An Analysis of Rice Import Demand in Selected Asian Economies.

Abstract

The primary purpose of this study was to analyze the rice import demand behavior in seven Asian countries namely, India, Indonesia, the Republic of Korea, Malaysia, the Philippines, Sri Lanka and Bangladesh. Three rice import demand models were estimated, the direct demand model which treats imports as a function of border prices and income, the residual import demand model where the imports are the excess of consumption over domestic production, and the reduced form model which incorporates both domestic demand and supply variables. Import demand functions were estimated by the ordinary least squares (OLS) method. Seemingly unrelated regression (SUR) was used in the estimation of the residual model hypothesizing that government policies contemporaneously influence both supply and demand. Structural stability tests were performed on all equations assuming that a structural break may have occurred in the early 1970s. The same standard model specifications were used for all countries in order to make inter-country comparisons possible. The data for the analysis were mainly from the Food and Agriculture Organization of the United Nations, the International Rice Research Institute in the Philippines and selected publications of the World Bank. The effects of government policy on rice import demand were explicitly incorporated into the models by including subsidized input-output price ratios and a technology index. The results indicated evidence of structural changes in demand and supply variables for all the countries analyzed. Instabilities were pronounced in the supply variables such as the input-output price ratio and the technology index indicating effects of government policy incentives to domestic production. Further, structural changes were observed in income, wheat prices and foreign exchange reserves. Out of the three models the reduced form model performed best in terms of forecasting efficiency. The residual import demand model estimates were the least efficient; however in this model SUR estimates were only marginally better than the OLS estimates. As Asian rice importers reach near self-sufficiency a shift away from input-output subsidies was noticeable but technology incentives continued unabated inducing shifts in the supply curve.