² On page 292 the author finally mentions satirically the fact that traditional economics may get around to monetary phenomena, “in Vol. II, or more often in a separate treatise”.

³ Chapter 19, entitled “Changes in Money-Wages”, deals with the effects of wage increases.

⁴ Mr. Keynes quotes Mill on Say's Law, but does not mention either Mill's explicit exception for crisis conditions which occurs a few pages previously in his Principles, or, of course, Mill's doctrine that the demand for products is not a demand for labour, which (however absurd) was one of his chief bids for fame.

⁵ The first difficulty in following up and interpreting this statement is the confusion between what is dependent upon the actual magnitude of a variable and what is dependent on changes in that variable. It is no exaggeration to say that the book is “packed” with examples of this confusion. If we interpret the statement in accord with what it actually says, the questions raised have to do with speed of change and differences in speed of change between independent and dependent variables, i.e., with “lags” in response, and the length of time required to establish a new equilibrium of the same sort which must be assumed as the starting point of the initial change, to make sense of the statement. But this view is contradictory to the conception of equilibrium in terms of which the theory as a whole is couched. The main assumption as to the psychology is repeatedly referred to in the book as a “law”. The statements alternate more or less at random between the form of a relation between changes (almost always increases) in income and changes (increases) in “non-consumption” (as to saving, see below) and the form of a relation between income itself and non-consumption. For the latter, see page following quotation (p. 28 at bottom) and the apparently crucial definition on page 90, which calls the propensity to consume the functional relation between income and expenditure on consumption out of that income. But on page 96 “the fundamental psychological law” is again a relation between increases, and on page 97 the two formulations are apparently identified. (Cf. also pp. 115, 121, 251, 247.)

It is to be noted as a separate source of confusion that a relation between changes in one direction does not necessarily hold for changes in the other. A sufficiently industrious and painstaking reader will finally discover that in this case the reverse change, decrease in monetary flow, is supposed not to occur. (Cf. p. 307, at middle, and discussion below in this review.)

⁶ Secondly, as we shall presently see, the thesis is specifically that the failure occurs in the demand for labour for use in connection with investment, not in connection with consumption.

⁷ Immediately following the last long quotation above (General Theory, p. 28), the author reiterates his special-case accusation (without using the words “classical economics”), asserting that the equilibrium level of employment cannot correspond to more than full employment, since wages cannot exceed marginal productivity, but that there is no reason for expecting employment to be as much as full, that this will be the case only “when the propensity to consume and the inducement to invest stand in a particular relationship to one another”. He means when they stand in a particular relationship to wages, the interest-rate, and general prices, which is obvious. What is mysterious and difficult to state clearly is the manner in which Mr. Keynes sets up an economic system on the basis of assumptions which imply that these variables or variable-complexes are either fixed or are determined by other forces than the mutual adjustment of supply and demand, i.e., by “bargains” or public authority, or “psychology”, or some other deus ex machina.

⁸ As already remarked, the question of the reversibility of functional relations predicated for change in one direction is a confusing feature of Mr. Keynes's argument as a whole. The most general and pervasive example is the fact that the whole work explains unemployment by showing why increase in employment is brought to a stop, or blocked before it can get started. Except in chapter 22, “Notes on the Trade Cycle”, which is really an appendix with a different point of view, little or no intimation is ever given that unemployment might result from a decrease in employment. In historical fact, as far as I know, unemployment on the scale of a serious social problem is not a typical state of affairs, and in every known case such a situation has followed at no long remove a period of relatively full employment—and has followed upon a sequence of change fairly uniform and familiar in its more general features and, similarly, periods of serious unemployment have in due course come to an end. But the question of how unemployment comes to pass is excluded from this work by the predetermination to make it a “normal” phenomenon, characteristic of an enterprise economy in stable equilibrium. It always follows upon equal or greater unemployment, never upon more employment. In this connection the interpretation of Mr.Keynes, by ProfessorHansen, Alvin H. (Journal of Political Economy, 10, 1936)Google Scholar is interesting in that the position of equilibrium is established on the way down and not on the way up, as in the book itself.

⁹ It is difficult to tell what is Mr. Keynes's conception of the relation between short-run and long-run conceptions and of their role in managers' decisions. The weakness of chapter 5 is again in point. It should be recognized that in the shortest short-run all, or virtually all, production of goods is for stock (in possession of some one) and all sales are sales from stock, hence that both are a matter either of speculative conversion of investment between goods and money, or of choice between consumption and investment. On the other hand, in the ultimate long-run there are no fixed costs, and for a system in equilibrium, stationary, or with growth (i.e., unless the system as a whole is decadent), there is no capital charge except interest. In the “theoretical” long-run, moreover, there is no speculative factor; but in reality the farther ahead plans must look the greater this factor becomes.

¹⁰ As already noted, it is expressly stated (p. 249) that prices rise, in terms of wage-units, with increasing employment—which seems to be the same as increasing income—in consequence of increasing cost (diminishing returns) in the short period.

¹¹ On page 83 (quoted above) it “may” fail. When we come to Mr. Keynes's theory of interest, we shall see that there is no indication of any way in which monetary saving, though it “is” an equal investment, can lead to any investment in the sense of technical production. The questions whether money savings are made by entrepreneurs as well as “owners of productive factors” (and rentiers?) and whether “owners of factors” means simply labourers, become important in connection with the effort to form any inclusive picture of the motives of saving and the way in which they operate; but I have not been able to find answers to them.

¹² As already suggested, Mr. Keynes's whole argument in connection with labour apparently assumes that it bargains as a unit, and that the complete unemployment of particular individuals (leaving them with no income? or none except “relief”?), will affect the supply price of labour in the same way as a fractional reduction in the employment and wages of a given group of employed men.

¹³ Economic Journal, June, 1931.

¹⁴ In the first of the chapters on “The Propensity to Consume” (ch. 8, p. 93), the rate of interest is referred to as being nearly the same thing as the ratio of exchange between present and future goods. But in the text no move is made to integrate this notion with the theory of interest. There is no indication of any causal relation either way between the interest rate and the exchange ratio or between either and the general price level. (Cf. General Theory, 140-1, reference to Fisher.)

¹⁵ It will be noticed that Mr. Keynes's discussion of the interest rate (the terms of investment) comes in between the treatments respectively of the two alternatives compared by the entrepreneur who makes a real investment, namely, the incentive to invest and the incentive to hold cash, the latter called “the incentive to liquidity”. But in fact neither of these alternatives has any reality apart from the other, or from the necessity of comparing them and making a choice.

¹⁶ It would surely have been in accord with Mr. Keynes's line of attack to emphasize the fact that at a time of deep depression there is little relation between the prices of capital goods or even securities (relative to yield) and any market rate of interest. Interest rates and capital values are both abnormally low. See above, p. 113, and below, p. 116.

¹⁷ This is probably more or less in accord with the general thinking of the business community, which fact, and its relation to the realities of the situation, might have been worth noting. Mr. Keynes makes no reference to the patent fact of the business cycle that men rarely borrow money to hold money, but do so to hold other forms of wealth (or to pay off some other debt) and that the rate of interest is highest when exchange medium is most abundant and its velocity of circulation most rapid (with the exception of the brief period of acute crisis, when the demand for cash rests primarily on actual, prospective, or feared needs to meet contractual or other obligations fixed in monetary terms.)

¹⁸ I think this idea fantastic, but the issue cannot be argued here.

¹⁹ This indeed is qualified to apply “up to the point where full employment prevails” (p. 372), but the text of this chapter, as well as the book as a whole, makes it clear that the qualification is essentially “theoretical”.

²⁰ One difficulty which may be mentioned is that if modern technology, with specialization and large scale organization of production, is not to be simply scrapped, great concentration of authority in the hands of individual human beings, or committees or “boards”, is unavoidable, and the issue is one of methods of selecting, motivating, and remunerating such functionaries, and of maintaining “responsibility” in the face of social objectives which must also be formulated through the workings of the social system itself. Reformers seem characteristically to pass somewhat lightly over the fact that these are human problems, essentially political problems, that there is no way which men will generally agree upon as valid to call in God and the angels to make the decisions and carry out the policies.

²¹ As to the import of the “sometimes” I have no inkling. Why the national income is measured in wage units is also obscure to me; presumably there is some connection with the dictum in the next section of the chapter, where it is explained that an increase in employment will increase the demand for money because of increased quantity and value of output, the latter in turn being due to rising wages and diminishing returns from labour “in the short period”. Why either money or real wage rates should rise before unemployment is absorbed is not explained and the increase in labour cost under conditions of unemployment is dubious; and granting both, the rise in prices rests on the dogma that they “must” equal or correspond to wage cost, which is the kind of reasoning we have been told earlier (p. 12) would have been expected of the classical school. More interesting is the fact that in the formal classification itself, prices were not mentioned, either as given, as independent variable, or as dependent variable.

²² This, of course, is a line on which a number of thinkers have been working and writing in recent years. I am thinking especially of the work of Mr. Hawtrey, the Swedish school, and Mr. Robertson; but only an authority on subject-matter can be an authority on the literature.

In the very first paragraph of his Preface, Mr. Keynes says: “Those, who are strongly wedded to what I shall call ‘the classical theory’, will fluctuate, I expect, between a belief that I am quite wrong and a belief that I am saying nothing new. It is for others to determine if either of these or the third alternative is right.” The prediction has been largely correct in my own case, though I should say that my difficulty (and no little annoyance) has been that of choosing between interpretations, one apparently nonsensical and the other more or less commonplace. “It is for others to determine” whether such a result proves that the one who arrives at it is “wedded” to some antique mode of thought. This, of course, is one of two “arguments” regularly hurled by revolutionary thinkers at those who do not immediately join up, the other being that the refusal is based on a vested interest. This the revolutionary is sometimes “polite” enough to imply is done unconsciously (i.e., blindly instead of intelligently); Mr. Keynes may be thanked for omitting the second. Since it has become quite the fashion to account for differences in intellectual position by psycho-analysing, or somehow “explaining”, one's opponent (and the example of following the fashion having in this case been set by Mr. Keynes), it may be permissible to note that our civilization of to-day, being essentially romantic, loves and extols heretics quite as much as its direct antecedent a few centuries back hated and feared them. The demand for heresy is always in excess of the supply and its production always a prosperous business. Where once it was necessary in writing to pose as merely restating and interpreting doctrine handed down from the Fathers, the surest way to public interest and acclaim now lies through pulling down and overturning everything established or accepted.

²³ See Quarterly Journal of Economics, 11, 1936, p. 175.Google Scholar

²⁴ Perhaps a constructive suggestion from a “mere theorist” may not be entirely out of order. It has long been in my mind that in the welter of cycle theories (most of which have merit in pointing to real factors in the problem) one point is still neglected which must be of some importance. It has been recognized for at least a century that within some limits speculative psychology tends to give rise to a kind of momentum or cumulative tendency in price changes. The equilibrium point being uncertain, the tendency of speculation for a rise to create a rise in the price of any commodity within limits outweighs the “force” tending toward equilibrium—and conversely. Reasoning which cannot be developed here would show that this tendency should be especially strong in the case of money, the essential function of which is to be held speculatively. I should not be surprised if this is the most important factor in the general tendency to oscillation in an economic system—in contrast with specific “cycles” affecting particular commodities, which according to the laws of chance should be distributed in periodicity and phase and so cancel out for the system as a whole.