- 學位論文
台灣砂糖政策有效性之檢定-向量最適理論之應用
The Efficiency Test of Taiwan Sugar Policy－An Application of Vector Optimization Theory
摘要
糖業是我國農業部門除了稻米以外政策性保護色彩濃厚的一項重要產業。在農產品貿易自由化的時代潮流下，我國糖業的保護政策，亦隨即受到了強烈的質疑與討論。然而，從福利的觀點來看，受政策干預下的糖業政策是否真得無效率呢？截至目前為止並無文獻作正式之檢定，故本文欲針對台灣糖業政策之有效性進行檢定。 在檢定政策有效性的方法上，若應用Gardner（1983）所提出的方法，將會面臨兩項重大的限制：（1）對於產業政治經濟模型的設定，只能包含兩個利益團體和和一個政策工具，並不符合真實世界的情況。（2）無法利用統計的大樣本來對政策之有效性作更嚴謹的統計檢定。故Bullock（1995）即利用向量最適理論（vector optimization theory）和拔靴複製法（bootstrap）發展出另一套不同的檢定方法，Bullock所提出的方法，不但對於產業政治經濟模型設定的假設限制較少；且在bootstrap法的應用下，將可對政策之有效性作一較為嚴謹的統計檢定。所以本研究即運用Bullock（1995）所發展的統計檢定方法，來檢定1990年台灣砂糖政策之有效性，亦即檢定政府實際執行的糖業政策是否具柏瑞圖效率（Pareto efficiency）。 在實證的研究結果方面，對於台灣砂糖市場之政治經濟模型的建立，本文將砂糖產業相關的重要利益團體化分為，蔗農、台糖公司以及砂糖消費者三者；而政策工具變數的認定則包含了，蔗農糖外銷收購價格、內銷淨價以及貨物稅率三項。而本文所得到的檢定結果為，在5﹪的顯著水準下，拒絕了糖業政策有效的虛無假設。亦即，經由統計的檢定，1990年外銷收購價格政策、內銷淨價政策以及貨物稅率政策三項砂糖政策同時執行下並不具有柏瑞圖效率。
並列摘要
In addition to rice industry, sugar industry is one of the most important agricultural industries protected by agricultural and trade policies in Taiwan. However, the impact of agricultural trade liberalization attracts many researchers to discuss the argricultural intervention policies. However, there is no research to test the efficiency of Taiwan sugar policy. Thus, the purpose of this paper is to test the efficiency of Taiwan sugar policy. Gardner (1993) developed a method of testing policy efficiency. Applying Gardner’s method to testing policy efficiency will suffer from two important limitations: (i) the inability to include in the model of political economy all of the actually important interest groups and policy instruments; and (ii) the inability to statistically analyze the effects of policy on the welfare of interest groups and on policy efficiency. In order to improve the limitations of Gardner’s method, Bullock (1995) proposed an alternative method that uses vector optimization theory and bootstrap methodology to test policy efficiency. The advantage of Bullock’s method is that fewer restrictions are needed than Gardner’s method. Moreover, Bullock’s method allows formal statistical examination of policy efficiency. Therefore, this paper applies Bullock’s method to test the efficiency of Taiwan sugar policy in 1990. The results of this paper show that the efficient redistributive hypothesis of Taiwan sugar policy is rejected at the 5% significance level.