Imagens das páginas

hands about 24 times in the year, thus effecting 2.10 X 24 or 50 billions of exchanges; the volume of deposits subject to check was 11.88 billions and changed hands about 60.2 times in the year, thus effecting 11.83 X 60.2 or 712 billions of exchanges, the two together making a total of 50+ 712 or 762 billions. This paid for a volume of trade of 595 billion units (a "unit" of trade being that amount of goods which in the base year, 1909, was worth $1), at prices 28 per cent higher than the prices of said base year, so that 595 X 128 per cent is also 762.

The large diagram affords full comparison (both visual and numerical) between the above-named figures for 1916 and the corresponding figures for previous years. It will be noted that between 1915 and 1916 there was, as forecasted last year, a general expansion in all the magnitudes of the equation. This expansion, in the case of bank deposits, was more than 25 per cent. In the other magnitudes it ranged from 10 to 20 per cent.

As was the case last year, though not so markedly in 1916, a large part of the increase in the volume of trade was doubtless due to the great expansion of the trade in securities.

The smaller diagram continues that given two years ago, and gives in greater detail the same data at the six dates each year for which the comptroller's figures for national banks are available. The first three dates1 were before the outbreak of war in Europe.

As noted two years ago, the various magnitudes of the equation of exchange varied little in the first three periods of 1914, preceding the war, but after its outbreak, they fluctuated violently. There were sudden contractions followed by gradual expansions.

The volume of trade showed a sudden fall following the outbreak of the war and a subsequent full recovery distributed through the remainder of 1914. The velocities of circulation showed the same sudden contraction with subsequent rapid recovery.

The money in circulation showed an emergency expansion during the last half of 1914, with a restoration to normal in 1915, and a ten

1 The exact dates for the comptroller's data are as follows:

For 1914, Jan. 13, March 4, June 30, Sept. 12, Oct. 31, Dec. 31.
For 1915, March 4, May 1, June 23, Sept. 2, Nov. 10, Dec. 31.

For 1916, March 7, May 1, June 30, Sept. 12, Nov. 17, Dec. 27. As indicated in the heading of the smaller diagram, all the rates are rates per year, not per month. The method of calculation is the same as that used two years ago and described in this REVIEW for June, 1915.

2 As to money this is only inferential, on the assumption that its velocity followed that of deposits.

dency to expand towards the end of 1916-a tendency which has been continued in 1917,3 i.e., beyond the period covered by the diagram.

The deposits, on the other hand, executed an opposite movement, contracting soon after the war began, and expanding during 1916 (and 1917).*

These various changes above noted are closely associated with a third magnitude, namely, the money in banks, that is, the (cash) bank reserves which are not represented specifically in the diagram but which are given (in billions) in the following table:

[blocks in formation]

These figures and those in the diagram indicate that the shock of war produced a displacement of cash from banks into private pockets and tills, followed by a reaction.

The net effect of all the various changes above mentioned on the scale of prices was trifling during 1914 and the first two thirds of 1915. Since that time there has been a rapid rise to the present time.

The chief factor in this rise has been the quantity of money in circulation and in banks, and the increase in these has been chiefly due to the gold imported from abroad.

8 Although not all the data needed to establish this conclusion are as yet available it is a fair inference from the two facts (1) that the total money in the United States outside of the U. S. Treasury has increased since 1916 (the figures for the first months of 1917 being successively (in billions): 4.44; 4.50; 4.58; 4.70; 4.74; 4.73; as against a previous maximum of 4.44 in Dec. 1916), and (2) that the part of the above which consisted of the money in national banks did not decrease (being .89 in March, 1917, and .84 in May, 1917, as against an average in 1916 of .86).

♦ Inferable from the fact that the part of these individual deposits subject to check which were in national banks are known to have expanded (being 6.37 from March, 1917, and 6.63 for May, 1917, as against a previous maximum of 6.35 in Nov., 1916).

The money in circulation outside of banks is the M in the diagram and acts on prices through MV. The money in banks is the basis for deposits (M') and so acts on prices through M'V'. In both of these, gold has been the dominant element.

Ordinarily the variations in this total money are no greater than those in velocities of circulation or in volume of trade, so that there is, ordinarily, no special or very evident correlation directly observable between the price level and money. But during the war the quantity of money has changed so much more markedly than velocities or trade that it has come to dominate the situation almost completely.

Accordingly, we find that latterly changes in the price level have followed changes in the money quantity (and especially changes in that part of this quantity which consists of gold) with very surprising closeness.

I include a diagram which shows these comparisons month by month." It shows, at the left: (1) a speculative "mark-up" in prices—such as is usually found at the outbreak of wars and which, as it soon proves out of tune with fundamental conditions, disappears: (2) the reduced circulation from the export of gold; and (3) immediately thereafter, the increased circulation from the issue of emergency currency.

After the middle of 1915 the agreement is so close that one can identify the exact corresponding points. We can even measure the lag. We find that a change in price level follows a change in total money after a lag of about two months. This became evident from the time, August, 1915, when the quantity of money first began its rapid increase. One month later prices began to shoot upward keeping almost exact pace with the quantity of money. The striking correspondence has continued up to date.

This close agreement, point by point, between the two upper curves, is shown by dotted lines and the lag, in each case, by the figures in circles, viz., 1, 22, 22, 2, 2, 1, 1 and 12 months respectively. Besides these eight very obvious cases of point-for-point correspondence we may count about a half dozen other less obvious ones; and besides these there is the very obvious general correspondence between the


I have used Dun's index number inasmuch as that of the Department of Labor month by month is not available up to date. The plotting is on a "ratio chart" and therefore gives, in the relative slopes of lines, the exact relative percentage changes of the various magnitudes. For a full description see my "The 'ratio' chart," Publications of the American Statistical Association, June 1917.

Since at least as early as August, 1915, there are only two points in Dun's curve (viz., February and March, 1916) which seem definitely out of tune with the other curves. Curiously enough these two discrepancies seem to be peculiar to Dun's index number. The other index numbers available (viz., Bradstreet's and that of the United States Bureau of Labor Statistics) do not show these discrepancies, although each has some other discrepancies of its own.

It must be remembered, of course, that any index number is only an imperfect measure of the actual price level, so that even a much rougher agreement than we find might well be expected.

It would be strange indeed if so close a correspondence were accidental. The chances, on any reasonable method of calculation, that the point-for-point correspondences should occur by accident are much below one in a hundred.

The agreement seems, therefore, a sufficient answer to the few who are still unwilling to concede any truth in the so-called "quantity theory" of money. Professor Cassell finds a somewhat similar correspondence in Russia, and Professor J. S. Nicholson of Edinburgh writes me that he is soon to publish a paper showing that "the index numbers of the Statist and Economist follow the Treasury notes by about three months."

The war has thus supplied a decisive laboratory experiment to test the long contested theories of money.

The great outstanding result is, so far as the United States is concerned, that the war has brought gold to us and that gold-inflation lies at the bottom of the spectacular rise in prices which we have recently experienced.

In the future, now that America has entered the war, we may expect greater and more varied effects. I anticipate, that, with the increased war expenditures and the consequent war loans, there will be a great expansion of deposits and a further great rise in the price level. A more drastic tax policy, as suggested by Professor Sprague of Harvard, would go far to retard and lessen this effect.

Yale University.


The Federal Land Bank of Springfield, Massachusetts, has issued a leaflet on How to Secure a Loan (pp. 4).

6 "The Present Situation of the Foreign Exchanges," by Professor Gustav Cassell, Economic Journal, vol. XXVI, no. 103 (Sept., 1916).

« AnteriorContinuar »