increasing loans and deposits and thus a demand for more gold reserves. Consequently, in countries long ago established on the gold standard there will be a steadily increasing demand for gold as exchanges expand. We find thus a special characteristic of the demand for gold (certainly not existing in the demand for silver). The power of developing countries to soak up new gold is as marked a part of present conditions as is the power of a porous and sandy soil to soak up a heavy rainfall. We must, therefore, take full account of the noticeable fact that the recent demand for gold seems about to keep pace with the new supply; that a shipment of gold from the mines to London is today eagerly competed for, not only by European countries, but by Egypt, India, Turkey, Argentina, and Brazil.

Consequently it may be of interest to see which countries have taken the largest amounts of gold into their stocks since 1895:

Besides the demand for gold in the arts, and the apparent monetary demand, as thus already presented, we must not omit to take into account also the large stocks of gold held by banks and institutions which publish no statements. In the hands of large private institutions like those of the Rothschilds, Bleichroders, and others, great amounts of gold are carried. It is from such stores that the needs of states, such as Austria-Hungary, France, Italy, and even the United States (in Cleveland's administration), have been supplied without drawing down visible

reserves.

Thus far, then, we have examined the one factor of demand for gold, among the "other things" (which were supposed to remain equal). There is abundant evidence to show that the demand for gold, in this recent period of rising prices (18961909) has been as strong as, or even stronger than, the demand for gold in the previous period (1873-1896) of falling prices.

VI

It looks very much as if we must seek for the causes of rising prices since 1896 in some of the "other things" not yet examined.

There is no time, however, for extended discussion on these points [such as Part I, b, 2), 3), 4), 5)].

In regard to Part I, b, 2) the effects of Tariffs and Taxation, Unionism and higher Wages, and changing Agricultural Conditions in increasing expenses of production in all industries are so patent as to require no enlargement. Immediately after the passage of the Dingley Act in 1897, a large list of articles rose in price precipitously. Moreover, just so far as higher money wages for the same work, or the same money wages for a reduced number of hours, have been granted without a corresponding increase in the efficiency of the labor, the expenses of producing goods in general-and consequently prices-have risen. But, without doubt, one of the most important factors in raising prices directly and indirectly-has been the increased price of food due to the changing conditions of agriculture. This most influential cause of higher prices is one of the "other things" which has been at work quite independent of the quantity of new gold. Moreover, the indirect effect of high prices of food produces the most serious practical problem. It wipes out all the gain of previous increases of wages, and drives laborers to repeat their demands for higher pay, thus working again to increase expenses of production. It is not too much to say that the gains of industry, shown by the fall in prices, as they stood about 1890 have been lost to us by the high tariffs of 1897 and the wastes of bad farming and the recent high costs of agriculture.

Our analysis would be inadequate, however, if we stopped here with our examination of expenses of production. The really practical problem is still before us in trying to analyze the forces at work fixing prices in that vague and dangerous margin between actual expenses of production and the prices in fact paid by the consumer. It is in this margin that we find in operation the "other things" mentioned in Part I, b, 3), 4), and 5). On these points I must necessarily be brief.

The whole raison d'être of monopolistic combinations is to control prices, and prevent active competition. As every economist knows, in the conditions under which many industries are today organized, expenses of production have no direct relation. to prices. In such conditions, there is a field in which the policy of charging "what the traffic will bear" prevails; and this includes industries that are not public utilities.

Furthermore, Part I, b, 4), we must face the fact of increasing riches not only in this country, but all over the world. New wealth makes a liberal spender. The retail dealer finding his expenses increasing and-even when they are not tries the experiment of charging his richer customers an increasing price. The newly rich pay and do not feel it. But what can the poorer unorganized buyer do when retail prices are raised? What can he do if his meat bill, or his plumbing-repairs bill, rises enormously? The extravagance of the rich has increased the cost of traveling, the rates at hotels, the fees, the luxury of steamships and automobiles, the consumption of fruits and vegetables out of season once never thought of, and has generally raised the standard of expenditure. Those of smaller income find they also must pay the higher prices. Thus we have reached a point where we have to pay almost whatever anyone asks. Organized buyers are the only offset to organized sellers.

Moreover, rising prices due to high expenses of production, or to combinations of sellers, present a paradise for speculation (Part I, b, 5)). A movement upward based on facts can be easily converted into a further rise based only on speculative manipulation. A rise of prices which brings large profits to a combination, thus directly affects earnings and gives especial opportunity to speculation in the securities of industrials. Hence, the field of speculation spreads from commodities (Part I) to securities (Part II). The facts as to the movement of prices of securities are well shown in Brookmire's Economic Charts since 1885; and, while the presence of gold serves as a fund of lawful money in reserves, the spread of speculation has gone on seemingly unaffected by the new supplies of gold. That is, speculative conditions may arise and disappear antecedent to and seemingly independent of the gold supplies.

RECENT CHANGES IN PRICE LEVELS AND THEIR

CAUSES.

By IRVING FISHER.

I find myself unable to agree with most of the positions taken by Professor Laughlin in his able paper. In my opinion the old quantity theory is in essence correct. What it needs is to be restated, not rejected. I have attempted to make what I believe to be the needed restatement in a forthcoming book on “The Purchasing Power of Money", Chapter XII of which I have had reprinted for distribution at this meeting.' I shall confine myself chiefly in the paper this morning to describing the contents of this chapter. The chapter does not attempt to discuss the theoretical determination of price levels, as this has been discussed in previous chapters. The aim, rather, is to work out quantitatively the statistics pertaining to the rise of prices during the last fifteen years.

Before, however, considering these statistics I may state, in a few words, my creed as to the causation of price levels. This is, according to my view, entirely distinct from the causation of the price of any individual commodity. It is just as impossible to determine the general level of prices by the supply and demand of individual commodities as to determine the general tidal level of the ocean by the winds affecting individual waves. Waves and tides are distinct and require distinct explanations. Likewise prices and pricelevels are distinct and require distinct explanations. Just as each wave presupposes a general tidal level with reference to which it is measured, so the supply and demand of each individual commodity presupposes a general level of prices. Each supplier and demander expresses his supply and demand in terms of money and he does so on the assumption of a given purchasing power of money. With a change in the purchasing power there will be a change in his particular supply and demand. Thus the discussion of individual prices presupposes a general price level. The proper order of study is from a general price level to particular prices rather than from particular prices to a general price level.

1

By the courtesy of the Macmillan Company we are permitted to publish this material and to make use of the diagrams contained therein.

Supply and demand are terms which help in the discussion of individual prices but not in the consideration of the general price level. The latter is determined in a simpler way, namely, by the equation of exchange, MV + M'V' = PT. This equation expresses algebraically the old quantity theory of money, with some elaboration. It is discussed in Professor Simon Newcomb's able and interesting work "Principles of Political Economy", in President Hadley's "Economics", in Professor Kemmerer's "Money and Prices", and elsewhere. In this equation M signifies the quantity of money in circulation; V, its velocity of circulation, or rate of turnover per annum; M', the volume of deposits subject to check; and V', its velocity of circulation, or rate of turnover per annum. T signifies the volume of trade (considered irrespective of the price level; in other words, reckoned for a given price level as an assumed base); and P the general level of prices resulting from the other five magnitudes in the equation. While it is true that the equation does not enable us to judge which of the magnitudes in it are causes, and which are effects, it is possible by other considerations, which I shall not here attempt to discuss, to show that P is the one passive element in the equation—that it is not cause, but effect. Under normal circumstances, and apart from transition periods, an increase in M, or the quantity of money in circulation, will bring about a corresponding increase in M', the volume of deposits subject to check, but will not disturb V, V', or T. It follows, therefore, that P must vary in direct proportion to M. All the possible causal relations between the magnitudes in the equation are fully discussed in various chapters of the book to which I have referred, but the upshot of the matter is that the old quantity theory still remains fundamentally true and that, under normal conditions, the general price level will respond in substantial proportion to the volume of money in circulation. This does not assert, of course, that during any historical period M will be the only factor affecting P; for it usually happens that all of the five factors in the equation which do affect P will change simultaneously. Each of these factors is the result of innumerable antecedent causes but no cause can affect the price level except through one or more of the five factors M, M', V, V', T.

One object of this paper is to show historically what have been the changes in all of these five factors during the last fifteen