Structural breaks, energy consumption, and GDP in Taiwan：Revisited

• Main Authors: Ching-Chiang Cheng, 鄭清江
• Other Authors: Ho-Chyuan Chen
• Format: Others
• Language: zh-TW
• Published: 2006
• Online Access: http://ndltd.ncl.edu.tw/handle/72940669140473987008

碩士 === 國立高雄第一科技大學 === 金融營運所 === 94 === Abstract Departing from previous studies on the causal relationship between energy consumption and economic growth, this paper tries to investigate a series of unit root and causality tests to detect causality between the real GDP, energy consumption and CPI in Taiwan employing the unit root tests allowing for structural breaks and the Johansen’s multiple cointegration tests for the 1959-2004 periods. Secondly, by using a dynamic vector error-correction model, we then analyze the direction of Granger causation. Thirdly, the relative strength of the causality is gauged (through the dynamic variance decomposition technique) by decomposing the total impact of an unanticipated shock in the other variables, include its own, in the multivariate system. Finally, these response paths of shocks to the system are traced out using impulse response graphs. Our main findings are: First, the conventional unit root tests indicate the energy consumption and CPI are I(1) stationary variable but the real GDP is I(1) or I(2), whereas the endogenous break unit root tests proposed by Zivot and Andrews (1992) and Perron (1997) reveal that all variable are I(1) stationary with a structural break. Therefore, it is inappropriate to take the second difference of the real GDP to achieve stationary. Second, we find that different directions of causality exist between GDP, iii energy consumption and CPI. The empirical result shows unanimously in the long run that the causality runs from real GDP to energy consumption and CPI without feedback and it exist no causality between real GDP and energy consumption in the short run. It implies energy conservation may not harm economic growth. Third, shocks to the system seemed to have had a more sustained if not pronounced effect in Taiwan. Overall, we do find the structural breakpoints, and they look to match clearly with the corresponding critical economic incidents.