Construction of a Food Demand Sub-System - A Case Study of Fish-Livestock Products in Taiwan

• Main Authors: Lee-Jung Lu, 呂麗蓉
• Other Authors: Fu-Sung Chiang
• Format: Others
• Language: zh-TW
• Published: 2002
• Online Access: http://ndltd.ncl.edu.tw/handle/32088900623889554711

碩士 === 國立海洋大學 === 應用經濟研究所 === 90 === Annual per capita consumption data for the period from 1968 through 2000 were used in this study to estimate the demand relationships among fish, pork, beef, chicken and other meat products in Taiwan. The quantities consumed of these five meat products were calculated from the production data. Prices for fishery products and other meat products were the weighted average of the individual fish species or meat in each group. The purposes of this study are (1) to construct a database of fish and livestock products in Taiwan for the period from 1968 through 2000; (2) to select the best demand system from four differential demand systems using the data constructed; (3) to study how meat expenditures and prices influenced consumer demand for fish and livestock products in Taiwan. The four demand systems used in this study are: LES, AIDS, LA/AIDS, and the Rotterdam model. Seemingly unrelated regression method was used to estimate the demand parameters; and these demand parameters were then used to derive their respective demand elasticities. Results show that the Rotterdam model performs better than the other three demand systems. In addition, this study also tests whether the parameters and elasticities derived from Rotterdam model are sensitive to the data of different sources or not. The empirical results of this study are summarized as follows: 1) using the same data, different model specifications resulted in different demand parameter estimates; 2) using different data sources, the same demand system specification resulted in different demand parameter estimates; 3) the income elasticities for the five meat groups derived from the four models are all positive, and among the five groups, beef has the highest elasticity, which shows its consumption was sensitive to changes in fish/livestock expenditures; 4) beef consumption is very sensitive to its own price changes (own-price elasticity is —1.043), which indicates that when beef price decreases, beef consumption would increase and expenditures on beef would increased (I am not sure whether profits can be increased or not, your study probably did not address this problem). Pork, chicken and other meat (own-price elasticities are —0.652, -0.178, and —0.051, respectively) were less sensitive to their own-price changes; therefore, when price increases, expenditures on these meat products would increase; and 5) the results derived from the Rotterdam model suggest that the substitution and the compliment relationships among these five meat groups are fairly weak. Annual per capita consumption data for the period from 1968 through 2000 were used in this study to estimate the demand relationships among fish, pork, beef, chicken and other meat products in Taiwan. The quantities consumed of these five meat products were calculated from the production data. Prices for fishery products and other meat products were the weighted average of the individual fish species or meat in each group. The purposes of this study are (1) to construct a database of fish and livestock products in Taiwan for the period from 1968 through 2000; (2) to select the best demand system from four differential demand systems using the data constructed; (3) to study how meat expenditures and prices influenced consumer demand for fish and livestock products in Taiwan. The four demand systems used in this study are: LES, AIDS, LA/AIDS, and the Rotterdam model. Seemingly unrelated regression method was used to estimate the demand parameters; and these demand parameters were then used to derive their respective demand elasticities. Results show that the Rotterdam model performs better than the other three demand systems. In addition, this study also tests whether the parameters and elasticities derived from Rotterdam model are sensitive to the data of different sources or not. The empirical results of this study are summarized as follows: 1) using the same data, different model specifications resulted in different demand parameter estimates; 2) using different data sources, the same demand system specification resulted in different demand parameter estimates; 3) the income elasticities for the five meat groups derived from the four models are all positive, and among the five groups, beef has the highest elasticity, which shows its consumption was sensitive to changes in fish/livestock expenditures; 4) beef consumption is very sensitive to its own price changes (own-price elasticity is —1.043), which indicates that when beef price decreases, beef consumption would increase and expenditures on beef would increased. Pork, chicken and other meat (own-price elasticities are —0.652, -0.178, and —0.051, respectively) were less sensitive to their own-price changes; therefore, when price increases, expenditures on these meat products would increase; and 5) the results derived from the Rotterdam model suggest that the substitution and the compliment relationships among these five meat groups are fairly weak.