As noted on page 1 above, the question at issue is often discussed with special reference to items like (a) upper bracket personal sur-taxes, (b) loss carryover, (c) averaging of income, (d) capital gains taxation, (e) consolidated corporation returns, (f) taxation of dividends, (g) corporation undistributed profits, capital stock or excess profits taxation, (h) depreciation allowances, (i) payroll taxes, and (j) sales and other excise taxes. The following analysis will treat each of these matters in turn, dealing with both the consumption and the investment aspects of the national income problem, but in general covering only such special points as have not been given in Part I above.

(a) UPPER BRACKET PERSONAL SURTAXES. He there is some conflict between tax reform designed to increase immediate consumption and tax reform designed to increase direct investment. The conflict may however be avoided in part by cutting upper-bracket rates and making up the revenue loss, not through increased consumption taxes, but by increases somewhere in the $5,000 -- $10,000 brackets; at these levels, some, at least, of the increased tax may still come out of funds that, on the one hand, would not have been spent on consumption, but, on the other hand, would not likely be available for taking the considerable risks that attach to investment opportunities that are "strategic" from the point of view of the business cycle.

However, it may be doubted that substantial results in increasing investment will be show by thus reducing these upper- bracket taxes (a) unless they are reduced greatly, say to a maximum of 50 per cent of the income in the top bracket, or even less and, more important, (b) until they have been kept at these low levels long enough, possibly five or ten years, to convince investors that they are permanent. Here is the conflict between long-term and short- term policy noted in the preceding section. To undertake any such program would probably necessitate, for the near future, substantially heavier taxation of the kind of income that would otherwise be spent for consumption, and this shift in taxation might easily contribute to depress business conditions over the short term.

(b) LOSS CARRYOVERS. Allowance of loss carryover is one of the cheapest ways, as concerns revenue, of encouraging investment; much of the revenue loss does not occur until the national income increases enough to give profits, and such an increase will augment the tax revenue in other parts of the system. It may be questioned whether entrepreneurs are as deterred as they should be, by the present failure to allow a loss carryover for more than two years, but in theory at least it is a decided barrier to any kind of industry that is sensitive to the business cycle. As concerns the individual tax on corporate investors, the transmutation of corporation profits into dividends tends to average the income somewhat as a carryover device should do. Consequently it is the unincorporatecd concern of a type that is sensitive to cyclical changes or that must undergo an initial period of loss in developing. a new product that would be affected the most.

The loss carryover has little, if any, direct effect on consumption.

(c) AVERAGING OF INCOME. Averaging of income, considered as a step beyond a loss carryover, undoubtedly increases willingness to work in some special cases, such as prize fighting, moving picture acting, etc., and the Revenue Act of 1939 has been of importance in this respect. The averaging device has still greater potential importance in increasing willingness to invest; here it would serve to put capital gains, especially, into lower rate brackets than if they were simply entered as income at 100 per cent with no averaging (the latter policy which, however, is not now followed except for gains on assets held eighteen months or less).

The averaging device probably has, if anything, a slightly deterrent effect on consumption, since use of it would necessitate a higher rate structure to yield the same revenue, hence a higher tax on regular incomes, which, being in general the smaller incomes, are likely to contain in general a larger consumption element.

(d) CAPITAL GAINS TAXATION. The inclusion of capital gains in the income tax base affects consumption somewhat, since an appreciable part of stock market winnings seems to be spent fairly soon, but the more important effect is on investment. Here much depends on (1) the particular technique used in the taxing of gains and (2) the prospects for change in the law. The present Federal law must have little, if any, effect on investments made for more than two years; the maximum rate applicable to a capital gain from the sale of a share of stock or other capital asset held more than two years is only 15 per cent. In contrast, income from interest, dividends, or salary is taxable at progressive rates that reach a maximum of 79 per cent.

For shorter periods, higher rates apply to capital gains, but these short period gains are probably not of much consequence to the economy at large except in a few cases -- e.g. apartment house building, where the original promoter commonly sells out as soon as the building is completed.

The extent to which capital losses are allowed as a deduction is important with respect to incentive to invest. At present the loss deductions are restricted considerably, and are probably exercising an important deterrent effect on new investment in unincorporated, risky enterprises. A more satisfactory situation would be reached by allowing more or less full deduction of capital losses in one way or another, and raising the rates on capital gains, which seem to be far too low in proportion to the rest of the income tax rate structure. However, some concession in rate is needed because capital gains tend to be irregular, and hence, in a progressive rate system, to pay more tax than a regular annual income of the same total amount over several years (an averaging device (see (c) above) would be helpful here). Moreover, it is probable that, in general terms, a lowering of the rate on gains is a more effective stimulant to investment than a liberalizing of deduction for losses. The point of this paragraph is that the rate lowering seems to have gone to extraordinary lengths compared with the loss allowances.

Such retarding effect as the treatment of capital gains and losses exercises on investment is likely to be felt most severely in new, speculative businesses, when, as noted above, the original investor often expects to receive his reward by selling out when the concern has matured. If, however, it is desired to promote such ventures, a specially low rate of tax on capital gains is apt to be a somewhat wasteful method, since it also loses revenue from stock market speculations and other ventures having less -- or no -- influence in increasing national income (at least over the long run).

(e) CONSOLIDATING CORPORATION RETURNS. Although this question is important from many points of view, including the effect on corporate structure and the fairness with which the tax burden is distributed, it probably has relatively little effect on the volume of the national income. To some extent the refusal to allow consolidated returns deters investment, since it is another example (see (b) and (d) above) of refusing to allow losses to be set off against gains, but it is of importance only to the larger concerns, particularly those that like to restrict the effect of losses in new ventures by starting these ventures through separately incorporated subsidiaries. Otherwise, the effects of refusing to allow consolidated returns can to some extent be avoided by consolidating subsidiaries in lieu of consolidating the subsidiaries' returns.

(f) TAXATION OF DIVIDENDS. Although corporate profits are taxed at about 17 per cent on the average (Federal corporation net income tax) and although corporations pay capital stock and "excess profits" taxes that unincorporated concerns do not pay, so much of the corporation's profits as are paid out to individuals in dividends are taxed at the full individual rate just the same as income from interest, profits of unincorporated enterprises, rentals, salaries, etc. Moreover, dividends received by corporations are exempt only to 85 per cent.

Allowances for small corporations less en these differences somewhat but the genial effect is a substantial penalty against the corporate form so far as small investors are concerned and against the use of stock (especially preferred stock) rather than bonds or other fords of debt. The use of the corporate form for very small businesses is made extremely difficult.

Whether the mere fact of discrimination decreases the national income in total is not clear. Removal of discrimination by partially untaxing dividends and making up the loss by increasing individual rates, for instance, might well have no appreciable net effect on either investment or consumption. Moreover, the discrimination is not so great as it might seem; the corporation's ability to keep earnings away from the individual surtax for a long time (sometimes forever) -- the problem of undistributed profits taxation -- counteracts to some extent the retarding effect exercised by such difference in tax treatment.

The present degree of inter corporate dividend taxation is probably not serious enough to check materially the national income.

One factor to watch is the investment trust situation. Any further discrimination against dividend income is likely to react seriously against the general type of investment trust (as contrasted with the open-end Massachusetts type, which is given special treatment), and such reaction may have adverse repercussions on the market for securities of new enterprises.

On the whole, however, the readjustment of the taxation of dividends, clearly needed, is a matter of avoiding obvious unfairness and possibly undesired effects on type of corporate financing, rather than of increasing the national income.

(g) CORPORATION UNDISTRIBUTED PROFITS CAPITAL STOCK OR EXCESS PROFITS TAXATION. The undistributed profits tax as amended in the 1938 Act was so small that it presumably had no appreciable effect on either consumption or investment.

An undistributed profits tax as substantial as that of the 1936 law does, on the other hand, have important effects on the national income. However, some effects are in one direction, some in another, and whether they almost cancel out is a question that cannot be answered satisfactorily without further study.

So far as the tax does not force distribution of profits, it has most of the effects of an ordinary corporation income tax; so far as it does force them, it probably increases the national income somewhat over the short period, since some part of the distributed profits is spent on consumption, and such investment as would have been made directly with the profits can in large part be made through the banking system or even bar getting back some of the distributed money. There will surely be some cases, however, where the short-term national income is lessened -- e.g., a hard-pressed concern that must contract its activities if it loses much of its cash, owned in part by well-to-do individuals who hoard the dividend money. So far as the distribution can be made in taxable "tickets" -- stock dividends of certain types, scrip, etc. -- the undistributed profits tax probably has somewhat less effect on national income than if the distribution were in cash, if we may assume that investors are less apt to spend stock dividends, etc., than cash dividends.

Over the long term the effects are much more uncertain. A severe undistributed profits tax probably hinders the growth of medium-sized corporations (too large to count on cooperation by the stockholders to reinvest the dividend money, and too small to appeal to the capital market or get term loans from banks), but this effect may be counterbalanced in part by a larger-than-otherwise growth of big corporations and an increase in the number of small corporations. There is the further question of whether, investments made without passing the test of the capital market -- that is, made simply by using accumulated earnings -- tend to be more or less successful in increasing the national income than other investments. Until much more study has been given to these long-term phases of the problem, even a tentative answer can scarcely be formulated.

The capital stock tax is not large enough to have an important effect on national income. It may decrease dividends payments, and hence to some extent consumption. Linked as it is with the "excess profits" tax, it is something like a tax on net income or output, with a corresponding retarding effect on investment. The fact that the tax encourages the raising of capital by borrowing instead of issuing stock has probably no important effect on national income, at least at the present level of the tax.

The so-called excess profits tax, low in rate and not directly effective in practice, because of its link with the capital stock tax, can be disregarded in the present discussion. A true excess profits tax with a substantial rate would require extensive analysis.

(h) DEPRECIATION ALLOWANCES. presumably the question at issue here is not whether full allowance will be given for depreciation by the time the asset is scrapped (this principle seems too well settled to be discussed here) but how the allowance shall be distributed over time -- or possibly whether sore part of the allowance shall be transmitted into a capital loss (upon the scrapping of the asset). Since a carryover of loss has been allowed (see (b) above) by the Revenue Act of 1939, the question here loses most of its importance. Since the carryover is somewhat restricted, however, depreciation might be allowed "in excess" in years of profit and correspondingly restricted in years of loss. The effect on investment would probably be such as the effect of a loss carryover, though in milder form, and concentrated in heavy industry fields where depreciation is an important cost factor. The direct effect on consumption would scarcely be appreciable.

Excessive depreciation in the early life of the asset, counter- balanced by inadequate depreciation in its later life, might have a mildly encouraging effect in a few cases.

If the question is one of allowing depreciation to an aggregate in excess of the cost or other basis, it raises fundamental issues of subsidy through tax relief and involves consideration of subsidies to other than capital costs -- e. g,., is there more or less benefit to be derived from giving the subsidy for labor costs (allowing a deduction of, not merely 100 per cent, but, say 150 per cent, of wages paid) than for depreciation?

(i) PAYROLL TAXES. So far as the payroll taxes are a direct burden on employees, they doubtless restrict consumption almost to the full extent of the tax, but correspondingly have little direct effect on investment. So far as they do not rest on employees, they are a form of margin tax and hence tend to restrict expansion of production and thus investment, except, possibly, for a special counteracting effect in promoting investment in machinery. On balance, such part of the tax as rests on employers probably does not retard investment appreciably at present rate levels, although so little has been discovered as yet about the effects of this tax that such a statement is of highly uncertain value. Thus, the chief short- term concern over the present payroll taxes' effect on national income may reasonably center on the consumption aspect.

(j) SALES AND OTHER EXCISE TAXES. The general effect of sales taxes, excises, and other taxes that vary directly and proportionately with the volume of business has been discussed above.

It might be possible to discover which of the taxed industries are already utilizing plant at nearly full capacity and which are not; if the former were given tax relief, any resulting increase in demand might bring important short-run tertiary effects in investment in plant and equipment, While this result could not be anticipated in other industries.

Likewise, it might be possible (though this is highly doubtful) to ascertain in which of the taxed industries the factors of production would be most likely to respond quickly most of the increased income they would get if demand increased as a result of lowering the tax -- in other words, where the potential secondary effects are greatest.

(k) ESTATE AND GIFT TAXES. Although these taxes are not specifically mentioned in the question, attention is called to them here because of the likelihood that of all the important taxes available, they have the smallest retarding effect on national income through consumption and investment. Over the very long term at very high rates they might conceivably create a disadvantageous shortage of capital funds, but that danger seems so remote at present as to deserve little consideration.