Biased technological change and factor demand in postwar Japanese agriculture, 1958–84

The objective of this study is to gain a better understanding of factor substitutions in postwar Japanese agriculture by shedding special light on biased technological change. Such biased technological change is first measured, then associated with the movements of factor prices, and then tested for the Hicksian induced-innovation hypothesis. In addition, a decomposition analysis is carried out in order to quantitatively examine the importance of the biased effects for determining changes in factor-cost shares and factor proportions during the 1958-84 period. A slightly modified Stevenson-Greene model of the translog cost function is employed. This model has at least two important advantages over ordinary translog cost functions. First, it incorporates time into the model such that all coefficients of the ordinary translog cost function may change over time. This is more realistic than the ordinary translog cost function which assumes that all coefficients are constant over the period of estimation. Another attractive feature is that it enables us to test the induced-innovation hypothesis directly. The model is applied to the 1958- 84 period by making use of farm-level data. The results show that technological change was biased towards saving labor and other inputs and using machinery, intermediate inputs, and land. This biased technological change is found to be, in principle, consistent with the induced-innovation hypothesis. Furthermore, it is shown through decomposition analyses that the biased technological change had significant impacts on changes in factor-cost shares and factor proportions during these years. The empirical results of this study imply that technological change in Japanese agriculture has, in general, proceeded in a manner consistent with factor endowments conditions since the late 1950s. An implication of this study for agriculture in less-developed countries is that agricultural policies seeking development through technological progress should be carried out so as to take advantage of peculiar factor endowments conditions in the individual countries.