This chapter describes the adoption of the Cobb–Douglas regression by agricultural economists over the 1940-1960 period. These economists used the regression to explore a set of long-standing questions specific to agricultural economics. As a result, their defense and development of the method and the criticisms they attracted from their colleagues, while drawing on the prewar literature surrounding the Cobb–Douglas regression, had different emphases, and controversial questions that surrounded Douglas's applications of the regression became irrelevant. The agricultural economists were also the first to estimate the Cobb–Douglas regression using data generated by individual firms, as opposed to the more highly aggregated data used by Douglas and his coauthors, and they embedded it in a probability-based statistical framework. They also developed general models for applying regression analysis to panel data, models that had a profound impact on empirical economic research. All of this contributed to the process through which the Cobb–Douglas regression came to be seen as an empirical tool potentially suited to a broad list of applications.

This chapter describes a process in which a number of long-standing hypotheses about sources of economic growth came to be represented as a series of modifications to Solow’s basic neoclassical production function, which in turn led to attempts to test empirically these hypotheses using production function regressions. Particular attention is given to two empirical research programs that made use of variants of Douglas’s regression procedure: the literature on “embodied technical change” initiated by Robert Solow, and Zvi Griliches’s work measuring the causes of economic growth, both of which emerged in the late 1950s and early 60s. Griliches and Solow proposed empirically tractable ways of testing various hypotheses about causes of economic growth. Their proposals were attractive in part because they made use of an empirical procedure, production function estimation, that involved statistical methods that were coming to be the foundation of graduate training in econometrics, and because they were expressed in terms of the mathematical neoclassical theory that was becoming central to graduate training in economic theory.