BOSTON GAS COMPANY [Note 1] vs. DEPARTMENT OF TELECOMMUNICATIONS AND ENERGY.

COWIN, J. This case arises out of an order by the Department of Telecommunications and Energy (department) regulating the manner in which Boston Gas Company, doing business as Keyspan Energy Delivery New England (company), switches from a "cost of service/rate of return" (COS/ROR) ratemaking model to a "performance based regulation" (PBR) model. The company appeals from an order of the department adjusting the company's proposed rates by imposing an "accumulated inefficiencies" factor. According to the department, this factor compensates consumers for some of the inefficiencies the company built up under the old COS/ROR model. The company argues that the department's decision to add an accumulated inefficiencies factor and the quantification of that factor are not supported by substantial evidence. In addition, the company argues that the department's order constitutes prohibited retroactive ratemaking. While we conclude that the department had a sound basis for imposing some accumulated inefficiencies factor, the department's quantification of the accumulated inefficiencies factor is unsupported by substantial evidence. We therefore vacate the department's order. [Note 2]

1. Regulatory background. General Laws c. 164, § 94, grants the department the authority to regulate rates for gas and electric companies. Each utility separately establishes its rates in individual adjudicatory proceedings before the department. Prior to 1995, the department set rates by using a COS/ROR ratemaking model. COS/ROR "permits utilities to charge rates that allow them to recover reasonable operating expenses and to earn a fair return on investment." D.P.U. 94-158 at 3 (1995). Under this system, the department first determines a utility's reasonable expenses. Next, the department calculates a figure representing a reasonable return on the utility's investment. The sum of these two figures is the utility's "total revenue requirement." Retail rates are set so that the utility can generate revenue equal to the total revenue requirement.

Page 235

A drawback of COS/ROR regulation is that it does not provide sufficient incentives for utilities to reduce their costs; a utility can simply pass extra costs to the customers in the form of higher rates. In addition, the rate established in an adjudicatory proceeding before the department remains the same until the utility initiates a rate change proceeding. Therefore, under COS/ROR, there is a delay between the time that the utility experiences increased costs and the corresponding proceeding establishing increased rates. This delay, called a "regulatory lag," actually benefits consumers because they continue to pay at the lower rate until the conclusion of the rate setting proceeding. Despite the advantage to consumers of the "regulatory lag," the department determined that, on the whole, the COS/ROR system resulted in higher rates for consumers.

In 1995, the department issued an order expressing its preference for PBR ratemaking. The Legislature subsequently endorsed this decision by enacting G. L. c. 164, § 1E, inserted by St. 1997, c. 164, § 193, which authorizes the department "to promulgate rules and regulations to establish and require performance based rates for each . . . gas company . . . in the commonwealth." The form of PBR at issue in this case is called "price cap" regulation. Under a price cap model, the department first determines an initial "cast off rate." [Note 3] This rate is then increased annually by accounting for inflation over time. However, the cast off rate is also reduced annually to account for the company's expected improvements in productivity. This adjustment to the cast off rate is called the "productivity offset."

The theory behind the price cap model is that it encourages the utility to improve its efficiency. See MCI Telecommunications Corp. v. Department of Telecommunications & Energy, 435 Mass. 144, 147 (2001). If the utility improves its productivity by more than the amount anticipated by the productivity offset, it keeps the extra profits. However, if the utility does not achieve the expected productivity gains, it will face a revenue shortfall. It is because of this incentive-producing benefit that the department endorsed PBR.

Page 236

2. Facts and proceedings below. The company filed proposed changes to its rates with the department on May 17, 1996. In response to the department's shift to PBR, the company proposed a five-year plan based on a price cap model. The company proposed a "cast off rate" pursuant to the old COS/ROR model, and then adjusted the cast off rate by a price cap formula. The company's price cap formula included a productivity offset based on, inter alia, a "consumer dividend," which is a downward adjustment in rates intended to account for efficiency gains a utility is expected to achieve as a result of the shift to PBR.

The department issued its decision (initial order) after conducting an investigation of the company's proposal. The department made downward adjustments to the company's cast off rate and increased the productivity offset by doubling the company's proposed consumer dividend of 0.5 per cent and adding a 1.0 per cent "accumulated inefficiencies factor." The stated purpose of the "accumulated inefficiencies factor" was to compensate consumers for inefficiencies "in the Company's current operations that the Department was unable to discover in its earnings review and would be unable to discover in a traditional rate case." In response to the company's motion for reconsideration, the department lowered the consumer dividend to 0.5 per cent, but retained the 1.0 per cent accumulated inefficiencies factor.

The company appealed from the department's order to the Supreme Judicial Court pursuant to G. L. c. 25, § 5, and a single justice reserved and reported the matter to the full court. At this point in the proceedings, the department agreed that it had not made adequate subsidiary findings and moved to remand the matter for further proceedings. This court discharged the report and ordered the single justice to remand the matter to the department. In its order following remand (remand order), the department retained the accumulated inefficiencies factor, but reduced it from 1.0 per cent to 0.5 per cent. The company appealed once again to a single justice pursuant to G. L. c. 25, § 5, who reserved and reported the matter to the full court.

3. Standard of review. Although the department has "broad authority to determine ratemaking matters in the public interest,"

Page 237

we must overturn the department's decision if "it is based on an error of law, unsupported by substantial evidence, unwarranted by facts found on the record as submitted, arbitrary and capricious, an abuse of discretion, or otherwise not in accordance with law." Massachusetts Inst. of Tech. v. Department of Pub. Utils., 425 Mass. 856, 867-868 (1997).

The company claims that the department's decision to impose an accumulated inefficiencies factor is not supported by substantial evidence. Substantial evidence is "such evidence as a reasonable mind might accept as adequate to support a conclusion." Cobble v. Commissioner of the Dep't of Social Servs., 430 Mass. 385, 390 (1999), quoting G. L. c. 30A, § 1 (6). In determining whether an agency decision is supported by substantial evidence, we must consider the record as a whole and reverse the agency's decision if "the cumulative weight of the evidence tends substantially toward opposite inferences." Id. at 391. Although we must carefully consider any evidence in the record that detracts from the agency's conclusion, we "accord due weight to the 'experience, technical competence, and specialized knowledge' of the department." Martorano v. Department of Pub. Utils., 401 Mass. 257, 261 (1987), quoting G. L. c. 30A, § 14 (7). The appealing party carries the burden of showing that the order appealed from is unsupported by substantial evidence. Massachusetts Inst. of Tech. v. Department of Pub. Utils., supra at 867.

4. The imposition of the accumulated inefficiencies factor. The initial order justifies the department's decision by relying on NYNEX Price Cap, D.P.U. 94-50 (1995) (NYNEX), an earlier decision that imposed an accumulated inefficiencies factor in a rate setting proceeding for NYNEX, a utility in the telecommunications industry. There, the department stated that because the average utility will be more efficient under PBR regulation, the department must find that the company has built up inefficiencies in its years under COS/ROR. Without an accumulated inefficiencies factor, according to the department, consumers "would receive none of the benefits associated with eliminating these inefficiencies."

The department reasserts this theory in its remand order, and finds support for it from the testimony of two company

Page 238

witnesses. These two witnesses argued that the underlying theory of PBR creates greater incentives for utilities to reduce costs. [Note 4] The company asserts that these statements stand only for the proposition that PBR creates greater risks for the company, as well as increased financial incentive to compensate for that risk. It is precisely that risk and potential reward that drive a company to achieve greater efficiency. Thus, the statements of company witnesses support the underlying theory of PBR, and the company does not challenge the department's authority to choose PBR over COS/ROR.

The department's conclusion that inefficiencies are embedded in the company's cast off rate as a result of COS/ROR regulation is supported by substantial evidence. If one accepts the proposition that PBR is superior to COS/ROR because it encourages greater efficiency, the logical conclusion is that "there must be inefficiencies embedded in the cost structure of [COS/ROR] regulated firms." An agency's findings are supported by substantial evidence if they "could have been made by reference to the logic of experience." New Boston Garden Corp. v. Assessors of Boston, 383 Mass. 456, 466 (1981), quoting G. L. c. 30A, § 1. The conclusion that accumulated inefficiencies exist is one that logically follows from the adoption of a PBR model. See Arthurs v. Board of Registration in Med., 383 Mass. 299, 310 (1981) ("an agency's conclusions based on inferences will not be set aside by a reviewing court unless they are unreasonable"). If the underlying rationale of adopting PBR is that it drives a company to greater efficiency, the department can conclude, as a matter of common sense, that the company

Page 239

has built up inefficiencies under COS/ROR, and that the consumer should share in the gains as these inefficiencies are discovered and eliminated. Although the department may not "speculate as to the existence of some future fact without any proper basis or explanation as to how the exercise of judgment leads to a conclusion," the department's inference is based on a theory adopted after careful consideration of all the evidence. Blue Cross & Blue Shield of Mass., Inc. v. Commissioner of Ins., 420 Mass. 707, 714 (1995). Thus, we hold that the imposition of the accumulated inefficiencies factor was within the department's discretion and is supported by substantial evidence.

In addition to lack of substantial evidence, the company asserts two additional arguments for setting aside the accumulated inefficiencies factor. First, the company contends that the department duplicated the reduction in future inefficiencies by imposing both a consumer dividend and an accumulated inefficiencies factor. According to the company, both of these figures account for the same future reduction in inefficiencies. The department, however, has articulated a difference: the consumer dividend is intended to "capture" the increase in efficiency that gas companies are expected to achieve under PBR, while the accumulated inefficiencies factor is intended to account for the additional short-term increase in efficiency the company will experience in the early years of the shift from COS/ROR to PBR. The department's theory is that, in the early years of the PBR plan, the company will discover and eliminate more inefficiencies than it will after the PBR system has been in place for a period of time. Therefore, an accumulated inefficiencies factor should be applied only in the first term of a PBR plan. Although both numbers anticipate future reductions in inefficiencies, it is within the department's discretion to separate these figures into two separate concepts. Attorney Gen. v. Department of Pub. Utils., 392 Mass. 262, 268 (1984) ("the choice of methodology is a matter committed to agency discretion and is beyond the scope of our review").

The company further argues that the imposition of the accumulated inefficiencies factor is inconsistent with the department's finding that the company's cast off rate is just and reasonable. The rates of a utility can only reflect costs found by

Page 240

the department to be reasonably and prudently incurred. See Boston Gas Co. v. Department of Pub. Utils., 387 Mass. 531, 539 (1982); Fitchburg Gas and Elec. Light Co. v. Department of Pub. Utils., 375 Mass. 571, 575 (1978). In its review of the company's proposed rates, the department rejected a portion of the company's expenses as failing to meet this standard. Because the department already deducted significant costs from the company's proposed cast off rates, the company contends that the department already accounted for inefficiencies in its rates.

We conclude that the department may both deduct those expenses not reasonably and prudently incurred from the company's proposed cast off rate and impose an accumulated inefficiencies factor. These two adjustments to the company's proposed rates are not inconsistent. Having already discovered some inefficiencies in the company's proposed cast off rate, the department can reasonably conclude that more inefficiencies will be exposed in the early years of the shift from COS/ROR to PBR. The department can logically conclude that the COS/ROR model has resulted in the buildup of inefficiencies that the department cannot discover in its traditional rate base and earnings review.

5. The quantification of the accumulated inefficiencies factor. The company argues that the department's choice of 0.5 per cent as the proper quantification of this factor is not supported by substantial evidence. We agree. Neither the department's initial order nor its remand order cites substantial record evidence to support the decision to set the accumulated inefficiencies factor at 0.5 per cent.

The department's initial order set the accumulated inefficiencies factor at 1.0 per cent by concluding that the figure adopted for the telecommunications industry in NYNEX Price Cap, D.P.U. 94-50 (1995), is an "appropriate proxy" for accumulated inefficiencies in the gas distribution industry. The department offered no justification for this conclusion, and even noted that "the potential for efficiency improvements is greater in the telecommunications industry . . . due to rapid technological advances occurring in [that] industry."

The department's remand order cuts the accumulated inefficiencies

Page 241

factor in half, but does not remedy the deficiency of its initial order. It simply justifies the 0.5 per cent figure by stating that the accumulated inefficiencies factor should be equal to the consumer dividend, because, as in NYNEX, supra, "it is appropriate to estimate that the industry's productivity increases will be twice as great as the expected" consumer dividend. The department's remand order does not point to record evidence to support 0.5 per cent as an accurate assessment of the accumulated inefficiencies. Nor does the order adequately explain why the consumer dividend provides an accurate "estimate" of the accumulated inefficiencies factor. Even assuming that tying the accumulated inefficiencies factor to the consumer dividend is logical because both measure future gains in efficiency, the department adopted a consumer dividend of 0.5 per cent only because the company "conceded" that this amount was appropriate. [Note 5] The department offers no evidence or explanation as to why it should double the company's proposed productivity offset.

In fact, the department's initial order concedes that it lacked information on which to base its decision, stating "there is little information regarding the efficiency improvements that should result as regulated companies move from [COS] to [PBR]." "While we recognize that some uncertainties cannot be precisely quantified, we do require more than a conclusory statement to that effect." Massachusetts Inst. of Tech. v. Department of Pub. Utils., 425 Mass. 856, 870 (1997).

The department asserts that, despite its uncertainty, tying the accumulated inefficiencies factor to the consumer dividend is a "methodology" that should be left to the department's discretion unless it results in a confiscatory rate. [Note 6] See Attorney Gen. v. Department of Pub. Utils., 392 Mass. 262, 269 (1984). Although

Page 242

our previous cases state that the choice of methodology in rate setting is a matter of agency discretion, these cases involved choices between alternative methods where there was adequate support for each method. See id. at 268; Massachusetts Elec. Co. v. Department of Pub. Utils., 376 Mass. 294, 302 (1978) ("When alternative methods are available, the department is free to select or reject a particular method . . ."). The need for explanation is especially important where the case concerns a matter of first impression, as it does here. Massachusetts Inst. of Tech. v. Department of Pub. Utils., supra at 868. Because the department failed to offer any reason for the determination that the accumulated inefficiencies factor should be equal to the company's proposed consumer dividend, we find that the department's quantification of the accumulated inefficiencies factor is not supported by substantial evidence.

6. Disposition. Under G. L. c. 30, § 14 (7) (e), we "may remand, set aside, or modify a Department order which is not based on substantial evidence." Fitchburg Gas & Elec. Light Co. v. Department of Pub. Utils., 371 Mass. 881, 887 n.10 (1977). Although we often remand cases for further proceedings where substantial evidence is lacking, see Massachusetts Inst. of Tech. v. Department of Pub. Utils., supra at 878; Morris v. Board of Registration in Med., 405 Mass. 103, 113-114, cert. denied, 493 U.S. 977 (1989), we have vacated the department's decision on occasion. See Cobble v. Commissioner of the Dep't of Social Servs., 430 Mass. 385, 396 (1999). The department has been repeatedly presented with the argument that its decision on the quantification of the accumulated inefficiencies factor is unsupported by substantial evidence and, in response, has issued three decisions (the initial order, the order following the company's motion for reconsideration, and the remand order), each lacking sufficient record support. In addition, the department has stated that it lacks information accurately to determine the correct amount of accumulated inefficiencies, and has offered no indication that its decision can ever be supported by substantial evidence. Therefore, a remand to the department would be futile.

7. Conclusion. This case is remanded to the county court where a judgment shall enter vacating those portions of the

Page 243

department's order that imposed an accumulated inefficiencies factor at 0.5 per cent.

So ordered.