Economia Politica. Rivista di teoria e analisi Economia Politica. Rivista di teoria e analisi
Non-Technical Abstract
Harrod's dynamics underwent a strange fate. On the one hand, his celebrated 'growth equation' was incorporated in all textbooks on macrodynamics or growth up to the early 1980s, and his 1936 multiplier-accelerator mechanism was referred to as the pioneer of a number of dynamic models. On the other hand, although he was flattered by these interpretations he never completely agreed with them, and indeed often complained that he did not attempt to produce a 'model', that he was dealing with cycles rather than growth, that his dynamics was more 'fundamental' than the competing ('econometric') approach. Harrod's lament, however, passed completely unnoticed by his contemporaries.
A careful reading of Harrod's writings suggests that he had indeed some grounds for complaint. This raises the problem of why some ideas not blatantly absurd (some of them were actually rather full of good sense) can not gain acceptance by fellow scientists. Harrod's dynamics therefore offers an interesting case study for the conditions under which new ideas circulate, change and are integrated in scientific thought.
Historians of thought in various disciplines have recently pointed out that acceptance of scientific ideas does not depend only on their logical stringency or heuristic value, but also on their author's capacity to present them with the right balance of emphasis between novelty (which is what makes an idea interesting) and consistency within a certain tradition of thought, and to convince fellow scientists that the new approach is the rational solution to the problem at hand.
Harrod seems to have been aware of these requirements. He always insisted on the 'revolutionariness' of his own dynamics but also showed how it was compatible with 'traditional economics', which dynamics was meant to encompass. And, more interestingly, he tackled the problem of the trade cycle in terms of the rationality of the approach right from the beginning of his research.
Harrod's starting point was the criticism to the theories tacitly assuming stability of equilibrium and resorting, for explaining the cycle, to exogenous forces keeping the system out of equilibrium. He thought instead that in order to provide an explanation of persistent fluctuations one should introduce some source of instability at the outset. This argument was logically and chronologically prior to the devising of his multiplier-accelerator mechanism. He was fully aware that the issue at stake was the 'kind of explanation', not simply the development of a trade cycle model.
Yet, Harrod was not successful in convincing his fellows of the necessity of basing the dynamic approach on this postulate, which he saw as the 'fundamental cause' of the cycle. He criticised Robertson, Lundberg, Hicks and the econometricians for assuming that the cycle was generated by lags or maladjustments. He rejected the interpretations of the instability principle as a result to be proved and he refused to admit that his 'model' needed to be completed by the introduction of some assumption as to the connection between different states of the system. However, none of Harrod's readers took these complaints seriously, and his attempt to lay new foundations for economic dynamics passed completely unnoticed.
Some reasons for this neglect are contingent, others are more profound. Harrod's poor understanding of mathematics led him to misinterpret the 'econometricians'' approach and did not enable him to provide a formulation making his intuitions explicit. Moreover, after the war the problem of the cycle no longer interested researchers, who engaged instead on growth theory. The major obstacle, however, consisted in the fact that Harrod's results could also be obtained with the instruments of the competing approach, which therefore looked heuristically more fecund.
Harrod thus failed on his own terrain, as he could not prove that his approach was the most rational one. In the eyes of his contemporaries, he appeared as a man behind his times. Yet some of his ideas eventually proved correct: non-linear dynamics models show that persisting cycles are characterised by unstable stationary points surrounded by stable attractors. Although formulated in unconvincing terms, Harrod's intuition thus proves to be ahead of his time —or, at any rate, beyond the mathematical understanding of his contemporaries.
DANIELE BESOMI is research fellow of the Divisione della Cultura, Canton Ticino, c.p. 59, 6950 Tesserete, Svizzera (home address)
dbesomi@guest.cscs.ch
Go back to the Home Page 
Go back to the Index
Redazione: econpol@spbo.unibo.it
Webmaster: lrossi@spbo.unibo.it