Testing the Conventional and Islamic Financial Market Contagion: Evidence from Wavelet Analysis

A major issue facing the investors in the financial markets of the contemporary world is to identify whether the observed stock market fluctuations are due mainly to contagion or fundamentals. This is due to the fact that if the fluctuations are mainly due to a contagion, then it is something like a ‘virus’ which would disappear after a few days. In contrast, if the fluctuations are mainly due to fundamentals, then it is something like a ‘pneumonia’ which is likely to continue for a long time. This study is the first attempt at testing whether there has been any contagion among the Shari’ah-compliant stock indexes during the most recent international financial crisis: the US subprime crisis of 2007-2009 and the Lehman Brothers collapse in 2008, with the application of a time-frequency decomposition technique known as ‘wavelet approach’ both in discrete and continuous forms recently imported to finance from engineering sciences. We analyze the daily data covering the period from June 2005 to December 2011 for the 18 MSCI conventional and Islamic stock market indexes of the Islamic (Malaysia, Indonesia, Turkey, GCC ex-Saudi) and non-Islamic countries (Japan, China, Korea, Taiwan and Hong Kong). Our study is focused on investigating the following empirical question: are the co-movements of selective stock markets normal (interdependent) or excessive (contagious) during the first and the second wave of the financial crisis? Our findings based on the time-frequency decomposition of wavelet approach are as follows: i) The wavelet correlation analysis indicates that, there is no clear evidence of contagion at any time-scales during the subprime mortgage crisis in USA; ii) However, during the collapse of Lehman Brothers, in all conventional stock indexes of non-Islamic countries except Japan, the wavelet correlation coefficients changed significantly at the first time-scale implying that there was a clear evidence of contagion at the first time-scale iii) But the conventional stock indexes of Islamic countries did not suffer from contagion excepting Indonesian MSCI stock index due to overlapping of confidence intervals iv) In all Islamic stock indexes in both Islamic and non-Islamic countries excepting China and Hong Kong Islamic MSCI indexes, the wavelet correlation coefficients changed insignificantly at all scales. Therefore, we fail to reject the null hypothesis implying that there was no clear evidence of contagion at all time-scales excepting China and Hong Kong. Additionally, for robustness, we studied the dynamic correlations between two continuous wavelet transforms through wavelet coherency analysis. The findings whether the stock comovements triggered by the financial crises were showing interdependence or contagion are plausible and intuitive and have implications for both the conventional and Shari’ah-compliant stock markets in terms of asset allocation strategy of risk managers and for policymakers’ optimal policy response to crises.