CHAPTER XV

THE QUANTITY THEORY: THE "PASSIVENESS OF PRICES"

Is the price-level passive? Is it true that while change may occur from causes outside the equation of exchange in volume of money, volume of trade, and velocities of circulation, a change in the price-level from causes outside the equation is impossible? Must the average of prices be a passive function of M, the V's, M' and T? Such is the general contention of the quantity theory, and such, very explicitly, is Fisher's contention. The price-level is always effect, and never cause (with slight modifications of the doctrine for transition periods) in its relations to the other magnitudes in the equation of exchange.

Now in one sense, it is my own contention that the price-_level_ can never be a _cause_ of anything. The price-level is an _average_. Averages may be _indicia_ of causation, but they are not themselves causes. They are not, in reality, anything _at all_. Causation is a matter which pertains to the particulars of which the average is made. But this is not the doctrine of the quantity theory. The quantity theory does, in certain connections, assign causal influence to the level of prices, particularly in the theory of foreign exchange, where the explanation of international gold movements rests on the doctrine that a price-level in one country, higher than the price-level of another country, drives money away.[332] It will be seen, in a moment, that Fisher relies on this principle to prove that the price-level of a country cannot rise without an increase of money--if it did so rise, it would drive out the money, and so be forced down again. The point at issue may be stated in terms of particular prices. The quantity theory is that, while particular prices may rise from causes affecting them, as compared with other prices, without a change in money, velocities, etc., still there cannot be a rise in the general average, because other prices will be obliged to go down to compensate. The issue is as to the possibility of a rise in particular prices, uncompensated by a corresponding fall in other particular prices, without a _prior_ increase in money, or velocities, or decrease in trade. I take up the issue in this form. I shall maintain that particular prices can, and do, rise, without a _prior_ increase in money or bank-deposits, or change in the volume of trade, or in velocity of money or deposits and also without compensating fall in other particular prices. Putting it in terms of Fisher's equation, I shall maintain, as against Fisher, that P can rise through the direct action of factors _outside_ the equation of exchange, that as a _consequence of such rise_ the other factors readjust themselves, and that a new equilibrium is reached which, in the absence of new disturbances from causes outside the equation, tends to be as permanent and stable as the old equilibrium was.

In the argument which follows, I shall respect thoroughly the distinction between "normal" and "transitional" effects. I do not think that this distinction is properly drawn by Fisher. In my discussion of the relation between the volume of bank-credit and the volume of trade, and in other connections, I have shown that Fisher leaves out of his normal theory most of the concrete factors which do affect both the concrete magnitudes, and the long run _averages_, of the factors in his own equation. But for the present, I shall meet him on his own ground, give his distinctions their fullest weight, and carry my argument through the "transition" to a point where no further change among the factors in the equation can be expected as a consequence of the initial change assumed.

Fisher's argument to show the passiveness of prices takes the form of a _reductio ad absurdum_. "To show the untenability of such an idea let us grant for the sake of argument that--in some other way than as effect of changes in M, M', V, V', and the Q's--the prices in (say) the United States are changed to (say) double the original level, and let us see what effect this will produce on the other magnitudes in the equation."[333] Then, if the equation of exchange is to be maintained, either M or M' or their velocities must be increased, or trade must be reduced. But he holds that none of these is possible. (1) Money will be reduced. High prices drive money away to other countries. Nor can gold come in via the mints. "No one will take bullion to the mints when he thereby loses half its value."[334] On the contrary, men will melt down coin. Nor will high prices stimulate mining. Rather, by raising the expenses of mining, they will discourage mining. (2) Bank-deposits cannot increase. Bank-deposits depend on the amount of money, and as that is reduced, they must be reduced, to keep their normal ratio to the volume of money. (3) The appeal to velocities is no more satisfactory. These have been already adjusted to individual convenience.[335] (4) Nor can trade be decreased. Since the average person will not only pay, but also receive, high prices, there is no reason why he should reduce his purchases. "_The price-level is normally the one absolutely passive element in the equation of exchange._"[336]

"But though it is a fallacy to think that the price-level in one community can, in the long run, affect the money in _that_ community, it is true that the price-level in one community may affect the money in _another_ community. This proposition has been repeatedly made use of in our discussion, and should be clearly distinguished from the fallacy above mentioned. The price-level in an outside community is an influence outside the equation of exchange of that community, and operates by affecting its money in circulation and not by directly affecting its price-level. _The price-level outside New York City, for instance, affects the price-level in New York City only_ via _changes in the money in New York City_."[337]...

"Were it not for the fanatical refusal of some economists to admit that the price-level is in ultimate analysis effect and not cause, we should not be at so great pains to prove it beyond cavil." To explain this "fanatical refusal," Fisher alludes to the "fallacious idea" that the equation of exchange cannot determine the price-level, because the price-level has already been determined by other causes, usually alluded to as "supply and demand." He urges, however, that supply and demand, cost of production, etc., relate, not to the price-level, but only to particular prices: that the price-level is a factor prior to, and independent of, the particular prices, and is presupposed by theories like supply and demand, cost of production, etc.[338]

The _reductio ad absurdum_, at first blush, looks impressive. One obvious criticism suggests itself, however, and it will be found to give a clue to a much more fundamental criticism: is it reasonable to assume a doubling of _all_ prices? Above all, must the assumption involve the doubling of the price of gold bullion? Part of the argument to show that gold bullion would not be minted rests on that assumption. But, more fundamental, for such an all round doubling of prices, no _cause_ could be assigned. Of course the hypothesis of an increase in prices without any cause is absurd, and Fisher easily disposes of it. But suppose we assign some _concrete causes_, outside the equation of exchange, which might affect prices, and see how the thing works then!

Fisher states on p. 95 that "other elements in the equation of exchange than money and commodities[339] cannot be transported from one place to another." And in the passage quoted above he maintains that price-levels in one country can influence price-levels in another country, or even price-levels in one city can influence price-levels in another city, only _via_ changes in money, in the second country or city. But other elements in the equation are _directly_ transferable, in fact. _Deposits_, _e. g._, in London, to the credit of New York bankers, may be transferred to Paris, directly, by _cable_ or by _letter_, and _prices_ are constantly being directly passed from one country or market to another by the same media. Let us suppose a strong case, to put our principle in relief. Assume an island, which produces a staple widely used, whose chief centre of production is outside the island. Assume that this staple, an agricultural product, rises greatly in price, owing to a blight, which promises to be permanent, in the main producing region. The blight does not affect the island, however. Let this product be the main product of our island, which we shall assume to be small. Let the island have communication with the outside world by boat only once in three months. Let it be, however, in constant communication by cable. Word comes by cable of the rise in the price in the staple. The staple at once rises in the island. No new money has come in to cause it. Will this be a rise in the price-level? Will there be compensating reductions in the prices of other things to leave the price-level unchanged? What prices can fall? Not the prices of goods that have been imported to the island, surely. They will rather tend to rise, because everybody on the island will feel richer than before, and will be disposed to buy more freely. Meanwhile, merchants and bankers on the island will be more ready to extend credit than before, so that they will be able to buy more freely. What else can fall? Not the prices of the land! Rather, the land will rise in price greatly, because the increased price of the staple, expected to be permanent, will promise bigger rents, and the price of the land, being a _capitalization_ of the annual rental, will rise very much more than anything else--it will rise to the extent of the capitalized price of the increase in the rents. Wages, likewise, will rise, since the price of the product of labor has risen. And the capital instruments in use in producing the staple will also rise, though not so much as land and wages, inasmuch as they can be brought in from outside at the end of three months. What is there that can fall--except, perhaps, such goods as are exclusively designed for the construction of poorhouses! A significant particular price rises--that is the first step; then, from causes familiar to all students of economics, other related prices rise; there is a general _sympathetic_ rise in prices, the _price-level_ has risen independently, from causes _outside the equation of exchange_. But now, can this rise sustain itself? Well, what can bring it down? When the ship comes, at the end of three months, it will bring in additional supplies of the articles of import, and they will go down to their old level. Will they go any lower than the old level? What is there to cause them to do so? The outside price-level should be higher now, rather than lower, since the _stock_ of the staple in question is reduced, and nothing else increased to compensate. Nor can any reason be assigned why other prices on the island: the staple in question, lands, wages, etc., should fall at all from the level they reached when the news first came.

Incidentally, our ship may also bring in more gold. The bankers, finding their deposits expanding, may feel it well to cable orders for more gold to increase their reserves, especially as they have been subject to somewhat unusual calls for cash for hand to hand circulation--though this last need they might well have been meeting by expanding their note issue.

Is there anything else to be said? Is not the new equilibrium stable? And is not the causal sequence precisely the reverse of that assigned by the quantity theory? _First_. a rise in prices; _second_, an expansion of credit, book-credit, notes and deposits; _third_, money comes in. If anyone is particularly anxious about the equation of exchange in this process, he may add to my expansion of credit an increase in velocities to keep it straight!

I may add that I see nothing in the "transition" I have described to cause trade to be reduced. Rather, I should expect the rising prices to make trade more active--or better, I should expect the rising _values_ of goods, etc., of which rising prices are the symptom, to make trade more active, particularly as there would be an increase in speculation to bring about readjustments, and to "discount" the prosperity. Nor can I find any reason why trade should be reduced below the old level in the new normal equilibrium. It would make no difference, however, if trade were reduced either transitionally or normally, since the point at issue is the possibility of a rise in prices originating from causes outside the equation of exchange, and compelling a readjustment of a permanent character in the other factors of the equation. The quantity theorist is at liberty to make this readjustment in any way he pleases. My point is made if he has to make the readjustment, and if the price-level stays up!

I have put my illustration in an extreme form to throw the whole thing in relief, and to make the demonstration free from a host of complexities. But is not the causal process essentially the same if we substitute, say, the Southern States for our island, and cotton for our staple? So long as the telegraph bringing news of the ruin of cotton production in India and Egypt, with the higher price of cotton, can come in ahead of the money that the quantity theorist might imagine rushing in a race with it on the train to be offered for the cotton, my point is made. In point of fact, there would be a general rise in prices and wages in the South, which, leading to an expansion of credit, would only gradually and in no definite ratio lead to an increase in money drawn from outside. Buyers outside would pay, not with money, but with checks drawn on New York, and Southern bankers would use their discretion as to how much actual cash they would bring in. With the elastic note issue of our Federal Reserve system, I see no reason to anticipate that money would be drawn to the South in an amount proportionate to the increase in prices. Even if it were, the causation would not run from money to prices, and that is the point at issue. If _rising_ prices can cause increasing money, the whole quantity theory is upset, whatever the proportions involved.

It will be noted that my illustration might be put partly in the form of the supply and demand argument. Increasing demand for cotton in the South leads to higher price of cotton; higher price of cotton makes cotton-growers richer, and enables them to increase their demand for imported goods, for land, and for labor. Supply and demand comes into conflict with the quantity theory, and does not suffer in the conflict! Supply and demand determine particular prices, and particular prices determine the price-level!

Now I wish to generalize this point. I shall show that the quantity theory conflicts with most of our doctrines of prices, as worked out in our systems of economics. I shall show that, in important cases, the quantity theory conflicts with the law of supply and demand, with the doctrine of cost of production, with the capitalization theory, and with the doctrine of imputation as worked out by the Austrians, whereby the prices of labor, land, and other agents of production rise or fall with the prices of the consumption goods which they produce. I shall show the conflict in important cases, and shall show also, in those cases, that it is not the quantity theory which can be sustained.

The general form of the conflict may be stated for all these theories. They are theories of the _relations_ of particular prices, concerned with showing that individual prices are so related that they tend to _vary together_. A rise in one price, according to these theories, tends to bring about _rises_ in others, and _vice versa_. The quantity theory, on the other hand, asserts a relation among individual prices such that a rise in one tends to bring about a _fall_ in others--it requires a _compensatory_ fall at one point, if there has been a rise somewhere else.

Let us take some cases. I shall take, first, the conflict between the quantity theory and the capitalization theory, as I can use the illustration just given in connection with it. I have, in a preceding chapter, given a statement of the capitalization theory. It is a theory concerned with the prices of long-time goods and income-bearers, as lands, houses, capital goods of various sorts that give forth their services through a series of years, stocks, bonds, etc. The prices of things of this sort, according to the capitalization[340] theory, depend on two factors: one, the money income expected from the income-bearer, the other, the prevailing rate of interest. This money income, except in the case of bonds, commonly depends on the prices of the products of the income-bearer, or (in the case of stocks) of the products of the concrete capital-goods to which the income-bearer gives title. If we may follow the Austrian division of goods into higher and lower "orders," or "ranks," we may say that the prices of the goods of higher ranks are the capitalizations of the prices of the goods of lower ranks specifically produced by them. Thus, concretely, if the price of wheat rises, we may expect the prices of land to rise, if the rate of interest remains the same. If the price of steel rises, we may expect the stocks of the U. S. Steel corporation to rise, also. If the prices of smokeless powder, and other war munitions soar, we may expect the prices of the stocks of the corporations involved to do precisely what they have done in the recent course of the stock market. All this, on the assumption that the rate of interest does not change, and that the risk factor remains constant. If these factors vary, the results will not present the mathematical exactitude that the formula calls for, but the general tendency will remain the same. On the other hand, if the incomes remain unchanged, but the rate of interest rises, then we may expect the capitalized prices to fall, and if the rate of interest falls, we may expect the capitalized prices to rise. From the standpoint of the present discussion, I suppose it might be fairest and best to state the capitalization theory on this point as Fisher himself states it. In his _Elementary Principles of Economics_ (ed. 1912) after giving a table showing in figures the difference made in different capital prices by different rates of interest (p. 125) he states (126): "If the value of the benefits derivable from these various articles continues in each case uniform, but the rate of interest is suddenly cut down from 5% to 2-1/2%, there will result a general increase in the capital values, but a very different increase for the different articles. The more enduring ones will be affected the most." And in his book, _The Rate of Interest_: "The orchard whose yield of apples should increase from $1,000 worth to $2,000 worth would itself correspondingly increase in value from, say, $20,000 to something like $40,000 and the ratio of the income to the capital value, would remain about as before, namely, 5%." (P. 15.) On the next page, he generalizes his notion: "One cannot escape this conclusion (as has sometimes been attempted) by supposing the increasing productivity to be universal. It has been asserted, in substance, that though an increase in the productivity of one orchard would not affect the total productivity of capital, and hence would not appreciably affect the rate of interest, yet, if the productivity of all the capital in the world could be doubled, the rate of interest would be doubled. It is true that doubling the productivity of the world's capital would not be entirely without effect upon the rate of interest; but this effect would not be in the simple direct ratio supposed. Indeed, an increase of the productivity of capital would probably result in a decrease, instead of an increase, of the rate of interest. _To double the productivity of capital might more than double the value of the capital._" (_Rate of Interest_, p. 16.)[341] Fisher reiterates this doctrine in his reply to Seager, in the _American Economic Review_, Sept. 1913, pp. 614-615.

Now my concern here is not with the points at issue as between Fisher and Seager: the "impatience" vs. the "productivity" theories of interest. For the present, I shall accept Fisher's doctrine on that point as true.[342] I am here interested in Fisher's doctrine that a doubling of the general productivity of capital would double, or more than double, the prices of capital instruments, including land. How is such a general rise in prices possible, if the quantity theory be true? Is not this a rise in general prices from causes outside the equation of exchange? That Fisher means the _money-prices_ of capital goods when he speaks of capital-values is perfectly clear. In the second quotation, he speaks of "capital-value of $40,000", and in general, his definition of value runs in terms of _price_ (_e. g., Purchasing Power of Money,_ pp. 3-4, and _Elementary Principles_, p. 17). Fisher has no absolute value concept in his system. We have in the passages cited two doctrines, both of which contradict the quantity theory: (1) that a reduction in the rate of interest will raise capital-prices (which are the largest factor by far in the price-level), and (2) that an increase in the product of capital goods means, not only more money paid for the products, but also more money paid for the production-goods. Incidentally, the general imputation theory would call for more money paid to laborers as well. How can all this be, on the quantity theory? And what can the poor equation of exchange do in such a case, if money does not increase, if bank-credit is limited by money, if velocities of circulation are fixed by individual habits and convenience, if trade _increases_ as a consequence of the increased number of goods produced, and if prices rise? It will not help much to assume that the productivity of gold mines is doubled also. The quantity of money does not depend very much on the annual production of gold. Besides, money need not, from the standpoint of the quantity theory, be made of gold. It might be irredeemable Greenbacks, fixed in quantity by law, or even dodo-bones! Would not the capitalization theory apply in the Greenback Period? I shall not try to solve the riddle. I am not responsible for it!

The conflict between the capitalization theory and the quantity theory may be more simply stated. Assume that the prices of consumers' goods and services rise, quantity of money and volume of exchanges remaining unchanged. On the quantity theory, other prices, the prices of producers' goods and services, lands, and securities, would have to come down enough to compensate, in order that the price-level might remain unchanged. For the capitalization theory, however, the prices of lands, securities, and long time capital goods in general would have to rise, since the incomes on which they are based have risen. Wages of labor engaged in making consumers' goods would also have to rise, on the general imputation theory.

The quantity theory conflicts with the capitalization theory. The quantity theory as presented by Fisher conflicts with the capitalization theory as presented by Fisher. Which theory is true? Would prices rise thus, or would they be held down in some way by the limitations on the quantity of money? I hold that I have already proved, in the reasoning given in connection with my hypothetical island, and in the case of the South with its cotton, that the capitalization theory tendency would prevail. The prices of products rise, and then the prices of the labor, land, and other capital goods which have produced them, rise, the rise in the prices of the capital goods behaving in accordance with the laws of the capitalization theory, and all of the rises after the initial rise in products being in accordance with the imputation theory of the Austrians.

This conflict suggests an interesting point. Various elements in our economic theory, added from time to time by different writers, have necessarily come from different philosophical and sociological view-points, and have behind them different philosophical, psychological, and sociological assumptions. The quantity theory, developing, as shown in the chapter on "Supply and Demand and the Value of Money," largely in isolation from the general body of economic theory, has a background of psychological and sociological assumptions quite different from that of many other doctrines. In the chapter on "Dodo-Bones," I stated these assumptions. The quantity theory rests in a psychology of blind habit. It assumes a rigidity in the social system such that it might be likened to a machine, with a hopper into which money is poured, which grinds out prices at the other end. I set this in contrast with the psychological assumptions underlying the commodity theory of money. That theory rests on the "banker's psychology." It assumes a highly reflective and calculating attitude on the part of economic men, with the disposition to look behind appearances for the security, to test things out, to get to bedrock in business affairs. Now the capitalization theory likewise assumes this banker's psychology. In its refinements, as represented by the mathematical formulae in the appendices of Fisher's _Rate of Interest_, it assumes a degree of precision in business calculation which few experts in bond departments apply, and which the highly fluid and alert dealers in Wall Street certainly have not time for, even if they had that degree of mathematical knowledge! In practice, it need not be said, particularly in the case of the prices of lands, the capitalization theory finds its predictions very imperfectly realized! But the two theories, resting in such divergent psychological assumptions, may be expected, _a priori_, to conflict. That they do conflict is not remarkable.

I shall show a similar conflict between the quantity theory and the law of costs. In general, the quantity theorist thinks that he has reconciled his theory with cost theory by pointing out that reduced costs manifest themselves in increasing production, which means increasing trade, which should, on the quantity theory, mean lower prices.[343] I need not, for my purposes, analyze this doctrine in detail, though I am disposed to consider it an accident that the two theories converge at this point. For the present, I shall analyze a case where reducing costs actually come as a consequence of the _reduction_ in the volume of trade, and inquire whether such a case will lead, as the cost theory would assert, to lowered general prices, or, as the quantity theory would assert, to _higher_ general prices. The case is that where by improved methods of handling goods, it is possible to dispense with middlemen. Concretely, assume that retailers of milk get in direct touch with dairymen, so that middlemen are eliminated, and that as a consequence the price of milk is reduced two cents a quart. What of the general price-level? T (trade) is reduced. There are less exchanges. Volume of trade does not mean volume of goods _produced_, but volume of _exchanges_. With a reduced trade, the quantity theory must assert that prices of commodities other than milk must, on the average, rise, not merely enough to compensate for the fall in milk, but more than that, enough to compensate for the reduced trade as well. But how can the other prices rise? Well, a point comes up obviously: the buyers of milk save two cents a quart. They can spend it for something else. This will raise the prices of other things. But, on the other hand, the middlemen now have less to spend. They have _exactly as much less_ as the others have _more_, the extra money that milk buyers have being, in fact, the money that the middlemen would otherwise have had. The one offsets the other. There is, then, no reason for the average of other prices to rise. Suppose we carry the process one step further. After a while, the middleman will find other work to do. Then they will have incomes again to spend. But in going to work again, they will be engaged in production, and so will, in general, be increasing the volume of trade. The quantity theorist could not expect a rise in prices from this!

And here we are given a clue to a fundamental confusion in the quantity theory, a confusion which, accepted by the reader, gives the quantity theory much of its plausibility. I refer to the confusion between _volume of money_, and volume of _money-income_.[344] The two need not be the same. The two generally are not the same. In the case I have described, the one has changed without a change in the other. Now if one wishes to view the process of price-causation from the standpoint of money offered for goods,--an essentially superficial,[345] but frequently useful, view-point--it is clearly money-_income_, rather than mere quantity of money in the country that is important. Into the determination of volume of money-income, however, come factors of a high degree of complexity, among them, prices for which there is no possible place within the confines of so simple and mechanical a doctrine as the quantity theory.

In passing, I notice a point to which I called attention in discussing Fisher's factors in the equation of exchange. I refer to his definition of velocity of circulation as the average of "person-turnovers" of money.[346] In the illustration given, there is no reason to suppose that this average is changed. The middlemen simply drop out of the average. They have no money to turn over! But velocity of circulation, defined as "coin-transfer," (_cf._ _supra_, p. 204) has clearly changed. The course of money has been short-circuited. It goes through fewer hands in the course of a given period. This last concept of velocity of circulation is clearly the one that must be used, if the equation of exchange is to be kept straight. But this fact should make it clear that velocity of circulation, instead of being the inflexible thing that Fisher has described, resting in individual habits and practices, a true causal factor in the price making process, is really a highly flexible thing, in large degree a passive function of trade and prices.

With this distinction between volume of money and volume of money-income[347] clearly held, we are prepared to go further in our attack on the quantity theory, granting the quantity theorist all his most rigorous assumptions, and still demonstrating that prices can vary independently, without prior change in quantity of money, volume of trade, or velocity of money. Let us assume the extreme case of the quantity theory: a closed market; no credit; no barter; a fixed supply of money; a fixed volume of trade; a fixed set of habits affecting velocity, namely, that everyone spends, in the course of the month, all that he has accumulated by the first of the month. The quantity theorist could not ask a more iron-clad set of assumptions than this! If the quantity theory is not valid here, if the price-level is not absolutely fixed, helpless to change, with these assumptions, then the quantity theory, even as a minor tendency, must be surrendered, and the quantity theorist must admit that the whole line of thought has been fallacious. But is the price-level passive? Suppose we assume a combination of employers of maid-servants, which forces down the wages of maid-servants from $20 to $10 per month. Assume further that there is no alternative employment for the maid-servants, so that they all remain at work.[348] So far, we have made a change in _one_ price, the price of domestic service. What of the general average of prices, the price-_level_? Well, so far, the price-level is down. If nothing else takes place, we have reduced the price-level by reducing one price. What else can take place? Two things: (1) the masters now have $10 per month each more to spend for other things than before. That tends to raise prices in their other channels of expenditure. (2) The maid-servants now have $10 each less to spend,--the same ten dollars! That lessens prices in the lines of their expenditure. These last two changes exactly neutralize one another. The first change, in the price of domestic service, remains unneutralized. The general price-level is, then, lowered--by a cause acting from outside the equation of exchange, directly on prices. The first change comes in one price. In the final adjustment, that change remains unneutralized. How is this possible? Is the equation of exchange still valid? As a mathematical formula, yes. As expressing a causal theory, in which prices are effect, and money, trade, and velocity causes, no. The equation is kept straight by a reduction in velocity. _Because_ the wages of maid-servants are reduced, _less_ money goes through their _hands_; $10 per month per maid are short-circuited. But the _cause_ is with the _prices_. The price-level, even under these absolutely rigorous assumptions, is not passive.

In general, I conclude that the price-level, under the laws governing particular prices, supply and demand, cost of production, the capitalization theory, the imputation theory, etc., can vary of its own initiative, independently of prior changes in the quantity of money, or of volume of trade, or other factors that the quantity theory stresses; and that these changes in the price-level (or in the particular prices which govern the price-level) can maintain themselves, and compel a readjustment in trade, credit, money and velocities, to correspond. This conclusion strikes at the very heart of the quantity theory, and, if valid, leaves the quantity theory disproved. More fundamentally, I should put it, prices can change because of changes in the psychological values of goods. These values are _social_ values, and are to be explained only by a social psychology. But for the present it has seemed best to me, as a means of attracting sympathetic attention from a wider circle of economists, to make use of the less debated doctrines of the science in attacking the quantity theory. It is not necessary to rest the case on my own special theory of value. Supply and demand, cost of production, the capitalization theory, the imputation theory--the general laws of the concatenations and interrelations of prices--are quite adequate for the confutation of the quantity theory. They are laws concerned with particular prices, and the price-level is nothing but the average of particular prices. Whatever explains, really explains, the particular prices, also explains the price-level.

Fisher, as we have seen, is not of this opinion. Although he has defined the price-level as an average of particular prices[349] he none the less exalts this average into a causal entity, prior to and master of the particular prices out of which it is derived, of which it is a mere average.[350] This average, he maintains, is presupposed in the determination of all particular prices.[351] This seems to me a wholly untenable position. _Ex nihilo nihil fit._ There cannot be _more_ in the average than there is in the particulars from which it is derived. In point of fact, there is necessarily vastly less. All the concrete causation is lost. The average, in itself, is nothing but a _statement_, a summary of _results_. I know nothing more metaphysical in the history of economic theory than this hypostasis of an average.[352]

I reject Fisher's notion that the average of prices is an independent entity. But I do not consider that the idea lying behind this untenable doctrine is absurd. Cost of production, supply and demand, and the other price theories _do_ presuppose something more fundamental. They do presuppose _money_, and the _value_ of money, as has been shown at length in Part I. The trouble with Fisher's notion comes in his definition of the value of money in purely relative terms as the _reciprocal of the price-level_, and his contention that the study of the value of money is identical with the study of price-levels.[353] Value is not a mere exchange relation.[354] Rather, every exchange relation involves _two_ values, the values of the two objects exchanged. These two values _causally_ determine that exchange relation. In the case of particular prices, then, we must consider not only the value of goods, but also the value of money. And the causes determining the general price-level will therefore include not alone the values of goods, but also the value of money. In the foregoing arguments by which I have shown that the price-level can vary independently of the other factors in the quantity theory scheme, I have been concerned only with changes in the values of goods, measured by a constant unit of value. If the value of money should also be varying, the concrete results on the price-level would have been different. On the face of things, there was nothing in the cases I discussed to require us to suppose that the value of money would also vary. The argument ran on the assumption of a fixed value of money. I have shown, in earlier chapters, that the assumption of a fixed value of money is fundamental to the laws of supply and demand, cost of production, and the capitalization theory. In point of fact, this assumption is rarely true--never strictly true. For causes which are in considerable degree independent of the causes governing the values of goods (as the causes governing their values are in considerable degree independent of one another), the value of money varies, now in the same direction as the values of goods in general, now in an opposite direction. Further, money itself does not escape the general laws of concatenation of values. The value of money has causes which are bound up with the values of other goods. Thus, when prices are rising and trade expanding, there is a tendency--commonly a minor tendency--for money also to rise in value, and so prices do not go quite as high as they would have gone had money remained constant. This tendency arises from the fact that there is more work for money to do in a period of active trade and rising prices. Gold also tends to rise in value in the arts, with prosperity. The reverse tendency manifests itself when prices are falling: money tends, in some measure, to fall in value with the goods,[355] and so prices do not fall as far as they would fall if money remained constant. But in general, the causes governing the values of goods, and the causes governing the value of money, are sufficiently independent to justify us in studying each separately, in abstraction, on the assumption that the other is unchanged. Hence, supply and demand, cost of production, and the other price theories, which assume a fixed value of money, are proper tools of thought for the study of the prices of goods.