Modelling Volatility Spillover Effects Between Developed Stock Markets and Asian Emerging Stock Markets

• Author: David E. Giles,Yanan Li
• DOI: http://doi.org/10.1002/ijfe.1506
• Published date: 01 March 2015
• Date: 01 March 2015

MODELLING VOLATILITY SPILLOVER EFFECTS BETWEEN

DEVELOPED STOCK MARKETS AND ASIAN EMERGING STOCK

MARKETS

YANAN LI and DAVID E. GILES*

,†

Department of Economics, University of Victoria, Victoria, BC, Canada

ABSTRACT

This paper examines the linkages of stock markets across the USA, Japan and six Asian developing countries: China, India, In-

donesia, Malaysia, the Philippines and Thailand over the period 1 January 1993 to 31 December 2012. The volatility spillover is

modelled through an asymmetric multivariate generalized autoregressive conditional heteroscedastic model. We find significant

unidirectional shock and volatility spillovers from the US market to both the Japanese and the Asian emerging markets. It is also

found that the volatility spillovers between the US market and the Asian markets are stronger and bidirectional during the Asian

financial crisis. Further, during the last 5 years, the linkages between the Japanese market and the Asian emerging markets be-

came more apparent. Our paper contributes to the literature by examining both the long-run and the short-run periods and fo-

cusing on shock and volatility spillovers rather than return spillovers, which have been the primary focus of most other

studies. Copyright © 2014 John Wiley & Sons, Ltd.

Received 24 September 2013; Accepted 5 November 2014

KEY WORDS: volatility; spillovers; stock markets; multivariate GARCH; asymmetric BEKK model

1. INTRODUCTION

The increasing economic integration of international stock markets has become especially important over the last

two decades. The substantial development of technology and the increased flow of capital between countries are

among the main factors contributing to this observed globalization. So understanding the nature and extent of link-

ages between different financial markets is important for portfolio managers and financial institutions. The volatil-

ity of returns is often used as a crude measure of the risk of holding financial assets (e.g., Brooks, 2002), so wh en

referring to international equity markets integration, researchers not only investigate returns causality linkages, but

they also measure volatility spillover effects. Information about volatility spillover effects is useful for the applica-

tion of value at risk and hedging strategies.

Recently, with the role of emerging markets becoming more important, economists have not only focused on

developed countries (e.g., Bae and Karolyi, 1994; Karolyi 1995; Theodossiou and Lee, 1993), but they have also

paid attention to emerging markets (e.g., Goetzmann et al., 2005; Lin and Wu, 2006; Ng, 2000; Wang et al., 2004;

Worthington and Higgs, 2004). For instance, the extent of the linkages between emerging stock markets and devel-

oped stock markets has implications for investors in both developing and developed countries. If emerging finan-

cial markets are only weakly integrated with their developed counterparts, external shocks may have limited

influence on the emerging markets, and then the developed market investors can benefit by including the emerging

market stocks in their portfolio, as this diversification should reduce risk. On the contrary, if the emerging stock

markets are fully integrated with the developed stock markets, the volatility in the emerging markets may decrease

as it will be mainly determined by the developed markets’volatilities, and the domestic emerging investors will

benefit from a low cost of capital (Li, 2007).

*Correspondence to: David E. Giles, Department of Economics, University of Victoria, Victoria, BC, Canada.

†

E-mial: dgiles@uvic.ca

Copyright © 2014 John Wiley & Sons, Ltd.

International Journal of Finance & Economics

Int. J. Fin. Econ. 20: 155–177 (2015)

Published online 1 December 2014 in Wiley Online Library

(wileyonlinelibrary.com). DOI: 10.1002/ijfe.1506

Thanks to recent developments in econometrics, and the associated software econometrics software,

1

in addition

to examining returns spillovers between equity markets, we can now also estimate volatility spillovers between dif-

ferent markets by using multivariate generalized autoregressive conditional heteroscedastic (MGARCH) models.

We take advantage of these developments by examining linkages between some emerging Asian stock markets

and two developed stock markets. The six emerging Asian markets used in this study are China, India, Indonesia,

Malaysia, the Philippines and Thailand.

2

We use the USA and Japan to represent developed countries in two dif-

ferent geographical regions. We employ an asymmetric BEKK

3

model proposed by Engle and Kroner (1995) and

improved by Kroner and Ng (1998) to examine both shock and volatility spillovers among each emerging market

and the two developed markets. Our data sample spans 20 years and includes both the 1997 Asian financial crisis

and the 2007 subprime financial crisis.

Although the empirical finance literature is rich in studies focusing on the transmissions and dynamic

linkages between major stock markets, our study distinguishes itself from these in three major respects. First,

instead of analysing transmissions only between developed stock markets, we examine the linkages between

the developed markets and Asian emerging markets. Second, we not only examine the long-run relationship

between different markets but also compare the results with different sample periods, and the two short-run

periods are chosen on the basis of two recent financial crises. Third, and importantly, we consider past shock

and volatility spillovers across different markets, whereas the previous literature focuses on returns

transmissions.

The remainder of this paper is organized as follows. The next section introduces the background of markets’

indices and provides some preliminary statistical analysis of our data. Section 3 outlines the methodology used

to analyse the volatility spillover effects, and the empirical results are discussed in Section 4. Finally, Section 5

summarizes and makes concluding remarks.

2. DATA ANALYSIS

We used widely accepted benchmark indices for the eight selected countries, with each index describing the overall

performance of large-capitalization firms in the respective country, taken to represent the overall equity market in

that country. The six Asian emerging stock market indices that we consider are the following: Shanghai Stock Ex-

change composite (CHSCOMP) for China; S&P BSE 30 SENSITIVE (IBOMSEN) for India; IDX composite

(JAKCOMP) for Indonesia; FTSE BURSA MALAYSIA KLCI (FBMKLCI) for Malaysia; PSE composite

(PSECOMP) for the Philippines; and BANGKOK SET 50 (BNGKS50) for Thailand. The two developed market

indices are the S&P 500 composite (S&PCOMP) for the USA; and TOPIX (TOKYOSE) for Japan. The daily data

are obtained from DATASTREAM.

4

The full sample period under study is from 1 January 1993 to 31 December 2012, except for Thailand

5

be-

cause of data limitations. The reason for selecting this period is that both the Asian financial crisis and the

subprime financial crisis are covered, so in addition to investigating the volatility performance over the long

run, we can also explore dynamic linkages across markets over the short run or over an economic recovery

period.

There are 5217 observations in our sample. By using daily data, we can capture more information than with

weekly or monthly data. In this study, we set prices on non-trading days to be the same as on the previous trading

day.

6

Also, our markets are located in different time zones, resulting in different opening and closing times, as is

shown in Figure 1 for two consecutive days.

Figure A1 in the Appendix shows the adjusted closing prices of our eight indices. Each index has a trough dur-

ing 1998 when the Asian financial crisis happened. The troughs in Japan, Indonesia, Malaysia, the Philippines and

Thailand are more obvious than those in the USA, China and India. However, when the US subprime financial cri-

sis occurred in 2007, each index has a clear trough. Visually, the markets’integrations are more apparent over the

last decade than during the 1990s.

The continuously compounding daily returns for each stock market are expressed in percentages, computed by

multiplying the first difference of the logarithm of the market closing value of the index by 100. Figure A2 in the

Appendix shows the returns series for each index. These series exhibit the typical ‘volatility clustering’of high-

YANAN LI AND DAVID E. GILES156

Copyright © 2014 John Wiley & Sons, Ltd. Int. J. Fin. Econ. 20: 155–177 (2015)

DOI: 10.1002/ijfe