Date 17-18 March 1937
Author George Haas
Title Rationale of the undistributed profits tax
Description Staff memo, Division of Tax Research, Treasury Department
Location Box 14; Corporations, Partnerships, and Sole Owners to Excess Profits and War Profits: 1936-1940; Records of the Office of Tax Analysis/Division of Tax Research; General Records of the Department of the Treasury, Record Group 56; National Archives, College Park, MD.

Rationale of the undistributed profits tax

(Memo. from Mr. Haas to Mr. Magill, March 17-18, 1937)

DATE: March 17, 1937

To: Mr. Magill

FROM: Mr. Haas

SUBJECT: Rationale of the Undistributed Profits Tax

1. The primary purpose of the undistributed profits tax was to obtain some $600 millions of additional annual revenues by making effective the Federal Government's prevailing progressive taxation of individual incomes. The estimated revenue yield was based upon 1936 calendar year conditions; it is recognized that the yield in the future may fluctuate substantially above and below this figure. It was thought that new taxes or the raising of rates on old taxes could hardly be considered at a time when the removal of an outstanding source of tax avoidance under the existing laws would provide all the new revenues needed. By eliminating this source of tax avoidance, we would also be correcting important sources of unfair discrimination in the distribution of the tax burden.

2. When, prior to the Revenue Act of 1936, corporations distributed their earnings to their stockholders, the dividends were subject to the surtax rates named in our income-tax law. When corporate earnings were not so distributed, the individual stockholders, while enjoying the benefit of these earnings in the form of more valuable investments, were enabled to avoid all payment of surtaxes thereon. Between 1923 and 1929, inclusive, more than 45 percent of the compiled net profits, after income and excess-profits taxes, of all corporations reporting net income was not distributed by the corporations and was therefore not subject to the individual income taxes, or their approximate equivalent, applicable to their stockholders. 3. Very large proportions of the incomes accruing for the benefit of members of the upper income groups in the United States had previously escaped the individual income surtaxes for long periods of forever in this fashion. Henry Ford, for example, has been liable for individual income surtaxes on only such portions of the profits of the Ford Motor Company as have actually been paid out to him. By allowing his profits to accumulate in and to be reinvested by the Ford Motor Company, he has been enabled to avoid individual income-tax liability for most of his share of such profits. The Federal Government need never get anything like the equivalent of the taxes avoided in this fashion by Mr. Ford (to continue the example), for at Mr. Ford's death his estate will pass to his heirs without any tax liability for the huge capital gains incorporated in this estate, part of which arose through the Ford Motor Company's retention of earnings; nor will the rates of tax applicable to the estate be at all higher than on estates of similar size that may have been created by individuals whose equitable incomes had been subject in much greater measure to our individual income and capital gains taxes.

4. Our high individual surtax rates were obviously ineffective in large part when their application could be postponed for long periods or avoided altogether through corporate retention of earnings. Further, even when corporate earnings were fully distributed over a period of years, there was unjust discrimination in the distribution of the tax burden. The ability of corporations and of their controlling stockholders to choose the timing of dividend distributions, without any effect upon the corporation's tax liability and without reference to current earnings, often resulted in a loss of revenue to the Federal Government and an unjust avoidance of taxation by stockholders of large personal incomes. The current earnings withheld by a corporation would often, if distributed, have raised the surtax brackets of many stockholders, thereby subjecting such earnings to the higher surtax rates. When withheld for a time and then paid out in years when the other income of important stockholders was smaller, such earnings escaped the higher rates to which they would have been subject. Individual businessman and partnerships possessed no corresponding choice for the timing of the distribution of earnings for income-tax purposes. All of the income of individuals and members of partnerships is taxable in the year earned, whether saved and reinverted in the business or not.

Moreover, shareholders in corporations that pursued liberal dividend policies were discriminated against as compared with shareholders in corporations pursuing niggardly dividend policies; for the corporation taxes took no account of the differences in dividend policies, while the individual income-tax liabilities were greater for stockholders in corporations making liberal distributions. Stockholders receiving large dividends commonly reinvest much of as these, but they are subject to individual income taxes thereon before such reinvestment. There the corporation made the reinvestment directly for stockholders, such investment took place tax free.

Again, an individual who reinvests in his business the large profits of one year and subsequently experience losses, is nevertheless subject in full to the income taxes on the profits of his good year; whereas the stockholders of a corporation that similarly reinvested the large earnings of one year and subsequently suffered equivalent losses, escaped individual income taxes on the profits of the good year.

5. It is clear that the most direct method of achieving equality of taxation of all income, corporate and other, would be to include in the taxable incomes of corporate stockholders their pro rata shares of the current earnings of corporations, whether distributed or not. For various reasons, economic as well as legal, this is impracticable. An increase in the schedule of normal corporation income tax rates might produce the mathematical equivalents in revenue, but would be unfair to stockholders with small incomes and to stockholders of corporations pursuing liberal dividend policies. The President therefore proposed to meet these problems of tax avoidance and tax inequity by repealing altogether the normal corporation income, capital stock and excess profits taxes; and to substitute therefor very substantial rates of tax on the undistributed profits of corporations. The though was that these rates of tax would encourage the distribution of the great mass of current earnings of corporations (no attempt was made to tax previously accumulated surpluses), thereby subjecting them to the individual income taxes; and that the taxes paid by corporations on the retained portions of their earnings would fully compensate the Treasury for the individual income taxes avoided through such retention.

The Congressional committees were disinclined to abolish the corporation income, capital stock, and excess profits taxes, partly because of the legal and economic advantages enjoyed by corporations over individuals and partnerships (particularly by large corporations), partly because such taxes were already well established and therefore reflected in the price of corporate securities, and partly because of the uncertainty attaching to revenue estimates for the new substitute tax. In consequence, the normal corporation income, capital stock, and excess profits taxes were retained in the Revenue Act of 1936, the two former at lower rates; and the rates of tax proposed on undistributed profits in the House Bill were reduced. Despite the retention of the normal tax on corporation incomes, the previous exemption of dividends from individual normal tax was eliminated.

As finally incorporated in the Revenue Act of 1936, the rates of tax on undistributed corporate earnings were designed to yield the Treasury, on the average and in conjunction with the normal corporation income tax, substantially the same revenues from undistributed corporate earnings as would be obtained from individual income taxes if these earnings were fully distributed; and in so doing substantially to equalize, on the average, the income tax burdens borne by owners of incorporated business enterprises, on the hand, and by individuals deriving income from other sources, on the other.

The Revenue Act of 1936 imposed the following rates of surtax on the undistributed net incomes of corporations (net income minus normal tax, interest on Governments obligations, taxable dividends paid, and sundry other credits. Adjusted net income as used below is defined as net income minus normal tax, interest on Government obligations, and sundry credits):

Undistributed net income not in excess of 10 percent of the
adjusted net income - percent.

Undistributed net income in excess of 10 percent and not in excess of
20 percent of adjusted net income - 12 percent.

Undistributed net income in excess of 20 percent and not in excess of
40 percent of the adjusted net income - 17 percent.

Undistributed net income in excess of 40 percent and not in excess of
60 percent of the adjusted net income - 22 percent.

Undistributed net income in excess of 60 percent of the adjusted net
income - 27 percent.

The first $5,000 of the undistributed net income of corporations with
adjusted net incomes of less than $50,000 was made taxable at the
first bracket rate only - 7 percent.

Banks, insurance companies, etc., were exempted from the
undistributed profits tax provisions principally on the ground
that, as fiduciary institutions, using other people's money
mainly, the greater their capital resources the greater the
protection to the public.

Domestic corporations in bankruptcy, or insolvent and in
receivership, for any portion of the taxable year were likewise
exempted, as were foreign corporations and corporations organized
under the China Trade Act.

Credits against adjusted net income are allowed corporations for such
portions of the earnings as cannot be paid in dividends by reason
of a written contract executed prior to May 1, 1936; or as are
required by such a written contract to be irrevocably set aside
within the taxable year for discharge of a debt.

Corporations whose dividend distributions are in excess of their
adjusted net incomes in any year may credit such excess as a
dividend distribution in a subsequent year, provided such excess
was paid in either the first or second immediately preceding
taxable year; but no such credit is allowed for dividends paid
prior to the first taxable year under the Act.

The Act specifically includes as a taxable dividend, and therefore
subject to credit, a dividend payable at the option of the
stockholder in either stock or money (or other property).

5. Prior to the imposition of the surtax on undistributed profits, the Treasury had estimated that less than $3,000 millions of net cash dividend disbursements would be made in 1936, After the enactment or the tax, this estimate was raised to more than $5,000 millions; and such evidence as we have to date supports such an estimated increase. (The November 1936 dividend declarations, as computed by The Annalist, were more than twice as large as in November 1935; these figures included intercorporate dividends.) The great bulk or the additional dividends went, of course, to members of the higher income groups. Between 1922 and 1934, inclusive, 81.8 percent of the dividends reported by individuals subject to income tax were received by individuals reporting net incomes of $10,000 or more.

6. During the Congressional debates on the tax bill, and subsequent there to, a number of criticisms were made of the measure. The more important of these are discussed below. A number of the critics, including some Congressmen and Senators, proposed as a substitute an increase in the normal corporation income tax to an absolute or average level of 20 to 25 percent. It is interesting in this connection to examine the following table which shows the combined normal &rid surtax as a percentage of net income for corporations of selected sizes under varying assumptions as to the percentage of income, after deducting the normal tax, paid out in dividends.

It will be observed that corporations with very large incomes may, in addition to limiting their taxes to 15 percent by paying out all of their adjusted net income, pay out only 30 percent of their earnings after normal tax and still pay total normal and surtaxes of 25.5 percent or less; or pay out only 50 percent of their earnings after normal tax and pay total taxes of only 21.4 percent or less on their net incomes. Corporations with net incomes between $2,000 and $20,000 can limit their dividend disbursements to 50 percent of their adjusted net incomes and still pay total normal and surtaxes ranging between 11.2 and 15.8 percent of their net incomes.

7. Although the new law removes major sources of tax avoidance and tax inequality, it is not altogether free from inequities. The retention of the corporate normal tax and the elimination of the exemption of dividends from individual normal tax may be defended on the ground that corporate income is properly taxable as such, because or the legal and economic advantages enjoyed by corporations, regardless of income differences among their stockholders. The corporate surtaxes, however, while defensible in their aggregate results, will involve some inequities because they will be paid at the same rates, in effect, by stockholders of all income groups. Such inequities will be eliminated to the extent that the corporate surtaxes actually bring about the disbursement of earnings in dividends; and to the extent that shareholders shift their investments from companies that penalize stockholders with small incomes by undue retention of earnings and the resultant payment of surtaxes thereon.

  Combined normal tax and surtax on undistributed profits expressed
   as a percentage of income, for corporations of selected sizes,
    assuming different percentages of income (after deducting the 
                normal tax) are paid out in dividends

                         Percentage paid out in dividends
Income /1/        0       10       20       30       40       50
                                Effective tax rate

     2,000      14.4     13.3     13.2     12.5     11.9     11.2  
    10,000      20.3     18.8     17.1     15.6     14.4     13.5 
    20,000      26.2     23.8     21.4     19.3     17.4     15.8 

    50,000      30.3     23.0     25.6     23.3     21.0     19.0
    75,000      31.2     28.9     26.5     24.2     21.8     19.9
   100,000      31.5     29.2     26.9     24.5     22.2     20.3
   200.000      32.0     29.7     27.3     25.0     22.7     20.8
   300,000      32.1     29.8     27.5     25.2     22.9     21.0
   500,000      32.2     29.9     27.6     25.3     23.0     21.2

 1,000,000      32.3     30.0     27.7     25.4     23.1     21.3
10,000,000      32.4     30.1     27.9     25.5     23.2     21.4

                          (TABLE CONTINUED)

                    Percentage paid out in dividends
Income /1/        60      70       80       90      100       
                            Effective tax rate

     2,000      10.6      9.9      9.3      8.6     8.0      
    10,000      12.9     12.3     11.7     11.0    10.4 
    20,000      14.3     13.1     12.4     11.3    11.2 

    50,000      17.2     15.7     14.3     13.3    12.7   
    75,000      18.0     16.5     15.1     14.1    13.5 
   100,000      19.4     16.9     15.5     14.4    13.8
   200.000      19.0     17.5     16.0     15.0    14.4
   300,000      19.1     17.7     16.2     15.2    14.6 
   500,000      19.3     17.8     16.4     15.4    14.3

 1,000,000      19.4     17.9     16.5     15.5    14.9  
10,000,000      19.5     18.0     16.6     15.6    15.0
/1/ It is assumed that the income did not include any interest on Government obligations or dividends received from other corporations.


The principal objections which have been advanced against the undistributed profits tax are:

(1) Corporations that use part or all or their earnings for
the payment of short-or long-term debts are subjected to a tax

(2) Corporations are deprived of a necessary source of
capital for expansion, except on payment of a tax penalty.

(3) The tax on undistributed earnings will discourage
corporations from creating the corporate surpluses deemed
necessary to maintain wages, employment, dividends, and business
solvency, through periods of depression.

(4) The tax will prevent small corporations from growing
into big ones, and therefore operate to protect established
corporations with accumulated surpluses from new competition.


This objection is usually stated as a self-evident truth requiring no logical or other support. In some cases, however, reference is made to existing or future contracts restricting the payment of dividends or requiring stated amounts of debt to be retired periodically, the implication being that the retirement of such debts can take place only out of earnings. It is not usually suggested, however, that individuals owing short-term debts contracted for consumption purposes, or longer-term debts such as mortgage debts on their homes, be released from income-tax liability for the portion of their incomes used for the payment of such debts. Nor is it usually proposed that members of partnerships be freed from individual income-tax liability on those portions of their partnership incomes employed for the retirement or partnership debts. The underlying assumption is that corporate debts are of such special character as to merit exceptional treatment. This assumption does not withstand analysis.

Corporate indebtedness consists of two chief types: commercial debts such as bank borrowings for inventory and other working capital purposes; and funded debt, usually represented by mortgage or debenture bonds.

With respect to the first type, that is, short-term loans for the purposes of acquiring inventories, paying wages, et cetera, it is not usually contemplated that such loans should be repaid out of EARNINGS. On the contrary, it is expected that such loans will be repaid out of the GROSS PROCEEDS of the business. A department store which borrows $100,000 for the purchase of seasonal merchandise for its Christmas trade is expected to repay this loan out of the gross retail proceeds of perhaps $l30,000, and not out of the $2,000 net profit remaining to it after all costs. It is for this reason that such loans are welcomed by banks as self-liquidating.

As a class, moreover, such loans are never completely liquidated because new loan transactions are usually being entered into by the same or other business enterprises as past loans are being liquidated -- though not necessarily, of course, in offsetting amounts. They constitute a revolving fund whose aggregate volume has normally exceeded the total statutory net income of American corporations. Their volume for any particular enterprise bears no necessary relationship to the amount of its net profits, and is frequently several times the amount of the invested capital of such enterprises.

It is true that exceptionally profitable enterprises sometimes reduce the average volume of their commercial borrowings by substituting capital funds derived from reinvested earnings. This, as stated above, is not the normal source of repayment of such borrowings. Other enterprises reduce or eliminate their commercial borrowings by the sale of additional securities to their stockholders and others. In general, however, the most economical utilization of the country's capital takes place when short-term needs are supplied by temporarily borrowed rather than by owned and permanent capital funds. Hence, no public purpose would be served by the extinguishment of short-term commercial debts.

Finally, it is obvious that the exemption from corporate surtaxes of earnings employed for the repayment of commercial debts would open wide an enormous door for tax avoidance: Debts could be contracted solely for the purpose of retiring them out of earnings and thereby indirectly reinvesting earnings and adding to the proprietorship capital tax-free.

With respect to bonded and other forms of longer-term indebtedness, somewhat different considerations apply. Much of the proceeds of such loans has normally been employed for plant, machinery, and other fixed assets. Is it fair or wise to place a surtax on earnings employed to retire such debts?

It should be observed in the first place that prudent lenders and prudent borrowers expect a fixed investment financed by such a loan to at least pay for itself during the life of the property. That is, funds for the complete retirement of the investment, if no replacement of the physical property is contemplated, will be provided by depreciation charges deducted from the gross receipts over a period of years. Prudent lenders and prudent borrowers are not likely to contract for loans extending over a longer period than the economic life or the property to be constructed or acquired with the proceeds; though loans are often contracted, perhaps by reason of temporary money market stringency, for periods shorter than the economic life of the property, with the expectation that such loans will be subsequently refinanced. In all such cases, depreciation charges should provide the necessary funds for the extinguishment of the debt; no net earnings should be necessary for this purpose.

It is more commonly true that physical properties financed by bond issues are continuously maintained or replaced. Where this is done by the use of funds allocated from gross receipts for depreciation and similar charges, the retirement of the indebtedness normally takes place in any one or more of three ways: (1) By refunding the debt at maturity or the earliest call period by means of a new issue of obligations (This is by far the most common method among railroad and public-utility companies, which account for about three-fourths of the aggregate funded debt of all non-financial corporations.); (2) by funds raised through the sale of additional common or preferred stocks (a method commonly used by industrial corporations); and (3) by funds derived from the undistributed earnings of the corporation.

The character and importance of the third method is the one at issue in this connection. It is clear that this third method differs from the second only in form. In both cases, stockholders' capital is used to retire the debt. In the one case, there is a formal issue of securities to recognize the increased equity of the stockholders in the capital assets. In the other case, the increased equity of the stockholders is recognized by a commensurate addition to the corporate surplus account. The essential identity of the two methods may be seen by contrasting a corporation which refrains from dividend payments for five years and employs the $25 per share of earnings accumulated during that period for the retirement of debt, with another corporation that pays out the $5 per share earnings annually, but which, through an issue of stock rights at the end of that period, obtains $25 per share from its stockholders to retire indebtedness. The fundamental fact is that the use of corporate earnings to retire corporate indebtedness adds to the equity of the stockholders.

It has been suggested that all corporate earnings that are employed for debt retirement be exempted from the application of the surtax. Such proposals possess an irrational emotional appeal because the supporting argument usually runs in terms of a debt-ridden individual -- although, as noted above, the law makes no concession to such an individual. The fundamental fact that corporate debts are parts of corporate capital structures and that the debt element in these structures is voluntary, for the most part, is ignored. Corporations of similar size in the same industry -- competitors -- vary greatly in the character of their corporate structures. Shall one company, because its stockholders desired to increase their own earnings by obtaining part of the corporate capital through the issue of fixed-interest securities, be given an additional tax advantage over the stockholders of another corporation in the same line of business who chose to supply all the capital through common stock? Corporations that have obtained part of their capital through the issue of debt obligations already enjoy a tax advantage over debt- free corporations in that the interest charges on the debt are deductible from the total earnings of the corporation before arriving at the net income subject to normal tax. To allow corporations to retire funded debt through reinvested earnings without payment of the surtax would be further to discriminate against competing and other corporations with debt-free capital structures.

We have already called attention to the fact that railroads and other public utility enterprises, which account for about three- fourths of the long-term debt of all non-financial corporations of the United States, commonly regard such debt as permanent elements in their capital structures, meeting the maturities of particular obligations by refunding issues. Industrial corporations that desire to reduce their long-term debt may employ funds provided by depreciation allowances and new capital funds obtained by the sale to stockholders and others of additional equity securities, as well as by funds directly obtained from corporate earnings. Under the Revenue Act of 1936, moreover, a corporation need not distribute its current earnings in cash in order to avoid liability for the surtax on undistributed earnings. It may retain the cash and employ it for debt retirement, and nevertheless avoid the surtax by the distribution of taxable dividends consisting of securities, such as preferred and common stocks, where the effect of the distribution is to change the pro rata interest of the shareholders. There would appear to be no logical or practical ground therefore for exempting from the surtax on undistributed earnings such portions or corporate earnings as are used for debt retirement. In order, however, to avoid hardship for corporations which had previously entered into contracts specifically restricting the payment of dividends or requiring a portion of the earnings to be paid in discharge of a debt, the Revenue Act of 1936 exempted from the surtax that portion of corporate earnings the distribution of which was so restricted.

Mr. L. H. Parker, in a manuscript study entitled "Investigation of possible revisions of the surtax on undistributed profits," asserts that the exemption just cited is discriminatory because such provisions in bond indentures are no more common than other requirements not recognized by the Act, which equally restrain the payment of dividends; such as (1) the maintenance of quick assets equal to or higher than a stated percentage of the bonds outstanding; (2) the maintenance of quick assets above a stated minimum ratio to current liabilities; (3) the prohibition of such dividends on common stock as would reduce the net quick assets below a certain number of dollars per share.

These three types of cases cited by Mr. Parker are quite different in character from those given special treatment under the Act. The amount or proportion of quick assets in relation to current liabilities or total bonded debt or number of shares of common stock bears no direct or necessary relationship to the amount of earnings available for dividends. Earnings rarely constitute the primary source of quick assets, and an improving quick asset position often takes place during the period when a company is steadily losing money. The amount or proportion of quick assets is primarily a function of the form in which the corporation desires or is required to keep its TOTAL ASSETS. Taxable dividends in the form of preferred stocks or longer-term debenture bonds, for example, could be paid in any of the foregoing three cases cited by Mr. Parker. In other cases cited by Mr. Parker, such as the requirement that a stated number of dollars be placed in a sinking fund each year, the same considerations apply. These sinking fund payments can be made from funds representing charges for depreciation or from the general assets of the corporation.

The practical effects upon the tax revenues of any substantial exemption from the corporate surtax of earnings used for debt retirement, apart from the possibilities of tax avoidance by the creation of new debts, may be guessed from the following figures, which show the aggregate indebtedness of all corporations submitting balance sheets with their income-tax returns in 1928 and 1933:

                                        1928            1933
                                        ____            ____
                                         (Billions of dollars)

Notes and accounts payable              27.4            19.3 
Bonded debt and mortgages               42.9            45.9
                                        ____            ____
Total                                   70.3            65.2