390. Harrod to G. Haberler , 1 November 1934 [a]

[Follows on from 388 , replies to 384 , answered by 395 ]

Christ Church, Oxford

1 November 1934

My dear Haberler

May I revert to some other points in your letter of October 25. I must explain that I do not wish to deal with these in a spirit of controversy; I do feel controversial about the point in my last letter, because I think that I have attained as absolute intellectual certitude about that point as I am ever likely to obtain about anything in economics. In what follows I merely throw out suggestions in the hope that they may be of some use. I shall not necessarily expect any answer, since I recognize that you cannot waste your time in correspondence, tho' of course I am always interested to hear your views. So much by way of preface.

1. I still think that the type of trade cycle theory discussed in your pp. 24-27 is analytically distinct from theories which turn on changes in roundaboutness. [1] The cycle might be characterized by more resources being devoted to higher stages in the boom and less in the depression, without there being any change in roundaboutness. I know that it is difficult to measure roundaboutness at a single point of time. But I suggest that there is no increase of roundaboutness compared with a former point of time, if the productive methods implied by the use of resources in the way they are actually being used at the point of time are not more roundabout than they were at the former point of time.

You first take the case (with increased consumers' demand) where there are unused resources and an elastic credit supply and concede that there would be no change of roundaboutness. [2] In so far as the credit supply was not perfectly elastic, interest rates would tend to rise, and the production process would pro tanto tend to be shortened, despite the fact that the amount of resources used at higher stages would tend to increase (as according to the arguments of your pp. 24-7).

Then you take the case of no unused resources. Strictly in this case there could be no increase of output and the hypothesis is illegitimate in the analysis of a boom. Taken less strictly you may mean that there is a potential increase in the supply of capital, and, the supply of labour being absolutely inexpansible, production might become more roundabout. But even this case is hardly legitimate, since by hypothesis the output of consumers' goods could not in the circumstances increase and it is the effects of such a postulated increase that we are analysing.

It is clearly not right to take a case in which there is no unused labour available initially for the analysis of a boom process. Therefore I dont think your second case can be taken as the more general.

I think the general case is this. Assume an increase in the demand for consumers' goods, when there is some unused labour. In accordance with the princ[iple]s. of pp. 24-7 the demand for factors at the higher stages will be initially disproportionately great, the initial demand for the waiting factor will be disproportionately great and the price of waiting is more likely to rise or likely to rise more than the price of the other factors. This would tend to make the production process actually become less roundabout at the time that more resources were employed at the higher stages.

2. I dont think you quite understood my problem connected with an increase in MV. [3] Of course so far as the price level is concerned it makes no difference whether the increase is in M or V. But this is hardly true when we are considering an increase of investment. To meet this, funds can only come from (i) the savings (ordinary sense) of the public and (ii) additional bank loans. But if M is constant there is nothing coming from (ii). \ the whole of investment must come from (i), i.e. from actual savings.

It does not help to talk about saving "enforced" through a rise of prices (due to rise in V). How does this get into the hands of entrepreneurs? The position is quite clear when the enforced saving is due to the issue of extra loans. These extra loans increase the number of units of purchasing power, thus causing enforced saving and at the same time put an equivalent of purchasing power (= that lost by the community) into the hands of entrepreneurs to whom the loans are issued. The case is exactly parallel to a government paying its way by inflation. It illicitly taxes the people by issuing new purchasing power, which it provides itself with from the printing press. But if prices rise by an increase in V, does that give a government an increased purchasing power? Not in the least. Then why should you suppose it to give entrepreneurs increased purchasing power? No new loans are issued to them. But in that case investment is equal to saving.

And it is. Yet one has the feeling that in a boom due to an increase of V and \ of prices, there is in some sense over-investment. This is where the Keynesian argument comes in. He holds that there is over-investment (whether the banks are making new loans or not) if a large part of the funds for that investment come from windfall profits. The windfall profits are by nature impermanent, and the effect of their use to finance an increase of investment may be analogous to the use of funds provided by additional bank creations.

I do not know if all this fits into your way of thinking about the matter sufficiently to be of any use.

Yours very sincerely

Roy Harrod

  1. 1. Haberler, Systematic Analysis of Business Cycle Theories (1934). Refers to Haberler's discussion of the Principle of Accelerated Demand: see letter 378 , [jump to page] and note 4 , and letter 384 , [jump to page] .

    2. Haberler qualified his conclusion as to the working of the Acceleration of Derived Demand due to the Existence of Durable Means of Production with respect to the case of excess capacity:

    • In the case of goods which have been manufactured for a long time, a qualification must be made with regard to excess capacity [...]. If there is excess capacity, then the acceleration of derived demand will start only at a later point, namely, after the excess capacity has been absorbed. (This is reminiscent of Spiethoff's distinction between the first and the second phase of the upswing in the course of the typical business cycle. The first phase of the upswing he calls the period from the low-point to the point where all excess capacity is absorbed. The second phase is characterised by the fact that new plant and fixed capital equipment is needed. In practice, of course, it is difficult to draw a sharp line for business as a whole between the first and second phases of the upswing, owing to the different conditions existing in different branches of industry) (Haberler, Systematic Analysis, p. 26).

    Analogous considerations on excess capacity passed into the final version of Haberler's study: Prosperity and Depression (1937), p. 91.

    3. See the "poser" in letter 378 , [jump to page] , and Haberler's reply in letter 384 , [jump to page] .

    1. a. ALS, three pages on three leaves, in GH Box 66. TLCc transcription, in LoN 10B.12653.12653. Marked "Y1", "copy" and "Original kept by Mr Haberler, 2.XI.34". Further Cc in NK3/5/11-12.


Welcome page

top of page

Return to index of this section

Go to previous page

Go to next page