台灣菲利普曲線非線性之實證研究

Abstract

2008-2009年金融海嘯發生後全球許多先進的經濟體出現了 “Missing Disinflation”，也就是通貨緊縮消失的局面，面對高失業率，由菲利普曲線估計出的通膨率應該要更低，然而當時的通膨率變化卻是相對穩定的。並且近年來許多文獻指出在歐洲等國家菲利普曲線的斜率發生改變，他們實證發現相較於傳統的單一斜率線性模型，使用非線性的模型更具解釋力。因此本文使用門檻迴歸模型 (Threshold Regression Model)，找出台灣菲利普曲線斜率發生改變的門檻值，觀察該門檻值前後曲線的斜率變化，並討論台灣的菲利普曲線是否也有 “Missing Disinflation” 的情形。另外，Coibion and Gorodnichenko (2015) 指出利用密西根大學消費者信心指數可以合理解釋 “Missing Disinflation ”，Doser and Nunes (2017) 也實證發現預期通膨中消費信心比起其他通膨預期項目佔有較大的比重。因此本文也納入由台經中心所調查出的消費者信心至預期通膨項目。本文實證發現，儘管台灣的資料消費者信心在預期通膨的占比不大，卻可以改善整個模型的解釋力。並且在大部分的模型設定中，台灣的菲利普曲線確實有 “ Missing Disinflation ” 的情形。

Since the financial crisis of 2008-2009, many advanced economies have experienced little decline in inflation , a phenomenon known as the “Missing disinflation” - as the unemployment rate reached high levels, inflation did not decline nearly as much as a linear Phillips curve predicted. Coibion and Gorodnichenko (2015) argue that using household inflation expectations as measured by the Michigan Survey of consumers, can account for the absence of strong disinflationary pressures since 2009, and Doser and Nunes (2017) prove that regressions which include both household and professional forecasts systematically point to a larger role for household forecasts than any other measure of inflation expectations, so we put Taiwan Consumer Confidence Index into the model and find that even Taiwan Consumer Confidence isn’t a dominant component of inflation expectations, it truly improve the explanation power of the model. In this paper, we use nonlinear model – Threshold Regression Model to see how the slopes of the Taiwan Phillips Curve has changed, and found that in many cases Phillips Curve of Taiwan existed “Missing Disinflation”.