根據經常帳現值理論模型，經常帳餘額由預期未來國民現金流量的變動決定，兩者之間有反向關係。本文以台灣與南韓為研究對象，藉由因果關係檢定、最適化均衡條件檢定與向量自我迴歸模型之係數檢定，來驗證經常帳現值模型的實證配適能力。研究結果顯示，台灣過去經常帳序列Granger causes當期國民現金流量變動，南韓則無；最適化均衡條件與VAR之係數限制式檢定上，台灣資料均不拒絕虛無假設，南韓則皆強烈拒絕各條件限制式之虛無假設。因此根據以上結果，本文可以判定經常帳現值模型對台灣資料的配適度優於南韓資料。 The present value model of the current account states that current account (CA) balances is inversely determined by the discounted value of expected changes in future national cash flows. This article empirically investigates how well the theory fits the data of Taiwan and Korea by conducting three sequential tests, including the causality test of a forward-looking CA, the orthogonality test of CA forecasting errors, and the theory-based cross equation restriction test on VAR coefficients. The results show that the past current accounts Granger causes the national cash flows in Taiwan but not in Korea. The Taiwan data also supports the present value model via the orthogonality and cross equation restriction tests, while Korea again fails. We conclude that the theory is useful in describing the determination of current accounts in Taiwan although Korea may require alternative models for the task.
