633. J. M. Keynes to Harrod , 18 February 1937 [a]

[Replies to 632 , answered by 634 ]

46, Gordon Square, Bloomsbury #

18 February 1937

Dear Roy,

Your conundrum is a very interesting one. But I am not quite sure how far it is a matter of words.

The phenomenon I am trying to describe has certainly absorbed savings. But it may be that I have misdescribed it.

The facts are as follows:--

For the period I am thinking of the value of capital equipment has increased from 100 to 265; the number of workers from 100 to 150; and the value of output from 100 to 240. [1] Now the increase in capital from 100 to 265 has certainly required that amount of saving, since the community is that much richer. What I said the other night was to the effect that the value of capital has increased only a little more than the value of output. My point was that the increase in capital from 100 to 240 was required to keep pace with output, and only the small excess of that was due to periods of production being longer in the sense that more capital is employed to produce a unit of output.

It would not be far from the facts to assume that the rate of interest has been the same throughout. At any rate it may facilitate our debate to assume that.

My argument is that, assuming a steady rate of interest, the nature of inventions is such as to require a greater amount of capital per worker, but that the increase is not much more than in proportion to the output. If the amount of capital required for this purpose falls short of our normal savings, then we have to try and stimulate the use of capital by a substantial reduction in the rate of interest.

Have I made the wrong distinction, or what? Would you relate the facts that I am trying to describe in your own language according to your own schematism?

It occurs to me as possible that there is a confusion between the period of transition and what happens after the transition is effected.

I am also not sure quite what my units of measurement are. But I imagine that the present money value of capital and the present money value of output have been deflated by reference to the same, probably a wholesale, index number.

To the best of my memory, I plead not guilty to the invention of the term "reflation". I fancy it comes from America, or did Hartley Withers think of it? [2] I will make enquiries. But however that may be, it certainly is not a conflated form of "reinflation". It means something quite different from that. "Reflation" is so to speak the blowing of oneself up from a state of deflation to a normal condition. "Inflation" is the state of blowing oneself up from a normal condition to being indecently distended. The Oxford Dictionary is overlooking the fact that, whether or not it is etymologically correct, the root word is "-flation". [3]

Yours ever,

J M Keynes

Are you not assuming falling prices & efficiency increases, whereas I am assuming rising incomes? If my income rises as a result of an invention, that does not make the excess of my income over my consumption any the less saving.

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

  1. 1. Data refer to the period 1860-1913: see "Some Economic Consequences of a Declining Population", in Keynes's Collected Writings, vol. XIV, p. 128. There is a slight discrepancy in the index of real capital for 1913, which in his lecture Keynes estimated to be 170.

    2. The word probably comes from America. The second edition of the Oxford English Dictionary agrees with the supplement to the first edition in attributing the first occurrence in print of the word "reflation" to an unsigned article in The Economist of 20 February 1932 on "Reflation in the United States". This describes a Bill concerning the concession of advances by the Federal Reserve Bank to a member bank, which was passed by the House of Representatives and was before the Senate at the time of writing. The term "reflation" was probably first used in the congressional debates or during the preliminary discussions in the drafting committee.

    3. See, for the viewpoint of the OED, letter 627 R.

    1. a. TLS with autograph addition and corrections, three pages numbered from the second, on three leaves, in HP II-67. Cc, without autograph marks, in JMK CO/2/2/76-79. 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