Obviously, under this plan a problem would arise in view of the traditional practice of corporate retention of earnings for internal use. The simple elimination of the corporate tax would result then in undistributed income escaping at least immediate tax liability. Various methods have been devised to solve this problem of untaxed income.

The most facile solution, of course, is the institution of an undistributed profits tax. Such a device would be highly unpopular with businessmen, in view of the experience of 1936-39, and can be expected to meet with stiff and well-founded opposition from that quarter. It is argued that such a tax on retained earnings would tend to place the corporation under pressure to distribute dividends, thereby militating against the potential investment projects and financial structure of the firm. Further, new, growing firms and smaller corporations would be placed at a serious disadvantage, since they usually require greater retention of earnings to finance expansion. Another and much-discussed device which could be utilized to insure complete taxation of all corporate income, in the absence of a separate corporate levy, is the partnership method. According to this solution, all stockholders are regarded as partners. They would be required to include as taxable income their pro rata share of corporate income, both distributed and undistributed. In this manner, all corporate income would be subject to full tax at personal normal and surtax rates. Needless to say, in theory, this would represent a virtually ideal solution to the problem of double taxation of corporate income. However, the administrative problems posed in allocating profits among stockholders, special treatment of various sources of income, and adjustments of reported income of prior years are generally considered to be overwhelming. Furthermore, the use of the partnership method would, in effect, require assessing a tax liability against individuals with respect to income not actually received. It would, therefore, be open to serious legal objection.

Approach No. 2: Credit for dividends paid

This approach, essentially, is based on the conception of the corporation as a separate and distinct entity, rather than merely an income conduit for stockholders. According to this view, the tax laws permit deductions for virtually all resource payments; therefore, a deduction for dividends, which in reality are payments for financial resources, should also be allowed. Under this proposal the corporation would deduct dividend payments in determining taxable income. Dividends, consequently, would be fully taxed at the personal level in the hands of stockholders, and the corporation, in effect, would incur tax liability on undistributed profits only.

This method would be fairly effective in relieving the burden of double taxation. To the degree that the corporate tax was reduced and dividend payments were unchanged, the plan would increase the amount of investment funds in the hands of the corporation and have a salutary effect on its incentive to invest. By the very nature of the method, it would tend to eliminate the discrimination against equity financing and would reduce any forward shifting which might exist. Since this approach to the problem is, in essence, an undistributed profits tax, it would, as mentioned earlier, tend to place undue pressure on management to pay out dividends and would tend to discriminate

against both new and smaller firms, which rely traditionally on internal financing to a much greater extent than older and larger corporations. Special provisions might be devised to eliminate this latter disadvantage, but this could very easily lead to the creation of serious administrative difficulties which might render the approach impractical.

Approach No. 3: Tax credit for dividends received

The methods which fall under this heading are the best known of all. Briefly, they include the British "withholding" method and the practices currently in use in the United States and Canada. Since both methods were dealt with in some detail earlier in this paper, present consideration will be limited to a brief review of their salient features.

The allowance of a tax credit for dividends received traditionally permitted in Britain represents an effective and, in its simplest form at least, a workable solution to double taxation. Essentially, the system is a type of tax withholding. The stockholder includes the value of dividends received and his pro rata share of tax paid on those dividends by the corporation as part of his taxable income. He then calculates his tax liability and deducts from it the tax paid by the corporation. If the amount of tax paid by the corporation exceeds the stockholder's liability, then he is granted a refund.

In Great Britain, however, the normal taxes on corporate and personal incomes are equal. The very inequality between the corporate rate of 52 percent and the personal normal rate of 20 percent in the United States, therefore, presents a problem. However, a withholding tax on dividends can be combined with any basic tax rate on corporations. In other words, the total tax on corporate income could be subdivided into two component taxes, one, a tax on corporate profits per se of 32 percent, and the other a withholding tax of 20 percent which would be pro rated among the stockholders of the corporation. Such a method would not eliminate double taxation completely, in view of the separate corporate tax. Eventually, however, the total tax on corporate income could be lowered gradually, while the withholding rate was increased, to the extent that refund experience did not become too burdensome and to the degree that revenue needs permitted. In the long run, complete elimination of double taxation of dividend income might be anticipated, as the withholding tax rate applied to dividends gradually approached equality with the total corporate tax rate.

Although this method would not eliminate the bias against equity financing as far as the corporation is concerned, it would provide an important stimulus to stock ownership since, if dividends were maintained, stockholders would obtain a greater net yield after tax. Furthermore, it would readily eliminate completely double taxation of distributed current income of corporations, while undistributed profits would be subject to corporate tax thereby reducing the possibility of tax postponement. Despite its generally beneficial effect on investment and equity, this method would not reduce the possibility of shifting the tax forward to consumers, and the tax itself would still tend to affect management decisionmaking.

Some concern has been expressed with the administrative problems implicit in this form of withholding. In its simplest form, stockholders would merely increase their dividends by a uniform proportion, in line with the rate of withholding at the corporate level. Although this might result in a certain unevenness with respect to equity, this would be more than compensated for by its relative administrative simplicity. A more sophisticated method, which sought to trace dividends to their various sources and to determine the composition of a dividend payment to a particular shareholder at a particular time, would create a very serious administrative burden.

The credit system currently practiced in the United States and Canada by comparison is much simpler in nature. In effect, it provides a partial credit to the stockholder against tax paid by the corporation. Accordingly, the individual stockholder includes the value of taxable dividends (i.e., in excess of $50 in the United States) received by him in computing taxable income and then applies a credit equal to some percentage of taxable dividends received against his tax liability. As previously explained, the U.S. provision for exclusion merely exempts a small absolute amount of dividends received, thereby, in effect, granting a tax credit at the individual stockholder's marginal rate with respect to dividends of $50 or less.

The credit and exclusion provisions of the 1954 code have been heralded in many quarters as a significant step forward in the development of a more realistic tax policy with respect to corporate income. With respect to distributed corporate income, the 1954 code granted a lesser percentage tax reduction to high income stockholders, as compared with those in the lower brackets. However, if total tax liability at both the corporate and individual level is taken into account, the present system of taxation of corporate income is uneven in impact and still tends to benefit higher bracket stockholders more than lower bracket investors, an advantage that is realized at least partially through capital gains." Nevertheless, this positive measure of relief, modest though it may be, to thousands of stockholders made stockownership more attractive and generally had a beneficial impact on investment levels and possibilities. The present Canadian and U.S. methods do not remove the corporate tax as a factor affecting management decisions; in particular, they do not reduce the bias against equity financing; nor do they relieve pressure on price, if such pressure does, in fact, exist. Their most outstanding advantage is their relative administrative simplicity. This is the simplest method by far of all those advanced to date.

In view of the obvious, though modest, benefits of our current method of relieving double taxation of corporation income and its ease of administration, Congress should abandon it only in the event that another method is considered more beneficial and workable. The potentialities of reforming the tax system as it relates to savings and investment are much too important to be forestalled at this late date through capricious or arbitrary action.

17 See particularly Shoup, Carl S., "The Dividend Exclusion and Credit in the Revenue Code of 1954." National Tax Journal, vol. III, No. 1, March 1955, pp. 136-147.

REVENUE IMPACT

Obviously, all possible approaches to the problem, to the extent that they reduce double taxation of corporate income, result immediately and directly in lower revenue yields. Although no attempt has been made to work out detailed estimates of these short-term revenue losses for the plans discussed earlier, their relative impact can, nonetheless, be ascertained. Other things being equal, the credit for dividends paid approach, the British "withholding" system, and the partnership method tend to produce the greatest revenue losses, in the short run, while the present U.S. tax credit and exclusion for dividends received results in a much more moderate impact on tax receipts, primarily because it fails by a wide margin to accomplish as complete a correction of double taxation as do the other plans. However, other things cannot reasonably be expected to remain equal. As mentioned earlier in this paper, the stimulus to economic activity generated by any sincere atempt to eliminate double taxation of corporate income will actually increase revenue yields in the longer run.