concurring in part, dissenting in part.
I concur with the majority in all but section III of the opinion which holds that the Utilities Commission properly increased Duke’s accumulated depreciation account, thereby reducing the rate base, by the amount of $3,879,000 as an offset to an adjustment in that same amount made by Duke to reflect its probable future depreciation expense. This portion of the majority decision is, in my view, patently erroneous and I respectfully dissent therefrom. The reasoning of the majority is strained and untenable as I shall demonstrate below.
The primary reasoning of the majority, as I understand it, is that a reading of all the sections and subsections of G.S. 62-133 together justifies the Commission’s action, relying on the principle that statutes in pari materia are to be construed together. The majority acknowledges the well-established principle of statutory construction that a statute dealing with a specific situation controls with respect to that situation, but then rejects that principle (and G.S. 62-133(b)(l) — the specific statute here involved) as being inapplicable here. The majority concludes, “the legislature intended to establish an overall scheme for fixing rates, and it must be interpreted in its entirety in order to comply with the legislative intent.” With that conclusion I wholly agree. My problem is not with the principle of statutory construction relied on by the majority; it is with its application, or rather with its inconsistent application, here. This is so because after stating the broad construction approach, the majority then proceeds to pick and choose among the numerous subsections of the statute (subsections other than but like G.S. § 62-133(b)(l) dealing with specific matters) in determining what it considers to be the “overall scheme for ratemaking” while completely ignoring the clearest admonition found in all of the ratemaking statute —that the only depreciation to be deducted from the original cost of the utility’s property is that “which has been consumed by previous use recovered by depreciation expense,” G.S. § 62-133(b)(l). The majority never rationalizes its refusal to apply this clear language except to hold that when read in conjunction with other parts of the statute, this subsection does not mean what it says!
*32I am unable to follow this kind of reasoning. Neither the majority nor I, of course, can be certain of what the Legislature meant to say in the ratemaking statute. We can, however, be certain as to what the Legislature did say, and, given the choice, I opt to believe that it said what it meant. That a legislative body said what it meant is, I believe, the first rule of statutory construction an appellate court should apply. When the language of a statute is clear and unambiguous, an appellate court should not strain to apply other canons of statutory construction to render that statute ineffective. Phillips v. Phillips, 296 N.C. 590, 596, 252 S.E. 2d 761, 765 (1979). As Chief Justice Sharp stated in Phillips, “It is true that statutes dealing with the same subject matter must be construed together. ‘When, however, the section dealing with a specific matter is clear and understandable on its face, it requires no construction.’ ” Id. (quoting Utilities Commission v. Electric Membership Corp., 275 N.C. 250, 260, 166 S.E. 2d 663, 670 (1969)). I shall proceed to analyze the applicable statute with that principle foremost in mind. This, incidentally, does absolutely no violence to the “in pari materia” rule on which the majority relies. Indeed, it causes the correct application of that rule because, in determining the “overall scheme” for ratemaking, I shall not, as has the majority, blatantly ignore a plainly worded subsection of a statute. That is my understanding of the “in pari materia” rule — that all parts of an act should be considered, not just those parts which might lead to a desired result. By applying these rules correctly, I shall attempt to demonstrate below that the “overall scheme” established by statute did not authorize the Commission to increase Duke’s accumulated depreciation beyond that actually recovered at the end of the test year and, hence, that the Commission improperly reduced the rate base.
While there are several positions taken by the majority with which I disagree, as noted below, its most serious error in determining the “overall scheme” from our statute, in my view, is its failure to understand the bifocal thrust of G.S. 62-133. The statute is not, as the majority apparently believes, limited to a single time factor for rate determination. Two separate factors, an understanding of which is absolutely critical to the issue here posited, is clearly contemplated by the statute.
The critical point I wish to make can best be explained by first summarizing the ratemaking procedure provided by G.S. *3362-133. I am the first to concede that this statute is disjointed, somewhat inartfully drawn and confusing. Careful analysis, however, reveals that the Legislature provided a methodical scheme of ratemaking.
The first point to be made is that G.S. 62-133 contemplates that the fixing of rates to be charged by a public utility is based on two separate factors. The failure of the majority to recognize this critical dichotomy leads, I think, to the majority’s failure to understand the two factors contemplated by the statute and how those factors affect the rate determination. The two factors which must be determined in order properly to fix rates are (1) the rate base, and (2) the utility’s probable future expenses. These factors differ both in terms of the type of data making up each and in terms of the relevant time at which each is measured. The majority opinion completely confuses the time periods and, thus, treats the entire ratemaking process as if only the test year period were involved. Again, failure to understand the bifocal nature of rate determination leads to the misunderstanding that only one time period is involved. The following brief summary of the statute will, I believe, confirm this conclusion.
For the purpose of the question presented by this appeal, the following is an accurate summary of the relevant provisions of G.S. 62-133, the ratemaking statute.
The Commission, as a first step in setting rates, must ascertain the rate base. This is done by determining the reasonable original cost of the public utility’s property in service as of the end of the test period. From that amount is subtracted accumulated depreciation “which has been consumed by previous use recovered by depreciation expense.” To this amount is added the reasonable original cost of investment in plant under construction. The resulting figure is the rate base. G.S. 162-133(b)(l). Thus, the rate basé factor involves data concerning unrecovered cost of property as of the end of the test period. The rate base, which for convenience I shall call Factor 1, is then multiplied by what the Commission has determined to be a fair rate of return and the product represents a fair return on the property in service at the end of the period. The time frame for the computation of Factor 1, the rate base, is “the end of the test period,” as provided in both G.S. 62-133(b)(l) and G.S. 62-133(c). The sole excep*34tion, also provided by subsection (c), is that in determining the rate base the Commission shall consider evidence of actual cost changes based upon circumstances and events occurring after the test period but before the hearing. This exception has no application in the case before us. The events and occurrences which led to the depreciation adjustments in question here took place only during the test period itself.
The second factor to be determined is that of the utility’s probable future operating expenses. This figure is then added to the product of the rate of return and the rate base in order to fix the appropriate and lawful rates to be charged by the utility. This second factor, however, involves separate accounting data and has a timing consideration different from that of the rate base. As stated above, the rate base is determined as of the end of the test period based on events which have already occurred and, in this sense, is a fait accompli. The expenses making up the second factor, however, are those likely to occur in the future and involve an estimation. This estimation of probable future expenses is based on the level of plant in service at the end of the test period. While Factor 2 relies on the level of plant in service as established by Factor 1 for its estimation, the two factors encompass two entirely different periods of time. The first factor looks to the past, the second to the future.
The statutory dichotomy is thus clear; the first factor, the rate base, concerns the original cost of the plant in service at the end of the test year (events which have already occurred); the second, the probable future expenses (projections based on the level of plant in service at the end of the test year). When Factor 1 (rate base) is multiplied by the rate of return and that product is added to Factor 2 (future expenses), their sum represents the level of revenue deemed adequate by the Commission to pay the utility’s reasonable operating expenses and to give its investors a fair return on their investment. This sum is the amount of revenue the utility will receive through sale of its services. In tabular form, I would summarize the ratemaking procedures as it applies- to this case1 thusly:
*35Statutory Formula
(Factor 1 x Rate of Return) + Factor 2 = Revenues to be earned through charges for utility’s service
Factor 1 (Rate Base) = Original cost of plant in service at end of test year
less: accumulated depreciation “which has been consumed by previous use recovered by depreciation expense”
plus: reasonable original cost of plant under construction.
Factor 2 = Probable future expenses based on level of plant and equipment in operation at the end of the test period.
The Statutory Time Frame
For Factor 1
End of the test period (except for adjustments for events occurring after test period as allowed by G.S. 62-133(c)).
For Factor 2
Determine probable future expenses based on the plant and equipment in service at the end of the test period.
Applying the foregoing to the record before us, I think the Commission erred at one crucial point in the ratemaking process. In its application, Duke properly submitted a figure representing the original cost of its plant in service on 31 December 1979, the end of the test period. Subtracted from this figure is the amount of accumulated depreciation reserve on the utility’s books at the close of the test period. The accumulated depreciation reserve is the book entry representing property “which has been consumed by previous use recovered by depreciation expense” as of the end of the test period, the precise computation prescribed by G.S. 62-133(b)(l). Duke also submitted a figure representing its probable future (not test period) depreciation expense based on the plant and equipment in service at the end of the test period.2 The *36Commission, however, subtracted not only the accumulated depreciation on Duke’s books as of the end of the test period but also the sum of $3,879,000 (the same amount as Duke’s estimation of its future depreciation expense) as a pro forma adjustment to the accumulated depreciation for the test year. As I understand it, the Commission’s reasoning is that it regards the pro forma adjustment to accumulated depreciation to be a necessary accounting corollary to the pro forma adjustment which Duke made for future depreciation expense. (This figure represents what the depreciation expense for the full year would have been had the depreciated property in question been in service throughout the year. It had actually come into service at various times throughout the year.) The Commission acknowledged that the additional $3,879,000 added to the accumulated depreciation, thereby reducing the rate base, “ha[d] not been collected from the company’s customers.” The Commission also stated in its order that it was relying on its precedent in a previous rate proceeding. The Commission concluded that the adjustment to the rate base achieved “a proper and equitable matching of revenues and costs.” In so acting, I believe the Commission committed serious error.
As explained above, I think the majority’s error in affirming the Commission’s action results from its failure to understand the different factors established by our statute and the relevant time frames encompassed by those factors. G.S. 62-133(b)(l) is plain and unambiguous: it requires that the rate base (Factor 1) be based on plant in service at the end of the test period, less only that depreciation which represents cost previously consumed and recovered by depreciation expense. The Commission violated this portion of the statute here by subtracting from the company’s original plant cost accumulated depreciation which had not been consumed and recovered by depreciation expense as of the end of the test period. Duke, however, was entitled to the additional depreciation expense by way of an adjustment under Factor 2 because that factor is based on an entirely different time frame —the projection of probable future expenses (including, of course, depreciation expense).
This Court has spoken to this issue before. In State ex rel. Utilities Commission v. Edmisten, 291 N.C. 327, 230 S.E. 2d 651 (1976), it was said:
*37Apparently the Attorney General is arguing that the Commission must assume the utility’s operating expenses will remain the same as they were during the test period in setting rates for some future period. This is not the law. Rate schedules are set with an eye no less toward the future than to the past. General Statutes 62-133(b)(2), (b)(3) and (c) contemplate that the Commission will consider “probable future revenues and expenses” in setting rates for the future. “Obviously, conditions do not remain static.” . . . The company’s experience during the test period regarding revenues produced and operating expenses incurred “is the basis for a reasonably accurate estimate of what may be anticipated in the near future if, but only if, appropriate pro forma adjustments are made for abnormalities which existed in the test period and for changes in conditions occurring during the test period. ...” ... Estimates regarding probable future revenues and expenses, however, must be based upon the utility’s plant and equipment actually in operation at the end of the test period.
291 N.C. at 342, 230 S.E. 2d at 660 (emphasis added) (citations omitted).
The Commission’s action here, condoned by the majority, plainly violates the underlined portion above. This is so because of the requirement that estimates regarding probable future revenues and expenses must be based upon the utility’s plant and equipment actually in operation at the end of the test period. Here, by reducing the rate base through the pro forma adjustment to accumulated depreciation, the result is to produce a smaller net plant in service than that “actually in operation at the end of the test period.”
The majority stresses that G.S. 62-133(b)(l) must be read in conjunction with G.S. 62433(c). The view taken in this dissent does indeed read these statutes in conjunction with one another. Both clearly provide that the original cost of a public utility’s property is to be determined as of the end of the test period and provide no support for the Commission’s action.
Nor am I able to find any support for the Commission’s action in G.S. 62433(d), which allows the Commission to consider “all other material facts of record that will enable it to determine *38what are just and reasonable rates.” This provision, in my opinion, is a catch-all intended to cover unusual situations which may arise in future rate cases and are too numerous and too complex to be dealt with by legislation. Depreciation expense is not such an “unusual situation.” Utilities will always have probable future depreciation expense, and most plants do not initially come into service on January 1. I believe that the Legislature fully and explicitly dealt with the situation of a plant’s coming into service in mid-year with G.S. 62-133(b)(l) and (c). These subsections make clear the time frames within which the rate base and expense components are to be measured. I cannot believe that the Legislature relegated the treatment of such a common situation to a general catch-all provision.
When adjustments are made within the test period, as here, the applicable portion of G.S. 62-133 is that referring to “probable future revenues and expenses.” And, the first sentence of the statute clearly provides that these probable future expenses shall be based on plant and equipment in operation at the end of the test period. On this point, the statute is about as clear and unambiguous as it could possibly be.
I might add also that this view of the statute seems appropriate to me from an economic standpoint as well. As stated in J. Bonbright, Principles of Public Utility Rates 197 (1961):
What [the accumulated depreciation reserve] represents is the amortized costs of the assets in the sense of that part of the costs which has already been charged, or which should have been charged, to previous periods of operation. “Cost minus depreciation” is therefore a shorthand expression for costs remaining to be amortized by future charges to operation and hence indirectly by future charges against the consumers of public utility service.
In other words, the purpose of the depreciation reserve in utility regulation, as I understand it, is to recognize that amount of the cost of plant which has already been recovered through rates. This Court acknowledged this principle in State ex rel. Utilities Commission v. Heater Utilities, Inc., 288 N.C. 457, 219 S.E. 2d 56 (1975). Based on this understanding of the purpose of an accumulated depreciation reserve, it obviously follows that when a pro forma adjustment is made to that reserve to reflect *39something which has not been recovered through previous rates, then the utility will never recover depreciation on that increment and the rates are, therefore, deficient under our statute.
This Court has decided numerous cases which hold that the Commission is not free to devise its own principles of ratemaking but must comply with the requirements of Chapter 62 of our General Statutes. E.g., State ex rel. Utilities Commission v. Duke Power Co., 285 N.C. 377, 206 S.E. 2d 269 (1974). Moreover, a ratemaking practice of the Commission, even one which it has followed for many years (as I suspect is the case here) and is commonly accepted in other jurisdictions, is unlawful if contrary to G.S. 62-133(b)(l). I would also add that my reading of the appellee’s brief compels the conclusion that the Commission has taken the action in question here primarily on the basis of what it considers to be a fundamental accounting principle, that of double entry bookkeeping. While the Commission’s action may accord with sound accounting principles, it does not necessarily follow that those accounting principles apply to ratemaking. Ratemaking is a statutory matter; in setting up a ratemaking procedure our Legislature is free to adopt any process it so chooses, even one which does not follow accounting principles. I believe it has done so here. While accounting principles may be helpful in ratemak-ing, they do not control the process.
I wish also to make these additional observations:
(1) The majority devotes several pages to explaining that various subsections of G.S. 62-133 must be‘ read in conjunction with one another. As noted above, I agree that the subsections of a statute should be construed together. Duke itself does not contest this canon of statutory construction. The majority’s lengthy discourse lends absolutely no support to the majority’s ultimate conclusion that the Commission’s pro forma offset to accumulated depreciation “should be made.”
(2) I also think it worthwhile to note that the majority further strains to rationalize its decision by quoting at length from Public Staff Witness Winters’ testimony. Mr. Winters testified as to the Public Staff’s reasoning behind seeking the offset to accumulated depreciation. After giving extensive excerpts from Winters’ testimony on this point, the majority simply states, “We find that the Commission was fully justified in that conclusion.”
*40I find it a novel approach to judicial decision-making to rely on the testimony of a witness to determine whether the action of an administrative agency is affected by error of law. It was my understanding, apparently mistaken, that questions of law were matters for the courts and not for a witness’s speculation.
(3) Without any real explanation, the majority opinion also notes that to hold in favor of Duke concerning the issue before us would result in rates which would yield a “windfall” to Duke at the expense of the ratepayers. In this instance, any resulting unfairness is to Duke, not the ratepayers. This is so first because the depreciation expense definitely will be incurred by Duke. This is acknowledged by the Commission itself. It seems logical to me that, without the adjustment for future expenses, the expenses would be understated and the rates therefore deficient. An analogous situation would be an increase in wage rates. Should employee wages be increased during the test period, obviously the higher wages being paid at the end of the test period must be considered in predicting probable revenues and expenses for the future even though they were not paid throughout the test period.
The rate base, on the other hand, must include all property serving the public and should not be artificially reduced. When the rate base is decreased by adjusting the accumulated depreciation by an amount which exceeds that recorded on the company’s books, the company will not earn the fair rate of return on its total plant. I simply find no logic or fairness in the position that the rate base should be reduced by taking depreciation on utility property which has not yet occurred at the end of the test period when our statute clearly provides that the rate base is supposed to be figured on the basis of that precise point in time.
When G.S. 62-133 is correctly understood as explained above and in light of the completely unambiguous language of G.S. 62-133(b)(l), it is crystal clear that the Commission erred as a matter of law in making the pro forma offset. Indeed, without quarreling with the majority as to the appropriate standard of review, I believe the Commission exceeded its statutory authority in violation of G.S. 62-94(c)(2). This is so because with respect to the rate base I find absolutely no authority in our statutes to make pro forma adjustments to the rate base for events occurring *41within the test period. In this connection, the plant is to be considered as it actually existed at the end of the test period.
In fairness to the majority, I must confess that my initial study of the briefs submitted led me to the same conclusion which it has reached. I was persuaded at that time by the very appealing argument that normal accounting practices would call for the offset ordered by the Commission. Having now determined, after painstaking study of the statute, that the Legislature had a different methodology for ratemaking in mind, as explained above, I must respectfully dissent. The methodology for ratemaking is the prerogative of the Legislature, not that of the Commission or this Court.
For the reasons stated above, I vote to reverse the Court of Appeals’ decision on the point here discussed. The case should be remanded to that court with instructions that it remand to the Utilities Commission with directions that it recompute the rates after removing the pro forma adjustment for accumulated depreciation.
Chief Justice BRANCH and Justice Exum join in this dissenting opinion.. Again, it must be noted that although G.S. 62-133(c) allows adjustments to the rate base for actual changes in cost occurring after the close of the test year, that subsection is inapplicable to this case. The formula set forth in this opinion is for those cases in which no adjustments are made for events occurring after the test year.
. The majority’s assertion that Duke made a pro forma adjustment to its test year depreciation expense is incorrect; this figure represents probable future expenses.