391. R. F. Kahn to Harrod , 1 November 1934 [a]

[Replies to 389 , answered by 392 ]

London so from King's College, Cambridge #

1 November 1934

My dear Roy,

I very much sympathise over this question of definitions. We have to suffer a good deal in Cambridge. But you refer to "standard terms and concepts". How many people who glibly use the "savings and investment" of the Treatise are at all conscious of the highly peculiar nature of its definition of saving? It is not merely a question of convenience, but also one of truth. "The difference between savings and investment" is universally interpreted as having a real meaning, in the sense of affording a causal interpretation of observed phenomena, and not merely as the difference between actual income and some arbitrarily chosen "normal" income, which is all that it really is. The fault may be partly that of Keynes (of the Treatise), but if so, Keynes was definitely guilty of an error of logic and not only of a badly chosen definition.

The two main arguments for the new definition of saving are:--

(a) that it is what is ordinarily meant by saving,

(b) that the truism "savings = investment" points to some important truths, e.g.

(i) investment is always self-financing: there need never be any question of "where the money comes from". (Cf. my "M r Meade's relation" of my article on "Home Investment and Unemployment" and Meade's own book. [1] "Savings equal investment" sums up all this stuff in three words.) [b]

(ii) All references to "forced saving", etc. are meaningless.

(iii) It makes no difference what part the banking system is playing in supplying credit or taking charge of hoarded funds.

(iv) The rate of interest cannot be determined by the "supply and demand of savings".

To elucidate them and similar doctrines in terms of the "savings" of the Treatise is an awful job. As a matter of fact, 99 out of every 100 readers of the Treatise are firmly convinced that "savings - investment = new hoarding - new bank credit".

As regards the trade cycle, the new definition is also very helpful. Employment is always such as to provide that income out of which people, with their given propensity to save, will in part save an amount equal to investment. A change in investment simply alters output by an amount determined by the "multiplier". The whole thing thus reduces to a study of the causes of changes in the rate of investment.

Meade's references to "neutral money" suggest slight infection. [2] But I think he has abstracted. After all, "M r Meade's relation" is the forebear of "savings = investment."

A recent letter to me from Robertson reveals that he fails to realise that the "difference between savings and investment" is merely a product of the queer definition of the Treatise and that it is something to do with the banks. If I succeed in convincing him that " ", I shall have radically altered his state of mind. That illustrates what I mean when I applaud the "simple-minded" definition of saving.



  1. 1. R. F. Kahn, "The Relation of Home Investment to Unemployment", Economic Journal XLI, June, pp. 173-98; "Mr. Meade's relation" is introduced on pp. 188-91. J. E. Meade, The Rate of Interest in a Progressive State (1933).

    2. Meade, The Rate of Interest in a Progressive State (1933), chapter 2.

    1. a. ALI, two pages on one leaf, in HP IV-586-668b.

      b. Ms: bracket not closed.

Welcome page

top of page

Return to index of this section

Go to previous page

Go to next page