Cross-sectional analysis of demand for labor and capital inputs in manufacturing industries : a case study of Korea

Typescript.Thesis (Ph. D.)--University of Hawaii at Manoa, 1984.Bibliography: leaves 97-106.Microfiche.xi, 106 leaves, bound ill. 29 cmThis study addresses questions concerning factor demand, labor-labor substitution, and labor-capital substitution in factor markets for overall manufacturing and 2-digit manufacturing industries of Korea. The objectives are twofold: 1) to analyze input demand, and labor-labor and labor-capital substitution where labor heterogeneity is captured by disaggregating labor by occupation, and 2) to test the null hypotheses of alternative specification of production structure on nonhomothetic model. This study is different from other studies in that (1) it examines the input interrelationships for 2-digit industries, (2) it utilizes Korean cross-section by subindustries in studying aggregate manufacturing, and (3) it utilizes pre- and post-embargo data. The study is further differentiated from other studies in that the nonhomothetic production structure is postulated. Finally, it relates a developing economy. We know of no existing study for developing countries which explicitly disaggregates labor types. The objectives of this study require the estimation of production relations such that there are no a priori constraints on the elasticities of substitution between factor inputs. Thus, translog cost functions are specified as quadratic approximations to the production process; where the relevant inputs are labor, disaggregated by occupation, namely, production labor (blue collar workers) and nonproduction labor (white collar workers), and capital. As for the estimation procedure, Zellner's "seemingly unrelated regression" technique is employed to estimate the cost function coupled with the factor share demand equations, accounting for cross-equation disturbance correlations and adding-up parametric restrictions. The features of production technology are characterized by the nonhomothetic production structures not only for overall manufacturing but also dominantly for 2-digit manufacturing industries in Korea. Our empirical results indicate that each factor input is a substitute for the other factor inputs in overall manufacturing and that capital is more substitutable for production labor than for nonproduction labor. When we proxy the amount of human capital embodied in labor by• the occupation, the labor with a greater amount of human capital is less easily substitutable for physical capital. This finding is consistent with the Griliches hypothesis that, "skill or education is more complementary with physical capital than unskilled or raw labor." In addition, as labor has more human capital, the derived demand for labor becomes less elastic to its own prices. This implies that as a worker accumulates human capital, both the employer and the worker have a greater incentive to continue his employment regardless of possible minor wage fluctuations. The low overall elasticity of substitution between capital and labor has an important employment-wage implication for Korean industrialization in that it has contributed to the rapid growth of employment and wages rather than impeding their growth. However, while the limited substitution among the factor inputs may temporarily moderate the effect of the wage increase on employment, it may not prevent the higher wage rate from forcing the producers toward labor-augmenting technologies in the long-run. Investment tax credits will bring about a desired increase in capital accumulation, and there will be generated a substitution effect from labor to capital, reducing the demand for both types of workers given the level of output. Since the elasticity of substitution of production workers for capital is higher than that of nonproduction workers, production workers will be more adversely affected. In other words, employment of less skilled, young workers would suffer under investment incentives. Finally, the own price elasticities of demand for production labor are quite sizable, while those for capital are smaller than those for labor. The evidence that demand for labor is relatively more price responsive than capital input suggests that greater attention be given to policy incentives, such as employment tax credits, which seek to stimulate employment by inducing movements along demand schedules