in effect that, until tax rate reform had proceeded to the point where the structure was reasonably compatible with the need for and requirements of optimum economic growth, use of revenue gain for this purpose should have priority over all other uses barring a grave new military emergency. In operation, such a policy would mean that except for the military contingency there would be no increases in any sector of Federal spending which were not offset by decreases in others.

The possibilities for rate reform through this approach are rather spectacular in a program extending over a relatively few years.

It is generally estimated that the revenue gain from economic growth will run about $1 billion for each percentage point of growth. The reason that the gain for a percentage point of growth is more than 1 percent of present total revenues is that, without material change in exclusions, deductions, exemptions and credits, the base of the personal tax, taxable income, increases roughly 50 percent faster than growth in personal income. Vigorous growth probably would be somewhat more productive of revenue gain, relatively, than sluggish growth.

Thus, if the economy should grow no more than the generally used conservative estimate of 3 percent a year-or in the range of growth for the 5 years preceding the recent recession-there would be a revenue gain of $3 billion annually. It is now evident, however, that wage-push on the one hand, and punitive tax rates on the other, have retarded growth since the Korean war. I for one believe that it would be sound national policy to anticipate that, upon enactment of a thorough program of tax rate reform, the growth rate would move upward to at least the range of 4 percent-or the average for 60 years before the depression of the 1930's. Actually, in view of the greater productiveness of capital as the result of rapid technological advance, and the obvious fact that a higher income economy is able to save more than a lower income economy if not restrained by punitive tax rates, there is no reason why the economy should not grow more rapidly than in the predepression years as the dynamics of growth are reenergized under moderate tax rates.

With the availability of up to $4 billion annually for tax rate reform, two interrelated procedures must be brought into play in order to make certain that the goals of reform are achieved at the earliest possible date.

The first is the forward scheduling of the rate reductions in one legislative package. This not only would provide assurance of consistency in national policy over the life of the legislation, but would in itself amount to a declaration against using revenue gain to underwrite increased spending.

This protection against increased spending, however, would be greatly strengthened by the second procedure, namely, provision to force postponement if budget unbalance becomes a threat. Such a provision would insure against a return to deficit financing if some unforeseen contingency should force up the spending level. At the same time, the two procedures, in combination, would enable the public and the Government to make a clear choice each year between less repressive tax rates or more spending.

SETTING GOALS FOR TAX RATE REFORM

In considering what goals would be appropriate for tax rate reform, the lack of rudder or compass makes it necessary to use makeshift for dealing with graduation. We are bound in by what can be done in a particular year, but not in the number of years over which rate reductions might be forward scheduled. In a practical sense, therefore, we are free to set goals for reduction within a considerable range of judgment.

We have already found one mooring post, namely, the linking of the top rate of personal tax with the top rate of corporate tax in a reform program. The question therefore becomes: To what point or range should the top rates of both taxes be moored?

I have long felt that a reasonable, though arbitrary, goal would be one-third of the base, or 33 percent. In considering such a goal, we must overcome the normal tendency to appraise the personal tax in a vacuum, as if it were the only tax borne by persons in the middle and higher income brackets. It has been estimated that people in these income circumstances bear a burden of taxes, direct and indirect other than the Federal income taxes, which will range up to 15 to 20 percent of their incomes. Thus, with a 33 percent top rate of Federal individual tax, the total tax burden of people in the highest brackets would still run up toward 50 percent of their incomes. Another arbitrary, but eminently reasonable approach in regard to initial legislation, would be to set the goal at or in close relation to the top rate of corporate tax between World War II and the Korean war, namely, 38 percent. It may be argued with considerable force that in a period of indefinite duration when economic security and growth are the basic armaments of a cold war, we should not contemplate continuance of a top rate of tax on the income of any individual or business in excess of the rate on corporate income in the period of readjustment after the most costly and destructive war known to history.

Or, these two approaches might be combined in the sense that any top rate goal within the range of 33 to 38 percent would provide a structure which would be moderate in relation to the present structure and contemporary fiscal circumstances. If there were no other benchmarks which could be brought into play there certainly should be no hesitancy in casting initial tax rate reform legislation on this basis alone. However, the concepts involved do lack flexibility if extended beyond contemporary fiscal circumstances, or considered in the light that fiscal circumstances must inevitably be subject to ups and downs over the long pull. For example, in the event of a new grave military emergency, on the one hand, or movement into an era of substantial disarmament on the other, these approaches singly or in combination would provide inadequate guides in the readjustment of the rate structure which would or should take place.

I have given considerable thought to the problem of how, on the one hand, to contain and control graduation and excessive taxation of corporate income and, on the other, to permit flexibility as fiscal conditions change.

I find that there is a way, a mooring post for permanent top rate policy, which as of the present would in general produce a result which coincides with the range of the pre-Korean war corporate rate,

and the one-third concept, and which would still permit substantial variation from such a range, both down and up, under changed cir

cumstances.

This idea for a permanent top rate policy guide is simply that the top rates of tax be loosely tied, in a 2-to-1 ratio, to the percentage relation of total Federal revenues to national income. In other words, the tax rate under any method should not exceed twice the average burden of taxes under all methods. At the present time, Federal revenues are running in a percentage relationship to national income of around 19 to 20 percent. Thus, if this guide, or mooring post for top income tax rates were in effect at the present time, those rates would be in the range of 38 to 40 percent.

By far the most attractive feature of this mooring post for top rates of income tax is its flexibility. For example, if the ratio of Federal tax revenues to national income should decline to 15 percent, top rates in the range of 30 percent would be in order. By the same token, if a new national emergency should force up the Government take as a percent of national income to say 25 or 30 percent, the guide would permit top rates in the range of 50 or 60 percent. If there should be another world war, and if it should be fought on our part on a payas-you-go basis, such a guide would permit top rates of tax which might approach the level of confiscation. But, when the war was over, there inevitably would be a drastic downward adjustment in the rates to accord with the problems of economic reconstruction and resumed growth of the private economy.

It should be noted moreover that when applied to the purpose of drafting thorough-tax rate reform legislation at this time, this flexible guide corroborates the general range of top rate goals of the preKorean war and one-third concepts.

Even though the current relation of Federal revenues to national income may be slightly higher than 38 percent, the pre-Korean top corporate rate, the very fact of launching a thorough program of taxrate reform, with forward scheduling and postponement procedures, would insure that Federal spending and retained revenues would not increase as national income grows over the year immediately ahead. Thus, over the life of such legislation, the revenue to national income ratio would decline significantly.

A specific program

Putting all of the factors which I have recounted together, I would suggest for the consideration of this panel and of this committee a program which, over a 5-year period, would bring the top rates of both individual and corporate tax down to 38 percent. This would mean a 14-percentage point reduction in the combined top rate of corporate tax, or an average of 3 percentage points a year minus 1. The first bracket rate of individual tax would be reduced 1 percentage point a year, or from 20 to 15 percent.

The reason for choosing 38 percent, instead of 37 percent which would permit an even 3-percentage point reduction a year in the corporate rate, is that a 38-percent top rate in regard to the individual tax would permit a uniform progression of 1 percentage point between each bracket without tampering with the taxable income bracket system itself.

I am attaching table II to illustrate how the rates could be reduced, and compressed insofar as graduated rates are concercned, in five annual steps in a reasonably consistent and uniform pattern.

I am not offering here a final conclusion in regard to the distribution of the reductions in the corporate tax between the normal and the surtax rates, now 30 to 22 percent, respectively. The existence of these two rates, instead of a flat rate on all corporate income, is a cause of complexity in the code and enforcement difficulties. Ideally, especially in regard to any possibility which might exist for ultimate coordination of the individual and corporate taxes with the same flat rate of tax, it would be sound policy to apply most of the 14 percentage points to reduction in the surtax rate. However, reduction of 7 percentage points in each of the rates would roughly maintain the present relationship.

Revenue effect

The revenue effect of the reductions in the individual tax rates, stated in terms of tax savings, is shown by taxable income brackets in table III attached. Based on 1957 income levels, the revenue effect of the individual tax reductions over 5 years would be $10.8 billion, of which 49.7 percent would fall in the first bracket of taxable income and 68.9 percent in the first three brackets combined. For purposes here, it is assumed that the revenue effect of putting these reductions into effect would be about $2.2 billion a year over a 5-year period.

Taking the estimate of $450 million of revenue from each percentage point of corporate tax, the reduction in the corporate rates over 5 years would have a total revenue effect of $6.3 billion. Thus, the total revenue effect over 5 years of the entire package would be in the range of $17.1 billion, or $3.4 billion a year.

If, upon enactment of such a program, the economy should grow at an average rate of 4 percent a year, producing $4 billion in revenue gain, this program could be put into effect with an annual average surplus of gain over reform dollars in the order of $600 million.

In actual operation, the revenue effect of such a program would be spread out over 6 fiscal years instead of 5. For practical reasons involving the relations of forward scheduling and postponement, the statutory reductions should be as of January 1, not July 1. The total revenue effect of a corporate tax reduction effective January 1 is felt in the fiscal year beginning the following July.

Hence, all of the effect of one annual reduction in the corporate rates, would be pushed ahead to the sixth instead of fifth fiscal year after enactment.

If such legislation were enacted retroactively to January 1, there also would be some pushing forward of the revenue effect in regard to the individual tax. However, the adjustment in this respect would take place completely in the following fiscal year, meaning that the total revenue effect for the individual reductions in the second year would be higher than in other years. This additional revenue impact probably would be completely offset by limiting the first reduction in the corporate tax to 2 percentage points, leaving the 3-percentage point reductions for each of the following years.

Precautions

In considering the relationships set forth in this program, there are four factors which should be kept in mind.

1. The impact on the income of individuals of the reformed rate structure must be found in the relative burden of taxes after reform and not in the relative percentage cuts applied to preexisting rates. In this respect, it is sometimes overlooked that the measure of regressivity, proportionality or progressivity of a tax is the relation of tax to total income; not the tax base, or taxable income as regards the personal income tax. The relation is known as the "effective" rate of tax. For a given rate scale, effective rates of tax as compared with statutory rates are relatively smaller in the lower income levels than in regard to higher incomes. This is so because exemptions and dependency credits provide substantial protection from payment of tax to people in the lower income brackets; a protection however which becomes relatively unimportant as income increases.

The following table shows the effective rates of tax which would result from a first bracket rate of 15 percent-without including the graduated rates-for incomes of a four-person family:

Under the program which I have outlined, the graduated rates above the basic rate shown in the table would greatly increase the upward sweep of effective graduation. Actually, there would be a range of effective tax from 1.5 percent of income at the bottom up toward 35 percent at the top.

2. There should be no misunderstanding as to what is meant in reference to shifting of tax burdens. The word "shifting" as used here means that the absolute tax burdens of some taxpayers are increased to correspond with reductions afforded to other taxpayers. In referring to the reshuffling approach to tax rate reform, without overall reduction in tax burdens, I have noted that the results would be meaningful in relation to economic growth only if there was a downward shift in tax burdens. No such shifting would take place in a program preempting revenue gain for income tax rate reform. It is true that, in the effectuation of the program, there would be a change in the relative position of taxpayers. In other words, more of the revenue from the individual tax would come from the first bracket rate, and less from the graduated part of the structure. But there would be no shifting of tax burdens as such. To the contrary, over the life of the program which I have outlined herein, there would