規模經濟、不完全競爭與可計算一般均衡分析：以增加政府支出的經濟影響為例

Abstract

Harris (1984)首開先河，透過一個 加拿大的可計算一般均衡分析 (Computable General Equilibrium，簡稱 CGE) 模型的建構、模擬與分析，強調 是否考量規模報酬與不完全競爭的市 場結構，對貿易自由化經濟影響評估 的結果有很顯著的差異。本研究的主 要研究目的為在考量規模報酬與不完 全競爭的市場結構下，透過台灣動態 CGE 模型與資料庫的建立，以增加政 府支出對我國總體經濟與產業經濟的 影響評估為例，在短期與長期的情境 下探討是否考量規模報酬與不完全競 爭的市場結構對模擬結果有很顯著的 不同。基本模型的延伸主要參考 Abayasiri-Silva and Horridge (1997, 1999)。在短期的情境下，廠商的個數 固定，可能存在有淨利潤的現象；而 在長期的情境下，由於廠商的自由進 出（廠商的個數為內生變數），淨利 潤將不存在。在模型中，規模報酬將 分廠商層級與產業層級兩種情況來探 討，而廠商的定價法將分完全競爭、 獨佔競爭、及Harris 主觀設定等法則 加以分析。模擬情境將因對生產技 術、定價法與廠商進出的不同假設而 有所不同。若是市場結構的差異對模 擬結果有很顯著的影響，則目前國內 相關CGE 的經濟影響評估，例如兩岸 入會、洽簽自由貿易協定、產業結構 預測，甚至於溫室氣體排放基線預測 等，均有必要做進一步檢視。

In a pioneering paper, Harris (1984) emphasized the importance of imperfect competition and economies of scale in understanding the effects of trade liberalization on the Canadian economy within an applied general equilibrium framework. The objective of this research is to examine the short run and long run effects of the increase of government expenditure on the Taiwan economy using an applied general equilibrium model, which incorporates economies of scale and imperfect competition. Economies of scale are incorporated in the model at the industry level and the firm level. The pricing behavior is modeled as perfectly competitive, monopolisitically competitive and in other ad hoc ways, as in Harris (1984). The different assumptions about technology, pricing behavior and firm entry are combined in various ways to produce a variety of scenarios in our simulations. If the market structure like imperfect competition and economies of scale is important in understanding the effects of government expenditure expansion on Taiwan economy within an applied general equilibrium framework, an overhaul on economy-wide analysis on Taiwan’s WTO accession, free trade agreement, industrial structure forecasting and GHG emission baseline forecasting is indispensable.