Global Volatility Transmission and Portfolio Management: The Case of Saudi Arabia

The Saudi stock market’s performance has changed considerably over the past two decades, thus strengthening this market’s position in not only the Gulf Cooperation Council (GCC) region but also the Arab world. The GCC countries’ financial development and Saudi Arabia’s economic growth have also influenced this market. The market index, termed Tadawul All Share Index (TASI), has changed through many economic and financial crises resulting from globalisation. The Saudi stock market that commenced in 1985 is a major market in the Middle East and North Africa region, and since it has been maturing, it is defined enough to be assessed empirically. Therefore, this thesis explores the changing patterns of the Saudi stock market volatility over time, the effects of volatility pattern changes on optimal portfolio choices and the evolution of these choices. Significantly, given the increasing globalisation and financial integration in recent years, this thesis tests a new hypothesis that the Saudi stock market’s volatility is affected by global volatility spillovers (i.e. the volatility in global markets, in commodity markets during the sample periods, including the 2008 global financial crisis [GFC] and the 2014–2016 oil price shocks) with important outcomes for portfolio management. To the best knowledge of the author, this type of comprehensive analysis that offers insights on volatility origin, although critical and novel in the finance and portfolio management fields, has not been undertaken for the purpose of policy analysis in the literature and, in particular, for Saudi Arabia. The research uses advanced econometrics: cross-correlation function (CCF) and multivariate generalised autoregressive conditional heteroscedasticity (GARCH) models (i.e. Baba, Engle, Kraft and Kroner [BEKK], constant conditional correlation [CCC] and dynamic conditional correlation [DCC] models) and high-frequency daily data for 2007– 2018. The multivariate models are reliable and effective for addressing volatility transmission interactions and correlations. These models are flexible in revealing the changes in conditional variance and covariance and simplify the estimation process. These approaches are appropriate and significant in volatility origin studies. In addition, Kroner and Ng’s approach is used to determine the optimal weight and Kroner and Sultan’s approach to determine the optimal hedge ratio. The major findings and contributions of this thesis to the volatility origin area of finance and portfolio management are summarised as follows: (i) This empirical research examines the causality relationship of the TASI with global stock markets and major commodity markets during the full and crisis/shocks periods, using daily data for determining the causality in variance movements of the TASI and stock/commodity pairs. The CCF test shows that overall, the current relationship between TASI and the global stock markets is significant but finds no link between TASI and the commodity markets for all periods except the full period in which it is obvious that TASI is unidirectionally causal to crude oil. This thesis documents bidirectional causality in variance in global stock markets between TASI and some global stock markets over certain periods. (ii) In the second empirical analysis, the Saudi stock market’s trading partner stock markets and major commodity markets are analysed to ascertain volatility transmission and correlations with TASI. This analysis, based on the estimation results of the multivariate GARCH models (BEKK, CCC and DCC), shows that the volatility transmission effect and the conditional correlation behaviours in the full and crisis/shock periods differ. The findings show that transmission channel of shock and volatility spillover varies from market to market over the three periods and that TASI also reacts differently over the three periods. Over the GFC and oil decline periods, some global equity markets influenced TASI volatility to varying degrees, as revealed by the statistically significant model coefficients. This thesis documents that TASI is more integrated with global stock markets throughout the periods of GFC and oil decline. In addition, the results show highly positive conditional correlation at least at the 5% significance levels between TASI and the international stock markets during the three sample periods. (iii) The thesis findings for optimal portfolio weights and hedge ratios recommend that portfolio risk can be minimised without reducing portfolio efficiency by combining some indices of global stocks and commodities into a well-diversified portfolio with TASI. Since international portfolio diversification has become more popular worldwide, investors, portfolio managers and policymakers can apply this information to make new financial investments, recommend cautious financial regulations and implement quick, efficient policy tools.