439 U.S. 522 (1979)
THOR POWER TOOL CO.
COMMISSIONER OF INTERNAL REVENUE.
Supreme Court of United States.
Argued November 1, 1978.
Decided January 16, 1979.
CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE SEVENTH CIRCUIT.
Mark H. Berens argued the cause for petitioner. With him on the briefs were Lee N. Abrams and Douglas A. Poe.
Stuart A. Smith argued the cause for respondent. With him on the brief were Solicitor General McCree, Assistant Attorney General Ferguson, and Ann Belanger Durney.[*]
Eric Neisser filed a brief for the Taxation With Representation Fund et al. as amici curiae urging affirmance.
MR. JUSTICE BLACKMUN delivered the opinion of the Court.
This case, as it comes to us, presents two federal income tax issues. One has to do with inventory accounting. The other relates to a bad-debt reserve.
The Inventory Issue. In 1964, petitioner Thor Power Tool Co. (hereinafter sometimes referred to as the taxpayer), in accord with “generally accepted accounting principles,” wrote down what it regarded as excess inventory to Thor’s own estimate of the net realizable value of the excess goods. Despite this write-down, Thor continued to hold the goods for sale at original prices. It offset the write-down against 1964 sales and thereby produced a net operating loss for that year; it then asserted that loss as a carryback to 1963 under § 172 of the Internal Revenue Code of 1954, 26 U. S. C. § 172. The Commissioner of Internal Revenue, maintaining that the write-down did not serve to reflect income clearly for tax purposes, disallowed the offset and the carryback.
The Bad-Debt Issue. In 1965, the taxpayer added to its reserve for bad debts and asserted as a deduction, under § 166 (c) of the Code, 26 U. S. C. § 166 (c), a sum that presupposed a substantially higher charge-off rate than Thor had experienced in immediately preceding years. The Commissioner ruled that the addition was excessive, and determined, pursuant to a formula based on the taxpayer’s past experience, what he regarded as a lesser but “reasonable” amount to be added to Thor’s reserve.
On the taxpayer’s petition for redetermination, the Tax Court, in an unreviewed decision by Judge Goffe, upheld the Commissioner’s exercise of discretion in both respects. 64 T. C. 154 (1975). As a consequence, and also because of other adjustments not at issue here, the court redetermined, App. 264, the following deficiencies in Thor’s federal income tax:
calendar year 1963—$494,055.99 calendar year 1965—$59,287.48
The United States Court of Appeals for the Seventh Circuit affirmed. 563 F. 2d 861 (1977). We granted certiorari, 435 U. S. 914 (1978), to consider these important and recurring income tax accounting issues.
The Inventory Issue
Taxpayer is a Delaware corporation with principal place of business in Illinois. It manufactures hand-held power tools, parts and accessories, and rubber products. At its various plants and service branches, Thor maintains inventories of raw materials, work-in-process, finished parts and accessories, and completed tools. At all times relevant, Thor has used, both for financial accounting and for income tax purposes, the “lower of cost or market” method of valuing inventories. App. 23-24. See Treas. Reg. § 1.471-2 (c), 26 CFR § 1.471-2 (c) (1978).
Thor’s tools typically contain from 50 to 200 parts, each of which taxpayer stocks to meet demand for replacements. Because of the difficulty, at the time of manufacture, of predicting the future demand for various parts, taxpayer produced liberal quantities of each part to avoid subsequent production runs. Additional runs entail costly retooling and result in delays in filling orders. App. 54-55.
In 1960, Thor instituted a procedure for writing down the inventory value of replacement parts and accessories for tool models it no longer produced. It created an inventory contra-account and credited that account with 10% of each part’s cost for each year since production of the parent model had ceased. 64 T. C., at 156-157; App. 24. The effect of the procedure was to amortize the cost of these parts over a 10-year period. For the first nine months of 1964, this produced a write-down of $22,090. 64 T. C., at 157; App. 24.
In late 1964, new management took control and promptly concluded that Thor’s inventory in general was overvalued. After “a physical inventory taken at all locations” of the tool and rubber divisions, id., at 52, management wrote off approximately $2.75 million of obsolete parts, damaged or defective tools, demonstration or sales samples, and similar items. Id., at 52-53. The Commissioner allowed this writeoff because Thor scrapped most of the articles shortly after their removal from the 1964 closing inventory. Management also wrote down $245,000 of parts stocked for three unsuccessful products. Id., at 56. The Commissioner allowed this write-down, too, since Thor sold these items at reduced prices shortly after the close of 1964. Id., at 62.
This left some 44,000 assorted items, the status of which is the inventory issue here. Management concluded that many of these articles, mostly spare parts, were “excess” inventory, that is, that they were held in excess of any reasonably foreseeable future demand. It was decided that this inventory should be written down to its “net realizable value,” which, in most cases, was scrap value. 64 T. C., at 160-161; Brief for Petitioner 9; Tr. of Oral Arg. 11.
Two methods were used to ascertain the quantity of excess inventory. Where accurate data were available, Thor forecast future demand for each item on the basis of actual 1964 usage, that is, actual sales for tools and service parts, and actual usage for raw materials, work-in-process, and production parts. Management assumed that future demand for each item would be the same as it was in 1964. Thor then applied the following aging schedule: the quantity of each item corresponding to less than one year’s estimated demand was kept at cost; the quantity of each item in excess of two years’ estimated demand was written off entirely; and the quantity of each item corresponding to from one to two years’ estimated demand was written down by 50% or 75%. App. 26. Thor presented no statistical evidence to rationalize these percentages or this time frame. In the Tax Court, Thor’s president justified the formula by citing general business experience, and opined that it was “somewhat in between” possible alternative solutions. This first method yielded a total write-down of $744,030. 64 T. C., at 160.
At two plants where 1964 data were inadequate to permit forecasts of future demand, Thor used its second method for valuing inventories. At these plants, the company employed flat percentage write-downs of 5%, 10%, and 50% for various types of inventory. Thor presented no sales or other data to support these percentages. Its president observed that “this is not a precise way of doing it,” but said that the company “felt some adjustment of this nature was in order, and these figures represented our best estimate of what was required to reduce the inventory to net realizable value.” App. 67. This second method yielded a total write-down of $160,832. 64 T. C., at 160.
Although Thor wrote down all its “excess” inventory at once, it did not immediately scrap the articles or sell them at reduced prices, as it had done with the $3 million of obsolete and damaged inventory, the write-down of which the Commissioner permitted. Rather, Thor retained the “excess” items physically in inventory and continued to sell them at original prices. Id., at 160-161. The company found that, owing to the peculiar nature of the articles involved, price reductions were of no avail in moving this “excess” inventory. As time went on however, Thor gradually disposed of some of these items as scrap; the record is unclear as to when these dispositions took place.
Thor’s total write-down of “excess” inventory in 1964 therefore was:
Ten-year amortization of parts for
discontinued tools $22,090
First method (aging formula based
on 1964 usage) 744,030
Second method (flat percentage
Thor credited this sum to its inventory contra-account, thereby decreasing closing inventory, increasing cost of goods sold, and decreasing taxable income for the year by that amount. The company contended that, by writing down excess inventory to scrap value, and by thus carrying all inventory at “net realizable value,” it had reduced its inventory to “market” in accord with its “lower of cost or market” method of accounting. On audit, the Commissioner disallowed the write-down in its entirety, asserting that it did not serve clearly to reflect Thor’s 1964 income for tax purposes.
The Tax Court, in upholding the Commissioner’s determination, found as a fact that Thor’s write-down of excess inventory did conform to “generally accepted accounting principles”; indeed, the court was “thoroughly convinced . . . that such was the case.” Id., at 165. The court found that if Thor had failed to write down its inventory on some reasonable basis, its accountants would have been unable to give its financial statements the desired certification. Id., at 161-162. The court held, however, that conformance with “generally accepted accounting principles” is not enough; § 446 (b), and § 471 as well, of the 1954 Code, 26 U. S. C. §§ 446 (b) and 471, prescribe, as an independent requirement, that inventory accounting methods must “clearly reflect income.” The Tax Court rejected Thor’s argument that its write-down of “excess” inventory was authorized by Treasury Regulations, 64 T. C., at 167-171, and held that the Commissioner had not abused his discretion in determining that the write-down failed to reflect 1964 income clearly.
Inventory accounting is governed by §§ 446 and 471 of the Code, 26 U. S. C. §§ 446 and 471. Section 446 (a) states the general rule for methods of accounting: “Taxable income shall be computed under the method of accounting on the basis of which the taxpayer regularly computes his income in keeping his books.” Section 446 (b) provides, however, that if the method used by the taxpayer “does not clearly reflect income, the computation of taxable income shall be made under such method as, in the opinion of the [Commissioner], does clearly reflect income.” Regulations promulgated under § 446, and in effect for the taxable year 1964, state that “no method of accounting is acceptable unless, in the opinion of the Commissioner, it clearly reflects income.” Treas. Reg. § 1.446-1 (a) (2), 26 CFR § 1.446-1 (a) (2) (1964).
Section 471 prescribes the general rule for inventories. It states:
“Whenever in the opinion of the [Commissioner] the use of inventories is necessary in order clearly to determine the income of any taxpayer, inventory shall be taken by such taxpayer on such basis as the [Commissioner] may prescribe as conforming as nearly as may be to the best accounting practice in the trade or business and as most clearly reflecting the income.”
As the Regulations point out, § 471 obviously establishes two distinct tests to which an inventory must conform. First, it must conform “as nearly as may be” to the “best accounting practice,” a phrase that is synonymous with “generally accepted accounting principles.” Second, it “must clearly reflect the income.” Treas. Reg. § 1.471-2 (a) (2), 26 CFR § 1.471-2 (a) (2) (1964).
It is obvious that on their face, §§ 446 and 471, with their accompanying Regulations, vest the Commissioner with wide discretion in determining whether a particular method of inventory accounting should be disallowed as not clearly reflective of income. This Court’s cases confirm the breadth of this discretion. In construing § 446 and its predecessors, the Court has held that “[t]he Commissioner has broad powers in determining whether accounting methods used by a taxpayer clearly reflect income.” Commissioner v. Hansen, 360 U. S. 446, 467 (1959). Since the Commissioner has “[m]uch latitude for discretion,” his interpretation of the statute’s clear-reflection standard “should not be interfered with unless clearly unlawful.” Lucas v. American Code Co., 280 U. S. 445, 449 (1930). To the same effect are United States v. Catto, 384 U. S. 102, 114 (1966); Schlude v. Commissioner, 372 U. S. 128, 133-134 (1963); American Automobile Assn. v. United States, 367 U. S. 687, 697-698 (1961); Automobile Club of Michigan v. Commissioner, 353 U. S. 180, 189-190 (1957); Brown v. Helvering, 291 U. S. 193, 203 (1934). In construing § 203 of the Revenue Act of 1918, 40 Stat. 1060, a predecessor of § 471, the Court held that the taxpayer bears a “heavy burden of [proof],” and that the Commissioner’s disallowance of an inventory accounting method is not to be set aside unless shown to be “plainly arbitrary.” Lucas v. Structural Steel Co., 281 U. S. 264, 271 (1930).
As has been noted, the Tax Court found as a fact in this case that Thor’s write-down of “excess” inventory conformed to “generally accepted accounting principles” and was “within the term, `best accounting practice,’ as that term is used in section 471 of the Code and the regulations promulgated under that section.” 64 T. C., at 161, 165. Since the Commissioner has not challenged this finding, there is no dispute that Thor satisfied the first part of § 471’s two-pronged test. The only question, then, is whether the Commissioner abused his discretion in determining that the write-down did not satisfy the test’s second prong in that it failed to reflect Thor’s 1964 income clearly. Although the Commissioner’s discretion is not unbridled and may not be arbitrary, we sustain his exercise of discretion here, for in this case the write-down was plainly inconsistent with the governing Regulations which the taxpayer, on its part, has not challenged.
It has been noted above that Thor at all pertinent times used the “lower of cost or market” method of inventory accounting. The rules governing this method are set out in Treas. Reg. § 1.471-4, 26 CFR § 1.471-4 (1964). That Regulation defines “market” to mean, ordinarily, “the current bid price prevailing at the date of the inventory for the particular merchandise in the volume in which usually purchased by the taxpayer.” § 1.471-4 (a). The courts have uniformly interpreted “bid price” to mean replacement cost, that is, the price the taxpayer would have to pay on the open market to purchase or reproduce the inventory items. Where no open market exists, the Regulations require the taxpayer to ascertain “bid price” by using “such evidence of a fair market price at the date or dates nearest the inventory as may be available, such as specific purchases or sales by the taxpayer or others in reasonable volume and made in good faith, or compensation paid for cancellation of contracts for purchase commitments.” § 1.471-4 (b).
The Regulations specify two situations in which a taxpayer is permitted to value inventory below “market” as so defined. The first is where the taxpayer in the normal course of business has actually offered merchandise for sale at prices lower than replacement cost. Inventories of such merchandise may be valued at those prices less direct cost of disposition, “and the correctness of such prices will be determined by reference to the actual sales of the taxpayer for a reasonable period before and after the date of the inventory.” Ibid. The Regulations warn that prices “which vary materially from the actual prices so ascertained will not be accepted as reflecting the market.” Ibid.
The second situation in which a taxpayer may value inventory below replacement cost is where the merchandise itself is defective. If goods are “unsalable at normal prices or unusable in the normal way because of damage, imperfections, shop wear, changes of style, odd or broken lots, or other similar causes,” the taxpayer is permitted to value the goods “at bona fide selling prices less direct cost of disposition.” § 1.471-2 (c). The Regulations define “bona fide selling price” to mean an “actual offering of goods during a period ending not later than 30 days after inventory date.” Ibid. The taxpayer bears the burden of proving that “such exceptional goods as are valued upon such selling basis come within the classifications indicated,” and is required to “maintain such records of the disposition of the goods as will enable a verification of the inventory to be made.” Ibid.
From this language, the regulatory scheme is clear. The taxpayer must value inventory for tax purposes at cost unless the “market” is lower. “Market” is defined as “replacement cost,” and the taxpayer is permitted to depart from replacement cost only in specified situations. When it makes any such departure, the taxpayer must substantiate its lower inventory valuation by providing evidence of actual offerings, actual sales, or actual contract cancellations. In the absence of objective evidence of this kind, a taxpayer’s assertions as to the “market value” of its inventory are not cognizable in computing its income tax.
It is clear to us that Thor’s procedures for writing down the value of its “excess” inventory were inconsistent with this regulatory scheme. Although Thor conceded that “an active market prevailed” on the inventory date, see 64 T. C., at 169, it “made no effort to determine the purchase or reproduction cost” of its “excess” inventory. Id., at 162. Thor thus failed to ascertain “market” in accord with the general rule of the Regulations. In seeking to depart from replacement cost, Thor failed to bring itself within either of the authorized exceptions. Thor is not able to take advantage of § 1.471-4 (b) since, as the Tax Court found, the company failed to sell its excess inventory or offer it for sale at prices below replacement cost. 64 T. C., at 160-161. Indeed, Thor concedes that it continued to sell its “excess” inventory at original prices. Thor also is not able to take advantage of § 1.471-2 (c) since, as the Tax Court and the Court of Appeals both held, it failed to bear the burden of proving that its excess inventory came within the specified classifications. 64 T. C., at 171; 563 F. 2d, at 867. Actually, Thor’s “excess” inventory was normal and unexceptional, and was indistinguishable from and intermingled with the inventory that was not written down.
More importantly, Thor failed to provide any objective evidence whatever that the “excess” inventory had the “market value” management ascribed to it. The Regulations demand hard evidence of actual sales and further demand that records of actual dispositions be kept. The Tax Court found, however, that Thor made no sales and kept no records. 64 T. C., at 171. Thor’s management simply wrote down its closing inventory on the basis of a well-educated guess that some of it would never be sold. The formulae governing this write-down were derived from management’s collective “business experience”; the percentages contained in those formulae seemingly were chosen for no reason other than that they were multiples of five and embodied some kind of anagogical symmetry. The Regulations do not permit this kind of evidence. If a taxpayer could write down its inventories on the basis of management’s subjective estimates of the goods’ ultimate salability, the taxpayer would be able, as the Tax Court observed, id., at 170, “to determine how much tax it wanted to pay for a given year.”
For these reasons, we agree with the Tax Court and with the Seventh Circuit that the Commissioner acted within his discretion in deciding that Thor’s write-down of “excess” inventory failed to reflect income clearly. In the light of the well-known potential for tax avoidance that is inherent in inventory accounting, the Commissioner in his discretion may insist on a high evidentiary standard before allowing write-downs of inventory to “market.” Because Thor provided no objective evidence of the reduced market value of its “excess” inventory, its write-down was plainly inconsistent with the Regulations, and the Commissioner properly disallowed it.
The taxpayer’s major argument against this conclusion is based on the Tax Court’s clear finding that the write-down conformed to “generally accepted accounting principles.” Thor points to language in Treas. Reg. § 1.446-1 (a) (2), 26 CFR § 1.446-1 (a) (2) (1964), to the effect that “[a] method of accounting which reflects the consistent application of generally accepted accounting principles . . . will ordinarily be regarded as clearly reflecting income” (emphasis added). Section 1.471-2 (b), 26 CFR § 1.471-2 (b) (1964), of the Regulations likewise stated that an inventory taken in conformity with best accounting practice “can, as a general rule, be regarded as clearly reflecting . . . income” (emphasis added). These provisions, Thor contends, created a presumption that an inventory practice conformable to “generally accepted accounting principles” is valid for income tax purposes. Once a taxpayer has established this conformity, the argument runs, the burden shifts to the Commissioner affirmatively to demonstrate that the taxpayer’s method does not reflect income clearly. Unless the Commissioner can show that a generally accepted method “demonstrably distorts income,” Brief for Chamber of Commerce of the United States as Amicus Curiae 3, or that the taxpayer’s adoption of such method was “motivated by tax avoidance,” Brief for Petitioner 25, the presumption in the taxpayer’s favor will carry the day. The Commissioner, Thor concludes, failed to rebut that presumption here.
If the Code and Regulations did embody the presumption petitioner postulates, it would be of little use to the taxpayer in this case. As we have noted, Thor’s write-down of “excess” inventory was inconsistent with the Regulations; any general presumption obviously must yield in the face of such particular inconsistency. We believe, however, that no such presumption is present. Its existence is insupportable in light of the statute, the Court’s past decisions, and the differing objectives of tax and financial accounting.
First, as has been stated above, the Code and Regulations establish two distinct tests to which an inventory must conform. The Code and Regulations, moreover, leave little doubt as to which test is paramount. While § 471 of the Code requires only that an accounting practice conform “as nearly as may be” to best accounting practice, § 1.446-1 (a) (2) of the Regulations states categorically that “no method of accounting is acceptable unless, in the opinion of the Commissioner, it clearly reflects income” (emphasis added). Most importantly, the Code and Regulations give the Commissioner broad discretion to set aside the taxpayer’s method if, “in [his] opinion,” it does not reflect income clearly. This language is completely at odds with the notion of a “presumption” in the taxpayer’s favor. The Regulations embody no presumption; they say merely that, in most cases, generally accepted accounting practices will pass muster for tax purposes. And in most cases they will. But if the Commissioner, in the exercise of his discretion, determines that they do not, he may prescribe a different practice without having to rebut any presumption running against the Treasury.
Second, the presumption petitioner postulates finds no support in this Court’s prior decisions. It was early noted that the general rule specifying use of the taxpayer’s method of accounting “is expressly limited to cases where the Commissioner believes that the accounts clearly reflect the net income.” Lucas v. American Code Co., 280 U. S., at 449. More recently, it was held in American Automobile Assn. v. United States that a taxpayer must recognize prepaid income when received, even though this would mismatch expenses and revenues in contravention of “generally accepted commercial accounting principles.” 367 U. S., at 690. “[T]o say that in performing the function of business accounting the method employed by the Association `is in accord with generally accepted commercial accounting principles and practices,'” the Court concluded, “is not to hold that for income tax purposes it so clearly reflects income as to be binding on the Treasury.” Id., at 693. “[W]e are mindful that the characterization of a transaction for financial accounting purposes, on the one hand, and for tax purposes, on the other, need not necessarily be the same.” Frank Lyon Co. v. United States, 435 U. S. 561, 577 (1978). See Commissioner v. Idaho Power Co., 418 U. S. 1, 15 (1974). Indeed, the Court’s cases demonstrate that divergence between tax and financial accounting is especially common when a taxpayer seeks a current deduction for estimated future expenses or losses. E. g., Commissioner v. Hansen, 360 U. S. 446 (1959) (reserve to cover contingent liability in event of nonperformance of guarantee); Brown v. Helvering, 291 U. S. 193 (1934) (reserve to cover expected liability for unearned commissions on anticipated insurance policy cancellations); Lucas v. American Code Co., supra (reserve to cover expected liability on contested lawsuit). The rationale of these cases amply encompasses Thor’s aim. By its president’s concession, the company’s write-down of “excess” inventory was founded on the belief that many of the articles inevitably would become useless due to breakage, technological change, fluctuations in market demand, and the like. Thor, in other words, sought a current “deduction” for an estimated future loss. Under the decided cases, a taxpayer so circumstanced finds no shelter beneath an accountancy presumption.
Third, the presumption petitioner postulates is insupportable in light of the vastly different objectives that financial and tax accounting have. The primary goal of financial accounting is to provide useful information to management, shareholders, creditors, and others properly interested; the major responsibility of the accountant is to protect these parties from being misled. The primary goal of the income tax system, in contrast, is the equitable collection of revenue; the major responsibility of the Internal Revenue Service is to protect the public fisc. Consistently with its goals and responsibilities, financial accounting has as its foundation the principle of conservatism, with its corollary that “possible errors in measurement [should] be in the direction of understatement rather than overstatement of net income and net assets.” In view of the Treasury’s markedly different goals and responsibilities, understatement of income is not destined to be its guiding light. Given this diversity, even contrariety, of objectives, any presumptive equivalency between tax and financial accounting would be unacceptable.
This difference in objectives is mirrored in numerous differences of treatment. Where the tax law requires that a deduction be deferred until “all the events” have occurred that will make it fixed and certain, United States v. Anderson, 269 U. S. 422, 441 (1926), accounting principles typically require that a liability be accrued as soon as it can reasonably be estimated. Conversely, where the tax law requires that income be recognized currently under “claim of right,” “ability to pay,” and “control” rationales, accounting principles may defer accrual until a later year so that revenues and expenses may be better matched. Financial accounting, in short, is hospitable to estimates, probabilities, and reasonable certainties; the tax law, with its mandate to preserve the revenue, can give no quarter to uncertainty. This is as it should be. Reasonable estimates may be useful, even essential, in giving shareholders and creditors an accurate picture of a firm’s overall financial health; but the accountant’s conservatism cannot bind the Commissioner in his efforts to collect taxes. “Only a few reserves voluntarily established as a matter of conservative accounting,” Mr. Justice Brandeis wrote for the Court, “are authorized by the Revenue Acts.” Brown v. Helvering, 291 U. S., at 201-202.
Finally, a presumptive equivalency between tax and financial accounting would create insurmountable difficulties of tax administration. Accountants long have recognized that “generally accepted accounting principles” are far from being a canonical set of rules that will ensure identical accounting treatment of identical transactions. “Generally accepted accounting principles,” rather, tolerate a range of “reasonable” treatments, leaving the choice among alternatives to management. Such, indeed, is precisely the case here. Variances of this sort may be tolerable in financial reporting, but they are questionable in a tax system designed to ensure as far as possible that similarly situated taxpayers pay the same tax. If management’s election among “acceptable” options were dispositive for tax purposes, a firm, indeed, could decide unilaterally—within limits dictated only by its accountants—the tax it wished to pay. Such unilateral decisions would not just make the Code inequitable; they would make it unenforceable.
Thor complains that a decision adverse to it poses a dilemma. According to the taxpayer, it would be virtually impossible for it to offer objective evidence of its “excess” inventory’s lower value, since the goods cannot be sold at reduced prices; even if they could be sold, says Thor, their reduced-price sale would just “pull the rug out” from under the identical “non-excess” inventory Thor is trying to sell simultaneously. The only way Thor could establish the inventory’s value by a “closed transaction” would be to scrap the articles at once. Yet immediate scrapping would be undesirable, for demand for the parts ultimately might prove greater than anticipated. The taxpayer thus sees itself presented with “an unattractive Hobson’s choice: either the unsalable inventory must be carried for years at its cost instead of net realizable value, thereby overstating taxable income by such overvaluation until it is scrapped, or the excess inventory must be scrapped prematurely to the detriment of the manufacturer and its customers.” Brief for Petitioner 25.
If this is indeed the dilemma that confronts Thor, it is in reality the same choice that every taxpayer who has a paper loss must face. It can realize its loss now and garner its tax benefit, or it can defer realization, and its deduction, hoping for better luck later. Thor, quite simply, has suffered no present loss. It deliberately manufactured its “excess” spare parts because it judged that the marginal cost of unsalable inventory would be lower than the cost of retooling machinery should demand surpass expectations. This was a rational business judgment and, not unpredictably, Thor now has inventory it believes it cannot sell. Thor, of course, is not so confident of its prediction as to be willing to scrap the “excess” parts now; it wants to keep them on hand, just in case. This, too, is a rational judgment, but there is no reason why the Treasury should subsidize Thor’s hedging of its bets. There is also no reason why Thor should be entitled, for tax purposes, to have its cake and to eat it too.
The Bad-Debt Issue
Deductions for bad debts are covered by § 166 of the 1954 Code, 26 U. S. C. § 166. Section 166 (a) (1) sets forth the general rule that a deduction is allowed for “any debt which becomes worthless within the taxable year.” Alternatively, the Code permits an accrual-basis taxpayer to account for bad debts by the reserve method. This is implemented by § 166 (c), which states that “[i]n lieu of any deduction under subsection (a), there shall be allowed (in the discretion of the [Commissioner]) a deduction for a reasonable addition to a reserve for bad debts.” A “reasonable” addition is the amount necessary to bring the reserve balance up to the level that can be excepted to cover losses properly anticipated on debts outstanding at the end of the tax year.
At all times pertinent, Thor has used the reserve method. Its reserve at the beginning of 1965 was approximately $93,000. See 64 T. C., at 162. During 1965, Thor’s new management undertook a stringent review of accounts receivable. In the company’s rubber division, credit personnel studied all accounts; a 100% reserve was set up for two accounts deemed wholly uncollectible, and a 1% reserve was established for all other receivables. Ibid. In the tool division, credit clerks analyzed all accounts more than 90 days past due with balances over $100; a 100% reserve was established for accounts judged wholly uncollectible, and an identical collectibility ratio was applied to accounts under $100 of the same age. A flat 2% reserve was set up for accounts more than 30 days past due, and a 1% reserve for all other accounts. Id., at 162-163. These judgments, approved by three levels of management, indicated that $136,150 should be added to the bad-debt reserve, bringing its balance at year-end to a figure slightly below $229,000. Id., at 162. Thor claimed this $136,150 as a deduction under § 166 (c).
The Commissioner ruled that the deduction was excessive. He computed what he believed to be a “reasonable” addition to Thor’s reserve by using the “six-year moving average” formula derived from the decision in Black Motor Co. v. Commissioner, 41 B. T. A. 300 (1940), aff’d on other grounds, 125 F. 2d 977 (CA6 1942). This formula seeks to ascertain a “reasonable” addition to a bad-debt reserve in light of the taxpayer’s recent chargeoff history. In this case, the formula indicated that, for the years 1960-1965, Thor’s annual chargeoffs of bad debts amounted, on the average, to 3.128% of its year-end receivables. 64 T. C., at 163. Applying that percentage to Thor’s 1965 year-end receivables, the Commissioner determined that $154,156.80 of accounts receivable could reasonably be expected to default. The amount required to bring Thor’s reserve up to this level was $61,359.20, and the Commissioner decided that this was a “reasonable” addition. Accordingly, he disallowed the remaining $74,790.80 of Thor’s claimed § 166 (c) deduction. Both the Tax Court, 64 T. C., at 174-175, and the Seventh Circuit, 563 F. 2d, at 870, held that the Commissioner had not abused his discretion in so ruling.
Section 166 (c) states that a deduction for an addition to a bad-debt reserve is to be allowed “in the discretion” of the Commissioner. Consistently with this statutory language, the courts uniformly have held that the Commissioner’s determination of a “reasonable” (and hence deductible) addition must be sustained unless the taxpayer proves that the Commissioner abused his discretion. The taxpayer is said to bear a “heavy burden” in this respect. He must show not only that his own computation is reasonable but also that the Commissioner’s computation is unreasonable and arbitrary.
Since it first received the approval of the Tax Court in 1940, the Black Motor bad-debt formula has enjoyed the favor of all three branches of the Federal Government. The formula has been employed consistently by the Commissioner, approved by the courts, and collaterally recognized by the Congress. Thor faults the Black Motor formula because of its retrospectively: By ascertaining current additions to a reserve by reference to past chargeoff experience, the formula assertedly penalizes taxpayers who have delayed in making writeoffs in the past, or whose receivables have just recently begun to deteriorate. Petitioner’s objection is not altogether irrational, but it falls short of rendering the formula arbitrary. Common sense suggests that a firm’s recent credit experience offers a reasonable index of the credit problems it may suffer currently. And the formula possesses the not inconsiderable advantage of enhancing certainty and predictability in an area peculiarly susceptible of taxpayer abuse. In any event, after its 40 years of near-universal acceptance, we are not inclined to disturb the Black Motor formula now.
Granting that Black Motor in principle is valid, then, the only question is whether the Commissioner abused his discretion in invoking the formula in this case. Of course, there will be cases—indeed, the Commissioner has acknowledged that there are cases, see Rev. Rul. 76-362, 1976-2 Cum. Bull. 45, 46—in which the formula will generate an arbitrary result. If a taxpayer’s most recent bad-debt experience is unrepresentative for some reason, a formula using that experience as data cannot be expected to produce a “reasonable” addition for the current year. If the taxpayer suffers an extraordinary credit reversal (the bankruptcy of a major customer, for example), the “six-year moving average” formula will fail. In such a case, where the taxpayer can point to conditions that will cause future debt collections to be less likely than in the past, the taxpayer is entitled to—and the Commissioner is prepared to allow—an addition larger than Black Motor would call for. See Rev. Rul. 76-362, supra.
In this case, however, as the Tax Court found, Thor “did not show that conditions at the end of 1965 would cause collection of accounts receivable to be less likely than in prior years.” 64 T. C., at 175. Indeed, the Tax Court “infer[red] from the entire record that collectibility was probably more likely at the end of 1965 than it was [previously] because new management had been infused into petitioner” (emphasis added). Thor cited no changes in the conditions of business generally or of its customers specifically that would render the Black Motor formula unreliable; new management just came in and second-guessed its predecessor, taking a “tougher” approach. Management’s pessimism may not have been unreasonable, but the Commissioner had the discretion to take a more sanguine view.
For these reasons, we agree with the Tax Court and with the Court of Appeals that the Commissioner did not abuse his discretion in recomputing a “reasonable” addition to Thor’s bad-debt reserve according to the Black Motor formula. Thor failed to carry its “heavy burden” of showing why the application of that formula would have been arbitrary in this case.
The judgment of the Court of Appeals is affirmed.
It is so ordered.
[*] Briefs of amici curiae urging reversal were filed by Donald E. Egan, Francis X. Grossi, Jr., Robert S. Connors, Laurence B. Kraus, and Stanley T. Kaleczyc, Jr., for the Chamber of Commerce of the United States; and by Crane C. Hauser, Arthur I. Gould, Richard D. Godown, and John Lucas for the National Association of Manufacturers of the United States.
 In August 1964, Stewart-Warner Corp., Thor’s principal shareholder (owning approximately 20% of petitioner’s outstanding common shares), agreed with Thor to purchase substantially all of Thor’s assets. Its ensuing examination and audit led Stewart-Warner to conclude that petitioner’s assets were substantially overstated and its liabilities understated. The purchase agreement then was rescinded and Stewart-Warner agreed, instead, to provide management assistance to Thor.
 Both in his brief, Brief for Respondent 6, 17, 30-31, and at oral argument, Tr. of Oral Arg. 24-25, the Commissioner has maintained that the reason for the allowance of Thor’s $2.75 million writeoff was that the items were scrapped soon after they were written off. The Court of Appeals accepted this explanation. 563 F. 2d 861, 864 (1977). Thor challenges its factual predicate, and asserts that 40% of the obsolete parts in fact remained unscrapped as late as the end of 1967. Reply Brief for Petitioner 8. The record does not enable us to resolve this factual dispute; in any event, we must accept the Commissioner’s explanation at face value.
 The inventory items broke down as follows:
Raw materials 4,297
Finished parts and accessories 33,670
Finished tools 4,344
Total number of inventory items 44,092
364 T. C., at 158.
 The operation of Thor’s aging formula is well illustrated by a chart set forth in the opinion of the Tax Court. Id., at 159. The chart assumes that 100 units of each of five hypothetical items were on hand at the end of 1964, but that the number of units sold or used in that year varied from 20-100:
Units on Units sold Percent of
hand at or used 0-12 13-18 19-24 +24 write-down
Item 12-31-64 in 1964 Months Months Months Months
A 100 20 20 10 10 60
0% 50% 75% 100%
____ ____ ____ _____
0 5 7.5 60 = 72.5
B 100 40 40 20 20 20
0% 50% 75% 100%
____ _____ _____ _____
0 10 15 20 = 45.0
C 100 60 60 30 10 0
0% 50% 75% 100%
____ ____ ____ _____
0 15 7.5 0 = 22.5
D 100 80 80 20 0 0
0% 50% 75% 100%
____ ____ ____ ____
0 10 0 0 = 10.0
E 100 100 100 0 0 0
0% 50% 75% 100%
____ _____ _____ _____
0 0 0 0 = 0.0
 “So here is where I fell back on my experience of 20 years in manufacturing of trying to determine a reasonable basis for evaluating this inventory. In my previous association, we had generally written off inventory that was in excess of one year. In this case, we felt that that would be overly conservative, and it might understate the value of the inventory. On the other hand, we felt that two years . . . would be too optimistic and that we would overvalue the inventory [in view of] the factors which affect inventory, such as technological change, market changes, and the like, that two years, in our opinion, was too long a period of time.
“So what we did is we came up with a formula which was somewhat in between . . . writing off, say, everything over one year as compared to writing everything [off] over two years, and we came up with this formula that has been referred to in this Court today.” App. 57.
 This write-down was formulated as follows:
Type of Inventory
(1) tool parts and motor
parts at plant A 5 $26,341
(2) raw materials, work-in-process,
and finished goods at plants A and B 10 99,954
(3) hardware items at plant A 50 34,537
64 T. C., at 159-160; App. 209.
 The Tax Court found that the finished tools were too specialized to attract bargain hunters; that no one would buy spare parts, regardless of price, unless they were needed to fix broken tools; that work-in-process had no value except as scrap; and that other manufacturers would not buy raw materials in the secondary market. 64 T. C., at 160-161.
 It appears that 78% of the “excess” inventory at two of Thor’s plants was scrapped between 1965-1971. Id., at 161; App. 218.
 For a manufacturing concern like Thor, Gross Profit basically equals Sales minus Cost of Goods Sold. Cost of Goods Sold equals Opening Inventory, plus Cost of Inventory Acquired, minus Closing Inventory. A reduction of Closing Inventory, therefore, increases Cost of Goods Sold and decreases Gross Profit accordingly.
 The Regulations define “method of accounting” to include “not only the over-all method of accounting of the taxpayer but also the accounting treatment of any item.” Treas. Reg. § 1.446-1 (a) (1), 26 CFR § 1.446-1 (a) (1) (1964).
 See 64 T. C., at 166; Tr. of Oral Arg. 17-19. Even if Thor had made a timely challenge to the Regulations, it is well established, of course, that they still “`must be sustained unless unreasonable and plainly inconsistent with the revenue statutes,’ and `should not be overruled except for weighty reasons.'” Bingler v. Johnson, 394 U. S. 741, 750 (1969), quoting Commissioner v. South Texas Lumber Co., 333 U. S. 496, 501 (1948).
As an alternative to his argument that Thor’s write-down was inconsistent with the Regulations, the Commissioner argues that he was justified in disallowing the write-down in any event because it constituted a “change of accounting method” for which Thor failed to obtain the Commissioner’s prior consent, as required by § 446 (e), 26 U. S. C. § 446 (e). The Regulations define a change of accounting method to include “a change in the treatment of a material item.” Treas. Reg. § 1.446-1 (e) (2) (i), 26 CFR § 1.446-1 (e) (2) (i) (1964). In view of our disposition of the case, we need not reach this alternative contention.
 E. g., D. Loveman & Son Export Corp. v. Commissioner, 34 T. C. 776, 796 (1960), aff’d, 296 F. 2d 732 (CA6 1961), cert. denied, 369 U. S. 860 (1962). See Schnelwar & Jurgensen, The New Inventory Regulations in Operation and Other Inventory Valuation Considerations, 33 N. Y. U. Inst. on Fed. Tax. 1077, 1093-1094 (1975); AICPA Accounting Principles Board, Accounting Research Bulletin No. 43, ch. 4, Statement 6 (1953), reprinted in 2 APB Accounting Principles 6016 (1973). Judge Raum emphasized in D. Loveman & Son that “market” ordinarily means the price the taxpayer must pay to replace the inventory; “it does not mean the price at which such merchandise is resold or offered for resale.” 34 T. C., at 796.
 Thor seeks to justify its write-down by citing Space Controls, Inc. v. Commissioner, 322 F. 2d 144 (CA5 1963), and similar cases. In Space Controls, the taxpayer manufactured trailers under a fixed-price contract with the Government; it was stipulated that the trailers were suitable only for military use and had no value apart from the contract. The taxpayer experienced cost overruns and sought to write down its inventory by the amount by which its cost exceeded the contract sales price. The Court of Appeals, by a divided vote, held that the write-down was authorized by Treas. Reg. § 1.471-4 (b), reasoning that the taxpayer in effect had offered the trailers for sale by way of the fixed-price contract. 322 F. 2d, at 151. While not necessarily approving the Fifth Circuit’s decision to dispense with the “actual sale” rule of § 1.471-4 (b), we note that that case is distinguishable from this one. In Space Controls, the fixed-price contract offered objective evidence of reduced inventory value; the taxpayer in the present case provided no objective evidence of reduced inventory value at all.
Petitioner’s reliance at oral argument on United States Cartridge Co. v. United States, 284 U. S. 511 (1932), is, we think, similarly misplaced. The taxpayer in that case manufactured ammunition for the Government during World War I. In 1918 the taxpayer was instructed to stop production immediately, with a provision that settlement of its claims for unfinished and undelivered ammunition would be negotiated later. At the end of its taxable calendar year 1918, the ammunition was unsalable at normal prices and settlement negotiations had not yet begun; the taxpayer, accordingly, wrote down its 1918 closing inventory to “market,” which was agreed to be $232,000. Id., at 519. The question was whether the taxpayer, in computing its 1918 taxable income, should value its inventory at that figure, or at $732,000, the sum it ultimately realized upon settlement of its claims with the Army in 1920-1922. This Court held that, in accordance with the annual accounting principle, market value controlled, noting that the taxpayer at the end of 1918 “had no assurance as to what settlements finally would be made or that it ever would receive more than the then market value of the inventories.” Id., at 520. This case, we think, may be said to support, rather than to conflict with, the result we reach here. Just as Thor cannot write down its inventory, in the absence of objective evidence of lower value, because of an anticipated future loss, so the taxpayer in United States Cartridge could not be required to write up its inventory, in the absence of objective evidence of higher value, because of an anticipated future gain. In this respect, at least, tax accounting travels a two-way street.
 See, e. g., H. R. Doc. No. 140, 87th Cong., 1st Sess., 14 (1961) (the President’s tax message); B. Bittker & L. Stone, Federal Income, Estate, and Gift Taxation 843 (4th ed. 1972); Skinner, Inventory Valuation Problems, 50 Taxes 748-749 (1972); Schwaigart, Increasing IRS Emphasis on Inventories Stresses Need for Proper Practices, 19 J. Tax. 66, 69 (1963).
 The Commissioner also contends that Thor’s write-down of “excess” inventory was prohibited by Treas. Reg. § 1.471-2 (f), 26 CFR § 1.471-2 (f) (1964). That section states:
“The following methods . . . are not in accord with the regulations in this part:
“(1) Deducting from the inventory . . . an estimated depreciation in the value thereof.
“(2) Taking work in process, or other parts of the inventory, at a nominal price or at less than its proper value.
“(3) Omitting portions of the stock on hand.”
See Rev. Rul. 77-364, 1977-2 Cum. Bull. 183 (percentage write-down of “slow” and “doubtful” inventory violates § 1.471-2 (f) (1)); Rev. Rul. 77-228, 1977-2 Cum. Bull. 182 (deduction from closing inventory of “excess” items still retained for sale violates § 1.471-2 (f) (3)). The Court of Appeals and the Tax Court did not consider these contentions. In view of our disposition, we need not consider them either.
 Until 1973, § 1.471-2 (b) of the applicable Regulations provided in pertinent part:
“In order clearly to reflect income, the inventory practice of a taxpayer should be consistent from year to year, and greater weight is to be given to consistency than to any particular method of inventorying or basis of valuation so long as the method or basis used is substantially in accord with §§ 1.471-1 to 1.471-9. An inventory that can be used under the best accounting practice in a balance sheet showing the financial position of the taxpayer can, as a general rule, be regarded as clearly reflecting his income.”
The inventory Regulations were amended in 1973 to require most taxpayers engaged in manufacturing to use the “full absorption method of inventory costing,” currently set forth in § 1.471-11. T. D. 7285, 1973-2 Cum. Bull. 163, 164; 26 CFR § 1.471-11 (1978). As part of these amendments, the final sentence of § 1.471-2 (b)—containing the “as a general rule” language —was deleted; further, the requirement that inventory practices be “substantially in accord with §§ 1.471-1 to 1.479-9” was revised to require that such methods be “in accord with §§ 1.471-1 through 1.471-11.” 26 CFR § 1.471-2 (b) (1978) (emphasis added). The Tax Court and the Court of Appeals both determined that the 1973 amendments to § 1.471-2 (b) were inapplicable to this case. 64 T. C., at 167; 563 F. 2d, at 866 n. 11. We agree.
 “I think it is pretty obvious that [inventory representing a 10-year supply] has inherently less value [than inventory representing a 1-year supply] because of the things that can happen to the inventory. Some of it will be lost. Some of it may become damaged. Some of it will become obsolete because of the technological change. Some won’t be sold because of the fact that you have market changes. So we were confronted with the problem, as anybody in the manufacturing field [would be], of trying to develop a relationship between inventory quantity and anticipated usage.” App. 56-57 (testimony of Thor’s president).
 AICPA Accounting Principles Board, Statement No. 4, Basic Concepts and Accounting Principles Underlying Financial Statements of Business Enterprises ¶ 171 (1970), reprinted in 2 APB Accounting Principles 9089 (1973). See Sterling, Conservatism: The Fundamental Principle of Valuation in Traditional Accounting, 3 Abacus 109-113 (1967).
 Accord, Raby & Richter, Conformity of Tax and Financial Accounting, 139 J. Accountancy 42, 44, 48 (Mar. 1975); Arnett, Taxable Income vs. Financial Income: How Much Uniformity Can We Stand?, 44 Accounting Rev. 482, 485-487, 492-493 (July 1969); Cannon, Tax Pressures on Accounting Principles and Accountants’ Independence, 27 Accounting Rev. 419, 419-422 (1952).
 See, e. g., McClure, Diverse Tax Interpretations of Accounting Concepts, 142 J. Accountancy 67, 68-69 (Oct. 1976); Kupfer, The Financial Accounting Disclosure of Tax Matters; Conflicts With Tax Accounting Technical Requirements, 33 N. Y. U. Inst. on Fed. Tax. 1121, 1122 (1975); Healy, Narrowing the Gap Between Tax and Financial Accounting, 22 Tulane Tax Inst. 407, 417 (1973); A Challenge: Can the Accounting Profession Lead the Tax System?, 126 J. Accountancy 66, 68-69 (Sept. 1968).
 E. g., Raby & Richter, supra, at 44; Arnett, supra, at 486; 126 J. Accountancy, supra, at 68.
 Arnett, supra, at 492 (noting that there are “many and diverse `acceptable’ practices in valuing inventories, depreciating assets, amortizing or not amortizing goodwill,” and the like); 126 J. Accountancy, supra, at 69 (nothing that “methods of determining inventory costs vary widely and various methods, if consistently applied, will be acceptable for accounting purposes”); Eaton, Financial Reporting in a Changing Society, 104 J. Accountancy 25, 26 (Aug. 1957); Cox, Conflicting Concepts of Income for Managerial and Federal Income Tax Purposes, 33 Accounting Rev. 242 (1958); Cannon, supra, at 421 (suggesting that accountants “are quite prone to define `generally accepted’ as `somebody tried it'”).
 Thor’s experts did not testify that the company’s write-down procedures were the only “generally accepted accounting practice.” They testified merely that Thor’s inventory needed to be written down, and that the formulae Thor used constituted a “reasonable” way of doing this. App. 166, 184, 196.
 The details of the calculation are set out in Black Motor Co. v. Commissioner, 41 B. T. A., at 302. See 2 CCH 1978 Stand. Fed. Tax Rep. ¶ 1624.0992; Whitman, Gilbert, & Picotte, The Black Motor Bad Debt Formula: Why It Doesn’t Work and How to Adjust It, 35 J. Tax. 366 (1971).
 Malone & Hyde, Inc. v. United States, 568 F. 2d 474, 477 (CA6 1978); Business Dev. Corp. of N. C. v. United States, 428 F. 2d 451, 453 (CA4), cert. denied, 400 U. S. 957 (1970); United States v. Haskel Engineering & Supply Co., 380 F. 2d 786, 789 (CA9 1967); Patterson v. Pizitz, Inc., 353 F. 2d 267, 270 (CA5 1965), cert. denied, 383 U. S. 910 (1966); Ehlen v. United States, 163 Ct. Cl. 35, 42, 323 F. 2d 535, 539 (1963); James A. Messer Co. v. Commissioner, 57 T. C. 848, 864-865 (1972).
 Atlantic Discount Co. v. United States, 473 F. 2d 412, 414-415 (CA5 1973) (citing cases); Consolidated-Hammer Dry Plate & Film Co. v. Commissioner, 317 F. 2d 829, 834 (CA7 1963).
 E. g., Malone & Hyde, Inc. v. United States, 568 F. 2d, at 477; First Nat. Bank of Chicago v. Commissioner, 546 F. 2d 759, 761 (CA7 1976), cert. denied, 431 U. S. 915 (1977).
 See, e. g., Rev. Rul. 76-362, 1976-2 Cum. Bull. 45, 46 (“[A]s a general rule, the Black Motor formula may be used to determine a reasonable addition to a reserve for bad debts” under § 166 (c)).
 E. g., Atlantic Discount Co. v. United States, 473 F. 2d, at 413, 415; Ehlen v. United States, 163 Ct. Cl., at 45, 323 F. 2d, at 540-541; James A. Messer Co. v. Commissioner, 57 T. C., at 857, 865-866.
 See § 585 (b) (3) of the 1954 Code, 26 U. S. C. § 585 (b) (3) (using “six-year moving average” formula as alternative method of computing reasonable addition to bad-debt reserve for banks); § 586 (b) (1) (using “six-year moving average” formula to compute reasonable addition to bad-debt reserve for small business investment companies).
 E. g., Westchester Dev. Co. v. Commissioner, 63 T. C. 198, 212 (1974), acq., 1975-2 Cum. Bull. 2 (Commissioner abused discretion in invoking Black Motor where taxpayer’s recent bad-debt experience was “wholly unrepresentative” given its “comparatively brief operational history”).
 E. g., Calavo, Inc. v. Commissioner, 304 F. 2d 650, 651-652, 654 n. 4, 655 (CA9 1962) (extraordinary addition to reserve to cover losses on accounts due from debtor who recently became insolvent).
 Indeed, as has been noted, a significant portion of Thor’s addition to its reserve reflected blanket aging of accounts. Both the Treasury, Rev. Rul. 76-362, 1976-2 Cum. Bull. 45, 46, and the courts, United States v. Haskel Engineering & Supply Co., 380 F. 2d, at 787, 789; James A. Messer Co. v. Commissioner, 57 T. C., at 857, 866, have held that such mechanical formulae are inadequate to overcome the Commissioner’s discretionary invocation of Black Motor under § 166 (c).
Debt & Collections Case Law THOR POWER TOOL CO. v. COMMISSIONER OF INTERNAL REVENUE – Supreme Court of United States
439 U.S. 522 (1979)