The Demand for Meats in Indonesia: A Censored Regression Approach

In this study, meat consumption and socio-demographic data from the 1990, 1993 and 1996 SUSENAS Household Food Expenditure and Consumption Surveys were employed to estimate the demand for meats in Indonesia. The provinces of DKI Jakarta and West Java were chosen as the areas of study because of the population, level of meat consumption and the availability and quality of information in these two provinces. Several statistical and econometric procedures were performed. Firstly, a cluster analysis (Nicol, 1991) was used to aggregate the 16 meat types recorded in the SUSENAS into four Meat Groups (MG-1, 2, 3 and 4). Secondly, a double truncation procedure was used to estimate the linear approximation of the Almost Ideal Demand System (LA/AIDS) because of the large number of zero observations in the data as well as the fact that budget shares lie between zero and one. It is expected that the results of the study are more robust than previous censored regression approaches which only considered one-sided truncation at zero. The estimated expenditure elasticities show that MG-1 (with the dominant meat, beef) and MG-3 (with the dominant meat, untrimmed bones) are income-inelastic, whereas MG-2 (with the dominant meat, commercial and native chicken) and MG-4 (with the dominant meat, beef liver) are income-elastic. The estimated uncompensated own-price elasticities are negative, as suggested by economic theory. The estimated own-price elasticity of MG- 1 is -0.92, therefore, it is inelastic whereas MG-2, 3, and 4 have elastic estimated own price elasticities, -1.09, -1.16 and -1.03 respectively. The estimated uncompensated crossprice elasticities suggest that all the meat groups tend to be substitute goods as expected.