606. D. H. Robertson to Harrod , 23 December 1936 [a]

[Replies to 603 , answered by 607 ]

Trinity [College, Cambridge]

23 December 1936

My dear Roy,

Many thanks for your letter, which reached me at Geneva, where I have been, at Loveday's request, keeping touch with Tinbergen's attempts to throw statistical light on these troubled waters. [1]

I'm so glad you didn't dislike my "Notes" as much as you feared. [2] (Such signs of irritation as I display in conversation and correspondence are largely due to the fact that I am not the research professor, seated in an ivory tower, of your article, [3] but a workaday supervisor and examiner, who finds himself obliged to read quantities of what seems to him great rubbish,--and also the chairman of a Faculty Board, who would fain preserve some order and progression in the studies of the young).

On the points in your letter (returned for reference [b] ).

(i) [4] I feel sure you have momentarily misunderstood JMK's apparatus. There is no doubt, I think, that even at the margin he regards DZ as = DZ + DP [c] . The extra profit which occurs as the result of expanding output under diminishing returns is regarded as, so to speak, hanging automatically on to the extra factor cost. Z is simply supply price multiplied by volume of output, and DZ is the increment in this,--the whole shaded area in the figure [fig. 1].

A nonsense apparatus you may say,--and so at first I was inclined to think it,--so, I think, it would be if each branch of industry were integrated. But since they are not, marginal costs to all (or most) producers consists partly of costs of materials, which include the profits of other producers. So I'm prepared for the sake of argument to accept the apparatus, without much liking it. If you dislike it, you must have it out with him, not me!

(ii) [5] This of course is a much more difficult matter. Experience seems to show that it is very difficult to make the assumptions precise, as to avoid over-stating the case in one direction or another.

Your argument (2) [6] is that already discussed, & put in what I think to be its proper place, in my §7. [7] Your argument (1) seems to me to imply partial but not complete pre-adaptation to changes in demand, and also that instantaneous operation of the "multiplier" at its full normal strength which I dispute. The real crux seems to me to lie in argument (3). I conceive of the extra supply of saving as coming up against two demand curves, one representing the productivity use, one the liquidity. For there be no increase in investment either must be completely inelastic or completely elastic or both [fig. 2]. It can be plausibly argued that both are likely to be the case in face of a large sudden increase of saving, occurring in a depression, and/or expected to be reversed. It does not seem to me plausible to suppose that this is a correct account of the operation of a gradual increase in the propensity to save in a normally prosperous community with a well organised capital market. That the elaboration of the capital market should actually operate against the effective utilisation of desire-to-save seems to me a paradox too great to be acceptable.

A propos of this whole topic, and of your book, [8] I feel that there is probably an important distinction to be drawn between forms of investment which are closely "geared" by the "principle of acceleration" to the demand prices of particular kinds of consumption goods, & those--especially power and transport in all their numerous forms--which are highly susceptible to invention on the one hand, and dependent on rather vague estimates of ultimate consumers' prosperity on the other. In general I sympathise with your restoration of the principle of acceleration to its proper place in the theory of fluctuation (but if you were handing out bouquets, why not one to Pigou for his extremely clear statement of the case in Ind. Fluc. 1929 ed. Part I, ch. IX [9] ? [d] ): but to elevate it into the position of a dragon which devours all long period thrift seems to me to do it too much honour. [10]

You will see from these latter ill-thought-out observations that I have been reading your book. It turns out that I promised (or so it is said) months ago to review it for the Toronto Journal. [11] So after Christmas I must settle down to do so. You must, I fear, expect a mixture of like and dislike: I hope in any case I shall continue to earn your good marks for tone & temper.

Meanwhile a happy Christmas.



I see I have omitted to comment on your (iii)--the non-intelligibility of p. 190. [12] It doesn't seem much use to do so without knowing more precisely what is obscure. My general point is that JMK seems to admit in the end the long-period plasticity of wages etc, whence it follows that in the long period a high liquidity preference (large value of Marshallian K) (1) will manifest itself not in unemployment but in low money incomes (2) will only manifest itself in high interest rates if the money is such that people can successfully waste their energies in increasing its amount,--the old Ricardian point about the wastefulness of a gold currency. Hence I am v[ery] sceptical about JMK's excursion into mediaeval history.--But I'm afraid I'm only repeating what I've [e] written, not explaining it!

  1. 1. After Haberler completed his survey and synthesis of trade cycle theories, the League of Nations appointed Tinbergen to devise a statistical test applicable to them (the resulting publications were Haberler's Prosperity and Depression, 1937, and Tinbergen's Statistical Testing of Business Cycle Theories, 1939). Loveday, the Director of the Financial Section and Economic Intelligence Service of the League of Nations, felt that such an enquiry "needs something more than a specialised mathematical work. It needs to be watched and directed by a really competent economist." He therefore suggested to approach Robertson as "probably the best qualified person" among the small number of distinguished economists who took part in the discussion of Haberler's work. Loveday suggested that Robertson should visit Geneva every quarter for two or three weeks (Loveday to the Secretary General (through the Treasurer) of the League of Nations, 16 September 1936, in LoN 10B/25630.12653).

    At a later stage Harrod was also invited to take part in the discussions of Tinbergen's statistical inquiry on the trade cycle: see letters 686 , 687 , and 772 .

    2. Robertson, "Some Notes on Mr. Keynes' General Theory of Employment" (1936).

    3. Harrod, "Mr. Keynes and Traditional Theory" ( 1937:4 ).

    4. Letter 603 , [jump to page] .

    5. Letter 603 , [jump to page] .

    6. Refers to the numbers Robertson pencilled in the margin of Harrod's letter (the precise location is indicated in the footnotes to letter 603 ).

    7. Robertson, "Some Notes on Mr. Keynes' General Theory of Employment" (1936), pp. 187-88.

    8. Harrod, The Trade Cycle ( 1936:8 ).

    9. A. C. Pigou, Industrial Fluctuations (1929). Chapter IX of part I on "Instrumental and Consumption goods" critically discusses the view that the cycles originates from fluctuations in the instrumental industries, as compared to the opposite thesis that the origin of industrial fluctuations lies in the actual and expected demand of consumption goods (pp. 107-8).

    10. This point was restated in Robertson's review of Harrod's book "The Trade Cycle, by R.F. Harrod" (1937), p. 126.

    11. Robertson, "The Trade Cycle, by R.F. Harrod" (1937), appeared in the Canadian Journal of Economics and Political Science, February.

    12. Letter 603 , [jump to page] .

    1. a. ALS, three pages on three leaves, numbered from the second, in HP IV-990-1069d/45.

      b. Harrod's letter 603 is preserved among Harrod's papers.

      c. Robertson's emphasis.

      d. Robertson seems to have added the question mark as an afterthought.

      e. Ms: «Ive».

Welcome page

top of page

Return to index of this section

Go to previous page

Go to next page