Return, Volatility and Equity Fund Flows Linkages: Evidence from an Emerging Market

Transcription

1 Retur, Volatility ad Equity Fud Flows Likages: Evidece from a Emergig Market Ros Zam Zam Sapia, Jig Qua Lee To Lik this Article: DOI: /IJARBSS/v8-i7/4333 Received: 02 Jue 2018, Revised: 07 Jue 2018, Accepted: 29 Jue 2018 Published Olie: 18 July 2018 I-Text Citatio: (Sapia & Lee, 2018) To Cite this Article: Sapia, R. Z. Z., & Lee, J. Q. (2018). Retur, Volatility ad Equity Fud Flows Likages: Evidece from a Emergig Market. Iteratioal Joural of Academic Research i Busiess ad Social Scieces, 8(7), Copyright: 2018 The Author(s) Published by Huma Resource Maagemet Academic Research Society ( This article is published uder the Creative Commos Attributio (CC BY 4.0) licese. Ayoe may reproduce, distribute, traslate ad create derivative works of this article (for both commercial ad o-commercial purposes), subject to full attributio to the origial publicatio ad authors. The full terms of this licese may be see at: Vol. 8, No. 7, July 2018, Pg JOURNAL HOMEPAGE Full Terms & Coditios of access ad use ca be foud at 172

2 Retur, Volatility ad Equity Fud Flows Likages: Evidece from a Emergig Market Ros Zam Zam Sapia, Jig Qua Lee Faculty of Ecoomics ad Maagemet, Uiversiti Kebagsaa Malaysia, UKM Bagi, Selagor, Malaysia zamzam@ukm.edu.my Abstract This study examies the dyamic likages amog market retur, market volatility, ad equity fud flows of istitutioal ad retail ivestors both foreig ad local ito Malaysia Stock Exchage, Bursa Malaysia. Usig a total of 1661 daily observatios of aggregate trade data for a period from 1 st October 2009 to 30 th Jue 2016, this study fids that market retur has a effect o buy trades of local ivestors, sell trades of foreig istitutios ad local retailer as well as et flows of foreig istitutios ad retailers. O the equity flows-market retur relatio, the fidig shows that the buy trades of foreig retail ivestors, et flows of foreig istitutios ad foreig retailers affect market retur. This study also provide evidece that market volatility is sigificatly impact foreig istitutioal ivestors buy trades as well as local retailer et flows of equity. Furthermore, this study reveal that there are isigificat results for the equity flows-market volatility relatio. The fidigs of this study is crucial ad will beefit most to portfolio fud maagers, traders, foreig ad local ivestors dealig with Bursa Malaysia. Keywords: Market Retur, Market Volatility, Equity Fud Flows, Istitutioal Ad Retail Ivestors, Grager Causality Itroductio Equity fud flows have bee a typical idicator used by ivestors ad market speculators to gauge other ivestors behaviour or market setimets. The chages i fud flows reflect the chages i market setimets. It has bee a ature that security market roller-coaster would ever stop, ad for that reaso, market retur ad market volatility amog others, have bee caused of cocer for global ivestors as determiats of ivestmet decisios. Fiace literature also demostrates that differet arrays of ivestors egage with differet tradig styles of which may affect market price ad/or volatility differetly. For istace, a study by Bae et al. (2008) state ivestors tradig styles might lead to further icreases i prices or greater volatility. This is due to mometum traders demad liquidity as price icreases. 173

3 There are ample empirical studies o the relatios betwee market retur ad fud flows of equity, ad documet two mai fidigs. Firstly, retur of the market has a impact o equity flows. Prior literature supports the positive associatio betwee retur of the market ad equity flows i large umber of the studies (Froot et al. 2001; Jijarak et al. 2011). Moreover, there is dissimilar tradig behaviour of fud flows to market retur ad volatility betwee coutries (Paek 2014). Secodly, fud flows of equity also have a effect o market retur. Past studies that show fud flows of equity have a impact o market retur icludes Bekaert et al. (2002), Dahlquist ad Robertsso (2004) ad Frech ad Li (2012). Hece, prior studies show the existece of bidirectioal relatioships betwee market retur ad fud flows of equity. Stock retur volatility is fudametal to fiace. There are also may studies have bee carried out o the relatioships betwee market retur volatility ad fud flows of equity. The fidigs of past studies show that there is a associatio betwee market volatility ad fud flows of equity, ad vice versa (Lee et al. 2015; Cao et al. 2008; Nguye ad Le 2013). Eve though may studies have bee carried out o the relatios betwee market retur, market volatility ad equity flows o developed market, the same area of study has bee relatively uderexplored for the Malaysia equity market. Hece, it is crucial to research the dyamic likages amog market retur, market volatility, ad equity fud flows of which is the focus of this study. This study aims to preset a ew uderstadig o the relatioships amog market retur, market volatility, ad equity fud flow of a emergig equity market, Bursa Malaysia which is the official stock exchage of Malaysia. This study would like to aswer the subsequet two research questios: First, does retur of the equity market affects equity fud flows or vice versa? Secod, does volatility of the equity market affects equity fud flows or vice versa? The fidigs of this paper cotributes to the limited empirical literature o ivestors tradig behaviour particularly for emergig equity market. The fidigs of this study are beeficial to various parties such as equity ivestors, fud maagers ad policymakers of emergig markets like Malaysia equity market. Uderstadig the behavior of equity flows is crucial to both ivestors ad fud maagers as this will assist them to develop better tactical ivestmet strategies to maximize retur. A sudde capital outflow by equity ivestors particularly foreig ivestors may dampe ad destabilize local equity markets particularly durig periods of recessios (Tsai 2009) or period of fiacial upheaval. Thus, it is crucial to have a better kowledge o the ivestors tradig behaviour i order to meet ay uaticipated evets. Not oly that, havig a better kowledge o the ivestors tradig behaviour may assist the local authority to ehace its equity market liquidity ad cosequetly stocks marketability amog ivestors. The remaider of this paper is as follows: Sectio 2 presets the literature review, Sectio 3 presets the data ad methodology ad the tested hypothesis. Sectio 4 presets the results while Sectio 5 provides the summary ad coclusios of this study. Literature Review Fiace literature documets volumious studies o the likages betwee market retur ad fud flows of equity. Furthermore, differet arrays of ivestors are perceived to egage with differet tradig styles of which may affect market price ad/or volatility differetly. For example, a study by Bae, Yamada ad Ito (2008) reveal that ivestors tradig styles may cause prices to icrease further or volatility to become greater. This is due to mometum traders who demad liquidity whe there 174

4 is a icrease i prices. The behavior of ivestors tradig differs ot oly itra but also iter equity markets. For istace, Paek ad Ko (2014) documets differet tradig patter of equity fud flows to market retur ad volatility betwee coutries. Previous studies show mixed results o the relatios betwee retur of market ad equity fud flows. Earlier studies which focus o the likages betwee market retur ad aggregate mutual fud flow icludes a study by Warther (1995) which shows that there is a egative relatio betwee returs ad subsequet flows. Edele ad Warer (2001) also fid a very strog relatio betwee retur ad fud flows. Eve though i geeral ivestors take a overight period to react, their study reveal that lagged retur has a impact o flows. Furthermore, Cha ad Kim (2010) ad Kim ad Kim (2016) fid evidece that there is a positive uidirectioal causal relatio from stock retur to fud flows at a macro level. Mishra (2011) coducts a similar study i Idia ad his fidig supports empirical evideces that there is a uidirectioal causal relatio from stock market retur to mutual fuds ivestmet. I aother study, Watso ad Wickramaayake (2012) state that share market returs Grager-cause maaged fud flows i Australia. Oh ad Parwada (2007) study the relatioship betwee fud flows ad stock market returs i Korea. They fid that i terms of et tradig flows, mutual fud ivestors are cotraria traders. Moreover, Ha et al. (2015) demostrate that market retur shocks have a cotemporaeous positive effects o cash iflows while little effect of o cash outflows. Meawhile, i the cotext of Greece capital market, studies by Caporale et al. (2004) ad Alexakis et al. (2005) fid that there is a bidirectioal causal relatio betwee mutual fud flows ad stock returs. A study by Aydoga et al. (2014) reveals similar fidig o Turkish capital market. Be- Rephael et al. (2011) o the other had reveal that lagged returs do ot have a impact o mutual fud flows. Frech ad Naka (2013) obtai similar fidigs. Their study reveals a isigificat impact of market retur o U.S. equity iflows ito Chia ad Idia markets. Istead, the U.S. ivestors attract to the Chia ad Idia market is based o the uique characteristics ad fudametal values of these two coutries. Aother study by Richards (2005) examies the ivestors tradig patters of six Asia emergig equity markets o daily basis. The fidigs reveal that idividual ad foreig ivestors execute differet tradig strategies. Foreig ivestors ted to be mometum traders. O the other had, idividual ivestors ad less sophisticated istitutios ted to be cotrarias. O the Malaysia equity market perspective, there exists a positive associatio betwee market retur ad fud flows of equity i fiace literature. The studies iclude Bekaert et al. (2002), Boh ad Tesar (1996), Froot ad Ramadorai (2008), Froot et al. (2001), Jijarak et al. (2011), Li ad Swaso (2004), Li ad Swaso (2008) ad Sapia ad Auzairy (2015). Prior literature also documets the presece of flow-retur relatioship i equity market. For istace, Warther (1995) fid evidece of a positive relatioship betwee flows ad subsequet returs. Edele ad Warer (2001) examie the relatioship betwee market returs ad aggregate equity flow by usig daily flow data of U.S. equity. Evidece of their study reveals a associatio betwee the variables which the causality ruig from fud flows to market retur. Be-Rephael et al. (2011) also fid similar results whereby lagged flows seem to predict future returs. Other studies that show a sigificat impact of fud flows o market retur icludes Bekaert et al. (2002), Dahlquist ad Robertsso (2004), Frech ad Li (2012), Froot ad Ramadorai (2008), Froot et al. (2001) ad Ülkü ad İkizlerli (2012). Chadra (2012) study also provides similar fidigs, but the associatios are 175

5 oly i short period of time. I short, prior studies provide evideces of bi-directioal relatioships betwee market returs ad fud flows of equity. All of the above metioed studies focus o the dyamic relatioship betwee market returs ad equity fud flows. However, prior literature also provides evideces that there is a associatio betwee equity flows ad market retur volatility. High market retur volatility reflects higher risk ad ucertaity which may dampe a coutry s ecoomic performace. Most of the studies show the directio of causality ruig from fud flows of equity to market volatility. For istace, Li (2006) demostrates that foreig ivestors ivestmet decisios affect market volatility sigificatly across time as foreigers play a icreasig importat role i Taiwa s market. Pavabutr ad Ya (2007) ivestigate the impact of foreig portfolio flows o the volatility of Thailad stock market. Their study reveals that foreig flows have a effect o market retur volatility of which due maily by the uexpected shocks to foreig flows. Usig a VAR approach, Cao et al. (2008) documet a egative correlatio betwee volatility ad cocurret ad lagged flow. A shock i iflow foresees a decrease i volatility, while a shock i outflow foresees a icrease i volatility. Bae et al. (2008) examie the tradig impact of various ivestor types o retur volatility of equity. The fidigs of their study reveal that the iteractios of mometum ad cotraria ivestors trades affect market volatility. Market volatility icreases above the average level whe there is greater mometum buy trades ad less cotraria sell trades. The mometum ivestors are said to demad liquidity while the cotraria or profit-takig ivestors to supply liquidity. O the other had, volatility dimiishes by more tha half, whe there are greater sell trades by cotraria ivestors ad less buy trades by mometum ivestors. Nguye ad Le (2013) aalyse the impacts of daily foreig flows o Vietamese stock market volatility ad fid that foreig flows have a impact o market volatility but oly o short term basis. Li ad Wag (2010) examie a short ru dyamic relatio betwee trades of istitutioal ivestors ad stock price volatility i a retail ivestor-domiated emergig market. They fid that there is a egative relatio betwee volatility ad et istitutioal trades which is sigificat. Moreover, the fidigs show that volatility ad istitutioal trade relatioship differs for buy ad sale trades, ad for small ad large stocks. Ahmed (2016) examies the relatioship betwee market volatility ad equity flows of foreig ad local ivestors tradig activities i Qatar Stock Exchage. His study demostrates that sales of foreig istitutios ad idividuals have a positive impact o market volatility while purchases of foreig istitutios ted to reduce market volatility. O the cotrary, purchases of local istitutios impact market volatility egatively while sales of local istitutios have o impact o volatility. Aother study by Umutlu, Akdeiz ad Salih (2013) demostrate that there is a positive relatio betwee volatility ad foreig ivestors et equity flows eve after cotrollig for market-wide price impacts, market developmet, liquidity ad persistecy i volatility. Wag (2007) study documets strog cotemporaeous relatioship betwee foreig equity tradig ad market volatility i Idoesia ad Thailad ad foreig sale trades explai most for the market volatility i both coutries. Past studies that documet market volatility-fud flows relatioships amog others are Paek ad Ko (2014) ad Lee et al. (2015). Paek ad Ko (2014) explores the dyamic relatioships betwee market volatility, market retur ad equity flows of both U.S. ad Japa equity market. The result of their 176

6 study demostrates the existece of a positive feedback tradig for the U.S. ad Japa market, however with completely differet causality. The result reveals that the positive feedback tradig is due to the egative effect of lagged market retur o outflows for the U.S. market ad the positive effect of market retur o iflows for the Japa market. Furthermore, there is a egative cotemporaeous effect of market volatility shocks o et flows for the U.S. market. There is a positive (egative) cotemporaeous effect of retur shocks o et flows for the U.S (Japa) respectively. The fidigs also reveal that volatility shocks rather tha retur shocks explai more the cash flows for the U.S. market, ad there is a opposite result for the Japa market. Lee et al. (2015) examie the dyamic relatios amog market volatility, market retur, ad aggregate equity fud flows o a mothly basis i a iteratioal cotext. Their fidigs amog others show that Wester ivestors give emphases o market volatility whe they buy ad redeem equity fuds. Cao et al. (2008) also fid similar fidigs. Their study shows that there is a egative impact of volatility o daily mutual fud flows. Aother study by Ha et al. (2015) demostrates that market volatility shocks affect cash outflows (et flows) i positive (egative) way cotemporaeously. However, the impact of market volatility shocks o iflows is very miimal. I short, prior studies also documets mixed results o the market volatility-fud flows relatios. Data ad Methodology Measuremet of Market Retur, Market Volatility ad Net Flows This study employs time series data of market retur, market volatility ad equity flows. The time series data is from 1 st October 2009 to 30 th Jue 2016, of which i total comprises of 1661 umber of observatios to examie the relatioships amog market retur, market volatility ad equity flows. This study focuses maily o three variables which are market retur, market volatility ad equity flows i the form of trades quatity (buy trades, sell trades ad et flows). Table 1 presets the descriptios of the abbreviatios for the above three mai variables utilize i this study. Table 1. Abbreviatios Variables Abbreviatios Variables Abbreviatios Foreig Equity Flows Local Equity Flows Istitutios Istitutios Buy Trades FIBUY Buy Trades LIBUY Sell Trades FISELL Sell Trades LISELL Retail Retail Buy Trades FRBUY Buy Trades LRBUY Sell Trades FRSELL Sell Trades LRSELL Net Flows Net Flows Istitutios NFFINST Istitutios NFLINST Retail NFFRET Retail NFLRET Market Retur RET Market Volatility RETVOL This study employs Equatio (1) to compute daily market retur. 177

7 where Rt = (Pt Pt-1) / Pt-1 (1) Rt = Daily market retur, Pt = Closig price of market idex o day t, ad Pt-1 = Closig price of market idex o day t-1 Cosistet with Giles (2008), this study uses retur squared as a proxy for market volatility. While, this study employs Equatio (2) to compute daily et flows i the form of quatity of trades. where VBuy VSell F t = (V Buy V Sell ) (V Buy +V Sell ) Ft = Daily et flows, = Quatity of buy trades, ad = Quatity of sell trades (2) Uit Root Before proceeds with VAR Grager Causality test, this study adopts a uit root test to esure that all of the time series data are statioary. This study employs Augmeted Dickey Fuller (ADF) test (Dickey ad Fuller (1979; 1981) to determie the statioarity of market returs, market volatility ad fud flows data. This study will ot be able to carry out further aalysis usig Vector Autoregressive (VAR) model if the variables have bee foud to be ot statioary. The ull hypothesis to test the statioarity of the data is that the variable cotais a uit root which meas the time series data are ot statioary, ad the alterative hypothesis is that the variable has o uit root. Moreover, this study employs Akaike iformatio criterio (AIC) for the determiatio of the appropriate umber of lag legth. VAR Grager Causality This study employs VAR Grager Causality test to examie the dyamic relatioships betwee market retur ad equity fud flows, ad betwee market volatility ad equity fud flows. Cosistet with Darwish (2012), a VAR model which icludes market returs ad equity fud flows is as follows: R t = α R + i=1 α i R t i + i=1 β i F t i + ε R,t (3) F t = α F + i=1 α i F t i + i=1 β i R t i + ε F,t (4) While, a VAR model which icludes market volatility ad et flows is as follows: V t = α V + i=1 α i V t i + i=1 β i F t i + ε V,t (5) F t = α F + i=1 α i F t i + i=1 β i V t i + ε F,t (6) where R t is market retur, F t is Buy Trades, Sell Trades or Net Flows i the form of quatity of trades ad V t represet market volatility, ε R,t, ε F,t ad ε V,t are orthogoal error terms ad idicate autoregressive lag legths. 178

8 The ull hypothesis i Grager causality is that Ft does ot Grager-cause Rt i Equatio (3). The ull hypothesis for Equatio (4) is that Rt does ot Grager-cause Ft. Likewise, the ull hypothesis of Equatio (5) ad Equatio (6) is that Ft does ot Grager-cause Vt ad Vt does ot Grager-cause Ft respectively. The ull hypothesis above ca be addressed by H0: β i = 0 for all i i stated equatio ad the test statistic is a Chi-square test. Rejectio of ull hypothesis supports the presece of Grager causality betwee the variables. Results ad Discussios Descriptive Statistics Table 2 reports the summary of descriptive statistics of buy ad sell trades, et flows, market retur ad market volatility. The statistics show that local retailers cotribute the most to the tradig volume, followed by local istitutios ad the foreig istitutios. This support the cojecture that Malaysia equity market is retail ivestors domiated market. The average et flows of foreig ad local ivestors are egative except for local retailers. This meas that foreig ivestors both istitutioal ad retail as well as local istitutioal ivestors sell more tha buy of domestic shares. Table 2. Descriptive Statistics of Daily Market Retur, Market Volatility ad Equity Flows Variable Mea Std. Dev. Skewess Kurtosis Jarque- Bera ADF FIBUY ** ** FISELL ** ** FRBUY ** ** FRSELL ** ** LIBUY ** ** LISELL ** ** LRBUY ** ** LRSELL ** ** NFFINST ** ** NFFRET ** ** NFLINST ** ** NFLRET ** ** RET ** ** RETVOL 3.46E E ** ** Note: **Sigificat at 1% level The Jarque-Bera ormality test idicates that all of the variables are ot ormally distributed. The fidigs i Table 2 also report the results of ADF test for uit root. The results show that the ull hypothesis where variables metioed are o-statioary at level is all rejected. Thus, further aalysis usig VAR Grager causality test is permissible. 179

9 Correlatio Coefficiets Table 3 presets the correlatio coefficiets betwee market retur ad equity et flows, ad betwee market volatility ad equity et flows. The results i Table 3 demostrate that market retur is egatively correlated with et flows of foreig retailers, local istitutios ad local retailers with the correlatio of , ad respectively at a sigificace level of 1%. Table 3. Spearma Correlatio Coefficiets betwee Market Retur, Market Volatility ad Net Flows. Variable NFFINST NFFRET NFLINST NFLRET RET RETVOL NFFINST NFFRET ** NFLINST ** NFLRET **.053 * ** RET.264 ** ** ** ** RETVOL *.075 **.050 * ** Note: ** Sigificat at 1% level, * Sigificat at 5% level. This study uses Spearma Correlatio test due to the o-ormality of the time series data. The Spearma correlatio betwee market retur ad et flows of foreig istitutios is which idicates that a positive relatioship exists betwee these two variables. The statistics i Table 3 also reports that market volatility is positively correlated to et flows of foreig retailers, local istitutios ad local retailers with correlatio of 0.075, ad respectively. The Spearma correlatio betwee market volatility ad et flows of foreig istitutios is , which meas that whe market volatility icrease, et flows of foreig istitutios decrease. This is cotrary to the study of Umutlu Akdeiz ad Altay-Salih (2013) where they fid that there is a positive relatioship betwee average total volatility ad foreig ivestors et equity flow. I additio, the results i Table 3 reveal that there is iverse relatioship for the correlatios betwee market retur ad et flows; ad market volatility ad et flows for all categories of ivestors. VAR Grager-Causality Test The earlier sectio shows that there is a correlatio betwee market retur, market volatility; ad et flows but these correlatios do ot ecessarily mea that there is a causal relatio betwee the variables. This sectio reports the results of Grager causality betwee market retur ad equity fud flows, ad betwee market volatility ad equity fud flows accordig to VAR framework as per Table 4 ad 5 respectively. Relatioship betwee Market Retur ad Equity Flows This sectio reports the causal relatio betwee market retur ad equity fud flows of foreig ad local ivestors, both istitutioal ad retail. The fidigs i Pael A, B ad C of Table 4 show the retur-flows relatio i terms of buy trades, sell trades ad et flows respectively. 180

10 The fidigs demostrate that the testig hypothesis that market retur does ot grager-cause equity flows is rejected for the buy trades of local istitutios, local retailers; ad the sell trades of both foreig istitutios ad local retailers at sigificace level of 1% ad 5% respectively. These fidigs suggest that market retur has a impact o local istitutios ad retail equity purchases. Moreover, market retur also impacts the sell trades of foreig istitutios ad local retailers. The results also report the rejectio of the ull hypothesis that equity flows do ot grager-cause market retur for the buy trades of foreig retailers. These results sigify that there is a sigificat impact of foreig retail ivestor buyig activities o market returs. This study exteds the aalysis o the market-flows causal relatios but this time is betwee market retur ad et flows. The results i Pael C illustrate that the ull hypothesis that market retur does ot grager-cause et flows is rejected for both et flows of foreig istitutios ad retailers at the sigificace level of 1% respectively. These fidigs sigify that market retur sigificatly affect et flows of both foreig istitutios ad retailers. Adaoglu ad Katircioglu (2013) fid that stock returs affect et foreig ivestor flows o a mothly basis for the pre-eu accessio egotiatios period which is cosistet with the fidigs of this study. Ahmed (2016) demostrate that Qatari market daily returs grager cause idividual ad istitutioal ivestmet flows of both local ad foreig ivestors. Based o weekly frequecy data, Frech (2011) establishes sigificat relatioship betwee past market retur ad foreig flows for South Africa equity market. Aother study by Hog ad Lee (2011) shows that Korea market retur affects istitutioal, idividual ad foreig ivestors et ivestmet flows. Table 4. Grager-Causality Test betwee Market Retur ad Equity Flows Equity Flows Market Retur does ot Grager-cause Equity Flows Equity Flow does ot Grager-cause Market Retur Chi-sq Prob. Chi-sq Prob. Pael A: Buy Trades Foreig Istitutios Foreig Retailers * Local Istitutios ** Local Retailers * Pael B: Sell Trades Foreig Istitutios * Foreig Retailers Local Istitutios Local Retailers * Pael C: Net Flows Foreig Istitutios ** ** Foreig Retailers ** ** Local Istitutios Local Retailers Note: ** Sigificat at 1% level, * Sigificat at 5% level. 181

11 The ull hypothesis that et flows do ot grager-cause market retur is also rejected for et flows of both foreig istitutios ad retailers at 1 percet sigificat level. These fidigs also provide evidece that there is a bidirectioal causal relatioship betwee market retur ad et flows of both foreig istitutios ad retailers. Prior study by Bekaert et al. (2002) provides evidece that capital flows affects retur of Malaysia equity market o mothly basis. Hog ad Lee (2011) also documet similar fidigs. Their study shows that foreig ad istitutioal ivestors et ivestmet flows as well as govermet drive Korea market retur. Frech (2011) however, fids o relatios betwee et flows ad returs of South Africa equity market. Relatioship betwee Market Volatility ad Equity Flows Table 5 presets the results of Grager causality test betwee market volatility ad equity fud flows. Pael A, B ad C report the results of causal relatios betwee market volatility ad equity flows i term of buy trades, sell trades ad et flows respectively. The fidig of this study demostrates the ull hypothesis that market volatility does ot grager-cause equity flows for buy trades of foreig istitutios is rejected at 5% sigificat level. The fidig idicates that market volatility has a impact o the buy trades of foreig istitutioal ivestors. Meawhile the ull hypothesis that market volatility does ot grager-cause equity flows for sell trades of all categories of ivestors is accepted at 5% sigificat level. Prior study such as Paek ad Ko (2014) reveal that past market volatility sigificatly ad positively affect both iflows ad outflows for the US market. Meawhile, lagged market volatilities has isigificat (sigificat) egative effect o iflows (outflows) for the Japa equity market. The results i Pael C idicate the ull hypothesis that market volatility does ot grager-cause equity flows is rejected for et flows of local retail ivestors at 1% sigificace level. Paek ad Ko (2014) demostrate that there is o impact of lagged market volatilities o et flows. Meawhile, the ull hypothesis that equity flow does ot grager-cause market volatility is accepted for the buy ad sell trades; ad et flows of all categories of ivestors eve at 5 percet sigificat level, thus is ot cosistet with fidigs of prior studies. A study by Li ad Wag (2010) demostrate that istitutioal buy (sell) trades affect price volatility at the market level egatively (positively). Aother study by Ahmed (2016) aalyses the associatio betwee market volatility ad flows of equity, both foreig ad local ivestors o Qatar Stock Exchage. The fidigs of his study show that there is a positive impact of foreig istitutios ad idividual sell trades o market volatility. O the other had, foreig istitutios buy trades have a egative impact o market volatility. Meawhile, local istitutios buy trades affect market volatility egatively but there is o impact for the sell trades of local istitutioal ivestors. Nguye ad Le (2013) examie the flowsmarket volatility relatios o Vietamese stock market. They fid that there is a sigificat shortterm impact of past foreig flows o market volatility. The foreig flows-market volatility relatio is stroger ad sigificat i the bull market as compared to bear market for gross buy ad sell; ad gross et buy. Li ad Wag (2010) demostrate that there is sigificat ad egative associatio betwee istitutioal buy-sell imbalace ad price volatility; ad it is due to uexpected compoet. Aother study by Umuthu et al. (2013) show that foreig ivestors et equity flows ifluece average total volatility i the Istabul Stock Exchage. A study by Cao, Chag ad Wag (2008) reveal that there is bidirectioal relatioship betwee mutual fud flows ad market volatility. 182

12 Table 5. Grager-Causality Test betwee Market Volatility ad Equity Flows Equity Flows Market Volatility does ot Grager-cause Equity Flows Equity Flows does ot Grager-cause Market Volatility Chi-sq Prob. Chi-sq Prob. Pael A: Buy Trades Foreig Istitutios * Foreig Retailers Local Istitutios Local Retailers Pael B: Sell Trades Foreig Istitutios Foreig Retailers Local Istitutios Local Retailers Pael C: Net Flows Foreig Istitutios Foreig Retailers Local Istitutios Local Retailers ** Note: ** Sigificat at 1% level, * Sigificat at 5% level. Summary ad Coclusios This study examies the dyamic likages amog market retur, market volatility ad equity flows of Malaysia stock market mai ivestor groups. By usig VAR approach, this study examies market retur-flows ad market volatility-flows causal relatios, ad vice versa. The results of VAR Grager causality test provide evidece of sigificat relatioships betwee market retur ad buy trades of local ivestors, sell trades of foreig istitutios ad local retailer as well as et flows of foreig istitutios ad retailers. These fidigs show that the iformatio cotet of earlier market returs is valuable to predict the tradig behaviour of ivestors participate i local equity market (Ahmed, 2016). Meawhile, equity flows affect market retur for the buy trades of foreig retail ivestors, et flows of foreig istitutios ad retail ivestors categories. This study also fids evidece that market volatility is sigificatly impact the buy trades of foreig istitutioal ivestors ad local retailer et flows of equity. I additio, the fidigs of this study also reveal that oe of the equity flows iflueces market volatility. This study cotributes a uderstadig o the dyamic relatioships amog market retur, market volatility ad equity flows of Malaysia Stock Exchage. The results of this study ca be used as guidace to various parties such as fud maagers, istitutioal ad retail ivestors, market speculators ad policy makers to uderstad the ivestors tradig behaviour as well as to gauge the market setimets. As Umuthu et al. (2013) states i their paper, it is very crucial to have a uderstadig o the costs ad beefits of equity flows i.e. foreig equity ivestmet i stock market as this matter has importat policy implicatios. Oe of the costs that ca arise is a icrease i retur volatility i emergig market brought by foreig equity ivestmets. If this situatio occurs, the policy 183

13 makers have to develop ways to stabilise the stocks. This is because high volatility i local capital markets ca impedes the growth ad developmet of fiacial markets i the log ru (Aimpichaimogkol ad Padugsaksawasdi 2013). Thus, it is much easier to take ecessary actios if they have a better uderstadig o the behavioural patters of equity ivestors. Ackowledgemet The author would like to thak Yayasa Tu Ismail Edowmet Fuds, Uiversiti Kebagsaa Malaysia for the fiacial support to coduct this research (Research grat #: EP ad Research Grat # EP ). Refereces Adaoglu, C. & Katircioglu, S.T. (2013). Foreig ivestor flows ad blue chip stock returs. Iteratioal Joural of Emergig Markets, 8(2), Ahmed, Walid M.A. (2016). Cross-border equity flows ad market volatility: the case of Qatar Exchage. Iteratioal Joural of Emergig Markets, 11(3), Aimpichaimogkol, H. & Padugsaksawasdi, C. (2013). Foreig Owership ad Stock Retur Volatility: Evidece from Thailad. Iteratioal Review of Busiess Research Papers, 9(5), Alexakis, C., Niarchos, N., Patra, T., & Poshakwale, S. (2005). The dyamics betwee stock returs ad mutual fud flows: Empirical evidece from the Greek market. Iteratioal Review of Fiacial Aalysis, 14, Aydoga, B., Vardar, G. & Tuç, G. (2014). The iteractio of mutual fud flows ad stock returs: Evidece from the Turkish capital market. Ege Academic Review, 14(2), Bae, K. Yamada, T. ad Ito, K. (2008). Iteractio of ivestor trades ad market volatility: evidece from the Tokyo Stock Exchage. Pacific-Basi Fiace Joural, 16, Bekaert, G., Harvey, C. R., & Lumsdaie, R. L. (2002). The dyamics of emergig market equity flows. Joural of Iteratioal Moey ad Fiace, 21, Be-Rephael, A., Kadel, S., & Wohl, A. (2011). The price pressure of aggregate mutual fud flows. Joural of Fiacial ad Quatitative Aalysis, 46(2), Boh, H. & Tesar, L.L. (1996). U.S. Equity ivestmet i foreig markets: Portfolio rebalacig or retur chasig? AEA Papers ad Proceedigs, 86(2), Cao, C., Chag, E. C., & Wag, Y. (2008). A empirical aalysis of the dyamic relatioship betwee mutual fud flow ad market retur volatility. Joural of Bakig ad Fiace, 32, Caporale, G. M., Philippas, N. & Pittis, N. (2004). Feedbacks betwee mutual fud flows ad security returs: evidece from the Greek capital market. Applied Fiacial Ecoomics, 14, Cha, H. J. & Kim, J. (2010). Stock returs ad aggregate mutual fud flows: a system approach. Applied Fiacial Ecoomics, 20, Chadra, A. (2012). Cause ad effect betwee FII tradig behaviour ad stock market returs: The Idia experiece. Joural of Idia Busiess Research, 4(4), Dahlquist, M. & Robertsso, G. (2004). A ote o foreigers tradig ad price effects across firms. Joural of Bakig & Fiace, 28, Darwish, M. (2012). Testig the cotemporaeous ad causal relatioship betwee tradig volume ad retur i the Palestie Exchage. Iteratioal Joural of Ecoomics ad Fiace, 4(4), Dickey, D. A., & Fuller, W. A. (1979). Distributio of the estimators for autoregressive time series with a uit root. Joural of the America Statistical Associatio, 74(366),

14 Dickey, D. A., & Fuller, W. A. (1981). Likelihood ratio statistics for autoregressive time series with a uit root. Ecoometrica 49(4), Edele, R. M., & Warer, J. B. (2001). Aggregate price effects of istitutioal tradig: A study of mutual fud flow ad market returs. Joural of Fiacial Ecoomics, 59, Frech, J.J. (2011). The Dyamic Iteractio Betwee Foreig Equity Flows ad Returs: Evidece from The Johaesburg Stock Exchage. The Iteratioal Joural of Busiess ad Fiace Research, 5(4), Frech, J. J., & Li, W.-X. (2012). A ote o US istitutioal equity flows to Brazil. Review of Accoutig ad Fiace, 11(3), Frech, J. J., & Naka, A. (2013). Dyamic relatioships amog equity flows, equity returs ad divideds: Behavior of U.S. ivestors i Chia ad Idia. Global Fiace Joural, 24, Froot, K. A., & Ramadorai, T. (2008). Istitutioal portfolio flows ad iteratioal ivestmets. The Review of Fiacial Studies, 21(2), Froot, K. A., Coell, P. G. J. O., & Seasholes, M. S. (2001). The portfolio flows of iteratioal ivestors. Joural of Fiacial Ecoomics, 59, Giles, D. E. (2008). Some properties of absolute returs as a proxy for volatility. Applied Fiacial Ecoomics Letters, 4(5), Grouard, M. H., Levy, S. & Lubochisky, C. (2003). Stock market volatility: from empirical data to their iterpretatio. Fiacial Stability Review, Ha, Y., Lee, B.S., Paek, M. & Ko, K. (2015). Structural VAR Approach to Mutual Fud Cash Flows: Net Flows, Iflows, ad Outflows. Asia-Pacific Joural of Fiacial Studies, 44, Hog, G. & Lee, B.S. (2011). The tradig behavior ad price impact of foreig, istitutioal, idividual ivestors ad govermet: Evidece from Korea equity market. Japa ad the World Ecoomy, 23, Jijarak, Y., Wogswa, J. & Zheg, H. (2011). Iteratioal fud ivestmet ad local market returs. Joural of Bakig & Fiace, 35, Kim, JB., & Kim, J. M. (2016). Stock returs ad mutual fud flows i the Korea fiacial market: A system approach. BOK Workig Paper No , Lee, B. S., Paek, M., Hab, Y. & Ko, K. (2015). The dyamics of market volatility, market retur, ad equity fud flow: Iteratioal evidece. Iteratioal Review of Ecoomics ad Fiace, 35, Li, W. & Wag, S.S. (2010). Daily istitutioal trades ad stock price volatility i a retail ivestor domiated emergig market. Joural of Fiacial Markets, 13, Li, A.Y. (2006). Has the Asia crisis chaged the role of foreig ivestors i emergig equity markets: Taiwa s experiece. Iteratioal Review of Ecoomics ad Fiace, 15, Li, A.Y. & Swaso, P.E. (2004). Iteratioal equity flows ad developig markets: The Asia fiacial market crisis revisited. Joural of Iteratioal Fiacial Markets, Istitutios ad Moey, 14, Li, A.Y. & Swaso, P.E. (2008). U.S. ivestors ad global equity markets. Iteratioal Review of Fiacial Aalysis, 17, Mishra, P. K. (2011). Mutual fuds ivestmet flow ad stock market returs i Idia. Visio, 15(1), Nguye, L. & Le, N. (2013). Impact of Foreig Portfolio Flows o Stock Market Volatility - Evidece from Vietam. Joural of Emergig Issues i Ecoomics, Fiace ad Bakig, 2(5),

15 Paek, M. & Ko, Kwagsoo. (2014). Aggregate et flows, iflows, ad outflows of equity fuds: The U.S. versus Japa. Japa ad the World Ecoomy, 32, Pavabutr, P. & Ya, H. (2007). The Impact of Foreig Portfolio Flows o Emergig Market Volatility: Evidece from Thailad. Australia Joural of Maagemet, 32(2), Oh, N. Y., & Parwada J. T. (2007). Relatios betwee mutual fud flows ad stock market returs i Korea. Joural of Iteratioal Fiacial Markets, Istitutios ad Moey, 17, Richard, A. (2005). Big Fish i Small Pods: The Tradig Behavior ad Price Impact of Foreig Ivestors i Asia Emergig Equity Markets. The Joural of Fiacial ad Quatitative Aalysis, 40(1), Sapia, R.Z.Z. & Auzairy N.A. (2015). Foreig Equity Flows ad Market Retur Likages: Evidece of Malaysia Stock Market. Global Busiess Review, 16(5S), 1S-14S. Ülkü, N. & İkizlerli D. (2012). The iteractio betwee foreigers tradig ad emergig stock returs: Evidece from Turkey. Emergig Markets Review, 13, Umutlu, M., Akdeiz, L., & Altay-Salih, A. (2013). Foreig equity tradig ad average stock-retur volatility. The World Ecoomy, 36, Wag, J. (2007). Foreig equity tradig ad emergig market volatility: Evidece from Idoesia ad Thailad. Joural of Developmet Ecoomics, 84, Warther, V. A. (1995). Aggregate mutual fud flows ad security returs. Joural of Fiacial Ecoomics, 39, Watso, J., & Wickramaayake, J. (2012). The relatioship betwee aggregate maaged fud flows ad share market returs i Australia. Joural of Iteratioal Fiacial Markets, Istitutios ad Moey, 22,