470. J. M. Keynes to Harrod , 27 August 1935 [a]

[Replies to 459 , 460 , 462 , 464 , 468 , and 469 , answered by 471 ]

Tilton, Firle, Sussex. #

27 August 1935

My dear Roy,

Here is a copy of the remaining chapter 25 which you have not already had.

I am still too much engrossed in re-writing the earlier chapters to have given as close an attention as I wish to all you have written me about chapters 14, 15 and 16. But I have now read what you say carefully enough to feel fairly convinced as to the broad position.

It is, I am sure, a big question of substance, not of manners or controversial fairness. If you are right as to what I have established, it is certain that you are also right that a good many of my criticisms need reconsideration. But I am still of the opinion that if my constructive sections are correct, my critical sections are more than justified. And I think that your acceptance of my constructive parts can only be partial if you do not accept my critical sections.

I will, of course, go carefully through your detailed points and be careful not to make criticisms of which I do not feel doubly sure. But the general effect of your reaction, apart from making me realise that I must re-write all this drastically if I am to make myself clear, is to make me feel that my assault on the classical school ought to be intensified rather than abated. My motive is, of course, not in order to get read. But it may be needed in order to get understood. I am frightfully afraid of the tendency, of which I see some signs in you, to appear to accept my constructive part and to find some accommodation between this and deeply cherished views which would in fact only be possible if my constructive part has been partially misunderstood. That is to say, I expect a great deal of what I write to be water off a duck's back. I am certain that it will be water off a duck's back unless I am sufficiently strong in my criticism to force the classicals to make rejoinders. I want, so to speak, to raise a dust; because it is only out of the controversy that will arise that what I am saying will get understood. Take your own case (on the assumption that I am right). If I had left out all the parts you object to about the classical school, you would have simply told me that you were largely in sympathy and liked it. But my attack on the classical school has brought to a head the fact that I have only half shifted you away from it. Your preoccupation with the old beliefs--and much more so in the case of most other people--would prevent you from seeing the half of what I am saying unless I moved to the attack.

But all this, of course, brings us back to the point as to whether I am right and in particular as to what it is which I accuse you of not having seen.

Let me take one or two passages from your script which I had better quote since I think you have no copy. You say, "I do not think Marshall failed to appreciate the distinctness of the concepts of marginal efficiency of capital and the rate of interest. He knew that there was a schedule of marginal efficiencies and he thought that interest was determined by this and the schedule of propensity to save. Now you use both these schedules, but to determine the level of income, the relevance of whose variations in this nexus he did not appreciate." [1] To deal with a minor point first, I think this is overstating a great deal the clearness of Marshall's thought, since it overlooks the fact that my definition of marginal efficiency of capital is quite different from anything to be found in his work or in that of any other classical economist (except for a passage which he makes little subsequent use of in Irving Fisher's latest book [2] ). I emphasise this, because the discovery of the definition of marginal efficiency of capital looks very slight and scarcely more than formal, yet in my own progress of thought it was absolutely vital. [3] Let us, however, assume that you are rightly interpreting Marshall. What the above passage then comes to is what you emphasise in numerous other places, namely, that my discovery consists in showing that the classical theory forgets the part played by changes in the level of income, but that this omission can be repaired, and we are then left with the rate of interest determining at what point, so to speak, the schedule of the propensity to save cuts the schedule of marginal efficiencies. This appears to be a view to which I have converted you and which you think to be my view. If this was true, then the modification called for in the classical theory would be much milder than that which I appear to demand. But the fact is that the above is not my view as to what happens--far from it, and it is to that substantial point that we have to direct our minds.

I shall try to re-write the chapters in question so as to make it more clear what it is I am saying. I can scarcely hope to achieve this in a letter. But I must obviously give a few indications of what it is I am driving at.

In the first place the statement that the classical theory would hold if incomes were constant needs a good deal of elaboration before it is intelligible. Obviously by income you cannot mean money income. If, on the other hand, you were to mean employment being constant you would be more on the right track, but still the conditions would not be sufficient.

The main point is, however, that my theory is essentially not a theory that the rate of interest is the factor which, allowing for changes in the level of income, brings the propensity to save into equilibrium with the inducement to invest. My theory is that the rate of interest is the price which brings the demand for liquidity into equilibrium with the amount of liquidity available. It has nothing whatever to do with saving. You say in one passage that I am doing great violence to the fundamental ground-work of thought in getting rid of supply and demand analysis. [4] But I am doing no such thing. I am substituting demand and supply analysis for liquidity instead of that for savings. Marshall you say "thought interest was determined by the schedule of marginal efficiencies and the schedule of the propensity to save"; [5] he forgot that incomes could change. I am saying something totally different from this when I say that interest is determined by the demand and supply for liquidity. You imply in passage after passage that, if the schedule of marginal efficiencies is known and if the propensity to save is known, the rate of interest can be deduced. But my whole point is that it cannot be deduced from these factors and has to be found from a different source. This is where your ingenious analogy of the private German lessons fails. [6] You say quite correctly that the price of the lesson equates the propensity to give German lessons with the marginal efficiency of German lessons. In the case you have chosen price does play that function. But suppose that whenever the price of a German lesson went down the demand schedule also shifted its position, the whole thing would have no meaning. You get your point of equilibrium because the demand and supply schedules for German lessons do not shift their position whenever the price changes. If for any price whatever the demand and supply schedules shifted so that they intersected at the point in question, clearly they would not be determined by the price. Now it is the characteristic of the demand and supply schedules for savings that they have this unfortunate characteristic. In truth there are no such things as these schedules. They are completely bogus. Without bringing in liquidity preference the position of equilibrium is entirely indeterminate, and any method , such as the classical one, which endeavours to arrive at the rate of interest without bringing in liquidity preference is bound to be circular in the worst possible sense of the word.

I fancy that the misunderstanding partly comes from the fact that, although you have lately accepted the view that saving and investment are equal, you still think of them as being different things; whereas I regard them as being merely different names for the same phenomenon looked at from different points of view. Saving is a name given to a certain quantity looked at as the excess of income over consumption. Investment is the name given to the same quantity regarded as the constituent of income other than consumption. The demand for German lessons is quite a different thing from the supply of German lessons even though the two may be equal. But saving and investment are the same thing. It is vital for the application of the supply and demand analysis with interest as the equilibrating factor that interest should cause saving to increase and investment to diminish, as the classicals all believe. But if the rate of interest operates on the amount of saving in the same direction as on the amount of investment, if, that is to say, a rise in the rate of interest necessarily diminishes saving, then the independence of the two things necessary to allow interest to be an equilibrating factor is non-existent.

I can scarcely expect you to be much impressed by all this rigmarole, and you must wait for my revised version where I shall try to be more convincing. But my own firm conviction is that your mind is still half in the classical world, and that you ought to be accusing me, not of bad manners, but of faulty theory.

What you say about the last chapter but one is somewhat connected with this. It is certainly not my object in this chapter unduly to depreciate the classical school, and I will see if I can put in a passage to make that clear. What I want is to do justice to schools of thought which the classicals have treated as imbecile for the last hundred years and, above all, to show that I am not really being so great an innovator, except as against the classical school, but have important predecessors and am returning to an age-long tradition of common sense. It would help me if you were to mark passages which you especially object to, or which you think could be curtailed. I should certainly like to reduce the space given to the mercantilists , but feel that I must give chapter and verse. I cannot but think, however, that you would feel rather differently about this chapter if I were able to convince you that the classical theory of the rate of interest has to be discarded in toto, and is incapable of rehabilitation in any shape or form.

Yours ever,

J M Keynes

P.S. Will you ponder the following propositions:

1. Saving and investment are merely alternative names for the difference between income and consumption.

2. The supply curve of savings and the demand curve for investments have no determinate point of intersection, since they lie along one another in all conditions throughout the whole of their length. This applies equally in equilibrium or in dis-equilibrium. There are no conceivable circumstances in which the one curve does not occupy the same situation as the other throughout its length;--provided, of course, the same conditions, whatever conditions are assumed, are taken as applying to both alike.

3. The propensity to save and the schedule of marginal efficiencies are two curves which do not intersect anywhere, because they are not in pari materia and do not relate to the same variables. The propensity to save is a curve which relates the amount of investment to the amount of income. The schedule of the marginal efficiencies of capital is a curve which relates the amount of investment to the rate of interest. There is no sense in which they can be said to intersect.

What then remains? What are the demand and supply (other than those for liquidity) which the rate of interest is in certain hypothetical positions supposed to equilibrate?


I have rewritten effective demand, income, and user cost radically.

R. F. Harrod Esq., Christ Church, Oxford.

  1. 1. Letter 462 , [jump to page] .

    2. Keynes refers to I. Fisher's notion of the "rate of return over cost," in The Theory of Interest, New York: Macmillan, 1930, pp. 155-61.

    3. On the place of the discovery of the marginal efficiency of capital in the progress of Keynes's thought see letter 581 , [jump to page] .

    4. Letter 460 , [jump to page] .

    5. Letter 462 , [jump to page] .

    6. Letter 459 , [jump to page] , and letter 464 , [jump to page] .

    1. a. TLS with autograph corrections and additions, 10 pages on 10 leaves, in HP II-55. CcI, also corrected in Keynes's hand, in JMK GTE/1/320-29. Printed in Keynes, Collected Writings, vol. XIII, pp. 547-53. Reproduced by kind permission of the Provost and Scholars, King's College, Cambridge.

Welcome page

top of page

Return to index of this section

Go to previous page

Go to next page