||This dissertation respectively uses 1978-2004 and 1982-2010 annual data of China and Asia-Pacific countries as the empirical basis and employs three different empirical models to explore the relationship between financial development and economic growth. First, we perform a panel regression and take 31 provinces , eastern, central, and western regions of China as the discussion subjects. We apply the 3SLS panel method and a 3-year rolling average of this method to examine the linear relationship between economic growth and financial development. During the 1978-2004 period the following findings that have positive influences on economic growth: (1) the panel of 31 provinces for the growth rate of ratio of domestic deposits to GPP and domestic loans to GPP; (2) the two panels of eastern and central regions in China for the growth rate of ratio of domestic deposits to GPP. Therefore, the financial development indicator of the growth rate of ratio of domestic loans to GPP in the western region has a negative influence on economic growth. Moreover, we conduct a robust test by setting up a dummy variable in 1989 and gain a same result. |
Due to the 3SLS panel method, we ignore the dynamic correlation between variables. To solve the above problem, we adopt the Wavelet-based measure of comovement, incorporating different time and different frequency domains in the model and re-examine the correlative analysis of the three regions of China between the two variables. According to the Wavelet-based measure of comovement in Rua (2010), after inspecting our findings from different frequency domains, we note that the positive comovement effect between financial development (domestic deposits to GPP and domestic loans to GPP) and economic growth is more significant in the long run than in the short run. The evidence for a short-run effect showing a negative correlation since 1989 is likely due to the impact from the Tiananmen Square incident. Furthermore, inspecting our findings from different time domains, whether in the first or the fourth term of China’s financial reforms, the direct effects released from the two indices of financial development show positive influences on economic growth. Furthermore, as well by extending the sample time 1978-2008 to re-perform a robust test, we find that in domestic loans to nominal GPP and economic growth, the results of the nation, the east and west region are all the same, apart from the central region, especially the domestic deposits to nominal GPP and economic growth support that the former results hold robust evidences.
We next follow the framework of the Wavelet-based measure of comovement from the Wavelet coherency analysis in Aguiar-Conraria and Soares (2011) to examine the coherency and phase difference in the relationship between economic growth and financial variables in 12 Asia-Pacific countries from 1982 to 2010. We observe the comovement phenomenon among variables and the causality of a lead-lag relationship. The empirical results identify the existence of a significantly positive comovement in the short run (1-4 years’ frequency) in the growth-finance relationships for all 12 countries. For nine countries (Singapore, South Korea, Malaysia, Indonesia, Philippines, Thailand, and Fiji, Samoa, Papua New Guinea) in their localized periods the wavelet model shows causality in that economic growth leads financial development. In addition, in 6 out of the 12 countries, which is Japan, New Zealand, Australia, Thailand, Malaysia, Samoa, a growth-finance relationship is found to have a significantly positive comovement in the long run (5-8 years’ frequency). And these six countries in their localized periods that the wavelet model shows a causality relationship in that financial development leads economic growth, hence, the paper has verified that monetary neutrality are not supported. Finally, we note that the causality relationships go in both directions in the long run and short run and observed the difference-phase for developed countries is more significant than for developing countries. This is the first paper to apply the time-frequence wavelet-based measure of comovement and wavelet coherency analysis to study the relationship between financial development and economic growth in China and 12 Asia-Pacific countries. Moreover, it systematically considers the comovement, the extent of correlation, and the long-run and short-run dynamic correlative paths under different time-frequency domains between financial indicator variable and economic growth, offering new contributions to domestic empirical studies in this field.