Kenneth J. Arrow's 1962 article “The Economic Implications of Learning by Doing” is considered a seminal contribution to endogenous growth theory. However, no history of its origins has been written yet. We aim to fill this gap by studying the genesis of Arrow's article, showing how its context of discovery conditioned its form, the concepts Arrow used and the way he defined them, and the modeling strategy he employed. We first show that Arrow's work on scale economies was part of a research project carried out with Hollis Chenery on technology and resource allocation funded by the National Science Foundation, a project that enabled Arrow to develop a theoretical and macroeconomic theory of increasing returns. We then examine the influence of John Dewey's experiential approach to education on Arrow's use of the notion of “learning by doing” and its approximation by cumulative investment. We finally study the different strategies for modeling learning developed in the 1950s and 1960s, showing the influence of RAND's and Armen Alchian's contributions on Arrow's use of a log-linear equation of learning and its introduction into an economic growth model.

The text of this article is only available as a PDF.
You do not currently have access to this content.