765. N. Kaldor to Harrod , 3 May 1938 [a] , [1]
[Answered by 766 ]
I think your example proves my point--at any rate on my assumptions.
In this case, the two methods are equally profitable.
B is superior to A, for it yields 7.2%, while A yields only 4.1%.
B is superior to A. B is adopted if rate of interest is less than 5%.
A is superior to B. Excess of profits on A: 20%.
A is superior to B. Excess of profits on A: 75%.
Moreover, as you see, the difference in profitability is all the greater, the greater the change in wages.
You are right, on the other hand that excess receipts on B are always equal to 10% of the additional capital required. This would be only relevant, however, if receipts were entirely independent of the amount of capital invested, i.e. when the elasticity of demand is zero, e.g. building a bridge, or a tunnel.
Assuming perfect competition, receipts increase at the same rate as the amount of capital invested. Take case V, assuming equal amounts of cap. invested in both cases:--
By investing an extra 250 in method A, extra profit remains 70% of additional capital. So long as the r[ate]. of i[nterest]. is less [than] 70 per cent, this method will be adopted; B is never reverted to, however much the interest rate might fall.
{You might ask, how is equality achieved in this case between the marginal efficiency of capital (70%) and the rate of interest (which might be anything) in other words, what determines the output of the firm? Answer: it might be increasing marginal risk, which can either be deducted from the m.e. of cap. or added to the rate of interest. [2] [fig. 1]
Fall in the rate of interest, cet. par. leads to an expansion of production, but it leaves the method of prod[uction]. unaffected}.
My provisional conclusions (which leaves room for both of us to be "right"):--
1) Infinitely elastic demand (& constant physical returns to scale):--method depends entirely on the level of real wages, & not on the r[ate]. of i[nterest]. at all.
2) zero elasticity of demand:--method depends alone on the r. of i. & not on the level of real wages at all;
3) Imperfect
competition:--(
) method adopted depends on both. (It is possible, however, that with
joint-stock company-organisation, firms should aim at maximizing
profit per unit of cap invested, rather than aggr[egate].
profits, in which case again it only depends on real wages)
It is not clear what stimulated this exchange. The topic, however, was taken up again by Kaldor in "Capital Intensity and the Trade Cycle" (Economica, NS VI, February 1939, pp. 39-66), an analysis of what determines the optimum degree of capital intensity depending on the elasticities of the marginal efficiency curves corresponding to different alternative methods of production and of the schedule of the marginal cost of borrowing (pp. 47-50). The correspondence with Harrod is acknowledged on p. 50n. See letter 803 , [jump to page] .
Kaldor was engaged in writing this paper in mid-July; he described it as dealing "with the factors that determine the degree of labour-saving-ness of equipment & its variations during the cycle" (letter to Keynes of 14 July, in JMK EJ/1/5/225). The article was ready by 3 September (letter to Keynes, JMK EJ/1/5/228).
2. Reference is to Kalecki's "principle of increasing risk", as expounded in an article of the same title in Economica, NS IV, November 1937, pp. 440-47 (misquoted by Kaldor as "The Principle of Increasing Marginal Risk" on p. 47 of the article cited in note 1 ).
- a. ALI, three pages on three leaves (the last of which halved), in HP IV-669-688. Photocopy in NKP NK/3/30/86/84-86 (photocopies of the part of the Harrod-Kaldor correspondence housed in HP were donated to Kaldor by Chiba University of Commerce in occasion of his visit in Japan in 1985. Among Kaldor's papers there is also a copy of a catalogue of the Harrod's papers in Japanese, probably also donated on the same occasion, in NKP NK/2/164).
=4.1%
=7.2%
=0%
=5%
=16.6%
=13.3%
=70%
=40%