CHAPTER X

"NORMAL" VS. "TRANSITIONAL" TENDENCIES

The Quantity Theory, as a causal theory, is, then, little altered by the passage from a hypothetical, creditless economy to the actual world, where a vast deal of credit is used,--particularly in Professor Fisher's hands. Of the different kinds of credit, only deposits subject to check are recognized as directly influencing prices, and deposits subject to check are controlled by the volume of money. The causal theory[183] remains, then, as follows: if M be increased, it will increase M' proportionately; it will not change the V's; it cannot increase T; to keep the equation straight, therefore, P must rise in proportion to the rise in M. A decrease of M, reducing M' proportionately, leaving V's and T unchanged, must proportionately reduce P. P is passive. A change in P cannot sustain itself, unless it be due to a prior change in T, the V's, M or M'.

This theory is set forth with the qualification that these effects are the "normal" effects of the changes in question. The proportion between quantity of money and price-level is not strictly maintained during "transition periods." I now approach the most difficult question which I shall have to answer as to the meaning of Fisher's terms. The same problem arises for all quantity theorists. Precisely what is the distinction between "transition periods" and "normal periods"? What limitations and qualifications does he admit to the rigorous statement of his theory so far given? I may first express the opinion that the line shifts greatly in his own mind, or at least shifts greatly in the exposition. I do not find an explicit statement in which definitions are given. The matter is chiefly discussed by Fisher in ch. 4,[184] which is called "Disturbance of Equation and of Purchasing Power during Transition Periods." There we find, as I have stated, no definitions, but the initial statements would suggest the following: a transition period is the period following a change in any one of the factors in the equation during which a readjustment among all the others is taking place; the normal period is the period preceding such a change, or following the transition after such a change, and is characterized by the fact that all the factors are at rest, in stable equilibrium. Equilibria during transition periods are unstable. During the transition, the relations among the factors vary: M and M' need not keep their fixed ratio; P need not be wholly passive; M and P need not keep the same proportion. But until M and M' get back into the normal ratio, until P becomes proportional to M (in the proportion prior to the initial disturbance), there is no rest; the equilibrium is unstable. How long is a transition period? How realistic is the notion of a transition period? Is the transition period a theoretical device, to aid in isolating causes, or is it supposed to be a real period in time? Is the normal period a real period in time, or is it merely a theoretical hypothesis? It is not easy to answer these questions. Thus (p. 72) the seasonal fluctuations are declared to be "normal and expected," and, at the same time, one gets the impression that Fisher considers them illustrations of his "transitions," in which the normal theory does not strictly hold (pp. 72, 169). What is described chiefly in the chapter on transition periods is the business cycle--a theory of the business cycle, based primarily on the notion that the failure of interest to rise as fast as prices rise causes the "boom," and that the draining of bank reserves precipitates the crisis. I shall not discuss this theory, as a theory of business cycles, further than to say that Wesley Mitchell's study would indicate that the interest rate is a minor factor, and that, while as a theoretical possibility, the drains on bank reserves may check prosperity if something else doesn't do it first, practically something else always does come in ahead, so far as his studies have gone.[185] My interest here is primarily in seeing the limitations Fisher imposes on his theory, and the qualifications he admits. If the business cycle is the typical transition period, during which his normal theory doesn't hold, when does the normal theory hold? When are the "normal periods"? There is no concrete period during which prices are neither rising nor falling, during which no important changes are taking place among the factors.[186] At times, Fisher seems to indicate that the normal period is imaginary (pp. 56, 159). Is, then, the contrast between a realistic "transition period" and a hypothetical "normal period" or are both hypothetical? Is the equation of exchange, too, a mere hypothesis? It should be, if it is to set forth a merely hypothetical theory. But no, Fisher insists on putting concrete data into it, and, indeed, gives an elaborate statistical "proof" of the equation. It, at least, is realistic. I confess that my certainty as to Fisher's meaning grows less, as I study his book with greater care. If the typical transition period be the business cycle, then the normal period could come only once, say, in ten years--or whatever period, regular, or irregular, one chooses to assign to the business cycle. The concrete price-levels for the greater part of the time are then surrendered to other causes. And the one-year cycle described in the equation of exchange is quite irrelevant. The equation of exchange should cover the whole business cycle, to fit in with the theory. Indeed, a realistic equation of exchange would then have no meaning at all, as the average price-level during the business cycle, played upon by a host of causes other than the factors described in the quantity theory, would not be the same as the average price-level which _would have_ obtained had only the "normal" causes been in operation.[187]

The distinction between "normal" and "transition" _periods_ suggests a dangerous fallacy: namely, that during one period one sort of causation is working, with the other in abeyance. In fact, whatever causes there are are working all the time. The only legitimate thing is to abstract from one set of causes, and see what the other set, if left to themselves, will bring about. But this sort of abstraction has many dangers, one of which is that the causes abstracted from are frequently thought of as non-existent. The chemist, in his laboratory, can in actual physical fact abstract impurities from his chemicals, and see what they will do. He can even perform experiments in what is practically a vacuum. But the economist has no right to _think in vacuo_! All that he has a right to do is to assume the factors which he does not wish to study _constant_. And even that he must not do if (1) changes in the factors which he wishes to study do in fact lead to changes in the factors abstracted from, or (2) if the factors which he wishes to study can only change _because_ of prior or concomitant changes in the factors from which he is abstracting. Is it, for example, legitimate to assume an increase in M' apart from its usual accompaniment, an increase in PT?

The notion, too, that causation can be seen in a state of stable equilibrium should be critically analyzed. Causation is only _revealed_ by a _course of events_, when mechanical causation is involved. The relation of cause and effect may be a contemporaneous relation in fact, and it is possible, where conscious, psychological phenomena are involved, to discern causal relations among the elements in a mental state by direct introspection. It is the not uncommon practice, also, in the theory of mechanics, or in theoretical economics, where the method of investigation is deductive rather than inductive, to abstract from the temporal sequence, and to construe causal relations as timeless, logical relations. But even here, the cause of a _change_ in the general situation precedes the change in time, and it is only by abstraction that the time element is left out. If there is no question as to the causal relations, this abstraction is legitimate, but if all that one knows about the situation be that in a stable equilibrium certain constant ratios obtain, then the question as to which term in the ratio is cause and which is effect remains unanswered. In Fisher's situation, then, assuming that it be true--which I shall deny--that the only stable equilibrium is that which the normal theory requires, it still remains true that the causal relations among the factors can only be revealed by a study of the transitions, by seeing the temporal sequence of changes in the factors of the equation. Even if it be granted that M, M' and P tend to keep a constant relation to one another, the quantity theory falls if, for instance, it can be shown that a change may first occur in P, spread to M', and finally reach M last of all, leading to a new normal equilibrium which is stable. I shall later show cases of this sort.[188]

The abstract formulation of Fisher's contrast will not, I believe, give us an answer as to the extent to which he thinks his quantity theory realistic. I find myself particularly in genuine uncertainty as to the point mentioned above: would an actual equation of exchange for the whole business cycle, made up of the averages of M, M', V, V', P and T for the whole period, exhibit the "normal" relations among these factors? Or would this "normal" relation only emerge concretely at some moment of time in the course of the cycle when the abnormal causes affecting the price-level happened to offset one another? Or is it true that no actual figures which might be found, either for a moment of time, or as averages for any given period, will exhibit the relations required, and that only a hypothetical equation, based on the figures for M, M', V, V', P and T that _would have been realized_ had there been no "disturbing" causes, will show these "normal" relations? If, as Fisher at times indicates--as in his reference to Boyle's Law (p. 296)--he is stating only an abstract tendency, which may be neutralized by other tendencies in the situation, so far as concrete results are concerned, then it is this last doctrine which we must take, and the concrete equation of exchange has little if any relevance. If, moreover, this last interpretation be given, then the whole of Fisher's elaborate statistical "proof" is pointless. The only sort of statistical proof which would be relevant would be of a much subtler sort, not a mere filling out of the equation of exchange by means of annual figures, but an effort to disentangle and measure the _importance_ of his tendency, as compared with other tendencies. But we have the other tendencies merely mentioned in qualitative terms, and we never find any definite statement, of mathematical character, as to how important they are.

It seems pretty clear, however, that on the whole, despite occasional suggestions that his theory is abstract, Fisher means his theory to be the overwhelmingly important point in the explanation of actual price-levels. He is particularly insistent on the high degree of the generality of his contention that P is passive. Thus: "So far as I can discover, _except to a_ LIMITED _extent during transition periods, or during a passing season_, (_e. g._, _the fall_) (capitals mine, italics Fisher's), there is no truth whatever in the idea that the price-level is an independent cause of changes in any of the other magnitudes, M, M', V, V', or the Q's."[189] On p. 182 he enumerates in a series of propositions his general normal theory, and adds, as the first sentence of proposition 9: "Some of the foregoing propositions _are subject to_ SLIGHT _modification during transition periods_." (Italics and capitals mine.) And the general drift of the argument, particularly in chapter 8, where the heart of Fisher's causal theory is presented, would indicate that the concessions he is disposed to make are very slight, indeed.

The question as to how long a _time_ is required, in Fisher's view, for a transition to occur, and for his normal tendencies to dominate, is nowhere made clear. The quantity theory, in the hands of some writers, is a very long run theory, for others, it is a short run theory. Thus, Taussig would make the "run" exceedingly long.[190] Mill makes it a short run theory. "It is not, however, with ultimate or average, but with immediate and temporary prices, that we are now concerned. These, as we have seen, may deviate widely from the standard of cost of production. Among other causes of fluctuation, one we have found to be, the quantity of money in circulation. Other things being the same, an increase of the money in circulation raises prices, a diminution lowers them. If more money is thrown into circulation than the quantity which can circulate at a value conformable to its cost of production, the value of money, so long as the excess lasts, will remain below the standard of cost of production, and general prices will be sustained above the natural rate."[191] I pause to note that it is really strange that a single name should describe theories so different, resting on such essentially different logic. Long run or short run theories, all are "quantity theories," whether "money" be defined as gold, or as all manner of media of exchange, or as only those media of exchange which pass from hand to hand without endorsement. Fisher would doubtless call his theory a long run theory. From the standpoint of the notion that "prices ... lag behind their full adjustment and have to be pushed up, so to speak, by increased purchases,"[192] however, we get a short run quantity theory doctrine. The logic of these two is very different. The short run doctrine seeks to explain the actual process of price-making in the market. Money is offered against goods, and the actual quantities on each side determine the momentary price-level, concretely. Or, when credit is considered, money and credit offered against goods, at a given time, or in a given short period, determine the actual price-level reached. This is the logic of the equation of exchange--actual money paid is necessarily equal to actual money received. The long run doctrine is fundamentally based on a different notion. Surrendering the actual or average of price-levels to other causes, in part, it still asserts that, given time enough, and barring new disturbing tendencies, a price-level will ultimately be reached which will bear it out. I find no recognition, on Fisher's part, of the fact that these two doctrines are different, and, in fact, I find them blended and confused in the course of his argument. He would doubtless maintain that his is a long run doctrine. But how long is the "run"? Sometimes it seems to be, as already shown, a whole business cycle. Sometimes a passing season, as the fall. When he undertakes to apply his theory to a practical proposal for regulating the value of money, he relies on the quantity theory tendency to bring about adjustments so quickly that it is worth while to make _monthly_ adjustments in anticipation of it.[193] When discussing the changes in gold premium on the Greenbacks during the exciting times of the Civil War, he relies so thoroughly on his theory that he will not allow even the rapid change of four per cent in a single day following Chickamauga to occur except in conformity with the quantity theory. This last statement is so remarkable that I must quote Fisher himself: "It would be a grave mistake to reason, because the losses at Chickamauga caused greenbacks to fall 4% in a single day, that their value had no relation to their volume. This fall indicated a slight acceleration in the velocity of circulation, and a slight retardation in the volume of trade" (263). It would be indeed remarkable if the changes in the gold market, which got war news before the newspapers got it, and where changes in gold premium occurred before the rest of the country could possibly react to the war news, should be controlled by V and T! I had not supposed that the most rigorous of short run quantity theorists would make any such demands on his theory as that. Indeed, I had not supposed that the quantity theory would feel called on to explain the gold premium, as such, except in so far as the gold premium is an index of general prices.

Finding it impossible to limit Fisher to any single statement of the quantitative importance of his normal theory as compared with the other tendencies at work, but concluding that, on the whole, he considers it of high importance, I shall now proceed to an analysis of the reasoning by which he seeks to justify it as a _qualitative_ tendency. I shall maintain that, however long or short the period required, however strong or weak the tendency he defends, the reasoning by which he seeks to justify it is unsound, and that even as a qualitative tendency, the quantity theory is invalid. At a later part of the book, as in an earlier part,[194] I shall undertake to find the modicum of truth which the quantity theory contains, and shall show that no quantity theory is needed to exhibit this modicum of truth.