Transit’s rate base; the capital structures of Transit and other utilities; Transit’s cost of capital and that of other utilities; the degree of risk borne by Transit shareholders, as compared with other regulated utilities; dividends and earnings of Transit and its parent, and those of other utilities; Transit’s estimated debt expense for the future annual period; actual and authorized operating ratios of Transit and other utilities; and the average market prices of the stock of Transit’s parent. Having thus reviewed the record in somewhat greater detail than in its original order, the Commission then simply reaffirmed its earlier conclusion on the rate of return in language strikingly reminiscent of that which when remanding we had characterized as “stock generalizations which serve only to frustrate, rather than illuminate, judicial review:” “We have before us then an abundance of testimony relating to returns authorized and/or earned by transit companies, telephone companies, electric companies, and gas companies. This information allows us to examine earnings on investments carrying, to some degree, a comparable risk, and are of some value when considered in relation to this particular company. No single rate of return is universally applicable to all transit companies in the United States. A fair rate of return varies with the times and conditions of a particular company as these conditions and opportunities exist at the time of the weighing of the facts in the making of a determination. What constitutes a reasonable rate of return is a question of fact, the solution of which calls for the exercise of sound judgment and common sense by the Commission. “We have weighed very carefully all testimony and exhibits in this proceeding, along with supplemental data where noted. We are of the opinion that fares producing an operating ratio in the range of 95 to 95.5 constituted a fair and proper return. The projected net operating income of $1,-480,000 was the amount necessary to enable Transit to service its debt, pay reasonable dividends, retain a reasonable portion in its business, and to attract investments of private investors. It was, in addition, the amount necessary to maintain investor confidence and to protect the company from the risks peculiar to the transit industry and to Transit itself.” Dividend Payout The Commission’s opinion is almost wholly devoid of any indication of the method by which it calculated that a margin of return of $1,480,000 was fair and reasonable. The Commission did account for a portion of that sum by finding that a dividend payout of $500,000 would be reasonable — apparently in response to our observation that “Transit’s annual dividend pay-out of about $500,-000 appears to have been treated as if it were a cost of operation, like the annual expenditure for gas and oil, with no examination of, or conclusion about, its appropriateness.” Our concern, however, is hardly alleviated by the Commission’s treatment of this point on remand: “While there was no direct testimony as to the dollar amount required to be available for dividends, the record does show that $500,000 has been paid out annually over past years to the investors. In view of the fact that in the past three years the market price of the parent company has been relatively stable, it appears to us that the margin of profit allowed by the Commission was fair and reasonable to both the investor and the consumer.” The Commission’s stock price stability theory cannot survive close scrutiny. A stable market price could indicate that a constant dividend yield has been consistently too high or too low, as well as that it has been fair and reasonable. Moreover, the theory finds little support in the record, and we have been referred to no authority buttressing the validity of this method of establishing the appropriateness of a dividend. Thus we are left, as before, without an acceptable rationale for a dividend payout of $500,-000, but only the fact that over a period of years Transit has paid annual dividends in that sum. Return on Rate Base The Commission’s ratemaking responsibilities are by no means exhausted by any determination it may make as to a fair dividend. Even were we able to sustain the Commission in that regard, we would not be enabled thereby to affirm the Commission’s allowance of a total return of $1,480,000. What is needed is a reasoned justification of the end_. result — the entire excess of gross operating revenues allowed over operating expenses. Here, however, the Commission does not provide any suitable explanation of or support for its ultimate conclusion that a margin of return of $1,480,000 is fair and reasonable. Save for its verdict with respect to the dividend payout, the only concrete finding that the Commission made is that the authorized margin of return would be equivalent to a return on rate base of approximately 6%%. We express no view on the reasonableness of such a return on rate base, nor on the accuracy of the Commission’s statement that such a return is “conservative.” We do stress, as we did in our former opinion, that testing the reasonableness of an operating ratio by reference to the return on rate base, while sometimes useful, “can never be conclusive. The important thing is not that the result reached by the operating ratio method be compared with that which would have been reached under some other method, but that the operating ratio method be applied rationally and intelligibly in the first instance.” Absent a calculation of operating ratio responsive to the considerations enumerated in our opinion remanding the Commission’s reference to return on rate base could not provide independent support for the result it reached. Costs of Capital On our prior reviéw, we took note of the Supreme Court’s statement in FPC v. Hope Natural Gas Co. that “in balancing investor and consumer interests, it is essential to consider the capital costs of the enterprise.” We emphasized that the operating ratio method does not make an inquiry into capital costs irrelevant, and that indeed “it was in part the failure of rates established by the rate base method to cover the capital costs of most motor carriers” that gave birth to the operating ratio method. During the course of the 1963 hearing, bofh Transit and the protestants submitted evidence relating to Transit’s cost of capital. Transit’s expert witness testified to a combined cost of debt and equity capital for Transit of 7.8%, derived from industry averages. He stated further that this figure could not itself be taken to represent a fair return, but would have to be adjusted upward, to about 10%, to allow for a reasonable addition to surplus. The 10%, he said, could be applied to the “physical value of the plant that is devoted to public service” in order to determine a fair return. Protestants’ expert recommended that Transit be permitted to earn gross revenues sufficient to provide a return on its equity capital of 12% after allowing for operating expenses and interest on its debt. He reached this figure after study of the earnings-price ratios of Transit’s corporate parent and 12 urban transit companies that he felt were “most comparable” to Transit. Over the ten-year period from 1952 through 1961, he found, the average of the 12 companies was approximately 11%. This he increased to 12% to compensate for the increased risk borne by Transit’s shareholders emanating from Transit’s abnormally high debt-equity ratio. He stated that the 12% could be applied to Transit’s average book equity for 1963 in order to provide a fair return on equity. He then added the resulting figure to Transit’s estimated debt expense for the test period, and arrived at a sum which he felt should be allowed as the margin of return over operating expenses. The Commission made no findings based on any of this evidence, and did not analyze the cost of capital testimony of Transit’s expert witness. With respect to the testimony of protestants’ expert, the Commission had only this to say: “Moreover, we have considered but placed little emphasis on protestant’s recommended rate of return utilizing the cost-of-capital method. It should be noted that more traditionally the computation of the cost of capital is done by determining the cost of debt and of equity and weighting the two for a composite rate. The protestants in this case disregarded this traditional method. We feel that the cost-of-capital method utilized is inappropriate in determining a proper rate of return for a transit company. Our view is reinforced when we consider the capital structure of this particular company. Other factors in addition to cost of capital should be considered in making the return allowance. These additional factors include the degree of risk involved, a comparison of earnings, the ability to attract capital, the economic conditions of both the market place and the community, the efficiency of management, and the contribution to surplus. We have seen that the recommendation of the pro-testante expert witness is based not on a mathematical formula but a weighting of any [sic] judgmental factors. Thus, the ‘judgment’ factors tend to undermine the objectivity of the end results. This is especially true when the answers given by the witness on his cross-examination are compared with his direct testimony.” We are unable to understand the Commission’s failure to make findings on the critical cost of capital factor. Nor are we persuaded by its stated reason for rejecting the approach taken by protestants’ expert. In the first place, that protestants’ expert “disregarded” the “traditional method” and testified to a cost of equity capital, rather than an overall or combined cost of capital, does not appear to us to deprive his testimony of utility. The witness explained why he talked in terms of a cost of equity capital: “I do not employ a composite rate of return, as I believe that my method of allowing the actual interest expenses, and a determination of the return on the actual equity investment is simpler and gives approximately the same result.” Had the Commission desired, the protestants’ figure for cost of equity capital could have been transformed quite easily into a figure representing a combined cost of debt and equity capital. As we have noted, Transit’s witness did testify to an overall cost of capital. We are also puzzled by the Commission’s statement that “the cost of capital method * * * is inappropriate in determining-a proper rate of return for a transit company,” and we do not understand how this view is “reinforced” by reference to “the capital structure of this particular company.” Indeed, it is difficult to see how else the Commission could hope to determine “the sum of money needed to attract the capital, both debt and equity, required to insure financial stability” than through an analysis of the sort proposed by the two expert witnesses. And though Transit’s unusual capital structure may have a bearing on its costs of acquiring debt and equity capital, it is unclear why that factor precludes ascertainment and consideration of such costs at all. Comparable Earnings By including “a comparison of earnings” in its list of “additional factors” the Commission may have had reference to FPC v. Hope Natural Gas Co., where the Supreme Court said that ‘‘the return to the equity owner should be commensurate with returns on investments in other enterprises having corresponding risks. That return, moreover, should be sufficient to assure confidence in the financial integrity of the enterprise, so as to maintain its credit and to attract capital.” As various commentators have noted, implicit in this language are two methods for determining a fair rate of return. One, commonly referred to as the “attraction of capital” standard, attempts to do so by looking to the capital costs of the enterprise as reflected in earnings-price ratios. The other, frequently denominated the “comparable earnings” standard and suggested by the italicized portion of the language quoted, is based on a comparison of the return on equity of the subject company with the returns afforded by other companies. Thus, while we are unable to accept the Commission’s reasons for rejecting the cost of capital evidence presented to it, we certainly do not intend to intimate any disagreement with the Commission’s statement that “a comparison of earnings” is an additional factor to be considered. On the contrary, we strongly admonish that a comparison of the returns being afforded Transit’s shareholders with those of other companies of corresponding risk is necessary to a responsible determination of the proper margin of return which Transit should be allowed. Notwithstanding the Commission’s allusion in this case to the comparable earnings standard, it is clear that it made no meaningful evaluation of the level of returns being afforded Transit shareholders. In remanding, we noted that, at the time Transit began operations, the company’s heavy debt capitalization “made inevitable large returns on equity to the extent that the relatively heavy interest requirements were safely covered and the gamble on traffic levels after the resumption of operations was won.” But we added that: “The urgency of the problem of traffic uncertainty has, however, receded appreciably into the past. Although it continues to exist to some degree by reason of the seemingly inexhaustible enthusiasm of motorists and highway builders, on the one hand, and persistent speculation about rail rapid transit, on the other, we think the time has come for the Commission to make a careful review and analysis of the earnings experience of Transit from its inception, and of what that experience has meant to the equity owners, both by way of dividend payments and growth in book values through retained earnings. We do not see how current fare increases can properly be appraised apart from such a study, and the reflection of its results in the articulation of the Commission’s decision as to the margin of revenues over expenses appropriate to today’s needs, as distinct from the crisis conditions of nearly a decade ago.” The Commission responded to these comments by setting forth gross statisties on Transit’s total , and average-per-year retained earnings over a nine-year period, and the range of net operating income figures over that period. It observed that “the rate of dividend growth has been zero since 1960,” noted “little growth trend in the earnings per share,” and stated that “[t]he degree of stock price stability in the market place indicates to us that earnings in the past have not been excessive.” The Commission’s approach thus appears to have been to justify current earnings and dividends by reference to those of the past, in clear contrast to our mandate that the Commission articulate its “decision as to the margin of revenues over expenses appropriate to today’s needs as distinct from the crisis conditions of nearly a decade ago.’’ There is much information in the records of both the 1963 and 1965 rate proceedings relative to Transit’s return on equity, as well as the returns being earned by companies with which Transit might be compared, but the Commission made no findings based on any of this evidence. We respect the Commission’s primary fact-finding responsibility, but we cannot ignore the substantial indications in the record that the Commission has permitted Transit to earn extraordinarily high returns on its investment. The protestants in the 1963 proceeding introduced evidence that Transit’s average book equity for 1963 would be $4,206,990; and its debt expense for that period $602,089. On this basis the total return of $1,480,746 allowed by the Commission represents a return on equity of $878,657, or approximately 21% . Transit introduced data showing the average earnings on common stock equity of 15 transit companies for 1959-61 to be 8.7%, and the protestants showed average earnings on book equity of 12 transit companies for 1952-61 to be 7.5%. Transit’s evidence in the 1965 proceeding discloses the average return on book equity for 18 urban transit companies for the period 1960 to 1964 to have been approximately 4.5%, while Transit’s average return on book equity for the same period was 23.3%. Protestants introduced exhibits revealing that, for the period 1959 to 1964, Transit’s average return on book equity was 24.01% while the average for ten transit companies for that period was 9.45%. One witness in the 1963 proceeding testified that Transit had experienced “abnormally high earnings on the equity investment,” and calculated that over a six-year period Transit had accrued earnings in excess of “fair annual earnings on book equity,” which he set at 14%, in the amount of $3,882,000. Another witness in the 1965 proceeding testified that “when one looks at the figures [representing Transit’s return on equity] * * * one wonders whether or not he is looking at the figures for a regulated utility. * * * [T]here are many highly speculative industries which would like to be able to hold out such high returns to the equity investor. I know of no other utility company that can show as high a return on equity, on average, as this one.’’ We do not decide, whether, based on this record, the Commission could have made a sustainable finding that the return allowed by Order No. 245 was “commensurate with returns on investments in other enterprises having corresponding risks.” In view of the Commission’s failure to make a finding one way or the other on that question, no such inquiry on our part is called for. We do think, however, that the Commission’s failure to deal meaningfully with the evidence we have summarized is significant as a further indication that it has misconceived the nature of its responsibilities and the kind of inquiries and determinations that must be made in setting rates under the operating ratio method. In the absence of any inquiry into the appropriateness of the returns being afforded Transit shareholders in the light of returns being earned by other companies of comparable risk, the peculiar risks borne by Transit shareholders, and any other relevant factors, we are unable to conclude that the Commission has responsibly complied with its statutory mandate to prescribe just and reasonable fares. In sum, we find in the Commission’s supplemental opinion little in the way of “particularized reference” to the matters discussed in our remand opinion, none of the essential findings called for, and no manifestation of inquiry into any of the considerations “which must be prayerfully explored by the conscientious regulator before he can begin to say why he fixed upon 4.8 rather than 6.5 or 3.2.” Instead, it has clearly treated Transit’s $500,000 dividend payout as a “cost of operation” without any inquiry into its appropriateness. It has dismissed evidence as to Transit’s cost of capital as “inappropriate.” And it has made no meaningful evaluation of the returns being afforded Transit shareholders. We think it evident that, though purporting to equate a fair return with “the amount necessary to enable Transit to service its debt, pay reasonable dividends, retain a reasonable portion in its business, and to attract investments of private investors,” the Commission has neglected to apply that principle in practice. Under these circumstances, the disposition we should make of the orders under review is a matter of some difficulty, to which we now turn our attention. Disposition On our original review of this case in Transit I, we were unable to discern from the Commission’s opinion the method by which it had computed the return to be allowed Transit. Because the Commission had “described its deliberations and conclusions * * * [solely] in terms of stock generalizations which serve only to frustrate, rather than illuminate, judicial review,” we were left without “an adequate basis for knowing why it did what it did.” But notwithstanding our uncertainty as to whether the Commission had actually made the inquiries and the concomitant decisions we held to be required by the statute, we were unwilling to conclude, merely from the absence of findings in its order, that the Commission had not performed its duties. Thus we did not disturb the effectiveness of the fare increase granted by Order No. 245, nor did we make provisions for restitution by Transit of increased fares collected pursuant to that order. Instead, we remanded to the Commission to enable it to clarify the grounds of its action or, if necessary, to formulate a new order. The case now before us, however, is quite different. The Commission, once again, does not advance a rational basis for its determination of rate of return— despite the opportunity afforded it to do so, after we had enunciated the applicable principles and emphasized the need for findings. Moreover, it now appears affirmatively that at no time in this proceeding has the Commission made the investigations and the resolutions essential to a legitimate exercise of its authority to prescribe just and reasonable fares. Since we cannot escape the conclusion that the Commission’s approval of the fare increase was based upon a mistaken view of its responsibilities in setting rates under the operating ratio method, we hold that Orders Nos. 245 and 563 must be set aside The question nonetheless remains whether we should again remand to the Commission, to permit further consideration and a new order. For even where agency action must be set aside as invalid, but the agency is still legally free to pursue a valid course of action, a reviewing court will ordinarily remand to enable the agency to enter a new order after remedying the defects that vitiated the original action. A remand for this purpose, however, necessarily assumes continuing power in the agency to deal with the subject matter of the proceeding. Where, because of changed circumstances, or because of the decisional grounds nullifying the initial order, the agency does not possess that authority, a remand is manifestly unwarranted. Orders Nos. 245 and 563 have been superseded by later fare orders entered by the Commission subsequent to our remand in Transit I- The propriety of another remand thus hangs on the Commission’s power to devise a new order, nunc pro tunc, governing the years intervening between Order No. 245 and the entry of a subsequent order prescribing the “lawful fare * * * to be in effect.” The Commission, however, possesses no authority to fix rates for the past. An order prescribing the lawful fares to be charged by a public utility, being essentially legislative in character, ordinarily speaks only for the future. And we find nothing in the statutory provisions governing the Commission’s regulatory responsibilities that indicates an intent to depart from this “customary pattern of fixing rates prospectively.” Hence, we conclude that the Commission lacks power to enter a new rate order in this proceeding, and that a remand for further consideration is not called for. With Orders Nos. 245 and 563 invalid and further remand now futile, it follows that Transit must be compelled to make appropriate restitution for the increased fares it collected. This conclusion is unaffected by the fact that we do not decide that the fares authorized are unjust or unreasonable as a matter of law. Our role as a reviewing court is not to make an independent determination as to whether fares fixed by the Commission are just and reasonable, but rather to insure that the Commission, in exercising its rate-making power, has acted rationally and lawfully. Our function is normally exhausted when we have determined that the Commission has respected procedural requirements, has made findings based on substantial evidence, and has applied the correct legal standards to its substantive deliberations. Our task is likewise at an end when we have ascertained that the Commission has not done so. In this case, given our conclusion that the Commission failed to apply appropriate criteria, and failed to make the inquiries prerequisite to valid exercise of its rate-setting authority, we could not permit Transit to retain the increased fares, since to do so would be to give legal effeet to the Commission’s invalid order. This is so notwithstanding that we have held neither that the Commission lacked power to order a fare increase, nor even that the fares authorized are, as a mat£er 0f jaw> unjust or unreasonable. Thus we are confronted with the necessity of formulating the criteria by which the amount of restitution is to be measured. Ordinarily, of course, the proper disposition on setting aside a rate increase unlawfully ordered by the Commission would be to compel the regulated company to restore the entire difference between the higher fares collected under the invalid order and the amount that it would have received from the fare schedule previously in effect. More fundamentally, however, our decision in this regard is to be governed by the equitable considerations which apply to suits for restitution generally The basic question is whether “the money was obtained in such circumstances that the possessor will give offense to equity and good conscience if permitted to retain it,” and is “no longer whether the law would put him in possession of the money if the transaction were a new one.” Since restitution is not a matter of right, but is “ex gratia, resting in the exercise of a sound discretion,” it lies within our authority to direct restitution in an amount less than the whole sum of the increased fares collected under the invalid order, or to deny it altogether, if compelling equitable considerations so dictate. The exercise of such an equitable discretion by this court is by no means an usurpation of the administrative powers of the Commission nor is it an arbitrary extension of judicial authority; it is “mere inaction and passivity in line with the historic attitude of courts of equity for centuries.” Because we have found the Commission’s action in approving the fare increase to have been invalid, and because we have no basis in later valid action of the Commission for inferring that the rates set by Order No. 245 were in fact, just and reasonable, despite the defects in the Commission’s deliberations leading to their original approval, we could not, as we have said, give legal effect to those rates by withholding restitution altogether. At the same time, we are confronted by circumstances indicating a substantial probability that it would be inequitable to compel Transit to restore the entire amount it realized from the fare increase. More specifically, we see that by Order No. 245 the Commission found that the net income of $937,669 which Transit would earn under the then existing fare schedule would not constitute a “fair and adequate return,” and nowhere in this proceeding has that finding been challenged. Indeed, the evidence presented by protestants at the 1963 hearing was to the effect that a return of $1,107,000 would be fair and reasonable. Additionally, we note that the net income which the Commission found Transit would earn in consequence of the preexisting fares would have represented a return on equity of less than 8%, while protestants’ expert witness recommended a return on equity of 12%. And in a later rate proceeding, after determining Transit’s cost of capital and taking into account the returns afforded by other companies of comparable risk, the Commission found to be fair and reasonable a return on equity of 14%, a determination today affirmed by this court as based on adequate findings and reasons supported by substantial evidence. We could not permit the Commission’s latter conclusion to control the amount of restitution now to be ordered, since it was based on a record reflecting conditions in years later than the period relevant here, and it did not purport to be a determination of a fair return for the years with which we are concerned. On the other hand, it has obvious significance as a factor suggesting the inequity in ordering Transit to repay the entire amount of the fare increase. In sum, we find substantial and unmistakable indications, in the record under review as well as in a subsequent determination of the Commission which we have affirmed, that it would be unfair to order Transit to restore the full amount it realized under the invalid fare increase. In so concluding, we have placed particular reliance upon the Commission’s finding as to the amount of net income which would have been earned by Transit under the fare schedule in effect prior to Order No. 245. That finding was properly grounded, conformably with the Commission’s role in fixing rates to operate in the future, upon estimates of expenses to be incurred and revenues to be received during the future test period. But we perceive no justification for permitting such a prediction to control the course we are to follow in equitably disposing of funds collected during years already past. “There are times, to be sure, when resort to prophecy becomes inevitable in default of methods more precise. At such times, ‘an honest and intelligent forecast of probable future values made upon a view of all the relevant circumstances’ is the only or-ganon at hand. * * * But prophecy, however honest, is generally a poor substitute for experience.” We think these observations are applicable here, and we are unable to see how any proper resolution of the matter of restitution in the circumstances presented could ignore the reality of Transit’s financial experience during the years in question. In laying down a standard by which to measure Transit’s right to retain funds collected under the fare increase, we are aware that we are ill-equipped, even were we authorized to do so, to search the record and reach our independent conclusions as to what would have constituted reasonable fares for the period in question. Nevertheless, the duty to reach a just decision in this regard cannot be shirked, and our effort must be to find a solution which lies within our competence as a reviewing court, while at the same time responding in the fullest possible measure to the equitable considerations that must guide us. The disposition we deem most consonant with this objective would compel Transit to restore the amount realized by the fare increase only to the extent that its actual return is not reduced to an amount which all parties have agreed would be unreasonably low. Thus Transit will be permitted to retain any portion of the higher fares necessary to preserve its actual earnings during the years in question at the level conceded by the protestants to represent a fair return. In so doing, we do not say that the Commission erred in failing to adopt the testimony of protestants’ expert witness, nor that fares designed to produce the return proposed by the protestants would have been the only lawful fares, nor even that they would have been just and reasonable. We decide only that in the circumstances of this case it clearly does not offend “equity and good conscience” to permit Transit to retain that part of the fare increase essential to avoidance of an undisputedly unfair return. One circumstance upon which we place considerable weight in reaching this conclusion is the availability of the Commission’s 1967 decision permitting a return to the equity holder quite close to that recommended by the protestants in this proceeding. Though, for reasons already explained, we are unable to regard that determination as controlling with respect to the question before us, we think it is entitled to weight as “the opinion of a body of experts upon matters within the range of their special knowledge and experience.” Being cognizant of the dangers in attributing finality to positions taken by the parties in a proceeding which so broadly affects the interests of the public as this one, we do not decide whether, in the absence of that later determination by the Commission, the course taken here would be open to us. II The Acquisition Adjustment Account Petitioners in No. 20,201 complain of a fundamental change by the Commission in the method of its treatment of Transit’s acquisition adjustment account, which has engaged our attention before. The event giving rise to this account was Transit’s purchase in 1956 of properties from its predecessor, Capital Transit Company (Capital), at a price lower by $10,339,041 than the net original cost of those properties to Capital. Transit’s allowances for depreciation thereon could, of course, have been related to its own acquisition cost; but this would have required the development of new depreciation rates computed on remaining life, and new depreciation bases derived in part from distribution of the purchase price among the items of property acquired. To save the labor incidental to that process, however, the Public Utilities Commission of the District of Columbia (PUC), the Commission’s predecessor, ordered that two things be done. One was the establishment of the acquisition adjustment account to accommodate an amortization, over a ten-year period beginning August 15, 1956, of the $10,339,041 difference in acquisition cost to Capital and Transit, respectively. The other was a direction that depreciation be accrued on the basis of Capital’s original cost and at the rates previously fixed for Capital, with ten annual offsetting credits to operating expenses of $1,033,904 derived from the amortization. PUC selected the ten-year period for amortization in order to link it to an annual accrual of $1,044,196 over the identical period to a reserve designed to absorb Transit’s estimated future expenses for track removal and street repaving incidental to its franchise-required conversion from Capital’s trolleybus operation to Transit’s eventual all-bus operation. With the amortization and the accrual in almost the same annual amount for exactly the same period, any material impact from either upon Transit’s income posture would be avoided. In Order No. 245, however, the Commission suspended the annual accruals to the reserve as of January 1, 1963, concluding that its existing balance would cover all of Transit’s removal and repaving expenses anticipated for the immediate future. The Commission stated that it would also reexamine, the period for liquidation of the acquisition adjustment balance in light of PUC’s decision to coincide its amortization with the period established for accruals to the reserve for removal and repaving. The Commission observed that “the proper [original] solution [by PUC], completely equitable to both applicant and the ratepayers, would have required the total sum represented by the acquisition adjustment account to be distributed ‘over all items of depreciable property’ as recorded on the books as of August 15, 1956,” and it felt that such treatment “would have been the only completely accurate and equitable method.” The Commission would, it said, “relate the acquisition adjustment balance to the properties acquired on August 15, 1956, which are still in service and subject to depreciation at original cost” but, noting that additional evidence was needed for this purpose, it ordered that amortization of the acquisition adjustment account continue without change “until adequate evidence supporting a different rate [of amortization] is presented to the commission.” In Transit I, we upheld the Commission’s determination on this point. Shortly before we did so, however, the Commission, by Order No. 385, had proceeded to alter radically the scheme for amortization of the acquisition adjustment balance. Without taking further evidence, it decided that as of January 1, 1964, the period for amortization of the balance of $2,519,484 would be extended to August 15, 1976, the duration of Transit’s franchise. The Commission thus enlarged the remaining original period, which would have expired on August 15, 1966, for the write-off of this balance from approximately two and one-half years to about twelve and one-half years, and reduced the amount of the annual amortization from the previous $1,033,904 to the much lower figure of $199,561. In directing this change, the Commission did not relate the liquidation of the acquisition adjustment balance to any new schedule for depreciation of the involved properties, a procedure it had promised in Order No. 245 and one which we had regarded as “most sensible.” Instead, it said that amortization of that balance within the remaining two and one-half years of the period originally prescribed might so reduce net operating profits after the terminal point on August 15, 1966 as to require another fare increase. Thus arising is the question whether the Commission was at liberty to alter the mode of amortization, with a view to pegging Transit’s fares, without correlating in some reasonable fashion, the new method in time and amount to ongoing depreciation of the acquired properties, for the inflated book values of which the acquisition adjustment account was established as an offset. As a preliminary matter, though, we face, but reject, a common protest by Transit and the Commission that petitioners are foreclosed from litigating that issue because they did not seek a review of Order No. 385, by which the change complained of was directed. The Commission promulgated Order No. 385 without a hearing of any sort, in spite of its claim, in defending its earlier Order No. 245 before us in Transit I, that it had to have more evidence before it could intelligently alter the amortization of the acquisition adjustment balance. And when the Commission adopted Order No. 385, it proceeded upon a completely new theory — the need to stabilize Transit’s fares. There was no opportunity for the protestants to combat the change, or the basis upon which it was made, before it became a reality. Petitioners made a timely application to the Commission for reconsideration of Order No. 385. In denying it, the Commission proclaimed ■ that the order was “subject to complete analysis and review in any future rate proceeding.” This very clearly left the matter open to reexamination, and indeed the Commission reexamined it in Order No. 564, this time rejecting petitioners’ objection on the merits. It is the determination made in the latter order, presently under review, that petitioners now attack, and this they may properly do. Reverting to the merits of petitioners’ contention, we find that the Commission’s new plan for amortization of the acquisition adjustment balance largely sets for naught fundamental considerations the Commission was obliged to observe. One was the intended function of the amorization as the counterweight to excessive depreciation charges. While, on the assets Transit bought from Capital, Transit was annually taking the same depreciation Capital was allowed to take, the amount of the annual amortization was to balance out the difference between Transit’s purchase price and their higher value on Capital’s books. Quite obviously, Transit’s annual depreciation expense would become inflated if the compensatory value of the amortization was reduced. And such a reduction was necessarily destined when the Commission ordered its drastic changes in the amortization term and the amount of the annual amortization. This we see in vivid outline as we again consult the record before us. Of $46,534,172 in original cost of the de-preciable properties Transit acquired from Capital in 1956, a total of $35,979,-793, or 77.3% had been retired from service by January 1, 1964, the effective date of the Commission’s change. By Transit’s projection, not significantly variant from the plan the Commission adopted in Order No. 385, 61.1% of the remaining $10,554,379 would be retired at the close of 1966, the future test period, and 89.3% thereof at the end of 1970 — approximately the midpoint of the new amortization period. At the same time, the acquisition adjustment amortizations, which would occur annually in level amounts, would quite expectably maintain a steady course. The ever-growing divergence of property retirements from acquisition adjustment amortization is revealed in the following computation: The Commission’s stated reason for modification of the acquisition adjustment amortization was the avoidance of a fare increase in 1966, when under the existing arrangement the amortization would come to an end. Nonetheless, it did not undertake to support, either by evidence or findings, its theory that this circumstance would create a probable need for such an increase, and on the record its view in that regard seems conjectural. This is especially so since at the time of the changeover there existed a court-ordered reserve of more than $2,000,000 to serve precisely the Commission’s professed purpose. And we detect, as serious consequences of the Commission’s modification of the acquisition adjustment amortization artificial reductions of Transit’s net earnings for ensuing years which ostensibly would justify the very fare increase that the change was predicted to forestall. For the calendar year 1964, Transit stated a return on equity of 10.56%, reflecting $511,115 in net income and $4,837,992 in equity. Without the change, Transit’s net income would have been $760,232 greater, and the return on equity would have soared to more than 26%. Similarly, for the 12 months ending June 30, 1965 — the historical test period — Transit reported net operating income of $1,024,855, representing a 3.23% rate of return; but without the change the return on equity, with another $760,232 in additional income for that year, the rate of return would have been almost 75% greater. Again, for 1966 — the future test period— Transit’s net earnings were estimated at only $648,357, but with the change its projected depreciation expense was $475,-145 higher than otherwise it would have been. Judicial decisions have long censured depreciation plans breeding excessive charges which in turn exaggerate the costs of serving the public. We ourselves have held, in a conceptually indistinguishable situation, “that the substantial inflation of operating costs due to excessive depreciation * * * is unlawful.” The Commission’s handling of Transit’s acquisition adjustment account does violence to the wholesome principle those cases espouse. Moreover, the costs of the service a regulated utility provides should, as far as possible, be borne by those who are served as they are being served. The disproportion between the amortization of the acquisition adjustment balance- and remaining depreciable life of the acquired properties charges present riders with substantial depreciation expense properly assignable only to those who will ride in the future. If Transit’s customers are to pay for the rides they themselves take, operating expenses must be projected within a range that is reasonable in the historical as well as the mathematical sense. We do not suggest that Transit should be required to make up at one fell swoop the differences reflected by possible percentage variations between the amount of depreciation accruing on the acquired properties before January 1, 1964, the Commission’s changeover date, and the portion of the acquisition adjustment balance written off before that date. Prior to that time, the period for amortization of that balance was related, not firmly to depreciable life of the properties Transit got from Capital, but rather, as we have said, to the ten-year period for accruals to the reserve for track removal and street repaving; and this arrangement has never been challenged. The resolution of any problem in that connection is in the first instance, of course, a matter for the Commission. Now, however, the track removal and repaving accruals have been halted, at least temporarily, and the Commission has decided that it will put the amortization of the remaining acquisition adjustment balance on a different footing. We hold that, in doing so, it must maintain, subsequent to the changeover date, a reasonable relationship between the amortization and accruals of depreciation of the properties remaining in service. III Deficiency In Depreciation Reserve In 1964, the Commission found a large deficiency, existing as of August 15, 1963, in Transit’s reserve for depreciation on operating properties. It then required the depreciation reserve to be adjusted upward to its proper level and as an offset, the amount of the deficiency to be placed in a suspense account. In Order No. 564, the Commission decided to close out the deficiency balance of $1,099,627 in the latter account. The blueprint it followed in doing so, however, gave rise to the cluster of problems we reach next. Of the deficiency balance, $806,168 was the aggregate of underaccruals of depreciation on properties then still in service. The Commission directed that this amount be taken out of the suspense account and placed above the line in Transit’s depreciation expense account for 1966 and treating that much of the deficiency as a decrement Transit might legitimately recover from its fare-payers, the Commission granted Transit “[restoration” by authorizing the removal of $806,168 from the court-ordered reserve that we have mentioned and will shortly consider in greater detail. The remaining $293,459 of the deficiency balance represented underaccruals on properties which, after the deficiency was ascertained, were' transferred to nonoperating status. Holding that Transit could not collect that portion of the deficiency from its farepayers, the Commission decreed that it be transferred from the suspense account to the depreciation expense account as a below-the-line entry. The $806,168 Deficiency Petitioners in No. 20,201 argue that the $806,168 should not have been charged off in a single year, and should not have been deducted from the court-ordered reserve. They point to the fact that some of the properties as to which this part of the deficiency arose have service lives exceeding 30 years, and urge that any recoupment be éffected through annual deductions from operating income over the remaining life of Transit’s franchise. The problem, as we analyze it, breaks down into three facets, each of which we will discuss. The first concerns the handling of this item in the projection of Transit’s operating costs preliminarily to establishment of a margin of return. The second is the question whether Transit may reclaim this portion of the under-accrued depreciation for its investors. If so, the third is whether its deduction from the court-ordered reserve, in lieu of some other type of adjustment, is in order. Treatment for Ratemaking Purposes The Commission viewed the $806,168 as a loss to Transit’s investors for which they had not been compensated, and felt that “the techniques of preferring to accomplish [reimbursement] by a charge against a court-ordered reserve in lieu of charging a particular period’s profit and loss account, is merely a matter of semantics.” By its appraisal, “whether the deficiency * * * is written off directly on the operating statement of a particular period or periods, or whether it flows first through the court-ordered reserve, the impact on the rate-paying public is still the same.” We are not able to concur in the Commission’s reasoning. As we interpret the record, the above-the-line debit of the $806,168 to Transit’s depreciation expense account added just that much in pre-1966 depreciation expense to Transit’s operating costs for 1966. The calendar year 1966, it will be recalled, was the future annual period for which Transit’s operating expenses were estimated, and we understand that the Commission’s estimates incorporated this adjustment. If this is so, the projection of depreciation expense was inflated to the extent of $806,168 by a past, nonrecurring item. This, in turn, would necessarily misshape the margin of return that Transit was ultimately allowed. We have, in another context, delineated the principle that, to avoid this very kind of distortion, deductions for depreciation must be maintained in reasonable relationship with the service period of the property depreciated. To reach the closest possible accord with that principle, depreciation deficiencies, once ascertained, must ordinarily be amortized over such remaining life as the properties involved may happen to have. We speak now of the treatment of the $806,168 for ratemaking purposes, and not of its possible defrayal by Transit’s riders. In the understanding that the Commission included that part of the depreciation deficiency in its computation of Transit’s expenses for the future test period, we say that its action in doing so was erroneous. Recovery from Farepayers Petitioners do not contend that a depreciation deficiency cannot normally be levied against ratepayers, but encompassed, we thing, within their objection to deduction of the $806,168 from the court-ordered reserve is the query whether Transit’s farepayers now have such an obligation. It was “clear to the Commission that this deficiency in depreciation charges should equitably be made up by charges to the ratepayer,” but to us that is not nearly so evident. We would agree that if Transit’s investors have not recouped the portion of their investment reflected by the deficiency, the Commission, weighing the equities, might in some reasonable manner impose some or all of the burden upon the riders. We encounter difficulty, however, because although the Commission undertook to determine whether the burden should “equitably” fall upon investors or farepayers, it nowhere appears to have ascertained whether Transit’s investors have already been reimbursed for the diminished value of their investment. We faced a cognate problem in Washington Gas Light Company v. Baker, where PUC, in fixing the utility’s rate base, deducted an allocable portion of its book reserve for depreciation rather than straight-line depreciation accruals. PUC acted upon the conclusion that the utility’s revenues had been reduced because the annual charges for depreciation during most of the life of the property were too low, and that “it would be inequitable for investors to bear the burden a second time by, for example, charging the undepreciated amount to earned surplus.” We pointed out that “Despite this premise, the record contains no evidence as to whether earnings during the life of this property sufficiently exceeded the fair rate of return to compensate investors for the inadequate depreciation charges. We recognize that the legality of past rates may not be challenged, and that past excessive earnings belong to the Company just as past losses must be borne by it. That is not to say, however, that when the Commission purports to act on the equities of the situation, and awards higher rates because of past inequities to investors, it must not support the factual premise upon which it builds by evidence in the record. ‘Elaborate calculations which are at war with realities are of no avail.’ ” The same considerations are involved here. The Commission made no finding as to whether Transit’s investors have received remuneration for the depreciation undercharges from actual earnings over and above fair return while the properties were in service. Nor has our attention been directed to any evidence upon which a considered judgment in that regard could be attained. We cannot approve as equitable an arrangement based upon a determination that does not reflect a consideration of factors crucial to an informed decision as to where the equities really lie. Our remand of Order No. 564 will provide an opportunity for the Commission to explore the matter, make findings, and reach a just conclusion responsive to the guidelines we have discussed. Resort To The Court-Ordered Reserve Should the Commission find that Transit’s investors remain unreimbursed for the underaecrued depreciation, it would again face the question whether repayment could be accomplished by resort to the court-ordered reserve — a course petitioners vigorously oppose. We reiterate, however, that the $806,168 represented depreciation accruing during an era gone by, and that to no extent was it an advance charge for depreciation to arise only in the future. And we rule that the court-ordered reserve is a legitimate source for full and immediate rectification of any unreimbursed deficiency that might be found. In our 1963 Bebchick decision, we held that a fare increase PUC had granted Transit was unwarranted, and this made necessary a disposition of the excess in fares that Transit had collected while the increase was in effect. It was “not feasible,” we said, “to require refunds to be made to individuals who paid the increase,” but “[nevertheless, the amount realized by Transit from the increase must be utilized for the benefit of the class who paid it, that is, those who use Transit.” To accomplish this, we continued Transit must establish a fund, or set up an account or reserve, in an amount equal to the increase. We specified that “[t]he utilization and disposition of the fund, or the special account or reserve, as the case may be, shall be left to the discretion of the Commission having regulatory authority with respect to Transit, provided such discretion is exercised consistently with the purpose of benefiting Transit users in any rate proceedings pending or hereafter instituted.” Thus the special reserve, to which the Commission within these limitations was at liberty to resort, soon thereafter came into being. The deficiency in Transit’s depreciation reserve developed because too little depreciation had been charged to fare-payers in 1963 and prior years. The Commission, in correcting the deficiency of $806,168 from the court-ordered reserve, endeavored to place the charge, as nearly as could be, where it should always have been. As the Commission said, “[t]he logic of charging this deficiency off against the court-ordered reserve lies in the fact that just as the credits in the court-ordered reserve represent a build-up over a period of years prior to the current year, so does the depreciation reserve deficiency represent a build-up in past years’ depreciation charges to be ma