Robinson

["Neri Salvadori" <nerisal@xxxxxxxxxxx>]

I have inserted in the directory "incoming" the UNCODED Postscript file of the
paper "'Productivity curves' in _The Accumulation of Capital_" (I hope that Ric
will like to move it to the directory "Salvadori") The title of the file is
"Robinson.ps". (People who got difficulties with the coded version of the file
are welcome to get the present version.) The paper has been delivers at a
Conference on Joan Robinson in Turin and will be delivered again at the
Convention of the History of Economics Society in June. Comments are very
welcome. The file is very long (707K) since there are several figures (the
papers is of 16 pages only). A description of the paper could be the following.

In _The_Accumulation_of_Capital_ Joan Robinson (1956) developed a description of
technology in terms of "productivity curves". After the publication of
_Production_of_Commodities_by Means_of_Commodities_ by Sraffa, Joan Robinson
preferred to abandon that description of technology. In the four years between
the publication of _The_Accumulation _of_Capital_ and that of
_Production_of_Commodities_by Means_of_Commodities_ there was no mathematical
formalization of the Robinsonian productivity curves put forward. In the years
afterwards the preference for the Sraffian construction by Joan Robinson herself
was certainly not inviting a job like that. This is a pity for at least one
reason: the description of technology in terms of productivity curves is much
more workable for economists with a Neo-Classical background and interests in
macroeconomics. This fact becomes especially relevant in these days when growth
theory is again fashionable in terms of endogenous growth. The paper presents a
formalized version of the Robinsonian "productivity curves".

A "productivity curve" is indeed a sort of production function which is built
upon the assumption that the rate of profit is kept constant. This assumption on
the rate of profit allows one to measure capital in terms of the consumption
good correctly (consumption is assumed to be proportional to a given basket of
commodities). Of course a productivity curve has interest only in the point(s)
of the function in which the slope equals the given rate of profit. Since this
analysis can be performed for each feasible rate of profit, the wage rate-profit
rate relationship, the capital-profit rate relationship, and the output-profit
rate relationship can be determined.

The paper can be read also as a comment to the first part of an appendix to The
Accumulation of Capital called "Diagrams" (pp. 411-423). Each section of the
paper starts with references to this appendix which are useful to grasp the
relevant concepts as stated by Joan Robinson. These concepts are then analyzed
with the help of the mathematical tools which have been considered appropriate.
The relevant concepts mentioned are those of "productivity curve" ($ 2), "family
of productivity curves" ($3), and "pseudo-production function" ($4). The
concluding remarks ($5) are devoted to possible generalizations and
applications.




Neri Salvadori

Dipartimento di Scienze Economiche, Universita' di Pisa
Via Ridolfi 10, i56100 PISA (Italy)

FAX: (39)(50)598040
e-mail: nerisal@xxxxxxxxxxx
Tel: (39)(50)549215