Arkwright Mills v. Commissioner

SOPER, Circuit Judge. This petition for review of a decision of the United States Processing Tax Board of Review involves a claim for refund of money paid as a processing tax under the Agricultural Adjustment Act of May 12, 1933, 48 Stat. 31, 7 U.S.C.A. § 601 et seq. The question is whether the. Board was correct in finding that the taxpayer was not — 466 — entitled to any refund 'because the taxpayer had not established that it had borne the burden of the tax. The taxpayer is a South Carolina corporation with its principal place of business at Spartanburg. It was engaged in the first domestic processing of cotton. It paid the sum of $257,415.97 in processing taxes on the cotton processed at the rate of 4.2c per pound. It seeks to recover a part of this sum, that is, a minimum of $79,152.17, on the ground that it bore the burden of the tax to at least that extent. The taxing statute was held unconstitutional on January 6, 1936, by the decision of the Supreme Court in United States v. Butler, 297 U.S. 1, 56 S.Ct. 312, 80 L.Ed. 477, 102 A.L.R. 914. See, also Rickert Rice Mills v. Fontenot, 297 U.S. 110, 56 S.Ct. 374, 80 L.Ed. 513. The statute was repealed by Section 901 of the Revenue Act of 1936, 49 Stat. 1648, 1747, 7 U.S.C.A. § 623 note, which was held constitutional in Anniston Mfg. Co. v. Davis, 301 U.S. 337, 57 S.Ct. 816, 81 L.Ed. 1143. The repealing statute prescribes the conditions in allowance of refunds of the tax. Section 902, 7 U.S.C.A. § 644, provides that no refund shall be allowed unless the claimant proves that he bore the burden of the tax and has not been relieved' thereof nor reimbursed therefor, not shifted the burden directly of indirectly in any manner whatsoever. Section 906, 7 U.S.C.A. § 648, forbids any suit or proceeding for the refund of the tax except a proceeding before a Board of Review established in the Treasury Department, and confers jurisdiction upon the Board to review the action of the Commissioner of Internal Revenue upon a claim for refund and to determine the amount of the refund. The Circuit Courts of Appeals are empowered to review the decisions of the Board and modify or reverse them if not in accordance with law. Section 907, 7 U.S.C.A. § 649, makes the following provision as to evidence and presumptions : “Sec. 907. Evidence and presumptions “(a) Where the refund claimed is for an amount paid or collected as processing tax, as defined herein, it shall be primafacie evidence that the burden of such amount was borne by the claimant to the extent (not to exceed the amount of the tax) that the average margin per unit of the commodity processed was lower during the tax period than the average margin was during the period before and after the tax. If the average margin during the tax period was not lower, it shall be prima-facie evidence that none of the burden of such amount was borne by the claimant but that it was shifted to others.” Section 907(b) provides that the average margins for these periods shall be the average of the margins for the months within the periods; that the margin for each month in the tax period shall be computed by deducting the cost of the commodity processed plus the tax from the gross sales value thereof; and that the margin for each month in the period before and after the tax period shall be computed by deducting the cost of the commodity processed from the gross sales value thereof. The cost of the commodity processed is the actual cost and the gross sales value is the quantity sold multiplied by the 'current sales price.. Section '907(c) provides that the tax period is the period during which the taxpayer paid the tax and the “period before and after the tax” is the twenty-four months preceding the effective date of the tax, and the six months, February to July, 1936. Section 907(e) is in part as follows: “(e) Either the claimant or the Commissioner may rebut the presumption established by subsection (a) of this section by proof of the actual extent to which the claimant shifted to others the burden of the processing tax. Such proof may include, but shall not be limited to— “(1) Proof that the difference or lack of difference between the average margin for the tax period and the average margin for the period before and after the tax was due to changes in factors other than the tax. * * * ” The tax period in the pending case was from August 1, 1933 to February 28, 1935, except September, 1934, when the taxpayer’s plant was idle. The period before and after the tax was from August 1, 1931 to July 31, 1933, and from February to July, 1936. The Board found that from the earliest to the latest of these dates the taxpayer purchased cotton in the open market at prevailing prices, processed it and sold the manufactured goods for the highest price it could obtain. The average margin per unit of the commodity processed during the tax period was approxi — 467 — mately 11.5c, and the average margin per unit for the period before and after the tax was approximately 8c. The taxpayer's expenditures, exclusive of cotton, were approximately 10c per unit of the commodity processed during the tax period and approximately 6.8c per unit during the period before and after the tax. The Board also found that in determining the cost of the commodity processed during the tax period, the petitioner treated the processing tax as one of the elements entering into such cost. On contracts of sale entered into prior to August 1, 1933, but on which delivery was made after August 1st, the processing tax was billed separately. On contracts of sale entered into after August 1, 1933, the goods were invoiced at a lump sum price and in about 1% of such contracts there appeared after the price the notation “including processing tax”, but in such instances the selling price was the same as when this phrase was omitted. The final decision of the Board was that no part of the burden of the amount paid or collected as tax with respect to its processing of cotton was borne by the petitioner, but such burden was shifted to others in its entirety. It thus appears that the average margin per unit during the tax period exceeded the average margin in the base period before and after the tax, and therefore in conformity with Section 907(a) the presumption is that none of the burden of the tax was borne by the taxpayer but that it was all shifted to others. As we have seen, there was an increase of expenditures during the tax period of 3.2c per unit over the period before and after the tax, but it is agreed that even if this is taken into account, it fails to overcome the presumption. The taxpayer, in order to overcome the presumption, offered the testimony of witnesses to show that one cause of the increased margin in the tax period was a greater demand for its goods which could be expressed in dollars; and that even if there had been no tax, this greater demand would have existed and would have increased the taxpayer’s profits. This testimony was rejected by the Board. The entire discussion of the Board upon this crucial point is found in the following paragraphs of its opinion by member John W. Edwards: “This testimony was thus directed in an endeavor to ascribe a value to the increased demand for the petitioner’s goods which existed in the tax period, and involved the consideration and measurement of so many varying, intangible factors that it is believed such opinions amounted to no more than conjecture. The effort to establish what the petitioner’s margin would have been in the absence of the processing tax is an attempt to prove what would have occurred in an imaginary and nonexistent market. “Section 907(e) provides that the presumption established by section 907(a) may be rebutted ‘by proof of the actual extent to which the claimant shifted to others the burden of the processing tax.’ Actualities — not hypotheses — are thus prescribed by the statute as necessary to overcome the presumed extent to which the claimant shifted the burden of the tax. The testimony in this case departs from the actualities thus contemplated by the statute and seeks to overcome the statutory presumption by speculation as to what would have occurred in an imaginary world. This is not believed to be the type of proof authorized by section 907(e) to be used in rebuttal of the presumption, or as fulfilling the conditions on allowance of refunds prescribed by section 902 of the statute.” A dissenting opinion which sharply defines the issue under consideration was filed in the following terms: “Matthews dissenting: In my opinion petitioner proved through its expert witnesses that the increased demand in the tax period was a factor other than the tax which accounted for a part of the margin for that period, and by computations based on sound statistical analyses determined the measure of such factor. Petitioner’s witnesses were economists and statisticians of unquestioned ability. One of such witnesses has had years of experience in analyzing statistical data and has on many occasions appeared as an expert witness for the Government. The respondent did not on cross-examination or by the introduction of other evidence attempt to show that the statistical method used by petitioner’s witnesses was not valid. On the evidence presented, therefore, petitioner has shown that it bore the burden of the tax at least to the extent of $79,-152.17.” We understand the Board to hold that the evidence as to change in demand in the periods chosen by Congress for — 468 — comparison should he ignored because the evidence represents merely an attempt to show what the taxpayer’s margin would have been during the tax period if there had been no tax; and because this effort involves the setting up of a hypothetical or imaginary situation which has no tendency to prove the actual extent to which the taxpayer bore the burden of the tax. This point of view in our opinion cannot be maintained. In the first place it is manifest that the statutory test, which Congress itself devised, involves a comparison of the taxpayer’s' actual position during the tax period with what it would have been if there had been no tax.. Section 907(a) requires a comparison of the margin in the tax period with the margin in the base period before and after the tax. If the margin is lower in the tax period, the presumption is that- the taxpayer has borne at least a part of the tax; but if it is not lower, the presumption is that the taxpayer has paid no part of the tax. Qearly, this test assumes that all the factors except the tax which affect the margins are the same in both periods, for otherwise the comparison would not show whether the taxpayer bore the burden or not; and this assumption involves the very sort of hypothesis which the Board condemns. That this is true is recognized by high authority. See Statement of Secretary of Agriculture Wallace before Finance Committee on Title VII of the Revenue Act of 1936, Hearings before Committee on Finance, United States Senate, 74th Congress 2nd Session, on H.R. 12, 395, p. 859; Ferger on Windfall Tax and Processing Tax Refund, American Economic Review, March, 1937, p. 52. Moreover, there is no novelty or vice in the hypothetical approach to the ascertainment of a person’s rights or the extent of damage wrongfully imposed upon him. Valuation of property involves the ascertainment of a hypothetical selling price. Breach of contract often involves a hypothetical loss of profits. Personal injury often requires an estimate of earnings that would have been made had the injury not occurred. Frequently the investigation of these matters is of such difficulty that mathematical certainty in the determination of amounts is impossible; but rights are not lost thereby, and it becomes the 'duty of the courts to find the relevant facts and to base a reasoned judgment upon' them. See Palmer v. Connecticut Railway & Lighting Co., 312 U.S. 713, 61 S.Ct. 609, 85 L.Ed. 1143. Of course care must be taken to give consideration to all the factors that affect the calculation to be made. Congress recognized this fact by giving only prima facie value to the presumption described in Section 907(a) ; and by permitting the parties to offer proof that the actual incidence of the tax differed from that indicated by the presumption. Such a qualification of the statutory method was necessary, for otherwise the requirements of due process would not have been met and the constitutionality of the statute would have been' endangéred; and the Act has indeed bée'n interpreted to afford a “full opportunity to the claimant to present any evidence which may be pertinent to the questions to be determined by the Board of Review and which may be appropriate to overcome any presumption which might be indulged either under Section 907(a), * * * or otherwise.” Anniston Mfg. Co. v. Davis, 301 U.S. 337, 351, 355, 356, 57 S.Ct. 816, 825, 81 L.Ed. 1143. In Epstein v. Helvering, Commissioner, 4 Cir., 120 F.2d 427, it was held that the statutory presumption could be rebutted by showing that a smaller margin in the post tax period was caused by a substantial change in the character of the taxpayer’s business which then took place; and in Cones & Son Mfg. Co. v. United States, 7 Cir., 123 F.2d 530, it was held permissible to show that a'larger margin in the tax period was due to the realization by the taxpayer during that period of the increased market value of raw cotton on hand on August 1, 1933, over and above its original cost. In the pending case, the taxpayer offered evidence tending to show the effect of an increased demand for cotton in the tax period. This evidence was given in part by economic and statistical experts and in part by one with business experience in the textile industry. ' According to the latter’s testimony, the experience of men in the cotton industry indicates that there is a greater margin of profit in the manufacture and sale of textiles when the price of cotton is relatively high, so that textile prices rise faster on a rising cotton market than- the price of cotton, while on a falling' cotton- market the price of textiles falls faster than the price of cotton. This 'evidence'-also tended to show that the demand' 'for goods was much'- less in the — 469 — pre-tax period of twenty-four months from August 1, 1931 to July 31, 1933, when the full effect of the financial depression was realized, than it was in the tax period" from August 1, 1933 to February 1, 1936, and in the post-tax period from February to July, 1936, when the recovery program of the federal government had begun to exert its influence. The experts gave corroborating testimony, basing their opinions upon economic and statistical data. They concluded that a correct determination of the merits of the taxpayer’s claim could not be reached by the use of the statutory formula alone.. Accordingly, they made an effort to find a period similar to the tax period in respect to demand and other influential factors. They found such a period equal in length to the statutory base period before and after the tax, in the thirty months from January 1, 1936 to June 30, 1938, during which the average cost of cotton was almost the same and the manufacturing costs were, for all practical purposes, identical to those in the tax period. A study was also made of the relationship of the price of cotton and the price of manufactured goods; and it was testified that during periods in which manufacturing costs are approximately the same, there is a constant and dependable ratio, between the selling price of cotton cloth and the cost of raw cotton. This ratio differs from the marginal comparison prescribed in Section 907, for in that comparison the assumption is that there is a constant margin between the cost of the cotton and the selling price of the cloth. For example, if cotton costs 8c per pound and cloth sells for 16c per pound, the assumption is that if the price of cotton goes to 10c, the price of cloth will go to 18c. " The finding of the experts testifying for the claimant, on the other hand, was that the margin is not constant but proportional, so that in the example given, where the selling price of the cloth is twice that of the cost of cotton, then if the cost of cotton rises 2c, the price of cloth rises 4c or twice as much. This economic tendency, according to the evidence, was verified, in the light of data on cotton and cloth prices published by the United States Department of Agriculture, by tests' which covered the period from February 1, 1936 to February 1, 1940, with respect to seventeen varieties of cotton cloth. The resulting expert opinion was that a definite ratio exists between cotton and cloth prices, and that the influence of the factor of demand upon margins is so great that it cannot be left out in any calculation or comparison designed to determine the incidence of the tax. The next step was to apply these conclusions to the facts of the case by adjusting the statutory formula for marginal comparison so as to make allowance for the greater intensity of demand that existed in the tax period as compared with the statutory base period. The demand was the same in the tax period as it was in the thirty month period, and, therefore, the demand in the latter period was compared with the demand in the "statutory base period. This was done by comparing the net margins, i. e., the gross margins, less manufacturing costs, in the two periods. The net margin in the statutory base period was subtracted from the net margin in the thirty month period, and the difference was attributed to the greater demand in the last mentioned period. By this method the amount by which the greater demand increased the margin in the tax period was calculated, and tfie opinion was finally formed that the taxpayer bore the burden of the tax in the minimum sum of $79,152.17. This expert testimony, as we have seen, was not controverted. The reliability of the statistical data upon which the conclusions were based was not questioned. The method employed in drawing the conclusions from the data, and in applying them to the facts of the case, was not challenged. The evidence was rejected simply on the ground that in the judgment of the Board it was not the type of proof required by the statute to rebut' the statutory presumption. In this action there was error. The statute permits the use of any evidence which is pertinent to the questions to be determined. The evidence offered was pertinent and possessed of probative force, and would have justified a finding in the claimant’s favor. It is true that it is the Board’s function to determine the weight and credibility of the evidence offered by the claimants; but in this instance the Board, acting under a mistaken view of its duties under the Act, failed to give to the evidence that consideration which it deserves. The decision of the Board must therefore be reversed, and the case — 470 — remanded for rehearing, at which either party will be at liberty to produce additional testimony. Reversed and remanded.