International Journal of Islamic and Middle Eastern Finance and Management Emerald Article: Financial market risk and gold investment in an emerging market: the case of Malaysia Mansor H. Ibrahim Article information: To cite this document: Mansor H. Ibrahim, (2012),"Financial market risk and gold investment in an emerging market: the case of Malaysia", International Journal of Islamic and Middle Eastern Finance and Management, Vol. 5 Iss: 1 pp. 25 - 34 Permanent link to this document: http://dx. doi. org/10. 1108/17538391211216802 Downloaded on: 26-09-2012
References: This document contains references to 13 other documents To copy this document: permissions@emeraldinsight. com This document has been downloaded 335 times since 2012. * Users who downloaded this Article also downloaded: * Mohamed Hisham Yahya, Junaina Muhammad, Abdul Razak Abdul Hadi, (2012),"A comparative study on the level of efficiency between Islamic and conventional banking systems in Malaysia", International Journal of Islamic and Middle Eastern Finance and Management, Vol. 5 Iss: 1 pp. 48 - 62 http://dx. doi. org/10. 1108/17538391211216820
Muhamad Abduh, Mohd Azmi Omar, (2012),"Islamic banking and economic growth: the Indonesian experience", International Journal of Islamic and Middle Eastern Finance and Management, Vol. 5 Iss: 1 pp. 35 - 47 http://dx. doi. org/10. 1108/17538391211216811 Samy Nathan Garas, (2012),"The control of the Shari'a Supervisory Board in the Islamic financial institutions", International Journal of Islamic and Middle Eastern Finance and Management, Vol. 5 Iss: 1 pp. 8 - 24 http://dx. doi. org/10. 1108/17538391211216794 Access to this document was granted through an Emerald subscription provided y ASSUMPTION UNIVERSITY OF THAILAND For Authors: If you would like to write for this, or any other Emerald publication, then please use our Emerald for Authors service. Information about how to choose which publication to write for and submission guidelines are available for all. Please visit www. emeraldinsight. com/authors for more information. About Emerald www. emeraldinsight. com With over forty years' experience, Emerald Group Publishing is a leading independent publisher of global research with impact in business, society, public policy and education.
In total, Emerald publishes over 275 journals and more than 130 book series, as well as an extensive range of online products and services. Emerald is both COUNTER 3 and TRANSFER compliant. The organization is a partner of the Committee on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive preservation. *Related content and download information correct at time of download. The current issue and full text archive of this journal is available at www. emeraldinsight. com/1753-8394. htm Financial market risk and gold investment in an emerging market: the case of Malaysia
Mansor H. Ibrahim Market risk and gold investment 25 Department of Economics, Universiti Putra Malaysia, Serdang, Malaysia Abstract Purpose – The purpose of this paper is to examine the relation between gold return and stock market return and whether its relation changes in times of consecutive negative market returns for an emerging market, Malaysia. Design/methodology/approach – The paper applies the autoregressive distributed model to link gold returns to stock returns with TGARCH/EGARCH error speci? cation using daily data from August 1, 2001 to March 31, 2010, a total of 2,261 observations.
Findings – A signi? cant positive but low correlation is found between gold and once-lagged stock returns. Moreover, consecutive negative market returns do not seem to intensify the co-movement between the gold and stock markets as normally documented among national stock markets in times of ? nancial turbulences. Indeed, there is some evidence that the gold market surges when faced with consecutive market declines. Practical implications – Based on these results, there are potential bene? ts of gold investment during periods of stock market slumps. The ? ndings should prove useful for designing ? ancial investment portfolios. Originality/value – The paper evaluates the role of gold from a domestic perspective, which should be more relevant to domestic investors in guarding against recurring heightened stock market risk. Keywords Malaysia, Emerging markets, Gold, Returns, Investments, Stock markets, Gold investment, Market return, Correlations, Market risk Paper type Research paper Introduction Over the past decades, the global ? nancial markets have witnessed a string of ? nancial crises, among them include the Mexican peso crisis in 1994, the Asian ? nancial ? in 1997/1998, the Russian crisis in 1998, the Brazilian crisis in 1999, the Argentine ? nancial crisis in 2001/2002 and most recently the US subprime crisis in 2007 and the Greece ? nancial crisis in 2009. Mentioning of these crises is likely to conjure up in the mind of many the images of excessive risk in stock market investment and to bring back interest in gold as an alternative investment asset. This interest is well-placed as gold used to be a standard of value, is still considered as a store of value and is universally accepted. Moreover, there seems to be a trong belief that gold can provide protection, as a hedge or a safe haven, against this heightened risk in the ? nancial markets. As noted by Baur and McDermott (2010), gold differs from other assets in that it reacts positively to adverse market shocks. As they mention, real gold value reached its historic high roughly in 1980 when the global economy faced the threat of stag? ation due to oil crises in 1970s. Likewise, at the time the US subprime crisis intensi? ed in September 2008, gold has responded with a surge in its value (Baur and McDermott, 2010). International Journal of Islamic and Middle Eastern Finance and
Management Vol. 5 No. 1, 2012 pp. 25-34 q Emerald Group Publishing Limited 1753-8394 DOI 10. 1108/17538391211216802 IMEFM 5,1 26 Against a backdrop of recurring ? nancial crises and contagion as well as emerging interest in gold, several studies have attempted empirical investigation of gold hedging property. Notable among these studies are recent works by Capie et al. (2005), Hillier et al. (2006), Baur and Lucey (2010) and Baur and McDermott (2010). Capie et al. (2005) investigate an exchange rate hedge of gold using weekly data of gold price and sterling-dollar and yen-dollar exchange rates from January 1971 to February 2004.
They ? nd supportive evidence for exchange rate hedging property of gold, although the strength of hedging tends to vary over time. Hillier et al. (2006) assesses the investment role of precious metals, namely gold, platinum and silver for the US market. They note low correlations between these three metals and stock market returns, which suggests diversi? cation bene? ts of gold investment. Baur and Lucey (2010) examines whether gold is a safe haven, i. e. maintaining its value in times of market stress or turmoil, for the US, UK and German markets.
They document evidence suggesting the ability of gold to hedge against ? nancial risks and to serve as a safe haven in extreme market conditions for these markets. Most recently, Baur and McDermott (2010) extend the work of Baur and Lucey (2010) to a larger number of markets, which include both major developed and emerging markets. They analyze the relations between gold return and returns of world and emerging market indexes, various regional market indexes, and 13 individual market indexes. Their results demonstrate the ability of gold to provide a hedge and a strong safe haven for European and US markets.
Thus, for developed markets, gold provides protection against losses during extreme market conditions. As they explain, investors in these markets sell stocks and buy gold when faced with heightened ? nancial risk. By contrast, the emerging markets seem to lack these properties indicating that investors tend to react differently to adverse shocks in emerging markets. Namely, they shift the composition of their portfolios by selling shares of emerging markets and seeking shelter in the developed markets, which are viewed to be relatively safe.
In the present paper, we take lead from these studies and examine the investment role of gold for an emerging Asian market, Malaysia. We attempt to contribute to this line of inquiry in several aspects. First, in Baur and McDermott (2010), the investment role of gold for emerging markets is examined by looking at the relation between gold return and emerging market index return and individual market returns of four largest emerging markets, i. e. Brazil, Russia, India and China. We add to their study by looking at a smaller emerging market.
Second, while the present study looks at gold investment from an international perspective, we look at the issue from a domestic perspective. All aforementioned studies employ gold price in US dollar in their analysis. Instead of using the dollar-denominated gold price and converting it into domestic currency unit as in Baur and Lucey (2010), we use domestic gold price instead. While we acknowledge that the Malaysian gold price may have depended on the global gold price, the use of gold price quoted domestically in ringgit screens out potential confounding effect of exchange rate movement and currency onversion. Finally, we bring out a new empirical perspective in evaluating the investment role of gold. Namely, we examine whether gold maintains its value or its relation with market returns when faced with consecutive negative daily returns. We focus on Malaysia due to deep interest in gold shown by Malaysian policymakers and academics in the face of 1997/1998 Asian ? nancial crisis. Tun Mahathir Mohamad, the then Prime Minister of Malaysia, voiced interest in this universally accepted asset and proposed the use of gold particularly in international trade settlement The News Strait Times, 2001). A series of international conferences have been organized on the subject of gold and gold Dinar[1], among them include International Conference on Stable and Just Monetary System and International Conference on the Gold Dinar in Multilateral Trade in 2002, International Conference on Gold in International Trade in 2003 and International Conference on Gold Dinar Economy in 2007. In July 2001, Malaysia became the 12th country in the world to have its own gold bullion coins through the launching of the gold bullion coins known as Kijang Emas by the Royal Mint Malaysia.
This is followed by the issuance of Royal Mint gold Dinar in 2003 and Kelantan State gold Dinar in 2006. While the introduction of these gold coins is to serve primarily as a store of value or an alternative ? nancial asset for investment, the gold investment performance for the case of Malaysia has hardly received any empirical attention. The availability of daily domestic gold bullion price since 2001 provides us an opportunity to examine the investment role of gold from a domestic market perspective and, at the same time, widens the literature on emerging markets. The rest of the paper is structured as follows.
In the next section, we provides the empirical framework used in the analysis. Then, we describes the data and present estimation results. Finally, we conclude with the main ? ndings and some concluding remarks. Empirical framework We specify our empirical model using an autoregressive distributed lag model along the line of Capie et al. (2005). Thus, we have: RG;t ? a ? rRG;t21 ? b1 RS;t ? b2 RS;t21 ? 1t ?1? where RG is the daily return of gold investment and RS is the corresponding return of stock investment. The lagged dependent is included to allow for autocorrelation structure in gold return.
Meanwhile, the incorporation of once-lagged stock return is based on our presumption that, in emerging markets, the transmission of information among markets may take time. That is, the changes in stock return may be impounded into the gold return with lag. The total sensitivity of gold return to stock market ? uctuations is based on the sum of stock market coef? cients, i. e. b1 ? b2. If this sum is signi? cantly positive and is far from unity or the model explanatory is close to zero, we may conclude that gold serves as a diversi? cation asset (Hillier et al. , 2006).
Meanwhile, if it is not signi? cant or is signi? cantly negative, then gold investment can provide a hedge against ? nancial market risk (Baur and Lucey, 2010; Baur and McDermott, 2010). We refer to equation (1) as our basic model. Based on equation (1), we ask further whether gold return dynamics remain similar under conditions of consecutive negative market returns. To this end, we adapt the framework used by Nam et al. (2005) in their analysis of stock return asymmetry by modifying equation (1) as: RG;t ? a0 ? a1 Nmt ? rRG;t21 ? ?b10 ? b11 Nmt ? ? RS;t ? ?b20 ? b21 Nmt ? ? RS;t21 ? 1t ?2? here Nmt is a dummy variable representing consecutive negative market returns. Five alternative dummies corresponding to days of consecutive negative returns are considered and they are de? ned as: Market risk and gold investment 27 IMEFM 5,1 N0 ? 28 " N1 ? N4 ? " " 1 if RS;t , 0 0 otherwise 1 if RS;t , 0; RS;t21 , 0 0 otherwise ?3? ?4? . . . 1 if RS;t , 0; 0 otherwise RS;t21 , 0; :::; RS;t24 , 0 ?5? Note that we include Nm as both intercept and interactive dummies. The intercept dummy is intended to capture the level effect of m ? 1 consecutive negative market returns, current return and the returns of last m days, on gold return.
Meanwhile, the interactive dummy is to capture the changing relations between stock return and gold return under conditions of consecutive negative market returns, the main interest of the paper. In the paper, we denote these models with alternative de? nition of dummies, respectively, as model N0, N1, N2, N3 and N4. In equation (2), the sum b10 ? b20 captures the relation between the two markets under normal market conditions while b10 ? b20 ? b11 ? b21 measures their relation when the stock market experiences m ? 1 days of consecutive negative returns. Accordingly, the signi? cance of b11 and b21 re? cts the changing relations between gold return and market return in times of market downturns. If they are signi? cantly positive, then the gold return tends to move in closer tandem to stock market movement, weakening gold investment role as a diversi? cation asset. However, if they are signi? cantly negative, then gold investment is said to provide at least a hedge against ? nancial losses during market downturns. Finally, if they are insigni? cantly different from 0, the dynamics of gold return tends to resist the slumps in stock prices and preserves its relation to the stock market regardless of the market conditions.
We believe that this perspective that we bring provides a nice complementary empirical exercise to the works of Baur and Lucey (2010) and Baur and McDermott (2010) that look at the relations between the two during extreme market conditions. In the implementation of equations (1) and (2), we take note of ample evidence that high-frequency asset returns tend to exhibit leptokurtic property or volatility clustering, the so-called autoregressive conditional heteroskedasticity (ARCH) effect. In ? nance literature, various error distributions have been assumed and variance equation speci? cations have been suggested.
The error distribution is assumed to be distributed according to either the normal distribution (N), t-distribution (T), or generalized error distribution (G). Among the time-varying variance speci? cations include the generalized autoregressive conditional heteroskedasticity (GARCH), threshold ARCH (TARCH), and exponentional GARCH (EGARCH). The latter two allow for asymmetric responses of volatility to positive and negative shocks. To avoid arbitrary model selection, we follow Capie et al. (2005) by basing on the maximum of log likelihood as a selection criterion. We ? nd asymmetric volatility speci? cation (TARCH or EGARCH) to best ? the gold return dynamics and generalized error distribution to best describe the error distribution. The suitability of asymmetric volatility modeling for gold return is in conformity with the behavior of other asset returns (Lobo, 2000; Koutmos and Martin, 2003). Data We employ 2,261 daily observations spanning from August 1, 2001 to March 31, 2010. The beginning date is dictated by data availability of gold bullion price. The selling prices of one troy ounce domestic gold bullion are used to represent domestic gold prices while the Kuala Lumpur composite index is used to represent aggregate prices of stock market investment.
The data on the two prices are sourced, respectively, from Malaysia’s central bank, Bank Negara Malaysia, and Data Stream International. We compute gold and stock market returns as the ? rst difference of the natural log of respective series. Table I provides descriptive statistics of the two returns. We also plot these series in level and ? rst-differenced forms in Figure 1. Both gold and stock prices experience an upward trend over the sample period. While the daily average gold return is relatively higher than the daily average stock market return (i. e. 0. 6 percent against 0. 03 percent), it is more volatile than the market return as re? ected their respective standard deviations. This is accounted by the more extreme positive values of gold return (0. 1246) than the stock market return (0. 0426). Meanwhile, the extreme negative value of stock market return (2 0. 9997) is only slightly higher than the corresponding value of gold return (2 0. 0782). From the plots, we also note marked reduction of stock market prices around years of the Argentine ? nancial crisis in 2001/2002 and of the US subprime crisis in 2007/2008.
While the gold return is positively skewed, the market return demonstrates a negative skewness. Both return series are characterized by excess peakness having kurtosis statistics to be substantially higher than 3. This suggests volatility clustering in the return series, which is apparent in the graphical plots. The Jarge-Bera statistics reported at the bottom of Table I soundly rejects the null of normality for both returns. These characteristics in the data seem to justify the use of GARCH-type models for model speci? cation. As a preliminary analysis, we report the cross-correlations between RG,t and RS,t for up to ? e lags. With the standard error in the order of 0. 021 in absolute value, the correlation of roughly 0. 042 and higher suggests signi? cance correlation between the two returns. We note very low and mostly positive correlations between gold return and contemporaneous and lagged stock returns. Among these correlations, only the DG Mean Median Maximum Minimum SD Skewness Kurtosis Jarque-Bera Probability Observations 0. 000305 8. 72 ? 102 5 0. 042587 2 0. 099785 0. 008518 2 0. 999659 15. 06466 14,082. 94 0. 000000 2,260 29 DS 0. 000561 0. 000000 0. 124645 2 0. 078182 0. 011909 0. 092587 12. 8588 8,656. 123 0. 000000 2,260 Market risk and gold investment Table I. Descriptive statistics IMEFM 5,1 8. 4 0. 15 0. 10 8. 0 0. 05 30 7. 6 0. 00 7. 2 6. 8 –0. 05 02 03 04 05 06 07 08 09 –0. 10 02 03 04 05 06 07 08 09 08 09 (b) Gold Return (a) Natural Log of Gold Price 7. 4 0. 08 7. 2 0. 04 7. 0 0. 00 6. 8 –0. 04 6. 6 Figure 1. Graphical plots of gold and stock prices and returns –0. 08 6. 4 6. 2 02 03 04 05 06 07 08 09 –0. 12 (c) Natural Log of Kuala Lumpur Composite Index 02 03 04 05 06 07 (d) Stock Market Return correlation between gold return and once-lagged stock return is signi? ant. Its correlation is positive, suggesting that the gold market tends to follow the stock market with one-day lag. The cross-correlations between gold return and lead stock returns indicate the absence of signi? cation correlations. Accordingly, the gold market does not lead the stock market. This preliminary analysis seems to provide a basis for our one-equation empirical approach with no feedback from gold return to stock return and with the inclusion of once-lagged stock return in the mean equation of gold return. As regards to our main interest, it indicates at best the diversi? ation property of gold investment since its noted positive correlation is far from unity. However, this ? nding is only suggestive and must be subject to a formal analysis, which we turn next (Table II). Estimation results This section conducts a formal analysis of gold return and its relation to stock market return as speci? ed in equations (1) and (2) using GARCH-type models. We experiment with various error distribution assumption and variance speci? cation and choose the one that maximizes the log likelihood. The values of log likelihood functions for alternative models are given in Table III.
This log likelihood criterion unequivocally suggests the generalized error distribution of error terms. It also suggests either TARCH or EGARCH speci? cation to best describe variance speci? cation. TARCH speci? cation is chosen for basic model, model N0 and model N1 while EGARCH speci? cation for other models. Note that the differences in the log likelihood values between the two speci? cations are marginal. Estimation of the TARCH (1, 1) model for the basic mean equation yields the following results (numbers in parentheses are p-values): RG;t ? ht ? 0:0004 20:0344RG;t21 20:0111RS;t ?0:016? ?0:046? 0:582? 0:0000014 ?0:008? ?0:07721221 t 31 ?0:0502RS;t21 ?0:014? 20:05351221 I t21 t ?0:000? Market risk and gold investment ?0:003? ?0:9413ht21 ?0:000? N ? 2; 259; GED Parameter ? 1:7025 ? 0:000? ; Log Likelihood ? 7; 168:42 where It ? 1 if 1t , 0 and 0 otherwise. The use of TARCH model implies that previous shocks have asymmetric effects on volatility. Since the coef? cient of 1221 I t21 is negative, t bad news (1t , 0) tends to dampen market volatility. In other words, once-lagged positive news (1t2 1 . 0) exerts a greater impact on gold return volatility than negative news does, which conforms to the ? ding of Capie et al. (2005). Moreover, gold return volatility tends to be highly persistent as suggested by large coef? cient of lagged volatility. Turning to our main theme, we note the signi? cance of only once-lagged stock return. This conforms to the correlation structure observed in the previous section. However, its coef? cient is small, in the order of 0. 05. Thus, a 10 percentage point k RG,t, RS,t-k RG,t, RS,t? k 0 1 2 3 4 5 0. 0032 0. 0579 2 0. 0224 0. 0127 2 0. 0085 0. 0173 0. 0032 0. 0240 0. 0151 0. 0254 0. 0258 2 0. 0167 GARCH Speci? cation Basic N0 N1 N2 N3 N4
GARCH-N GARCH-T GARCH-G TGARCH-N TGARCH-T TGARCH-G EGARCH-N EGARCH-T EGARCH-G 7,035. 569 7,146. 246 7,163. 378 7,046. 186 7,153. 767 7,168. 421 7,026. 377 7,158. 247 7,168. 083 7,035. 893 7,146. 520 7,165. 204 7,046. 458 7,154. 348 7,170. 701 7,026. 710 7,158. 82 7,170. 554 7,036. 291 7,146. 26 7,163. 645 7,046. 785 7,153. 782 7,168. 730 7,027. 169 7,158. 361 7,168. 641 7,034. 568 7,142. 140 7,159. 647 7,045. 231 7,149. 472 7,164. 399 7,031. 521 7,154. 147 7,164. 628 7,031. 221 7,138. 171 7,156. 706 7,043. 397 7,146. 017 7,162. 170 7,030. 436 7,151. 064 7,163. 104 7,030. 379 ,134. 302 7,152. 533 7,042. 447 7,141. 644 7,157. 886 7,031. 285 7,146. 542 7,159. 008 Table II. Estimated cross-correlations Model Table III. Log likelihood of alternative GARCH speci? cations IMEFM 5,1 32 reduction in stock returns is associated the decrease in stock return by 0. 50 percentage point on average and likewise for the stock market increase. Note that the coef? cient of lagged gold return is negative. This suggests that the gold return tends to exhibit a reversal pattern and that the long run impact on gold return of stock market variations is even smaller.
In order to evaluate the dynamics of gold return during times of consecutive negative market returns, we estimate the chosen GARCH models (Table III) for the consecutive negative returns ranging from one to ? ve days (equation (2)). Results of the estimation are provided in Table IV. Note from the table that there are no changes in the results for the variance equation. Gold return volatility depends mostly on its past volatility and positive shocks tend to propel higher volatility. In the mean equation, we generally observe no level effect of consecutive negative market returns on gold return except for model 3.
Similar to the basic model, we note signi? cant positive coef? cient of lagged stock return in all models except one, i. e. model N0. More importantly, there seems to be no changes in the relations between gold and stock returns in times of consecutive negative market returns. The coef? cients of interactive dummies are all indistinguishable from 0 except one, i. e. the N3 model. In the case of N3 model, the investment role of gold is further enhanced. In responses to four consecutive Estimated coef? cients Mean equation a0 a1 r b10 b11 b20 b21 Variance equation u0 u1 u2 u3 N0 (TARCH) 0. 0000 2 0. 0007 2 0. 315 * 0. 0465 2 0. 0602 0. 0352 0. 0254 N1 (TARCH) 0. 0003 2 0. 0004 2 0. 0320 * 2 0. 0054 0. 0263 0. 0545 * * 2 0. 0114 Model N2 (EGARCH) N3 (EGARCH) N4 (EGARCH) 0. 0004 * * 0. 0001 2 0. 0341 * * 2 0. 0093 0. 0110 0. 0474 * * 0. 0150 0. 0004 * * 2 0. 0025 * * 2 0. 0265 2 0. 0034 2 0. 0979 0. 0549 * 2 0. 2243 * * 0. 0004 * * 2 0. 0008 2 0. 0284 * 2 0. 0036 2 0. 0146 0. 0507 * * 2 0. 2640 0. 000001 * * * 0. 000001 * * * 2 0. 1156 * * * 2 0. 1064 * * * 2 0. 1261 * * * 0. 0809 * * * 0. 0776 * * * 0. 0858 * * * 0. 0830 * * * 0. 0923 * * * 2 0. 0575 * * * 2 0. 0539 * * * 0. 0595 * * * 0. 0603 * * * 0. 0592 * * * . 9402 * * * 0. 9410 * * * 0. 9942 * * * 0. 9950 * * * 0. 9936 * * * Notes: Signi? cant at: *10, * *5 and * * *1 percent, respectively; the estimated models are: Mean equation: RG;t ? a0 ? a1 Nmt ? rRG;t21 ? ?b10 ? b11 Nmt ? ? RS;t ? ?b20 ? b21 Nmt ? ? RS;t21 ? 1t Variance equations: TARCH: Table IV. Estimation results of extended models ht ? u0 ? u1 1221 ? u2 1221 ? I t21 ? u3 ht21 t t GARCH: p????????? log ht ? u0 ? u1 j1t21 = ht21 j ? u2 1t21 =ht21 ? u3 log ht21 negative market returns, current and last three-day returns, the gold market tends to move in the opposite direction of stock market slumps.
The coef? cient of interactive dummy-lagged stock return in the N3 model is signi? cantly negative and its magnitude (in absolute term) is substantially higher than the coef? cient of lagged stock return. Thus, there seems to be a movement of the gold market away from downward trend in the stock market. The evidence that we uncover, thus, supports strong resistance of the gold market to stock market downturns. This is in sharp contrast to the well-documented ? nding that national stock markets tend to have strong co-movements during times of market decline and turmoil, which limit potential diversi? cation bene? across national stock markets. The heightened reaction of domestic stock markets to downturns in other markets have been documented by Pagan and Soydemir (2001) and Bahng and Shin (2003) for several emerging markets. Moreover, the ? nancial crises are noted to propagate shocks more strongly through the contagion or domino effect (Dornbusch et al. , 2000; Hasman and Samartin, 2008; Markwat et al. , 2009). Thus, a ? ight to other markets for shelter during times of ? nancial crises may not help. In the case of gold investment, its diversi? cation bene? ts are not restrained in times of market downturns.
Indeed, there is some evidence that the stock market may surge in value when the stock market posts a negative trend. Conclusion A series of ? nancial crises that erupted in different parts of the world and their accompanying excessive risk have raised serious concern over investment in stock markets and are likely to bring back interest in gold as an alternative investment asset. In light of this, we examine the relation between gold and stock returns and investigate whether it changes during times of consecutive negative market returns for an emerging market, Malaysia.
Applying GARCH-type models to daily gold and stock returns over the period August 2001-March 2010, we uncover evidence indicating signi? cant positive relation between gold return and once-lagged stock return. However, the coef? cient of the once-lagged stock return in gold return equation is small and far from unity. We further note that, their relation has not strengthened during times of consecutive days of market declines. To the contrary, we ? nd some evidence that gold return tends to break from its positive relation with stock market return following four consecutive stock market returns. These ? dings are in sharp contrast to the observed strong co-movements among national stock markets in periods of market downturns, which are attributed to contagion or domino effect. Based on these results, we incline to suggest the favorable property of gold as an investment asset for the Malaysian emerging market. At least, gold provides a diversi? cation bene? t to investors in the Malaysian market. The domestic Malaysian gold market tends to have resistance to heightened risk in the stock market as its preserve its low positive relation with stock market variations regardless of the market conditions.
At best, with evidence pointing to the negative relation between gold return and stock market return after four consecutive negative market returns, gold tends to possess a hedging property in times of market declines. In short, our results seem to support the initiative by Malaysia in introducing various gold coins, namely Kijang Emas, Royal Mint gold Dinar and Kelantan State gold Dinar, as a vehicle for preserving wealth in the midst of recurring ? nancial turbulences during the present time. Market risk and gold investment 33 IMEFM 5,1 34 Note 1. Dinar refers to the name of gold coin used in Islamic history.
The interest in gold Dinar during the Asian ? nancial crisis is not only limited to its store of value role and its use in international trade settlement but also to the adoption of gold as a payment standard. References Bahng, J. S. and Shin, S. -M. (2003), “Do stock price indices respond asymmetrically? Evidence from China, Japan, and South Korea”, Journal of Asian Economics, Vol. 14 No. 4, pp. 541-63. Baur, D. G. and Lucey, B. M (2010), “Is gold a hedge or a safe haven? An analysis of stocks, bonds, and gold”, The Financial Review, Vol. 45 No. 2, pp. 217-29. Baur, D. G. and McDermott, T. K. (2010), “Is gold a safe haven?
International evidence”, Journal of Banking & Finance, Vol. 34 No. 8, pp. 1886-98. Capie, F. , Mills, T. C. and Wood, G. (2005), “Gold as a hedge against the dollar”, Journal of International Financial Markets, Institutions and Money, Vol. 15 No. 4, pp. 343-52. Dornbusch, R. , Park, Y. and Claessens, S. (2000), “Contagion: how it spreads and how it can be stopped”, World Bank Research Observer, Vol. 15 No. 2, pp. 177-97. Hasman, A. and Samartin, M. (2008), “Information acquisition and ? nancial contagion”, Journal of Banking & Finance, Vol. 32 No. 10, pp. 2136-47. Hillier, D. , Draper, P. and Faff, R. 2006), “Do precious metals shine? An investment perspective”, Financial Analysts Journal, Vol. 62 No. 2, pp. 98-106. Koutmos, G. and Martin, A. D. (2003), “Asymmetric exchange rate exposure: theory and evidence”, International Journal of Money and Finance, Vol. 22 No. 3, pp. 365-83. Lobo, B. J. (2000), “Asymmetric effects of interest rate changes on stock prices”, The Financial Review, Vol. 35 No. 3, pp. 125-44. Markwat, T. , Kole, E. and van Dijk, D. (2009), “Contagion as a dom? no effect in global stock markets”, Journal of Banking & Finance, Vol. 33 No. 11, pp. 996-2012. Nam, K. , Washer, K. M. and Chu, Q. C. 2005), “Asymmetric return dynamics and technical trading strategies”, Journal of Banking & Finance, Vol. 29 No. 2, pp. 391-418. (The) News Strait Times (2001), “Practices in Islamic banking”, News Strait Times, June, p. 26. Pagan, J. A. and Soydemir, G. A. (2001), “Response asymmetries in the Latin American equity markets”, International Review of Financial Analysis, Vol. 10 No. 2, pp. 175-85. Corresponding author Mansor H. Ibrahim can be contacted at: mansorhi@hotmail. com To purchase reprints of this article please e-mail: reprints@emeraldinsight. com Or visit our web site for further details: www. emeraldinsight. com/reprints
