227. Viner to Harrod , 23 February 1932 [a]
[Replies to a letter not found]
Department of Economics, University of Chicago #
23 February 1932
Deepest apologies for my delay in acknowledging both the reprint of your excellent article  and your most generous note of comment on my own cost article.  I was convalescing from an appendix operation when your article and letter arrived, and I postponed a reply until I had had time to give your argument careful consideration. There follow some comments, first, on your article, and next, on your letter.
P. 568--"If it be supposed that in a firm (marginal) costs other than competitive marketing costs are falling at the old equilibrium ..." I would contend this is inconceivable as a short-run phenomenon. Similarly for p. 569:--I cannot imagine any internal technological circumstances which would result in decreasing marginal prime costs in the short-run (marketing costs excluded), unless you assume that law of diminishing returns is not operative.
P. 571, Fig. 1. KC p and LC t are stated to represent total prime costs and total costs, respectively, where average prime costs and average total unit costs would be clearer. (In other words, I prefer my own terminology, including "marginal gross revenue" for what you call "increment of aggregate demand."  ) Your text is correct on LC t even if you slipped in the graph. 
Your graphical treatment on p. 571 is identical in substance with Marshall's treatment of pure monopoly. What you show is that under imperfect competition, the conditions of equilibrium for an individual concern are qualitatively identical with those for a complete monopoly.
Pp. 572-3. This is a case of regional monopolies with competition only at the frontiers between the regions. Under these circumstances a discriminatory price policy is in order for all the producers, i.e. "dumping".
P. 573. Next paragraph. The assumption you rightly reject would lead logically to denial of possibility of long-run competition under any circumstances except substantial net diseconomies of large-scale control, since monopoly can always profit through substitution of marginal gross revenue for price as regular of output, even when no economies in cost result from monopoly.
P. 576 to end. Your propositions hold for average costs, but not for marginal costs, if you grant that the average prime cost curve is positively inclined within the range of observation.
The article covers a great deal of ground in small space, and I find nothing in it that I would object to, except the tacit assumption that marginal prime costs may decrease in the short-run as output increases. It would probably be a valid retort even to that objection that the question is one of fact. Have you seen Hotelling's article on somewhat similar lines in the Q.J.E. of several years ago? 
Finally, as to my article, the draftsman and yourself are right, and I was wrong.  It was a natural but mistaken tendency to assume that optimum scale for a particular output and lowest cost output for that scale would be identical. But you saw what was wrong with my reasoning and my draftsman did not.
Could you spare another reprint or two for use with my students?
Most cordially yours,
R. F. Harrod, Esq., Christ Church, Oxford, England
2. J. Viner, "Cost Curves and Supply Curves" (1931).
3. The concept is defined in Harrod, "Notes on Monopoly and Quasi-Competition" (here reproduced as essay 6 , [jump to page] ) and "Notes on Supply" ( 1930:3 ), pp. 238-39. Harrod's terminology was also disliked by Haberler: letter 230 , [jump to page] . For further references see note 4 to essay 6 .
4. The mistake in the graph was also noted by Joan Robinson: letter 224 , [jump to page] .
5. The Quarterly Journal of Economics does not seem to have published contributions by Hotelling previous to 1932. Viner may refer to Hotelling's article on "Stability in Competition", Economic Journal XXIX, March 1929, pp. 41-57, in which the problem of marketing increments of produce is examined.
6. Viner, "Cost Curves and Supply Curves", p. 35. Reference to the argument with the draftsman is in footnote 2 to p. 36. Harrod later commented upon this in "Doctrines of Imperfect Competition" ( 1934:3 ), p. 451n, and "Imperfect Competition, Aggregate Demand and Inflation", Economic Journal LXXXII, March 1972, p. 396. The matter was taken up again in Harrod's review of Viner's The Long View and the Short (Economica, August 1959, p. 262); Harrod, however, forgot this exchange and that he sent Viner an offprint of his article. Harrod's review actually generated some correspondence, with Viner apologizing for not having seen Harrod's article and Harrod blaming himself for not having sent it; in a subsequent letter, Viner further discussed some details regarding the envelope curve (Viner to Harrod, 9 November and 2 December 1959, Harrod to Viner, 24 November 1959, all in HP IV-1202-1212).
- a. TLS with autograph corrections, two pages on two numbered leaves, in HP IV-1202-1212.
top of page
Return to index of this section
Go to previous page
Go to next page