Spillover Effect of the Global Financial Crisis on Turkey

Abstract

Purpose – The 2008 Global Financial Crisis took place in the USA and impacted the entire world to different degrees. Basically, the aim of this study is basically to analyse the possible mean, symmetric and asymmetric volatility spillover. Design/methodology/approach – The data is considered as the whole period, the pre-crisis period(2003-2008) and the post-crisis period(2008-2013). A particularly narrow range (2003-2013) is chosen for data analysis. Because when the sample range is kept too wide, other noise factors other than the crisis increase and the results may be biased. In the first part of the application, mean spillover and symmetric volatility spillover are analysed with GARCH models. In the second part of the application, the data is examined with GJR-GARCH models to detect possible asymmetric volatility spillovers. Findings-According to the findings of the research, mean spillover is detected in the post-crisis period and throughout the entire period according to the GARCH model. Again, positive volatility spillover is detected in the post-crisis period and throughout the period. According to the GJR-GARCH model findings, mean spillover is larger and more significant in the post-crisis period compared to the entire period. In addition, a negative volatility spillover is determined in the entire period with 5% significance. A positive and significant leverage effect is also found in the entire period and post-crisis period. Discussion-For periods that are significant in terms of spread in mean and volatility, it can be claimed that past movements in the US stock market are preliminary indicators of the fluctuation in mean and volatility in the local market for traders, investors and analysts in Turkey. It can also be stated that weak form efficiency can be rejected in Turkey for periods where there is a spread in the mean and volatility.