467. Harrod to J. M. Keynes , 19 August 1935 [a]
[Replies to 466 , continues at 468 ]
Sennen Cove Hotel, Sennen Cove, Cornwall #
19 August 1935
Your last chapters have just arrived. We intend to return to London on Thursday the 22 nd . But our plans are not rigidly fixed, so would you write to Christ Church, which always has the proper instructions about forwarding.
I thought the revised draft of Book I very clear & good. I have made one or two further notes on it, which I enclose.
With regard to interest & the classical theory, I will try to isolate the offending phrases or sentences & let you have them.
I have just discovered Angus Davidson,  who lives in a very small village here remote from the world.
2nd draft. 
Ch. I. line 5.  I should like you to say instead of "Classical Theory of Economics"--"Classical theory of the subject". It is less tremendous, but, I feel, a more accurate description of what you have done, and \ desirable. The rest of the chapter would follow on quite logically. 1
Galley 7. III. 2nd. sentence.  To make the final "it" grammatical, dont you want "who consent to a reduction of money wages" instead of "who weaken <in> [b] their demand for m-w"? 2
Galley 8. IV. Def. in italics. 3  I prefer your former definition. It seems to me clearer; also I think it is better because it does not bring in demand.
With the classical notion of a continuous supply curve, the proposition that any fall in real wages would lead to a withdrawal of labour, is tantamount to saying that at the existing wage no more labour will be forthcoming. You have made the requisite breach with classical theory by pointing out that at the existing real wage and even at a lower one more labour is available than employed. I think the introduction of demand is confusing at this point. 4
Galley 12. (I ought to have mentioned this before.) 4th paragraph  "If this were true ...." I find this sentence awfully difficult. Take the words "... will no longer be accompanied by any increase of output." Dont you have to add at existing money wage rates? But what does Say's Law say about that? It does not preclude the screwing up of output by inducing more labour by offering higher rates: Z will still be equal to D. Does output as a whole ever cease to have some elasticity? 5 Anyhow not till long after the "full employment" of your former definition has been reached. 6
The most that I can make of this reasoning is the following:--
Any increase of demand makes its own supply: there is no such thing as un-employment due to deficient aggregate demand. Competition will \ induce entrepreneurs to expand without limit. But they will not expand if so doing puts them at a competitive disadvantage; this can only happen if in endeavouring to expand they have to offer higher prices to attract factors from each other. But so long as there is involuntary un-employment, they will not have to do this. \ if there is involuntary unemployment there is no obstacle to expansion. \ Say's Law provides for full employment in your sense.
Is this a correct interpretation? Perhaps I have missed some point. I confess I can make nothing of a state in which output is absolutely inelastic. 7
But I am not sure that the argument as I have set it out is absolutely watertight, because tho' diminishing returns is no obstacle to a general expansion, demand price always being equal to cost, however high, yet it might provide a force keeping competitors in check, one entrepreneur being in a position in which he cant expand owing to increasing costs unless the others do.
I am in agreement with the concluding sentences of this paragraph.
Galley 13. II. line 7.  "costs". 8 This must be "prices of factors". Surely you dont want to assume here (in contradiction to preceding pages) that costs per unit of output are constant in ordinary sense. 9 Yet from your footnote 1 I judge that you do. Such an assumption does violence to the general picture you have given on preceding pages where you suppose increased costs and also is in direct contradiction to (b) below, for by it the m.p. of labour is the same for every value of N. 10
If you make the change suggested, footnote (1) must go.
Reference to the temporary assumption at the top of galley 14 suggests that you do mean what I mean. It was your footnote (1) that put me out and it must surely be a slip. 11
Galley 14a. 1st full para.  Steeped as I am in the mutual determinacy of the classical system, I very much dislike the first 2 sentences. I dare say you wont agree! They seem to be over-dogmatic, and I dont believe in the last resort that pure theory can yield a causal proposition of this kind! What I should like would be in place of "real wages" in line 2, "marginal disutility of labour measured in real terms", and in place of the 2nd sentence a statement to the effect that the prop. to con. & the rate of new inv. are uniquely related to a given level of employment and that in turn to a given level of real wages. 12
2. This draft of the General Theory (1936) is not extant: see Keynes's Collected Writings (1971-89), vol. XIII, p. 542.
3. General Theory (1936), p. 3. Harrod's suggestion was accepted.
4. General Theory (1936), p. 14 top. Harrod's suggestion was incorporated.
5. General Theory (1936), p. 15. The older definition of involuntary unemployment is in Collected Writings, vol. XIV, p. 366 middle.
6. General Theory (1936), p. 26, last paragraph.
7. Possibly General Theory (1936), p. 27, first paragraph (if this is the case, Harrod wrongly wrote "III" for "II").
8. General Theory (1936), p. 30, middle.
- a. ALS, one page, with three pages AN, pencilled, in JMK GTE/1/310-13. Printed in Keynes, Collected Writings, vol. XIII, pp. 542-44.
b. Possibly «to».
top of page
Return to index of this section
Go to previous page
Go to next page