Peoples Federal Savings and Loan Association of Sidney v. Commissioner of Internal Revenue

948 F.2d 289

• Court: Court of Appeals for the Sixth Circuit
• Decided: April 9, 1992
• Docket: 90-1939
• Status: Published

Opinion

MILBURN, Circuit Judge.

The Commissioner of Internal Revenue (“Commissioner”) appeals the decision of the United States Tax Court determining that the appellee, Peoples Federal Savings and Loan Association of Sidney, Ohio (“taxpayer”), owed a tax deficiency for 1978, but was entitled to refunds for 1979 and 1980.

From 1964 to 1978, the Commissioner had in force a regulation, 26 C.F.R. § 1.593-6(b)(2)(iv), which provided that, for purposes of the “percentage of taxable income” method of computing the deduction for additions to loan loss reserves under 26 U.S.C. § 593, a carryback year’s taxable income would not be reduced by a net operating loss carryback to such year under 26 U.S.C. § 172. However, in 1978, the Commissioner issued new regulations, 26 C.F.R. § 1.593-6(A)(b)(5)(viHvii), which, in effect, reversed the old regulation and provided that net operating loss carrybacks under section 172 reduced the taxable income of a carryback year before the “percentage of taxable income method” deduction was computed thereon. The sole issue in this case is whether the regulations issued by the Commissioner in 1978 are valid. For the reasons that follow, we reverse and remand.

I.

The facts in this case are stipulated and not in dispute. Taxpayer is a mutual savings and loan association which, during the taxable years 1971-1980, claimed deductions for additions to its reserve for loan losses. It calculated the amounts of those deductions using the “percentage of taxable income” method prescribed by 26 U.S.C. § 593(b)(2)(A) of the Internal Revenue Code. 1 Section 593(b) governs the computations of the amount of the deduction available for additions to loan loss reserves and provides that a deduction taken for an addition to loan loss reserves with respect to loans secured by real property may be computed using one of three different methods. One of these, the “percentage of taxable income” method established in section 593(b)(2)(A), allows a deduction for loan loss reserves computed as a particular percentage of the taxpayer’s “taxable income” for the year. It is this “percentage of taxable income” method that taxpayer used to compute all the deductions for additions to loan loss reserves here in issue. 2

In 1981, 1982, and 1983, taxpayer sustained net operating losses in the amounts of $410,791, $563,491, and $379,652, respectively. By virtue of section 172(b)(1)(F), each net operating loss could be carried back to the ten years preceding each of the loss years and treated as a deduction under section 172(a), thereby reducing the taxpayer’s taxable income, and accordingly its income tax liability, for the year. A net operating loss is carried first to the earliest permissible year where it is used to neutralize taxable income. If the net operating loss carryback exceeds the taxable income for that earliest permissible year, the excess of the net operating loss is carried forward year by year in chronological order to succeeding years until the net operating loss is expended in neutralizing each successive year’s taxable income or, stated another way, until the net operating loss is fully “absorbed” by the taxable income of the carryback year or years. Thus, a net operating loss carryback reduces the taxable income in the carryback year and usually results in an amended return recomputing the taxpayer’s income tax liability and seeking a refund of any taxes previously paid.

In this case, taxpayer took deductions for additions to loan loss reserves from 1971 through 1980 and in each instance used the “percentage of taxable income” method of section 593(b)(2)(A). When it carried back to those years the net operating losses it sustained in 1981, 1982, and 1983, it used the losses to reduce or eliminate taxable income, but it did not re-compute the loan loss reserve deductions it had previously taken based on the original, and therefore higher, taxable income levels. As a result of the net operating loss carrybacks, taxpayer had much less taxable income in the carryback years, and, accordingly, it filed amended returns claiming income tax refunds for those years. The Commissioner determined deficiencies for the tax years 1978-80, and this litigation followed.

The problem in this case has to do with accounting mechanics and arises out of the interaction between the “percentage of taxable income” method for computing loan loss reserve deductions under section 593 and the net operating loss carryback provisions of section 172. Because a net operating loss carryback reduces the taxable income for the year to which it is carried back, the question arises as to whether a taxpayer using the “percentage of taxable income” method for computing deductions for loan loss reserves should compute the deduction for the carryback year before or after the net operating loss carryback is applied to reduce the taxable income for that year. The Commissioner’s position is that the taxpayer must re-compute its loan loss reserve deductions to take into account a net operating loss carryback that has reduced the taxable income of the carry-back year. Because the deduction is computed as a percentage of taxable income, this position results either in a smaller loan loss reserve deduction or no loan loss reserve deduction, depending on whether the net operating loss carryback merely reduces or extinguishes the taxable income of the carryback year. The taxpayer’s position is that any loan loss reserve deduction it has taken in conformity with the “percentage of taxable income” method is presumptively reasonable and therefore need not be re-computed, even when a net operating loss is carried back to that year and results in a reduction or a complete neutralization of the taxable income for that year.

The statutes involved do not expressly provide a solution to the problem in this case. The Commissioner’s regulations do solve the problem, but there is a question raised concerning their validity.

The most recent regulations promulgated under the Commissioner’s broad authority to “prescribe all needful rules and regulations for the enforcement of [the Internal Revenue Code],” 26 U.S.C. § 7805(a), provide that taxable income, for the purposes of the “percentage of taxable income” method, is computed “by taking into account ... any other deduction or loss allowed under subtitle A of the Code, such as any deduction allowable under § 172.” 26 C.F.R. § 1.593-6A(b)(5)(vii). 3 Thus, the current regulations expressly require that taxable income reflect the section 172(a) net operating loss (“NOL”) deduction prior to the calculation of the deduction for an addition to loan loss reserves. This regulation was proposed, in substance, in 1971. Prop.Treas.Reg. § 1.593-6A(b)(5), 36 Fed. Reg. 15050, 15052-15053 (1971). The final regulation was issued in May 1978, T.D. 7549, 43 Fed.Reg. 21453, 21455 (1978), and amended in May 1979. T.D. 7626, 44 Fed. Reg. 31177 (1979).

Prior to the issuance of the current regulations in 1978, the Commissioner had issued regulations that provided for the opposite result. Promulgated in 1964, 26 C.F.R. § 1.593-6(b)(2)(iv) provided that, for purposes of the “percentage of taxable income” method, taxable income for the year would be computed “[without regard to any net operating loss carryback to such year under section 172.” This regulation (“old regulation”) was based on an even earlier revenue ruling, Rev.Rul. 58-10, 1958-1 C.B. 246, in which the Commissioner ruled that, for purposes of the “percentage of taxable income method,” taxable income would not be reduced by a net operating loss carryback.

From all of the foregoing, it is clear that the Commissioner changed course in 1978 when he issued 26 C.F.R. § 1.593-6(A)(b)(5)(vi)-(vii) (“new regulations”). The differences between the applications of the old regulation and the new regulation may be seen in the following computations involving a hypothetical net operating loss carryback of $1.0 million from the loss year 1981.

[[Image here]]

[[Image here]]

It may readily be seen how the application of the new regulation results in the quicker absorption of the net operating loss carry-back and the elimination of any excess net operating loss that might otherwise be carried forward to offset taxpayer’s income in subsequent carryback years.

The Tax Court, after reviewing the legislative history and purpose of the statutes involved, held that the Commissioner’s change of course in 1978 “violated Congressional intent and was therefore an invalid exercise of Treasury’s rule-making authority.” It held that this outcome was required by its recent holding in Pacific First Federal Savings Bank v. Commissioner, 94 T.C. 101 (1990), in which it had ruled that the same regulations at issue in this case were invalid, and it adopted the rationale of Pacific First Federal as its reasoning in this case. 4

As earlier indicated, the issue in this case is whether or not the new regulations, 26 C.F.R. § 1.593-6(A)(b)(5)(vi) and (vii) (1978) are valid.

II.

A.

Courts of appeal have jurisdiction to review the decisions of the Tax Court “in the same manner and to the same extent as decisions of the district courts in civil actions tried without a jury.” 26 U.S.C. § 7482. Because the facts in this case are stipulated and the only issues for review have to do with the Tax Court’s conclusions of law, this court will review the matter de novo. Walter v. Commissioner, 753 F.2d 35, 38 (6th Cir.1985).

B.

Operative Statutes and Regulations

The major questions in this case arise out of the interaction of three statutes. 26 U.S.C. § 63 defines taxable income as “gross income, minus the deductions allowed by this chapter.”

26 U.S.C. § 172(a) provides:

There shall be allowed as a deduction for the taxable year an amount equal to the aggregate of (1) the net operating loss carryovers to such year, plus (2) the net operating loss carrybacks to such year. For purposes of this subtitle, the term “net operating loss deduction” means the deduction allowed by this subsection.

26 U.S.C. § 593 specifies how deductions for additions to loan loss reserves are computed and taken. At all material times prior to 1986 it read as follows:

(a) Organizations to which section applies. — This section shall apply to any mutual savings bank not having capital stock represented by shares, domestic building and loan association, or cooperative bank without capital stock organized and operated for mutual purposes and without profit.

(b) Addition to reserves for bad debts.-

(1) In general. — For purposes of section 166(c), the reasonable addition for the taxable year to the reserve for bad debts of any taxpayer described in subsection (a) shall be an amount equal to the sum of—

(A) the amount determined to be a reasonable addition to the reserve for losses on nonqualifying loans, computed in the same manner as is provided with respect to additions to the reserves for losses on loans of banks under section 585(b)(3), plus

(B) the amount determined by the taxpayer to be a reasonable addition to the reserve for losses on qualifying real property loans, but such amount shall not exceed the amount determined under paragraph (2), (3), or (4), whichever amount is the largest, but the amount determined under this subparagraph shall in no case be greater than the larger of—

(i) the amount determined under paragraph (4), or

(ii) the amount which, when added to the amount determined under subpara-graph (A), equals the amount by which 12 percent of the total deposits or with-drawable accounts of depositors of the taxpayer at the close of such year exceeds the sum of its surplus, undivided profits, and reserves at the beginning of such year (taking into account any portion thereof attributable to the period before the first taxable year beginning after December 31, 1951).

(2) Percentage of Taxable Income Meth od.—

(A) In general. — Subject to subpara-graphs (B), (C), and (D), the amount determined under this paragraph for the taxable year shall be an amount equal to the applicable percentage of the taxable income for such year (determined under the following table):

(E) Computation of taxable income. — For purposes of this paragraph, taxable income shall be computed—

(i) by excluding from gross income any amount included therein by reason of subsection (f),

(ii) without regard to any deduction allowable for any addition to the reserve for bad debts,

(iii) by excluding from gross income an amount equal to the net gain for the taxable year arising from the sale or exchange of stock of a corporation or of obligations the interest on which is excludable from gross income under section 103,

(iv) by excluding from gross income an amount equal to the lesser of % of the net long-term capital gain for the taxable year or % of the net long-term capital gain for the taxable year from the sale or exchange of property other than property described in clause (iii), and

(v) by excluding from gross income dividends with respect to which a deduction is allowable by part VIII of subchapter B, reduced by an amount equal to the applicable percentage (determined under subparagraphs (A) and (B)) of the dividends received deduction (determined without regard to section 596) for the taxable year.

The Commissioner argues that the plain meaning of these statutes requires that any net operating loss carryback under section 172(a) must be used to reduce the taxable income in any carryback year before the taxpayer computes its deduction for loan loss reserves under section 593. The Commissioner contends that

the Code requires that the Section 172(a) net operating loss deduction be taken into account in determining “taxable income” for purposes of Section 63(a), and that the Section 63(a) definition of taxable income be used for purposes of Section 593(b)(2), except as specifically modified by Section 593(b)(2)(E).

Brief of Appellant at 7.

Since the taxpayer filing an amended return in order to take advantage of a net operating loss carryback will already have filed an original return containing a deduction for loan loss reserves, the Commissioner’s interpretation of the interaction among these statutes requires that the original deduction for loan loss reserves be re-computed because the taxable income, of which the deduction is always a percentage under this particular method of computation, has been reduced by the net operating loss being carried back. The regulations in effect since May 1978 have interpreted the workings of these statutes in that manner, and the current regulations thus require the re-computation of any “percentage of taxable income” method deduction under section 593.

Taxpayer argues, however, that these new regulations are the usurpers of an older, contradictory regulation that has been approved by Congress and therefore expresses Congress’ true intent in these matters. That older regulation, 26 C.F.R. § 1.593-6(b)(2)(iv), first issued in 1964, provided that, for purposes of computing the deduction for additions to loan loss reserves under the “percentage of taxable income” method of § 593, taxable income shall be computed “[w]ithout regard to any net operating loss carryback to such year under Section 172.” If the old regulation were applied in this case, the taxpayer would be allowed to carry back its net operating losses to reduce its taxable income in the carryback years without being required to re-compute the section 593 deduction for additions to loan loss reserves.

The old regulation apparently sprang from an even earlier revenue ruling issued in 1958, Rev.Rul. 58-10,1958-1 C.B. 246, in which the Commissioner ruled that a subsequently occurring net operating loss does not render unreasonable an addition to the loan loss reserves that was reasonable at the time it was established. Taxpayer challenged the new regulation in the Tax Court and argued that it was invalid because the old regulation better expressed the intentions of Congress. It pointed out that, from at least 1958, when Rev.Rul. 58-10 was issued, to 1978, when the new regulations were issued, the Commissioner’s position was that deductions for additions to loan loss reserves need not be re-computed when net operating losses were carried back. Congress amended section 593 in the Revenue Act of 1962, Pub.L. No. 87-834, § 6(a), 76 Stat. 960, after the issuance of Revenue Rui. 58-10, and amended section 593 again in the Tax Reform Act of 1969, Pub.L. No. 91-172, § 432(a), 83 Stat. 487, after the old regulation had been issued in 1964. Taxpayer contended that leaving the Commissioner’s old interpretation intact when it might easily have changed it showed Congress’ implicit approval of it.

On the other hand, the Commissioner argued that his new regulations, issued in 1978, were applicable. He argued that the old regulation had been wrong, that it required correction, and that Congress implicitly approved the 1978 correction when it re-enacted section 593 in the Tax Reform Act of 1986, Pub.L. No. 99-514, § 903, 100 Stat. 2085.

The Tax Court’s Opinion

The Tax Court tested the regulation’s validity according to a standard it found in National Muffler Dealers Association v. United States, 440 U.S. 472, 477, 99 S.Ct. 1304, 1307, 59 L.Ed.2d 519 (1979), that is, “whether the regulation harmonizes with the plain language of the statute, its origin, and its purpose.” As for the plain language of the statute, the Tax Court found it “not free of ambiguity” and thus not helpful. Pacific First Fed. Sav. Bank v. Commissioner, 94 T.C. 101, 108 (1990). It went on to consider, therefore, the origin and purpose of the statute to see whether the new regulation was “fundamentally at odds with the manifest congressional de sign.” United States v. Vogel Fertilizer Co., 455 U.S. 16, 26, 102 S.Ct. 821, 828, 70 L.Ed.2d 792 (1982).

Origin

The legislative history reveals that prior to 1952 mutual savings institutions were exempt from federal income tax. The Revenue Act of 1951, Pub.L. No. 81-183, 65 Stat. 452 (1951), repealed the tax exempt status of these mutual institutions and subjected them to the regular corporate income tax. It also provided them with a special deduction for additions to loan loss reserves equal to 100 percent of net income, provided the total reserves and surplus did not exceed 12 percent of deposits. This deduction was carried forward into the Internal Revenue Code of 1954 as section 593. Its terms were altered, however, to provide a deduction equal to the amount of the institution’s “taxable income” rather than “net income” as before. 26 U.S.C. § 593 (1954).

The Section 593 deduction for additions to loan loss reserves proved to be so large in practice that mutual savings institutions remained virtually tax exempt. In the Revenue Act of 1962, Pub.L. No. 87-834, § 6(a), 76 Stat. 960, Congress amended section 593 to limit the “percentage of taxable income” method deduction to 60 percent taxable income, 5 and further reduced the deduction by providing special exclusions from gross income to be applied in the determination of “taxable income” for the purposes of section 593. According to a House Report, Congress had two motives for making the foregoing changes. First, it wished to tax mutual savings institutions. Second, it wished to allow them a special deduction to insure that ample reserves for loan losses were maintained. H.R.Rep. No. 1447, 87th Cong., 2d Sess. 33 (1962).

In 1969, Congress decided that its 60 percent formula of 1962 had been too generous, and it modified the formula by incrementally reducing the deduction from 60 percent to 40 percent of taxable income over a ten-year period ending in 1979. 26 U.S.C. § 593(b)(2)(A) (1969). It also added new exclusions from gross income to the list it had begun in 1962, thereby further reducing “taxable income” for purposes of the deduction by contracting the base to which the percentage formula was applied. 26 U.S.C. § 593(b)(2)(E) (1969). Congress thought that “the present bad-debt reserve provisions are unduly generous as they have allowed these institutions to pay a much lower average effective rate of tax than the average effective rate for all corporations.” H.R.Rep. No. 91-413 (Pt. 1), 91st Cong., 1st Sess. 125 (1969), U.S.Code Cong. & Admin.News 1969, 1645, 1775. It retained the truncated deduction, however, because it wanted to “preserve the inducement for [mutual savings institutions] to continue investing in real estate mortgages,” and to encourage “reserves consistent with the proper protection of the institution and its policyholders in the light of the peculiar risks of long-term lending on residential real estate which is the principal function of these institutions.” Id.

In addition to amending section 593, Congress took the opportunity in 1969 to amend section 172 by extending the net operating loss carryback period from three to ten years for mutual savings institutions. Tax Reform Act of 1969, § 431(b) (adding 26 U.S.C. § 172(b)(1)(F)). It did this because it thought that the longer loss carryback period was “a better means to provide for large unexpected losses than to allow such institutions to build up their reserves tax free.” H.R.Rep. No. 91-413 at 125, U.S.Code Cong. & Admin.News 1969, at 1775. Congress apparently thus amended section 172 in order to counterbalance somewhat the cuts it had made when it amended section 593 so as to reduce the value of the “percentage of taxable income” method for computing deductions for additions to loan loss reserves.

The legislative history of the changes made in 1969 also discloses a congressional intent to raise the effective rate of taxation of mutual savings institutions. The House Report commented:

Since your committee’s bill increases appreciably the 23.2 percent effective rate of tax for commercial banks, it is your committee’s intention not only to bring the level of taxation of mutual savings banks (presently 6.1 percent) up to the level of savings and loan associations (16.9 percent), but also to provide an increase in the 16.9 percent rate somewhat comparable to the increase in the 23.2 percent rate for commercial banks. For this reason, the percentage deduction for additions to bad-debt reserves is being reduced from 60 percent to 30 percent, but this reduction is to take effect over a 10-year period. This percentage reduction in the formula will raise the effective rate of tax for these institutions, but will still leave some margin of tax advantage for them over commercial banks, which should preserve the inducement for them to continue investing in real estate mortgages.

H.R.Rep. No. 91-413, U.S.Code Cong. & Admin.News 1969, at 1775, 1969-3 C.B. at 278. The Senate, for its part, proposed a reduction of the applicable percentage of taxable income from 60 percent to 50 percent over a four-year period, S.Rep. No. 91-552, 1969-3 C.B. 423, 526, and Congress finally settled on a gradual reduction of the applicable percentage income to 40 percent over a period of ten years.

In 1986 Congress again revisited section 593 and further reduced the deduction in question to only 8 percent of taxable income. Tax Reform Act of 1986, Pub.L. No. 99-514, § 901(b)(2)(A), 100 Stat. 2085. This last curtailment of the “percentage of taxable income” method was made because regulatory changes had placed other financial institutions in direct competition with mutual savings institutions which, by virtue of the deductions available to them in section 593, had significant tax advantages. Congress decided that regulatory policies “are not promoted by providing a substantially lower effective tax rate for one competitor than for others,” H.R.Rep. No. 99-426, 99th Cong., 1st Sess. 581 (1985), and accordingly cut the deduction from 40 percent to 8 percent.

Purpose

As has been pointed out, the Tax Court found section 593 to be ambiguous with respect to the precise question in this case, and, therefore, it looked to the origin and purpose of the statutes involved to seek out Congress’ intent. It held that the new regulations did not harmonize with congressional intent, first, because the effect of the new regulation was to curtail the deduction for additions to loan loss reserves to a degree unintended by the Congress which had already curtailed the deduction in its 1969 amendments to section 593. Second, the Tax Court found that the new regulations violated congressional intent because they were at odds with Congress’ desire to ameliorate the effects of the 1969 curtailment by granting institutions the more generous ten-year carryback period for net operating losses under section 172.

Having decided what it thought Congress meant to do when it amended sections 593 and 172 in 1969, the Tax Court went on to reason as follows:

When Congress extended the NOL carry-back period from three to ten years in 1969, it was well-established that NOL carrybacks had no effect on deductions for additions to bad debt reserve for carryback years. The regulations so provided and there was no contrary authority specifically addressing the interplay between NOL carrybacks and the sections 166 and 593 deduction. Congress amended both (1) the bad debt reserve deduction provisions and (2) the net operating loss carryback provisions in order to achieve a limited increase in the income tax levels of mutual institutions. In order to determine the effect of proposed changes and, if necessary, to modify the changes, Congress must have examined the then-effective regulations and concluded that those regulations correctly reflected Congress’ intent as to how NOL carrybacks were to affect deductions for additions to bad debt reserve.

Pacific First Fed. Sav. Bank v. Commissioner, 94 T.C. 101, 112 (1990) (emphasis added). The core of the Tax Court’s holding, therefore, is that Congress “must have examined” the old regulations in effect in 1969 and must have approved them, though silently, when it amended sections 172 and 593 so as to reduce the loan loss reserves deduction while increasing the net operating loss carryback period.

The Tax Court rejected the Commissioner’s contention that the 1969 congressional amendments to section 593(b)(2)(E) were designed to comprise an exclusive list of things to be disregarded in arriving at the concept of taxable income for purposes of the deduction. The Commissioner had argued, as he does in this court, that section 593(b)(2)(E) is a tailored definition of “taxable income” for section 593 purposes and that conspicuously absent from it is a provision directing the disregarding of net operating loss carrybacks in the determination of “taxable income.” There was thus a “disregard” list set up in the very statute itself, but net operating losses, although they would have fit naturally and comfortably into that list, were not included in it. Therefore, the Commissioner argued, Congress must have intended that net operating losses not be disregarded in calculating “taxable income." The Tax Court, however, found that Congress never expressed any intention that its list of modifications to “taxable income” was to be viewed as an exclusive list, and the court felt that, had there been such an intent, there would be evidence of it in the legislative history. Chevron’s Rule of Deference

The parties agree that Congress has never directly addressed the precise issue in this case, that is, the order in which the deductions (1) for a net operating loss carryback under section 172 and (2) for an addition to loan loss reserves using the “percentage of taxable income” method under section 593 must be taken. Brief of Appellant at 9; Brief of Appellee at 8. The Tax Court found the statutes to be ambiguous on this question and turned to a review of a lengthy legislative history in the hope of glimpsing congressional intent. Because the parties and the Tax Court agree that Congress has not directly spoken to the precise question at issue, the rule of Chevron U.S.A., Inc., v. Natural Resources Defense Council, 467 U.S. 837, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984), should be applied to this case. 6

In Chevron, an environmental law case, the question at issue was the agency’s definition of “stationary source.” The court of appeals had found that the precise issue was not squarely addressed by the relevant statutes or their legislative histories, and it had turned to a consideration of the “purposes” of the congressional program to guide its decision. Holding this to be error, the Supreme Court stated the correct rule to be as follows:

When a court reviews an agency’s construction of the statute which it administers, it is confronted with two questions. First, always, is the question whether Congress has directly spoken to the precise question at issue. If the intent of Congress is clear, that is the end of the matter; for the court, as well as the agency, must give effect to the unambiguously expressed intent of Congress. If, however, the court determines Congress has not directly addressed the precise question at issue, the court does not simply impose its own construction on the statute, as would be necessary in the absence of an administrative interpretation. Rather, if the statute is silent or ambiguous with respect to the specific issue, the question for the court is whether the agency’s answer is based on a permissible construction of the statute.

Id., 467 U.S. at 842-43, 104 S.Ct. at 2781-82 (footnotes omitted) (emphasis added). The Court went on to say that a reviewing court

need not conclude that the agency construction was the only one it permissibly could have adopted to uphold the construction, or even the reading the court would have reached if the question initially had arisen in a judicial proceeding.

Id., 467 U.S. at 843 n. 11, 104 S.Ct. at 2782 n. 11. Thus, there may be several permissible constructions. If there are gaps left by silence or ambiguity of the statutes in question, agencies may fill the gaps with necessary rules, providing they are reasonable, and courts should not interfere with this process.

“The power of an administrative agency to administer a congressionally created ... program necessarily requires the formulation of policy and the making of rules to fill any gap left, implicitly or explicitly, by Congress.” Morton v. Ruiz, 415 U.S. 199, 231, 94 S.Ct. 1055, 1072, 39 L.Ed.2d 270 (1974). If Congress has explicitly left a gap for the agency to fill, there is an express delegation of authority to the agency to elucidate a specific provision of the statute by regulation. Such legislative regulations are given controlling weight unless they are arbitrary, capricious, or manifestly contrary to the statute. Sometimes the legislative delegation to an agency on a particular question is implicit rather than explicit. In such a case, a court may not substitute its own construction of a statutory provision for a reasonable interpretation made by the administrator of an agency.

Id., 467 U.S. at 843-44, 104 S.Ct. at 2781-83 (footnotes omitted) (emphasis added).

The regulations in question, both old and new, were issued by the Commissioner pursuant to the authority of 26 U.S.C. § 7805(a), which authorizes the issuance of “all needful rules and regulations for the enforcement of this title, including all rules and regulations as may be necessary by reason of any alteration of law in relation to internal revenue.” The regulations in this case were issued to fill a gap left by the statutes on the question of the ordering of these deductions, but there was no “express delegation of authority to the agency to elucidate a specific provision of the statute[s]_” Chevron, 467 U.S. at 843-44, 104 S.Ct. at 2781-83. Rather, the situation in this case is one in which the agency’s authority to fill this particular gap by regulation is implicit rather than explicit. Under these circumstances, a court may not substitute its own construction for the reasonable interpretation of an agency.

The Re-enactment Doctrine

In Pacific First Federal, the Tax Court found the statutes in question to be “not free of ambiguity,” and being thus unable to discern congressional intent from the faces of the statutes, it turned accordingly to an investigation of the origins and purposes of those statutes. Based on that investigation, the Tax Court concluded that Congress’ intent could be inferred from the 1969 amendments of sections 172 and 593, both made while the old regulation was indisputably in effect. Although the Tax Court recognized that Congress had never so much as mentioned the old regulation, it reasoned that Congress must have considered it and found it acceptable because Congress gave no contrary indication, as might have been expected had it disapproved of the old regulation.

This reasoning is a variant of the reenactment doctrine 7 whose fullest, most recent expression is found in NLRB v. Bell Aerospace Co., 416 U.S. 267, 94 S.Ct. 1757, 40 L.Ed.2d 134 (1974). In that case, the Court had to determine whether Congress, during the re-enactment of a labor statute, intended to adopt the National Labor Relations Board’s interpretation of the term “managerial employee.” The Board, in a series of decisions over two decades, had consistently interpreted the Taft-Hartley Act of 1947 to exclude “managerial employees.” When the Board certified a group of Bell Aerospace buyers as a bargaining unit, the company objected on the grounds that they were “managerial employees.” The Supreme Court noted the long series of previous adjudications in which the Board had taken a contrary position and further noted that Congress had amended the Act in 1959 without altering any of the Board’s earlier adjudicatory interpretations of the term “managerial employees.” The Supreme Court found that, among other things, “congressional failure to revise or repeal the agency’s interpretation is persuasive evidence that the interpretation is the one intended by Congress.” Id. at 275, 94 S.Ct. at 1762. In the decision itself, however, the “re-enactment doctrine” played a very minor role in the Court’s construction of the statute, because the legislative history of the Taft-Hartley Act was quite clear- in stating congressional preferences and referring directly and specifically to the Board’s well-known treatment of managerial employees as outside the coverage of the Act. The Court quoted from the House Managers’ statement in explanation of the Conference Committee Report as follows:

In the case of persons working in labor relations, personnel and employment departments, it was not thought necessary to make specific provision, as was done in the House bill, since the Board has treated, and presumably will continue to treat, such persons as outside the scope of the act. This is the prevailing Board practice with respect to such people as confidential secretaries as well, and it was not the intention of the conferees to alter this practice in any respect.

Id. at 282, 94 S.Ct. at 1765 (emphasis added). The Court went on to conclude that

[t]he legislative history strongly suggests that there also were other employees, much higher in the managerial structure, who were likewise regarded as so clearly outside the Act that no specific exclusionary provision was thought necessary.

Id. at 283, 94 S.Ct. at 1766 (emphasis added). The Supreme Court thus found that “the inference is plain that ‘managerial employees’ ” were to be excluded from the coverage of the Act. Id. at 284, 94 S.Ct. at 1767. While the Supreme Court noted the utility of the re-enactment doctrine as an interpretive tool, its decision is based primarily on its ability to read directly the clear record of legislative intent concerning the precise issue in the case. That is what distinguishes Bell Aerospace from the present case, where there is no reference by Congress in any of the legislative history to the Commissioner’s regulations, old or new.

Several considerations weaken the force of any arguments based on the reenactment doctrine. First, as has been pointed out, there is no evidence in the legislative history of the statutes in question that Congress ever had a specific or particular intent with respect to the ordering rule to be applied in this case. Second, in its various re-enactments and amendments of the statutes involved, Congress has never noticed the existence of, and has never referred to, the Commissioner’s old or new regulations. This is noteworthy because an important consideration in the application of the re-enactment doctrine is “the degree of scrutiny Congress has devoted to the regulation during subsequent re-enactments of the statute.” National Muffler Dealers Ass’n v. United States, 440 U.S. 472, 477, 99 S.Ct. 1304, 1307, 59 L.Ed.2d 519 (1979). The apparent degree of scrutiny in this case is zero. Finally, the legislative history in Bell Aerospace shows that Congress did not alter the statute so as to cast any doubt on its clearly expressed original intentions that the statute exclude managerial employees from its coverage. By contrast, Congress has steadily curtailed the deduction for additions to loan loss reserves under section 593 from the time it was codified in 1954, when a deduction of 100 percent of taxable income was allowed, to 1986, when the deduction dwindled to 8 percent of taxable income. Unlike the static situation in Bell Aerospace, the history of the statutes in this case shows that Congress has changed its policy with respect to this deduction on several occasions and always in the same direction: curtailment.

Congress’ pattern of changes in this deduction suggests that the agency charged with filling the gaps left by Congress during this process should have the freedom to alter its interpretations in the light of its experience and its perception of Congressional movement. This concept is well-recognized in the courts. For example, in National Muffler Dealers, the Commissioner interpreted the term “business league” in one way from 1919 to 1929, then changed directions 180 degrees and gave the term an opposite meaning from 1929 forward. The Court observed that

[n]othing in the regulations or case law directly explains the regulatory shift. We do know, however, that the change in 1929 incorporated an interpretation thought necessary to match the statute’s construction to the original congressional intent. We would be reluctant to adopt the rigid view that an agency may not alter its interpretation in light of administrative experience.

440 U.S. at 485, 99 S.Ct. at 1311 (citation and footnote omitted). The Court went on to find that “[t]he choice among reasonable interpretations is for the Commissioner, not the courts.” Id. at 488, 99 S.Ct. at 1312.

In Chevron, too, the Supreme Court noted “that the agency has from time to time changed its interpretation of the term ‘source’ ” and went on to state:

An initial agency interpretation is not instantly carved in stone. On the contrary, the agency, to engage in informed rulemaking, must consider varying interpretations and the wisdom of its policy on a continuing basis. Moreover, the fact that the agency has adopted different definitions in different contexts adds force to the argument that the definition itself is flexible, particularly since Congress has never indicated any disapproval of a flexible reading of the statute.

Chevron, 467 U.S. at 863-64, 104 S.Ct. at 2791-92.

The re-enactment doctrine, if rigidly applied, becomes a trap for the agencies. In tax situations, particularly, the misuse of this doctrine may

drastically curtail the scope and materially impair the flexibility of administrative actions; it would produce a most awkward situation. Outstanding regulations which had survived one Act could be changed only after a pre-view by the Congress. In preparation for a new revenue Act the Commissioner would have to prepare in advance new regulations covering old provisions. Their effectiveness would have to await Congressional approval of the new Act. The effect of such procedure, so far as time is concerned, would be precisely the same as if these new regulations were submitted to the Congress for approval. Such dilution of administrative powers would deprive the administrative process of some of its most valuable qualities — ease of adjustment to change, flexibility in light of experience, swiftness in meeting new or emergency situations. It would make the administrative process under these circumstances cumbersome and slow. Known inequities in existing regulations would have to await the advent of a new revenue act. Paralysis in effort to keep abreast of changes in business practices and new conditions would redound at times to the detriment of the revenue; at times to the disadvantage of the taxpayer. Likewise the result would be to read into the grant of express administrative powers an implied condition that they were not to be exercised unless, in effect, the Congress had consented.

Helvering v. Wilshire Oil Co., 308 U.S. 90, 101, 60 S.Ct. 18, 24, 84 L.Ed. 101 (1939). The re-enactment doctrine is merely an interpretive tool fashioned by the courts for their own use in construing ambiguous legislation. It is most useful in situations where there is some indication that Congress noted or considered the regulations in effect at the time of its action. Otherwise, the doctrine may be as doubtful as the silence of the statutes and legislative history to which it is applied. 8

Reversal of Agency Policy — Explanation

Appellees argue that an agency changing policy and direction may be required to explain its shift so that a reviewing court can determine that the change is deliberate and not arbitrary or capricious. Motor Vehicle Mfr. Ass’n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 41-42, 103 S.Ct. 2856, 77 L.Ed.2d 443 (1983). In this case, the Commissioner’s explanation for the issuance of his new regulations in 1978 is found in a memorandum written for the Secretary of the Treasury by his Acting Assistant Secretary for Tax Policy in 1978. 9 The reasons given for the proposed change were:

*As a legal and policy matter, it is clear that taxable income should be computed as it ordinarily is under the Code. The substantive rule in the regulation has been accepted for some time as the correct rule.

**The percentage of taxable income method is in substance a technique to lower the tax rate on thrift institutions.

* “Ignoring certain classes of deductions would have the effect of pyramiding benefits in an inappropriate and unintended manner.

“Publication of the final regulation was, we understand, deferred by former Secretary Simon at the urging of the industry, because of the poor economic condition of the thrift institutions in the early 1970’s.

“The industry has been aware of the new taxable income rule since 1971. There is no reason to defer the effective date of this rule beyond the date of publication of the proposed regulation.

The Acting Assistant Secretary went on to characterize the old regulation as “patently wrong.” The Commissioner also apparently believed that Congress’ 1969 additions to the list of exclusions in section 593(b)(2)(E) made that list into an exclusive one. Because the old regulation, then in effect, provided that net operating losses were excluded from the computations designed to establish taxable income for purposes of the section 593 deduction, the Commissioner believed that Congress tacitly disapproved the old regulation by not including its provisions in section 593(b)(2)(E) where they would so perfectly have fit. 43 Fed.Reg. 21454 (1978). These explanations are conclusive evidence that the Commissioner’s change of policy was a deliberate one.

As to its reasonableness, the Supreme Court has recently considered the question of an agency’s reversal of policy and has stated:

This Court has rejected the argument that an agency’s interpretation “is not entitled to deference because it represents a sharp break with prior interpretations” of the statute in question. In Chevron, we held that a revised interpretation deserves deference because “[a]n initial agency interpretation is not instantly carved in stone” and “the agency, to engage in informed rulemaking, must consider varying interpretations and the wisdom of its policy on a continuing basis.” An agency is not required to “ ‘establish rules of conduct to last forever,’ ” but rather “must be given ample latitude to ‘adapt [its] rules and policies to the demands of changing circumstances.’ ”

Rust v. Sullivan, — U.S.-, 111 S.Ct. 1759, 1769, 114 L.Ed.2d 233 (1991) (citations omitted) (emphasis added).

In Rust, the Court held that the Secretary of Health and Human Services “amply justified his change of interpretation” when he explained that (a) “prior policy failed to implement properly the statute,” (b) “the new regulations [were] more in keeping with the original intent of the stat ute...and (c) that the new regulations were “supported by a shift in attitude against the ‘elimination of unborn children by abortion.’ ” Id. The Commissioner’s explanation in this case is very similar to the Secretary’s in Rust. The Commissioner believed that the old regulations were “patently wrong” because they required an extraordinary method of computing “taxable income” not found elsewhere in the Internal Revenue Code, that the result of the old regulations was an unintended pyramiding of deductions, and that, by inference from Congress’ repeated curtailment of the section 593 deduction, there had been a “shift in attitude” away from special tax protections for mutual savings institutions. These explanations seem as reasonable as those in Rust, and, therefore, the Commissioner should be found to have made the explanation necessary under Motor Vehicle Manufacturers.

The Supreme Court also used Rust to emphasize Chevron’s holding on the degree of deference to be accorded agency interpretations in circumstances such as are found in this case.

While the petitioners’ interpretation of the legislative history may be a permissible one, it is by no means the only one, and it is certainly not the one found by the Secretary. It is well established that legislative history which does not demonstrate a clear and certain congressional intent cannot form the basis for enjoining the regulations.

Rust, 111 S.Ct. at 1770 (emphasis added). The Court concluded that because “the plain language and legislative history are ambiguous as to Congress’ intent in enacting Title X, we must defer to the Secretary’s permissible construction of the statute.” Id. at 1769.

Against the background of the silence of the statutes and their legislative history in the present case, and in view of the doubtful inferences that can be drawn from the application of the re-enactment doctrine in a situation in which Congress has never even acknowledged the existence of any of the regulations in question, the Tax Court’s conclusion that the new regulations are invalid as contrary to the intent of Congress seems strained. That conclusion may be a permissible one, but so is the Commissioner’s, whose view is reinforced by the fact that, throughout the decades of legislative history involved here, Congress continually tightened the reins on the section 593 deduction in question. It was reasonable and permissible for the Commissioner to follow that trend by correcting what he perceived to be a mistake. The fact that other reasonable or permissible interpretations of the statutory scheme exist is immaterial.

III.

The Tax Court, employing the “harmony” standard found in National Muffler Dealers, engaged in a plenary review of the legislative history of the statutory scheme without granting the Commissioner the degree of deference required by Chevron and Rust. 10 We agree with the Tax Court that section 593 is “not free of ambiguity,” but we believe that in doubtful situations we must take care not to substitute our own construction of a statutory scheme for a reasonable interpretation made by the agency.

In summary, we conclude that the Tax Court used the wrong standard to decide this case. Entirely ignoring Chevron (Rust had not been decided yet), the Tax Court employed the standard in National Muffler Dealers which seemed to allow a plenary review of the legislative history without the deference requirements found in Chevron and Rust. Because the Tax Court found the plain language of the statutes involved to be ambiguous, it apparently felt compelled to rely on a variant of the re-enactment doctrine, a doubtful application because Congress had never mentioned either regulation in question since the first was issued in 1964. Under these circumstances, we hold that the rule of decision for this case is supplied by Chevron and Rust, which command a greater degree of deference to administrative inter pretations of congressional legislation where Congress has not addressed the specific point in issue.

In light of Congress’ growing constriction of the “percentage of taxable income” method deduction in section 593, the Commissioner’s new regulation is reasonable. His desire to correct a previous ruling later determined to be wrong is not unreasonable, and the freedom of agencies to correct mistakes is a matter of recognized importance in the cases. Finally, after several decades of Congress’ complete silence and inaction regarding either of the regulations, it would be reasonable to infer a congressional purpose to leave the ordering rules in question in this case to the Commissioner as a matter of policy. Under such a grant of authority, the Commissioner’s deliberate action to correct what he perceived to be a previous mistake cannot be considered unreasonable.

Accordingly, the decision of the Tax Court is REVERSED, and this case is REMANDED for entry of judgment consistent with this court’s opinion.

1 . All section references in this opinion will be to the Internal Revenue Code of 1954 (26 U.S.C.) unless otherwise indicated.

2 . The other two methods of computing such deductions involve taking a specified percentage of outstanding loans as a deduction or basing the deduction on actual loan loss experience. See section 593(b)(3) — (4).

3 . 26 C.F.R. § 1.593-6A(b)(5)(vi) provides the same rule, though in different language, for certain other years material to this case.

4 . This court is, in substance, reviewing the Tax Court’s decision in Pacific First Federal, and references to the Tax Court’s findings, holdings or conclusions are to its opinion in Pacific First Federal unless otherwise indicated.

5 . Congress also added two other methods for computing a deduction for loan loss reserves, one based on a percentage of outstanding real estate loans, the other based on actual loss experience. Neither of these alternative methods was used in this case.

6 . The Tax Court did not mention Chevron in Pacific First Federal, and the parties have not cited it in their briefs.

7 . In a footnote to its opinion, the Tax Court stated that it did “not address the impact of the [re-enactment] doctrine on Treasury's authority to promulgate the challenged [regulations]." Pacific First Federal, 94 T.C. 101, 114 n. 5. This may be true as to the doctrine’s older and most rigid form which treats regulations extant during statutory re-enactment as having been promoted nearly to the status of statutes themselves. See, e.g., Helvering v. R.J. Reynolds Tobacco Co., 306 U.S. 110, 116, 59 S.Ct. 423, 426, 83 L.Ed. 536 (1939) ("force of law”). It is clear, however, that the Tax Court used the doctrine as an interpretive tool.

8 . Judge Learned Hand said, "[t]o suppose that Congress must particularly correct each mistaken construction under penalty of incorporating it into the fabric of the statute appears to us unwarranted; our fiscal legislation is detailed and specific enough already.” F.W. Woolworth Co. v. United States, 91 F.2d 973, 976 (2d Cir.), cert. denied, 302 U.S. 768, 58 S.Ct. 479, 82 L.Ed. 597 (1937).

9 . See Exhibit 130DZ in Pacific First Federal.

10 . Rust had not been decided when the Tax Court issued its opinion in this case.