能源稅的經濟效果－－考慮參數不確定的可計算一般均衡分析

ECONOMIC EFFECTS OF ENERGY TAXATION: A COMPUTABLE GENERAL EQUILIBRIUM ANALYSIS IN THE PRESENCE OF PARAMETER UNCERTAINTY

本文探討以能源含碳量多寡為課徵基準的能源稅， 並分析該稅制對臺灣總體經濟與產業結構的影響。目前文獻上已廣泛應用可計算一般均衡模型（computable general equilibrium model） 來評估氣候變遷議題， 惟此一模型在不同研究但相同議題上經常產生差異的結果，甚至是採用相同的模型。理論上，能源是一種投入，因此能源與生產結構的關係對評估能源稅效果應有相當的重要性， 此一關係特別在參數值的設定上尤其重要。本文除探討能源稅的經濟效果外， 我們擬進一步考慮生產結構的參數不確定性因素， 利用一個包含非固定參數值的可計算一般均衡模型， 來重新解釋課徵能源稅是否會因參數值的設定而產生差異的影響。首先，本文編制一個2004年臺灣社會會計矩陣表； 其次， 以該資料為基礎， 建立一個臺灣的可計算一般均衡模型； 接著，我們整理近30 年21 篇有關要素替代彈性估計的文獻， 進而設定本研究有關的要素替代彈性值區間； 最後， 透過此區間求解模型內生變數（總體經濟與產業結構）的極端值。本研究發現：能源稅對臺灣總體經濟與產業結構的影響效果，會隨著要素替代彈性的設定而有不同的影響，特別在消費面的民間消費與均等變量； 同時伴隨能源使用成本增加而帶動的CO2 減量幅度， 也會因參數值的差異而有不同的影響， 而且影響幅度差異甚大。本研究同時模擬在相同的CO2 的減量幅度， 會因本文所設定的參數值區間而有不同的能源稅課徵幅度， 最大值高於最小值達75%， 顯示要素替代彈性的設定對氣候變遷政策的擬定具相當影響性。學理上， 當能源成本增加， 較高的能源替代彈性代表廠商替換較有能源效率設備的調整程度也較容易，因此對CO2 減量成本與整體經濟會有較小的影響；由於跨國間的要素替代彈性差異頗大，這也顯示要素替代彈性對跨國間溫室氣體減量成本的比較應該占有相當重要的角色。

Computable general equilibriummodels (CGE) are nowwidely used to simulate energy and climate policies in both developed and developing countries. The simulation results and conclusions reached by these models depend on the structure of production factors and the size of the parameters specified. Some empirical studies on Armington’s elasticities in CGE models for energy analysis were found, but the parameter uncertainty casts doubt on the reliability of simulation results. In this paper,we have applied interval sensitivity methods for examining this type of parameter uncertainty. The objective in this paper is to provide sensitivity assessment of economic effects from changes in the parameter of elasticities in production function following the energy taxation increase. The estimates are derived from a 14-sector computable general equilibrium model calibrated to a 2004 social accounting matrix. We rely on building parameter interval sets, of which five parameters are considered. In order to obtain reliable simulation resultswe should employ empirical evidence gathered on the substitution elasticities from previous studies. Our results show the economic effects of energy taxations depend substantially on the degree to which capital and labor can substitute for energy. With the interval parameters incorporated, we find that extreme values of carbon emissions vary about 75%, and obtain evidence of large different results across industries and macro-variables. We also show that the economic effects of energy taxations depend on the degree of substitution between production factors. As we know, the elasticity of substitution is the degree of ease with which the firm can buy energy efficient equipments when energy costs increase. The CGE models that use a lower elasticity of substitution may have smaller economic impact by energy costs increase. Our results support the conclusion that the substitution elasticities between production factors have a major influence in evaluating energy-related policies.