384. G. Haberler to Harrod , 25 October 1934 [a]
[Replies to 378 and 381 , answered by 388 and 390 ]
The League of Nations, Geneva #
25 October 1934
My dear Harrod,
Many thanks for your two letters, of October 19th and 21st. Your comments on my memorandum  are exceedingly valuable, and I should be very grateful if you would let me have your further reflections [b] .
Your suggestions as to the future procedure of my enquiry are quite in harmony with my own plans, and I am glad to have my ideas confirmed by you. 
Concerning the first point of criticism in your letter,  I admit that the exact relationship between the principle of accelerated demand on the one hand and changes in the degree of "roundaboutness" of production on the other must be made clearer than I was able to do in my memorandum.  May I sketch my present position: Suppose there are unemployed resources and you get an inflationary increase in consumers' demand. If there is an elastic credit supply and the interest rate does not rise, this stimulus will be transmitted to the higher stages, workers will be employed there, and the production of producers' goods will rise faster than the production of consumers' goods. The new product will be produced in a roundabout way, but it is conceivable that the roundaboutness of production for the economy as a whole will not be changed, because the new production can be equipped, in all its stages, with unemployed resources.
Take now another extreme. Suppose there are no unused resources. Demand for some kinds of consumers' goods rises; [c] this stimulates the preceding stages to borrow from the banks. The flow of the new money directed to buy capital goods rises more than the initial flow of money which buys consumers' goods. There are no unused factors of production. Therefore factors must be hired away from other branches, consumption industries as well as capital goods industries, and the roundaboutness of production of the economy as a whole increases. This reasoning suggests that, except in some limiting and exceptional cases, the working of the acceleration principle leads to increased roundaboutness of production. I hope that I have made my point fairly clear.
Also in the second point you mention, I agree that I have to complete my memorandum. I honestly believe that Keynes' Treatise contains very valuable and new ideas. The difficulties I experience with his explanation of the trade cycle are connected with the theoretical groundwork (his "fundamental equations"  ), although I am convinced that what he says on the explanation of the business cycle is to a large extent independent of his theoretical framework. It would lead me too far to explain in this letter my objections against his equations. I hope I shall have an opportunity to come back to this point on another occasion.
As to the problem in the second para. on p. 3 of your letter,  of which I enclose a copy for your convenience, I should like to say this: The funds come from the increase in V. There is an inflationary increase in the flow of money. It makes no difference whether the flow of money increases [d] because M has risen or because V has gone up. The funds do not come from voluntary savings (although the increase in V leads to forced saving). Therefore the market rate has been lowered below the natural rate. This problem can, as you see, be tackled without entering Keynesian territory.
This brings me to your letter of October 21st. I think I will let my letter to "The Economist" stand as it is.  Your proposed answer (of which I enclose a typed copy) does not convince me at all. Frankly, it strikes me as if it were written in a strange language which I cannot understand. Some economic friends of mine read it, but they could not enlighten me; they felt the same way as I did. I should like to suggest that you ask Marschak for his opinion. He is a great authority on velocity of circulation, etc. 
Specifically, I should like to answer this: (1) My definition of saving is the one you give in your letter to "The Economist", viz., "income not spent on consumption". As an appreciation of the total stocks of money in terms of goods is not a part of the income, this appreciation does not absorb savings. If you have other definitions of "savings" and "income", it falls on you to make it clear; for a definition of income, which counts an appreciation of the money stocks as income, would be a very unusual one. 
(2) Your statement that my definition is not fruitful, for the reason that "the dictum referred to" (I take this to mean the statement that savings are equal to investments) "becomes entirely sterile, because it is satisfied whatever banking policy is adopted",  is, in my opinion, false. For, if banks create credit, investments become greater than savings (the newly created money is invested without having been saved), and, if banks contract credit, investments fall short of savings. I am not aware of any obscurity in this.
Would you mind my sending a copy of this letter and of your letter to Robertson? It would be worth while to have his opinion. 
With my best regards,
R.F. Harrod, Esq., Christ Church, Oxford, England. [e]
2. Haberler's memorandum was to constitute the first part of the study. The second part was A Synthetic Exposition Relating to the Nature and Causes of Business Cycles. Together they were published as Prosperity and Depression. A Theoretical Analysis of Cyclical Movements (1937).
3. Letter 378 , [jump to page] .
4. On Haberler's original discussion of the "principle of Accelerated Demand" see note 4 to letter 378 .
5. J. M. Keynes, A Treatise on Money (1930), in Collected Writings, vol. V, chapter 10.
6. Letter 378 , [jump to page] , paragraph beginning "May I put a poser ..." (pagination is referred to the typed transcription).
7. The letter, however, was considerably shortened for lack of space: see note 1 to letter 381 .
8. Haberler probably refers to an article Marschak had recently published on wealth and individual cash balances ("Volksvermögen und Kassenbedarf", Archiv für Sozialwissenschaft und Sozialpolitik, BD 68 H. 4, January 1933, pp. 385-419), and to "Vom Grössensystem der Geldwirtschaft", Archiv für Sozialwissenschaft und Sozialpolitik, BD 69, 1933, pp. 492-503, which are cited in this connection in Haberler's Prosperity and Depression (1937), p. 57-58.
9. The fact that Harrod's statement implied an unusual definition of savings was also pointed out by Kahn (letter 382 , [jump to page] ) and admitted by Harrod (letters 383 , [jump to page] , and 388 , [jump to page] ). This eventually led Harrod to modify the concluding paragraph of his second letter to The Economist ("Banking Policy and Stable Prices (2)", 1934:10 , press item 9 ) by acknowledging that he was referring to "a somewhat curious definition of savings" (see [jump to page] ).
10. Harrod, "Banking Policy and Stable Prices (2)" ( 1934:10 ), press item 9 , [jump to page] .
11. Harrod's Economist letters were the subject of an exchange between Haberler and Robertson. On 25 October Robertson thanked for a copy of Haberler's reply ("Banking Policy and Stable Prices", 1934, p. 8), which "exactly expresses [Robertson's] views", and sent to Haberler the typescript of his "Mr. Harrod and the Expansion of Credit" (1934) (Robertson to Haberler, 25 October 1934, CcTL transcription, in LoN 10B/12809/12653 and GH Box 66). In his reply, Haberler expresses the feeling that his "Banking Policy ..." "completes in one respect [Robertson's] article for the Economica". He suggests that Harrod's mistake consisted in mixing up
Haberler also sent a transcription of Harrod's proposed reply to Haberler's "Banking Policy ..." (attached to Harrod's letter 381 to Haberler of 21 October; the differences between Harrod's draft and the version eventually published in The Economist are listed in the editorial notes to press item 9 ) (Haberler to Robertson, 30 October 1934, CcTL, in LoN 10B/12809/12653).
On 1 November, Robertson expressed the following opinion as to Harrod's position:
It seems to me that they are missing something of fundamental importance and abandoning much of the ground that has been won in monetary theory of recent years. The practical inference apparently drawn is that all levels of interest-rate are `equilibrium' levels, but that among these equilibrium levels there is one of unique importance, viz., that which secures full employment.
[...] My present conclusion is that while my Cambridge colleagues are wrong in saying that every position is one of equilibrium, there are certain positions of what I call quasi-equilibrium, i.e. positions where the rate of interest is such that the existing level of prices, employment, etc. will be maintained, and there will be no further creation or destruction of money (Robertson to Haberler, 1 November 1934, CcTL transcription, in LoN 10B.12653.12653).
A few weeks later, Haberler summarized as follows the state of his debate with Harrod and Kaldor on Banking Policy and Stable Prices:
top of page
Return to index of this section
Go to previous page
Go to next page