GREAT NORTHERN RAILWAY CO. v. WEEKS, STATE TAX COMMISSIONER, ET AL.
No. 178
Supreme Court of the United States
Argued January 6, 7, 1936. Decided February 3, 1936.
297 U.S. 135
The contention that the new regulation is retroactive is without merit. Since the original regulation could not be applied, the amended regulation in effect became the primary and controlling rule in respect of the situation presented. It pointed the way, for the first time, for correctly applying the antecedent statute to a situation which arose under the statute. See Titsworth v. Commissioner, 73 F. (2d) 385, 386. The statute defines the rights of the taxpayer and fixes a standard by which such rights are to be measured. The regulation constitutes only a step in the administrative process. It does not, and could not, alter the statute. It is no more retroactive in its operation than is a judicial determination construing and applying a statute to a case in hand.
Judgment affirmed.
Mr. Harold D. Shaft, with whom Mr. P. O. Sathre, Attorney General of North Dakota, was on the brief, for respondents.
MR. JUSTICE BUTLER delivered the opinion of the Court.
This suit was brought in the federal district court for North Dakota by petitioner against the state tax commissioner and the auditors and treasurers of 30 counties, to enjoin collection of about 40% of 1933 taxes laid upon its railroad properties in each county. The assessed value of all petitioner‘s railroad property in the State is $78,832,888. The total of the tax is $1,508,352.34, of which petitioner has paid about 60%. The suit is grounded upon the claim that the taxes are based on a valuation that
The state law requires that all property subject to taxation be assessed at its true and full value in money.
Petitioner claims that the board made the assessment by attributing to North Dakota too great a proportion of a grossly excessive system valuation. More specifically its contentions are: (1) That, by reason of the methods employed for the ascertainment of the percentage of system value to be assigned to North Dakota, the assessment
The full and true value of the property is the amount that the owner would be entitled to receive as just compensation upon a taking of that property by the State or the United States in the exertion of the power of eminent domain. That value is the equivalent of the property, in money paid at the time of the taking. Olson v. United States, 292 U. S. 246, 254. The principles governing the ascertainment of value for the purposes of taxation, are the same as those that control in condemnation cases, confiscation cases and generally in controversies involving the ascertainment of just compensation. West v. C. & P. Telephone Co., 295 U. S. 662, 671.
In determining the amount of the assessment the board was not bound by any formula, rule or method, but for guidance to right judgment it was free to consider all pertinent facts, estimates and forecasts and to give to them such weight as reasonably they might be deemed to have. Courts decline to disturb assessments for taxation unless shown clearly to transgress reasonable limits. Overvaluation is not of itself sufficient to warrant injunction against any part of the taxes based on the challenged assessment; mere error of judgment is not enough; there must be something that in legal effect is the equivalent of intention or fraudulent purpose to overvalue the property and so to set at naught fundamental principles that safeguard the taxpayer‘s rights and property. Rowley v. Chicago & N. W. Ry., 293 U. S. 102, 109-111. The assessment is presumed to have been rightly made on the basis of actual value. Its validity must be tested upon consideration of the facts established by the evidence and of those of which judicial notice may be taken.
As to methods employed by the board. Respondents called as a witness Mr. Lyman A. Baker, who had been with the North Dakota tax commissioner for 19 years and during the last 13 years, ending January 1, 1933, had been deputy commissioner in charge of the valuation of railroads and other utilities. He was engaged in this litigation in behalf of respondents and spent much time in making computations and in the preparation of exhibits that were put in evidence by respondents. In substance he testified:
The first step in the valuation of railroad property within a State is to determine the value of the entire system. There are two classes of evidence ordinarily considered: The average market price of stock and bonds, and the past earnings over a period of years. As the stock and bond prices reflect value of the entire railroad, it is necessary to eliminate the value of non-operating property. The method requires a definite period over which to average price quotations and that must of necessity be somewhat arbitrarily fixed. It assumes that the average price reflects value, but rarely is controlling interest bought or sold on the exchange. Where control is sought prices advance sharply. The method also assumes that purchasers act on accurate knowledge of conditions; it ignores the influence of pure speculation. In applying the method, taxing boards, economists and railroad men have always adopted five-year periods immediately pre-
He further testified: Capitalization of net railroad operating income is generally recognized as an important element in estimating the value of railroad operating property. The average net income, usually for five years, is capitalized at a reasonable rate of return. The method assumes the amount so ascertained to be the value of the property. The income of a single year is seldom, if ever, used. As 1931 and 1932 were the worst years in railroad history since the panic of 1893, the use of the three-year period ending in 1932 has but little justification. The five-year period has been given the same weight by the state board for a good many years. It is generally considered that the rate of return that a company is allowed to earn under state and federal law is a fair rate to use. A rate of six percent. is justifiable under present conditions. Economic return from farm property, being about 65% of all that is taxable in North Dakota, decreased about 75% from the 1924-1928 average. Assessments on values indicated by capitalization of average net income for the five years ending in 1932 would result in giving preferential treatment to railroads as compared with other properties. Where properties have been operated at a loss over a period of years, there are no earnings to capitalize. Yet they have present and prospective value which would be reflected by use of the stock and bond valuation method. The criticisms made of these methods are not sufficient to render them valueless. “On the contrary the two methods are universally approved as the two best evidences of the value of a railroad which are available.”
The average of the valuations computed for 1932 as alleged in the answer is $78,735,110; the assessment was $78,850,024. The difference is less than one-sixth of one per cent. The trial court found that for 1933 the board assessed the property at “the sum of $78,832,888, which sum was the same as the 1932 assessment except for a
It is clear that the computations prepared by the tax commissioner for the use of the board served as guides to, if indeed they were not used as the measure of, value for the assessments made by the board for 1932 and 1933. Respondents’ answer and evidence require a finding that these computations were in fact the controlling bases and constituted the very foundation of the 1932 and 1933 assessments. In view of the relatively small differences between the 1933 assessment and the figures produced by the methods approved by respondents and universally as the best available evidence of value of a railroad, the latter rightly may be used in testing the validity of the former.
Petitioner‘s claim that the board‘s apportionment of system value to North Dakota operated to assess and tax property in other States cannot be upheld.
Petitioner maintains that the amount attributed to that State was found by the use of the factors above referred to. More fully described, they are: 1. Miles of all track (as of December 31, 1931)—20.19%. 2. Physical property as measured by cost of reproduction less depreciation (as of December 31, 1931)—13.84%. 3. Car and locomotive miles (average for five years ending with 1931)—19.90%. 4. Ton and passenger miles (average for five years ending with 1931)—18.65%. 5. Gross earnings (average for five years ending with 1931)—18.45%; the amount assigned to North Dakota was found by adding to the intrastate revenue earned in that State its mileage proportion of revenue derived from traffic moving partly within and partly outside her boundaries. The average of these percentages is 18.206%.
The principal grounds on which petitioner assails the board‘s apportionment are: Petitioner‘s railroad in North Dakota consists largely of relatively cheaper branch lines.
The problem of apportionment is a difficult one. It is impossible to formulate a rule generally applicable. Controlling conditions vary greatly from time to time. Allocations to be sufficiently accurate for practical purposes must be arrived at by the exercise of sound judgment based on facts that fairly reflect the relation between value of the system as a whole and value of the part within a State. Rowley v. Chicago & N. W. Ry., supra, 109, 110.
The methods proposed by petitioner discredit its objections to the apportionment made by the board. While the board had the 1933 assessment under consideration, petitioner submitted to it a brief in which it proposed two methods of apportionment. One was by use of the above described factors 1, 3, 4 and 5, and the other by use of 3, 4 and 5, producing, respectively, as calculated by petitioner, 19.35% and 19.14%. These percentages are higher than the average of those that petitioner says the board used. There is no evidence that the assessment was arrived at by use of apportioning percentages higher than those then used by petitioner.
Petitioner‘s bill suggests a method of apportionment that would assign to North Dakota less than 12½% of system value. Briefly, the method is this: To the revenue derived from North Dakota intrastate traffic add a part of the revenue received from the interstate moving in North Dakota that is equal to its proportion of the property used in, and the cost of, carrying that traffic; deduct the operating expenses incurred in moving it from the total revenue. The relation of the state net earnings
And here petitioner argues that the apportionment should be made on the basis of physical property, i. e. factor 2 above described. That would assign to North Dakota less than 15% of the system value. Petitioner did not submit it to the board as the measure or even include it in the group of factors upon which it based the calculations included in its brief. It is not sustained by evidence.
When regard is had to the size of the Great Northern system and the variety of things that affect values to be attributed to its railroad in different States, and the numerous matters as to which there may be wide difference of opinion, it must be held that percentages lower than, or substantially the same as, those petitioner itself used and submitted to the board are not confiscatory or arbitrary. Helvering v. Taylor, 293 U. S. 507, has no bearing.
There remains to consider whether the assessment may be sustained against petitioner‘s claim that, assuming validity of apportionment, the valuation was arbitrarily made and is grossly excessive.
No testimony was given by the tax commissioner or any other member of the board. They could not be compelled to submit to examination as to the operation of their minds in making the challenged assessment. Chicago, B. & Q. Ry. Co. v. Babcock, 204 U. S. 585, 593. Bohler v. Callaway, 267 U. S. 479, 491. It results that resort must be had to determinations of the board, find-
From the answer and evidence it unquestionably appears that the board valued petitioner‘s system and made apportionment to North Dakota. The assessments for the years 1929-1933 are: $83,294,677, $83,294,688, $82,999,997, $78,850,024 and $78,832,888. The trial judge found no overassessment for 1932 but refused to find the value of the property in 1933 or whether the assessment for that year was excessive; and he found that there is no evidence, other than that offered to show overvaluation, that the assessment was fraudulent or made with intent to defraud petitioner. The Circuit Court of Appeals held that petitioner failed to prove the method employed by the board to make the assessment or that it was grossly excessive and arbitrary.
Respondents’ witness prepared tabulations which were put in evidence as Exhibit P. It shows: 1. The value for each year, based on averages for preceding five years, from 1929 to 1933, inclusive, of the system operating property as indicated by: (a) stock and bond value less value of non-operating property; (b) net operating income capitalized at six per cent; (c) the average of (a) and (b). 2. Apportionment of system value to North Dakota by percentages reflected by average of: (a) three use factors, i. e. transportation service, traffic units and gross operating revenue; (b) mileage of tracks operated and the three use factors; (c) mileage of tracks owned and operated and leased and operated and the three use factors; (d) mileage of track operated, physical property and the three use factors; (e) mileage of track operated, physical property, the three use factors and net revenue. 3. Apportionment to North Dakota of: (a) stock and bond value; (b) capitalized value; (c) average of (a) and (b). 4. The North Dakota assessments. Ratio of assessments to apportionments made on (a) stock and
The composite of the five year average of stock and bond value and of net operating income capitalized for 1932 was $415,278,961; admittedly that was produced by the use of the two methods shown by respondents’ witness to be universally approved as the best methods for finding system value. The 1932 assessment is 99.93% of that figure apportioned on the basis of the three use fac-
If the system value for 1933 had been computed on the basis of the stock and bond and capitalized income methods used for 1932, it would have been $345,188,820, about 83% of the corresponding figure for 1932, less than 76% of like figures for 1931 and 1930 and about 79% of that for 1929. See footnote 2. The 1933 assessment exceeds what would have been made on system valuation based, on the five-year average of stock and bond values, apportioned by use of the five factors above described (advocated by the State as “the fairest basis of allocation“) by 19.97%, exceeds that based on net operating income capitalized by 29.44%, on the composite of both by 24.52%. The testimony and computations made by respondents’ witness show that the 1933 assessment could not have
| 1 | 2 | 3 | 4 | 5 | 6 | 7 | |
|---|---|---|---|---|---|---|---|
| Apportionment to North Dakota on basis of mileage of tracks owned and operated and leased and operated and the three use factors—Continued. | |||||||
| 1931 | 455,777,674 | 19.673 | 89,665,142 | 82,999,997 | 99.41 | 86.60 | 92.57 |
| 1932 | 415,278,961 | 19.775 | 82,121,415 | 78,850,024 | 98.35 | 93.79 | 96.02 |
| 1933 | 345,188,820 | 19.812 | 68,388,809 | 78,832,888 | 111.05 | 119.82 | 115.27 |
| Apportionment to North Dakota on basis of trackage operated, physical property and the three use factors. | |||||||
| 1929 | 437,789,980 | 18.05 | 79,021,091 | 83,294,677 | 116.91 | 95.97 | 105.41 |
| 1930 | 460,989,128 | 18.102 | 83,448,252 | 83,294,688 | 111.25 | 90.51 | 99.82 |
| 1931 | 455,777,674 | 18.136 | 82,659,839 | 82,999,997 | 107.84 | 93.94 | 100.41 |
| 1932 | 415,278,961 | 18.206 | 75,605,688 | 78,850,024 | 106.82 | 101.88 | 104.29 |
| 1933 | 345,188,820 | 18.34 | 63,307,629 | 78,832,888 | 119.97 | 129.44 | 124.52 |
| Apportionment to North Dakota on basis of trackage operated, physical property, the three use factors and net revenue. | |||||||
| 1929 | 437,789,980 | 18.783 | 82,230,092 | 83,294,677 | 112.34 | 92.22 | 101.29 |
| 1930 | 460,989,128 | 18.773 | 86,541,489 | 83,294,688 | 107.27 | 87.28 | 96.25 |
| 1931 | 455,777,674 | 18.710 | 85,276,003 | 82,999,997 | 104.53 | 91.06 | 97.33 |
| 1932 | 415,278,961 | 18.725 | 77,760,985 | 78,850,024 | 103.86 | 99.05 | 101.40 |
| 1933 | 345,188,820 | 18.763 | 64,767,778 | 78,832,888 | 117.26 | 126.52 | 121.72 |
The long period through which, even in 1933, the depression had extended compelled the conclusion that it was not temporary. Judicial notice must be taken of the fact that late in 1929 there occurred a great collapse of values of all classes of property—railroads, other utilities, commodities and securities, and that the depression then commenced progressively became greater. In making assessments in that period, the board was bound to take into account and give due weight to the sudden, progressive and enormous declines of value.
Respondents’ witness testified: The purchasing power of money greatly increased and correspondingly values decreased from 1929 to 1933. The Dow-Jones & Company average of 30 industrial stocks fell from 383.17 in September, 1929, to 41.22 in July, 1932. The average of 20 railroad stocks fell between the same dates from 189.11 to 13.23; total market value of all common and preferred stocks listed on the New York Stock Exchange fell from eighty-nine and one-half billion to fifteen and one-half billion dollars. Assessing officials and equalizing boards were confronted with a very difficult situation.
Changed business conditions affecting petitioner‘s traffic, coupled with competition from new methods of transportation, precluded belief that prospective improvement in petitioner‘s business and earnings would within a reasonable time, if ever, be sufficient to justify the assessment in question. Cf. Nashville, C. & St. L. Ry. v. Walters, 294 U. S. 405, 423. But from 1929 to and including 1933 the board reduced assessments of petitioner‘s North Dakota railroad properties by less than six per cent. It is everywhere known that the general decline in values in that period was very much greater than that. The evidence conclusively shows that the value of petitioner‘s system and of its North Dakota railroad properties declined several times six per cent. Its traffic, gross earnings and net income from operation fell off enormously.3 The 1929 collapse and the decline progressively following it resulted in much lower levels of prices and values which at least as early as 1933 were to be regarded
In cases such as this, courts are not permitted to weigh evidence of value. They may not substitute their opinions for the findings of assessing officers or boards. But, when the jurisdiction of the district court is appropriately invoked, it is its duty to decide upon the merits of the taxpayer‘s claim that the assessment of his property was arbitrarily made and is grossly excessive. It clearly appears that the board failed to give reasonable weight to the falling off of petitioner‘s traffic, gross earnings, operating income, the extraordinary shrinkage in values of railroad properties, the prices of commodities and securities generally. The value of petitioner‘s property varied with the profitableness of its use, present and prospective. Cleveland, C., C. & St. L. Ry. Co. v. Backus, 154 U. S. 439, 445. Southern Ry. Co. v. Kentucky, 274 U. S. 76, 81-82. Petitioner‘s system net operating income was for 1929 in round figures $32,457,000; in the following years $21,912,000, $12,669,000, $1,290,000. The board persistently disregarded known conditions essential to the just ascertainment of value. While the vanishing of values described by respondents’ witness, the reduction of the tax base from 75% to 50%, and the established limitations upon rates of taxation, justify diligence on the part of the assessing authorities that taxable property be assessed at full value, neither these nor any other conditions warrant or excuse arbitrary or excessive valuations.
The facts alleged in respondents’ answer, and those shown by the testimony of their witness and his computations above described, compel the conclusion that, by reason of changed conditions affecting value, the methods or system by which the board arrived at the 1933 value of petitioner‘s railway as a whole were plainly calculated
The judgment of the Circuit Court of Appeals will be reversed. The case will be remanded to the district court with directions to enter a decree for petitioner, the plaintiff below, that respondents, the defendants below, be
Reversed.
MR. JUSTICE STONE, dissenting.
I think the judgment should be affirmed.
The decision of the Court turns on the constitutionality of the valuation, for 1933 taxation, of so much of petitioner‘s railroad as is located in North Dakota.
The Court finds that the valuation of the railroad within the state is not so disproportionate to the value of petitioner‘s entire railroad system as to transcend due process. See Fargo v. Hart, 193 U. S. 490, 500; compare Rowley v. Chicago & Northwestern Ry., 293 U. S. 102, 109-111. It does not find, and there is no contention, that there has been any discrimination in the valuation of petitioner‘s property as compared with that of other property in the state. Its decision that the tax is invalid rests on the single ground that the valuation is excessive.
This conclusion is based on an elaborate examination of the evidence produced before the trial court, evidence which it is assumed affords the only basis for the valuation of the Board of Equalization. Emphasizing as the important, if not controlling factors, in determining taxable value, the depressed market value of the securities of the entire railroad, representing its property in many states, its diminished earnings, its capitalized value based
We may lay aside any consideration of the numerous uncertain and imponderable elements involved in valuation of a railroad which may well make the use of such a formula untrustworthy in times like the present, see Rowley v. Chicago & Northwestern Ry., supra, 109; which would seem to make it impossible for a court to say that the rejection of the results of such a formula by the taxing officials involved anything more than the exercise of an authorized judgment, which courts cannot pronounce arbitrary merely because it does not conform to their own.
The feature of the decision which is especially a matter of concern is that for the first time this Court is setting aside a tax as a violation of the
Assessment for taxation is a quasi-judicial act and the tax assessment has the quality of a judgment. Hagar v. Reclamation District, 111 U. S. 701, 709; Gallup v. Schmidt, 183 U. S. 300; Londoner v. Denver, 210 U. S. 373, 386; Turner v. Wade, 254 U. S. 64. Even if the valuation of the Board be erroneous, the errors of a state
It has long been recognized that discrimination between taxpayers, if intentional or so persistent as to be systematic, is a denial of equal protection, whether the discrimination is in the application of different rates to property in the same class or in inequality in its valuation. Iowa-Des Moines Bank v. Bennett, 284 U. S. 239, 245; Cumberland Coal Co. v. Board of Review, 284 U. S. 23, 25ff; Chicago G. W. Ry. Co. v. Kendall, 266 U. S. 94, 98; 99; Sioux City Bridge Co. v. Dakota County, 260 U. S. 441, 445; Raymond v. Chicago Traction Co., 207 U. S. 20, 37. But to hold that a tax is unconstitutional because based upon an assessment which is too high, as compared with the value of the same property for purposes of condemnation, overlooks the principle upon which property taxes are laid and collected. Taxation is but a method of raising revenue to defray the expenses of government, and of distributing the burden among those who must bear it. The taxpayer cannot complain of the tax burden which he has to bear, who shows no inequality in the application of it. And plainly he does not show inequality merely by proving that the valuation of his property for taxation is much higher than its market or its condemnation value.
The burden of a property tax like the present is distributed by applying a rate of tax to the assessed valuation of all taxable property. Variation of either, without
Recently we held that a claim that the rate of a nondiscriminatory tax is excessively high presents no constitutional question. Magnano Co. v. Hamilton, 292 U. S. 40, 44. No reason has been advanced at the bar, or given in the opinion of the Court, why a tax valuation, excessive when compared with condemnation or market value, should have any different legal consequences. In neither case is inequality of the tax burden established. It is for that reason that this Court has held, without exception, that valuation of property for tax purposes, however excessive, not shown to be discriminatory, infringes no constitutional immunity. Rowley v. Chicago & N. W. Ry., supra, 111; Southern Ry. Co. v. Watts, 260 U. S. 519, 526; and see Cumberland Coal Co. v. Board of Review, 284 U. S. 23, 25ff; Sunday Lake Iron Co. v. Wakefield, 247 U. S. 350.
Cases setting aside an excessive allocation of railroad system value to the taxing state, Fargo v. Hart, supra; Rowley v. Chicago & Northwestern Ry., supra, or setting aside improper valuation made for purposes of condemning property, Monongahela Navigation Co. v. United States, 148 U. S. 312, or for determining whether public utility rates are confiscatory, Southwestern Bell Telephone Co. v. Public Service Comm‘n, 262 U. S. 276, 287, 288; Bluefield Waterworks Co. v. Public Service Comm‘n, 262 U. S. 679, 692; McCardle v. Indianapolis Water Co., 272 U. S. 400, 408, 412, do not support the decision now made. In such cases the complainant, because the valuation is too high or too low, suffers a harm from which the Constitution guarantees immunity. But the Constitution guarantees no immunity from taxation even though the tax, because of its amount, may be burdensome, see Magnano Co. v. Hamilton, supra, or because it is as high in a year of depression and falling property values as in years of prosperity. Beyond this, petitioner does not show that it is harmed, or present any case for invoking the protection of the Constitution.
MR. JUSTICE BRANDEIS and MR. JUSTICE CARDOZO join in this opinion.
Notes
Column 1, system value according to average of stock and bond value and capitalized value.
Column 2, percentage of system value in North Dakota as reflected by the apportionment factors used in the exhibit.
Column 3, values apportioned to North Dakota according to average of stock and bond values and capitalized values.
Column 4, the North Dakota assessments.
Column 5, ratio of assessments to assigned values on the basis of stock and bond values.
Column 6, ratio of assessments to assigned values on the basis of capitalized values.
Column 7, ratio of assessments to assigned values on the basis of composite of both.
| 1 | 2 | 3 | 4 | 5 | 6 | 7 | |
|---|---|---|---|---|---|---|---|
| Apportionment to North Dakota on basis of three use factors, i. e., transportation service, traffic units and gross operating revenue. | |||||||
| 1929 | $437,789,980 | 18.826% | $82,418,342 | $83,294,677 | 112.09% | 92.01% | 101.06% |
| 1930 | 460,989,128 | 18.80 | 86,665,956 | 83,294,688 | 107.12 | 87.15 | 96.11 |
| 1931 | 455,777,674 | 18.853 | 85,927,765 | 82,999,997 | 103.73 | 90.37 | 96.59 |
| 1932 | 415,278,961 | 19.00 | 78,903,003 | 78,850,024 | 102.36 | 97.62 | 99.93 |
| 1933 | 345,188,820 | 19.14 | 66,069,140 | 78,832,888 | 114.95 | 124.03 | 119.32 |
| Apportionment to North Dakota on basis of trackage operated and the three use factors. | |||||||
| 1929 | 437,789,980 | 19.113 | 83,674,799 | 83,294,677 | 110.40 | 90.63 | 99.55 |
| 1930 | 460,989,128 | 19.148 | 88,270,198 | 83,294,688 | 105.17 | 85.57 | 94.36 |
| 1931 | 455,777,674 | 19.185 | 87,440,947 | 82,999,997 | 101.94 | 88.81 | 94.92 |
| 1932 | 415,278,961 | 19.298 | 80,140,534 | 78,850,024 | 100.78 | 96.11 | 98.39 |
| 1933 | 345,188,820 | 19.345 | 66,776,777 | 78,832,888 | 113.73 | 122.71 | 118.05 |
| Apportionment to North Dakota on basis of mileage of tracks owned and operated and leased and operated and the three use factors. | |||||||
| 1929 | 437,789,980 | 19.59 | 85,763,057 | 83,294,677 | 107.72 | 88.42 | 97.12 |
| 1930 | 460,989,128 | 19.632 | 90,501,386 | 83,294,688 | 102.58 | 83.46 | 92.04 |
| Year | Railway Operating Revenue | Net Railway Operating Income |
|---|---|---|
| 1922 | $103,452,937 | $17,276,598 |
| 1923 | 120,077,771 | 24,731,992 |
| 1924 | 110,243,104 | 24,201,287 |
| 1925 | 114,924,960 | 28,276,183 |
| 1926 | 117,383,909 | 31,280,429 |
| 1927 | 117,904,005 | 29,202,540 |
| 1928 | 126,737,091 | 31,294,069 |
| 1929 | 125,932,808 | 32,457,523 |
| 1930 | 104,996,076 | 21,912,508 |
| 1931 | 77,087,454 | 12,669,420 |
| 1932 | 55,549,247 | 1,290,551 |