Office of Consumer Counsel v. Department of Public Utility Control

905 A.2d 1, 279 Conn. 584, 2006 Conn. LEXIS 319

• Court: Supreme Court of Connecticut
• Decided: August 22, 2006
• Docket: SC 17465
• Status: Published

Opinion

Opinion

BORDEN, J.

The principal issue in this administrative appeal 1 is whether the named defendant, the department of public utility control (department), properly allowed rate recovery to the defendant Connecticut Light and Power Company (company), in a 2004 rate proceeding for an expense that had not been allowed in a 1993 rate proceeding. The plaintiff, the office of consumer counsel, claims that the department should not have allowed the rate recovery because it had not expressly promised the company, when it declined to allow the expense in the 1993 proceeding, that it could recover the expense in a later proceeding. Alternatively, the plaintiff claims that, even if the department promised deferred recovery in 1993, allowing the rate recovery in the 2004 proceeding was unjustified in light of intervening circumstances. The trial court concluded that the department properly allowed the rate recovery and dismissed the plaintiffs appeal. We affirm the judgment of the trial court.

The department, in a rate proceeding, allowed the company to recover certain projected employee pension costs. The plaintiff appealed from that decision to the trial court, which dismissed the plaintiffs appeal. This appeal followed.

The record reveals the following relevant facts and procedural history. In 1992, the company filed a request for a rate increase with the department (1992 rate filing). The 1992 rate filing included the company’s projected pension costs for fiscal years 1993 to 1995. Those costs were developed by the company’s pension actuaries and included the amortization over 18.3 years of the company’s share of the pension plan’s unrecognized net gain of $68.3 million as of December 31, 1992. 2

The department issued a draft decision on the 1992 rate filing on June 1, 1993 (draft decision). In the draft decision, the department amortized the $68.3 million in unrecognized pension gain over an eight year period instead of the 18.3 years requested by the company. As a result, the company’s pension expense for rate-making purposes was reduced by approximately $5 million per year for the three fiscal years, 1993 to 1995, for a total reduction of $15.7 million.

The company submitted written comments on the draft decision in which it advised the department that, because of the department’s reduction of pension expenses for rate-making purposes, “the [cjompany now faces a situation where its pension expense for book purposes . . . will be over $5 million higher [annually] than has been allowed in rates. To avoid the significant financial impact of this conflict, the [cjom-pany requests language to be added to the final decision that would allow recognition of a regulatory asset for the difference in pension expense created by the eight year vs. [eighteen] year amortization periods for the net unrecognized gain. That is, the final decision should specifically state that because the gain will be amortized over eight years for ratemaking purposes, all other things being equal, ratemaking pension expenses in years [nine through eighteen] will be higher than the [cjompany’s . . . booked pension expense. This will allow the [cjompany to avoid an earnings hit for the $5 million mismatch between what has been provided in rates for this item and what is reflected in its financial statements.”

The department issued its final decision on the 1992 rate filing on June 16, 1993 (1993 decision). In the 1993 decision, the department noted that, “[i]n the history of regulation in the [s]tate of Connecticut, 1993 is the worst year to be confronted with rate cases since the depression of the 1930’s. [The company], in recognition of the state of the economy, offered an alternate multi-year rate proposal intended to mitigate the rate shock of a potential 13.9 [percent] one-year rate adjustment.

“Even with a multi-year proposal, the [company’s] rate request of $358 million is the highest in its history. Could the ratepayers and the economy of the [s]tate survive such an increase? . . . Could this be accomplished in this 1993 recessionary economy? Could we fairly service the ratepayers and also fulfill our responsibility to the [c]ompany under [General Statutes] § 16-19e? These questions were uppermost in our minds as we proceeded with the hearings.” (Citation omitted.)

Addressing the issue of pension expenses, the 1993 decision stated that, “[g]iven the magnitude of the unrecognized net gain, the persistence of past earnings in excess of forecasted long-term rates of return, and the projection of negative . . . direct pension expenses over the next five years, the [department] deems it appropriate that [the company’s] $68.3 million net gain be amortized over eight years rather than 18.3 years.” The decision also included the following language which the company had requested in its written comments on the draft decision: “The [department] recognizes the fact that as a result of using an eight-year amortization of the unrecognized net gain, the [c]om-pany’s pension expense for book purposes . . . will be over $5 million higher [annually] than the amount to be included in rates. The [department] recognizes that, all other things being equal, ratemaking pension expenses in years [nine] through [eighteen] will be higher than the [c]ompany’s . . . booked pension expense.”

In 1998, the legislature enacted the 1998 Electric Restructuring Act, Public Acts 1998, No. 98-28 (act), which required the company to sell its electricity generation assets and to focus solely on transmission and distribution. The act allowed the company to recover from its customers the difference between the net book value of certain of its assets under the act and their previous market value, or its “stranded costs . . . .” See Connecticut Light & Power Co. v. Dept. of Public Utility Control, 266 Conn. 108, 110, 830 A.2d 1121 (2003). In 1999, the company attempted to recover as stranded costs a portion of the pension expenses that had been the subject of the 1993 decision. The plaintiff argued that allowing recovery of the expenses as stranded costs would result in retroactive rate making. Citing the “all other things being equal” language of the 1993 decision, the department concluded that “it would be inappropriate to eliminate the regulatory obligation to recover the expense difference of the 1993-1995 unrecognized pension gain.” It denied recovery of the expense, however, stating that the issue would be “addressed in the [c]ompany’s next rate case.”

The company filed its next request for a rate increase in 2003 (2003 rate filing). 3 In that filing, it again requested recovery of the disallowed pension expense. On December 17, 2003, the department issued an initial decision denying recovery of the expense (initial decision) . The company then filed a petition for reconsideration of that issue, among others, on the ground that the department had not properly considered its 1993 decision promising future recovery of the expense. The department granted the request for reconsideration and, on August 4, 2004, it issued its final decision (2004 decision) in which it reversed the initial decision. The department stated that it agreed “with the [plaintiff] that the [department did not in [the 1993 decision] establish a regulatory asset for the unrecognized pension gain. There is, it has been held, ‘a fundamental difference between a decision to record deferred charges and a decision to recover deferred charges .... A decision to permit recording of deferred charges has only an indirect impact on rates, and, therefore, is not subject to the same scrutiny as a decision to raise rates.’ [Business & Professional People for the Public Interest v. Illinois Commerce Commission, 146 Ill. 2d 175, 246, 585 N.E.2d 1032 (1991)]. The [1993 decision] permitted the recording of deferred charges, leaving all ratemaking considerations of the deferred charges (including regulatory asset treatment and rate recovery) to be decided in a future rate case, at which all parties could present evidence on such issues. In this proceeding, [the company] requested recovery of the deferred charges, and a full and complete record was created. Upon review of the record of the rate case and this reopener, the [department concludes that it is appropriate to allow [the company] to establish a regulatory asset to recover the unrecognized pension gain. As [the plaintiff] states, [the company’s] actual pension expense over 1993, 1994 and 1995 was below the allowed pension expense; however, the fact remains that, but for the [department's actions in reducing the unrecognized pension gain assumptions, the gap between allowed and actual would have been greater.” The company had asked the department to allow it to recover the $15.7 million as a regulatory asset over four years, but the department allowed recovery over thirteen years with no return.

The plaintiff appealed from the department’s decision to the trial court pursuant to General Statutes §§ lb-35 and 4-183. The plaintiff argued on appeal that the department’s 1993 decision had not established a regulatory asset and that such assets cannot be created “after the fact.” It also argued that the department had failed to consider whether denying recovery would harm the company and, therefore the decision constituted impermissible “single-issue” rate making. Finally, the plaintiff argued that allowing recovery constituted impermissible retroactive rate making.

In its memorandum of decision, the trial court concluded that it was “not necessary to pigeonhole or separate the analysis of a regulatory asset from that of deferred charges, as [the plaintiff] attempts. In 1993, the [department] concluded that [the company] must amortize its pension responsibilities over eight years rather than the standard 18.3 years. This created a book value that had to be considered in a later rate case.” Accordingly, the court concluded that the department properly had determined in the 2004 decision that the deferred pension expense could be treated as a regulatory asset.

The trial court also rejected the plaintiffs claim that the 2004 decision constituted impermissible single issue rate making. It noted that the prohibition against such rate making “ensures that when a utility seeks agency approval of its revenue requirements, the utility fully discloses aggregate costs, rather than certain specific costs related to a component of its operation. . . . The [department] must not consider revenue requirements in isolation. . . . Here the [department] was furnished the various costs that formed the basis of [the company’s] proposed rate schedules and the [department] conducted a review of these costs. The [department’s 2004] decision allowing a regulatory asset for the pension gain was a portion of a multitiered process and cannot be labeled as single issue rate making.” (Citation omitted.)

Finally, the trial court rejected the plaintiffs claim that the 2004 decision constituted impermissible retroactive rate making. It stated that “[t]he [department] had properly reserved the $15.7 million amount and ‘red flagged’ it [in the 1993 decision] for further action. . . . [T]he [department] was not illegally revising a rate previously set; rather the agency was giving finality to an issue previously left unresolved.” (Citation omitted.) Accordingly, the trial court dismissed the plaintiffs appeal.

On appeal to this court, the plaintiff claims that the trial court improperly concluded that the 1993 decision created a regulatory asset and, therefore, allowing recovery for the pension expense in the 2004 decision did not constitute impermissible retroactive rate making. The plaintiff further claims that, even if the 1993 decision did create a regulatory asset, the trial court improperly failed to consider whether the company had been harmed by the initial denial of the expense and, without such consideration, the decision constituted impermissible single issue rate making. We disagree with both claims.

We begin by setting forth the standard of review. “In reviewing [an] administrative rate decision, the court must . . . ensure that the agency’s decisionmaking process was conducted pursuant to the appropriate procedures and that the outcome of the process reflects reasoned decisionmaking — a reasonable application of relevant statutory provisions and standards to the substantial evidence on the administrative record. Section 4-183 (g) coupled with the presumption of validity that attends a . . . rate order . . . establishes a standard for judicial review that is appropriately deferential to agency decisionmaking, yet goes beyond a mere judicial rubber stamping of an agency’s decisions. . . .

“Within this context, judicial review of [the department’s] action is governed by the [Uniform Administrative Procedures Act]; General Statutes §§ 4-166 through 4-189; and the scope of that review, the substantial evidence rule, is restricted. . . . Substantial evidence exists if the administrative record demonstrates a substantial basis of fact from which the fact in issue can be reasonably inferred. . . . With regard to questions of fact, it is neither the function of the trial court nor of this court to retry the case or to substitute its judgment for that of the administrative agency. . . . Judicial review of conclusions of law reached administratively is also limited. The court’s ultimate duty is only to decide whether, in light of the evidence, the [agency] has acted unreasonably, arbitrarily, illegally, or in abuse of its discretion. ... In the specialized context of a rate case, the court may not substitute its own balance of the regulatory considerations for that of the agency, and must assure itself that the [department] has given reasoned consideration to the factors expressed in § 16-19e (a).” 4 **4 (Citations omitted; internal quotation marks omitted.) Connecticut Light & Power Co. v. Dept. of Public Utility Control, 219 Conn. 51, 57-58, 591 A.2d 1231 (1991).

“This broad grant of regulatory authority carries with it the necessarily equally broad discretion, to be exercised within legal limits; see General Statutes § 4-183 (g); to take into account such varying factors as economics, public policies, accounting principles, fairness to the parties, and the context and intent of any prior agreements entered into by the utility, and to balance those factors by a process of reasoned decision-making. . . . The [department] is not required, however, to place determinative weight on any single factor.” (Citation omitted.) Id., 64. “Because the plaintiff[’s] appeal to the trial court is based solely on the record, the scope of the trial court’s review of the [department’s] decision and the scope of our review of that decision are the same.” (Internal quotation marks omitted.) Quarry Knoll II Corp. v. Planning & Zoning Commission, 256 Conn. 674, 726 n.29, 780 A.2d 1 (2001).

We next review the law governing a utility commission’s treatment of regulatory assets for rate-making purposes. “A regulatory asset is a liability of a utility’s ratepayers. Utility companies may incur large expenses in various ways — storm damages, installation of new facilities, increased taxes and so forth. These expenses, if passed immediately on to ratepayers, could create havoc. An immediate recovery of such expenses could cause sudden upward increases in rates, commonly termed ‘rate shock.’ In order to avoid rate shock, [public utility] commissions often will permit utility companies to recover their expenses from ratepayers on a deferred basis, listing the ratepayers’ debt as a ‘regulatory asset.’ A regulatory asset is, therefore, a future debt of the ratepayers that can be passed on, together with interest, to the ratepayers.” Office of Consumer Counsel v. Dept. of Public Utility Control, 252 Conn. 115, 126-27, 742 A.2d 1257 (2000).

For general accounting purposes, regulatory assets are governed by financial accounting Statement No. 71, issued by the financial accounting standards board. See 3 Financial Accounting Standards Board, Accounting Standards — Original Pronouncements, Statement of Financial Accounting Standards No. 71, p. 4738 (State ment No. 71). 5 Statement No. 71 recognizes that, “[i]n many cases, a regulated company will be required to capitalize certain costs that have been incurred, while generally accepted accounting principles require that the costs be expensed.” B. Jamigan, Financial Accounting Standards — Explanation and Analysis (18th Ed. 1996) p. 1287. Statement No. 71 allows the regulated company to capitalize incurred costs when the following two conditions have been met: (1) “It is probable that revenues in the future will be at least equal to the costs that are capitalized”; and (2) “[t]he revenues are intended to recover incurred costs, rather than anticipated future costs.” Id.

Because this court has had little experience in determining the significance and application of these accounting standards in rate-making proceedings; but see footnote 13 of this opinion; we seek guidance from our sister jurisdictions. The Illinois Supreme Court considered that issue in Business & Professional People far the Public Interest v. Illinois Commerce Commission, supra, 146 Ill. 2d 175. That case involved extremely complex rate-making proceedings for three newly constructed nuclear electrical generating facilities. Id., 190. The Illinois regulatory commission (commission) had granted the utility’s petition to defer certain expenses related to costs of construction and depreciation for the period between the date that the facilities went into service and the effective date of the first rate order, a period of approximately three years. Id., 231-32. In the later rate proceeding, the commission allowed the utility to recover the full amount of the deferred expenses. Id., 237. On appeal, the plaintiff argued that the rate order violated the prohibitions on both retroactive rate making and single issue rate making. Id., 243-47.

The court recognized that the Illinois statutes did “not permit refunds if the established rates are too high, or surcharges if the rates are too low. . . . This rule is consistent with the prospective nature of the [c]om-mission’s legislative function in ratemaking. In addition, this rule promotes stability in the ratemaking process.” 6 ****6 (Citation omitted.) Id., 243. It rejected the plaintiffs argument that allowing full recovery constituted retroactive rate making, however, because the deferred expenses never had been included in a valid rate order. 7 Id., 244.

Addressing the single issue rate-making claim, the court stated that “[t]he rule against single-issue rate-making recognizes that the revenue formula is designed to determine the revenue requirement based on the aggregate costs and demand of the utility. Therefore, it would be improper to consider changes to components of the revenue requirement in isolation. Oftentimes a change in one item of the revenue formula is offset by a corresponding change in another component of the formula. For example, an increase in depreciation expense attributable to a new plant may be offset by a decrease in the cost of labor due to increased productivity, or by increased demand for electricity.” (Emphasis in original.) Id., 244; see also State ex rel. Utility Consumers Council of Missouri, Inc. v. Public Service Commission, 585 S.W.2d 41, 56 (Mo. 1979) (permitting utility to use fuel adjustment clause impermissibly permits one factor to be considered to exclusion of all others in determining whether rate should be increased); Philadelphia Electric Co. v. Pennsylvania Public Utility Commission, 93 Pa. Commw. 410, 422, 502 A.2d 722 (1985) (“ [t]he general rule is that there may be no line by line examination of the relative success or failure of the utility to have accurately projected its particular items of expense or revenue and an excess over the projection of an isolated item of revenue or expense may not be, without more, the subject of the [Commission’s order of refund or recovery, respectively, on the occasion of the utility’s subsequent rate increase requests”). The court found that, in the case before it, “much, if not all, of the increased expenses recorded as deferred charges were offset by decreases in other operating expenses combined with higher revenues derived from increased demand.” Business & Professional People for the Public Interest v. Illinois Commerce Commission, supra, 146 Ill. 2d 245. It concluded, therefore, that allowing recovery of the full amount of the deferred expenses without consideration of the offsetting amounts constituted single issue rate making. Id.

In response to the utility’s arguments to the contrary, the court stated that “ [t]here is a fundamental difference between a rate case and a case involving only accounting procedures. A decision to permit recording of deferred charges has only an indirect impact on rates and, therefore, is not subject to the same scrutiny as a decision to raise rates. It is improper to assume . . . that having met the criteria for recording deferred charges, [the utility] is automatically entitled to recovery of the full amount recorded. . . .

“The purpose of the accounting variance is to protect [the utility] from adverse financial impact caused by the regulatory delay period, and to afford [the utility] the opportunity to recover these charges.” Id., 246-47. “In order to protect [the utility] from adverse financial impact while at the same time protecting the ratepayers, it is necessary to determine the extent of the harm [that the utility] suffered as a result of the delay [between the date that the plants went into service and the effective date of the rate order]. It is possible . . . that the amount [that the utility] lost due to increased depreciation charges was offset by decreases in other expense items.” Id., 247. Thus, the court in Business & Professional People for the Public Interest concluded that full recovery of the previously recorded deferred expense in the rate-making proceeding was not automatic, but must be considered in light of intervening circumstances. 8

Applying these principles to the present case makes it clear that the department, in the 2004 decision, recognized that the 1993 decision was a “decision to permit recording of deferred charges” rather than a “decision to raise rates.” Id., 246. It is also clear that the department found that the recording of the deferred charges did not create a regulatory asset. The plaintiff seizes upon this finding in support of its claim that, whatever representations were made in the 1993 decision, they did not amount to a promise of future recovery. The plaintiff has provided no authority, however, for its implicit claim that that the department may grant recovery for a deferred expense only when it has consistently characterized the expense as a regulatory asset. The distinction made by the department between deferred expenses and regulatory assets is not that the latter are recoverable in rate proceedings, while the former are not; it is that the latter actually have been included in a rate order, while the former have not. See id. (“there is a fundamental difference between a decision to record deferred charges and a decision to recover deferred charges”); id. (“[tjhere is a fundamental difference between a rate case [involving recovery of a regulatory asset] and a case involving only accounting procedures [involving the creation of a regulatory asset]”). The very purpose of allowing a deferred expense is to provide an opportunity for recovery in a future rate proceeding. See id., 247. Although the distinction made by the department may be useful for purposes of linguistic clarity, for general accounting purposes the opportunity to recover an expense in a future proceeding is, by definition, a regulatory asset. Moreover, nothing in Business & Professional People for the Public Interest supports the plaintiffs claim that a utility may recover deferred expenses only when the regulatory commission explicitly promises future recovery of a specific dollar amount in a specific rate-making proceeding. 9 Indeed, our research has not revealed and, despite its claim that the department may create a regulatory asset only by using “explicit and unmistakable” promissory language, the plaintiff has not pointed to, any statutory, regulatory or common-law standards in Connecticut governing the creation of a regulatory asset for rate-making purposes. 10

In the absence of any more specific standards, we must conclude that the creation of a regulatory asset— or, in the department’s parlance, a deferred expense— is governed by Statement No. 71, while the recovery of a regulatory asset in a rate proceeding is a matter within the department’s broad discretion as limited by the considerations set forth in § 16-19e (a) (4). Cf. Connecticut Light & Power Co. v. Public Utilities Control Authority, 176 Conn. 191, 215, 405 A.2d 638 (1978) (“[t]he determination of the time when [an] amortization period begins to run, for rate-making purposes, is a matter that, in the absence of a statute or regulation or order specifically making that determination, has to be left to the sound discretion of the [department], to be exercised under the guidelines in ... § 16-19e”). Thus, a regulatory asset has been created when the department, at the time of the original deferral, gave reasonable assurances that recovery of the expense would be allowed at some future date; see B. Jamigan, supra, p. 1287 (company may record regulated asset when it is probable that revenues will be recovered equal to costs that are capitalized); and rate recovery ultimately may be allowed when the department has determined that the requirements of § 16-19e (a) (4) and (5) have been met. In addition, we agree with the court in Business & Professional People for the Public Interest, and the defendants do not dispute, that, as a general rate-making principle, retroactive rate making and single issue rate making are not permissible.

In the present case, the company filed written comments on the draft 1993 decision in which it expressly requested “language to be added to the final decision that would allow recognition of a regulatory asset for the difference in pension expense created by the eight year vs. [eighteen] year amortization periods for the net unrecognized gain.” The company also requested that the following language be included in the final decision in order to create the regulatory asset: “[A]ll other things being equal, ratemaking pension expenses in years [nine through eighteen] will be higher than the [c]ompany’s . . . booked pension expense.” The department explicitly recognized in the 1993 decision that the company’s pension expense for book purposes would be more than $5 million higher annually than the amount included in the rates and included in the decision, almost verbatim, the language intended to create a regulatory asset that the company had requested in its written comments. In light of this history, we conclude that the department’s determination in the 2004 decision that it had provided assurances in the 1993 decision that the company ultimately would be allowed to recover the disallowed pension expense through higher rates was supported by substantial evidence and was not contrary to the law. 11 We also con- elude, therefore, that consideration of the expense in the 2004 rate proceeding did not constitute retroactive rate making. See Business & Professional People for the Public Interest v. Illinois Commerce Commission, supra, 146 Ill. 2d 244 (when treatment of expense has not been conclusively decided in prior rate order, allowing recovery in later proceeding does not constitute retroactive rate making).

Having reached this conclusion, it remains for us to determine whether allowing rate recovery for the deferred pension expense was within the discretion of the department, as limited by § 16-19e (a) (4) and (5). In making that determination, we consider both the reasonableness of the decision allowing the initial deferral and the reasonableness of the ultimate decision allowing recovery. As the company points out, the plaintiff makes no claim that the company improperly calculated the unrecognized pension gain in the 1993 rate proceeding or that its proposal to amortize the gain over 18.3 years did not meet the governing accounting and actuarial standards. Accordingly, it is undisputed that it would have been within the department’s discretion to allow the company to recover all of the projected pension costs in the 1993 rate proceeding as prudently incurred operating costs. We conclude, therefore, that it was within the department’s discretion to assure the company in the 1993 decision that, in the absence of some countervailing consideration, recovery of the expense would be allowed at a later date.

The plaintiff argues, however, that the department’s failure to take a fresh look in the 2004 rate proceeding at the assurances made in the 1993 decision in light of intervening circumstances constituted impermissible single issue rate making. Specifically, the plaintiff points out that the department found in the 2004 decision that the company’s actual pension expenses for the years 1993 through 1995 were lower than the pension expenses allowed in the 1993 rate proceeding and argues that the department should have taken that fact into account in determining whether it should allow recovery for the full amount of the deferred expense. In support of this argument, the plaintiff contends that the court’s decision in Business & Professional People for the Public Interest stands for the proposition that, when a regulatory commission creates a regulatory asset by promising a utility that disallowed expenses may be recovered in a future proceeding, the commission’s failure to reconsider the scope of that promise in the future proceeding in light of intervening circumstances and to adjust recovery accordingly always constitutes impermissible single issue rate making. See Business & Professional People for the Public Interest v. Illinois Commerce Commission, supra, 146 Ill. 2d 247 (commission was required to determine whether denying full recovery of deferred expense to utility would harm utility in light of offsetting decreases in other expenses in interim between recording of expense and recovery of expense in rate proceeding).

We are not persuaded that the department violated the prohibition against single issue rate making in the present case. We note that the regulatory commission in Business & Professional People for the Public Interest allowed the recording of the deferred expenses not within the context of a rate-making proceeding, but in response to the utility’s petition to record the expenses in “a case involving only accounting procedures.” Id., 246. Thus, the commission did not have the opportunity to consider the propriety of allowing full recovery of the expense in light of the utility’s entire financial picture before allowing the utility to record the deferred expenses. In contrast, in the present case, the department made its initial determination that any deferred expenses could be recovered in full in a later proceeding within the context of the 1993 rate proceeding. Thus, the department had the opportunity to reduce or eliminate the expense for rate-making purposes in light of all of the evidence presented at that proceeding and chose not to do so.

Moreover, the department stated in the 2004 decision that it had reviewed “a full and complete record” in the 2004 rate proceeding and had seen no reason to reevaluate its earlier promise. As we have indicated, there is no dispute that the actual amount of the unrecognized pension gain was $68.3 million and the company’s proposed 18.3 year amortization of that amount was justified actuarily. Moreover, the deferral of the expense was initiated by the department for the primary purpose of avoiding “rate shock” for the company’s ratepayers, 12 not, as in Business & Professional People for the Public Interest, by the utility for the sole purpose of preserving an opportunity to recover expenses that otherwise would not be considered in any rate proceeding. The department gave no indication at that time that the amount of the deferred expense should be subject to adjustment on the basis of the company’s actual pension expenses. We conclude, therefore, that, once the objective of avoiding “rate shock” had been achieved, the department was not compelled to reconsider its original promise solely because actual pension expenses were lower than anticipated. See Connecticut Light & Power Co. v. Dept. of Public Utility Control, supra, 219 Conn. 64 (department has broad discretion under § 16-19e [a] [4] to consider wide variety of facts, including prior agreements entered into by utility, and is not required to place determinative weight on any single factor). 13

In summary, we conclude that, because the company’s deferred pension expense had not been denied conclusively in the 1993 decision, the decision to allow recovery in the 2004 decision did not constitute retroactive rate making. We further conclude that, because the department considered the expense both within the context of the 1993 rate proceeding, when it initially allowed recording of the deferred expense, and within the context of the 2004 rate proceeding, when it allowed full recovery, its action did not constitute single issue rate making and was within the department’s discretion. 14 Accordingly, we conclude that the trial court properly dismissed the plaintiffs appeal.

The judgment is affirmed.

In this opinion the other justices concurred.

1 The plaintiff appealed from the judgment of the trial court to the Appellate Court, and we transferred the appeal to this court pursuant to General Statutes § 51-199 (c) and Practice Book § 65-1.

2 The pension plan’s unrecognized net gain represented the amount of actual pension gains in excess of previously forecasted gains. The 18.3 years amortization period represented the average expected future working life of the company’s employees as of 1992, and was established in accordance with generally accepted financial accounting standards.

3 The company and the department entered into an agreement in 1996 whereby the department’s adjustment of the amortization period terminated at the end of the 1995 fiscal year. The company resumed the 18.3 year amortization of the unrecognized pension gains for rate-making purposes on July 1, 1996. The company did not seek to recover the deferred pension expense for the three years ending July 1,1996, until the eight year amortization period had expired because the company believed that, under the 1993 decision, it could recover the expense only during years nine through eighteen of the original 18.3 year amortization period. Although the company did not request recovery of the expense until 2003, it listed the difference between its pension expenses for accounting purposes and the pension expenses allowed in the 1993 proceeding as a regulatory asset in its 1998 rate filing.

4 General Statutes § 16-19e provides in relevant part: “(a) In the exercise of its powers under the provisions of this title, the Department of Public Utility Control shall examine and regulate the . . . establishment of the level and structure of rates in accordance with the following principles . . . (4) that the level and structure of rates be sufficient, but no more than sufficient, to allow public service companies to cover their operating and capital costs, to attract needed capital and to maintain their financial integrity, and yet provide appropriate protection to the relevant public interests, both existing and foreseeable which shall include, but not be limited to, reasonable costs of security of assets, facilities and equipment that are incurred solely for the purpose of responding to security needs associated with the terrorist attacks of September 11, 2001, and the continuing war on terrorism; (o) that the level and structure of rates charged customers shall reflect prudent and efficient management of the franchise operation . . . .”

5 The relevant portion of Statement, No. 71 provides: “Rate actions of a regulator can provide reasonable assurance of the existence of an asset. An enterprise shall capitalize all or part of an incurred cost that would otherwise be charged to expense if both of the following criteria are met: “a. It is probable that future revenue in an amount at least equal to the capitalized cost will result from inclusion of that cost in allowable costs for rate-making purposes.

“b. Based on available evidence, the future revenue will be provided to permit recovery of the previously incurred cost rather than to provide for expected levels of similar future costs. . . .” 3 Financial Accounting Standards Board, supra, Statement of Financial Accounting Standards No. 71, ¶ 9, p. 4740.

The standards further provide that the word “probable” is used “with its usual general meaning, rather than in a specific technical sense, and refers to that which can reasonably be expected or believed on the basis of available evidence or logic but is neither certain nor proved ...” Id., p. 4740 n.6.

6 See also, e.g., Detroit Edison Co. v. Public Service Commission, 416 Mich. 510, 523, 331 N.W.2d 159 (1982) (“the rule against retroactive ratemaking is that when the estimates prove inaccurate and costs are higher or lower than predicted, the previously set rates cannot be changed to correct for the error”); Boston Edison Co. v. Dept. of Public Utilities, 375 Mass. 1, 6, 375 N.E.2d 305 (“a rate increase may not be awarded retroactively as matter of law”), cert. denied, 439 U.S. 921, 99 S. Ct. 301, 58 L. Ed. 2d 314 (1978); In re Petition of Elizabethtown Water Co., 107 N.J. 440, 452-53, 527 A.2d 354 (1987) (utilities may not recover revenue deficiencies caused by insufficiency of prior rates because rate making is necessarily present and prospective); Providence Gas Co. v. Burke, 475 A.2d 193, 197 (R.I. 1984) (“[o]ne of the central principles of ratemaking is that rates must be prospective”); Railroad Commission v. Lone Star Gas Co., 656 S.W.2d 421, 425 (Tex. 1983) (recognizing “fundamental principle that utility rates are set for the future, and not the past”); MGTC, Inc. v. Public Service Commission, 735 P.2d 103, 107 (Wyo. 1987) (“[t]he rule against retroactive ratemaking is a generally accepted principle of public utility law which recognizes the prospective nature of utility ratemaking and prohibits regulatory commissions from rolling back rates which have already been approved and have become final”).

7 Cf. Philadelphia Electric Co. v. Public Utility Commission, 93 Pa. Commw. 410, 502 A.2d 722 (1985). In Philadlelphia Electric Co., the utility submitted a request for the creation of a deferred expense after the expense had been incurred. Id., 415. The utility granted the request for accounting purposes, but made it clear in the order granting approval that it ultimately might deny recovery of the expense and that the utility must include a statement to that effect in its financial statements. Id., 415-18. The commission ultimately denied recovery of the expense and the utility appealed. Id., 418-19. The Commonwealth Court of Pennsylvania concluded that, in the absence of any proof that the previous expenses were extraordinary and nonrecurring, the commission properly had determined that the utility was barred from seeking retroactive recovery of the expenses. Id., 422.

8 Statement No. 71 also recognizes that recovery of a deferred expense is not always guaranteed. It provides that “[r]ate actions of a regulator can reduce or eliminate the value of [a regulated] asset. If a regulator excludes all or part of a cost from allowable costs and it is not probable that the cost will be included as an allowable cost in a future period, the cost cannot be expected to result in future revenue through the rate-making process. Accordingly, the carrying amount of any related asset shall be reduced to the extent that the asset has been impaired. Whether the asset has been impaired shall be judged the same as for enterprises in general.” 3 Financial Accounting Standards Board, supra, Statement of Financial Accounting Standards No. 71, ¶ 10, p. 4741.

9 The plaintiff argues that “when a regulatory commission validly creates a regulatory asset for certain utility costs or obligations, two points have been established for that asset. First, its dollar amount has been determined and is not subject to further regulatory adjustment. Second, the possibility that the utility will be denied rate recovery for the asset has vanished.” If that were the case, however, then the court in Business & Professional People for the Public Interest — upon which the plaintiff relies — would not have ordered the commission to consider whether recovery of the deferred expenses should be reduced or eliminated. Moreover, the plaintiff’s argument is contradicted by the portion of Statement No. 71 that explicitly recognizes that regulatory assets may be reduced or eliminated. See footnote 8 of this opinion.

The plaintiff also relies on the department’s decision in the company’s Application to Increase Rates and Revenues — Millstone 3 Deferred Accounting Treatment, Docket No. 85-10-22 (April 1,1986), in support of its argument that a regulatory asset must be created by clear and unmistakable language. In that proceeding, the company sought deferred treatment of the costs of operating “Millstone 3,” a newly constructed nuclear generation plant, incurred between the date of commercial operation and the effective date of the rate orders. The department found that “[t]he [cjompany’s proposal to defer the operating cost of Millstone 3 and the associated carrying cost is reasonable and appropriate under the circumstances and subject to subsequent review prior to inclusion in rates. ” Accordingly, it concluded that “ [t]he [cjompany shall defer all costs associated with the operation of Millstone 3 from the date of commercial operation of the unit until the effective date of a final decision in the instant docket.” Even if it is assumed, however, that this language is somewhat clearer than the language of the 1993 decision — a point that we do not necessarily concede — nothing in the Millstone 3 decision suggests that the department must use extraordinarily clear and unmistakable language in creating a regulatory asset.

10 Cf. Ohio Rev. Code Ann. § 4928.01 (A) (26) (West 2006 Pocket Part) (“ ‘[r]egulatory assets’ means the unamortized net regulatory assets that are capitalized or deferred on the regulatory books of the electric utility, pursuant to an order or practice of the public utilities commission or pursuant to generally accepted accounting principles as a result of a prior commission rate-making decision, and that would otherwise have been charged to expense as incurred or would not have been capitalized or otherwise deferred for future regulatory consideration absent commission action”).

11 The plaintiff argues that the expense was flatly disallowed because of “the magnitude of the unrecognized net gain, the persistence of past earnings in excess of forecasted long-term rates of return, and the projection of negative . . . direct pension expenses over the next five years . ...” If the department had believed that these reasons justified permanently disal lowing recovery of the “mismatch” between the company’s pension expenses for book purposes and its pension expenses for rate-making purposes, however, it would not have included the “all things being equal” language requested by the company in the 1993 decision. Accordingly, we conclude that the reasons were provided in order to show why deferring the expense would be reasonable, and not to justify denying the expense.

12 The plaintiff argues that the pension expense was not deferred to avoid “rate shock,” but was flatly disallowed. We have rejected this argument. See footnote 11 of this opinion.

13 The company relies on our decision in Connecticut Light & Power Co. v. Public Utilities Control Authority, supra, 176 Conn. 191, in support of its argument that it is always arbitrary for the department to deny even a portion of a regulatory asset. In that case, the authority granted the plaintiff utilities’ request to amortize certain storm related and power outage expenses over three years. Id., 208. When it granted the request, it stated that “ ‘rate-making treatment of this matter can only be adjudicated in formal rate-making proceedings.’ ” Id., 214 n.32. By the time of the next rate-making proceeding, the expenses had been fully amortized on the utilities’ books. Id., 213. The authority disallowed recovery of the expenses on that ground. Id. The utilities appealed from the ruling to the trial court and the trial court sustained the appeal, except as to any portion of the claim that had been denied in previous proceedings. Id.

In the utilities’ further appeal to this court, we concluded that, by authorizing the utilities to amortize the expenses, the authority implicitly had promised the utilities that they could recover the expenses in a rate-making proceeding because “otherwise there would be no purpose for seeking, and granting, the authorization.” Id., 214. “The ultimate purpose of the amortization authorization is oriented toward the rate-making process, not the internal bookkeeping of the company.” Id. Accordingly, we concluded that the authority’s ruling, that the expense could not be allowed if the expense had been fully amortized on the utilities’ books, was inconsistent with the order authorizing amortization and was arbitrary. Id., 214-15. We therefore affirmed the trial court’s order remanding the case to the authority to determine when the amortization period for rate-making purposes should commence. Id., 215.

We do not agree with the company’s argument in the present case that Connecticut Light & Power Co. v. Public Utilities Control Authority, supra, 176 Conn. 214-15, stands for the proposition that, once the department has created a regulatory asset, “it is arbitrary for the [department] to deny even a portion of recovery” in a subsequent rate proceeding. Our ruling in that case was premised on our determination that the authority’s decision to allow amortization of the expense over three years itself had constituted a promise as to how the expense would be treated for rate-making purposes. In contrast, in the present case, the department’s decision to amortize the unrecognized pension gain over eight rather than 18.3 years created the uncompensated expense. The contingent disallowance of an expense by the department, pending further consideration in a future proceeding, does not, in and of itself, constitute an irrevocable promise that full recovery of the expense ultimately will be allowed. As we have indicated, however, we conclude that, in light of the particular circumstances surrounding the 1993 decision and the 2004 decision, it was not an abuse of discretion for the department to allow full recovery of the pension expense.

14 In light of this conclusion, we need not address the company’s alternate ground for affirmance that the plaintiffs claims are barred by principles of res judicata because they were decided in the 1999 stranded cost decision when the department ruled that the 1993 decision had created a “regulatory obligation.” See Zizka v. Water Pollution Control Authority, 195 Conn. 682, 687, 490 A.2d 509 (1985) (claim of res judicata does not implicate subject matter jurisdiction of court).