企業儲蓄與景氣循環：台灣實證研究1962-2012

Abstract

企業儲蓄的變化在經濟環境的復甦上扮演關鍵性的角色，同時也受經濟環境的榮衰影響，近五、六年來企業儲蓄率攀升的現象已被廣泛關注中，但對於成因，是法規的改變、央行的低利政策、銀行與企業間的信貸信賴關係，抑或是景氣循環所致，尚無定論。 本研究探討企業儲蓄率變化的影響因素，特別聚焦在景氣波動因素上，試圖驗證企業利用企業儲蓄作為一種避險工具以因應經濟環境的改變。使用1962年至2012年間總體經濟景氣循環相關變數作為主要研究變數，如：實質國內生產毛額(Gross Domestic Product，以下簡稱GDP)成長率、貨幣總計數變動率、通貨膨脹率，研究透過Autoregressive-Distributed Lag(以下簡稱ARDL)共整合分析法，結合誤差修正模型(Error Correction Model，以下簡稱ECM)，分別解讀企業儲蓄率與各變數間長期均衡及短期動態調整。 為避免遺漏其他影響企業儲蓄的變數，並關心加入其他變數後，景氣循環變數的估計值是否穩定，再以政府消費率、政府移轉性支付率、勞動份酬，以及亞洲金融風暴(虛擬變數)作為控制變數進行穩健性測試。研究發現，落後一期的實質GDP 成長率、落後二期的貨幣總計數變動率和通貨膨脹率對企業儲蓄率有顯著負向關係，貨幣總計數變動率在落後一期及通貨膨脹率落後三期時對企業儲蓄率有顯著正向關係。短期動態調整部分，ARDL模型的誤差項係數在ARDL-ECM各模型全部為負值，且在政府部門、重大事件影響及全部變數模型下對企業儲蓄率變動具有顯著影響。 本研究的貢獻在於確立企業儲蓄與景氣循環變數之間具有長期/共整合關係，而落後一期的實質GDP成長率對企業儲蓄率有顯著負向關係更證實了企業儲蓄是企業用來作為因應景氣循環的調節工具，因而對企業儲蓄率的上升能提出合理解釋，並可期待未來隨景氣復甦，企業儲蓄率將會有適度的下降調整。

Corporate saving plays an important role in the aggregate economic activity. According to previous research, various factors may affect corporate saving, such as regulations, corporate-bank relationship, etc. Among all, this study focuses on how business cycles could have an impact on corporate saving. In order to understand the fluctuation of corporate saving, this study investigates the relationship between corporate saving and the following explanatory variables: annual growth rate of Monetary Aggregates and CPI, and growth rate of real GDP, these three variables are traditional regarded as important indicators of business cycle. We accept ARDL approach with co-integration analysis and perform robustness check with some control variables: i) government consumption rate, ii) transfer payments rate, iii) labor share, and iv) “Asian Financial Crisis” dummy variable that reflects corporate hedging against tail-risk. Empirical result show that business cycle variables are related to corporate saving, e.g. lag one period of growth rate of real GDP and lag two period of annual growth rate of Monetary Aggregates and CPI are both significantly negatively related to corporate saving. Robustness checks indicate that, (i) the business cycles related explanatory variables are stable; (ii) government consumption rate has significant negative effect on corporate saving. Finally, under the ECM within the ARDL framework, all the error correction terms are negative, indicating that the feedback mechanism is very effective.