And of course, it is the consumers' durable goods, primarily, that are affected by the income that would be released to consumer channels by this tax reduction.

(3) Changes in net exports: Similarly, production for net export creates current purchasing power unmatched by a civilian supply. Even with the full Marshall plan, these net exports will drop by about $1,500,000,000 in 1948.

(4) Changes in the value of residential construction: While the consumer customarily pays down only a small part of the purchase price of a new home, the whole value of the construction represents income to someone. The value of residential building was about $5,100,000,000 in 1947, it will rise to about $6,000,000,000 in 1948 if the Knutson bill becomes law, and clear indications are given that rent control is nearing an end. It will drop to $4,500,000,000 if present taxes are continued and strict rent control is extended for a long period of time.

(5) Changes in hourly earnings or wage level:

Wages and salaries form important parts of personal income. When fully reflected in personal income.

In preliminary negotiations with management, labor is asking for substantial increases. At the same time, there is an attitude on the part of management not to grant substantial increases. Increases of about 4 percent over the year-end level could be absorbed by most industries without price increases, provided labor output improved sufficiently to reflect the heavy capital expenditures of 1947. And you will recall from the chart that these expenditures were about 17.7 billion dollars, or 10 times the 1933 level.

Such an increase in hourly earnings, that is, 4 percent over the year end, without price increases or increased output per man-hour would substantially lower profits and consequently the demand for producers' durable equipment and plant construction late in 1948.

An increase in hourly earnings of 4 percent over the year-end level under conditions of continued full employment would add about 4.5 billion dollars to personal income in 1948.

(6) Changes in the difference between income and expenditures of Government: Deficit spending tends to add to incomes in excess of the amount required to purchase consumer goods and services at current prices. On the other hand, debt retirement converts current income into savings which may or may not find an immediate outlet if the debt of individuals or corporations other than banks is retired. If bank debt is retired, the purchasing power may simply be extinguished. Fiscal policy becomes commandingly important whenever the sum of the changes (1) to (5), as listed above, is about zero.

Summing up, we find, under the assumption that the Knutson bill becomes law, that the changes (1) to (5) will be plus 2.8 billion dollars. This means that Federal debt could be retired at this rate without contributing to inflation, or, for that matter, to deflation. A higher rate of debt reduction would be deflationary, and might be disastrous. A rate equal to $10,000,000,000 per year which would be the starting rate for 1948 without reduction in tax rates, would be so deflationary as to cause severe unemployment and a decline in personal income.

Today the United States economy is in excellent health except for present fiscal and credit policies. Both threaten severe deflation and

unemployment. In other words, the economic future of this country is peculiarly today in the hands of this committee.

We have made forecasts of income and employment under the two assumptions regarding taxes: (1) If the Knutson bill is passed by the Senate and becomes law; and (2) no tax reduction materializes.

In the first case, that is, the Knutson bill becomes law, we forecast that personal income, which was at an annual rate of about 209.7 billion dollars in December 1947, will average about 210 billion dollars during the calendar year 1948. In the second case, that is, if there is no tax reduction, personal incomes will average about 200 billion dollars, or $250 per family less. The wide difference is due to the collateral adverse effect on employment in the durable-goods industries, brought about by the sharply deflationary effect of the debt retirement at a suicidal rate.

The CHAIRMAN. Are there any questions?

Senator MARTIN. Mr. Chairman, I would like a little further explanation as to the statement where you indicate that if we reduce the debt at the rate of $10,000,000,000 per year, there would be great deflation and unemployment.

What amount, in your opinion, could be reduced from the public debt per year without danger of deflation?

Dr. Roos. That is a figure which I have already given by implication. You can calculate it by summing of the figures (1) to (5). This sum is 2,800,000,000 for 1948, sir. That would be the outside figure. Anything faster rate of debt retirement than that would create unemployment.

The CHAIRMAN. It is a rather interesting coincidence, and I think a coincidence only, that the Senate viewpoint is that we ought to retire not less than $2,600,000,000 a year.

of

Dr. Roos. That is interesting. That rate would be right. But course, the real problem lies in the correct forecast of income.

The CHAIRMAN. Do you agree, Dr. Roos, that it is impossible in advance to set out a rigid formula of debt retirement? I do not mean that it is impossible, but that it is unwise to set out in advance a rigid debt retirement formula to run over a period of years?

Dr. Roos. I think it is very unwise, because your conditions are going to change from year to year. For example, my organizations would never attempt to forecast incomes much beyond a year and a half. On the other hand we would not hesitate to forecast production for 4 or 5 years, but in the case of income there are many collateral factors that can cause inflation or deflation; that is, a balance one way or the other could change the ideal amount of debt reduction, throw it, you see.

While these unpredictable factors do not affect production very much, you see, they do affect income materially.

Let me just illustrate what I mean by that statement.

In our postwar forecasts, which I mentioned here, we have the Federal Reserve index at 170. Now, a little after that forecast of production was made, that is about November 1944, we assumed that wages would be advanced between 5 and 10 percent, or about 72 on the average, after the war. On that basis, we forecast the personal incomes on the old series.

You remember, the income series was revised last June. For the old series, we had personal income declining to about $145,000,000,000 within 6 months after the war ended, and then rising again to around $150,000,000,000 or thereabouts.

Now, about a year after that forecast was made, the idea was advanced that labor could get substantial wage advances without affecting prices. That was, I believe, early in November of 1945, after the war was over. At that point, 2 days after the President's discussion of this proposition we advised clients that they would have to prepare for substantial inflation lasting 2 to 3 years.

At that point, we raised our forecast of income for 1946 to $165,000,000,000, which was a very substantial raise. No change was made in our production forecast despite this very substantial change in the income estimate.

The CHAIRMAN. Out of your wide experience, have you ever known of any instances where there were substantial wage increases in competitive industry that did not reflect in higher prices?

Dr. Roos. Not substantial increases, no. There have been moderate ones from time to time which were offset by (1) increased outlays for machinery, better and more efficient machinery; and (2) better selection of workers.

The CHAIRMAN. But wide-scale? Substantial increases? Is there any known experience in history, in a so-called free economy, where that did not reflect in increased prices?

Dr. Roos. No, sir. The timing, though, is important. Remember, in 1937 industry experienced a big wage increase; but it came after there were already excessive inventories, and after we were already deflating bank credit. And the wage increase was offset by declining activity due to excessive inventories in the other industries, and by the elimination of high cost producers and unemployment.

The CHAIRMAN. I notice your forecast for calendar 1948. Have you made any forecasts for the first 16 months of 1949?

Dr. Roos. Yes; we have several forecasts based on different assumptions with respect to taxes.

The CHAIRMAN. Out of your memory, could you give us something as to the first 6 months of 1949 ?

Dr. Roos. Yes; if there is no tax reduction, our forecast is that by June of 1949, income will have declined 10 percent from the present level. It would take such a decline to bring about an economic balance again. That would mean personal income of about $190,000,000,000.

The CHAIRMAN. And with tax reduction?

Dr. Roos. Personal income would be about $210 to $212 billion under the Knutsen bill.

The CHAIRMAN. About $210 to $212?

Dr. Roos. That is right.

Senator MARTIN. What effect will that have upon the tax return on the present tax basis?

Dr. Roos. Well, the yield would be much higher with a higher income than it is today.

Senator MARTIN. But you are talking about a decrease in income to $190,000,000,000.

Dr. Roos. That would eliminate the Federal surplus. I mean that the economy itself would correct the attempt to retire debt too rapidly through lower employment and lower employment. The decline in income and employment would be the result of failure to reduce taxes. The CHAIRMAN. In other words, the income-tax reduction takes that jeopardy out of the system.

Dr. Koos. That is right, sir.

Senator GEORGE. What, Doctor, based on your studies, would be the effect of increasing the corporate tax burden, say, $3,200,000,000? Dr. Roos. Well, sir, if you refer to chart 2, here you see that an increase in the corporate tax of $3,200,000,000, which I think was your figure, would have the effect of reducing corporate earnings by a little better than 20 percent. And that, in turn, according to the calculation of the chart, would lower the producers' durable equipment expenditures by the same amount, or about 20 percent, that is by about $3,500,000,000.

There would be a similar adverse effect on plant construction, which is not included in the above calculation, because lower profit would also lower the construction activity. This would be only the net effect, aside from the deflationary effects, which would remain if you only shifted the tax burden and would take in the same amount of revenue income. You would tend, however, to place the burden of adjustment to lower income levels on the durable goods and construction industries. The real problem is not where you put the taxes, but how much they are in relation to income.

Senator MARTIN. Doctor, do you feel that by a proper regulation of debt reduction and taxation we could avoid deflation or inflation? Dr. Roos. Yes, sir. I am convinced, sir, that industry has reached the point now, where if you give it a reasonable break on fiscal policy it will do a good job of inventory control.

Now, that is quite different than it was years ago. Years ago industry did not have figures necessary for this control. Also it did not know how to use the figures that were available. There was no way of industry knowing what it was doing to create economic unbalances. But I am convinced that today industry could do a pretty good job. And if you added fiscal policy which is designed to stabilize, I believe that the economy will stabilize around an upward trend.

That is, you will avoid the ups and downs of income and unemployment and these dramatic harmful changes in conditions.

Senator MARTIN. Well, for example, to go entirely outside the tax field, we have appealed to industry to police their output, to keep it out of the gray and black markets. But they have not done it.

Dr. Roos. Without appearing to defend industry but merely to enlighten, I would like to suggest something.

Senator MARTIN. I am just asking for information. Because on the sound economy of America depends the position of the world. Outside of two or three little countries, we are the only country that is in anywhere near sound financial position.

Dr. Roos. Yes, sir.

Now, the real problem is this: While there is good national information, local market information is practically lacking. For example, there is no publicly available series on county income or income of a sales region. There is no publicly available information on build

ing contracts or many other series by local areas by means of which industry can make decisions.

For lack of that information, industry tends to allocate sales, in a tight situation, on the basis of past demand. That very method alone can establish these gray markets. Here is how it would work:

A dealer is located in an area that has lost population, and so has no need at all for housing, let us say. And the building-matèrials suppliers look at the situation only as it was prewar, when the sales were so-and-so. On the basis of this obsolete information they allocate product. Since the companies have no current information, they ship to the dealer on basis of what he used to buy. The dealer in turn has no local market, and so he advertises, for example, so many thousand feet of this, and so much of that, for sale.

A buyer comes along, say, all the way across the United States and picks up the materials. It is then in the gray markets.

Local market information is a crying need of industry. The Government bureaus do not supply it,

Senator MARTIN. Doctor, do you not believe that if the Government would get out of that field of information, of necessity industry would furnish itself? And such information, in my humble judgment, is very much better than the information that is put out by these bureaus here in Washington.

Dr. Roos. That is true to a certain extent. I am rather on the spot, in speaking here, because my organizations do get out much private information of a local nature.

Senator MARTIN. I do not want to put the doctor on the spot, but he is giving me, to my mind, some very valuable information, which I cannot only use in this committee but in other committees on which I am serving.

This use of historical production, of the production serving a certain community, is not very sound.

Dr. Roos. It is not very sound, sir; no.

Senator MARTIN. Take, for example, if you may permit, Mr. Chairman, this situation:

In the central west there has been a great improvement in farm machinery which saves labor, a very necessary thing on the farm. But there was no historic formula whatsoever. So those people do not have any steel, and we are trying now to work out some way in which those new plants can be supplied. They are going to employ a good many men, and they are going to help agriculture in the great central

west.

That is where we get the greatest amount of our food. So the historical plan, the historical formula, does not work so well in many cases, we have found.

Dr. Roos. That is correct.

The CHAIRMAN. Doctor, the Secretary of the Treasury has described a $200,000,000,000 income as representing current levels. Is there any support for that?

Dr. Roos. No, sir. I mean, the current level is about $209,700,000,000. That is quite a way from $200,000,000,000.

The CHAIRMAN. What was our last monthly report?

Dr. Roos. $209,700,000,000 for December.

The CHAIRMAN. 209.7 billion dollars for December. So it is unrealistic to talk about a $200,000,000,000 current level.