In this action in the District Court for the Southern District of New York, United States Steel Corporation sought large refunds of federal income and excess profits taxes for 1950. The issue with respect to income tax involved solely a question of law concerning the respective rights of the lessor and the lessee of Minnesota iron mines to percentage depletion; it was decided adversely to U.S. Steel on the Government's motion for partial summary judgment, 270 F. Supp. 253 (1967). The other claims concerned the Korean excess profits tax for 1950; it will be as well to reserve statement of them until we have discussed the issue of percentage depletion.

The facts giving rise to this portion of the case are sufficiently stated in three paragraphs of the complaining:

In 1950, Plaintiff was lessor and lessee of iron ore mining properties in the State of Minnesota. Pursuant to the terms of the leases, for 1950 the lessee paid Minnesota ad valorem taxes with respect to the mining properties and Minnesota royalty taxes with respect to royalties paid by the lessee.

In determining the Federal income tax for 1950 which Plaintiff has paid, for purposes of 114(b)(4) and the determination of the depletion deduction allowable under 23(m), the Minnesota ad valorem taxes and the Minnesota royalty taxes have been included in the lessor's 'gross income from the property,' and have been excluded from the lessee's 'gross income from the property.'

These adjustments were erroneous, and by reason thereof, in determining the Federal income tax for 1950 which Plaintiff has paid, deductions for depletion have been allowed in the amount of at least $304,323.54 less than the amount which should have been allowed.

Both sides moved for partial summary judgment on this issue; the district court granted the Government's motion, 270 F. Supp. 253 (1967).

A deduction for depletion, based broadly on 'the theory that the extraction of materials gradually exhausts the capital investment in the mineral deposit,' 4 Mertens, Law of Federal Income Taxation 24.02 at p. 7 (1966 rev.), has been allowed in one form or another since the Revenue Act of 1913. Id. 24.06-24.15b. The general problem of allocating the depletion deduction between lessor and lessee was first dealt with by Congress in the Revenue Act of 1918. Section 214(a)(10) of the 1918 Act, 40 Stat. 1068, provided that 'in the case of leases the deductions * * * shall be equitably apportioned between the lessor and lessee.' See 234(a)(9), 40 Stat. 1078-1079, for the identical provision with respect to the corporate income tax. Percentage depletion, i.e., depletion based on a stated percentage of 'gross income from the property' (defined, insofar as here relevant, as 'gross income from mining') rather than on the cost of the property, originated, but then only for oil and gas wells, in the Revenue Act of 1926, 204(c)(2), 44 Stat. 16.1 Six years later in the Revenue Act of 1932, Congress addressed itself specifically to the problem of apportioning the depletion deduction between lessor and lessee with respect to percentage depletion. In addition to retaining the 'equitable apportionment' provision, 23(l), 47 Stat. 181, Congress, in rewriting the provision for depletion of oil and gas wells with respect to a percentage of the taxpayer's gross income from the property and in extending this to coal, metal and sulphur mines, 114(b)(3) and (4), 47 Stat. 202-03, inserted the phrase:

excluding from such gross income an amount equal to any rents or royalties paid or incurred by the taxpayer in respect of the property.

The statutes here relevant, 23(m) and 114(b)(4) of the 1939 Code, do not differ in any significant way.

In a case decided in 1934 with respect to the tax years 1925, 1926 and 1927, the Supreme Court held in effect that the provision of the 1932 Act for the deduction of royalties from the gross income of a lessee stated a result that would have been reached without it, since Congress could not have intended one consequence when the lessee turned over royalty oil in kind to the lessor and a different one when the lessee sold the oil and paid the lessor in cash. Helvering v. Twin Bell Oil Syndicate, 293 U.S. 312, 321, 55 S. Ct. 174, 79 L. Ed. 383 (1934).2 The decision did not speak to the question here at issue, whether the deduction from the lessee's gross income (and the inclusion in the lessor's) should comprehend the payments of taxes which the lessee had contracted to make and would otherwise have been payable by the lessor.

The Regulations under the 1939 Code said nothing about excluding from the lessee's gross income amounts it was obligated to pay as taxes. They read, as had their predecessors going back to those under the 1932 Act:3

In all cases there shall be excluded in determining the 'gross income from the property' an amount equal to any rents or royalties which were paid or incurred by the taxpayer in respect of the property and are not otherwise excluded from the 'gross income from the property.'

The taxpayer asserts, and the Government has shown nothing to controvert this, that for many years the Commissioner made no attempt to require lessees to exclude from gross income taxes they were required to pay for the account of the lessor, and lessors did not seek to include such amounts in their depletable gross income.

In the early 1950's several lessors, doubtless stimulated by the established rule that 'taxes paid by a tenant to or for a landlord for business property are additional rent and constitute a deductible item to the tenant and taxable income to the landlord, the amount of the tax being deductible by the latter,' Reg. 45, Art. 109 (1921), approved in United States v. Boston & Maine R.R., 279 U.S. 732, 49 S. Ct. 505, 73 L. Ed. 929 (1929), began to press for the inclusion of taxes, which lessees were required to pay on the lessors' behalf, in the lessors' gross income as 'rents or royalties,' thus requiring, in principle at least, a corresponding exclusion of these amounts from the lessees' depletable income-- since 'at all events (the depletion provision) must be read in the light of the requirement of apportionment of a single depletion allowance.' Helvering v. Twin Bell Oil Syndicate, supra, 293 U.S. at 321, 55 S. Ct. at 178. The Bureau, doubtless responding to conflicting pressures, adopted a somewhat Solomonic course, providing for the apportionment of such taxes.4 Dissatisfied with not getting the whole hog, Burt, a lessor whose lease there involved is also one of those in question in the instant case, brought the issue with respect to the Minnesota ad valorem and royalty taxes before the Court of Claims.5 The court ruled unanimously in his favor, Burt v. United States, 170 F. Supp. 953, 145 Ct.cl. 282 (1959). Speaking of the ad valorem tax, it reasoned:

We are unable to escape the conclusion that under the terms of this particular lease the payment of the ad valorem taxes in the minerals in place was a part of the royalty compensation to plaintiffs. But for the provision in the lease that the mineral taxes were to be paid by the lessee, the levy would have been an in rem tax against the land itself, of which plaintiffs were the actual owners. Undoubtedly if the lessee had not agreed to pay these taxes the plaintiffs would have asked for and been entitled to a larger royalty payment in cash or in an increased percentage or payment of some kind. It seems to us, in essence, that it was a part of the total production income which the plaintiffs received and therefore they are entitled to the statutory depletion allowance on their part of the total production income which includes the ad valorem tax on the minerals as a part of the compensation, rent, or royalty.

Although unnecessary to its decision,6 it applied the same reasoning to the royalty tax, and added:

Of course, taxes as such are not a part of royalty production for either the lessor or the lessee. In this particular case we are not holding that as taxes they are the subject of depletion but as part payment to the lessor of his part of the depletion allowance, it being a part of the yardstick to measure lessor's share of the income from the production of iron ore in connection with the grant or privilege to the lessee to conduct the mining operations on the property.

The same logic does not apply to the lessee's payment of the in rem taxes on the property. They are not a part of his income from the property, but are a part of the payment of the cost of securing the mining lease privilege on plaintiff's property. He may be entitled to charge it off as a part of his expense or as a part of the cost of the property, but such taxes are certainly not a part of his production income. He is entitled to a depletion allowance on his income from the production of the property but this income certainly does not include the taxes which he may have paid.

Apparently concerned with the loss of revenue threatened in suits by lessors challenging the partial or total disallowance of taxes paid by lessees as depletable gross income, although recovery against the lessee would probably be barred by limitation,7 see 26 U.S.C. 1311-1315, the Government sought reconsideration of Burt, so far as concerned the Minnesota ad valorem tax, in Handelman v. United States, 357 F.2d 694, 174 Ct. Cl. 1042 (1966).8 Its argument was that whether or not the ad valorem taxes were 'rents or royalties,' they were not depletable gross income to the lessor since they were payable in any event regardless of production. After extensive review of the Supreme Court decisions on depletion, none of which it found to be particularly relevant, the court unanimously adhered to Burt with respect to any year where there was sufficient production to cover the payment of the taxes. 357 F.2d at 703.9

U.S. Steel's multi-pronged attack on these precedents may be reduced to three essential thrusts: that for percentage depletion purposes only royalties 'as such,' determined by the agreement of the parties, should be treated as the lessor's gross income from the property; that even if the lessee's assumption of the lessor's tax obligations be deemed an additional 'rent or royalty' within the meaning of the depletion provisions, under Minnesota law both the royalty and ad valorem taxes are in rem and hence may not be treated as primarily the lessor's obligation; and that in any event the ad valorem taxes are not dependent on production and hence payment of them by the lessee can under no circumstances be considered depletable income to the lessor excludable from the lessee's share of the depletion allowance.

With respect to the first point, we, like other courts that have considered them, have not found the multitude of cases on which U.S. Steel relies to be of much assistance. Neither Helvering v. Mountain Producers Corp., 303 U.S. 376, 58 S. Ct. 623, 82 L. Ed. 907 (1938) nor Helvering v. Twin Bell Oil Syndicate, supra, which U.S. Steel suggests to mandate the conclusion that only royalties 'as such' must be excluded from the lessee's gross income from the property, is truly apposite. Mountain Producers dealt not with the narrow problem whether payments of taxes should be considered as added royalties, but with a broader contention that the contractual assumption by the lessee of all the mining functions 'as part of the purchase price of the oil, 303 U.S. at 380, 58 S.Ct. at 624, should affect the allocation of the depletion allowance by allowing the lessor to include in its gross income from the property 'the cost of production defrayed by the (lessee) under its contract.' 303 U.S. at 379, 58 S.Ct. at 624. When it rejected that contention and said that under the circumstances the cash payments to the lessor represented its return on the extraction of the resources, the Court was hardly meaning to lay down a general rule that only cash payments made directly to the lessor may be considered as rents or royalties. Twin Bell, as indicated above, likewise did not deal with the question before us; indeed it may well be that in Twin Bell itself, some of the payments held excludable from the lessee's gross income had been made to the tax authorities, see fn. 2 although in diminution of rather than in addition to those labeled strictly as rents and royalties.

The issue in Callahan Mining Corp. v. C.I.R., 428 F.2d 721 (2 Cir.), cert. denied, 400 U.S. 903, 91 S. Ct. 141, 27 L. Ed. 2d 140 (1970), on which U. S. Steel also relies, was whether the lessor had an economic interest in the working interest supporting a depletion deduction based on gross income from the property or merely a net-profits, non-operating interest supporting a depletion deduction 'only on the amounts he actually received as the ore is mined and sold.' 428 F.2d at 725. U.S. Steel quotes this language and apparently relies on the case for the proposition that the court cannot construct a gross income based on anything other than the agreement of the parties and that the lessor is entitled to a deduction for depletion 'only on the amounts he actually received.' What neither party mentions, however, is that the very issue here facing us-- at least insofar as the 'royalties as such' argument is concerned-- was determined by the Tax Court consistently with the decisions in Burt and Handelman, which it cited. 51 T.C. 1005, 1021-1022. While there appears to have been no dispute over the question, the fact is that the lessee's payment of the portion of the tax-- there an Idaho 'net profits' tax-- for which the lessor was liable was held to be includible in the lessor's gross income. Finally, despite appellant's attempt to depreciate its force, Louisiana Land and Exploration Co. v. Donnelly, 394 F.2d 273 (5 Cir. 1968),10 constitutes another precedent in favor of the Government in this case. The court there upheld the claim of an oil and gas lessor for inclusion in depletable gross income of a Louisiana severance tax which the lessee had paid.

U.S. Steel next maintains that even if payment by the lessee of the lessor's taxes may in some circumstances be considered as additional 'rents or royalties,' Burt and its progeny were wrong in treating the Minnesota ad valorem tax as an obligation of the lessor which was discharged by the lessee, since the tax is merely in rem against the property, as, it asserts, is the royalty tax. Marble v. Oliver Mining Co., 172 Minn. 263, 215 N.W. 71 (1927). While U.S. Steel appears to be correct in contending that these taxes are treated as in rem obligations under Minnesota law, it does not follow that the burden of paying them is not on the lessor's interest and that the assumption of this burden does not constitute additional rents or royalties within the meaning of the depletion statute. As stated by the Supreme Court of Minnesota:

The royalty tax is an imposition upon the lessor's right, title, and interest in the lands demised. * * * In absence of (a) covenant (to pay the tax), the tax would, no doubt, have to be paid by the lessor to protect his title to the land. The intent of the lawmakers was to lay the burden of this tax upon the interest and estate of the one who granted the permission to mine.

Marble v. Oliver Mining Co., supra, 172 Minn. at 266, 215 N.W. at 72. In Fraser v. Vermillion Mining Co., 175 Minn. 305, 221 N.W. 13, 14, appeal dismissed, 278 U.S. 572, 49 S. Ct. 93, 73 L. Ed. 512 (1928), the same court, in speaking of the royalty tax, noted that 'the shifting of the burden of the tax from lessor to lessee is a matter of contract between the two. * * *' And in State v. Rea, 189 Minn. 456, 250 N.W. 41, 42 (1933), the Supreme Court of Minnesota was confronted with a case which was

the only one that has come to our notice where the state has not collected royalty tax from the recipient of the royalty, or from the lessee who may have stipulated to pay it as part consideration for the lease or permit under which exploring for or mining ore is done.

We have no reason to believe that the Minnesota courts would regard the agreement of the lessee to pay ad valorem (as distinct from royalty) taxes of the lessor as anything other than 'part consideration for the lease.' In any event, we think it rather fanciful to assume that in entering a lease agreement the lessee's contractual assumption of state taxes, which is obviously of value to the lessor, does not correspondingly reduce the amount of cash payments which the lessor receives from the lessee upon the extraction of the ore. Indeed, we can do little better than recite the hypothetical put by Judge Jones, speaking for the court in Burt, 170 F. Supp. at 959:

The writer is familiar with one oil lease contract covering 1,000 acres, only a small part of which was believed by the parties to be an area of production. The lease contract provided that the owner, for a three-year period of production should receive only a payment of all taxes by the lessee. After three years he was to receive the payment of all taxes or 1/8 of the proceeds of the oil produced, whichever was larger.

Does anyone believe that in such a contract the lessee would be entitled to all the depletion allowance and the lessor to none; or that in the event the 1/8 royalty exceeded the amount of taxes, the owner's part of the depletion allowance should be limited to the excess of payments above the amount of the taxes?

Finally, U.S. Steel's contention that the payment of ad valorem taxes by the lessee cannot be depletable income to the lessor since they are not dependent solely on production was dealt with at length, and, we believe, correctly, in the Handelman opinion so that it is unnecessary to repeat those considerations here.

If the question were arising for the first time, we should find the lessee's position somewhat more impressive, and the lessor's correspondingly less so, then the other courts that have considered the problem. One reason, not argued by U.S. Steel, is that when Congress wished to authorize a lessee to deduct sums other than rents properly so-called, it demonstrated its knowledge of how to say so. Section 23(a) of the 1932 Act, later 23(a)(1)(A) of the 1939 Code and now 162(a)(3) of the 1954 Code,11 permitted the deduction of 'rentals or other payments required to be made as a condition to the continued use or possession, for purposes of the trade or business, of property to which the taxpayer has not taken or is not taking title or in which he has no equity.' Cf. Duffy v. Central R.R., 268 U.S. 55, 45 S. Ct. 429, 69 L. Ed. 846 (1925). If Congress had used the italicized words in 114(b)(4)(A) of the 1939 Code, the Government would surely be entitled to prevail; since Congress used a narrower phrase and has allowed the variance to persist for forty years, arguably it desired 'rents and royalties' to mean only payments specifically denominated as such. Although courts often indulge in such an approach to statutory interpretation, often for want of any better way to 'find' an intent where probably there was none, there may be something rather irrational about it, especially in regard to revenue acts. The depletion provisions in the 1932 Act were doubtless drafted by specialists on the subject, quite without regard to the longstanding provision in the deductions section; and the spectacle of the average Congressman, or even of the knowledgeable members of the Ways and Means and Finance Committees, scrutinizing the language with the powerful magnifying glass needed to detect the different phrasing of the two sections and appreciate its possible significance is somewhat unrealistic.12

Again there is something essentially grotesque in the notions that a lessor's depletable interest grows as a state increases rates under existing tax laws and imposes new ones (as Minnesota did with the royalty tax here), or that it is 'equitable' that a lessee, in addition to paying these heavier taxes, should suffer a loss in the proportion of the gross income on which it can take percentage depletion. We are also somewhat impressed by U.S. Steel's argument that many mineral lands in Minnesota are owned and leased by the State, that an ad valorem tax is nonetheless imposed, that payment of such a tax by the lessee cannot be regarded as discharging an obligation of the lessor, and that the Government's position would thus result in a lessee of state lands being able to include a larger proportion of gross income as subject to depletion than the lessee of privately owned land.13 Yet it may be said against such arguments that the logical underpinnings of percentage depletion are so elusive-- some would use a stronger adjective-- that it is idle to attempt principled reasoning about its division between lessor and lessee.

Moreover, despite these considerations we are bound to agree that there is much force in the arguments in favor of the lessor, well expressed in the extracts from the Burt opinion we have quoted. There would seem to be little sense in a rule that would attribute different tax consequences to a situation where the lessor accepted a lower royalty and required the lessee to pay the taxes and another where the lessor insisted on a royalty higher by his estimate of what the taxes would be. The arguments here forcefully made by U.S. Steel have been considered and rejected on more than one occasion by two courts particularly expert in tax matters and by another court of appeals having much experience with oil and gas leases, as well as by the able district judge. While we are not bound by any of these decisions and would not hesitate to reject them if we were truly convinced they were in error, our beliefs simply do not rise to that height. Burt has been on the books for a dozen years, during which Congress has adopted a number of corrective tax measures but has not chosen to alter Burt. Under the circumstances we think it best to follow the uniform line of authority and leave correction, if that is called for, to the Supreme Court or, for the future, to the Congress.

The Excess Profits Tax Act of 1950, 64 Stat. 1137 (1951), sometimes referred to as the Korean Excess Profits Tax, imposed a heavy tax on 'excess profits' for years ending after June 30, 1950, and beginning before January 1, 1954. Broadly speaking, it followed the general scheme of the Excess Profits Tax Act of 1940 (the World War II Excess Profits Tax) in determining excess profits by comparing net income during the war years either with the actual average net income during a 'base period' consisting of the four preceding taxable years, 435, or with a stipulated percentage rate of return on invested capital, 436, 437, whichever was higher. A taxpayer using the average income method, as U.S. Steel did here, was allowed to eliminate the worst of the four years in determining its general average base period net income, 435(d)(2); the company eliminated 1946. Under the Korean Excess Profits Tax only 83% Of the average base period net income, plus and minus certain adjustments, could be used as a credit against wartime profits, 435(a)(1); the corresponding figure under the World War II Act had been 95%, 713.

Knowing that such a comparison might work unfairly in certain cases. Congress provided for a number of specific adjustments. But it recognized also that these might not be sufficiently comprehensive. It therefore, included what has come to be known as a general relief provision, 442(a), which reads in pertinent part:

If a taxpayer which commenced business on or before the first day of its base period establishes that, for any taxable year, within, or beginning or ending within its base period:

(1) normal production, output or operation was interrupted or diminished because of the occurrence, either immediately prior to, or during such taxable year of events unusual and peculiar in the experience of such taxpayer, or

(2) the business of the taxpayer was depressed because of temporary economic circumstances unusual in the case of such taxpayer, the taxpayer's average base period net income determined under this section shall be the amount computed under section (c) or (d), whichever is applicable.

In any year for which this section applied, a constructive income was to be computed for the affected months by applying the industry's yearly rate of return as determined by the Secretary of the Treasury, 442(e).14 If only one year was affected by an abnormality within the meaning of 442(a) and the constructive income exceeded 110% Of the actual, the constructive income was to be substituted for that year alone, 442(c). If 442(a) applied to two or more years, constructive income (if it exceeded 110% Of the actual) was to be used for all three years of the base period, 442(d).15

U.S. Steel claimed that each of the three base years remaining after elimination of the worst year, 1946, came within the general relief provision of 442(a). The grounds for this, summarily stated, were as follows:

(1) With respect to 1947 and 1948, because its production was curtailed and its costs increased by abnormalities in the supply of raw materials due to causes hereafter described (the 'materials' issue); (2) With respect to 1947 and 1948, because the price controls that had expired on November 13, 1946, and President Truman's request of July 15, 1947, that coal and steel companies forego price rises, prevented its making price increases it would other wise have made to meet rising costs of goods and services, resulting in an abnormally narrow cost-price gap (the 'cost-price' issue); (3) With respect to 1948, because its production was significantly curtailed as a result of two strikes, one from March 15 to April 22 and the other from July 5 to July 14, by the United Mine Workers which shut down coal producing facilities (the 'coal strike' issue) and an explosion at one of its Chicago works; (4) With respect to 1949, because of an interruption of production by a strike of the United Steel Workers from October 1 to November 11 (the 'steel strike' issue).

Judge Levet found against the company on the first three issues and for it on the fourth. The first two issues were decided on motions by the Government for summary judgment, 305 F. Supp. 497, 508 (1969); U.S. Steel has appealed with respect to the first of these and apparently also with respect to the second. But see, fn. 33 infra. The third and fourth issues were decided after trial, 316 F. Supp. 990 (1970). U.S. Steel has appealed from the decision against it on the coal strike issue, but the Government has withdrawn its appeal from the decision in the taxpayer's favor with respect to the 1949 steel workers' strike. The questions on the coal strike issue are largely of fact, on the materials and cost-price issues, at least at the present stage, of law. We shall deal initially with the issue concerning the 1948 coal strikes.

The judge and the parties were in agreement with respect to the basic legal principles determining whether the 1948 coal strikes brought U.S. Steel within 442(a)(1).16 Strikes are listed in the Regulations' definition of 'events,' Regs. 130, 40.442-2(3). On the other hand, the event must be 'unusual and peculiar in the experience of the taxpayer'; 'its occurrence' must not be 'ordinarily encountered in the taxpayer's business operations.' Regs. 130, 40.442-2(3). Also 'the interruption or diminution' of normal production by the unusual and peculiar event 'must be significant and not trivial.' Regs. 130, 40.442-2(2).

The first breach in the harmony on this issue concerns the judge's ruling that, in determining whether the 1948 strikes were 'unusual,' the comparison should not be simply with previous coal strikes, of which there apparently had been none of serious consequence during the period of comparison, but rather with all sorts of work stoppages due to labor disputes. We see no basis for faulting this ruling on the facts of this case. U.S. Steel was not a seller of coal in any substantial degree; its business was the integrated production of finished iron and steel products and coal mining was simply a part of the process. It would be unreasonable to read 442(a)(1) as entitling a taxpayer to relief because it had suffered a strike from a previously peaceful craft in one of the base years, although it had been the victim of equally or more serious strikes by other crafts in a period appropriate for comparison. The court selected the years 1941-45 as such a period. While U.S. Steel asserts that the court 'simply disregarded the testimony of qualified witnesses about the taxpayer's experience in the period prior to 1942 because no 'statistics' were kept' for those years-- testimony to the effect that during the 1930's strikes and other labor difficulties did not interrupt steel production in any serious way-- it does not challenge Judge Levet's findings that 'the only relevant years prior to 1948 for which plaintiff has presented any statistical evidence regarding strike losses are 1941, 1942, 1943, 1944, and 1945' and that 'plaintiff started in 1941 to record estimates of production losses due to strikes.' 316 F. Supp. at 1008. Apart from any question of remoteness, given the lack of reliable data for earlier years, Judge Levet was justified in limiting his comparison to the 1941-45 period.

The court assumed the production losses from the two 1948 coal strikes were those alleged by U.S. Steel on the basis of its exhibits, finding it unnecessary, in light of its ultimate determination, to consider the Government's contention that these figures were excessive. 316 F. Supp. at 1011. These exhibits indicated the following loss figures:

                        Thousands of Tons
                                Equivalent
                     Ingots  Finished Products
March-April Strike  502,986       358,135
July Strike          76,870        55,210
                    -------  -----------------
       Total        579,856       413,345

The difficulty in making a comparison between these figures of lost production in 1948 and the results for the years in the period chosen for comparison stemmed from the lack of similar reports for the earlier years. During those years the company had generally prepared 'Strike or Work Stoppage Reports,' called Form B, which showed the loss for each type of production for each of its plants. The problem was twofold. After so much time had elapsed, some of the Form B reports were missing for some of the years. Moreover, since they showed the loss of each type of product and in a number of cases a product loss of more than one type was recorded on individual or related reports, mere addition would result in pyramiding.

Faced with these problems Judge Levet engaged in what U.S. Steel properly characterizes as 'a painstaking analysis.' For two of the comparison years, 1943 and 1945, he arrived at firm figures of losses in net shipments (which the parties agreed was the appropriate loss indicator since 'in the stell-making industry production and total shipments are generally correlative,' 316 F. Supp. at 1008) of steel products due to work stoppages; for the other three, 1941, 1942 and 1944, he was able to arrive only at a range. These findings were as follows:

Reasonable Approximation of Potential Tonnage

1941 ................  13,456 --  270,852
                          (Min.)   (Max.)
1942 ................   6,552 --   45,223
                          (Min.)   (Max.)
1943 ............................ 246,389
1944 ................ 343,985 --  553,370
                          (Min.)   (Max.)
1945 ............................ 828,639

Comparing these figures with those for 1948 cited above, he concluded that U.S. Steel had failed to show that the 1948 coal strikes had constituted 'events unusual and peculiar in the experience of such taxpayer.'17

Apart from the arguments that coal strikes should have been considered as something separate and apart from other work stoppages, and that the comparison should have included still earlier years, which we have already rejected, U.S. Steel does not seriously contest that this conclusion would follow if the factual findings were correct. It insists, however, that the judge did not succeed in doing what he plainly intended, namely, to eliminate the duplication or worse that would result from simply adding up the figures in the Form B Reports. It insists that the only way in which double or triple counting could be avoided would be to base the calculations on loss in ingots, the 'bottleneck' in the production of finished steel, or-- what amounts to the same thing-- on a loss of finished products determined by applying to ingot losses an apparently recognized ration of approximately 70%. On these bases U.S. Steel comes up with the following tables:

Year           Ingot Loss
                     Coal Strikes
        All Strikes      Only
1941 ....... 18,715
1942 ........ 9,441
1943 ....... 48,000
1944 ...... 118,721
1945 ...... 478,500
1948 ................. 579,856

Year       Loss of Equivalent
            Finished Product
                     Coal Strikes
        All Strikes      Only
1941 ....... 13,456
1942 ........ 6,552
1943 ....... 31,920
1944 ....... 81,917
1945 ...... 333,036
1948 ................. 413,345

Mere statement of the theory behind U.S. Stell's calculations is almost sufficient to demonstrate its unsoundness, even ignoring the fact that some of the lost ingot figures are based on Form B reports for years in which they were concededly incomplete. Unlike the 1948 coal strikes, may of the work stoppages in the comparison years did not affect ingot production at all. Hence a comparison solely in terms of loss of ingots or of finished products determined on a ratio basis could not possibly be fair, since it fails to consider that losses of net shipments may result from interruptions in the production schedule at a stage subsequent to ingot production. While U.S. Stell baldly asserts that 'loss of finished net product production does not, could not, and did not translate into loss of shipments unless equivalent ingot production was * * * lost,' it points only to the rather general statements of two of its witnesses concerning the ability of the company to 'shift' its production from one finishing plaint to another in support of that proposition. Assuming that some shifting is possible U.S. Steel has referred us to no evidence that such shifting in fact occurred during the strike periods so as to eliminate all but reported ingot losses-- let alone, with the few exceptions noted below, see fns. 19 and 20, a substantial portion of those losses which Judge Levet carefully determined.

No useful purpose would be served by our endeavoring to restate the methods whereby the judge determined production losses for 1941-45, an analysis filling some 20 printed pages of his opinion, 316 F. Supp. 1011-1030.18 Suffice to say that we have carefully examined this and that while we recognize some validity in certain of U.S. Steel's criticisms,19 the company has not come forward with any more accurate figures, as it had the burden of doing, and we surely are not prepared to brand the judge's careful findings as 'clearly erroneous.'20

Paragraph 11 of the complaint, relying on both subsections of 442(a) claimed that U.S. Steel's normal production in 1947-48 was diminished by reason 'of events unusual and peculiar in the experience' of the taxpayer and its business 'was depressed because of temporary economic circumstances unusual' to it. Among the 'temporary abnormalities' were 'low quality of coking coal and iron ore, resulting in loss of production' and, also apparently, in higher costs.

Since this issue was decided against the taxpayer on a motion for summary judgment by the Government, we must consider assertions in the company's affidavits and depositions to be true, except insofar as these are invalidated by admissions or uncontradicted proof. Since we consider that the judge's findings, 305 F. Supp. at 501, drawn mainly from a joint statement under Rule 9(g) of the district court which he required, are not sufficiently generous to the taxpayer under this standard, a fuller statement of plaintiff's contentions is needed.

The problem concerned U.S. Steel's northern plants comprising 80% Of its total production. Before World War II the iron ore for these plants, principally of high grade,21 came from mines on the Mesabi Range in Minnesota. The coking coal came principally from mines in western Pennsylvania (Frick District coal), which was of high grade,22 and to a lesser degree from southern West Virginia (Pocahontas coal) and western Kentucky (the Lynch District coal). Such washing of the Frick District coal which needed further 'beneficiation' was done at a facility at plaintiff's Clairton, Pa. plant. Although it was foreseeable before World Was II that the high grade Frick District coal would ultimately be exhausted, it was thought that the problem could be met for some years by diminishing the relative share furnished by such coal and blending this with lower grade coal to obtain a suable mixture-- although ultimately beneficiating facilities to utilize lower grade coal would have to be installed. Similarly, although U.S. Steel foresaw before World War II that its reserves of high grade iron in the Mesabi Range would not last forever, the war produced problems concerning sources of future iron production more quickly than could have been anticipated.

The enormous demand for steel induced by the advent of World War II made it necessary for U.S. Steel to maximize coal and iron ore mining and production regardless of the effect of such action on post-war operations.23 For reasons about to be developed, the effect of this was to deteriorate the quality of coal and iron available for use in 1947 and 1948.

The Palmer and Colonial coal mines in the Frick District which had been the company's principal source of high grade coal were mined during World War II to such a degree that their reserves were rapidly depleted. Idle mines in the Frick District, on whose production plaintiff had counted for future mixing with lower grade coal, were brought into production, but this also did not suffice. Accordingly the company had to begin tapping reserves of low quality coal in Greene County. In 1944 it brought the Robena Mine in that county into full production. Although this turned out to have relatively high proportions of sulphur and ash, it was still possible to achieve satisfactory results by blending it with the high quality coals from the reopened mines in the Frick District and also by using the Clairton washer, although this had not been designed for coal of such poor quality. Realizing that a new washer would be required for the Greene County coal, plaintiff prepared specifications in 1944, but when the bids came in, it was found that the proposed plant (which was similar in concept to the Clairton washer) would be inadequate for the job due to insufficient reduction of the sulphur content in the coal. In an effort to develop a beneficiation process sufficiently effective to reduce the impurities in the Greene County coal to appropriate levels, U.S. Steel then conceived the plan of using a method (a heavy media process) that had been used for beneficiating iron ore but was novel in the coal industry. After preliminary tests indicated that this had promise, U.S. Steel took bids in the latter part of 1945, but warinduced shortages of men and materials delayed completion of the Robena washer until the latter part of 1948. When it came on line, the quality of the beneficiated Robena coal was greatly improved.24 Because of these conditions and those described in footnote 24, plaintiff was obliged in 1947 and 1948 to purchase coal from other producers. Due to scarcity in supply, it was necessary to purchase relatively small amounts from many sources, with consequent lack of uniformity. In summary, as stated in plaintiff's brief, 'the quality of coal available to charge plaintiff's coke ovens and the quality of resultant coke to charge to plaintiff's blast furnaces was the worst in history.'

The iron story was much the same, although less specifically documented. During World War II plaintiff used its best grades of iron ore in order to maximize production. For the same reason it deferred its normal stripping program on the Mesabi Range. It was necessary to catch up on this thereafter. Also, by 1950 and 1951, plaintiff had added sintering and washing facilities that in the average iron content in iron-bearing materials. Again, to quote plaintiff's brief, 'as a result of the conditions described above, the quality of the furnace burdens charged to plaintiff's blast furnaces in 1947 and 1948 was inferior to that experienced at any other time outside the base period.'

Plaintiff avers that the combined effect of the conditions summarized was equivalent to the daily charging to its blast furnaces of a hundred 60-ton railroad cars of inert or harmful materials, more than had ever been charged before or since, with a consequent loss of 2,000,000 tons in annual production of hot metal. Efforts to remedy this through the purchase of scrap were only partially successful, due, among other things, to a shortage of good quality scrap and a premium price. In consequence, finished steel production was at least 900,000 tons below normal in 1947 and 850,000 tons below normal in 1948,25 with resulting adverse effect on earnings of at least $36,000,000 and $42,000,000.

On these facts we do not think it was proper to render summary judgment against the plaintiff, whatever the right decision may be after a full development of the circumstances at a trial.26

The district court's adverse ruling with respect to 442(a)(1) seems to have rested primarily on the judge's belief that U.S. Steel had failed to show an 'event' such as that section requires. The Regulations, 40.442-2(a)(3), say:

Unusual and peculiar events contemplated in section 442(a)(1) consist primarily of physical rather than economic events or circumstances. Such physical events include floods, fires, explosions, strikes, and other exceptional and uncommon circumstances hindering production, output, or operation.

Apart from the use of the word 'primarily,' we do not see why such effects of the greatest war in the world's history as we have described above should not be regarded as unusual and peculiar 'physical' events.²⁷ Neither do we perceive why it should be material that the effects of World War II should take the form of a loss of access to suitable raw materials as distinguished, say, from the physical destruction of plants in the area of hostilities.28 We can see no significant difference between the latter case, which surely should qualify under 442(a)(1), and one where World War II deprived a manufacturer of access to a needed raw material, e.g., Malayan tin or rubber, and he had not been able to restore his imports to the accustomed scale in one of the base years.29 Coming even closer to the instant case, if the Government had seized all of U.S. Steel's high grade coal and iron mines because it thought the company was not mining them as effectively as the war effort required, see 54 Stat. 885, 892 (1940), as amended by 57 Stat. 163, 164 (1943), had exhausted them, and had left U.S. Steel completely dependent on low grade or purchased coal and iron with consequent diminution of normal production through 1947 or 1948, we cannot see why the statutory test would not be met. The company should not fare worse for having behaved as a good citizen and voluntarily done what the Government might have compelled.

Alternatively, we fail to see why, on the affidavits and depositions before the district court, U.S. Steel did not qualify under 442(a)(2). Subsection (2) covers situations rendered abnormal by factors other than interruption or disruption of normal production output or operation, e.g., a taxpayer producing usual volume at unusually high cost or selling at unusually low prices. However, the two subsections are closely related; as the Bulletin relating to the World War II Excess Profits Tax recognized, pp. 16-17, 'the same basic cause might lead to a section 722(b)(1) claim in the case of one taxpayer and a section 722(b)(2) claim in the case of another.' 7A Mertens, supra, 42.121 at 719. It would seem enough to qualify under 722(b)(2) (or 442(a)(2)) that a taxpayer's business was depressed because of being obliged to use low quality or high cost raw materials, if he had consistently followed a contrary course before the year in question and followed it again thereafter, at least when these 'temporary economic circumstances' were not ones for which he was at fault-- which the Government has not here asserted.

The Bulletin issued by the Treasury with respect to 722 of the World War II Excess Profits Tax, at p. 18, quoted in 7A Mertens, supra, 42.121 n. 42 at p. 726, gave as an example of a taxpayer entitled to relief because of temporary economic circumstances a brush company, normally importing its bristles from China, which was obliged to seek them elsewhere and accept inferior substitutes at higher prices because of the Japanese invasion of China in 1937. While in that case there was an 'event,' 422(a)(2) does not require one. Presumably the taxpayer in the case put in the Bulletin would have fared equally well if the Chinese producers had simply decided to sell elsewhere but an alternate source of supply was in sight a few years hence.

We have had some concern whether the business of U.S. Steel could be said to have been 'depressed' in 1947 or 1948. The annual reports received in evidence at the trial on the coal-strike issue show income after taxes of $127,100,000 in 1947 and $129,600,000 in 1948, which were the highest since 1929.30 As against this, however, the Bulletin issued in connection with the World War II Act said, p. 16, that 'In order for the depression of the business of a taxpayer to come under 722(b)(2), it is not necessary that the taxpayer's earnings be reduced to any particular level, nor even that they be less than the average for the taxpayer's industry. It is necessary only that they be less than they normally would have been for the taxpayer in the absence of the temporary economic circumstances shown.' 7A Mertens, supra, 42.121 at 720. While the Senate Finance Committee report on the World War II Act, quoted at n. 10 on the same page, says that 'As a general rule, high costs of production because of high costs of material, labor, capital, or other elements' would not qualify, this scarcely covers the case of temporary high costs unusual to a particular taxpayer and not incurred by the industry generally. Furthermore, despite U.S. Steel's relatively large profits in 1947 and 1948, a circumstance taken into account at least in part by Congress' decision to allow only 83% Of base year profits, whether actual or constructive, as a credit against the Korean excess profits tax in contrast to the higher percentage permitted for World War II, there are indications summarized in the margin,31 that something may have been going wrong at U.S. Steel during the base period in comparison to the rest of the industry.

The Government comes up with a number of other arguments which do not impress us. It says the taxpayer can point to no precedent where relief has been accorded in such a situation as this; the taxpayer makes the rather obvious response that the Government has cited none where it has been denied. The Government argues that the increase in capacity from the new beneficiating plants did not measure up to the quantitative tests for increase in capacity for production specified in 444. Taxpayer answers that it was the inadequacy of that and other special relief provisions that led Congress to enact 422. The Government points to 450, wholly exempting from excess profits tax the portion of adjusted excess profits net income attributable to the mining of certain strategic minerals, the bulk of which had previously been imported. U.S. Steel answers that this extraordinary measure of relief does not negate the right to a proper credit for the producer of such old-fashioned but essential substances as coal, iron and steel.

In the final analysis, the difference between the Government and the taxpayer on this issue is in approach. The Government, in a commendable desire to protect the revenue and curtail further litigation (although one would suppose that Korean Excess Profits Tax litigation must be nearing its end), would read the general relief provisions with peculiar, almost eviscerating, narrowness. On its view an 'event' must be a catastrophe affecting a particular taxpayer and perhaps a few others; 'unusual' must be not merely unusual in the ordinary sense but unexpectable. As in Oxford Paper Co. v. C.I.R., 302 F.2d 674, 678 (2 Cir. 1962), 'we find no basis for the glosses' the Government 'would impose upon what seems the straightforward language of the statute.' While the general 'relief provisions are not considered to be a means of conferring general equitable relief on taxpayers, or to confer relief just because base period earnings of a taxpayer may be low or depressed,' still 'it is incorrect and unrealistic to say that equities play no part in the decision of such cases. Their weight is not to be ignored, even if it cannot always be measured, and it is well within the legislative intent that they be viewed sympathetically in relation to the purposes intended to be advanced by these relief provisions.' 7A Mertens, supra, 42.117 at 632.32 Whether U.S. Steel can show at a trial that World Was II in fact created unusual problems that led to diminution of 'normal production output or operation' in 1947 or 1948, or that its business during those years 'was depressed because of temporary economic circumstances unusual' in its experience, we cannot now say. We hold only that summary judgment should not have been rendered against it.³³

This opinion would be incomplete if we failed to pay tribute to the quality and quantity of effort devoted by Senior Judge Levet to the resolution of this difficult case or, more accurately, cases. We also express our gratitude for the helpful briefs and argument of counsel for both parties.

The orders with respect to the depletion, coal strike, and cost-price issues are affirmed. The order granting summary judgment to the Government on the materials issue is reversed and the case remanded for further proceedings consistent with this opinion.34 No costs.