113. D. H. Robertson to Harrod , [May or June 1926] [a]
[Replies to 112 , answered by 114 ]
[May or June 1926]
You have convicted me of a good deal of muddle with your very searching criticism, and justified, I think, the self-abasement of my introductory quotation!  I do not think there is anything left for me but unconditional surrender of the main fortress which you attack.
I, D.H.R., do thereby admit & acknowledge
(i) that, capitalistic variations being defined as those which the enlightened self-interest of capitalists would tend to bring about, the said capitalistic variations are a function of the banking policy in force. This is true (a) because the banks, by stealing Short Lacking from the public, can lower artificially the average price at which entrepreneurs get their Short Lacking; 1 (b) because the rise in prices thus engendered offers a bonus to entrepreneurs at the expense of wage-earners.
(ii) To say therefore that "that banking-policy is good which brings about capitalistic variations" leaves the answer to the question "what banking-policy is good?" indeterminate. There might be many "good" policies. (In point of fact there will, I think, be none: for, as shown, every banking policy tends to bring about larger variations of output than those which capitalistic self-interest, given that banking-policy, would dictate.)
(iii) If I say, therefore, that a banking policy which brings about capitalistic variations is good, I must be held to mean that a banking-policy which brings about those capitalistic variations which would occur under a barter economy, and if there were no difficulty about getting hold of additional supplies of Short Lacking, is good.
(iv) It is not possible to urge in support of this proposition what might be urged in support of the proposition in (ii) above if that proposition had any precise meaning,--namely that if we are not prepared to replace the entrepreneur we shall be wise, if we want progress, to pander to his whims: for ex hyp. we could pander more lavishly to his whims by choosing a more inflationary banking-policy.
Nonetheless I still feel (a) that the case for promoting the type of capitalistic variation defined in (iii) is prima facie stronger than the case for promoting any other type of capitalistic variation, (b) that, if one attaches great importance (as I do) to the considerations set out on p. 22, there is a prima facie case for promoting some kind of capitalistic variation rather than trying to form a direct judgment as to what degree of variation will produce the best long-period results. Personally however I should be prepared even now to take some risks of slowing down progress in the interests of stability. In a stationary state (i.e. one in which population and desires were stagnant and fluctuation was about a constant normal) the case for preferring capitalistic to any other kind of variation would perhaps vanish: but so, in my view, would the greater part of the problem of the trade cycle.
I still incline to stick to my two main propositions: 1) that capitalists have much greater real independent inducements to vary output (as opposed both to fictitious inducements and to real inducements generated by changes in banking policy) than the "psychological school" would lead me to suppose: 2) that, given the need for material progress, the consequent fluctuations are much more in the true long-period interest of society as a whole than a contemplation of the immediate disharmony between the interests of entrepreneurs and those of workpeople would lead one to suppose.
Further, I think it is possible that my instinct to glaze over (i)(a) above was not far wrong! I.e. I suspect a good deal too much has been made of the direct stimulative or deterrent effects on the entrepreneur of changes in the bank-rate. The main consideration, from the entrepreneur's point of view, is whether or not he can get hold of the requisite Short Lacking at all; & I doubt whether the rate of output which, provided he can get the requisite Short Lacking, it is rational for him to establish, varies very much with the price which he has to pay for that Lacking.
A minor point on attached half-page.
I return all of your letters for reference, and in case you ever think the original worth while criticising in print anywhere. If not, I hope you'll keep them or let me have them back sometime, as I should like to refer to them again if ever it comes to a re-writing. 
Your p. 3, 3rd paragraph. 
Your supposition that in certain conditions the elasticity of supply of labour might be greater than (might for instance, so as to justify (c), be ) seems to be quite possible. It is however (isn't it?) on the face of it inconsistent with my proposition (so far, I think, accepted by you) that "capitalistic variations" will be greater than "workmen's variations". But that proposition (p. 20), and the whole argument of Ch. III, was meant to apply to variations from an assumed normal, and does not necessarily hold up upward expansions from a very sub-normal point which has been arrived at at the dictation of capitalists. Starting from such a point, the elasticity of supply of labour might well, I think, be greater than that of business enterprise,--& both might well be >1.
Is this right?
Harrod pointed out that "what Mr. Robertson seeks to deny, it turns out, is that price stabilization is universally, or even usually, the best policy" ("Mr. Robertson's Views on Banking Policy", Harrod 1927:8 , p. 224).
2. Harrod's letters are preserved among Harrod's papers in HP. Harrod recast his criticism in "Mr. Robertson's Views on Banking Policy" ( 1927:8 ). This was first sent to D. H. Macgregor (who forwarded it to Keynes) for publication in the Economic Journal (see letter 119 R), but was eventually printed in Economica. A revised version of Robertson's book was published in 1932; the alterations, however, do not regard the matters under discussion with Harrod, due to Robertson's unreadyness to undertake a thorough revision (see Robertson's note to the 1932 impression). Harrod's review article is not cited, not even in the preface to the 1949 Kelly edition.
3. Letter 112 , [jump to page] , paragraph beginning "The problem as set ...".
- a. AN, four pages on four numbered leaves (the last one obtained by halving a sheet of the same size as the first three); in HP IV-990-1069d/6.
top of page
Return to index of this section
Go to previous page
Go to next page