Edhec-Risk
Features
Volatility - October 23, 2013

The Local Volatility Factor for Asian Stock Markets

Industry surveys indicate growing investments in Asian passive investment equity products commonly motivated by the expected economic growth of the region relative to the rest of the world and the resulting equity premium. Investors are typically interested more in regional and country cap-weighted indices and ETFs tracking such indices rather than sector or style indices (see Amenc et al. (2012)).

On the other hand, investors and asset managers increasingly use OTC or exchange-traded volatility derivatives using volatility indices as underlyings to alleviate losses during market downturns, based on the negative correlation between equity returns and volatility which has been well-documented in the academic literature. From an investor perspective, the negative correlation presents hedging and diversification opportunities. In addition, negative correlation and high volatility are particularly pronounced in stock market downturns, offering protection against stock market losses when it is most needed and when other forms of diversification do not provide very effective exposure.

Although the market for volatility derivatives in Asia is still immature, it has been slowly developing. Hong Kong and Japan launched implied volatility futures contracts in 2012 and Australia and Korea have revealed plans to launch such contracts in the near future. However, given that US VIX is the most popular implied volatility index with VIX futures easily available, an important question is whether an exposure to a local modelfree option-implied (MFOI) volatility indicator has better hedging properties than an exposure to US VIX. On one hand, the academic literature finds empirical support for the presence of local volatility factors in Asian equity markets, implying that investors would be better off with an exposure to a local MFOI volatility index. On the other hand, there is also evidence of spillover effects from the US to the Asian markets suggesting that in times of market turbulence an exposure to VIX can reduce risk.

In contrast to most of the academic studies which are model dependent (e.g. GARCH-type volatility models or vector-auto-regressive models), we study these questions directly with MFOI volatility indices. In Asia, there are three such indicators with more than 9 years of history the Hong Kong (VHSI), the Japanese (VNKY), and the Korean (VKOSPI) indicators. MFOI volatility indices are available in India and Australia but have insufficient history. Our approach is to compare the hedging properties of hypothetical exposures to these three Asian MFOI volatility indices to the hedging properties of an exposure to VIX assuming an exposure to an Asian country-specific or regional equity index. We include in the study 12 local markets and 9 regional equity indices.

As far as hedging costs are concerned, we do not compare the costs of achieving the volatility exposures. In any event, the volatility futures markets in Hong Kong and Japan are rather illiquid and any such comparison would favour VIX strongly. Our main objective is to verify if an exposure to a local MFOI volatility index provides fundamental benefits to investors which could be a driver of higher demand for volatility derivative products in the future.

The empirical analysis in the paper is based on both conditional in-sample and out-of-sample analysis. We find strong evidence for a very significant local volatility factor in the Asian market index returns captured by VHSI, VNKY, and VKOSPI. In particular, stand-alone and multivariate analysis reveals that the relationship between the Asian equity index returns and the aforementioned MFOI volatility indices is significantly stronger than the relationship between Asian equity index returns and VIX. The multivariate analysis suggests either a weaker or insignificant relationship between the Asian equity market returns and the US VIX in the presence of Asian volatilities, implying that the Asian volatility indices can absorb the information content of the VIX.

Our conclusions are even stronger during the financial crisis of 2008 (the sample being from September 2008 to February 2009) which is surprising bearing in mind the spillover effects reported in the academic literature. A model selection algorithm shows a preference for an exposure to Asian MFOI volatility indices rather than an exposure to VIX which, in spite of all deficiencies of step-wise selection, does indicate a strong presence of an Asian volatility factor.

In addition to the in-sample analysis, we carry out an out-of-sample analysis with two hedging horizons of 5 and 15 days, which correspond to the average holding period of VIX futures, in two versions using daily and weekly data for parameter estimation. The out-of-sample analysis confirms the conclusions from the insample analysis for both horizons.

There are two implications of this work for investors with an exposure to Asian equity markets and for Asian volatility derivative markets. Firstly, an exposure to VIX is efficient most likely in very brief periods around significant volatility spillovers from the US market. Generally, an exposure to VIX for hedging an exposure to Asian equity markets is significantly less efficient than an exposure to an Asian MFOI volatility indicator. Secondly, and in relation to the previous point, from a practical perspective constructing an exposure to an Asian MFOI volatility indicator is presently very difficult because of the limited liquidity of the volatility futures market. Nevertheless, the results suggest that investors can greatly benefit in the future as the market grows further and liquidity increases, especially if their horizon is from a few days up to one week.


URL for this document:
http://www.edhec-risk.com/features/RISKArticle.2013-10-23.4808

Hyperlinks in this document: