NICHOLS, Justice.
The Plaintiff, New England Telephone and Telegraph Company (NET), seeks review of a decision and order of the Public Utilities Commission (Commission), the Defendant, which denied NET’s proposed rate increase and instead authorized NET to file a revised schedule of rates designed to increase revenues by a lesser amount. NET challenges the Commission’s rulings on (1) the adjustment of test year expenses; (2) the amortization of deferred tax reserve; (3) the imputation of interest on certain plant; and (4) the capitalization of research and engineering expenses.
We affirm the Commission’s decision and order.
Pursuant to 35 M.R.S.A. § 64 (Supp. 1982-83), NET on July 27, 1982, filed with the Commission revised tariffs to become effective August 26, 1982, seeking a $49.8 million increase in its annual gross revenues. This would augment revenues by approximately 30% over the previous year and would mean that the average telephone-related costs to a Maine family of four would rise, according to the Commission’s estimate, by $200 per year. The effective date of the tariffs was suspended twice by orders of the Commission dated August 4, 1982 and November 2, 1982.
Beginning on October 18, 1982, approximately thirty days of hearings were conducted. Over 160 exhibits were introduced, and more than 4,200 pages of testimony were recorded.
The Commission afterward concluded in its decision and order dated April 26, 1983, that NET was entitled to a rate increase of $11.4 million, which represented a 7% increase over 1981 rates. This computation was premised on a finding that a fair rate of return on NET’s investment was 11.31%. In order to establish what rate structure would yield the requisite return on investment, it was necessary to compute three items: NET’s expected gross utility revenues, its expected operating expenses, and its property providing service for which rates are charged and comprising the “rate base” on which a return should be earned.
In the course of calculating these variables, the Commission made,
inter alia,
the following determinations: First, in projecting NET’s wage costs (a major component of total expenses), the Commission took into account employee reductions occurring beyond the test year. Second, the Commission decided to amortize excess accruals in NET’s deferred tax reserve, thereby increasing NET’s operating income. In addition, the Commission reduced the rate base to reflect the smaller reserve. Third, in calculating NET’s federal income tax expense, the Commission imputed an interest expense on the amount of investment financed by the Job Development Investment Tax Credit (often referred to as JDITC), thus reducing its income for tax purposes. Finally, rather than treating research and systems engineering costs as a current expense, the Commission capitalized and amortized them over a ten-year period.
On May 25, 1983, NET filed a complaint seeking judicial review of that decision and order. The complaint alleges that the rates and charges authorized by the decision and order are too low for NET to obtain what the Commission had determined to be the
fair rate of return and are therefore confiscatory. NET contends that the Commission acted arbitrarily and abused its discretion in its treatment of (I) work force reductions, (II) excess of deferred taxes, (III) JDITC-fi-nanced plant, and (IV) research and engineering costs, and consequently understated its revenue requirement by about $6.5 million in all.
Furthermore, NET contends that the Commission’s handling of the excess deferred taxes could jeopardize NET’s qualification for taking the federal income tax benefit of accelerated depreciation.
We shall consider each of the challenged actions
seriatim.
I.
A basic principle of ratemaking is that a utility’s revenues, expenses and plant in the near future may be estimated by matching these items within a recent twelve-month operating period, called the “test year.” The parties to this proceeding agreed that the test year would be the calendar year 1981. It is well established that adjustments in test year results should be made to reflect subsequent known changes that would be certain to alter significantly the assumed balance among revenues, expenses and plant.
Camden and Rockland Water Co. v. Public Utilities Commission,
432 A.2d 1284, 1287 (Me.1981);
Central Maine Power Co. v. Public Utilities Commission,
153 Me. 228, 234-38, 136 A.2d 726, 731-33 (1957).
NET made three adjustments relating to wage and benefit costs: annualization of a wage and benefit increase that went into effect for the latter part of 1981, which reduced net income by $1.6 million; annual-ization of 1982 wage increases, which reduced net income by $2 million; and an adjustment for the transfer of employees to American Bell, Inc., occurring on January 1, 1983, which increased income by $619,000. These were accepted by the staff and the Commission.
The staff proposed, and the Commission accepted, a fourth adjustment, however, to reflect employee reductions occurring in 1982. This adjustment was calculated as decreasing payroll costs significantly and thereby resulting in an increase in net income of $1.7 million. NET objects to this calculation as an arbitrary departure from test year “matching” methodology, but it fails to distinguish in any persuasive manner this adjustment from the other three adjustments. NET cannot consistently maintain that adjustments for wage hikes are reasonable, in that they provide for a known change from which it may be predicted with certainty that wage costs will rise, and at the same time argue that employee reductions are
not
a known change from which it can be predicted with certainty that wage costs will fall.
NET asserts that it conclusively refuted the assumption that employee reductions would tend to diminish wage costs by introducing evidence of its expenses for July, 1982, and showing that these expenses, when annualized, exceeded the staff’s projected calculations for 1982. It also presented evidence of its expenses for a twelve-month period ending September, 1982. Such evidence is far from conclusive. It would be impossible to prove the precise extent of NET’s expenses for 1982 before the year had ended. The months examined may have been atypical of the calendar year as a whole.
Even if it should be assumed that the financial data for the months examined were representative of the entire year, it does not follow that the data prove the fallacy of the Commission’s common-sense assumption that wage costs would decline as the number of wage earners declined. Under the test year methodology, the Commission must be guided by certain facts occurring outside the test year in preparing
its estimates, see
Central Maine Power Co. v. Public Utilities Commission, supra,
153 Me. at 236-42, 136 A.2d 726; but its estimates need not be consistent with all facts outside the test year in order to be reasonable. By necessity, the Commission deals with estimates and hypothetical constructs. Such estimates must be based on reasonable formulations. The Commission has broad discretion in selecting among various rate-making methodologies, provided that they are reasonably accurate.
See New England Telephone & Telegraph Co. v. Public Utilities Commission,
390 A.2d 8, 49 (Me.1978).
The Commission is not required to manipulate its methodologies to eliminate every shred of suggested inaccuracy.
The Commission found, quite reasonably, that NET was really trying to obtain an attrition allowance through the backdoor without having presented a proper attrition analysis.
If NET’s figures were indeed representative of the 1982 year, the most that can be said is that for unexplained reasons NET’s 1982 expenses were greater than the pattern for the preceding year, taken together with relevant known variables, had foretold. NET would overlook the circumstance that there was a known change from the test year and would not make the downward adjustment in wage expenses that recognition of the change would logically compel — on the theory that this change is offset by an unproven and inconclusive increase in overall expenses caused by unpredicted variables. Such an approach would be inconsistent with the test year methodology which we have long favored.
See Central Maine Power Co. v. Public Utilities Commission,
153 Me. 228, 136 A.2d 726 (1957).
Typically, this Court applies a limited and deferential standard of review to the Commission’s fact-finding and choice of rate-making techniques.
Central Maine Power Co.
v.
Public Utilities Commission,
405 A.2d 153, 182 (Me.1979). We have no reason to do less in the appeal now before us. The Commission made a reasonable and technically proper decision, which was supported by substantial evidence.
II.
Section 167 of the Internal Revenue Code provides two basic methods of calculating the annual depreciation deduction allowed with respect to a taxpayer’s business property. Under the straight-line method, the taxpayer deducts an equal amount during each year of the asset’s useful life. Under the accelerated method, he may deduct greater amounts during the early years of the asset’s useful life. The taxes saved during these early years constitute “deferred taxes,” which will eventually be collected in the form of reduced annual deductions, but the taxpayer derives a benefit by receiving what amounts to a governmental interest-free loan. The option to take accelerated depreciation derives from the national policy of encouraging business investment.
In 1969, Congress became aware that the federal government was losing unforeseen additional revenue through accelerated depreciation in cases where a public utility used this method and regulatory agencies required a “flow-through”
of the tax sav
ings to ratepayers. Because flow-through reduces the rate charged to customers, the utilities derived less income and hence paid lower taxes. Congress therefore enacted § 167(7) of the Internal Revenue Code, the effect of which is to deny the availability of accelerated depreciation to utilities that employ an accounting technique which would result in flow-through.
In 1977-78, the Commission sought to flow through NET’s deferred taxes. NET was taking accelerated depreciation and using the accounting technique (normalization)
authorized under § 167(7). In an earlier case we held the Commission’s action to be arbitrary and an abuse of discretion because it put NET in jeopardy of losing its ability to take accelerated depreciation.
New England Telephone & Telegraph Co.
v.
Public Utilities Commission,
390 A.2d 8, 24 (Me.1978).
In January, 1979, the corporate income tax rate declined from 48% to 46%. As a result, a portion of the taxes in NET’s deferred tax reserve was deferred at a 48% rate but would be paid at a 46% rate. This meant that deferred taxes which had been accumulated at a rate higher than 46% exceeded the amount necessary to meet future tax liabilities associated with accelerated depreciation.
In 1981, the Commission rejected a proposal to flow through this excess in deferred taxes, citing the 1978
New England Telephone
case.
Re New England Telephone & Telegraph Co.,
42 PUR 4th 182, 200-01 (Me.P.U.C.1981). In the proceedings below, however, the Commission decided to flow through the excess by amortizing it over a two-year period. The Commission argues that, contrary to NET’s assertion on appeal, this adjustment is distinguishable from the flow-through adjustment of which we disapproved in the 1978
New England Telephone
case.
At the outset, it should be noted that there is indeed a critical difference between adjustments for ordinary deferred taxes and those for excess deferred taxes. The rate-making and accounting technique of normalization generally will yield rates that are in the long run proportionate to the utility’s tax expenses, because ratepayers at first contribute to the utility’s deferred tax reserve and later, when the utility’s tax expenses rise due to accelerated depreciation, are relieved from having to contribute a commensurately greater amount, owing to the presence of the tax reserve. Normalization does not compensate for
excess
deferred taxes caused by a decrease in tax rates. In this situation, ratepayers have contributed to the payment of future tax obligations, a portion of which will never become due. In order to avoid a windfall to the utility at the ratepayers’ expense, some sort of flow-through device must be used.
NET argues, however, that under section 167(7), to do so would jeopardize its ability to take accelerated depreciation.
Section 167(7 )(3)(G)(ii) provides that if a utility uses a method of depreciation other than the method used for rate-making purposes, it “must make adjustments to a reserve to reflect the deferral of taxes resulting from the use of such different methods of depreciation.” There is no reason, under this section, why a reserve should have to carry amounts in
excess
of deferred taxes. NET relies, however, upon Treas.Reg. § 1.167(1) — 1(h)(2), which NET interprets as specifying the exclusive circumstances in which the amount in the deferred tax reserve can be reduced. The regulation reads in pertinent part:
With respect to any account, the aggregate amount allocable to deferred tax under section 167(1) shall not be reduced except to reflect the amount for any taxable year by which the federal income taxes are greater by reason of the prior use of different methods of depreciation under subparagraph (l)(i) of this paragraph. An additional exception is that the aggregate amount allocable to deferred tax under section 167(1) may be properly adjusted to reflect asset retirements or the expiration of the period for depreciation used in determining the allowance for depreciation under section 167(a).
Treas.Reg. § 1.167(l)-l(h)(2)(i). Amortization of the excess deferred taxes would constitute a reduction of the reserve for a purpose not authorized by the foregoing Treasury Regulation, which, taken literally, would mean that normalization was not properly carried out, and, hence, accelerated depreciation is unavailable.
The Commission counters by directing attention to Treas.Reg. § 1.167(l)-(l)(h)(l)(b), which makes clear that the purpose of the reserve is “to reflect the total amount of deferral of federal income tax
liability
resulting from the use with respect to all of its public utility property of such different methods of depreciation.” (Emphasis added). The Commission reads section 1.167(l)-(l)(h) as a whole to prohibit generally reductions in the deferred tax reserve
only insofar as the reserve reflects tax liability.
The Code provision and accompanying regulations are ambiguous. Uncontradicted testimony below revealed that the Internal Revenue Service had indicated it would issue regulations to clarify the matter. The witness who so testified urged that it would be wise for the Commission not to act before those regulations are promulgated.
Like that witness, we are troubled by the Commission’s willingness to rush to action before the Internal Revenue Service has ruled on this matter. For the sake of expeditiously returning to the ratepayers a comparatively small sum ($911,000), the Commission saw fit to risk NET’s eligibility to take accelerated depreciation. Rather than speculate as to how the Internal Revenue Service may rule, the more prudent course may well have been to await a ruling.
However, we do not review the Commission’s decision today to determine whether we agree with the agency’s decision to act before it receives the Internal Revenue Service ruling; instead we review that decision for legal error, arbitrariness or abuse of discretion. We repeat that the regulation of public utilities is a function in the first instance, not of the judiciary, but of the Legislature, which has delegated its entire authority over the matter to the Commission.
Mechanic Falls Water Co. v. Public Utilities Commission,
381 A.2d 1080, 1090 (Me.1977). There are three possible ways that the Commission’s treatment of the excess deferred taxes could be construed as reversible error: (1) if where federal tax benefits are at stake, the construction by the Commission of ambiguous tax
provisions, on which the Internal Revenue Service has not yet ruled, is
per se
an abuse of discretion; (2) if the Commission erred in its legal conclusion that NET would not lose any tax benefits as a result of the amortization of excess deferred taxes; or (3) if the Commission erred in its factual finding that NET was not in reasonable jeopardy of losing its tax benefits. After careful consideration, we reject all three possible grounds for reversal.
In addition to its powers expressly conferred by statute, the Commission has implied powers to the extent necessary to fulfill its obligations effectively.
New England Telephone & Telegraph Co. v. Pnblic Utilities Commission,
362 A.2d 741, 754 (Me.1976). One of these implied powers is the power to interpret laws that may affect, or be triggered by, a rate-making decision. Although it may at times be more prudent for the Commission to await a definitive interpretation by an agency with greater authority in the area, it is obvious that the Commission would be paralyzed if it were forced in every instance to defer action whenever an ambiguous question of law was involved. Particularly in the area of tax law, which so often impinges on rate-making decisions, both the Commission and the public have an interest in the expeditious resolution of ambiguities that affect the setting of utility rates.
About a year ago, in
New England Telephone & Telegraph Co. v. Public Utilities Commission,
448 A.2d 272, 304-09 (Me.1982), we were confronted with a Commission adjustment which arguably jeopardized NET’s eligibility for a tax benefit. NET contended then that the Commission’s treatment of JDITC-financed plant violated I.R.C. § 46(f) and would render NET unable to take the investment tax credit. Under the Commission’s analysis, however, the treatment was consistent with the language and legislative history of section 46 and with the applicable treasury regulations. Rather than holding that the Commission should have left this determination to the Internal Revenue Service, we reviewed the pertinent tax provisions ourselves and resolved the apparent ambiguity in favor of the Commission. We follow the same approach today.
The Commission’s interpretation of the pertinent revenue provisions is, in our view, correct. Reading the provisions as a whole persuades us that neither Congress nor the Internal Revenue Service intended to predicate the availability of accelerated depreciation on the perpetual maintenance in the deferred tax reserve of monies that had been rendered supernumerary to the satisfaction of future tax liability. Not only do the provisions omit any reference to excess funds but Treas.Reg. § 1.167(l)-(l)(h)(l)(b) positively links the reserve to tax liability. Moreover, to disregard these considerations and to read Treas.Reg. § 1.167(l)-(l)(h)(2) very narrowly, as NET urges us to do, to prohibit all reductions in the deferred tax reserve except as specified, would mean that even accruals in the reserve due to outright error could never be corrected.
When it is recalled that the
raison d’etre
of these provisions was to protect the government from double loss of revenue,
amortization of excess deferred taxes will not seem inconsistent with the legislative purpose. To the extent that amortization results in a downward revision of the utili
ty’s revenue requirement and, hence, its taxable income, the government does incur a slight decrease in revenue. That decrease, however, is only in the amount that it had previously realized an unanticipated gain, because the government had been able to tax earlier yearly revenues which were achieved by setting utility rates so as to cover certain costs of future tax liability, but these costs (and, hence, the rates charged) were over-estimated. Thus, the net effect of amortization on the government’s revenue is essentially the same as if the utilities had never accrued an excess in deferred taxes in the first place.
We conclude, therefore, that amortization of a utility’s excess deferred taxes is not inconsistent with the preconditions for taking accelerated depreciation under section 167(7) of the Internal Revenue Code and the applicable treasury regulations. We are not aware of any court that has heretofore ruled upon this issue, but we note that the decisions of numerous public utilities commissions are in accord with our construction.
In
New England Telephone & Telegraph Co. v. Public Utilities Commission,
390 A.2d 8, 24 (Me.1978), we found “a reasonable likelihood that [NET] would lose its ability to take accelerated depreciation or, at least, be subject to federal action with respect thereto,” as a result of the Commission’s arbitrary action, which, therefore, constituted an abuse of discretion. In its decision and order the Commission expressly determined that its flow-through of the excess deferred taxes would not place NET in jeopardy of losing its ability to take accelerated depreciation. This finding of fact may not be upset on appeal if it was supported by substantial evidence in the record.
Central Maine Power Co.
v.
Public Utilities Commission,
414 A.2d 1217, 1232 (Me.1980).
In addition to concluding that the contemplated amortization would not run afoul of section 167(7), the Commission was also influenced by the fact that in the four (now five) years since the corporate tax rate has been lowered, the Internal Revenue Service has taken no action on the matter. During that period several state commissions have made the adjustment challenged here.
Our own Commission approved a similar adjustment in a 1982 proceeding, upon the advice of the utility involved.
Re Bangor Hydro-Electric Co.,
46 PUR 4th 503 (Me.P. U.C.1982).
The assumption that amortization of excess deferred taxes will not jeopardize accelerated depreciation has become increasingly widespread in the face of prolonged nonresponsiveness by the Internal Revenue Service.
In sum, the Commission was entitled to determine whether it would violate section 167(7) to amortize NET’s excess deferred taxes. In contrast to the 1978
New England Telephone
case, the Commission’s action was not arbitrary, because it found, and had an adequate basis for finding, that NET would not be placed in jeopardy of losing the tax advantage of accelerated de
predation. Therefore, as in the 1982
New England Telephone
case, we reach the issue of whether the Commission’s interpretation of the applicable tax provisions contained legal error and again conclude that it did not.
III.
In 1971, Congress enacted provisions for an investment tax credit. I.R.C. §§ 38, 46.
One purpose of the credit is to stimulate investment in new plant by providing as a credit against tax liability a statutory percentage of the investment during the tax year.
See New England Telephone & Telegraph Co. v. Public Utilities Commission,
448 A.2d 272, 304 (Me.1982). When a public utility uses this credit, known as JDITC (Job Development Investment Tax Credits), to finance a portion of its rate base, it incurs no capital costs for this portion and hence pays no interest.
In determining a utility’s federal income tax expense, as an element of cost of service, the Commission must determine interest expense, a deductible item. In the proceeding below, the Commission imputed an interest expense on the portion of NET’s rate base financed by JDITC. The rationale for doing so is essentially to extend part of the benefits of JDITC to the ratepayer. If the absence of interest expense on JDITC-financed plant is recognized for rate-making purposes, the result is to charge the consumer more for this plant, for which the utility incurs no capital costs, than for ordinary plant. By imputing interest, the Commission merely lowers the burden on ratepayers to the same level as for all other plant in the utility’s rate base.
NET had objected to this approach during the hearings on the grounds that it is wrong to impute expenses never actually incurred and that to do so might jeopardize continued eligibility to receive JDITC. On appeal, NET has abandoned its second argument, which we expressly rejected last year.
New England Telephone & Telegraph Co. v. Public Utilities Commission,
448 A.2d 272, 304-09 (1982). On the other hand, it pursues its first argument, although our 1982 decision forecloses this line of attack as well.
The fact that NET does not actually incur an interest expense was no less plain in last year’s appeal.
Id.
at 304. We also noted that if an interest expense is imputed, “the ratepayers bear the same cost of service, excluding the ratable reduction permitted by the statute, as they would pay in the absence of JDITC.”
Id.
at 309. We found that “one of the dominant themes of the legislative history [of the Revenue Act of 1971 is] ... that the benefits of the tax credit be shared by both investors and consumers. Under the Commission’s interest-synchronization method, both the shareholders and the rate-payers benefit from the credit.”
Id.
at 307. We concluded that the Commission’s treatment of JDITC was “reasonable.”
Id.
at 309.
That precedent controls in the case presently before us.
IV.
Under the License Contract between NET and its former parent, American Telephone and Telegraph Company, NET must pay an allocated share of the costs the latter company incurs in providing certain services. These include research and systems engineering, provided by Bell Telephone Laboratories. Research consists of attempts to understand natural laws and phenomena fundamental to communications technology and to investigate future application of such knowledge to the communications field. Systems engineering is devoted to the formulation of methods of planning, operating and maintaining the nationwide telecommunications network.
Although NET contributes to the funding of these operations as a current expense, the Commission chose to capitalize and amortize this expense over ten years. The basic reason was that these operations
provide benefits to NET that endure for well over a year. NET calls this an “unsupported characterization,” yet even its own witness admitted that these operations generally yield long-term benefits. Thus, the Commission’s finding of long-term benefits was supported by “substantial evidence in the record.”
New England Telephone & Telegraph Co. v. Public Utilities Commission,
448 A.2d at 278.
As we observed in
New England Telephone & Telegraph Co. v. Public Utilities Commission,
390 A.2d at 23, “In order to carry out its rate-making obligation, a regulatory commission must be able to go beyond the utility’s books of account.” Its rate-making obligation is principally to ensure that rates are just and reasonable, balancing the essential revenue needs of the utility with the value of service to the consumer and his ability to pay.
Central Maine Power Co. v. Public Utilities Commission,
150 Me. 257, 109 A.2d 512 (1955); 35 M.R.S.A. § 51. The value of service to the consumer varies over time. Therefore, it is important that costs and benefits of services should be distributed over time so that, as nearly as possible, the same ratepayers who are charged for a service will receive the benefit therefrom.
See Re Central Maine Power Co.,
26 PUR 4th 388, 402 (Me.P.U.C.1978),
cited in Central Maine Power Co. v. Public Utilities Commission,
433 A.2d 331, 343 (Me.1981). By capitalizing the expenses for research and systems engineering, the Commission reasoned, it would be spreading the costs to future ratepayers who doubtless would share in the benefits. Given the latitude that the Commission enjoys in selecting among rate-making techniques,
Mars Hill & Blaine Water Co. v. Public Utilities Commission,
397 A.2d 570, 576 (Me.1979), we conclude that the Commission did not abuse its discretion.
The entry is:
Decision and order of the Commission affirmed.
All concurring.