A Problem with the Empirical Neoclassical Analysis of Multisector Growth

Abstract

One of John Cornwall’s major contributions to our understanding of the economic growth process is to remind us that even the postwar growth of the advanced countries cannot be understood without reference to the massive structural change that has occurred, and continues to occur, in these economies.1 This was brilliantly set out, inter alia, in his 1977 book, Modern Capitalism. This work showed the limitations of trying to analyse economic growth in terms of the aggregate one-sector Solow-Swan model, where (steady-state) economic growth is determined by the exogenous growth of the labour force and the rate of technical progress. Cornwall’s careful marshalling of the evidence showed conclusively that the growth of the labour supply was endogenous, not exogenous. Moreover, his estimations of the Verdoorn Law showed that technical change was largely induced by output growth — that is, determined endogenously, thus anticipating the results of the ‘new growth theory’. Factors affecting the demand for a country’s output cannot be ignored in any understanding of disparities in economic growth.