Reply: Complementary Innovations by Roy Harrod and Alvin Hansen

When the editor of HOPE asked my opinion about the merits for publication of the Heertje-Heemeijer manuscript, “On the Origin of Samuelson's Multiplier-Accelerator Model,” I replied: You should publish it in its exact submitted form, but do provide me with the opportunity (as a living testifier) to publish with it my following commentary.

I applaud these authors for calling attention to Roy Harrod's important post-1935 contributions to the literature on the acceleration principle and macroeconomic business cycles. Harrod was a brilliant innovator who felt—properly felt—that a number of his seminal novelties failed to receive their deserved recognition. (The logician genius Frank Ramsey, as Economic Journal referee, advised Keynes not to publish Harrod's early article defining what came to be known as “marginal revenue,” thereby robbing Harrod of deserved priorities. Just before Keynes completed the text of his 1936 General Theory, Harrod's correspondence convinces me that Roy better understood the nuances of that paradigm than Maynard then did: for instance, in the 1937 Irving Fisher Festschrift, Keynes nodded by proclaiming that in his theory the interest rate that got established by liquidity preference could not be a positive function of a rising “marginal efficiency of capital”; Harrod had earlier demurred but apparently without effect. Also, Hicks's [1937] famous IS-LM diagrams of Model-A Keynesianism owed more to pre-1937 Harrod than we readers of that time realized.)

It does not follow logically, or on the basis of any factual evidence presented by Heertje and Heemeijer, that rendering unto Harrod credits he deserves must serve to discount from Alvin Hansen the attributions I have claimed for his analyses on me in the 1937–39 Harvard years.¹

The development of science is a multiperson game. It is not necessarily a zero-sum game where granting fame to a Harrod must subtract it from a Hansen. I have never been a believer in the Carlyle-Schumpeter theory of great scholars being all important in the history of science. Hansen was no mathematician (but see Samuelson 1976 for my laudatory evaluation of his theoretical acumen and originality). But years earlier at the University of Minnesota he had steeped himself in the acceleration principle literature of J. M. Clark (1917), Charles Bickerdike (1914), and Albert Aftalion (1913). Ragnar Frisch's 1931 American visit, which included Minneapolis, alerted Hansen to the intricacies of replacement depreciation in connection with the acceleration principle.² I am not sure that Hansen had noticed the clear recognition of the acceleration principle in Ralph Hawtrey, especially in the latter's Good and Bad Trade (1913); my Bayesian probabilities are that he did not know of Gustav Cassel's complete anticipation of Harrod-Domar in Theoretische Sozialökonomie (1918), the fifth edition of which was translated into English in 1932. Part of Hansen's coolness toward Harrod's Trade Cycle (1936) came from his resentment that a scholar would in that year proclaim as his own novelty (with a new name, the “relation”) what had been a staple in the literature for more than twenty years. (Note: the multiplier-accelerator was not that staple; the widely known accelerator was only one [of the two] functional equations, the one that connected the level of investment to the rate-of-change of consumption [or income]. Quite possibly Hansen and I, at his suggestion, were the only two frontier workers who read Harrod's book with a magnifying glass.)

I reaffirm now, from clear memory, that my 1939 difference equation(s) came directly from the numerical sequence on which Hansen based his thesis—his imperfect thesis in his 1938 book—that the pre-1937 U.S. upturn was a consumption-based recovery and that such recoveries by their arithmetic and logical nature were doomed to be finite-lived and to end in a permanent downturn (until the next favorable exogenous shocks happen to occur). I...