679. E. F. M. Durbin to Harrod , 15 June [1937] [a]

[Follows on from 670 , answered by 689 ]

33 York Terrace, [London] N.W.1.

15 June [1937] [b]

Dear Harrod,

I was immensely interested by our talk--and greatly enjoyed the evening altogether. [1] I think I see now, more clearly what my problem is. It is something like this.

You say that the recession will set in because of a decline in the demand for capital while I think it sets in because the Banks for various reasons are forced to interrupt a general inflation that would go on into a high inflation if they (the Banks) remained entirely acquiescent to the demands of industry.

Now I thought you argued that the decline in the demand for capital was due to a fall in the fraction of the Consumer's Income spent upon consumption goods--a fall due to the rise in the Rate of Saving expressed as a fraction of net income. My difficulty then was that the Rate of Saving so measured falls throughout the depression and rises from the beginning of the recovery (a point I think to be established by the empirical evidence).

But I now understand that what you are concerned about, is not the expenditure on consumption goods treated as a fraction of net income, but the rate of growth of the physical output of consumption goods. A curve expressing the increase in their output must turn over from the high rate of growth set up early in the boom to the secular rate of growth appearing when full employment is reached: [fig. 1] And to this I reply that I am troubled over basing a whole theory of the crisis upon the assumption that at the end of a period of three years depression the percentage [c] of human unemployment is very much greater than the percentage of "unemployed" machines.

Then I understand you to say--what then is your explanation of the fact that we both regard as established--that the percentage fluctuations in the output of capital are much greater than those of consumption goods?

I accept the acceleration principle--so that my only problem is to explain why there should be a change in the rate of growth in the demand for, <and> output of, consumption goods.

I think my reply is something of this kind

1. During the early stage of the recovery (1933-35 say) an un-maintainable rate of growth of consumption good production is set up by the co-incidence of the re-absorption of unemployed labour with the secular rate of growth in productive power.

There is also a recovery in the output of capital because of the increase in the number of firms able and willing to renew their depreciating capital.

But the unmaintainable [d] part of this rate of growth does not give rise to the accelerated demand for capital because there is a sufficient supply of unemployed machines to cater for it.

2. We now come to the period in which the rate of growth of physical consumption and the derived demand for capital is at the secular rate of growth. By now the credit expansion is well under weigh and net money income is rising fast (I suggest cumulatively) and the expenditure on consumption goods is rising (also cumulatively) but not so rapidly as total net income. After a time prices will rise as expenditure is rising more rapidly than the output of consumption goods. [fig. 2] The later period of the recovery (1936-?) has been reached.

3. I do not see that I am then involved in any great difficulty. After a time it is decided to check the inflation. The Banks take action. The rate of growth of net income is checked. Profit expectations are not fulfilled and an absolute downward movement in expenditure begins. The acceleration principle will then operate. The demand for consumption goods will decline. The growth of their physical output will be checked and the demand for new capital will be disproportionately checked. The multiplier and all that goes with it then secure that the contraction will be cumulative and severe.

But it will nevertheless follow that the crisis was due to Bank action in the sense that if action had not been taken the cumulative growth in expenditure, the output of consumption goods and the derived demand for capital, would have continued indefinitely with rising prices. There would have been no independent contraction in the demand for capital from industry.

This means that the acceleration principle can only be used to explain the relative movements of capital and consumption good production in the Cycle and not as an explanation of the "inevitable" crisis. That is my contention.

I only set this forth to see whether you think my views are a tenable logical hypothesis. Whether your hypothesis or my hypothesis--supposing both to be logically consistent--is the right one is simply a question of empirical evidence. And I wholly agree with you that the experimentum crucis would be the occurrence of a depression without (a) bank action (b) an external major shock to confidence. I think there is some evidence on my side but it would take too long to set it all down here. I should like to know what you think of the logic of the thing before doing so.

Again may I thank you for your book [2] which stimulated me enormously and for your hospitality which I greatly enjoyed


Evan F M Durbin.

  1. 1. Durbin refers to a meeting which must have taken place on 12 or 13 June, as he suggested in his previous letter: see [jump to page] .

    2. Harrod, The Trade Cycle ( 1936:8 ).

    1. a. ALS, four pages on four leaves, in EDP 3/5(12-15).

      b. Year not supplied.

      c. Ms: «%age» (also in the next two occurrences, in this paragraph and the following).

      d. Ms: «unmaintable».

Welcome page

top of page

Return to index of this section

Go to previous page

Go to next page