Abstract
Monetary policy is a topic of enduring interest. This is partially due to the financial liberalisation that increases the volume of international capital flows and the influence of foreign monetary policy on the domestic economy. Additionally, the growth of Islamic banks has created interest in examining the role of monetary policy in these relatively new financial institutions. This dissertation examines the impact of these factors in different countries that have been exposed to these issues. The first essay (Chapter 2) explores the dynamic relationships that foreign direct investment, non-resident Indian deposits and portfolio investment have with three monetary transmission mechanisms (MTMs), which are the asset price channel, the exchange rate channel and the bank lending channel. Impulse response functions (IRFs) are simulated from a structural vector autoregressive (SVAR) model to fulfil this objective. Monthly data from May 1998 to March 2011 were analysed. Capital inflows were found to have limited influence on MTMs in India. Moreover, the large error bands of the IRFs at longer horizons suggest that the reactions of macroeconomic variables to monetary policy shocks in India have been unstable, which is plausible in a transitional economy. The findings imply that enhancing the depth of the Indian stock market could improve the effectiveness of monetary policy. Lastly, capital controls and macro-prudential policy could be effective in protecting the process of monetary policy in India from foreign influences.
The second essay (Chapter 3) examines the Islamic bank lending channel in Malaysia and compares the results with estimations of conventional data. An unbalanced panel dataset from nine banking entities, ranging from 2005 Q2 to 2012 Q4, is estimated. All banks are publicly listed on the stock exchange in Malaysia. The dataset is estimated using the fixed effect or least square dummy variable analysis. The results indicate that GDP and inflation are positively and negatively linked to loan supply from Islamic and conventional banking operations, respectively. In addition, relationship banking encourages more lending by Islamic banking operations to borrowers during an economic crisis. The estimations cannot provide unanimous evidence to support the existence of Islamic and conventional bank lending. Indeed, the positive coefficient of the expected real monetary policy rate in the estimation of Islamic data suggests that the structure of the banking industry affects the impact of the interest rate on the Islamic loan supply. Equally importantly, Islamic banks are found to be more sensitive to changes in real GDP and financial crisis, suggesting that Islamic finance is not stable compared with conventional finance. Additionally, the role of bank liquidity and bank capital in lending behaviour are under the influence of market regulations, binding liquidity, capital requirements and risk aversion.
The third essay (Chapter 4) investigates the impact of world outputs and shocks, world price shocks, world interest rate shocks and global liquidity shocks on six domestic economic variables in New Zealand. Tests are conducted on quarterly data from 1991 Q1 to 2013 Q1, and apply SVAR to simulate IRFs. The estimations are conducted in two stages. The first stage generates the structural foreign shock using a SVAR model that consists of foreign variables; the second stage involves the SVARs that include domestic variables and the structural foreign shocks that were generated from the first stage. The findings show that some domestic variables react to the remaining foreign shocks, apart from world interest rate shocks. Among these domestic variables, the real exchange rate is found to be relatively sensitive to foreign shocks. The insignificant impact of the world interest rate also implies that the hypothesis of ‘beggar-thy-neighbour’ is not supported in this chapter. Furthermore, global liquidity has, in general, positively influenced the real exchange rate. Finally, the IRFs from the robustness check indicate that the periods and sizes of statistically significant reactions are similar. However, the statistically significant impacts of these foreign shocks are different in some SVAR models.