229. R. F. Kahn to Harrod , 24 March 1932 [a]

[Replies to a letter not found; continues at 235 ]

King's College, Cambridge

24 March 1932

Dear Harrod,

Many thanks for your interesting letter. [1] The trouble appears to me to be that you are making that very same extreme assumption according to which I show in my footnote that imports and exports necessarily balance. [2] You are assuming, I think, that all disbursements are devoted to goods, so that a reduction on the part of any individual's disbursements necessarily causes an equal reduction in the disbursements of some other individual, except in so far as it is taken off imports. On this assumption, imports must necessarily decline automatically to the same extent as exports. But the assumption is an absurd one. In addition to spending money on consumption goods, people save money, and savings are normally devoted to buying securities and not commodities. A reduction in a man's receipts is only in the extreme case accompanied by an equal reduction in his expenditure on goods. Savings, in the simple-minded sense of the term, will diminish, and the reduction in imports falls short of the reduction in exports by the reduction in savings.

The assumptions on which your view would be correct are, I think, as follows:--

(a) That any increase in the dole is met entirely by taxation.

(b) That any increase in taxation falls entirely on consumption and not at all on savings.

(Conditions (a) and (b) are equivalent to assuming that there is no dole--the assumption mentioned in my footnote).

(c) That when a business man's receipts are diminished, he reduces his expenditure on consumption goods to an equal extent.

All these three conditions are of the most extreme possible kind, and each one of them will fail to be realised in practice. I would suggest, therefore, that imports and exports will balance automatically only in the absurd kind of case contemplated in my footnote.

You would rightly claim however, that it is still necessary to follow up the reactions of a reduction in disbursements on securities. The reason why there are no such reactions is explained by Mr. Meade's relation. [3] There is a diminution in foreign lending, and therefore a diminution of the supply of securities on the market exactly equal to the diminution of disbursements on securities.

This is only true, however, if there is no movement of gold, so that the diminution in foreign lending is exactly equal to the difference between the diminution in exports and the diminution in imports. And that brings me to your very interesting tautology about the relation between a movement of gold and the quantity of money. I call it a tautology because you called it a tautology, but I venture to suggest that, having called it a tautology, you then proceeded to imbue it with causal significance to which such tautologies are not entitled. You showed quite rightly, I think, that if there is no change in bank loans, an export of gold must be associated with an equal reduction in the quantity of money. But you then proceed to deduce that an export of gold could only be brought about as a result of people wanting to reduce their holdings of money faster than the banks were willing to reduce their loans. This last step seems to me to involve a logical error. It may very well be that the same cause that brings about a movement of gold quite independently brings about a reduction in the quantity of money.

How this comes about can easily be demonstrated. Let us suppose that there is a certain diminution in the foreign balance. 1 This is accompanied by an equivalent reduction in the volume of savings, (i.e., disbursements that are not devoted to commodities). If gold is flowing abroad, the flow of securities on to the market is not diminished to the full extent suggested in the last paragraph but one. And the difference has to be made up by somebody or other reducing their holdings of money. The simplest way to look at it is that business men meet part of their losses out of their bank balances. But if they insist in meeting their losses entirely by the sale of securities, they force other people to substitute securities for money at a rate equal to the rate at which gold is leaving the country. All this of course is on the assumption that the total amount of bank assets remains unaltered--a very peculiar assumption.

I am afraid that this must sound awful tosh. As you will realise, I have not got very much further than the stage of thinking aloud. But you will also realise what an inspiring effect your paper has had on at any rate one of your audience, and I know others besides myself are looking forward to a more convenient opportunity of digesting your views.

Yours sincerely,

R. F. Kahn.

P.S. I meant to add a word on the question of small local areas. Of course, the same principle must apply. The important differences are

(a) The very high elasticity of demand for <capitals>

(b) The automatic ease with which foreign lending adjusts itself to the foreign balance owing to the fact that the area forms part of a larger banking system.

R. F. Harrod Esq.,

  1. 1. This letter comments on a paper (not found) on the "Theory of Balance of Foreign Payments", read by Harrod before the Marshall Society on 10 March 1932, which seems to represent Harrod's first written attempt to apply Kahn's employment multiplier to international trade. This material was to become chapter VI of Harrod's International Economics ( 1933:10 ), a draft of which was sent to Robertson a few days later (see letter 231 ). Harrod's address is summarized as follows in the minutes of the Marshall Society (Marshall Library of Economics, Cambridge, Marsoc 1, page 188):
    • M r Harrod explained the way in which his theory of the balance of foreign payments differed from the orthodox theory.

      He compared international trade to the local trade between districts within a country and showed that a permanent tendency towards a passive trade-balance might exist which would not necessarily tend to be offset by long-term capital movements until the necessary changes in relative price-levels had been brought about.

      He made much use of formulae as a cleaner way of presenting his arguments, which were nevertheless a trifle obscure.

      In the ensuing discussion the details of M r Harrod's theory were much criticised, but his main arguments seem to have been accepted.

    2. R. F. Kahn, "The Relation of Home Investment to Unemployment", Economic Journal XLI, June, pp. 173-98. The footnote referred to is on p. 190: it examines, by means of "Mr. Meade's relation", the effects on a country's exports of a reduction of imports due to tariffs.

    3. Kahn, "The Relation of Home Investment to Unemployment", p. 188.

    1. a. TLS with autograph additions, three pages on three leaves, in HP IV-586-668b.


1. Balance of trade, on income account [Kahn's footnote].


Welcome page

top of page

Return to index of this section

Go to previous page

Go to next page