Technological Adaptation, Optimum Level of Backwardness and the Rate of per Capita Income Growth: An Econometric Approach

Abstract

Technological progress is achieved either by the domestic discovery and application of knowledge (invention innovation) or by its transfer from abroad through adaptation ac tivity. Technological adaptation is defined here as the process of assimilation of new production functions, used or usable in a more advanced economy, by a less developed economy. The role of technological progress as a major cause of economic growth, alongside physical capital stock and the services of labor, has been described, explained, and fairly well established by several authors including Abramovitz,1 Solow,2 Kendrick,3 Massell,4 Brown and Popkin,5 and Denison.6 But there has been some dis sent. Griliches and Jorgenson attempt to show that errors in measurement of the contribution of capital account for a substantial proportion of measured changes in total factor produc tivity. 7 However, this objection stems from definitional differences and is not a substantive one.8 In what follows we shall see that a favorable interpretation of Griliches and Jorgenson's con tribution classifies their hypothesis as pertaining to a special case. The relevance and validity of their results hold only for the most advanced economy (the producer of new technology). Abramovitz's Residual will re main for those countries where technology is adapted, unless we are prepared to assume the duplication of technology is as costly as its original production, and this contradicts actual economic behavior and historical observation. Accordingly, various cross-national studies have led to the rediscovery of Abramovitz's residual.9 Finally, Denison in his recent comprehensive cross national study observes that various Western European countries obtained more growth from increase in capital per worker than the United States.10 The concept of the advantages of lurks behind such a conclusion, although this study will indicate that the concept of an level of is more realistic, theoretically plausible, empirically verifiable, and thus more appropriate. Extreme backwardness is not bliss even from the point of view of technological adaptation. The highest rate of technological adaptation and the associated economic windfall is achieved not by the least developed but by the moderately developed economies. This observation partially explains the difference between the high rate of per capita in come growth of the middle income countries in contrast to the smaller rate achieved by the very poor (least developed) and the very rich (most developed) countries. Gomulka assumes the dif fusion of knowledge (adaptation) across inter national boundaries to be costless.11 In this study, however, we assume that the adaptation of a technique is a less costly activity than its discovery and development, implying a higher marginal ef ficiency of capital accruing to the adaptor of technology in contrast to the producer of technology. Therefore our main task in this paper is twofold: 1.) to measure the contribution of technological adaptation to the per capita income growth rate, thus explaining the observed income growth differentials among countries, and 2.) to determine the optimum level of backwardness defined by a level of income or stage of economic development for which economic windfall due to adaptation activity is maximized. Accordingly, an adaptation growth model is constructed; its properties are examined, and finally, it is em pirically tested.