增加政府固定投資的經濟影響：投入產出模型與可計算一般均衡模型的比較

The Economic Impacts of Increased Government Spending: A Comparison of Input-Output Model and Computable General Equilibrium Model

本研究建立了我國的投入產出模型（input-output model）和可計算一般均衡模型（computable general equilibrium, CGE, model）來評估增加政府固定投資對總體經濟與產業經濟（如農業經濟）的影響，就這兩個模型在理論架構上的差異和實證結果的不同，進行分析比較並提出深入的解釋。研究結果顯示政府增加固定投資對我國總體經濟的衝擊受到所選用的模型的影響相當大，而對個別產業（如農業經濟）也是如此。由本研究的成果我們不難發現，產品與要素（勞動與資本）市場是否存在價格的僵固性，要素的總供給是否出現限制，都是政府固定投資是否能夠刺激經濟成長與帶動產業發展的決定性因素。I-O模型模擬得到的經濟影響度僅僅能視為極短期的最大帶動效果。當一國的經濟體趨於成熟，市場價格機制扮演重要角色時候，以凱因斯封閉法所架構的CGE模型對於短期現象也有相當合理的解釋能力。而當研究高度開發之經濟體的長期現象時，由於勞動力與資本財已經達到有效利用，一個以古典封閉法架構出的CGE模型會比I-O模型具有更好的解釋能力。 In this research, I constructed a Taiwan input-output (I-O) model and a Taiwan computable general equilibrium (CGE) model to study the economic impacts of increased government spending on the Taiwan economy and its industries. The theoretical structure of these two models was compared and the differences in their simulation results were addressed in depth. I found that the economic impacts of an increase in government fixed investment are sensitive to the model selected. Especially, the rigidity of prices in commodity and factor (labor and capital) markets and the supply elasticity of factor inputs inherent in the structure of I-O and CGE models play an important role in the determination of the magnitudes of economic impacts. The research finding indicates that the policy simulation results of an I-O model are best interpreted as the maximal possible short-run impacts on the economy. For a mature economy where markets are well established and price signals are most important information in market transactions, the model simulations based on the CGE model with Keynesian closure may generate acceptable estimates of economic impacts on the economy and reasonably explain the short-run economic phenomenon of policy shocks. When studying the long-run phenomenon in a well-developed economy where all factor inputs are in fixed supply in the long run, a CGE model with neoclassical closure is more suitable than an I-O model in addressing policy issues.

In this research, I constructed a Taiwan input-output (I-O) model and a Taiwan computable general equilibrium (CGE) model to study the economic impacts of increased government spending on the Taiwan economy and its industries. The theoretical structure of these two models was compared and the differences in their simulation results were addressed in depth. I found that the economic impacts of an increase in government fixed investment are sensitive to the model selected. Especially, the rigidity of prices in commodity and factor (labor and capital) markets and the supply elasticity of factor inputs inherent in the structure of I-O and CGE models play an important role in the determination of the magnitudes of economic impacts. The research finding indicates that the policy simulation results of an I-O model are best interpreted as the maximal possible short-run impacts on the economy. For a mature economy where markets are well established and price signals are most important information in market transactions, the model simulations based on the CGE model with Keynesian closure may generate acceptable estimates of economic impacts on the economy and reasonably explain the short-run economic phenomenon of policy shocks. When studying the long-run phenomenon in a well-developed economy where all factor inputs are in fixed supply in the long run, a CGE model with neoclassical closure is more suitable than an I-O model in addressing policy issues.