416. R. F. Kahn to Harrod , 8 January 1935 [a]

[Follows on from 410 , the exchange continues at 418 ]

King's College, Cambridge #

8 January 1935

My dear Roy,

I enormously enjoyed reading your Rejoinder. [1] I think it is frightfully important that you should publish it whether there is anything to rejoin to or not. The ideas that it contains should most certainly be brought before the public eye and I think your note does exactly what is wanted in a very effectual manner.

I enclose a few observations, referring to the points marked by the Roman numerals which I have pencilled in the margin. [2] But they are almost entirely suggestions for addition or amplification, and I do not want to suggest that there is any necessity to take them seriously.

I am afraid that I do not know anything about a section of the Marshall Society composed mainly of B.A's. I would suggest that you write to the Secretary and find out exactly what it is. But in any case the subject which you suggest is an excellent one. Everybody in Cambridge is always very anxious to hear other people's views on this topic. [3] The fact that you may overlap with J.M.K. will only make it the more interesting.

Yours ever,


I. It might be worth adding that on this definition an individual's saving is equal to the addition to his wealth, which is what everybody outside lunatic asylums [b] means by saving.

II. It might be well to point out that saving is the excess of income over expenditure and that income is the value of output. Nobody but a lunatic would regard the value of output as enhanced by the fact that expansion takes place by means of borrowing rather than out of accumulated profits.

III. I cannot quite see this but it is all such nonsense anyhow that it does not really matter.

IV. Do they include borrowing from the Banks in income?

V. It is to be noticed that previous to Robertson nobody, so far as I know, had ever devised this definition of saving. [4] Many people, including Haberler, have talked a great deal of gibberish about the difference between saving and investment and it is perfectly clear from the context that they meant neither meanings II, III or IV.

VI. One way to keep investment equal to saving in this sense is to keep total income constant. But if income is not constant it is still possible to get investment equal to saving. [5] Do they suggest that this is what the banking system ought to try to do if income refuses to remain constant? If so, the nature of the most desirable policy depends on the length of the "period" which is arbitrarily chosen. [6]

  1. 1. Harrod, "Rejoinder to Drs. Haberler and Bode" ( 1935:1 ).

    2. The draft bearing Kahn's numbering was not found among Harrod's papers. His notes, however, refer to the four definitions of saving examined by Harrod on pp. 82-84.

    3. No records survive of a paper by Harrod before the Marshall Society (Marshall Society, minute books, Marshall Library of Economics, Cambridge, Marsoc 1 and Marsoc 2). This, however, was mentioned again by Kahn on 6 April: see letter 446 , [jump to page] .

    4. Harrod did not provide explicit reference to Robertson's definition.

    5. Kahn soon realized that this statement was wrong: see letter 418 .

    6. After having noticed that Haberler's and Bode's notion of saving as "the difference between expenditure on consumption goods in a particular period and the income earned in the previous period" (this is, of course, the definition propounded by Robertson: see letter 436 , [jump to page] ) may give rise to the paradoxical result that refraining from spending may not be counted as saving if income is falling, Harrod remarked--possibly on Kahn's advice--that "It may be urged that the paradox is mitigated if the `periods' are sufficiently reduced in length. But to render an absurdity less conspicuous because smaller is not to alter its nature" (Harrod 1935:1 , pp. 83-84).

    1. a. TLI, one page, with annexed one page of observations, typed, both with autograph corrections; in HP IV-586-668b.

      b. Ts: «asylum».

Welcome page

top of page

Return to index of this section

Go to previous page

Go to next page