貨幣政策對本國產出與物價之影響－向量自我回歸模型符號限制法之應用

本研究旨在使用向量自我回歸模型並搭配Uhlig(2005)符號限制法，來認定本國的緊縮性貨幣政策衝擊，主要藉由觀察兩項央行政策主要影響目標，即為產出與物價對貨幣政策衝擊之反應，來檢視認定貨幣政策衝擊的方式是否正確。本研究把資料期間大致拆成早期與近期，並首先跟隨Ho and Yeh(2010)的符號限制認定方式，使用利率與外匯存底來認定貨幣政策的衝擊，發現原本1981年至2005年較顯著且有效的符號限制認定方式應用到1981年至2018年與1994年至2018年的資料期間時，得出了貨幣政策衝擊對於產出與物價的影響是較不顯著且效果短暫。本文嘗試使用貨幣總計數M2當作符號限制之認定方式，並將之使用於1981年至2018年與1994年至2018年的資料，最後得出貨幣政策衝擊對於產出與物價的影響是較顯著且持續；且若將之使用於1981年至2005年資料時，其結果亦不遜於原本使用利率與外匯存底的情況。本研究亦將多種不同的符號限制認定方法進行比較，最後仍發現使用貨幣總計數M2作為符號限制來認定貨幣政策衝擊時其結果最佳。

關鍵字

貨幣政策；產出與物價；向量自我回歸模型；符號限制法

並列摘要

The purpose of this paper is implementing vector autoregression model with Uhlig (2005) sign restrictions to identify monetary policy shock in Taiwan. By observing the responses of output and price to monetary policy shock, we can examine whether the way in which monetary policy shocks are identified is right or not. This study roughly divide the data into early and recent period. We followed Ho and Yeh (2010) identification scheme, using interest rate and foreign reserve to identify a contractionary monetary policy shock. The empirical results show that insignificant and temporary decline in real gross domestic product and price level by implementing Ho and Yeh (2010) identification scheme to the first quarter of 1981 to the last quarter of 2018 and the first quarter of 1994 to the last quarter of 2018. This study try other identification scheme, using broad money to identify a contractionary monetary policy shock. Our findings show that significant and permanent decline in real gross domestic product and price level by using broad money identification to the first quarter of 1981 to the last quarter of 2018 and the first quarter of 1994 to the last quarter of 2018.We implement it to the first quarter of 1981 to the last quarter of 2015. The results are not inferior to Ho and Yeh (2010) identification scheme. This study also compares many different identification schemes, and we find that using broad money to identify a contractionary monetary policy shock is more optimal in Taiwan.