Opinion by
Judge Craig,
Two telephone utilities, Continental Telephone Company of Pennsylvania and Quaker State Telephone Company, appeal from orders of the Pennsylvania Public Utility Commission (PUC or commission) that approved rate increases for the utilities but allowed them to recover in rates only a portion of the federal tax expenses that they claimed in their rate filings.
The sole issue raised by the utilities on this appeal is whether the PUC erred as a matter of law by recognizing tax savings that they realized because of their participation in a consolidated tax return of their parent corporation, in view of the utilities’ claim that the PUC order is contrary to a private letter ruling issued by the federal Internal Revenue Service (IRS), after the appeal to this court, stating that such a reduction of their federal income tax expense would cause them to be in violation of certain provisions of federal tax law requiring public utilities to normalize benefits they realize from making use of accelerated depreciation provisions of the tax law.
[28]*28“Normalization” is a ratemaking concept designed to adjust a utility’s operating expenses in its test year by eliminating abnormal, non-annual events which are known and certain to change in a regularly recurring manner. As applied to the effect of accelerated depreciation on public utility income tax expenses, normalization permits the utility to include in its current rate calculations an income tax expense higher than what the utility has actually paid, on the theory that the taxes saved by accelerated depreciation are merely deferred. Under sections 167(1) and 168(e)(3) of the Internal Revenue Code, 26 U.S.C. §§167(1) and 168(e)(3), normalization means that utilities employing rapid depreciation of certain of their assets, for tax purposes, must compute their federal income tax expense for ratemaking as if they were employing a straight-line method of depreciation, and they must enter the difference between the large deductions actually taken and the lower straight-line deductions into a deferred táxes account, rather than passing these savings on to customers. Ratepayers realize some benefit from the deferral of taxes because the utility’s rate base is reduced by the amount of the adjustment to the deferred tax reserve. See Barasch v. Pennsylvania Public Utility Commission, 507 Pa. 496, 500 n.1, 505-06, 491 A.2d 94, 95 n.1, 98 (1985) (BaraschlPenn Power) and 26 C.F.R. §1.167(1)-1(h)(6).
The policy behind requiring normalization of deferred tax expense is to provide incentives for capital formation by channeling the benefits of accelerated depreciation to' the company, rather than to the ratepayers. Although the taxes deferred are credited to a reserve for taxes, those funds are available for any proper business purpose, and, in theory, they will be used to finance new capital expenditures until higher taxes fall due. Of course, if the ratemaking process passed the benefits of [29]*29accelerated depreciation through to the ratepayers, the described incentives would be negated. Barasch/Penn Power, 507 Pa. at 505 and n.6, 491 A.2d at 98 and n.6.
Background
When consolidated tax returns are used, each subsidiary of a parent corporation calculates its separate income, deductions, tax liability and tax credits on a stand-alone basis, but each subsidiary does not then file a separate federal income tax return or pay the calculated tax to the IRS. Rather, the subsidiary submits its calculations (and, typically, the amount of its standalone tax liability, if any) to the parent corporation. As is permitted by sections 1501-1505 of the Internal Revenue Code, 26 U.S.C. §§1501-1505, the parent corporation then offsets taxable income generated by some subsidiaries with tax losses and credits generated by other subsidiaries to arrive at a figure representing the taxable income of the consolidated group.
Continental and Quaker State are wholly-owned subsidiaries of Contel Corporation (Contel),1 a holding company with numerous subsidiaries throughout the nation, including both regulated telephone utilities and unregulated companies. The utilities do not dispute that, as a result of consolidated return offsetting of income, the federal income tax return of the parent, Contel, has reflected in recent years an aggregate tax liability that has been significantly less than the sum of the subsidiaries’ liabilities computed on a stand-alone [30]*30basis. Brief of Petitioners p. 14; R.R. 446a.2 Nevertheless, in their rate increase filings before the PUC, the utilities claimed federal income tax expenses equal to the full amount of their stand-alone tax liabilities.
After hearings and the issuance of recommended decisions by Administrative Law Judge Morris Mindlin, the PUC, on February 6, 1986, entered final opinions and orders approving rate increases for Continental and Quaker State. The commissions orders reduced the utilities’ claimed federal income tax expenses on the basis of the commission’s conclusion that Pennsylvania law requires the PUC to disallow as an expense for ratemaking purposes any federal income tax expense claimed by a utility that the utility does not actually pay to the federal government. The commission acknowledged that, if an adjustment to a utility’s claimed tax expense caused a violation of the normalization requirements of the Internal Revenue Code, thereby depriving the utility of the opportunity of benefitting from accelerated depreciation deductions, to the overall detriment of its ratepayers, then the adjustment would be “objectionable.”
The reductions to the claimed tax expenses that the PUC adopted were those proposed by the expert witness for the Office of Consumer Advocate (OCA), Thomas S. Catlin.3 Mr. Catlin explained his method of calculating the adjustments as follows:
The first step in my calculations was to compare the taxes which the members of the Contel [31]*31system would have paid on a separate return basis to the actual consolidated income taxes of the system to determine the savings from consolidation in each year. I then reduced these savings to exclude those amounts resulting from the losses of the regulated companies. I then determined Continentals and Quaker States share of the net savings in each year based on their tax liability on a separate return basis compared to the tax liability of all members of the system on a separate return basis. Finally, I averaged the tax savings for the years 1981 through 1983 and applied the intrastate factor for income tax deductions to determine my adjustment.
OCA Statement No. 3, p. 41; R.R. 422a.4
The commission expressly preferred the adjustments proposed by OCA over those proposed by the commissions Trial Staff, because OCAs proposals were “effectuated only to the extent of current taxes.” R.R. 162a, 271a. That is, the OCA proposal applied the consolidated tax savings adjustment to reduce the amount of current taxes to zero, if the savings were that great, but not beyond zero. The utility’s allowance for deferred federal income tax, calculated pursuant to the accelerated depreciation provisions, remained untouched. Even where [32]*32the consolidated tax savings were greater than the total current taxes, the excess consolidated savings were not applied to- reduce the; amount of deferred taws.
[33]*33Accounting and ratemaking would be impossible if dollars in a common fund could not be labeled and segregated into accounts.
From this view, a reduction to current taxes is not tantamount to a reduction to deferred taxes in the same pool of dollars. And the sequential reduction of current taxes can be validly followed by the addition of provisions for deferred taxes or ITC [Investment Tax Credits]. Therefore, on conceptual grounds, we reject Mr. Charins reasoning.
R.R. 162a, 270a.6 The commission found support for its .conclusion that consolidated tax savings adjustments are not per se violative of the normalization requirements of the Internal Revenue Code in the fact that, although many state regulatory authorities throughout the nation had used consolidated tax savings adjustments for over a [34]*34decade, while the normalization provisions were in effect, no formal IRS action (i.e., deficiency assessments) or legislative action had been taken to correct any perceived conflict.
The consolidated tax adjustments adopted by the PUC resulted in reductions of $138,124 to Continentals claimed federal income tax expense and $128,118 to Quaker States tax expense. R.R. 206a and 323a.
Continental and Quaker State filed petitions for review challenging the lawfulness of the PUCs consolidated tax savings adjustments on March 7, 1986. On June 25, 1986, Senior Judge Lehman of this court granted the utilities’ .petition for a stay of proceedings to permit Contel to seek a private letter ruling from the Internal Revenue Service on the question of whether the PUCs application of a consolidated fax savings adjustment to the utilities’ claimed federal tax expense allowances would cause violations of sections 167(1) and 168(e)(3) of the Internal Revenue Code, 26 U.S.C. §§167(1) and 168(e)(3), which require that public utilities normalize the benefits they realize by taking advantage of accelerated depreciation provisions of the Code with respect to public utility property.7
The IRS issued a private letter ruling to the utilities on December 15, 1986, stating that application of the consolidated tax savings adjustment would violate the normalization and consistency provisions of the Code. Following reinstatement of the briefing schedule by this court, briefs have been filed by the utilities and the PUC, by the Consumer Advocate as intervening re[35]*35spondent, and by the National Association of Regulatory Utility Commissioners (NARUC) as amicus curiae.
“Actual Taxes Paid” Doctrine
This courts scope of review of a final order of the Public Utility Commission is to determine whether there has been a constitutional violation or an error of law and whether the findings of fact are supported by substantial evidence. 2 Pa. C. S. §704; Bell Telephone Co. of Pennsylvania v. Pennsylvania Public Utility Commission, 107 Pa. Commonwealth Ct. 34, 528 A.2d 268 (1987), appeal dismissed, Pa. , 552 A.2d 1046 (1989).
The Pennsylvania Supreme Court considered the question of what treatment to afford tax savings of utilities resulting from participation in a consolidated tax return in Barasch v. Pennsylvania Public Utility Commission, 507 Pa. 561, 493 A.2d 653 (1985) (Barasch/ UGI). In that case the PUC had approved a natural gas utility’s full stand-alone federal tax expense claim despite the fact that the utility participated in a consolidated tax return with its parent and other subsidiaries, resulting in an aggregate tax liability significantly less than the sum of the members’ tax liabilities as individually computed.8
On appeal, this court decided to reverse and remand for adjustment to the tax expense allowance. Cohen v. Pennsylvania Public Utility Commission, 78 Pa. Com[36]*36monwealth Ct. 545, 468 A.2d 1143 (1983). We noted that tax expenses are merely one element of operating expenses, which are one of the four basic elements involved in the calculation of utility rates, and that our courts, for many years, had refused to permit utilities to include in their ratemaking equations tax expenses that, because of participation in a consolidated return, they did not actually pay to the government. Cohen, 78 Pa. Commonwealth Ct. at 559, 468 A.2d at 1150.
On further appeal, the Supreme Court affirmed our decision, noting that the arguments advanced by the commission and the utility failed to recognize the basic ratemaking maxim that only expenses that are actually paid or payable by the utility may be included for the purpose of ratemaking. The Court held:
All tax savings arising out of participation in a consolidated return must be recognized in rate-making, otherwise we would be condoning the inclusion of fictitious expenses in the rates charged to the ratepayers.
Barasch/UGI, 507 Pa. at 568, 493 A.2d at 656. Summarizing, the Court said:
It is a violation of basic rate-making principles to charge ratepayers for theoretical expenses [for] which in practice the utility bears no liability. This is true no matter the category of expense. The ratepayers are entitled to the benefits of reduced tax expenses accruing to the utility by participation in a consolidated tax return as was filed in this case.
Id. at 570, 493 A.2d at 657. The Barasch/UGI opinion is a strong statement of the “actual taxes paid” doctrine.
IRS Private Letter Ruling
The utilities respond to this statement of Pennsylvania law by contending that the Internal Revenue [37]*37Services private letter ruling in this matter, essentially adopting the position of the utilities before the PUC and consistent with two other recent private letter rulings, has “resolved” the question of whether the consolidated tax savings adjustment will cause a violation of the normalization requirements of the Internal Revenue Code.
The utilities note that the Barasch/UGI decision did not address the precise issue of possible conflict between the actual taxes paid doctrine and sections 167(1) and 168(e)(3) of the Internal Revenue Code. However, one month before BaraschlUGI was issued, in a decision concerning the proper ratemaking treatment of rapid depreciation deductions for three categories of public utility property covered by three separate federal rapid depreciation statutes (where the utility did not participate in a consolidated return) the Supreme Court held:
If normalization violates the ‘actual taxes paid’ doctrine, the rates approved by the Commission cannot be ‘just and reasonable’ as required by law unless some extraneous penalty, such as that which Congress imposed with respect to post-1980 property, would leave the ratepayer relatively worse off if normalization is not allowed.
Barasch v. Pennsylvania Public Utility Commission, 507 Pa. 496, 521, 491 A.2d 94, 107 (1985) (Barasch/Penn Power) (emphasis added). As noted above, the PUC has acknowledged the potential danger of that “extraneous penalty.”
The utilities here have mischaracterized the legal effect of the IRS private letter ruling. As noted by the PUC in its brief, the federal courts consistently have held that IRS private letter rulings do not have the force and effect of law. In a case involving two private letter rulings (reaching opposite results), and referring specif[38]*38ically to private letter rulings as such (with reliance upon a Court of Appeals decision which held that even published revenue rulings do not have the force of law), a federal court states:
The applicability of the de novo standard determines the weight given to the Commissioners private letter ruling issued to the plaintiff Foundation. Revenue rulings are merely the opinions of lawyers within the IRS. If a ruling is an improper and incorrect interpretation of the law, it is without force. Idaho Power Company v. Commissioner of Internal Revenue, 447 F.2d 688 (9th Cir. 1973, reversed on other grounds, 418 U.S. 1, 94 S.Ct. 2757, 41 L.Ed.2d 535 (1974). If, however, the prior revenue ruling is consistent with the courts separate and independent construction of the exemption provisions, the ruling can serve as some evidence that the courts independent evaluation is correct. Hanover Bank v. Commissioner, 369 U.S. 672, 686-87 and n.19, 82 S.Ct. 1080, 1088 and n.19, 8 L.Ed.2d 187 (1962).
National Right to Work Legal Defense and Education Foundation v. United States, 487 F. Supp. 801, 805 (E.D.N.C. 1979) (emphasis added). See also Dixon v. United States, 381 U.S. 68, 72-73 (1965) (“The Commissioners rulings have only such force as Congress chooses to give them and Congress has not given them the force of law.”). Also, the letter ruling itself has a notice on the first page required for all such rulings: “This document may not be used or cited as precedent. Section 6110(j)(3) of the Internal Revenue Code.” R.R. la.
Hence, the IRS private letter ruling issued to Contel, although a statement of the position of its author and perhaps of the IRS as a whole, is not controlling. In particular, the letter ruling in itself cannot provide a legal basis for invalidation of a PUC order.
[39]*39Although the private letter ruling cannot control this courts conclusion as to the legality of the PUCs adjustments in these cases, the ruling could influence our conclusion if it rested upon compelling law and logic in the present context. However, it does not.
Section 167(1), added to the Internal Revenue Code in 1969, provided a system of accelerated depreciation for public utility property known as Accelerated Depreciation Range (ADR), and required utilities implementing accelerated depreciation for the first time to normalize the savings, rather than to pass those savings through to ratepayers. Section 167(1)(3)(G) added:
(G) Normalization method of accounting. In order to use a normalization method of accounting with respect to any public utility property—
(i) the taxpayer must use the same method of depreciation to compute both its depreciation expense for purposes of establishing its cost of service for ratemaking purposes and for reflecting operating results in its regulated books of account, and
(ii) if, to compute its allowance for depreciation under this section, it uses a method of depreciation other than the method it used for purposes described in clause (i), the taxpayer must make adjustments to a reserve to reflect the deferral of taxes resulting from the use of such different methods of depreciation.
Thus, if a utility taking advantage of accelerated depreciation normalized with straight-line method pro forma, it had to set up a deferred tax reserve.
Amendments to section 168 in 1980 established a new and even more rapidly accelerated depreciation program known as Accelerated Cost Recovery System (ACRS). Section 168(e)(3) required any public utility making use of ACRS to normalize, regardless of its [40]*40previous practice. This section provided a definition of normalization functionally identical to that in section 167(1), and then presented a provision relating to consistency, section 168(e)(3)(C), as follows:
(C) Use of inconsistent estimates and projections, etc.
(i) In general. One way in which the requirements of subparagraph (B) [defining normalization] are not met is if the taxpayer, for ratemaking purposes, uses a procedure or adjustment which is inconsistent with the requirements of subparagraph (B).
(ii) Use of inconsistent estimates and projections. The procedures and adjustments which are to be treated as inconsistent for purposes of clause (i) shall include any adjustment for ratemaking purposes which uses an estimate or projection of the taxpayers tax expense, depreci-
ation expense, or reserve for deferred taxes under subparagraph (B)(ii) unless such estimate or projection is also used, for ratemaking purposes, with respect to the other 2 items and with respect to the rate base.9
The private letter ruling issued to Contel set forth the provisions of section 168(e)(3) in its entirety, and also quoted the following portions of the regulations promulgated pursuant to section 167(1):
The normalization requirements of section 167(1) with respect to public utility property defined in section 167(1)(3)(A) pertain only to the deferral of federal income tax liability resulting from the use of an accelerated method of depreciation under section 167 and the use of the straight line [41]*41depreciation for computing tax expense and depreciation expense for purposes of establishing cost of services and for reflecting operating results in regulated books of account.
26 C.F.R. §1.167(l)-l(a)(l).
. . . [T]he amount of federal income tax liability deferred as a result of the use of a different method of depreciation under subdivision (i) of this subparagraph is the excess (computed without regard to credits) of the amount tax liability would have been had a subsection (1) method been used over the amount of the actual tax liability. Such amount shall be taken into account for the taxable year in which such different methods of depreciation are used.
26 C.F.R. §1.167(l)-l(h)(l).
The letter ruling then analyzed the effect of consolidated tax savings adjustments on normalization requirements as follows:
Tax expense computed under a normalization method of accounting, as so defined, includes both current and deferred tax expense as those terms are commonly used for financial statement purposes. Each of these components is an integral part of tax expense and cannot be excluded. When the same method of depreciation that is used for tax calculations is used for establishing cost of service, the tax calculated is the sum of the two components.
. . . The consolidated tax savings adjustment achieves through current tax expense a cost of service reduction that would violate normalization requirements if achieved through a reduction of deferred tax expense, and the effect is the same as if there were a partial flow-through of [42]*42the benefits of accelerated depreciation as a result of deferring less than all the difference between straight line and accelerated depreciation.
The use of tax benefits attributable to other entities in the determination of deferred tax expense or current tax expense will cause Taxpayer to fail to comply with the normalization provisions of section 167(1) of the Code and the regulations thereunder.
The normalization requirements of section 168(e)(3) of the Code are similar to the normalization requirements in section 167(1). If a proposed method of accounting does not meet the normalization requirements of section 167(1) it will not meet the requirements of section 168(e)(3).
Private letter ruling, pp. 5-6; R.R. 6-7.
The reasoning of the private letter ruling is faulty in several respects. First, the ruling foils to recognize that the amount of taxes to be deferred under sections 167(1) and 168(e)(3) — the difference between the accelerated depreciation deduction and the straight line depreciation deduction times the applicable tax rate — is wholly derived from the value of the utility property that is being depreciated, not from current income or other factors affecting adjustments to current taxes.
The only matters which can be of concern here are those within the boundaries of accounting for depreciation. The Code mandates discrete accounting by requiring that the accelerated depreciation savings be entered into a tax reserve account — the deferred taxes item which, in the present context, refers only to deferral in connection with depreciation.
The letter ruling is correct in stating that any adjustment to such deferred taxes would cause a violation of the normalization provisions, but the ruling then goes [43]*43afield by asserting that a reduction of current taxes would have the same effect. The bottom line in current taxes depends, of course, on a multitude of factors aside from depreciation, including consolidated return savings unrelated to depreciation.
The regulation quoted above expressly states that the normalization requirements pertain only to the deferral of federal income tax liability (resulting from the use of accelerated depreciation for tax liability and straight line depreciation for ratemaking normalization). 26 C.F.R. §1.167(l)-l(a)(l). If the amount of taxes required to be deferred under the normalization provisions is properly calculated and protected from diminution, then the utility has complied with those provisions, and there is no basis for a claim that a reduction in current taxes for other reasons will result in noncompliance.
That protection of the deferred taxes amount insures the implementation of the federal policy requiring accelerated depreciation savings to remain with the utility rather than go to the ratepayer.
A related error in the letter rulings analysis is its statement that the use of tax benefits attributable to other entities in the determination of current tax expense causes the taxpayer to fail to comply with the normalization provisions of section 167(1) and its regulations. As the commission and NARUC point out, the IRS may fear that permitting accelerated depreciation deductions of other, nonregulated, affiliates to contribute to the consolidated tax savings used to reduce the current tax of a utility would violate sections 167(1) and 168(e)(3). However, such a proposition would be erroneous, because those sections require normalization for accelerated depreciation benefits generated by public utility property only. They simply do not apply to benefits generated by non-public utility property.
[44]*44The letter ruling also stated an “additional reason” why the consolidated tax savings adjustment would violate the normalization provisions, namely, the authors conclusion that the PUCs adjustments would violate the consistency provisions of section 168(e)(3)(C), quoted above, by “including] the transactions of all affiliates for the purpose of determining ratemaking tax expense without giving consideration to such transactions in any other ratemaking calculations.” R.R. 7a. This conclusion also is in error. As the commission and NARUC again point out, the commissions estimates and projections were internally consistent when the commission calculated the utilities’ depreciation expense, reserves for deferred taxes and rate base. As noted above, “current taxes” — either adjusted or unadjusted — are not an element in the calculation of the amount of taxes required to be deferred under the normalization provisions. Therefore, ,the commissions reducing the utilities’ allowances for current taxes could not, and did not, affect the calculation of deferred taxes for normalization purposes or violate the IRS normalization provisions.
Conclusion
As noted above, the federal policy basis for normalization of deferred tax expense is to encourage capital formation by placing the money currently diverted from payment of taxes into the hands of the company, rather than into the hands of the ratepayers. We conclude that the orders of the Public Utility Commission in these cases properly calculated the benefits accruing to the utilities involved because of their making use of accelerated depreciation provisions of the Internal Revenue Code and complied with the requirements of the Code that those benefits be normalized. Because the orders assured that the amount of taxps required to be deferred under the normalization provisions would not be [45]*45reduced by the application of a consolidated tax savings adjustment to the utilities’ claimed federal income tax expenses, the orders fully complied with sections 167(1) and 168(e)(3) of the Code.
Accordingly, we shall affirm.
Order
Now, September 26, 1988, the orders of the Pennsylvania Public Utility Commission at Docket Nos. R-850044, et al. and R-850045, et al., dated February 6, 1986, are affirmed.
President Judge Crumlish, Jr. did not participate in the decision of this case.
CTC Exhibit 1G
In Line 4, the Company makes no consolidated tax adjustment. The main difference between Mr. Catlin’s and Mr. Wilsons adjustments in Line 4 is that the former deducts his computed consolidated tax savings only to the extent that intermediate cur[33]*33rent taxes are reduced to zero, whereas the latter deducts the full amount to the extent that, in this case, intermediate current taxes are reduced to a negative amount. In both cases, however, the provision for deferred taxes in Line 6 is added back with positive results, but differ by the amount that the consolidated savings reduction fell below zero. In the bottom line, line 9, the difference of all three methods may be seen. R.R. 161a, 268a-69a.