Non-Linear Programming in Educational and Manpower Planning: a Case Study of Thailand.

Existing mathematical models for educational and manpower planning are usually based on restricted assumptions of elasticity of demand for labor and of substitution among different types of labor. Such models cannot identify the feedback of an expansion of the educational system upon the contribution of educational output. The purpose of this study is to construct a mathematical model taking into consideration the effect of an increased labor force on their marginal productivities. The demand for labor in the model is neither vertical nor horizontal as it is assumed to be in the manpower-requirements or rate-of-return approaches, but downward sloping which implies that marginal productivity of labor depends on the number of laborers employed. Non-linear programming is applied to the problem of simultaneously determining the optimal student enrollments by level of education, optimal number of newly educated and uneducated labor types, optimal number of teachers to be produced and resources to be spent in the educational system. The objective function which is to be maximized subject to linear constraints is the production function of the economy, assumed to be quadraticly dependent on the number of laborers. The model is applied to Thailand 's data and yields the optimal path for 1976-1991. The major results insist upon the decreasing trend of educational investment in all types of education, even primary education. Primary education is an economic extravagence so long as the model allows substitution between uneducated labor and primary-educated labor. If choices of educational level are the major area of concern, the study results confirm the fact that primary education is to be expand ed to a very large extent with the condition that labor is completely substitutable. Higher education, on the contrary, should be reduced if labor is completely substitutable and expand ed if it is not. The study sensitivity analysis implies that optimal policy variables determined from the planning model depend to a very large extent on the assumption of elasticity of substitution among different types of labor and on demand for labor. Different assumptions concerning these parameters yield different results.Ph.D.EconomicsUniversity of Michiganhttp://deepblue.lib.umich.edu/bitstream/2027.42/158762/1/8204787.pd