dissenting: This $525 million proposal is the largest utility financing in New Hampshire history. *649Moreover, it represents but a fraction of the $1.7 billion investment of Public Service Company of New Hampshire (PSNH or company) in the first unit of the ill-starred Seabrook nuclear plant, currently projected to cost $4.7 billion. In sanctioning this proposal, the court paves the way for an extraordinary and potentially disastrous increase in PSNH rates. The repercussions of this decision will extend for decades.
We are convinced that the order of the public utilities commission (PUC or commission) is “unreasonable” within the meaning of RSA 541:13. The PUC committed errors of law in (1) failing to assess the reasonableness of the projected rate increases, and (2) failing to fully evaluate the possibility of PSNH’s bankruptcy and reorganization. We would vacate the PUC’s order and remand this case to the commission for a thorough investigation of and comprehensive findings relating to these issues.
In evaluating the evidence presented in this case, we have been forced to maintain a critical — even skeptical — frame of mind. Experience dictates this approach. If Seabrook’s history has taught us anything, it is that the projections and estimates offered by PSNH officials and witnesses — and frequently relied upon by both the commission and the court — almost always prove to be wrong.
The following table illustrates the evolution of PSNH’s Seabrook project estimates:
Estimate Estimate Date ($ million) Commercial Operation Date
Unit ] Unit 2 Total Unit 1 Unit 2
Feb-72 486 486 973 11/79 11/81
Mar-73 570 570 1140 11/79 11/81
Aug-73 587 587 1175 11/79 11/81
Jun-74 650 650 1300 11/79 11/81
Mar-75 772 772 1545 11/80 11/82
Dec-76 1007 1007 2015 11/81 11/83
Jan-78 1360 995 2355 12/82 12/84
Jan-79 1309 1301 2610 4/83 2/85
Apr-80 1527 1593 3120 4/83 2/85
Apr-81 1735 1825 3560 2/84 5/86
Nov-82 2540 2580 5120 12/84 3/87
Dec-82 2540 2709 5249 12/84 7/87
Jan-84 5070 5030 10100 4/87 ?
Mar-84 4550 4452 9002 7/86 12/90
Apr-84 4100 2760 6860 2/86 7/88
4479
*650Exhibit 63, table 1.1. PSNH has not been held accountable for this abysmal record. Indeed, the company has used its wildly inaccurate predictions to intimidate the commission. For example, in a rate case before the PUC in 1977, PSNH witness D. N. Merrill testified that “even assuming low load growth estimates, the company would be unable to satisfy consumer demand for energy after 1982 if [both Seabrook units] were not completed. ‘The inevitable result,’ Merrill told the commission, ‘would be enforced power rationing with blackouts.’” LUCC v. Public Serv. Co. of N.H., 119 N.H. 332, 344, 402 A.2d 626, 634 (1979).
In the instant docket, as in the past, the company’s evidence did not go unchallenged. Expert witnesses testified that PSNH had once again underestimated costs. Several of these experts have proven track records as prognosticators in the Seabrook project. For example, in 1981 witness Richard Rosen, relying on a statistical technique based on average trends in the nuclear power industry, estimated that both Seabrook units would cost about $9 billion. In early 1982 witness Paul Chernick, using a similar methodology, projected an in-service date for Unit I of March 1986 and a total cost for both units of $9.6 billion. In contrast, at the time these forecasts were prepared, PSNH was predicting a Unit I in-service date of February 1984, and a $3.56 billion cost for both units.
The PUC nonetheless repeatedly chose to discount the testimony of intervenor witnesses, and to endorse PSNH projections. According to the commission, company estimates are inherently “more credible” than estimates based on the experiences of other nuclear units. See Re Public Service Company of New Hampshire, 66 PUR4th 400, 403, 406 (N.H.P.U.C. 1985). On nearly every significant issue, the commission found in the company’s favor. The PUC apparently refuses to learn from the past.
This case is fraught with uncertainty. The total project cost, completion date, and amount of future rate increase are unknowns. We are certain of two things, however. First, the proposed financing represents an enormous sum of money borrowed at a staggering cost. This sale of $525 million in third mortgage deferred interest bonds and pollution control revenue bonds involves a two-year capitalization of interest at a 23% rate. The financing will yield but $321.8 million in net cash proceeds; the remaining $203.2 million represents interest to investors and other financing costs. According to Kathlyne Hadley, assistant treasurer of PSNH, the “associated annual levelized revenue requirement” for the financing is $86.46 million, which amounts to $7.2 million per month in interest. In the parlance of the investment world, these are “junk bonds.” Moreover, *651this is not the last trip to the well. The company projects that between 1987 and 1989 it will need to borrow an additional $343 million through the issuance of debt instruments in the securities markets in order to meet its capital requirements.
Second, if Unit I is completed PSNH rates will escalate, and ultimately every person and organization in the State will be affected. It will become decidedly more expensive to rent an apartment, own a house, or run a business. As the commission itself found, the rate increase “will create hardship, some economic disruption, and may result in the loss of future load.” Re PSNH, 66 PUR4th at 423. Yet neither the commission nor the court has confronted these realities. The PUC and this court have permitted a foundering utility to borrow the better part of a billion dollars without determining whether the people of this State or the utility itself can withstand the harsh economic effects that lie ahead.
We differ with the majority of the court on four issues. The first is the standard of review. This court has an affirmative duty under RSA 541:13 to scrutinize commission orders to assure that they contain neither legal errors nor unreasonable factual findings. The second issue involves the role of the PUC. New Hampshire’s regulatory scheme obligates the commission to take action where necessary to protect consumer interests during each phase in the development of a power plant, including the financing phase. The commission can act now, before this money is borrowed, before it finds itself facing a choice between a utility bankruptcy and oppressive rate increases.
The third issue concerns the rate projections presented in this case. The commission failed to evaluate the implications of these projections, stating that it could do so only in the context of a prudence investigation in a future rate case. This failure constituted an error of law.
The final issue pertains to bankruptcy. A thorough inquiry into the feasibility of a PSNH reorganization under chapter 11 of the Bankruptcy Code was essential in this case. The commission assumed that it could evaluate the alternatives to the proposed financing without considering bankruptcy, and did not conduct a detailed inquiry into reorganization. This failure was also an error of law.
I. Standard of Review
The legislature has delegated to the commission the task of determining, based upon subsidiary findings of evidentiary facts, whether the proposed financing serves the public good. RSA ch. 369. In the present challenge to the commission’s findings and order, *652RSA 541:13 defines both the appellants’ burden of proof and the court’s power to review commission findings and orders. Orders may be vacated for errors of law or when it appears “by a clear preponderance of the evidence” that the order is “unjust or unreasonable.” RSA 541:13.
The reasonableness of the order depends upon the validity of the underlying evidentiary findings. Because of the commission’s specialized role, its findings are accorded a legislative presumption of reasonableness: “all findings of the commission upon all questions of fact properly before it shall be deemed to be prima facie lawful and reasonable.” Id. The commission nonetheless is obligated to set forth its reasoning and conclusions so that the court may effectively review its “methods, findings and order.” LUCC v. Public Serv. Co. of N.H., 119 N.H. 332, 341, 402 A.2d 626, 632 (1979). See RSA 363:17-b. On appeal, the burden
“rests with the appellants to satisfy the court that upon the evidence the findings are erroneous. The fact that they are merely prima facie disposes of the contention that they were intended to stand, like the findings of a jury, if supported by any evidence. The legislature plainly intended to submit to the appellate tribunal the validity upon the evidence of the commission’s findings .... The term ‘prima facie’ imports merely that there was evidence upon which the finding could be made. Whether the findings should be made is the question on appeal.”
Grafton Etc. Co. v. State, 77 N.H. 490, 504, 93 A. 1028, 1033 (1915) (citation omitted).
Among the court’s functions in this case is ensuring that the commission has exercised its regulatory discretion within the scope of statutory authority and consonant with legislative policy. Application of the statutory standard “consistent with the public good” requires interpretation of that phrase. Through judicial review, the court ensures that the commission’s understanding of the public good does not contravene the meaning of the public utilities statutes.
Upon review, we “may inquire whether the agency’s decision was fairly based on a consideration of all relevant factors.” Appeal of Concord Natural Gas Corp., 121 N.H. 685, 693, 433 A.2d 1291,1296-97 (1981). The utilities statutes do not prescribe a precise formula for making a public good determination. Prior judicial decisions interpreting the language and purpose of the utilities statutes have established, however, that the commission’s determination must rest on proper recognition of consumer and investor interests, see Petition of N.H. Gas & Electric Co., 88 N.H. 50, 184 A. 602 (1936), and *653must encompass specific findings of need and economic comparisons. See Appeal of Seacoast Anti-Pollution League, 125 N.H. 708, 490 A.2d 1329 (1984); Appeal of Easton, 125 N.H. 205, 480 A.2d 88 (1984). When a review of the commission’s methodology reveals a failure to consider all relevant factors, this court must vacate the commission’s order for error of law. Boston & Maine R.R. v. State, 97 N.H. 380, 384, 89 A.2d 764, 767 (1952). “The public, as well as the parties, is entitled to a finding of the public good on a hearing without error of law.” Parker-Young Co. v. State, 83 N.H. 551, 560, 145 A. 786, 791 (1929).
In this case, the commission’s misconception of its role and misinterpretation of the public good standard have led to decision-making tainted by errors of law with respect to rates and bankruptcy.
II. Role of the Commission
The PUC’s primary responsibility is to protect the ratepaying public’s interest in adequate utility service at reasonable rates. Its secondary responsibility is to ensure that utility investors obtain a reasonable return on their investments. The PUC has broad statutory powers to carry out these duties. It must fully investigate all aspects of a proposed security issuance, including the purpose;? of the financing. If the financing will not serve the public good, the PUC must reject the proposed financing or place necessary restrictions on its approval. In this case, the commission failed to thoroughly investigate two issues, the effects of rate shock and the likely results of a PSNH reorganization, the exploration of which was essential to a determination of whether the proposed financing is consistent with the public good.
RSA 369:1 provides chat a public utility may not issue long-term securities without the approval of the PUC. RSA 369:1 and :4 provide that the PUC is responsible for determining whether the proposed issue and sale of securities is “consistent with the public good.” PUC approval “shall extend to the amount of the issue authorized and the purpose or purposes to which the securities or the proceeds thereof are to be applied.” RSA 369:1.
The object of RSA chapter 369 is to avoid over-capitalization; i.e., to limit that part of a utility’s capital investment represented by its capital stock and debts oí a permanent character. The legislature’s main object in providing for PUC control of a utility’s capital structure is “to establish and preserve a proper base for regulation of rates and service.” State v. New Hampshire Gas & Electric Co., 86 N.H. 16, 24, 163 A. 724, 728 (1932).
Preserving a proper base for regulation serves two purposes. The primary purpose is protection of the ratepayers’ interest in proper *654service at fair cost. Petition of N.H. Gas & Electric Co., 88 N.H. 50, 57, 184 A. 602, 607 (1936). This interest is jeopardized when capitalization threatens to become so high that consumers will suffer a diminution in service or oppressive rate increases. Id.
The secondary purpose of preserving a proper base for regulation is protection of investors’ interests in obtaining a fair return. See generally Petition of Derry Electric Co., 180 A. 697 (N.H. 1935), withdrawn, 88 N.H. 46, 184 A. 356 (1936). An investor can obtain “only such return as regards the rights of the consumer to reasonable rates and service.” Id. at 702. Hence, investor interests are endangered when completion of a utility project threatens to cause such an extraordinary rate increase that a significant amount of the company’s investment in the project must be excluded from the rate base.
In this case, the commission ignored the distinction between its primary duty to the public and its secondary duty to investors. In its April report the PUC stated that its role was to “balance the interest of ratepayers for electric service at reasonable rates, the interest of investors who are entitled to a reasonable return on [the utility’s] prudent investment and the overall financial integrity of the public utility.” Re PSNH, 66 PUR4th at 434. Such balancing, which gives equal weight to consumer and investor interests, runs contrary to the primary purpose of RSA chapter 369. The scheme of utilities regulation demands that when competing interests cannot be reconciled, the commission must give precedence to the interests of the consumer.
The commission asserts that RSA 363:17-a requires a “balancing in the decisional process of investor, consumer, and utility aspects of the public interest.” Re PSNH, 66 PUR4th at 434. RSA 363:17-a provides that the commission must serve as “the arbiter between the interests of the customer and the interests of the regulated utilities.” An “arbiter” is a “person with power to decide a dispute: judge.” Webster’s Ninth New Collegiate Dictionary 99 (9th ed. 1983). The “PUC as arbiter” statute thus emphasizes that the commission has an affirmative duty to decide disputes between utilities and their customers. The statute does not relieve the commission of its fundamental obligation to protect the public.
The PUC is not “an umpire blandly calling balls and strikes for adversaries appearing before it; the right of the public must receive active and affirmative protection at the hands of the Commission.” Scenic Hudson Preservation Conference v. FPC, 354 F.2d 608, 620 (2d Cir. 1965) (discussing role of Federal Power Commission). If the completion of Seabrook Unit I creates a dilemma whose twin horns are unreasonable rates and an unreasonable rate of return, then the *655PUC must choose the course least harmful to ratepayers. This choice, moreover, would withstand constitutional scrutiny. The public interest can justify a good deal of financial harm to utility investors without violating the State or Federal Constitution. See Drobak, From, Turnpike to Nuclear Power: The Constitutional Limits on Utility Rate Regulation, 65 B.U.L. Rev. 65 (1985).
The commission assumes that approval of this financing and the subsequent completion of Seabrook will produce a rate structure that will not be oppressive to ratepayers and at the same time will ensure corporate survival. We believe that the amount of money invested in Seabrook and the high cost of borrowed capital will render such a rate structure elusive at best and probably unattainable. This is an extraordinary financing that demands extraordinary regulatory skills and safeguards.
In Appeal of Easton, 125 N.H. 205, 480 A.2d 88 (1984), this court directed the PUC to engage in a thorough investigation when determining whether to approve a proposed utility securities issue. We have long held that “the PUC has a duty to determine whether, under all of the circumstances, the financing is in the public good— a determination which includes considerations beyond the terms of the proposed borrowing.” Id. at 213, 480 A.2d at 91. “[Legitimate matters for consideration under RSA chapter 369” include “whether the uses to which the loan will be put can be economically justified compared to other options available to the [utility]” and “whether the capitalization of [the] utility is jeopardized” by the proposed financing. Id. at 212, 213, 480 A.2d at 91.
The PUC has statutory authority to deny a financing request or to put restrictions on a proposed financing. RSA 369:1 states:
“[PUC] approval shall extend to the amount of the issue authorized and the purpose or purposes to which the securities or the proceeds thereof are to be applied, and shall be subject to such reasonable terms and conditions as the commission may find to be necessary in the public interest....”
In Easton, we reiterated that in approving a utility financing the PUC may attach reasonable conditions to a securities issue where necessary in the public interest. 125 N.H. at 213, 480 A.2d at 91. Although the commission “may not directly determine and impose upon the utility a financial structure of its own devising,” Petition of N.H. Gas & Electric Co., 88 N.H. 50, 58, 184 A. 602, 607 (1936), it has long been the practice of the commission, in denying a utility’s financing request, to suggest alternative proposals that might be acceptable to the commission. See Grafton Etc. v. State, 77 N.H. 490, *656497, 93 A. 1028, 1030 (1915). Thus, the commission could have rejected PSNH’s request for authorization to issue $525 million in securities if it had found that the security issuance or the underlying purpose of the issuance, the completion of Seabrook I, was not in the public good. In addition, the PUC could have granted its approval subject to various terms and conditions.
The PUC’s power to impose “reasonable terms and conditions” on approved financing proposals can be applied creatively. The commission was at liberty to accept the recommendations of Commissioner Aeschliman and “authorize the financing under conditions which protect the ratepayers from full cost recovery through rates and from further risks if the Seabrook plant cannot be completed.” Re PSNH, 66 PUR4th at 443 (Aeschliman, Comm’r, sep. op.). Commissioner Aeschliman recommended that the PUC impose two conditions limiting ratepayer exposure:
“(1) Cost recovery from ratepayers will not be approved for the equity portion of the cost of financing the Seabrook investment that is determined to be excess capacity.
(2) Cost recovery from ratepayers will be limited to those expenditures which were prudently incurred prior to the date of this order in the event that Seabrook I does not become operational and that RSA 378:3G-a is found to be unconstitutional.”
Id. The commissioner further recommended that the PUC disallow the third mortgage on existing utility generating assets. Id.
In Appeal of Public Service Company of New Hampshire, 122 N.H. 1062, 454 A.2d 435 (1982), a majority of this court called into doubt the PUC’s power to reject or place restrictions on a proposed utility financing where the purpose of the financing was to permit continued Seabrook Unit II construction. The court held that the PUC’s decision to prohibit PSNH from using the proceeds of future securities issues for the construction of Unit II was a taking of PSNH’s property without just compensation in violation of part I, article 12 of the New Hampshire Constitution. The majority stated that since PSNH had been granted a site and facility certificate for construction of Seabrook Units I and II under RSA chapter 162-F, the company had a “vested” right to construct Unit II. Id. at 1068-69, 454 A.2d at 438-39. The majority noted:
“In this state, the common-law rule is that ‘an owner, who, relying in good faith on the absence of any regulation which would prohibit his proposed project, has made substantial construction on the property or has incurred substantial liabilities relating directly thereto, or both, *657acquires a vested right to complete his project in spite of the subsequent adoption of an ordinance prohibiting the same.’ ”
Id. at 1069, 454 A.2d at 439 (citing Henry and Murphy, Inc. v. Town of Allenstown, 120 N.H. 910, 912, 424 A.2d 1132, 1133-34 (1980)).
We dissented in that case, stating that “RSA chapter 162-F, which established a procedure for the issuance of certificates of site and facility, does not limit the PUC’s authority to impose conditions on a utility’s issuance of securities.” Appeal of PSNH, supra at 1078, 454 A.2d at 444 (King, C.J., and Batchelder, J., dissenting). We maintained that PSNH did not have a vested right to complete Seabrook Unit II. “PSNH received its certificate of site and facility after the adoption of RSA 369:1. Therefore, PSNH knew when it received the certificate that it could not issue securities to fund the construction of the Seabrook plant without PUC approval.” Id. at 1078, 454 A.2d at 445.
We also maintained that the PUC must have continuing control over the issuance of securities by utilities because of the relationship between the cost of a project and the fair rate of return to which utilities are entitled. We concluded: “The court’s decision renders the PUC a rubber stamp for any financing proposal advanced by a utility. It gives PSNH a blank check to obtain as much money as it desires from the issuance of securities and to spend it exactly as it wants.” Id.
The court’s holding in Appeal of PSNH had a short life. In Appeal of Easton, 125 N.H. 205, 480 A.2d 88 (1984), this court unanimously recognized that the PUC has the power to regulate utility financings even where the underlying purpose of the financing is construction of an approved power plant. The court stated: “Appeal of Public Service Company of New Hampshire does not stand for the proposition that the PUC’s authority under RSA chapter 369 is limited to the determination of whether the terms of the proposed financing are in the public good.” Id. at 213, 480 A.2d at 91. Thus, the court affirmed the PUC’s traditional power to consider whether the underlying purpose of a proposed financing is in the public good and to reject or restrict the proposed securities issuance if necessary.
The process of adding a bulk power facility to the rate base occurs in three phases: siting, financing, and rate-setting. The commission regulates utility activity during each of these phases. During the siting phase, the commission’s inquiry is governed by RSA chapter 162-F, which states that the commission may not issue a “certificate of site and facility,” RSA 162-F:6, I, unless it finds that “construction of the facility is required to meet the present and future need *658for electricity.” RSA 162-F:8, 11(a). The statute requires the commission to consider “economic factors” in evaluating electricity needs. Id. As discussed previously, RSA chapter 369 governs the financing phase.
During the rate-setting phase, RSA chapter 378 applies. In a rate case the PUC must determine and fix “just and reasonable or lawful rates.” RSA 378:7. The rates must be “sufficient to yield not less than a reasonable return on the cost of the property of the utility used and useful in the public service less accrued depreciation.” RSA 378:27.
Contrary to the view of both the majority of the court and the commission, the PUC has broad authority to assess the reasonableness of projected rate levels in each of the three phases. During the siting phase the commission may deny a certificate based on “economic factors,” such as the effect of projected rates. During the financing phase, the commission must safeguard the “public good,” which includes the public’s interest in avoiding the impact of extraordinary rate increases. During the rate-setting phase, the commission has direct authority to prescribe rates.
Thus, issuance of a site and facility certificate does not constrain the commission if a utility’s continued participation in a project later proves detrimental to the public interest:
“[A] commission that has been trusted with the power to approve new plants based on its assessment of future conditions at the time of the initial decision to build should have the power to change its original decision when new circumstances indicate that the plant will not be needed. The decision to cancel involves the same analytical process as the decision to approve.”
Pierce, The Regulatory Treatment of Mistakes in Retrospect: Canceled Plants and Excess Capacity, 132 U. Pa. L. Rev. 497, 535 (1984). New Hampshire’s statutory scheme encourages utilities to provide the commission with accurate cost and need estimates from the outset of a project; if the estimates are significantly inaccurate the commission can assert its chapter 369 power at the financing stage to protect ratepayers. The statutes ensure that the PUC will be able to take action to protect the public interest before the cost of a project becomes so high that the commission must choose between frightful rate increases and utility bankruptcy.
Thus, the PUC must act to protect ratepayers and investors in a public utility whenever the interests of either group are endangered. The PUC can act when a utility applies for a site and facility certificate for a power plant, when a utility seeks to include a com*659pleted project in its rate base, and when a utility proposes long-term financing to continue construction of a power plant.
Our conception of the PUC’s appropriate role in this case is hardly novel. Utility commissions in other New England States have not been passive in regard to Seabrook. As the appellants point out, Maine, Massachusetts, Vermont and Connecticut regulators have taken affirmative steps to protect consumers from the potential rate impact of Seabrook.
In 1984, the Maine Public Utilities Commission found that “completing Seabrook I is uneconomic for all three utilities under credible assumptions.” Re Investigation of Seabrook Involvements by Maine Utilities, No. 84-113, slip op. at 1 (Me. P.U.C. Dec. 13, 1984). In response to the commission’s order to withdraw from the project, those three utilities have attempted to sell their Seabrook interests at about fourteen cents on the dollar. It now appears that one of the three may be permitted to retain its interest in Seabrook. Re Investigation of Seabrook Involvements by Maine Utilities, No. 84-113, transcript of deliberations at 6 (Me. P.U.C. July 3,1985).
In April 1985, the Massachusetts Department of Public Utilities (DPU) stated that “‘the degree of risk associated with increased costs and potential future abandonment [of the project] . . . precludes a finding that the plant is reasonably necessary.’ ” Fitchburg Gas & Elec. L. v. Dept. of Pub. U., 483 N.E.2d 76, 81 (Mass. 1985) (quoting Order of Apr. 4, 1985, Mass. D.P.U. No. 84-152). The Massachusetts Supreme Judicial Court affirmed the DPU’s decision to place three conditions on Seabrook-related financing requests of three investor-owned utilities in that State. See id. The conditions provide:
“1. In the event Seabrook I does not become commercially operable, cost recovery from ratepayers will be limited solely to those expenditures which were prudently incurred before the date of this Order.
2. In the event that Seabrook I becomes commercially operable, cost recovery from ratepayers will be limited to the marginal costs of capacity and energy that would otherwise be faced by the utility, but in no event more than the amount which would be collected by placing the prudently incurred, used and useful portion of the cost of the plant in rate base and no less than the amount that the company would be entitled to collect if the plant were abandoned as of the date of this Order.
3. In the alternative, a company may choose to receive an as-available marginal cost rate for electricity produced *660throughout the life of Seabrook I, without a constraint on the minimum and maximum levels of cost recovery.”
Id. at 80 n.3. Since the utilities refused to comply with these conditions the financing requests were denied. The court also affirmed the DPU’s decision regarding the Massachusetts Municipal Wholesale Electric Company (MMWEC), which stated that MMWEC could issue “‘only such bonds as . . . [the DPU finds] are reasonably necessary to mitigate the adverse consequences of . . . rate shock associated with its investment to date,’ ” and that the DPU would not permit MMWEC “‘to issue bonds to pay for further construction costs’” of Seabrook Unit I. Id. at 80 (quoting Order of Apr. 4, 1985, Mass. D.P.U. No. 84-152).
Furthermore, the Vermont Public Service Board (PSB), in May 1985, ordered utilities in that State to attempt to sell their ownership interests in Seabrook, stating “that it will be cheaper for Vermont to write off all of its expenditures to date and to acquire new power from other sources than to stick with Seabrook to its completion.” Re Seabrook Station Units 1 and 2, 67 PUR4th 212, 216 (Vt. P.S.B. 1985). The board later modified this position in a proceeding to review the proposed sale by Central Vermont Public Service Corporation of its Seabrook ownership interest to Eastern Utility Associates, Inc. at fifteen percent of the investment amount. The board noted that changed circumstances required Central Vermont’s management, and not the board, to decide whether the sale is a prudent means of disengagement. Hence, it appears the Vermont utilities may now retain their Seabrook interests, at least until suitable offers are received. See In re Central Vermont Public Service Corp., No. 5045 (Vt. P.S.B. Dec. 4, 1985).
Finally, we note that the Connecticut Department of Public Utility Control (DPUC), pursuant to legislative mandate; imposed a cost cap of $4.7 billion on Unit I for purposes of the upcoming rate proceeding. See Conn. Gen. Stat. Ann. § 16-19v (West. Supp. 1985); Re Seabrook Unit No. 1, 62 PUR4th 673 (Conn. D.P.U.C. 1984). Also, in a recent proceeding the department approved a financing proposal on the condition that the Connecticut utility not help finance any other joint owner’s Seabrook I ownership interest. See Application of United Illuminating Co., No. 84-11-07 (Conn. D.P.U.C. Dec. 12, 1984). These conditions and the cost cap set an upper limit for ratepayer exposure in Connecticut.
In making this comparison we recognize that the Vermont utilities’ ownership shares are considerably smaller than PSNH’s, and that PSNH, as the principal owner, has the largest investment at risk. Nonetheless, Massachusetts, Maine and Connecticut regula*661tors, who oversee utilities that own substantial portions of Seabrook, have taken protective action on behalf of the public interest. The steps taken by other New England States to protect consumers from Seabrook rate shock highlight the New Hampshire commission’s inertia.
III. Rates
The PUC found that projected rate increases of 200% by the turn of the century are consistent with the public good. This “finding,” however, was not based on an analysis of the evidence. It was, from the beginning, a foregone conclusion. Despite the court’s efforts to force the commission to consider the likely effects of these rate increases at this juncture, the commission persists in its belief that projected rates can be evaluated only in a rale case.
The commission clings to this notion with remarkable tenacity. Before the hearings in this proceeding, we directed the commission to consider “the effect on the company’s future rates of this and any further investment.” Appeal of SAPL, 125 N.H. 708, 718, 490 A.2d 1329, 1337 (1984). In its April report the commission neither projected probable rate levels nor addressed the likely effects of substantial rate increases. The appellants identified these failings in their motions for rehearing. In denying the motions, the commission (1) claimed (without accompanying citation to the April report) that it had “evaluated the economic impact of projected rate increases,” and (2) stated that “the impact of rates on the New Hampshire economy [would] clearly be an issue to be addressed in depth in a subsequent rate proceeding.” Report and Tenth Supplemental Order No. 17,601, slip op. at 22 (N.H.P.U.C. May 10, 1985).
By order dated October 30, 1985, the court remanded this case to the commission for a, supplemental report containing
1) findings as to the reasonably probable range within which rates will be set if the proposed financing is approved and the plant completed; and
2) determinations as to the effect of such findings on the validity of the conclusions reached in the PUC’s April decision.
This effort at last bore fruit; in its Report and Fifteenth Supplemental Order No. 17,939 (N.H.P.U.C. Nov. 8, 1985), the commission specified the range of “reasonably probable rates” that will obtain when Unit I is completed. The commission provided two groups of projections. The three projections in the first group (Tables 1- 3) assume that all of PSNITs costs incurred in the construction of Sea-brook Unit I will be included in the rate base. PSNITs “baso case” *662assumes rate phase-in; inclusion of demand for power by UNITIL (the holding company of Concord Electric Company and Exeter and Hampton Electric Company); commercial operation date (COD) 10/31/86; total cost $4.6 billion; and mature availability 72%. Table 1, reflecting PSNH’s “base case” (Exhibit 99-A), shows that in nominal dollars rates will double by 1991 and nearly triple by 2002. Table 2 relies on Exhibit 99-B, which is identical to Exhibit 99-A except that no phase-in is assumed. This projection shows that in nominal dollars rates will double by 1994 and nearly triple by 2003. Table 3 is based on Exhibit 124-D (assuming no phase-in, UNITIL exclusion, COD 10/31/86, total cost $4.6 billion, mature availability 60%, Seabrook Unit II costs written off 10/31/86). It shows that in nominal dollars rates will double by 1989 and triple by 2003.
The projections in the second group, Tables 4 and 5, rely on testimony of Donald Trawicki, a financial consultant and accountant for Touche Ross & Co. Trawicki testified that up to $1.1 billion of PSNH’s Seabrook investment could be excluded from the rate base without precipitating PSNH’s bankruptcy. The rates in these projections represent “the minimum rates that could possibly be set” compatible with PSNH’s survival. Table 4, which is based on PSNH’s initial base case and which assumes a $1.0 billion exclusion, shows that rates in nominal dollars will increase by 50% by 1989 and double by 1999. Table 5, a six-year projection that assumes an optimistic 4% annual increase in sales and a $1.1 billion rate base exclusion, shows that rates in nominal dollars will increase by 50% by 1991.
Although the PUC adopted this range of projections, it reiterated its conviction that the reasonableness of projected rate levels could not be assessed in this proceeding and made no significant attempt to evaluate the implications of its findings.
In analyzing whether the projected rates were reasonable, the PUC apparently applied the chapter 378 rate-setting standards. According to the commission, reasonable rates are rates “sufficient to yield not less than a reasonable return on the cost of the property of the utility used and useful in the public service less accrued depreciation.” RSA 378:27. The commission listed four issues that must be addressed in applying this definition:
1) whether there is a need for the power to be supplied by the project;
2) if so, whether the proposed plant is the cheapest means of meeting that need;
*6633) whether the cost of the plant was prudently incurred; and
4) whether the return on the cost of the plant will be reasonable.
Report and Fifteenth Supplemental Order No. 17,939 at 38 (N.H.P.U.C. Nov. 8, 1985). The commission stated that three of these issues, 1, 2 and 4, had been resolved in PSNH’s favor in the April decision. The commission also stated that the third issue, prudence, was not properly before it, and emphasized that until a chapter 378 prudence investigation was completed the commission was powerless either to evaluate prudence issues or to determine actual rate levels. Id. at 31-32. Beyond noting that “sharp increases in rates may result in transfer of load off system or conservation induced reduction in demand,” id. at 8, the commission made little attempt to evaluate the possible consequences of the projected rate increases and concluded that its rate projections had no effect on its earlier decision.
The commission apparently believes that its chapter 369 power to impose conditions to minimize consumer exposure to rate shock is negligible. According to the commission, the concept of rate reasonableness has meaning only in the context of a prudence investigation under chapter 378:
“Electricity rate levels flow from revenues necessary to provide a reasonable rate [sic] on prudent investment— not the reverse process of first determining a rate level to derive the level of prudent investment to be substantiated by the rates. A finding in a prudency review that the market for electricity will not support the level of rates will be relevant to determine whether capital investment not fully recoverable by rates was prudent in the first instance. In this proceeding we cannot prejudge the prudency issue by imposing a cap on rates and rate base knowing that the prudency issue will be fully adjudicated in a future rate proceeding.”
Id. at 31-32.
In sum, the commission believes that reasonable rates necessarily result from application of the rate-setting formula set forth in chapter 378, and that apart from the chapter 378 framework no objective assessment of the feasibility of projected rate increases is possible. Despite Commissioner Aeschliman’s admonition that “any given level of rates is only reasonable or unreasonable relative to some standard of comparison,” Report and Fifteenth Supplemental Order *664No. 17,939 at 2 (N.H.P.U.C. Nov. 8, 1985) (Aeschliman, Comm’r, sep. op.), the commission majority expressly refused to employ a market standard. Report and Fifteenth Supplemental Order No. 17,939, supra at 31. Apparently, the commission would find any projected rate increase consistent with the public good as long as the increase was derived from the “reasonable return on prudent investment” formula.
This position is faulty. Rate increases of this magnitude cannot be evaluated in a vacuum. The protection of the public interest in just and reasonable rates cannot be left entirely to the rate-setting process under chapter 378. Nor can the commission assume that the market can bear whatever rate increases result from the rate-setting formula. As we demonstrated in part II, the commission is obligated under chapter 369 to assess projected rate levels and their effects, and to take steps to protect ratepayers from those effects where necessary in the public interest. At a minimum, the commission was required to adopt a standard of comparison to serve as a bench mark in evaluating the projected rates.
In a 1949 rate case, the public service commission relied on evidence pertaining to rates in effect or sought by the utility in other jurisdictions. Company v. State, 95 N.H. 353, 64 A.2d 9 (1949). The evidence was offered primarily to cast doubt on the reasonableness of the proposed rates and the credibility of the utility’s request. The commission noted that the rates sought would subject New Hampshire consumers to higher charges than those assessed against the same utility’s Massachusetts customers for comparable service. The commission stated that there was no difference in the value of the services, and no reason New Hampshire service should cost more. It also noted that the requested rates were higher than those filed in any other State for similar service. Id. at 362, 64 A.2d at 17.
On appeal the utility contended that this evidence was improperly considered. We found to the contrary:
“While the jurisdiction of the Commission is confined to intrastate rates, it is not required to fix them in a vacuum, or to close its eyes to the company’s conduct of its affairs in neighboring states where comparable conditions are to be anticipated. This is not to say that rates in New Hampshire may be fixed according to a standard of rates effective elsewhere. New Hampshire rates must reflect the proper and reasonable costs of business here. But a comparison of rates which reveals a differential not reasonably to be expected under conditions commonly regarded as comparable may be thought to make suspect the reasonableness of the various components relied upon to sustain the *665higher rates. No exposition of factors which might account for the apparent differential was undertaken by the company. ... In this situation, the evidence was pertinent in considering whether the company had sustained its burden of establishing the reasonableness of the rates requested in New Hampshire.”
Id. at 362-63, 64 A.2d at 17 (emphasis added).
In this case there is evidence that projected PSNH rates will be considerably higher than those prevailing outside the ambit of its monopoly. Gregory Palast, a utility rate economist for Union Associates of Boston, stressed the importance of comparing PSNH rates to those of neighboring jurisdictions, and submitted tables juxtaposing PSNH rates with commercial and industrial rates in selected metropolitan areas. Witness Trawicki agreed that the “Commission should be concerned with the comparative rates between Public Service here and the adjoining utilities in our neighboring states,” and submitted evidence that PSNH rates would substantially exceed projected NEPOOL (New England Power Pool) average rates following completion of Unit I. Commissioner Aeschliman compared the projected PSNH and NEPOOL rates and concluded that between 1987 and 1991 the differential would be 4-6<P/kwh. Report and Fifteenth Supplemental Order No. 17,939 at 10, table 2 (N.H.P.U.C. Nov. 8, 1985) (Aeschliman, Comm’r, sep. op.). Assuming a “cost to go” of one billion dollars and full cost inclusion in rate base, by 1987 PSNH customers will be paying electric bills fifty percent higher than those paid by customers of other New England utilities. Id. at 9, table 1.
The commission acted unreasonably in ignoring this evidence. The evidence provided the commission with an opportunity to test the results of its economic analysis against an objective standard. There are few fixed points of reference available to utilities forecasters. Chicken and egg problems abound; a set of rate projections is needed to generate a load forecast, and a load forecast is needed to generate a set of rate projections. No assumption can be regarded as immutable, and every assumption must be subject to reexamination and verification. The rate projections cast doubt on many of the commission’s assumptions. We are less confident than the commission, for example, that an average annual growth rate of 4.25% in industrial electricity sales between 1983 and 2003 (which the commission projects) and rate increases of 200% over the same period are compatible. In refusing to assess the implications of its rate projections, the PUC missed an important opportunity to verify the soundness of some of its other assumptions.
*666“It is . . . not sufficient to say that since you have found that the load forecast is reasonable, and that the plant is needed, that the level of rates that results is reasonable per se. It is necessary to test the results — the level of rates that result from a given level of investment — against an objective standard to see if in fact the results make any sense. This is really a process of validating your economic analysis and if you cannot validate that analysis against an objective standard, it should tell you that there is something wrong with your economic analysis.”
Id. at 3.
We do not contend that a comparison with NEPOOL average rates was required in this case, nor do we regard the ^S'P/kwh differential standard of reasonableness proposed by Commissioner Aeschliman as the sole feasible means of addressing this issue. The commission has considerable discretion in this regard. We are convinced, however, that the commission was required to adopt some standard of comparison, based on rates prevailing in actual markets, as a means of assessing the reasonableness of the projected rates. Without such a standard, both the commission and this court on review operate in a vacuum.
The majority of this court, in affirming the commission’s order, adopts the view that the PUC had no obligation in this proceeding to address the implications of the projected rate increases. The commission cannot act now to protect consumers, it seems, because it can do so later in a rate case:
“The term ‘reasonable rate’ must be understood as referring to the result of the ratemaking process... .
. . . [T]he reasonableness of a rate should not be determined either independently of the process by which expenses, rate base, and rate of return are set, or after that process has been completed ....
Indeed, any attempt to judge reasonableness apart from that process would entail redundancy and risk both illegality and unconstitutionality.”
We respectfully disagree. The commission’s chapter 378 power to set rates neither circumscribes its authority nor diminishes its obligation under chapter 369 to protect the public interest. We respond to the majority’s charges of redundancy and the risks of illegality and unconstitutionality in turn.
The majority asserts that our position entails redundancy because *667“it is difficult to think of any consideration bearing on reasonableness that may not be raised appropriately” in the rate-setting process. In other words, because the commission in the upcoming rate case will be able to consider all pertinent factors in setting reasonable rates, any PUC effort in this proceeding to prevent oppressive rate increases would be duplicative and premature.
Two errors taint this argument. First, the argument ignores the distinction between the PUC’s role in the rate case and its role in a financing case. In a rate case the commission deals post facto with utility investments. In most rate cases, the commission, by applying the prudent investment test, the used and useful test, and the rate-setting formula, is able to accommodate both consumer and investor interests. That is, the commission is in a position to permit investors to recover a reasonable return without subjecting consumers to excessive rate increases.
In the Seabrook rate case, however, the commission will find itself in a much more difficult position. The commission will have to choose between excluding a significant part of the Seabrook investment from the rate base and burdening consumers with large rate increases. A substantial rate base exclusion will result in utility investors failing to earn a return on their investments. Oppressive rate increases will result in job loss, industry migration, and reduced living standards. The commission will not be able to prevent these losses; it will only be able to allocate them between investors and consumers.
In a financing case the PUC has a different function. The money involved has not yet been borrowed. The PUC, with foresight, has the power to limit potential losses before they occur. Thus, the PUC should have balanced the consequences of approving the proposed financing against the consequences of disapproving it. The losses that would have resulted if the PUC had not approved this additional financing may have been smaller than the losses that will flow from approval. In order to make an informed judgment on this issue, the PUC had to assess the reasonableness of the projected rates.
An examination of rate projections was essential to provide an indication of the losses that may ensue in the later rate case. If after such an investigation the PUC had concluded that a substantial rate base exclusion was inevitable, the commission could have used this information in assessing the relative costs and benefits of approving the proposal. If a rate base exclusion on the order of one billion dollars will be the only available means of preventing the economic costs of oppressive rate increases, then the commission should have *668included this assessment in its report. Instead, the commission ignored the probable effects of the projected rate increases, and justified its position by invoking the possibility of significant rate base exclusions. This tactic is both disingenuous and irresponsible. If inevitable losses lie ahead, the commission should acknowledge them now, and plan accordingly.
A second error underlies the majority’s charge of redundancy. The charge is premised on the dubious assumption that up to $1.1 billion of PSNH’s investment can be disallowed without precipitating bankruptcy. This assumption is founded on an uncritical acceptance of witness Trawicki’s projections. According to the majority of the court, these projections “establish a significant datum for purposes of the commission’s Easton responsibility.” This is an extraordinary assertion. In our view the Trawicki projections are far too speculative to constitute a vital cog in the majority’s argument. Had •they been subjected to the same degree of scrutiny by the commission as were many of the appellants’ projections, they would have been discounted entirely. The Trawicki projections are based on unreasonably optimistic assumptions (including $4.5 billion total cost, COD 8/1/86, and mature capacity 72%), and assume, inter alia, that lenders will not call their loans when PSNH breaches financial maintenance tests and other protective covenants. Moreover, even under Trawicki’s projections, PSNH rates will increase by 50% within a six-year period. Report and Fifteenth Supplemental Order No. 17,939 at 23 (N.H.P.U.C. Nov. 8, 1985) (Aeschliman, Comm’r, sep. op). We are far less certain than the majority that the PUC will be able to sufficiently protect consumers in the upcoming rate case.
The majority next asserts that our position risks “illegality” because “the relevant statutes mandate the provision of a reasonable return on net cost of used and useful property. The application of any ratemaking standard without reference to such a return would be inconsistent with the statutory mandate.” This contention is premised on the erroneous notion that because the reasonableness of rates under chapter 378 can be evaluated only in terms of the “reasonable return on prudent investment” formula, reasonableness of projected rates under chapter 369 cannot be assessed at all. The only statute directly “relevant” in this case is RSA chapter 369, which provides an independent mechanism to ensure that rates do not become excessive.
Chapters 378 and 369 exist for different purposes and mandate different legal standards. This is a financing case, not an inchoate rate case. Chapter 378 requires the PUC to use hindsight; chapter 369, foresight. Chapter 378 mandates application of a formula designed to ensure investors a reasonable return; no such formula *669constrains the PUC in a “public good” inquiry under chapter 369. Rates that emerge from the chapter 378 process may be inconsistent with the public good. Apart from its chapter 378 function, the PUC had a duty under chapter 369 to consider the possible economic and social effects of the proposed security issuance, and had the authority to take steps to protect ratepayers from such effects. See Appeal of SAPL, 125 N.H. 708, 490 A.2d 1329.
The majority also asserts that our analysis risks unconstitutionality because “any criterion of reasonableness that might be applied independently from the [rate-setting process] would run the risk of unconstitutionality by inviting the fixing of rates without regard to the balancing of interests.”
We reiterate: this is not a rate case, and the money at issue has not been borrowed. Once a utility makes an investment, the constitution may require that rates be set without regard to “independent” criteria. In this case, however, there is no PSNH property interest subject to confiscation. At issue is a proposed security issuance. No constitutional provision guarantees PSNH the unlimited right to borrow money.
Finally, we take issue with the majority’s discussion of rate differentials. The majority concedes that a rate differential “has value as a critical tool,” but asserts that its use in this proceeding would be improper. According to the majority, “[t]he 4-5<P/kwh rate differential is really a standard for predicting tolerable price-demand effect.... If it is a sound measure, it should be incorporated into the analysis of need for power. If it were so incorporated, its reappearance as a separate criterion of reasonableness would be redundant.”
We disagree. There is nothing sacrosanct or exclusive about the issues — need for power, comparative incremental cost and financial feasibility — the PUC considered in this case. Each bears on the public good; each represents a different way of examining the same problem. As the majority states, “because of the interrelationship of all the issues, the conclusion on any one ... is necessarily dependent on the resolution of the others.” Adoption of rate level assumptions, a necessary step in assessing the need for power, does not preclude analysis of the projected rate levels. Protecting the consumer’s interest in reasonable rates has long been a primary concern of the PUC in financing cases. See Petition of the N.H. Gas & Electric Co., 88 N.H. 50, 57, 184 A. 602, 607 (1936). Nothing in chapter 369 or elsewhere prohibits the PUC from evaluating the possible effects of the projected rate differentials in this case. We see no point in requiring the commission to adopt rate projections if the projections can play no role in the evaluation of the proposed financing.
*670The commission has utterly failed to come to grips with the rate issues in this case. The PUC has abnegated its chapter 369 responsibility to protect the public interest in failing to adequately assess the reasonableness of rates, thereby committing an error of law within the meaning of RSA 541:13.
IV. Bankruptcy
Bankruptcy is an unavoidable issue in this case. Determination of whether a proposed financing serves the public good requires a comparison of the consequences of approving the proposal with the consequences of not approving it. All parties in this case agree that denial of this financing would result in a PSNH bankruptcy and reorganization under chapter 11 of the Bankruptcy Code. Such a reorganization would have profound effects on New Hampshire. Yet these effects may be less harmful than the consequences of the projected rate increases. Thus, any assessment of the relative economic desirability of further participation in Seabrook requires full consideration of a PSNH chapter 11 reorganization. In failing to fully evaluate the bankruptcy alternative, the commission committed an error of law.
In Appeal of SAPL, we stated that Easton “will require the commission to determine the relative economic desirability of allowing or disallowing the company’s continuing participation in construction of the first Seabrook reactor, before it rules on the anticipated third or Newbrook financing.” Appeal of SAPL, 125 N.H. at 718, 490 A.2d at 1336. “[T]he ultimate . . . determination,” we wrote, “must rest on the record of a substantial inquiry.” Id. at 718, 490 A.2d at 1337. At the outset of this proceeding, the commission recognized that “an evaluation of the long:term alternatives to completion of Seabrook Unit I” was required, and the parties agreed that the scope of the proceedings included the issue of bankruptcy.
We reject the majority’s position that consideration of bankruptcy is irrelevant and the majority’s concomitant endorsement of the “analytic standard” the commission employed to address bankruptcy. The commission stated that its “finding that PSNH’s proposed financing ... is in the public good is independent of the probable bankruptcy of the company if its petition is denied.” Re PSNH, 66 PUR4th at 426 (emphasis added). This standard is legally erroneous. The commission was charged with determining whether the financing was consistent with the public good. To accomplish this the commission had to compare the financing and the resulting rates to a potential PSNH bankruptcy and its effects, in detail and with specificity. The commission’s failure to perform this analysis creates an implicit corporate survival standard; the PUC apparently *671commits itself to preserve company solvency at all costs. We do not think the commission should ensure PSNH’s corporate survival at the expense of New Hampshire ratepayers.
The commission also failed to examine alternative sources of electric generation — small power production, cogeneration, and conservation — under economic forecasts that included the effects of reorganization. Instead, the commission examined generating alternatives only in the context of incremental costs, and stated that the uncertainties of reorganization precluded forecasting. A finding that the financing is in the public good independent of a full consideration of bankruptcy is insufficient under Appeal of SAPL. Such a finding renders the analysis that follows it an empty exercise.
We find further error in the commission’s allocation of the burden of proof. The commission stated that “[t]he company has sustained its burden to prove that the proposed financing will serve the public good. Intervenors have not proved that further delay will serve the public good or that bankruptcy at this or a later time is a rational alternative to going forward with Seabrook.” Re PSNH, 66 PUR4th at 436. The burden of proof on the issue whether bankruptcy is a rational alternative is not on the appellants, but on PSNH. Such proof depends on a showing that approval of the financing is superior to the alternative — a PSNH reorganization. In requiring the intervenors to prove that bankruptcy would be preferable to completion of the plant, the commission committed legal error.
We now consider the evidentiary record. The record includes five PSNH financial scenarios which demonstrate that denial of the financing and cancellation of the plant will result in bankruptcy, and testimony by various PSNH employees that bankruptcy is not a viable alternative. Although PSNH provided scores of financial forecasts containing various economic and regulatory assumptions, the company provided no financial scenarios or forecasts based on reorganization assumptions.
The record also includes independent bankruptcy studies of other electric utilities. Re PSNH, 66 PUR4th at 426. A review of this evidence discloses two facts. First, these studies disfavor reorganization as an alternative for an electric utility. Second, this evidence, although voluminous, is not specific to Seabrook.
In addition to this evidence, the commission heard testimony from four witnesses who are not PSNH employees — Mark Vaughn, Esq., Gregory Palast, Donald Trawicki, and Dean Robert Viles. Vaughn is the co-author of a 33-page bankruptcy report prepared by the Manchester law firm of Devine, Millimet, Stahl & Branch, P.A., at the request of the attorney general and the commission. The Devine, Millimet report discussed several issues, including jurisdiction of *672the commission and the bankruptcy court, priority debts, the delay and expense of administration, the bankruptcy “stigma,” possible tax effects on municipalities, and the debt burden on a reorganized PSNH.
Although this report constituted the primary legal analysis of a possible PSNH reorganization, Vaughn testified that it did not purport to be a complete study. The report states: “While pointing out certain problem areas of a Chapter 11 proceeding, this report does not attempt to forecast or speculate on the ultimate results of a Chapter 11 proceeding.” Moreover, in response to the question, “[d]o you think . . . there might be some areas in your report that the Commission should investigate further?” Vaughn remarked, “[w]ell, the economic issues were not really dealt with by our report and I think that’s what this proceeding [is] about.”
Palast, a utility rate economist for Union Associates of Boston, stated that reorganization would have a positive effect on rates, that post-reorganization borrowings would cost less, and that a post-bankruptcy PSNH would be a financially stronger company. He also testified that a comprehensive bankruptcy study was needed.
Trawicki, a financial consultant and accountant for Touche Ross & Co., prepared a financial scenario entitled “Newbrook plan not implemented.” He stated that rates after a PSNH reorganization would be lower than the rates PSNH predicts — assuming completion of Seabrook and a cost phase-in — until the mid-nineties, but would exceed PSNH’s predicted rates thereafter. Further, he identified several risks and uncertainties posed by bankruptcy, including the expense of raising capital for new construction, investor perceived risks, future rate setting, and the revaluation of PSNH property and rate base.
In addition, Dean Viles of the Franklin Pierce Law Center, an academic expert, offered his views on bankruptcy. He stated that service to customers during a reorganization would not deteriorate because the company’s cash flow would continue, enabling the company to operate while it restructured its debts, and that the 22 to 24% cost of capital is higher than any known in a chapter 11 reorganization. Dean Viles indicated that an in-depth study of the bankruptcy issue was imperative.
The following exchange occurred during Dean Viles’ testimony.
“Comm’r Nassikas: Is it your opinion that the Commission has an adequate evidentiary record to determine whether or not the Company should file for a Chapter 11 arrangement?
*673Dean Viles: I am afraid I have to say, no.
[T]he magnitude of the questions involved here and how much is at stake for the future not only of the company but for the ratepayers and the general welfare in this state, [suggest] that more [evidence] would be desirable.”
More specifically, Dean Viles stated: “I think you need to have more advice about how a bankruptcy would actually work, . . . [The Devine, Millimet] report does not tell you what will happen, it doesn’t give you a game plan .... For example, the Commission could profit from looking at successful examples of Chapter 11.” Dean Viles also testified that the Touche Ross report was “sketchy.”
“It does not go into what might happen in the case of the reorganization after approvals were made for the New-brook funding. It doesn’t give a very clear explanation why it doesn’t take into account salvage value or the claims . . . because of ... an unfinished Seabrook plant. The Touche-Ross report does not indicate with detail the long term consequences of a reorganization.”
The testimony of these independent experts establishes that the evidentiary record on bankruptcy was inadequate, yet the commission failed to observe their recommendations. Moreover, the commission reached unsubstantiated conclusions. For example, the commission relied on witness Vaughn’s statement to conclude that “[i]t is likely that capital markets will raise the cost of borrowed capital above prebankruptcy levels.” Re PSNH, 66 PUR4th at 428. No authority was cited in the Devine, Millimet report for this proposition. This finding is dubious in light of Dean Viles’ testimony that the cost of capital in a reorganization has never reached the cost associated with the present financing.
Although the evidence pertaining to bankruptcy was scant, the commission was not obligated to rely on the parties to supply further information or to rest on the inadequate record. The commission has broad investigatory powers. RSA 365:19 states: “In any case in which the commission may hold a hearing it may, before or after such hearing, make such independent investigation as in its judgment the public good may require. . . .” In its April report the commission emphasized that it regarded Seabrook-specific data as more credible than the results of industry-wide analysis, at least in *674the engineering context. Re PSNH, 66 PUR4th at 403, 406. When the commission addressed bankruptcy, however, it failed to obtain and review sufficient Seabrook-specific evidence. Rather than engaging in or procuring further economic analysis, the commission summarily concluded that “reorganization will be more costly to ratepayers regarding reliable electric services at reasonable rates over the long term than the proposed financing to put Seabrook on line:’Id. at 437.
For the commission to have made adequate findings on bankruptcy, a comprehensive economic analysis was required. Such an analysis would have included (1) detailed long-term projections of post-reorganization cost of capital and rates, and a comparison of these projections to the projected cost of capital and rates that will result from the present financing and completion of Seabrook Unit I; (2) analysis of the economics of alternative sources of electric generation assuming a PSNH reorganization; (3) consideration of reorganization assuming approval of this financing; and (4) as the appellants argued, “analysis of what the balance sheet or capital structure of a reorganized PSNH would look like, under different assumptions as to bankruptcy court treatment....”
One commentator recently stated that although bankruptcy may be a “draconian route” for a utility, it is “an option not to be lightly dismissed.” Samuels, A Consumer View on Financing Nuclear Plant Abandonments, Pub. Util. Fort., Jan. 10, 1985, at 24-26. We do not seek to minimize the risks and uncertainties inherent in a utility bankruptcy. See Flashen & Reilly, Bankruptcy Analysis of a Financially-Troubled Utility, 59 Am. Bankr. L. J. 135 (1985). Nonetheless, because independent bankruptcy witnesses recommended further study, and because the commission was required to fully consider the relative economic desirability of a PSNH reorganization as part of the alternative to the financing, we think the commission erred in reaching its conclusions.
V. Conclusion
Approval of this financing heralds an era of unprecedented rate increases for New Hampshire ratepayers. In light of the dismal history of this project, we believe the PUC has not properly scrutinized the company’s case for further Seabrook involvement. In delaying a public good inquiry until this late date, and in further deferring inquiry into the impact of projected 100-200% rate increases on the New Hampshire economy, the commission has failed to fulfill its responsibility to protect the people of this State.
The company has consistently claimed that if this financing proposal is not approved, PSNH will seek protection from its creditors *675under chapter 11 of the Bankruptcy Code. This ultimatum makes clear the signal importance of the bankruptcy issue. Bankruptcy is not an alternative to power supply; it is, however, an alternative to this financing. The commission did not make a detailed study of this alternative. Instead, the PUC on an inadequate record found that bankruptcy was not in the public good. Once the PUC made this finding there was no alternative to approval of the financing request. This methodology flawed the proceedings and ultimately the validity of the commission’s report. Before another half billion dollars were borrowed, the possibility of a reorganized PSNH should have been explored.
An investor is under no compulsion to put his money at risk in 23% junk bonds. The ratepayer within the zone of PSNH’s monopoly does not enjoy the same freedom of choice in determining from whom he will purchase electricity. Ratepayers are powerless to prevent the excessive rate increases that result from utility over-capitalization. A ratepayer who cannot pay his electric bill will take little comfort in the knowledge that his rate was determined according to the statutory formula.
The consequences of high utility rates have not been investigated by the commission, nor has the commission taken measures to protect ratepayers. The legislature intended that the hard questions posed by this case should have been addressed before the financing was approved. Yet the commission has avoided meeting these questions head-on. We would remand for a complete inquiry as to the impact of the projected rates on the New Hampshire economy and the consequences of PSNH bankruptcy, and for imposition of such ratepayer protection measures as are found to be necessary.