458. Harrod to J. M. Keynes , 31 July 1935 [a]

[Replies to 451 , continued in 459 ]

51, Campden Hill Square, W. 8. #

31 July 1935

Dear Maynard

I have read with great interest to the end of Bk II. [1] The enclosed are merely some jottings on particular points. But I am not sending back my galleys, as I shall probably want to re-read, when I have finished the whole of what I have. I cant yet make any sensible general remarks. I am enjoying it very much.



Galley 3. 1 2nd postulate. [2]

I do not find this very crucial section altogether clearly expressed. As it is so important, I do suggest re-drafting.

You are really dealing with 3 sorts of unemployment.

1. What I will call frictional unemployment, allowed for in classical theory and described by you down to the words "in addition".

2. Voluntary unemployment--at least you say "it cannot be described as involuntary", which according to classical theory is all the unemployment there is besides the frictional. This you deal with from the words "in addition" to the end of that para. [3]

3. Involuntary unemployment (in addition to the frictional) to be defined by you later.

I dont think this threefold division comes out clearly. Nor is it plain what you mean when you say that unemployment (II) "cannot be described as involuntary." I think it a great pity to introduce a reference to "involuntary unemployment", which is carefully defined later, by a side-wind. (Frictional unemployment (I) is also in a sense involuntary)

Moreover when you say "subject to these admittedly important qualifications", you are really, but not explicitly or clearly, referring back to matters mentioned before the words "in addition". For the unemployment due to the cause mentioned after these words is not a qualification but is precisely that determined by the classical postulates themselves. What follows explains how much unemployment of the kind referred to in the "in addition" sentence there will be according to classical theory. [4]

Galley 5. I find the 2nd para obscure. [5] "at variance with the facts" suggests dogmatism in rather a challenging way. I dont see that you would lose anything by omitting this. I think I see what you mean, but it seems to me that it might give a critical reader opportunity for cavilling.

Galley 7. III para 3. [6] I agree with the general drift of this section. But the para 3 is a howling paradox in its present form, unless you insert some such words at the end of line 1 as "albeit unconsciously." Industrial struggles are clearly not consciously struggles about the sharing between particular trades of a given average wage. Every labour man would repudiate this violently. 2

The industrial struggles are almost always intended to strike at capitalists not fellow workers. And if there is no strike against rising prices, that may be because it is thought (even if wrongly) that the whole community (including capitalists) is in the same boat. I press this point because I do think it important in this book that you should avoid un-necessary paradox.

Galley 13 IV paras 2 and 4. [7] Ambiguous [b] expressions: " will increase by less than the increase in ." "The part of D will increase more slowly than . [c] " What you mean is that the arithmetic addition to is less than the arithmetic addition to D. (I believe). But your expressions, especially the second, might be interpreted as "the proportionate increase in " (which starts, of course, by being smaller than D) "will be less than the proportionate increase of D." (And this, tho' it may be true, is, I believe going further than you intend.) 3

Galley 20. End of ch. 4. [8] D should surely read to be in conformity with the notation in Ch. 3. (II.) 4

Furthermore, while I agree that you can treat the matter perfectly well by relating supply price directly to N rather than output, I find this explanation very difficult to understand. What is the good of talking about it unless you can evaluate it in principle? But if you do, you introduce units of output. 5

Actually I cannot follow your last sentence--this may be mere stupidity--but I do suggest it is rather difficult and un-necessary. What I cannot directly intuit [d] is that is the ordinary Supply Curve.

Galley 25. Last para. 1st line. [9] 6 Ought not the word "consumable" to come in before "goods" [e] ? You remember that some newly finished goods are capital equipment. (cf. def. Galley 23. 2nd. para. [10] ) These are surely included in C.

Galley 26. 2nd para. [11] The words in brackets "hereafter defined as prime costs". Cant you drop these and leave prime costs till you come to define them? There is the grammatical difficulty that it is not perfectly clear that they refer, as I take it they must, to "the current outgoings of entrepreneurs" and not to the "balance of the current outgoings of entrepreneurs". And there is the further point of substance that while the outgoings of entrepreneurs in the aggregate may (but see below) [f] be regarded as equivalent to prime costs, they are certainly not necessarily the prime costs of the goods which the particular entrepreneurs produce. And this bracket confuses by suggesting that you wish them to be. 7

Moreover "current outgoings of entrepreneurs" is not equal to but probably greater than prime costs as defined in II below, [12] for some of those outgoings are in fact the profits of other entrepreneurs.

I am awfully sorry that you have had to define prime costs as not including user costs (user cost v[ery]. interesting!), because your definition cuts across the established usage as I understand it of the recent voluminous literature on imperfect competition. But I suppose it is inevitable! I should have liked you to use for your prime costs, some such expression as "prime factor costs" . 8

Galley 29. [13] I dont care for your last sentence in footnote 2, because it neglects the analysis of imperfect competition so laboriously achieved by many (including your Joan Robinson!). Quasi-rent is, to put it briefly, the excess of price over marginal 9 revenue (= marginal cost) ¥ no. of units produced. [14] The absurdity to which you refer just doesnt arise. (And in the exceptional case of perfect competition it is expected that marginal wage cost > av[erage]. wage cost.) Many of the younger generation have become keen on and familiar with these concepts and would regard your note as ignoratio elenchi! [15]

Galley 35. footnote. [16] Words [g] "in my sense" in brackets. It should be made plain that the reference is to the sense of this volume, as in the text to which the footnote refers you have been discussing the "Treatise" sense.

Galley 35. I, last para but 1. A little more than half way down. [17] "Without retaining anything in exchange". This suggests a gift and is confusing. I suggest instead "and reduces the total of his assets by that amount." 10

At the end of I I have the feeling that you ought to deal explicitly with how the creation and cancellation of credit by banks fits [h] in with this analysis, as this has been the subject of such keen debate and confusing misunderstanding. (Having read to end of Bk II I still feel this.)

Galley 39. 1st sentence of 1st new para. [18] Meaning [i] plain but wording needs revision? for "every net investment" the "aggregate of net investment"?

Line 17 of this paragraph. [19] Ought not the words "+ prime" to go in after "user" [j] ? On the other hand from footnote 1 [20] I gather that in speaking of gross incomes of firms you have already deducted prime costs. This seems odd. Have you done this consistently? 11

Last word of this para "saving". [21] This refers to ? Ought you not to add "by private individuals"?

  1. 1. The variorum of the proofs of the General Theory of Employment, Interest and Money are printed in volume XIV of Keynes's Collected Writings (1971-89). Page numbers of The General Theory (1936) refer to volume 7 of Keynes's Collected Writings.

    2. Collected Writings, vol. XIV, pp. 353-54; General Theory, p. 6.

    3. "In addition, there is, of course, also the so-called unemployment explained by the refusal or inability of a unit of labour, as a result of legislation or social practices or of combination for collective bargaining or of slow response to change or of mere human obstinacy, to accept a reward not greater in value than the product attributed to its marginal productivity. Such labour, however, as is unemployed because it is not on offer at the current wage, cannot be regarded as part of `involuntary' unemployment." (Galleys 3-4, in JMK GTE/6/3/1-2).

    4. Harrod's suggested changes are incorporated in the final version of the General Theory, p. 6. See Keynes's letter 463 of 9 August, [jump to page] .

    5. Collected Writings, vol. XIV, p. 357; General Theory, p. 9, second full paragraph, in particular lines 21-22. Harrod's suggestion was accepted.

    6. Collected Writings, vol. XIV, p. 365; General Theory, p. 14, first full paragraph. Keynes rephrased the paragraph along the lines suggested by Harrod.

    7. Collected Writings, vol. XIV, p. 375 (first and third paragraphs); General Theory, p. 29 bottom. The passage has been rewritten.

    8. Collected Writings, vol. XIV, pp. 390-91; General Theory, pp. 44-45. The symbol

    , which originally indicated the supply price (Collected Writings, p. 375), was not used in the final version.

    9. Collected Writings, vol. XIV, p. 401, first line of the second paragraph.

    10. Collected Writings, vol. XIV, p. 395, first full paragraph.

    11. Collected Writings, vol. XIV, p. 402 middle.

    12. Collected Writings, vol. XIV, p. 404.

    13. Collected Writings, vol. XIV, p. 412 note 1.

    14. For Harrod's treatment of quasi-rent see "Notes on Supply" (Harrod 1930:3 ), in particular p. 234.

    15. This comment was sent back by Keynes on 9 August 1935 (see letter 463 , [jump to page] ); Harrod further commented on 12 August (letter 464 , [jump to page] ).

    16. Collected Writings, vol. XIV, pp. 427-28, footnote 3. The comment on Robertson was rephrased and brought into the text on pp. 78-79 of the General Theory.

    17. Collected Writings, vol. XIV, p. 428. The passage was rephrased: see General Theory, pp. 81-82.

    18. Collected Writings, vol. XIV, pp. 437, line 14.

    19. Collected Writings, vol. XIV, p. 437, last line of text.

    20. Collected Writings, vol. XIV, p. 438, footnote 3.

    21. Collected Writings, vol. XIV, p. 438.

    1. a. ALS, one page, in JMK GTE/1/283. The accompanying notes, six pages on six leaves, with pencilled comments and marks in Keynes's hand, are in JMK GTE/1/284-289. The letter and the notes are printed in Keynes, Collected Writings, vol. XIII (1973), pp. 527-30.

      b. Ms: «ambiguous».

      c. Ms: in both occasions Harrod misquoted «D» for «


    1. d. Ms: «intuite».

      e. Inverted commas added for clarity.

      f. Ms: the remark here reproduced within brackets was added as an afterthought.

      g. Ms: «words».

      h. Ms: «fit».

      i. Ms: «meaning».

      j. Inverted commas added for clarity.

1. Keynes ticked the entire page (comments to Galley 3).

2. Keynes ticked the entire page (comments to Galleys 5 and 7)

3. "Rewritten" [Keynes's note in the margin].

4. "Yes" [Keynes's note in the margin].

5. "No not of aggregate output" [Keynes's note in the margin].

6. "Rewritten" [Keynes's note in the margin].

7. "Deleted" [Keynes's note in the margin].

8. This passage is ticked in the margin by Keynes.

9. Keynes circled the word "marginal", and commented in the margin: "I mean by quasi-rent the excess of price over average cost ¥ no. of units".

10. This passage is ticked in the margin by Keynes.

11. "Deleted" [Keynes's note in the margin].

Welcome page

top of page

Return to index of this section

Go to previous page

Go to next page