(Tables are not uploaded well. Something needs to be done for them.)
ＣＨＡＰＴＥＲ １４
THE PROOFING PROCESS OF THE ＧＥＮＥＲＡＬ ＴＨＥＯＲＹ (2)
Our examination in the preceding chapter of the galleys from the Pre-First Proof Typescript, produced in summer 1934, to First Galley III, which came out in June or July 1935, has so far omitted consideration of Chapters 2, 3, and 6-9 of what we are calling the 'Galley Table of Contents'. Let us now complete our examination of the galleys by looking at these chapters, all of which, with the exception of Chapter 2, Keynes altered considerably during the 'Great Revision' undertaken in August 1935.
In the context of the proofing process of these chapters, from 'The General Theory' to the General Theory, we need, in fact, to examine two important bursts of activity on Keynes's part. One is the Great Revision ; the other occurred in September 1935 and will be referred to as the 'Changeover'.
1. Chapters 2, 3, and 6-9 of the Galley Table of Contents
E. The Great Revision
With respect to the chapters concerned, the precipitate of the Great Revision is virtually the text of the General Theory, so that an examination of the texts produced by the Revision amounts essentially to an examination of the General Theory itself. What matters to us here, though, is that we understand how Keynes changed his ideas at this time.
The large-scale revision undertaken by Keynes in August 1935 encompassed Chapters 1-3, 6-11, 15, and 16 of the Galley Table of Contents. Of these, we have yet to examine Chapters 6-9 of Book II, 'Definitions and Ideas', and those which belong to 'Introduction' (Chapters 1-3). At this stage Keynes not only rearranges the chapters considerably, but also alters the definitions of some fundamental concepts.
(a) Chapters 6-9
Book II of the Galley Table of Contents is composed of Chapters 4-9, of which Chapters 4 and 5 were finalised in the Pre-First Proof Typescript (see Table 13-2). Let us now first look at Chapters 6 and 7, and then at Chapters 8 and 9.
<1> Chapters 6 and 7
(i) Revisions to the Formal Structure
Two points are worth noting about the revisions Keynes made, in terms of formal structure, to Chapter 6, 'The Meaning of Income', and Chapter 7, 'The Definitions of Quasi-Rent, Saving and Investment'.
First, the two chapters are at the stage of completion, not only in substance but also in style, as far as the text envisioned in the Galley Table of Contents is concerned. Chapter 6 (JMK.14, pp. 398-418), which is the revision of Chapter 6, 'The Definition of Income', of the Pre-First Proof Typescript, is only slightly revised in the Second and Third Galleys, and Chapter 7 (JMK.14, pp. 418-425) is only slightly revised in the Second Galley.
Secondly, it became Keynes's main work in August 1935 to rewrite these two chapters. As a result of the Great Revision, they became, together, Chapter 6, 'The Definition of Income, Saving and Investment', with its 'Appendix on User Cost', of the General Theory.
Chapter 6, which is composed of five sections, was revised from the Third Galley to the General Theory as follows: (i) part of Section I (JMK.14, pp. 401-402) becomes the corresponding passage (GT, pp. 52-53) in Section I, 'Income', of Chapter 6 of the General Theory; (ii) part of Section II (JMK.14, pp. 401-402) becomes the corresponding passage (GT, pp. 67-68) in Section 1 of Appendix on User Cost to Chapter 6 of the General Theory; (iii) Section III remains unchanged in terms of style, but is restructured to form Sections II, III and IV of the Appendix on User Cost to Chapter 6 of the General Theory; (iv) Sections IV and V are scrapped and do not appear in the General Theory (Section IV dealt with the calculation of income in extreme cases, such as that of 'a product prepared long beforehand [which] necessarily yields up its fruits now and is incapable of being stored up for a later date' (JMK.14, p. 413), and 'that part of the wastage of any piece of equipment which occurs by the mere passage of time and irrespective of whether it is used or not' (JMK.14, p. 414); Section V stressed the need to define income, saving and investment consistently).
Chapter 7, which is composed of three sections, is revised from the Second Galley to the General Theory as follows: (a) Section I, which deals with quasi-rent, is dropped in the General Theory; (b) some of the arguments developed in Sections II and III, which deal with fundamental concepts in relation to on the one hand long- and short-period expectation, as well as the actual result, and, on the other, to user cost, correspond to pp. 60-64 of the General Theory, which are almost completely equivalent to Section II, 'Saving and Investment', of Chapter 6 of the General Theory; (c) the greater part of Section I, 'Income', of Chapter 6 of the General Theory (pp. 54-60), in which user cost and supplementary cost are discussed, is newly written and added at this time.
Looking at the changes outlined above from the standpoint of the General Theory, we can sum up this part of the proofing process as follows: (i) Keynes writes Chapter 6 of the General Theory by changing considerably the contents of Sections II and III of Chapter 7 of the Second Galley and the contents of Sections I and II of Chapter 6 of the Third Galley; (ii) Keynes writes the appendix to Chapter 6 of the General Theory following the contents of Section III of Chapter 6 of the Third Galley.
(ii) Theoretical and Conceptual Revisions
Chapter 6 (Third Galley) and Chapter 7 (Second Galley) contain the core discussion, in terms of the Galley Table of Contents, of the definitions of fundamental concepts such as income, investment and saving. These chapters are closely related to Chapter 6 of the General Theory, which contains the core of the argument concerning fundamental concepts in the General Theory. Let us examine how Keynes changed the definitions of such fundamental concepts in order to arrive at the final form of the General Theory. It will be convenient to do this by comparing and contrasting the Pre-First Proof Typescript with the manuscripts written before and after it, as well as by comparing and contrasting First Galley I with Chapter 6 of the General Theory.
(α) The Pre-First Proof Typescript and the Immediately Preceding and Succeeding Manuscripts
The definitions of fundamental concepts adopted by Keynes in the General Theory can be traced back to those he tried out in 'The General Theory' (see Section 1 of Chapter 12). Let us look at how Keynes changed the definitions of these concepts as he proceeded from 'The General Theory', through the Pre-First Proof Typescript, to the Third Galley.
The most important characteristics these manuscripts have in common are: (i) income is defined as the realised value of the sale proceeds and effective demand as the expected value of the sale proceeds; (ii) the definition of income differs from that of effective demand, even if the difference between the realised value and the expected value of the sale proceeds is disregarded.
'The General Theory' - As discussed in Section 1 (A(a)) of Chapter 12, Keynes puts forward the following formulations in this manuscript:
F = Y - D (1)
D = Q + E (2)
E = N.W (3)
where D denotes effective demand, Y income, F entrepreneurs' windfall profits, Q quasi-rent, E prime cost, N the volume of employment, and W money wages.
In equation (1) income is defined in such a way that it is larger than effective demand by the amount of windfall profit, while in equation (2) effective demand is defined as the sum total of the quasi-rent (meaning normal profit) and the prime cost. These formulations are peculiar to 'The General Theory'; in particular, windfall profit (excess profit), which is defined as the difference between income and effective demand, does not appear again after this manuscript.
The relation between investment I, saving S, and consumption C is also peculiar１ to 'The General Theory'. This is connected to the difference in definition between income Y and effective demand D:
D = C + I (4)
Y = C + S (5)
From equations (4), (5), and (1) we can obtain:
F = S - I (6)
From the end of 1932 (that is, ever after 'The Parameters of a Monetary Economy') Keynes stands by the notion of investment-saving equilibrium. Equation (6), to which the peculiar definitions of effective demand and excess profit are vital, appears as a consequence of his argument, though he does not state this explicitly in 'The General Theory'. Keynes repeatedly stresses that investment and saving should be construed as 'gross' concepts. Gross investment is defined as inclusive of new investment, replacements, and repairs, while gross saving is defined as inclusive of sinking funds.２
The Summer Manuscript - In this manuscript Keynes puts forward the followi
ng formulations (the notation follows 'The General Theory', except where indicated):
D = Y + Uｓ (7)
Y = C + I (8)
Y = C + S (9)
Qｓ１ = D - E (10)
Qｓ２ = Qｓ１ - Uｓ (11)
where Uｓ is the user cost, and Qｓ１ and Qｓ２ respectively the return to and income of entrepreneurs, as defined in the Summer Manuscript.
The most significant differences between the Summer Manuscript and 'The General Theory' in the definitions of fundamental concepts are that in the Summer Manuscript: (i) windfall profit disappears; (ii) user cost, Uｓ, is considered to be the difference between effective demand and income (equation (7)); (iii) income is equal to the sum total of consumption and investment (equation (8)).
Keynes defines the difference between effective demand and the prime cost as the return to the entrepreneurs, Qｓ１ (equation (10)). The income of the entrepreneurs Qｓ２, which is the difference between the return to the entrepreneurs Qｓ１ and the user cost Uｓ (equation (11)), corresponds to quasi-rent in 'The General Theory'.
Qｓ２ is equal to the profit of the entrepreneurs P１ (see equation (37) below) in First Galley I. This can be shown as follows. Using equations (7), (10) and (11), Qｓ２ can be expressed as:
Qｓ２ = Y - E (12)
On the other hand, P１ can be expressed as
P１ = Y - E１ (13)
where E１ denotes the prime cost in First Galley I. (This follows from equations (27) and (37), to be written down and discussed in (β) below).
Since E = E１, we have:
Qｓ２ = P１ (14)
In the Summer Manuscript the fundamental concepts are defined in nearly the same way as in First Galley I. In both cases Keynes treats those concepts inclusive of user cost as 'gross' concepts, and stresses that effective demand can be defined as a gross concept. (The concept of user cost per se appears for the first time in the Summer Manuscript - see Section 2 (A) of Chapter 12.) It is difficult to know exactly how Keynes defines user cost, as the extant material does not show this clearly. However, because the phrase 'the user cost of the initial stock of capital goods' (JMK.13, p. 472) was used, and because Keynes explicitly adopts user cost in the definition of investment, we can be fairly safe in saying that this is a new concept, one which had not appeared in 'The General Theory'.３
For the sake of clarity, let us pause to go through this a bit more formally. If we denote new investment as conceived in the Summer Manuscript by Iｓｎ, replacements and repairs by R, investment as conceived in 'The General Theory' by Iｇ, and investment as conceived in the Summer Manuscript by Iｓ, we can write:
Iｇ =Iｓｎ - R (15)
Iｓ = (Iｓｎ + R) - Uｓ (16)
The definition of investment which appears in the Summer Manuscript (equation (16)) is not seen in 'The General Theory', in which Keynes emphasises Iｇ. In the Summer Manuscript, Keynes always uses 'investment' to mean Iｓ, and refers to Iｓｎ as 'net investment'.４
The Pre-First Proof Typescript - Keynes deals with the definitions of fund
amental concepts in Chapter 6, 'The Definition of Income'. The title does not appear in the table of contents of 'The General Theory' (see Table 12-1). This chapter corresponds to Chapter 6, 'The Meaning of Income', in the tables of contents from First Galley I to the Third Galley. That is, after some revision, this chapter becomes Chapter 6 of First Galley I (see JMK. 14, pp. 398-418). Unfortunately, in contrast to the Summer Manuscript and First Galley I, the parts in the Pre-First Proof Typescript which correspond to Chapter 7, 'The Definitions of Quasi-Rent, Saving and Investment' of First Galley I, in which the interrelation between the concepts concerned are discussed, are not extant. However, on the basis of material reproduced in JMK.14 (pp. 398-418) we may express the interrelation between the concepts concerned, as conceived in the Pre-First Proof Typescript, as follows５:
Y = A - U ｐ (17)
Uｐ = Sｐ - E (18)
Qｐ１ = A - E (19)
Qｐ２ = Qｐ１ - Uｐ (20)
where A is gross sales proceeds, Uｐ user cost, Sｐ the supply price, E the prime cost, Qｐ１ quasi-rent, and Qｐ２ net quasi-rent. Subscript p indicates the Pre-First Proof Typescript. (Note: Unlike in the Summer Manuscript and First Galley I, in the Pre-First Proof Typescript the argument is not couched in symbols. However, for the sake of convenience, we use them here.)
As in First Galley I, we might assume that the firm is completely integrated. It should be also noted that 'the cost spent on the maintenance and improvement of the initial capital equipment', B, which plays an important role in First Galley I (see (β) below), does not appear as yet.
Because the definitions of fundamental concepts used in the Summer Manuscript are almost all carried over to First Galley I, and because the Pre-First Proof Typescript was written in between the two, we can be fairly safe in saying that the definitions in the Typescript are almost all carried over from the Summer Manuscript. (One exception is equation (18), which defines user cost. This makes no appearance in the Summer Manuscript.) In that case, equation (17) becomes the same as equation (7), provided the definition of user cost is the same in the Typescript and First Galley I. Then, gross sales proceeds, A, becomes equal to effective demand, D, and the equivalence Q ｐ１= Qｓ１ can be obtained from equations (19) and (10). Moreover, Q ｐ２ = Q ｓ２ can be obtained from equations (20) and (11).
In the Typescript, the term 'user cost' is defined as 'the loss in the prospective value of a plant due to using it as compared with not using it'６, so that we can safely say that the definition of user cost per se is established in the Typescript. The supply price, Sｐ, is defined as 'the lowest price which the owner of the equipment will accept for its output rather than lay it up'.７ By transforming equation (18) into the following, we find that the supply price is, by definition, the sum total of the user cost and the prime cost:
Sｐ = Uｐ + E (21)
(β) First Galley I and the General Theory
To properly understand Keynes's definitions of fundamental concepts such as income, investment and saving, it is very important that we pay attention to how the concept of 'user cost' is defined.
In the case of Chapter 6, there is little or no difference in content between First Galley I and the Third Galley. In the case of Chapter 7, the same is true with respect to First Galley I and the Second Galley. We will examine Chapter 6 of the Third Galley and Chapter 7 of the Second Galley below. It should cause no confusion if, for the sake of brevity, we refer to these collectively as 'First Galley I'.
Let us begin by pointing out the differences in fundamental assumptions between First Galley I and the General Theory: (i) in First Galley I, Keynes assumes that firms are fully integrated, while in the General Theory he allows for the possibility that firms may not be fully integarated８; (ii) in First Galley I, but not in the General Theory, Keynes uses 'B' to denote the cost of the maintenance and improvement of the initial capital equipment. As we shall see, this is a vital point, and determines the differences between the two texts in the definitions of some fundamental concepts. In connection with (i), note that in the General Theory Keynes denotes a certain sum paid on purchasing finished output from other entrepreneurs by A１. In First Galley I, therefore, A１= 0 (in order to compare First Galley I with the General Theory on the same footing, in the discussion below we will assume that A１= 0).
The notation adopted in the General Theory is as follows:
A : the sale proceeds of finished output either to consumers or to other entrepreneurs. Assuming A１= 0, these are equal to the value of consumption.
G : the actual value of capital equipment at the end of the period. In First Galley I, this is denoted by C.
G′ : the greatest value of capital equipment at the end of the period which would have been maintained had it not have been used for production. In First Galley I, this is denoted by C ′.
B : the cost which must be incurred to maintain the value of capital equipment at G′.
In the discussion to follow we will follow this notation insofar as it is possible to do so.
User Cost - This is the opportunity cost of capital equipment which current production requires. We will denote the concept of user cost adopted in First Galley I by U１ and the concept of user cost adopted in the General Theory by U２. These are formulated as follows:
U１ = (C′- B′) - (C - B) (22)
U２ = (G′- B′) - (G - A１) (23)
where B denotes the cost of the maintenance and improvement of the initial capital equipment.
If we use the above notation and assume that A１ = 0, equation (22) becomes:
U１ = (G′- B′) - (G - B) (24)
Equation (23) becomes:
U２ = (G′- B′) - G (25)
The difference between the two equations lies in the fact that equation (24) contains B while equation (25) does not. The expenditure on capital equipment installed in the current period accounts for most of B, so that G - B is the value of capital equipment which contributed to current production at the end of the period. As a result of this, we know that U１ is the opportunity cost of the capital equipment which contributed to the current production. On the other hand, U２ is defined in such a way that the expenditure on capital equipment installed in the current period is deducted from it. Therefore, it is inappropriate for the definition of user cost, which is the opportunity cost of the capital equipment which contributed to current production. (The definition of user cost given in First Galley I would seem to be more appropriate than that of the General Theory). From equations (24) and (25) we can obtain:
U２ = U１ - B (26)
If we accept that U１ is a more appropriate conception of user cost than U２, then it turns out that Keynes makes an extra deduction of B, the cost of the maintenance and improvement of the initial capital equipment, from U１.
Income - Income as conceived in First Galley I, which we can denote by Y１,
is defined by:
Y１ = A + B - U１ (27)
Income as conceived in the General Theory is defined by equation (7) above. If we denote this conception of income by Y２, then:
Y２ = A - U２ (28)
By making use of equation (26), we can obtain:
Y１ - Y２ = B - U１ + U２ = 0
Thus we have shown that Y１ is equal to Y２.
Consumption, Investment and Saving - Let us compare First Galley I with the General Theory with regard to the definitions of consumption, investment and saving. Consumption is equal to A in the system in which firms are completely integrated (that is, A１ = 0). Saving is defined as the excess of income over consumption in both the integrated and non-integrated cases. Because the definitions of consumption and income are the same in both cases, it turns out that the definition of saving is also the same in both cases.
However, in the definition of investment there is some difference between the two cases. Let us denote the conception of investment in First Galley I by I１, and that in the General Theory by I２. Then these are defined by equations (29) and (30) respectively:
I１ = Y - A + U１ (29)
I２ = Y - A (30)
From these we obtain:
I１ = I２ + U１ (31)
Thus I１ is larger than I２ by the amount of user cost as understood in First Galley I, U１. By making use of equation (27), equation (29) can be transformed into the following (remembering that Y = Y１ = Y２):
I１ = B (32)
That is, the cost of maintenance and improvement of the initial capital equipment, B, is equal to investment as understood in First Galley I, I１. Keynes refers to the difference between I１ and user cost, U１, as 'net investment'. Let us denote this by I１ｎ． Then:
I１ｎ = I１ - U１ (33)
From equations (24), (32), and (33) we can obtain:
I１ｎ = G - (G′- B′) (34)
This is the definition of investment adopted in the General Theory, I２ (see GT, p. 55). On the one hand, in First Galley I, 'gross' concepts are defined as inclusive of user cost, U１, while 'net' concepts are exclusive of it. On the other hand, in the General Theory, 'gross' concepts are defined as inclusive of the supplementary cost, V; 'net' concepts as exclusive of it (the supplementary cost is defined as the 'depreciation of the equipment which is involuntary but not unexpected' (GT, p. 56)). It therefore turns out that net investment, I１ｎ, in First Galley I corresponds to gross investment in the General Theory, while net investment in the General Theory can be expressed as I１ｎ - V.
The Factors of Production - In First Galley I the prime cost, which we will denote by E１, is composed of the amount spent on finished goods (the main constituent of which is labour-cost９), and the cost of the maintenance and improvement of the initial capital equipment, B (the main constituent of which is the labour-cost required for production of investment goods within firms).１０ This means that the prime cost is equal to the amount of money which the fully integrated firm pays for both the investment goods which are produced and retained within the firm and the finished output. It does not contain the user cost, U１, and is equal to the factor cost, F, of the General Theory:
E１ = F (35)
In the General Theory, prime cost, which we will denote by E２, is defined as the sum total of the factor cost, F, and the user cost, U２１１:
E２ = F + U２ (36)
The profit of entrepreneurs, as understood in First Galley I, which we will denote as P１, is equal to the income of entrepreneurs as understood in the General Theory, which we will denote as P２. This can be shown as follows. P１ and P２ are respectively defined by:
P１ = A + B - U１ - E１ (37)
P２ = A - E２ (38)
From these we obtain:
P１ - P２ = B - U１ - E１ + E２ (39)
If we make use of the equation B - U１ = - U２ (derived from equation (26)), as well as equations (35) and (36), then equation (39) gives us:
P１ - P２ = - U２ - F + (F + U２) = 0 (40)
This shows that P１ = P２.
The concept of 'quasi-rent' appears in First Galley I, but is dropped in the General Theory. In the former, Keynes distinguishes three kinds of time elements in variables: the actual value, short-period expectation and long-period expectation. He uses one prime (′) to indicate the short-period expectation, two primes (″) to indicate the long-period expectation. In the case of quasi-rent, the actual value of quasi-rent is called 'profit', while its long-period expectation is called 'prospective yields'. The term 'quasi-rent' itself is retained to show the short-period expectation of quasi-rent, and is formulated as follows:
Q = P′+ U１′ (41)
where Q denotes 'quasi-rent'.
Effective Demand - The definition of 'effective demand' in the General The
ory is different from that in First Galley I, though the definition of 'income' is the same. In First Galley I, effective demand includes user cost, U１, but in the General Theory it does not include user cost, U２. If we denote effective demand as understood in First Galley I by D１ and effective demand as understood in the General Theory by D２, then they are conceived as follows:
D１ = Y′+ U１′ (42)
D２ = Y′ (43)
where Y denotes income. Again, one prime (′) indicates short-period expectation.
In First Galley I Keynes explains the concept of effective demand as follows:
...in contradistinction to income, effective demand is reckoned gross of user cost, so that D = Y′+ U′. It is essential to reckon effective demand gross, since it is the gross value of output which absorbs spending power (JMK.14, p. 422).
In the General Theory, effective demand is defined as the 'aggregate income which the entrepreneurs expect to receive, inclusive of the incomes which they will hand on to the other factors of production' (GT, p. 55).
From equations (42) and (43) we can obtain:
D１ = D２ + U１′ (44)
In First Galley I, Keynes argues that effective demand is important because it contains user cost, and distinguishes the concepts inclusive of user cost (such as quasi-rent, and gross investment) from those exclusive of it (such as income, profit and saving). On the other hand, in the General Theory he defines effective demand, investment, income, profit and saving as exclusive of user cost, and does not use quasi-rent. This suggests that the role of user cost in Keynes's theory recedes somewhat in the General Theory.
We can summarise the relation between the definitions of fundamental concepts adopted in First Galley I and in the General Theory as follows: (i) the definitions of effective demand, investment and prime cost differ between the two, the difference depending on whether they include user cost (First Galley I) or not (the General Theory); (ii) the definitions of income, profit and saving are the same in the two; (iii) the equation U２= U１ - B is central to the relation between the two sets of definitions.
<2> Chapters 8 and 9
Chapter 8, 'The Meaning of Saving', of First Galley I corresponds to Chapter 7, 'The Meaning of Saving and Investment, Further Considered' of the General Theory. Chapter 9, 'The Meaning of Saving', of First Galley I corresponds to Chapter 8, 'The Propensity to Consume: I. The Objective Factors', of the General Theory. Overall, the degree of change in the proofing of these chapters is less than that in the cases of Chapters 6 and 7.
Chapter 8 - This chapter is extant in its First Galley I form and in its Second Galley form. Each version of the chapter is composed of three sections. Section II of the First Galley I version, in which Keynes criticises Hayek's theory of forced saving, derives directly from Section VI of Chapter 8, 'Investment and Saving', of the Summer Manuscript. In the Second Galley, Keynes adds to this only a supplementary explanation, in which he states that net investment is necessarily equal to saving, and that the theory of forced saving is an inappropriate application of Bentham's theory to the state of underemployment. In the Second Galley Keynes rewrites his account of how a change in the quantity of money brings about a change in saving through a change in distribution (see JMK.14, p. 430), but the content of the theory is unaltered.
In terms of formal arrangement, the transformation from Chapter 8 of the Second Galley to Chapter 7 of the General Theory is considerable. The latter is composed of five sections. In the General Theory Keynes adds Section I (in which he states that the difference in terminology lies in the difference in the definitions of investment and income) and Section II (in which he defines investment, and criticises Hawtrey's definition of investment as exclusive of liquid capital, as well as the Austrian School's idea of capital formation and capital consumption).
Section III of Chapter 7 of the General Theory corresponds to parts of Section I of Chapter 8 of the Second Galley (JMK.14, pp. 425-427). There is no difference of content between the two. Section IV of Chapter 7 of the General Theory can be said to be virtually the same as Section II of the Second Galley. On the other hand, Section V of Chapter 7 of the General Theory was compiled out of several parts of Chapter 8 of the Second Galley (JMK.14, pp. 428-429, pp. 432-433 and 434-436), plus some new passages (GT, pp. 82-83) in which the relation between the granting of bank credit and the theory of investment-saving equilibrium is discussed.
Chapter 9 - In Chapter 9, 'The Meaning of Investment', Keynes argues that so-called 'financial provision', by functioning as saving, decreases net investment and brings about stagnation in the economy. This argument is already complete in Section IV of Chapter 8, 'Investment and Saving', of 'The General Theory', not only in substance but also in form.
Chapter 9 is extant in both First Galley I and in the Second Galley. The only difference between the two versions lies in the fact that in the Second Galley Keynes adds some material on Kuznets' study of gross capital formation in the United States. This is an addition to a discussion of Clark's study of gross capital formation in the United Kingdom, which is referred to in First Galley I, and does not change the nature of the argument. Chapter 9 of the Second Galley is incorporated into Section IV of Chapter 8 of the General Theory.
(b) The Introductory Chapters
The three chapters which belong to Book I, 'Introduction', are: Chapter 1, 'The General Theory', Chapter 2, 'The Postulates of the Classical Economics', and Chapter 3, 'The Principle of Effective Demand'. (These are reproduced at JMK.14, pp. 351-352, 352-369, and 359-379 respectively.)
Chapter 1 is almost identical to Chapter 1 of the General Theory. Let us here examine Chapters 2 and 3.
<1> Chapter 2
Chapter 2, 'The Postulates of the Classical Economics', is extant in the galleys from First Galley I to the Third Galley. Apart from a few slight changes of expression, there is no real difference between the succeeding versions. In the Great Revision, moreover, there is no change in substance although there are some stylistic changes.
Chapter 2 of the General Theory is comprised of seven sections. In Sections I and II, Keynes at this stage adds 'a decrease in the marginal disutility of labour' to the list of the means of increasing employment which are available to classical economics (GT, p. 7). Apart from this, only minor revisions of style and wording are made here. In Section III Keynes states that the struggle over money-wages does not determine the level of real wages, although 'it affects the distribution of the aggregate real wages between different labour-groups' (GT, p. 14). The section is the same in content as the corresponding part (JMK.14, pp. 363-365) of First Galley I, though there is some change in style. Section IV, which deals with involuntary unemployment, is the same in content as the corresponding parts (JMK.14, pp. 366-369) of First Galley I, though again there is some change in style. The only real difference is that in First Galley I Keynes discusses involuntary unemployment from the point of view of the supply side of labour only, while in the General Theory he discusses it from the point of view of the demand side for labour as well. The argument in Section V of the General Theory, in which Keynes discusses the implications of accepting the first postulate and rejecting the second, was formerly included in Section IV of First Galley I. The argument in Section VI, in which Keynes states that Say's Law underlies the entire classical theory, formerly 'appeared in an abbreviated form' (JMK.14, p. 368) as Section I of Chapter 3 of First Galley I. In the General Theory Keynes adds a quotation from Mill１２ to illustrate the classical economists's belief in Say's Law.１３ Section VII, in which Keynes refers to three assumptions, namely: (a) the real wage is equal to the marginal disutility of the existing employment; (b) there exists no involuntary unemployment; and (c) Say's Law, on which the classical theory depends, was newly written for the General Theory. It does not exist in First Galley I.
<2> Chapter 3
Chapter 3, 'The Principle of Effective Demand', has some relation to the definitions of various fundamental concepts. As far as this chapter is concerned, the galleys from First Galley I to the Third Galley are all extant. The three galleys have in common the following passage, which forms the core of Chapter 3:
By effective demand I mean the sum for which the current output can actually be sold; and by the state of effective demand I mean the schedule relating the sum, for which the current output resulting from employment of any given number of men can be sold, to the number of men employed. In other words if D is the effective demand for the output which results from employing N men...the state of effective demand can be expressed by D=f(N).
Similarly if D ′is the supply price of the output of N men, in the sense that an expectation of sale proceeds D′from their output will just make it worth the while of entrepreneurs to employ N men, the employment function...can be written D′= F(N).
It follows that so long as production is elastic in the sense that an
increase of D′will be associated with an increase in N, producers will tend to increase employment provided that D′is not greater than D, i.e. they will push employment to the point where D′= D (JMK.14, pp. 370-371).
Keynes's formulation here, which is the same one he used in the second of his 1934 Michaelmas lectures, is essentially as follows:
D = f(N) (45)
D′= F(N) (46)
D = D′ (47)
where D denotes effective demand, f(・) the state of effective demand, D′ the supply price, and F(・) the employment function.
The level of employment is determined by these equations. This formulation is basically the same as that in the General Theory, except for terminology. In the General Theory, equation (45) is called the 'aggregate demand function', equation (46) the 'aggregate supply function', D the 'aggregate demand price', D ′ the 'aggregate supply price', and the value of D at the level of employment determined by equation (47) 'effective demand'.
Let us compare this formulation with the formulations (for determining the level of employment) in 'The General Theory' and the Summer Manuscript. As we saw in Chapter 12, the employment function in 'The General Theory' and the Summer Manuscript was used not only as a supply concept but also as an equilibrium concept. This duality is absent in Chapter 3 (JMK.14, pp. 369-378) of the galleys from First Galley I up to and including the Third Galley, in which the function D = f(N) is treated as representing the demand side, and the function of D′= F(N) is treated as representing the supply side. However, in the General Theory the duality reappears, and characterises its theoretical structure. This is considered briefly in the next chapter. (A more detailed discussion can be found in my Reconstruction).
In the galleys from First Galley I to the Third Galley, Keynes argues that income as a realised value differs from effective demand as an expected value by the amount of user cost. According to Keynes, as we saw in (β) above, effective demand is a vitally important concept in the determination of the level of employment because it includes user cost. Moreover, user cost, as we also saw in (β), plays an important role in the definitions of some fundamental concepts such as investment, saving and profit. In equations (45) to (47) above, D and D′include user cost, which, Keynes maintains, has important significance for the formulation.
Although we say that equations (45) to (47) are the formulation used in the General Theory, the truth of this assertion is subject to one proviso, namely that it is permissible to overlook the fact that in the General Theory the definitions of effective demand, investment, income, profit and saving have been changed in such a way that they do not include user cost. In the General Theory Keynes explains the reason for this change as follows: 'since user cost is obviously dependent both on the degree of integration of industry and on the extent to which entrepreneurs buy from one another, there can be no definition of the aggregate sums paid by purchasers, inclusive of user cost, which is independent of these factors' (GT, p. 24, fn.2).
F. The Changeover
As we explained in Section 1 of Chapter 13, 'the Changeover' designates the revision work of First Galley III and the change from the Galley Table of Contents to the table of contents of the General Theory, which took place between September and October 1935. It is evident that up to September Keynes continued to structure his work in line with the Galley Table of Contents. In his letter to Harrod of 17 August 1935, for instance, he writes: 'Here are the last two chapters of my book.... But chapter 26 is too long,.... In chapter 27 the emphasis hasn't worked out as I intended' (JMK. 13, p. 542). Further, in his letter to Hawtrey of 4 September 1935, he reports: 'I have...completed my re-writing of the first three books, namely chapters 1-11' (〓JMK. 13, p. 576). These chapters, then, are patently in line with the Galley Table of Contents.
Even as late as September, Keynes continues to rewrite the chapters which are concerned with the rate of interest. In his letter to Harrod of 10 September 1935, we read: 'I shall be here from September 22 at least up to the end of the month.... It is probable that by then I shall have finished re-writing the chapters dealing with rate of interest' (JMK.13, p. 559). This bears fruit in Chapter 14, 'The Classical Theory of the Rate of Interest', of the General Theory.
Due to the paucity of our sources, however, we know neither when precisely Keynes compiled the table of contents of the General Theory, nor how he rewrote the galley after September. As we saw in (E) above, August was the period of the Great Revision. At that stage Keynes may well have been simultaneously engaged in rewriting the galley in a way which deviates markedly from the Galley Table of Contents, although we lack the evidence to confirm this. Apart from this, it may be surmised that it was in the period September/October that he rearranged the whole galley in line with the table of contents of the General Theory. What remains as a fact is that Hawtrey, in his letter to Keynes dated 19 December 1935, made comments on the proofs which is based on the table of contents of the General Theory, for we find such a passage as 'Own―rates of interest. A renewed study of chapter 17 leads me ...' (JMK.13, p. 625).
Tables 13-2, 13-3, and 13-4 in the preceding chapter show how Keynes rearranged the whole galley. It should be noted that Chapters 6-10 (in the Galley Table of Contents) undergo substantial further change to reach the form of the General Theory, and that there the arrangement again changes as a result of Chapter 22 (of the Galley Table of Contents) being shifted to become Chapter 10 of the General Theory.
In sum, the following points emerge from our analysis in Chapters 13 and 14:
(1) First Galley I represents the most considerable piece of revision work car
ried out on the topics covered in the Galley Table of Contents. The Second and Third Galleys represent only stylistic revisions of First Galley I, and introduce no changes of content.
(2) First Galley I is composed of Chapters 1-19.１４ Of these, Chapters 4, 5, 12
and 13 had been fashioned into completed texts of the corresponding parts of the General Theory prior to First Galley I. The other chapters are completed even on a stylistic level, as far as the Galley Table of Contents is concerned. However, the theory of consumption, the theory of liquidity preference, and the definitions of various fundamental concepts had been completed in substance before First Galley I.
(3) The largest of the changes made in the proofing process from First Galley
I to the Third Galley takes place in the Great Revision. The definitions of some fundamental concepts (such as effective demand, investment, and the prime cost) change considerably as a result of both the change in the definition of 'user cost', and the change in its treatment.
2. The Michaelmas Term Lectures of 1935
After 'The Great Revision' of August 1935 and the Changeover of September 1935, how did Keynes proceed to work on producing the General Theory before its publication in February 1936? Did he make no further changes, or did he rather continue to make substantial changes? It is these questions we must now address. Before doing so, however, we should remind ourselves that, apart from the lecture notes taken by Lorie Tarshis in the Michaelmas Term of 1935 (14 October - 2 December), there remains no material to document the development of Keynes's ideas during this period. (It is also worth noting here that, except for the lecture notes taken by Keynes's students during 1932-35, there remains very little material documenting the development of the theory of liquidity preference tout court.)
The truth is, Keynes did make substantial changes to the text even as late as the lectures of 1935. The lectures were not delivered exactly in accordance with the final text of the General Theory . This indicates that quite possibly Keynes rewrote the galleys during and even after the lectures. We will discuss this point below.
A. The Core Content
In the Michaelmas term of 1935 the theory of the determination of the volume of employment is formulated using the aggregate supply function and the aggregate demand function as the central concepts. The timespan kept in view here is confined to the short period in which capital equipment and the technology of production are given. The theory is formulated as follows:
Z =Ψ(N) (1)
D = f(N) (2)
Ψ(N) = f(N) (3)
Equation (1) is the aggregate supply function. This relates the volume of employment to the sale proceeds, the expectations of which induce entrepreneurs to employ the corresponding volume of employment (Z is the cost of production of the volume of output of N men). Equation (2) is the aggregate demand function. This relates the aggregate demand entrepreneurs expect to encounter for their output to the number of men they employ. The volume of employment is determined at the point at which the two curves intersect. The aggregate demand, D, is composed of demand for consumption, D１, and the demand for investment, D２. When employment increases, both aggregate real income and aggregate consumption increase, but the latter does not increase as greatly as the former. This is expressed by the relation D１ = X(N). Demand for investment is expressed by the relation I = D２ = F(N). It follows that Ψ(N) - X(N) = I, so that N depends on Ψ, X and D２.
The theory of investment Keynes presented in 1935 has the following two characteristics:
(1) The discussion of the marginal efficiency of capital here is less close to that in the General Theory than was the discussion in the lectures of 1934.
(2) Concerning the theory of liquidity preference, in the sixth lecture (18 November) the precautionary motive is made dependent on the rate of interest, while in the seventh lecture (25 November) it is made dependent on income. This change is particularly worthy of note.
Besides these, the following points also deserve mention:
(1) A list of the objective factors responsible for the propensity to consume makes its first appearance so far as the lectures are concerned, though it had already appeared in 'The General Theory' (1934). However, the list of the objective factors enumerated at this point differs somewhat not only from that in 'The General Theory', but also from that in the General Theory. The most salient differences are that in the General Theory changes in the wage-unit are added and changes in the rate of interest (which appears in 'The General Theory' and the lectures for this year) are replaced by changes in the rate of time-discounting.
(2) The qualified items which are listed in the calculation of the multiplier
are not always the same as those in the General Theory.
(3) The discussion of 'the relation between a change in the money wage and effective demand', which is the subject matter of Chapter 19 of the General Theory, differs from that in the General Theory, in both the order and number of items listed.
(4) To the identified means of curing unemployment in the case where the two classical postulates hold good, 'improvement in organisation' is added, though 'improvement in foresight' is not as yet mentioned.
(5) The definitions of various concepts, the key concept among which is user cost, attain the final forms found in the General Theory.
(6) Keynes lists a number of problems for the Quantity Theory of Money which remain insoluble even if many conditions necessary to make the theory hold good are added. The problems mentioned are virtually the same as those discussed in the General Theory.
B. Chronological Analysis
In the first lecture (14 October), Keynes begins by stating that the employment theory of classical economics is based on the two classical postulates. The second postulate is compatible with two kinds of unemployment (frictional and voluntary). If the two postulates hold good, there are only four means of overcoming unemployment. Among these, the first (improvement of the organisation of the labour market) is newly added (in the General Theory this item is described as improvement in organisation or in foresight) while the rest were already mentioned in the first lecture of 1934.
Keynes then goes on to criticise the classical theory from two standpoints. One is a 'prima―facie' criticism, virtually the same as that given in the first lecture of 1934, the only difference being an addition to the effect that 'within a certain range the demand of labour is in terms of a certain money rather than real wage' (Rymes, p. L-2. In the General Theory (p. 8) Keynes changes this to 'within a certain range the demand of labour is for a minimum money-wage and not for a minimum real wage').
The other criticism is more theoretical. The classical employment theory erroneously assumes that real wages depend on money wage bargains between employees and employers. Keynes argues that real wages depend upon other factors in the economic system, and cannot be changed by revising money wages. He then defines the third category of unemployment, involuntary unemployment, which was precluded by the classical theory. This is as in the General Theory. Decrease in the volume of employment may seem related to higher real wages. However, it is actually induced, Keynes argues, not by labour demanding higher real wages, but by other, quite different, forces.
Keynes next goes on to critically discuss the classical economists' acceptance of Say's Law, quoting passages from J.S. Mill and Marshall (GT, pp. 18-19), and mentions four assumptions on which classical economics depends. Of these, 'wage bargain determines real wage' is omitted from the General Theory (see GT, pp. 21-22).
In the second lecture (21 October), the theory explaining how the volume of employment is determined is put forward. We have already discussed this. The argument here is exactly the same as that given at GT (p. 29), the only change being a minor one of notation. Then Keynes refers to the problem of poverty amidst plenty. Accordingly as society becomes richer, Z - D１ becomes larger, while the marginal efficiency of capital generally falls because capital stock is already large (so that D２ tends to be deficient). To escape from this dilemma, Keynes feels that we 'may help by altering C instead of [the] rate of interest', by means of: (i) increasing consumption through more equitable distribution of income; (ii) causing a diminution in the attractiveness of saving. He concludes the lecture by suggesting that the government, which need not care much about the market rate of interest, should increase public investment in order to increase the total investment.
Thus far, the lectures follow the lines of Chapters 2 and 3 of the General Theory. Now, in the third lecture (28 October) Keynes skips over the material contained in Chapters 4 and 5 of the General Theory (he may have judged that he had already sufficiently discussed this material in the third lecture of 1934), and jumps to Chapter 6.
As we saw, the definition of user cost was put forward in the fourth lecture of 1934. It is in 1935 that the definition of user cost attains the form found in the General Theory, as do the definitions of all the other fundamental concepts. The main difference between the lectures of 1934 and those of 1935 is that whereas in the earlier lecture series the difference between 'gross' and 'net' is considered to be user cost, in the later lecture series the difference between 'gross' and 'net' is considered to be the 'supplementary cost', which at that time makes its first appearance.
The income of the community is defined as the sum of the income of the factors of production and the income of the entrepreneurs. In calculating this, user cost should be subtracted. This is the portion of depreciation due to entrepreneurs' voluntary decisions to produce, and is defined as the excess of the net value if the producers had not used their plant, over the net value if they had used it. It is represented as follows:
U = (G１ - B１) - (G - A１ )
where G１ is the value of the stocks of goods, equipment, and so forth, at the end of the period of production, after the entrepreneur has spent the optimum sum, B１, to prevent depreciation; G is the value of the stocks of finished goods, equipment and so forth at the end of the period; and A１ is the value paid out for goods bought during the period.
G１ and B１ correspond respectively to G′ and B′ in the General Theory, though in the latter the terms are defined in relation to the capital equipment only. Denoting the value of finished goods sold during the period by A, the income of the community and the income of the entrepreneurs are as follows:
The income of the community = A - U
The income of the entrepreneurs (gross profits) = A - U - F
Moreover, Keynes introduces expected depreciation, or supplementary cost, V, which is involuntary but foreseen, and by means of this he defines net income as:
Net income = A - U - V
Keynes then defines saving and investment as follows:
Gross saving = A - U - (A - A１) = A１ - U
Net saving = A１ - U - V
Gross investment = A - U - (A - A１) = A１ - U
He concludes the third lecture by critically discussing : (i) Hawtrey's emphasis on windfall increases in investment (this is closely related to Section II of Chapter 7 of the General Theory); (ii) Robertson's definition of saving as well as his own in the Treatise (which is closely related to Section III of Chapter 7 of the General Theory).
The fourth lecture (4 November) corresponds closely to Section IV of Chapter 7 (in which forced saving is critically discussed) and Chapter 8 (in which the objective factors behind the propensity to consume are discussed) of the General Theory. The criticism of forced saving is the same as that found in the General Theory. Keynes then proceeds to discuss the propensity to consume. After stating that real income can be approximated by income expressed in terms of wage-units, and briefly referring to the subjective influences (the subjective factors) in the short period (which he explained in detail in the fifth lecture of 1934), Keynes discusses the objective influences (the objective factors) for the first time as far as the lectures are concerned. Here he mentions five items:
(1) changes in the rate of interest;
(2) changes in fiscal policy;
(3) changes in the ordinary person's expectations as to the relation between present and future income;
(4) windfall changes in capital values which do not affect income, such as movements on the Stock Exchanges;
(5) the gap between income and net income.
As we have already seen, it is in 'The General Theory' (spring 1934) that Keynes for the first time mentioned 'objective factors'. On that occasion, he took up:
(a) the quantity of employment;
(b) the rate of interest;
(c) the state of confidence or long-term expectation.
Clearly, except for items (1) and (b), the two lists do not overlap.
Let us compare the objective factors given in the General Theory with those discussed in the fourth lecture. The items listed in the General Theory are:
(i) a change in the wage-unit (this is taken up not in the fourth lecture, but in the fifth);
(ⅱ) a change in the difference between income and net income (equivalent to item (5));
(ⅲ) windfall changes in capital-values not allowed for in calculating net income (equivalent to item (4));
(iv) changes in the rate of time-discounting (not necessarily equivalent to item (1). See GT, p. 93);
(v) changes in fiscal policy (equivalent to item (2));
(vi) changes in expectations of the relation between the present and the future level of income (equivalent to item (3)).
Comparing, we find that there exist some differences between the two lists, so that there is a considerable possibility that Keynes rewrote this part of the text after 4 November. The most salient differences are that in the General Theory changes in the wage-unit are added, and changes in the rate of interest are transformed into changes in the rate of time-discounting. Besides this, we can also note the following.
In the fourth lecture, item (5) is discussed in detail, and sinking funds are taken up as a key issue, while in the General Theory, item (ii) (which corresponds to item (5)) is only briefly mentioned. Instead, this item is discussed in detail in Section IV of Chapter 8 of the General Theory, in which 'financial provision' plays a central role. As we do not find the term 'financial provision' anywhere in the lecture, it is likely that Keynes wrote Section IV after the lecture. (It is interesting to note that the figures concerning gross investment and net investment mentioned at the end of this lecture are the same as those mentioned at GT, pp. 102-103.)
In the fifth lecture (11 November), Keynes begins by referring to the important influences that changes in income measured in terms of wage-units have on saving. This is indicative of rewriting, for whereas in the fourth lecture this does not constitute one of the objective factors, in the General Theory it is taken up as the first item, as indicated in the list above.
Keynes goes on to examine topics that later form the content of Chapter 10 of the General Theory: the marginal propensity to consume and the multiplier. He first puts forward the marginal propensity to consume as ΔCｗ/ΔYｗ = 1-1/k and the multiplier as k = ΔYｗ /ΔIｗ. This is followed by a discussion of the multiplier's relation to Kahn's employment multiplier. Keynes proceeds to point out allowances which should be made in calculating the multiplier. Although this corresponds to Section III of Chapter 10 of the General Theory, these parts of the text are not always identical to the later version, so that it is likely that Keynes rewrote these parts after the lecture. In the lecture, Keynes lists the following as allowances:
[1] the dole financed by borrowed money should be regarded as disinvestment;
[2] dissaving;
[3] the existence of imported goods.
By contrast, in the General Theory the following points, more general in nature, are mentioned:
[a] the possibility that investment may decrease due to a method's of financing in economic policy causing the rate of interest to rise;
[b] the possibility that investment may decrease due to a government programme's, through psychological confusion, causing liquidity preference to increase or the marginal efficiency of capital to decrease;
[c] the possibility that some portion of increased investment may contribute to an increase in employment abroad (corresponding to item [3]).
Keynes goes on to develop an argument similar to that found in Section V of Chapter 10 of the General Theory, using, if not identical, at least similar figures. He then proceeds to an argument which corresponds to Section VI of Chapter 10 of the General Theory. Historically, there are two types of investment which serve to increase employment: war and gold-mining. Keynes presents his famous story to the effect that a Treasury policy to fill bottles with old bank notes, drop them in disused gold-mines, and have entrepreneurs dig them up, would be an effective method of decreasing unemployment. The source of the difficulty lies in the fact that the causes of saving and investment are very far apart. The community has perhaps become rather timid about grandiose expenditures on investment, possibly on account of - in the case of Britain at any rate - Puritan thinking or Gladstonian ideas. In this connection, Keynes applauded the large-scale expenditures taking place in contemporary Russia, Germany and Italy. Though we do not find expressions like 'Puritan thinking' or 'Gladstonian ideas' in the General Theory, the basic idea is the same as in the lecture.
The main theme of the sixth lecture (18 November) is the inducement to invest, that is, the marginal efficiency of capital and the rate of interest. This corresponds to Chapters 11-13 of the General Theory. Asking the students to refer to Fisher's argument, Keynes defines the marginal efficiency of capital as the 'rate at which prospective yields must be discounted to equal the replacement cost... [the] rate of discount which makes [the] present value of expected yields equal to current replacement cost' (Rymes, p. L-24). He states that the marginal efficiency of capital in general is the rate which is the greatest among them (the same idea is expressed at GT, pp. 135-136).
If the marginal efficiency of capital is higher than the rate of interest, an equilibrium between the two is attained by (i) a fall in prospective yields due to the production of more capital, and/or (ii) a rise in the short-period supply price due to the production of more capital. Keynes states that, with regard to bringing about equilibrium, in the short period (ii) is more important, while in the long period (i) is. The explanation given here corresponds to that at p. 136 of the General Theory, but there is a subtle difference between the two: in the lecture the argument is made in relation to the determination of investment while in the General Theory it is made in relation to the downward slope of the schedule of the marginal efficiency of capital. (Note that the argument of the General Theory had already been discussed in the sixth lecture of 1934).
Keynes next examines Marshall's view of this concept. In the General Theory, after having discussed three ambiguous points surrounding the concept of the marginal efficiency of capital, Keynes states that Marshall's conception is very close to his own. (The three unresolved issues are: (a) Is the marginal efficiency of capital concerned with the physical output of capital or the value of output? (b) Is it an absolute value or a ratio? (c) Is it concerned with Q１ only, or with Q１, Q２, Q３,...Qｎ?) As we already know, this is the point discussed in the sixth lecture of 1934.
In the sixth lecture of 1935, on the other hand, Marshall's view is discussed in relation to (a), (b), and not (c) but instead a fourth issue, namely, (d) Is the marginal efficiency of capital capital's current addition to output? Keynes then proceeds to discuss, in relation to Fisher's theory, whether the marginal efficiency of capital is the real rate of interest. This point was discussed in greater detail in the sixth lecture of 1934, and corresponds to the passage at pp. 142-143 of the General Theory (in which Keynes argues that when a rise in prices is expected, output is stimulated due to a rise in the marginal efficiency of capital). Then Keynes refers to the demand schedule for capital.
As is clear from the above considerations, the marginal efficiency of capital in the sixth lecture of 1934 is, on the whole, closer to that in the General Theory than that in the sixth lecture of 1935 is.
Keynes next proceeds to consider the rate of interest. He begins by criticising the classical theory of the rate of interest. According to this theory, the rate of interest is determined by the interaction of the schedule of the marginal efficiency of capital with the propensity to save. However, saving depends on the level of income, which depends, in turn, upon the rate of investment. Furthermore, if investment is always equal to saving, the two functions will not intersect. (This criticism had already appeared in the seventh lecture of 1933, as well as in the fifth and seventh lectures of 1934.)
Stating that there are two kinds of psychological time preferences ((i) how much to spend/save, (ii) in what form to hold assets), Keynes argues that 'the rate of interest is [the] return [an individual] gets for being ready to lock up his reserves in a non-liquid form' (Rymes [Tarshis], p. L-26). The rate of interest is determined at the level at which individuals' desires to hoard are equal to the amount of money.
Keynes next discusses the motives for liquidity preference. Three are identified: the transaction, precautionary, and the speculative motives. The transaction motive depends upon the volume of business and prices in the short run, and upon changes in banking habits in the long run. The precautionary motive is defined as the desire for safety in certain matters. Keynes asks why people hold money as a precautionary measure. He answers this question, linking this motive to the rate of interest:
The rate of interest exists because of uncertainty as regards [the] future rate of interest. Suppose we always know what rate of interest should be....Then unless for some of the period [the rate of interest] is never negative - then rates of interest for debts of different maturities at which it is always advantageous to hold debt rather than money. ...That is, if you know the rate of interest for all future times - you need not hold any wealth in form of money for precautionary motives' (Rymes [Tarshis], p. L-27).
The speculative motive is defined as the desire to hold money because one believes that one has a more accurate opinion than the market as to the future prices of debts. In the short period this motive is the most sensitive, and depends upon the rate of interest.
Let us now trace out the historical development of the liquidity preference theory. The theory to the effect that the rate of interest is determined by the quantity of money and the operation of liquidity preference was put forward in the fourth lecture of 1932, and is the component which appeared earliest among the principal elements of the General Theory. However, very interestingly, the analysis of the motives for liquidity preference was quite often changed - even immediately prior to the publication of the book. This point, it should be noted, can only be confirmed from the lecture notes taken by Keynes's students. In the eighth lecture of 1933, Keynes divided the motives for holding money into the income, business, precautionary and speculative motives, and argued that the first three depend on the business cycle, overdraft facilities and the rate of interest, while the last depends on the state of bearishness. Then, in the seventh lecture of 1934, he considered the transaction motive and the store-of-wealth motive, and argued that the former depends on expected current operations, while the latter depends on uncertainty as to the future rate of interest. In the sixth lecture of 1935, Keynes states that the transaction motive depends on the volume of business and prices (in the short run), while the precautionary and speculative motives depend on the rate of interest (the precautionary motive might correspond to the store-of-wealth (motive) discussed in the seventh lecture of the preceding year). The story does not end there, however. Rather interestingly, in the seventh lecture of 1935, Keynes comes eventually to consider the precautionary motive as a function of income. This view is adopted in the General Theory, so it is likely that Keynes rewrote this portion of the text immediately prior to publication.
In the seventh lecture (25 November), Keynes considers the topics discussed in Chapters 14-16 of the General Theory. He critically examines the classical theory of the rate of interest. This corresponds to Chapter 14 of the General Theory. According to Keynes, the classical theory can only give the rate of interest when income is fixed at its full employment level, but cannot determine the rate of interest when income fluctuates.
Keynes next discusses the theory of liquidity preference. This corresponds to Chapter 15 of the General Theory. The motives for liquidity preference are identified as (1) the income motive, (2) the business motive, (3) the precautionary motive, and (4) the speculative motive. This is the same set that was given in the eighth lecture of 1933, but Keynes's thinking has changed, for now motives (1), (2) and (3) are stated to be functions of income in the form of L１(Y), and motive (4) to be a function of the rate of interest in the form of L２(R), which is the same position as in the General Theory (the notations are shown just below). This position also differs from the sixth lecture of 1935, in that in the latter the precautionary motive depends on the rate of interest:
Think psychol[ogy] of liquidity funct[ion] L being made of two liquidity functions L１ (motives (1), (2), and (3)) - largely a function of income, and L２ depending mainly on [the] rate of interest... (Rymes [Tarshis],
p. L-29).
Thus the liquidity preference function, L, is composed of functions L１ and L２.
M = L１(Y) + L２(R)
where M is the quantity of money, Y income, R the rate of interest.
In this connection Keynes considers open market operations to be among the most powerful policy tools available to the Bank. If the Bank is prepared to buy and sell at a certain point, the rate of interest would have to be at that point and M will vary accordingly. The argument here corresponds to that at GT, pp. 205-206. Keynes then criticises the classical view of saving as follows:
(1) The act of saving does not guarantee a compensatory increase in consumption in the future;
(2) The act of saving does not mean an increase in investment, but a reorganisation of ownership of wealth;
(3) The act of saving does not alter liquidity preference.
Except for point (3), the argument here relates to Section I of Chapter 16 of the General Theory.
Keynes next briefly refers to an argument which is to be developed in Section II of Chapter 16 of the General Theory. That is, it is preferable to speak of a piece of capital as having a yield in the course of its life in excess of what it cost, because it is scarce rather than because it is productive.
Keynes concludes the seventh lecture by telling a story which relates to Section III of Chapter 16 of the General Theory (the story about the situation in which the rate of interest would become zero).
In the eighth (final) lecture (2 December), Keynes begins by stating that money is the durable asset whose own rate of interest is the greatest (as far as lectures are concerned, he first referred to this point which has something to do with Chapter 17 of the General Theory). Keynes then considers issues to be discussed in Sections II and III of Chapter 21 of the General Theory . For example, he lists the insoluble problems the quantity theory of money would still face, even if all and any assumptions which are necessary for it to hold good were to be added. The points taken up are in substance the same in the lecture and the book. Strictly speaking, however, while in the lecture decreasing returns and a bottleneck in some goods are dealt with together, in the book they are explained separately. This suggests that Keynes rewrote this portion of the text after 2 December. Keynes proceeds to a discussion corresponding to Section VI of Chapter 21 of the General Theory ('a generalised statement of the quantity theory of money').
Finally, Keynes takes up the relation between changes in money wage and effective demand (the main theme of Chapter 19 of the General Theory). The discussion at pp. 262-264 of the General Theory corresponds to the items listed at this point in the lecture, though the order and number of items considered differs. This also indicates a strong possibility that Keynes rewrote this material after 2 December. Item (1) in the lecture (the effect of decreasing liquidity preference) becomes item (5) in the book; item (2) (unfavourable effect on consumption) becomes item (1); item (3) (a favourable effect on the schedule of the marginal efficiency of capital) becomes item (4); item (4) (a worsening of the terms of trade) becomes item (3). The following items considered in the General Theory are not mentioned in the lecture, however: (2) (an increase in the balance of trade), (6) (an optimistic effect on the psychology of entrepreneurs), and (7) (the effect of an increase in debts on the psychology of entrepreneurs).
If we compare the 1935 lectures with the 1934 ones, the following points emerge:
(i) As to the formulation of 'the principle of effective demand', there is little or no difference between the two, notwithstanding some differences in terminology, such as in the definition of effective demand, and the names of functions, and other notations.
(ⅱ) In 1934 the difference between 'gross' and 'net' centres around user cost, while in 1935 it centres around supplementary cost.
(ⅲ) In 1935 we see the first appearance in Keynes's lectures of the objective factors behind the propensity to consume.
(ⅳ) There are some difference between the two lecture series in the analysis
of the motives for liquidity preference.
(v) The material which appears in Chapters 16, 17, 19, 21 of the General Theory is discussed only in the lectures of 1935.
3. The Formative Process
A. Order of Establishment of the Major Theories Comprising the GeneralTheory
The earliest established of the key theories comprising the General Theory is the theory of liquidity preference, which states that the rate of interest is determined at the point at which the quantity of money equals the preference for liquidity. (The analysis of the motives for liquidity preference is, however, another matter). The theory makes its first appearance in 'The Monetary Theory of Production' (mid 1932; see Section 2 (D) of Chapter 8), and is put forward as the relation ρ = A(M), which has a negative slope, in the fourth Michaelmas lecture (31 October) of 1932. Next comes acceptance of the first postulate of classical economics with rejection of the second. This position is stated for the first time in either the Second Manuscript or the first Michaelmas lecture (16 October) of 1933. Then, in the fourth lecture (1933), the importance of adopting money value and employment as units in an analysis of the economy, and the importance of short-period and long-period expectations are discussed for the first time.
The next major development comes in the sixth lecture of 1933. The normal (fundamental) psychological law, which was then stated in the First Undated Manuscript at the end of 1933) and the proposition that income is determined in such a way that Y = C + I is satisfied are clearly stated. In the seventh lecture of 1933 the marginal propensity to consume and the multiplier theory are put forward.
Table14-1 Michaelmas Term of 1935
No.
Contents
1st
14
Oct.
・A formulation of the classical theory of employment and criticism thereof (`prima― facie' criticism and a theoretical criticism).
・Criticism of Say's Law and four assumptions upon which classical economics is
grounded.
2nd
21
Oct.
・The theory of the determination of the volume of employment by means of the
aggregate supply function and the aggregate demand function.
・Poverty amidst plenty - the means of escaping from this dilemma:
(i) More equitable distribution of income; (ii) Diminution in the attractiveness of saving; (iii) Increase in public investment.
3rd
28
Oct.
・Definition of user cost, income, saving and investment, and introduction of supplementary cost.
・Criticisms of Hawtrey's idea of investment, Robertson's definition of saving, and
Keynes's own definition in the Treatise.
4th
4
Nov.
・Criticism of forced saving.
・The propensity to consume.
・Subjective influences (factors).
・Objective influences (factors) - items differ from those in the General Theory.
5th
11
Nov.
・The importance of the influence which a change in income measured in terms of wage-units has on saving.
・The marginal propensity to consume and the multiplier.
・Allowances one should make in calculating the multiplier.
・Two types of investment for increasing employment ― war and gold-mining. ・Criticism of timidity towards grandiose expenditures, and applause
of grandiose expenditures in Russia, Germany and Italy.
6th
18
Nov.
・The inducement to invest - the marginal efficiency of capital and the rate of interest. ・An examination of Marshall's view of the marginal efficiency of capital. ・The demand schedule for capital. ・Criticism of the classical theory of the rate of interest.
・Two types of psychological time preferences of an individual.
・Motives for liquidity preference - the transaction motive which depends on the volume of business and prices (in the short run), the precautionary motive which depends on the rate of interest, and the speculative motive which depends on the
rate of interest.
7th
25
Nov.
・Criticism of the classical theory of the rate of interest.
・The theory of liquidity preference - the income motive, the
business motive and the precautionary motive depend on income,
while the speculative motive depends on the rate of interest.
・Criticism of the classical theory of saving.
・The story about the situation in which the rate of interest would become zero.
8th
2
Dec.
・Money - the durable asset whose own rate of interest is the greatest. ・All and any assumptions required in order for the quantity theory of money to hold good, and a list of still insoluble problems.
・A generalised statement of the quantity theory of money.
・The relation between changes in money wages and effective demand.
In the Second Undated Manuscript (between the end of 1933 and the first half of 1934), the investment theory as determined either by an equilibrium of the demand and supply prices, or by an equilibrium of the marginal efficiency of capital and the rate of interest, makes its first appearance. (The marginal efficiency of capital appears for the first time in this manuscript, though the definition is different from that in the General Theory). In 'The General Theory' (spring 1934), mention of subjective and objective factors first appears, together with the wage unit (though the objective factors differ from those in the General Theory ). In this manuscript, moreover, the investment theory is established. Then, in the Summer Manuscript, the term 'user cost' makes its first appearance (though used differently from that in the General Theory). The definition of user cost becomes the same as in the General Theory in the fourth lecture (12 November) of 1934.
The theory of employment, in tandem with the related concepts such as effective demand, employment function, user cost, income, 'gross', 'net', and so forth, continues to undergo change. The process begins with the formulation found in the First Manuscript of 1933, which then goes through three major reformulations: in the Michaelmas lectures of 1933; in 'The General Theory'; and in The Summer Manuscript. The theory goes through further reformulation in the Michaelmas lectures of 1934 and 1935, before finally reaching the state found in the General Theory.
The same is true of the analysis of the motives for liquidity preference. This continues to be altered through the Michaelmas lectures of 1933, 1934 and 1935. Even at the final stage two significant changes occur. (Keep in mind that the role played by user cost in defining fundamental concepts changes with the appearance of supplementary cost in the Michaelmas lectures of 1935.) Even in the fourth lecture of the Michaelmas terms of 1935, the objective factors are not the same as those given in the General Theory.
In the context of the developmental sequence we have outlined here, we need to comment on Keynes's own account of the genesis of his major ideas. In his famous letter to Harrod dated 30 August 1936 he explained:
To me, the most extraordinary thing regarded historically, is the complete disappearance of the theory of the demand and supply for output as a whole, i.e. the theory of employment, after it had been for a quarter of a century２８ the most discussed thing in economics. One of the most important transitions for me, after my Treatise on Money had been published, was suddenly realising this. It only came after I had enunciated to myself the psychological law that, when income increases, the gap between income and consumption will increase, - a conclusion of vast importance to my own thinking but not apparaently, expressed just like this, to anyone else's. Then, appreciably later, came the notion of interest as being the meaning of liquidity preference, which became quite clear in my mind the moment I thought of it. And last fall, after an immense lot of muddling and many drafts, the proper definition of the marginal efficiency of capital linked up one thing with another (JMK.14, p. 85).
His explanation runs as follows: the [fundamental] psychological law; the theory of the demand and supply for output as a whole; the theory of liquidity preference; the marginal efficiency of capital.
So far as we judge from the surviving material, however, the true sequence, which is quite different from Keynes's retrospective explanation, runs as follows:
(1) The theory of liquidity preference makes its first appearance in mid 1932, and is put forward asρ = A(M) on 31 October in 1932;
(2) The 'theory of the demand and supply for output as a whole', i.e. the theory of employment, appears for the first time in the First Manuscript of 1933 (prior to October, because it rejects the two postulates of the classical economics), followed by his presentation of the two systems of equations in which either income or the level of employment is determined in his eighth lecture on 4 December 1933 ;
(3) The fundamental psychological law appears for the first time in his sixth lecture (20 November) of 1933 and in the First Undated Manuscript;
(4) The marginal efficiency of capital, though different from that in the General Theory, appears for the first time in the Second Undated Manuscript written during the period between the end of 1933 and the first half of 1934;
(5) Subsequently, both the theory of employment and the analysis of the motives for liquidity preference as well as the concept of the marginal efficiency of capital were reformulated several times before their final forms were attained in the General Theory.
How can Keynes's statement be reconciled with what we know about the developmental sequence? What Keynes intended to emphasise in the statement１５ quoted above may have been that the principal stimulus that enabled him to progress from the Treatise to the General Theory was the discovery of the theory of income (or employment), whereas the discovery of the theory of liquidity preference played only a subsidiary role in this process. Moreover we may pay attention to a distinction between the conception of 'new' ideas and the elaboration or maturity of them which takes some period of time. Or there is a possibility that there disappeared some relevant manuscripts. After allowing for them, however, the following fact remains; the question of when each component in the General Theory came to Keynes's mind for the first time is one thing, and the question of how Keynes retrospectively thought of the sequence is another.
Table 14-2 Chronological Order of Development of the Major
Theories of the General Theory
Period
Main Theories
1932
・Liquidity preference theory ρ= A (M)
(The determination of the rate of interest.)
1933
・the theory of demand and
supply for output as a whole
・Changes (the theory of employment).
in the ・Acceptance of the first
analysis postulate and rejection
of motives for of the second.
liquidity ・Units and expectations.
preference. ・The fundamental psychological
law.
・The marginal propensity to
consume and the multiplier
theory.
・Changes in
the theory
of employment,
which
accompany
those in related concepts such as effective
demand,
employment
function,
user cost, income,
'gross' and 'net'.
1934
1935
・Subjective factors and wage unit.
・Objective factors.
・Establishment of the theory
of investment (the marginal
efficiency of capital).
・User cost.
B. Process of Development of the Chapters of the General Theory
It is difficult to establish precisely how each chapter of the General Theory evolved into its final form. Keynes frequently rewrote the table of contents before finally arriving at the table we have in the General Theory. This contributes to the fact that the individual chapters have complicated and intertwined histories. Briefly, the sequence of tables of contents is:
(1) Two tables of contents written in 1932 (see Table 8-1. The first is entitled 'The Monetary Theory of Production').
(2) Three tables of contents written in 1933 (see Table 10-1. The first is entitled 'The Monetary Theory of Employment'; the next two are both entitled 'The General Theory of Employment').
(3) The table of contents of 'The General Theory of Employment, Interest and Money' written in the spring of 1934 (see Table 12-1).
(4) The Galley Table of Contents (see Tables 13-2, 13-3 and 13-4 which cover the period between October 1934 and September 1935).
(5) The Changeover from the Galley Table of Contents to the table of contents of the General Theory (September 1935).
Of course, the question of when exactly Keynes's ideas originally came to him, and how he developed them, is quite different from the question of when each chapter of the General Theory was first drafted and how each evolved into its final form. In trying to understand the theoretical development from the Treatise to the General Theory the more important issue is, naturally, the former. We dealt briefly with this question in Section 3 (A) above and shall return to it in Chapter 16.
Having said that, it is nevertheless quite useful for understanding the origins of the General Theory to describe the gestations of the individual chapters. Table 14-3 presents our findings in this regard, based on the argument of this chapter and the last.
The chapter of the General Theory completed earliest in both substance and form was Chapter 12, 'The State of Long-Term Expectation'. This was accomplished in the 'Pre-First Proof Index Version' ('The General Theory'). Then came Chapter 10, 'The Marginal Propensity to Consume and the Multiplier', which was completed in substance in the First Undated Manuscript. This was followed by Chapter 9, 'The Propensity to Consume: II. The Subjective Factors', which was established in substance in 'The General Theory'. This in turn was followed by Chapter 4, 'The Choice of Units', Chapter 5, 'Expectation as Determining Output and Employment', and Chapter 11, 'The Marginal Efficiency of Capital', all of which were completed in the 'Pre-First Proof Typescript', written in the summer of 1934.
The chapters whose contents are complete in First Galley I are: Chapter 1, 'The General Theory'; Chapter 2, 'The Postulates of the Classical Economics'; Chapter 3, 'The Principle of Effective Demand'; Chapter 13, 'The General Theory of the Rate of Interest'; Chapter 14, 'The Classical Theory of the Rate of Interest'; Chapter 15, 'The Psychological and Business Incentives to Liquidity'; and Chapter 16, 'Sundry Observations on the Nature of Capital'.１６
Next came Chapter 17, 'The Essential Properties of Interest and Money', which was completed in the Second Galley. This was followed by the chapters completed in First Galley II: Chapter 18, 'The General Theory of Employment Re-Stated'; Chapter 19, 'Changes in Money-Wages'; Chapter 20, 'The Employment Function'; Chapter 21, 'The Theory of Prices'; and (probably) Chapter 22, 'Notes on the Trade Cycle'.
For the rest, two things should be borne in mind:
(1) The following chapters were completed after First Galley III: Chapter 23, 'Notes on Mercantilism, the Usury Laws, Stamped Money and Theories of Under-Consumption'; Chapter 24, 'Concluding Notes on the Social Philosophy towards which the General Theory Might Lead';
(2) The following chapters were established as a result of the Great Revision: Chapter 6, 'The Definition of Income, Saving and Investment'; 'Appendix on User Cost'; Chapter 7, 'The Meaning of Saving and Investment Further Considered'; Chapter 8, 'The Propensity to Consume: I. The Objective Factors'; Chapter 14, 'The Classical Theory of the Rate of Interest'; and 'Appendix on the Rate of Interest in Marshall's Principles of Economics, Ricardo's Principles of Political Economy, and Elsewhere'.
Having made these points, we should emphasise again that it is very difficult to describe accurately how each chapter of the General Theory came into being and evolved into its final form. This is largely because Keynes rewrote so many times, and continued to do so almost to the end. In fact, as we saw in Section 2 of this chapter, it is likely that he continued to amend and rewrite the galleys even after the Changeover of September and October 1935.
Table 14-3 The Process of Development of the Chapters of
the General Theory
Book
Chapter
The Process of the Development
Book
Chapter 1, “The General Theory"
・Finalised in First Galley I in substance and form.
Chapter 2, “The Postulates of the Classical Economics"
・The idea begins as early as the Second
Manuscript of 1933 or the Michaelmas lectures of 1933.
・Completed in fact in First Galley I.
Chapter 3, “The Principle
of Effective Demand"
・The idea first occurs as early as the First Manuscript of 1933 or the Michaelmas lectures of 1933.
・Various changes up to First Galley I or the Michaelmas lectures of 1934 and 1935.
Book
II
Chapter 4, “The Choice of Units"
・The idea first occurs as early as the Michaelmas lectures of 1933. ・Completed in fact in the Pre-First Proof Typescript.
Chapter 5, “Expectatio
n as Determining Outputand Employment"
・The idea first occurs as early as the Michaelmas lectures of 1933. ・Completed in fact in the Pre-First Proof Typescript.
Chapter 6, “The Definition of Income, Saving and Investment"
・The idea first occurs as early as the Michaelmas lecture of 1933. ・The original form first occurs in the Pre-First Proof Typescript. ・Has a corresponding part in First Galley I. ・Undergoes alteration in the Great Revision.
Appendix on User Cost
・The idea first occurs as early as the Summer Manuscript or the Michaelmas lectures of 1933.
・Has a corresponding part in First Galley I. ・Undergoes alteration in the Great Revision.
Chapter 7, “The Meaning of Saving and Investment Further Considered"
・Has a corresponding part in First Galley I. ・Undergoes alteration in the Great Revision.
Book
III
Chapter 8, “The Propensity to Consume: I. The Objective Factors"
・Has a corresponding part in `The General Theory'.
・Undergoes alteration in the Great Revision.
Chapter 9, “The Propensity to Consume: II. The Subjective Factors"
・Has a corresponding part in `The General Theory'.
・Finalised in substance in `The General Theory'.
・Completed in form in First Galley I.
Chapter 10, “The Marginal
Propensity to Consume and
the Multiplier"
・The idea first occurs as early as the Michaelmas lectures of 1933. ・Finalised in substance in the First Undated Manuscript. ・Finalised in form in First Galley II.
Book
Chapter
The Process of the Development
Book IV
Chapter 11, “The Marginal
Efficiency of Capital"
・The idea first occurs as early as Second Undated Manuscript or the Michaelmas lectures of 1934.
・Completed in substance and form in the Pre-First Proof Typescript.
Chapter 12, “The State of
Long- Term Expectation"
・The idea (volatility and irrationality of expected quasi-rents) first occurs as early as the Michaelmas lectures of 1933. ・Completed in substance and form in the Pre-First Proof Index Version (`The General Theory')
Chapter 13, “The General Theory of the Rate of Interest".
・The idea first occurs as early as `The Monetary Theory of Production'. ・Finalised in First Galley I.
Chapter 14, “The Classical
Theory of the Rate of
Interest"
・The idea first occurs as early as the Michaelmas lectures of 1932. ・Finalised in substance in First GalleyⅠ. ・Undergoes alteration in the Great Revision.
Appendix on the Rate of Interes in Marshall'sPrinciples of Economics,Ricardo's Principles of Political Economy, and Elsewhere.
・The idea first occurs as early as the Michaelmas lectures of 1932 or the Third Manuscript of 1933.
・Undergoes alteration in the Great Revision.
Chapter 15, “The Psychological and Business Incentives to Liquidity"
・The idea first occurs as early as the Michaelmas lectures of 1933. ・Finalised in substance in First Galley I. ・Finalised in form in the Second Galley.
Chapter 16, “Sundry Observations on the Nature of Capital"
・Has a corresponding part in the Second Undated Manuscript and the Pre-First Proof Typescript.
・Finalised in substance and form in First Galley I.
Chapter 17, “The Essential
Properties of Interest
and Money"
・Finalised in the Second Galley.
Chapter 18, “The General Theory of Employment Re-Stated"
・Has a corresponding part in `The General Theory' and The Summer Manuscript. ・Finalised in substance and form in First Galley II.
Boo
k
Chapter 19, “Changes in
Money-Wages"
・Finalised in substance and form in First
Galley II.
Appendix on Prof.
Pigou'sTheory of
Unemployment
・The idea first occurs as early as the
Michaelmas lectures of 1933.
Chapter 20, “The
Employment Function"
・The idea first occurs as early as `The
General Theory' or the Summer Manuscript.
・Finalised in substance and form in First
Galley II.
Chapter 21, “The Theory
of Prices"
・Finalised in substance and form in First
Galley II.
Boo
k
VI
Chapter 22, “Notes on
the Trade Cycle".
・(Probably) finalised in First Galley II.
Chapter 23, “Notes on
Mercantilism, the Usury
Laws, Stamped Money
and Theories of
Under-Consumption"
・The idea first occurs as early as the
Michaelmas lectures of 1932.
・Has a corresponding part in First Galley III.
Chapter 24, “Concluding
Notes on the Social
Philosphy towards Which
the General Theory Might
Lead"
・Has a corresponding part
in First Galley III.
(Note) Cf. Tables 13-2, 13-3 and 13-4.
Notes
1) Cf. JMK.13, p. 442 and 436.
2) Cf. JMK.13, p. 435.
3) However, we may find its origin in the remarks at JMK.13, p. 431, (iii).
4) Cf. JMK.13, p. 431.
5) Based on the statements at JMK.14, p. 400 (fn.2) and 412 (fn.2).
6) Cf. JMK.14, p. 400 (fn.2).
7) Cf. JMK.14, p. 400 (fn.2).
8) Cf. GT, p. 55.
9) Because we are assuming a fully integrated firm.
10) Cf. JMK.14, p. 404.
11) Cf. GT, p. 53.
12) As Becker and Baumol (1952) showed, and Blaug (1985, pp.150-157) explained very clearly, Say's Law comes in two versions: Say's Identity and Say's Equality. In spite of the fact that Mill refers to both of these (in Chapter 14 of Book III of his Principles), Keynes refers to Say's Identity only. See also Davis and Casey (1977). Blaug interprets Say's Law as assuming a monetary economy and Walras's Law as assuming a barter economy. This interpretation differs from that of Lange.
Recently Kates (1998) maintains, challenging Sowell's classics (1972), that Keynes's formulation of Say's Law and the (Becker and Baumol-type) modern interpretations influenced by this are different from what classical economists had in mind ('Recessions were not caused by a failure of demand, but rather were due to problems associated with the structure of demand relative to the structure of supply' (p. 4)).
13) Lange (1942) accurately formulates Say's Law after Keynes. Lange formulates Walras's Law as a proposition to the effect that the sum total of excess demand value of all goods inclusive of goods as means of payment, is necessarily equal to zero, and formulates Say's Law as a proposition to the effect that the sum total of excess demand value of all goods except for goods as means of payment, is equal to zero. Thus he clearly distinguishes Walras's Law from Say's Law. According to Clower, Keynes's attack on Say's Law amounts to an attack on Walras's Law in the sense of Lange. Cf. Leijonhufvud (1968), p. 68, fn.1. Negishi (1981, Chapter 7, 'Malthus and the Impossibility of Full Employment'), has an interesting analysis of Malthus's attack on Say's Law from the point of view of the problem of over-determination.
14) Robertson's letter to Keynes dated 19 January 1935 and Keynes's response to this letter (JMK.13, pp. 493-496) are useful for understanding First Galley I. Among the discussions on First Galley I between Keynes and Robertson, the most important occurs in Robertson's letter to Keynes of 3 February 1935 and Keynes's response to this letter (JMK.13, pp. 496-520), in which the principle of effective demand and the theory of liquidity preference are debated. Robertson's brief criticism of the principle of effective demand (and the multiplier theory) is found in Robertson (1940, Chapter IX). Robertson's most coherent criticism of the theory of liquidity preference is found in Robertson (1940, Chapter I). This is connected with the problem to which we referred in Note 15 to Chapter 13 of the present book. It should be also noted that at the same period there was a severe controversy between Harrod and Robertson. Roberston says in his letter to Harrod dated 27 September 1934 that 'if your line of reasoning [developed in Harrod (1934)] is right, it makes nonsense of everything which I (as well as Hayek and company) have been trying to say for the last eight years'. Harrod writes in his letter to Robertson dated 3 October 1935 that 'My only ray of sunshine is Maynard's book (GTE). I think I agree with him (as against you) that his book, if true, is pathbreaking'. For this see Young (1989), pp. 75-82.
In relation to the proofreading process we need to pay attention to Keynes's discussion with Hawtrey, together with Keynes's discussion with Robertson and Harrod to which we referred in Note 15 to Chapter 13 of the present book. The sources before the General Theory are contained in JMK.13 (pp. 567-633) and those after are contained in JMK.14 (pp. 2-55). The most important document is Hawtrey's comment to Keynes (JMK.13, pp. 567-576). In his letter to Hawtrey of 4 September ( JMK.13, pp. 576-577), Keynes states that he is revising the galley (this is the Great Revision), taking Hawtrey's comment into consideration. From an examination of the Variorum we can confirm that Keynes at this time revised as follows: (i) the quasi-rent was deleted (Chapter 7 of the Galley Table of Contents is the main place in which Keynes makes revision); (ii) the concept of the marginal efficiency of capital was applied to working capital. However, the debate between Keynes and Hawtrey continued to be fierce after the publication of the General Theory. In this connection, see especially Keynes's letter to Hawtrey of 24 March 1936 (JMK.14, pp.14-18).
15) In 1937 Keynes also states that 'the initial novelty lies in my maintaining that it is not the rate of interest, but the level of incomes which ensures equality between saving and investment. The arguments which lead up to this initial conclusion are independent of my subsequent theory of the rate of interest, and in fact I reached it before I had reached the latter theory'(JMK.14, p. 212). For the order of genesis of the main components of the General Theory, see Moggridge (1992, pp. 558-559), and Clarke (1988, pp. 259-264).
16) Chapter 16 is the most important one for Harcourt and O'Shaughnessy, maintaining that here we find the 'denial of a crucial aspect of Say's law in the saving-investment interrelationship which is vital for Keynes's theory of effective demand' (Harcourt ed., 1985, p. 24). They declare themselves to be chapter 16 Keynesians, showing their scepticism of the detailed arguments of Chapter 17.
