Are Momentum Strategies Profitable? Evidence from Singapore

Transcription

1 Are Momentum Strategies Profitable? Evidence from Singapore Vikash Ramiah a, Tony Naughton b and Madhu Veeraraghavan c a,b School of Economics, Finance and Marketing, RMIT, GPO Box 2476V, Melbourne, 3001, Australia. c Department of Accounting and Finance, Monash University, Clayton Campus, Victoria 3800, Australia. 1 Electronic copy available at:

2 Are Momentum Strategies Profitable? Evidence from Singapore Abstract We investigate the profitability of momentum investment strategies for equities listed on the Stock Exchange of Singapore (SGX-MAINBOARD), second board (SGX- SESDAQ) and a combined board. We also investigate the relationship between stock returns and past trading volume for equities listed in the Singapore stock exchange. We report significant momentum profits for the period investigated in this paper and document that momentum is a persistent feature of stock returns for all three boards. We also document that momentum portfolios earn returns as high as 14.1% per month on the secondary board, 13.29% on the combined board and 11.22% on the main board. JEL Classification: G11, G12, G15 Keywords: Momentum, Turnover Ratio, Past Returns, Singapore, Multifactor Model 2 Electronic copy available at:

3 I. Introduction Jegadeesh and Titman (1993) define a momentum strategy as one where arbitrageurs buy well-performing stocks and sell poor-performing stocks. Research on identifying profitable momentum strategies has expanded rapidly in recent years. For instance, Rouwenhorst (1998) reports significant momentum returns from a sample of 12 European countries and his subsequent 1999 paper confirms the existence of momentum strategies in emerging markets. Lee and Swaminathan (2000) find that both momentum and trading volume appear to predict subsequent returns for equities listed in the US equity market. Chan, Hameed, and Tong (2000) document significant evidence of momentum profits based on 23 stock market indices. Hameed and Kusnadi (2002) report statistically significant momentum profits from Asian markets. The relationship between volume and momentum is a particular feature of the literature. For example, Lee and Swaminathan (2000) and Connolly and Stivers (2003) show that past trading volume can provide an important link between momentum and value strategies as past trading volume can predict the magnitude and persistence of price momentum. In a similar vein, Blume, Easley, and O Hara (1994) and many others 1 argue that traders can obtain valuable information about stock returns by observing past price and past trading volume. Behavioural finance theories that explain momentum effects include expectation extrapolation (DeLong et al (1990)), conservatism in expectations updating (Barberis et al (1998)), biased self-attribution (Daniel et al (1998)), disposition effect (Grinblatt and Han (2001)), and selective information conditioning (Hong and Stein (1999)). 1 Brown (1999), Odean (1998), Dow and Gorton (1997), Blume and Easley (1994), Lakonishok, Shleifer, and Vishny (1992), Shefrin and Statman (1994), Karpoff (1986), Kyle (1985), Varian (1985) and Grossman and Stiglitz (1980). 3

4 Menkhoff, and Schmidt (2005) describe momentum traders as the least risk-averse professionals, going aggressively with the trend. Conrad and Kaul (1998) show that momentum profits can be explained by the cross sectional difference in individual stocks expected returns. However, Fama and French (1996), using the three-factor model, fail to explain the abnormal returns and conclude that momentum remains a puzzle requiring further investigation. Chordia and Shivakumar (2000) investigate the influence of macroeconomy on momentum and report that momentum strategies perform well during periods in which the macroeconomic state is favourable, while it does not during recessions. In this paper, we not only extend the literature on momentum trading by providing international evidence but we are the first to investigate the profitability of momentum strategies for a market with multiple boards. We investigate various momentum trading strategies and also consider the role of trading volume for equities listed on the Stock Exchange of Singapore. In a multiple board scenario, it is important for momentum investors to identify where momentum strategies are most profitable. The Stock Exchange of Singapore is a very good example of a multiple board system. With a stable political setting and economic fundamentals coupled with conducive business and regulatory environment, Singapore is widely regarded as a premier equity market in the Asia Pacific region. The equity market is split into two segments namely the SGX-MAINBOARD (main board) and the SGX-SESDAQ (second board). The main board primarily attracts large companies with criteria such as cumulative pre-tax profits of at least $7.5 million over the last three years or market capitalisation of at least $80 million at the time of the initial public offering. The second board attracts smaller companies, 4

5 foreign or local, in raising funds from the stock market. There are no quantitative requirements for a listing on SGX-SESDAQ. Companies listed on the second board may apply for transfer to the main board once the main board requirements are met. In addition to investigating the momentum strategies for main and second board we also investigate the momentum strategies for a combined board. In sum, we investigate whether investors can generate superior returns by investing in strategies unrelated to market movements. Our analysis suggests that returns are driven by momentum phenomenon. This is because all three boards exhibit a very strong momentum effect. We find that on average, a zero cost portfolio that invests in past winners and sells past losers earns returns as high as 14.1% per month. Not surprisingly this return was recorded in the second board and also the returns in the second board were consistently higher than the main board. Interestingly, when replicating the trading volume sort we find that trading volume does not play an explicit role in predicting future returns of stocks in medium term periods. However, trading volume does contain information about the extent of the continuation of stock returns in longer holding periods. In other words, trading volume can help predict the persistence and the reversal of momentum pattern in holding periods beyond one year. The rest of the paper is organised as follows: In Section II we present the data and methods used in this paper. Section III presents the empirical findings while Section IV concludes the paper. 5

6 II. Data and Methods Data Monthly stock return index, trading volume and the number of outstanding shares for the period 1990 to 2004 are obtained from Datastream. We have a total of 590 stocks in the combined board of which 437 firms are listed on the main board and the rest on the second board. The monthly average of these variables for the entire period is calculated and Table 1 shows the descriptive statistics of the above variables for the three different boards. On average the combined monthly return and main board monthly return in Singapore is not statistically different from zero 2 while the second board generates negative returns. A further analysis of the return distribution shows that the monthly returns are not normally distributed in any of these boards with evidence that returns are negatively skewed and leptokurtic. It is also evident from Table 1 that volume traded and number of shares is significantly higher on the main board. Datastream and Kenneth French s website 3 were used to collect data on size factor, book-to-market factor, risk free rate and Singapore market indices. We used the three-month Singapore Treasury Bill rate as the risk free rate, and Singapore Strait Times Price Index (new) as the proxy for the market. To overcome 4 the practical problem of applying the Fama and French Model to markets outside the United States, we used the Singapore SE UOB SESDAQ price index as a proxy for small firms and the Singapore equities mainboard price index as a proxy for big firms. We 2 The Strait Times Index was around the 1500 points in January 1990 and went up to 2066 points by end of Data on the Sesdaq and Main board were not available and through our index replication, it was possible to calculate the return on these two boards Faff (2004) argues that there exists useful proxies for the Fama and French factors that can be easily constructed from off the shelf index data. In his paper he uses the Russell Indexes to construct SMB and HML portfolios. 6

7 define SMB as the difference between the returns on the second board and the main board. Given that Kenneth French s website reports the returns for HML we downloaded the returns from French s website. Methodology We define monthly return as follows: ( SRI it SRI it MR = 1) i (1) SRI it 1 Where MR i is the monthly return for stock i. SRI it is the stock return index for stock i at time t. SRI it-1 is the stock return index for stock i at time t-1. Trading volume is defined as the average monthly turnover ratio where the monthly turnover ratio is obtained by dividing the monthly trading volume of a stock by the number of shares of the same stock at the end of the month. Many studies have used turnover ratio as a consistent measure of trading volume since raw trading volume is not scaled and highly likely to be correlated with size. 5 Our portfolio construction is similar to that of Lee and Swaminathan (2000). Portfolios are formed on a monthly basis. At the beginning of each month from March 1990 to December 2004, we rank all eligible stocks independently on the basis of past returns for the return momentum. The stocks are then assigned to one of ten decile portfolios based on their returns over the past J months (where J = 1, 3, 6, 9 and 12 months respectively). Next the portfolios are held for K months (where K = 1, 3, 6, 9 and 12 months). Returns for K-month holding period are based on equally weighted average returns of every stock in the portfolios. For example, the monthly return for 5 See, Campbell, Grossman, and Wang (1993) and Lee and Swaminathan (2000). 7

8 a three-month holding portfolio is the average of the portfolio return from this month s strategy, last month s strategy and strategy from two months ago. We focus on the extreme winner and loser deciles over the next K months and next 14 years. The strategies are to buy the winner portfolio and sell the loser portfolio for different holding and formation periods. The winner and loser portfolios are then subcategorized into three other portfolios namely, high volume (H), medium volume (M) and low volume (L). The stocks within each decile are split into three other tertiles (H, M, L) based on average monthly trading volume during the J-month estimation period. Our definition of trading volume and the criteria to classify high and low trading volume stocks are based on stock turnover ratios as described above. The high, medium and low portfolios within each tertile refer to stocks with smallest to largest trading volume. The strategy is to long the high volume traded portfolios and short the low volume traded in each decile. Therefore, H-L return can be calculated for each decile. When these returns are positive (negative) we can conclude that, conditional on past returns, high volume stocks generally perform better (worst) than low volume stocks. Liu et al. (1999) report the importance of the three-factor model in explaining momentum profits and found that the three-factor model captures momentum better than CAPM. A time series analysis similar to Fama and French (1996), Naranjo et al. (1998), Heston et al. (1999) and Faff (2004) is used in this approach. Thus, we regress the momentum portfolio returns on the overall market factor, size and book to market equity factors. We also regress the returns of winners and losers on the market, firm size and book-to-market factors. R pt -R Ft = α p + β p (R Mt - R Ft ) + S p (SMB t ) + H p (HML t ) + e pt (2) 8

9 Where R pt is the return of portfolio in month t, R Ft is the risk free asset in month t, and R Mt is the return on the market proxy in the month. R pt -R Ft is the excess return on the portfolio and R Mt - R Ft is the excess return on the market portfolio. SMB represents the mimic portfolio for the size factor and HML the mimic portfolio for the book-tomarket factor. III. Empirical Findings This section reports the returns for different momentum and volume-based momentum strategies. We confirm strong momentum behaviour in that momentum effects are present in all three boards with the strongest effect for equities listed in the second board. Interestingly, we find no evidence of a relationship between stock returns and trading volume over the medium term holding period. However, we do find evidence that trading volume can predict the timing of reversal of the momentum phenomenon. Simple Strategies Tables 2, 3 and 4 summarise the empirical results from several momentum strategies in the different markets. Following, Lee and Swaminathan (2000) we report the mean return from a dollar neutral strategy of buying extreme winners and shorting extreme losers, R10-R1. At the beginning of each month, stocks are ranked and grouped into deciles on the basis of their returns over the previous 1, 3, 6, 9 and 12 months. Thus, there are 10 portfolios ranging from top winners to worst losers every month from January 1990 to December We report results for the extreme losers (R1) and the winner (R10). In each month, we also long the winner portfolio and short the loser portfolio and the returns of this zero cost portfolio is shown as R10-R1. The results in Tables 2, 3 and 4 suggest a clear and consistent momentum effect for equities listed in the Singapore Stock 9

10 Exchange. With the exception of J=1 formation period, returns at the other formation dates for winner portfolios are significantly larger than those of stocks in the loser portfolio. These results are consistent across the three different boards. Columns 3 to 20 report the equal-weighted average monthly returns over the next K months (K=1, 3, 6, 9, 12) for portfolios formed based on J months. For example on the second board (see Table 4), when J=6 and K=6, with a six month portfolio formation period, past losers on average lose 1.78 % over the next six months while past winners on average gain 5.24% over the same period. The zero cost portfolio which short the loser and long the winner in this case earns 7.02% over six months. This return translates to an annual return of around 14%. With the exception of the J=1 formation period, the differences in monthly returns between winner and loser portfolios are positive and significant in every combination of K and J in the short run. This result is consistent across the different boards. On average, these differences are very high. For example on the second board, the zero cost portfolio earns 14.10% and 12.69% when K=1 and J=6 or K=1 and J=9 respectively. The last 10 columns of Tables 2, 3 and 4 report the monthly returns for each portfolio for up to fourteen years following the portfolio formation. We find that the momentum effect lasts up to one year for portfolios formed based on past 1, 3, 6, 9 and 12 months on the main board and combined board while it takes a longer period to disappear in the second board. From year 2 onwards, a reversal pattern is observed for these portfolios. The zero cost portfolios consistently produce around zero percent returns after one year for main and combined board whilst it returns around 2% even after 5 years in the secondary market. It can be concluded that the reversal pattern is slower in the secondary market than the other markets. For those portfolios formed based on 1 10

11 and 3 months on the main and combined board, a reversal pattern is seen straight after 9 months and the zero cost portfolios generate almost zero returns for the next fourteen years of holding. Thus, there is evidence that shorter the formation period, the quicker the reversal of the momentum effect in the main and combined board while the secondary board does not exhibit any of these characteristics. Furthermore, it seems that the reversal gets larger over time as R10-R1 generates lower returns for longer holding periods in all the three boards. The momentum returns based on our range of formation periods for the Main board are shown in Table 2 and then graphically in Exhibit IA. With the exception of the one month formation period (J=1), all other formation periods show the highest returns in the first month of holding followed by a steady decline. For formation period of 6 months to 12 months, the mean reversal process is slower than the remaining formation periods. Momentum returns are calculated as the returns on the extreme winner portfolios minus the returns on the extreme loser portfolios (R10-R1). For winner portfolios, the returns reach their best at K=1 (except for J=1) and decline if the holding period is extended to 6, 9 and 12 months. They all converge to about 1% return before year 5 and remain around this percentage return (see Exhibit 1B). It is noticeable from Exhibit 1C that when the holding period is extended to more than 1 month, returns for loser portfolios (R1) generally increase. In other words, for every J period other than 1, the returns of R1 portfolios are at their worst when they are held for one month (K=1). The returns for formation periods which are greater than 6, the negative returns turn into positive just after one year. As for the returns for J=3, it has a quicker reversal of just three months. The return for the one-month formation period is always positive and just like other formation periods approaches the 1% return. Another implication of these exhibits is that the Lee and Swaminathan (2000) momentum strategy does not work for the one-month formation period. Instead 11

12 investors should have a long position on both the winner and loser portfolios. Furthermore the same strategy should be adopted for holding periods of more than two years. As the main board heavily influences the combined board, similar findings are observed on the combined board. Table 3, Exhibits 2A, 2B, and 2C illustrate these results. We now proceed to analyse the second board. The momentum returns based on our range of formation periods are shown in Table 4 and then graphically in Exhibit 3A. With the exception of the one month formation period (J=1), all other formation periods show high returns in the first month of holding followed by a decline. All formation periods reverse to a mean return of about 3% in year 2 to then steadily decrease over the next twelve years. The one-month formation period exhibits a different pattern. It yields a negative return in the first three holding periods and then approaches the 3% return in year 2. Momentum returns are slightly higher in the second market than the main board. For winner portfolios, the returns reach their best at K=1 (except for J=1) and decline if the holding period is extended. Over a twelve-year period, we can still observe a return of about 1% to 2% (see Exhibit 3B). It is clear from Exhibit 3C that when the holding period is extended to more than 1 month, returns for loser portfolios (R1) generally increase. In other words, for every J period other than the one-month formation period, the returns of R1 portfolios are at their worst when they are held for one month (K=1). One month and three-month formation periods reverse to zero return in one year while it takes another year for the remaining periods to join them. We suggest that for a formation period of one month and holding period of one month, investors should reverse their strategy, i.e. go long on the loser and short on the winner portfolios. Furthermore for holding periods (i.e. for K>2 years), investors 12

13 can adopt a one side momentum strategy, i.e. they only have to buy the winner portfolios. Thus, we can conclude that there are different strategies in place in the main and second boards. While a zero-cost Lee and Swaminathan (2000) momentum strategy does work in both markets, it is not the optimal momentum strategy in any of these markets in this particular case. Another interesting strategy is to long the winners in the one market and short the losers in the other market. For instance, for formation period of 6 months, if we adopt a long position in the main board and the short position in the second board, we can earn 12% (i.e. 7.69% from the main board and 4.31% from the secondary board). On the other hand, a long position in the second board and short position in the main board will yield 13.32%. Stock Returns and Past Trading Volume In this section we examine whether there is any relationship between stock returns and past trading volume for equities listed in the SGX. Tables 5, 6 and 7 report returns for portfolios formed on the basis of a two-way sort between past returns and past trading volume on the three different boards. So far, most of our findings are consistent with previous studies in this respect. However, when we take trading volume into consideration, we find that trading volume does not help predict stock returns. In this respect our findings challenge prior research in this area. Several interesting results can be observed in these tables. First, Lee and Swaminathan (2000) find that conditional on past returns, lower volume stocks generally perform better than high volume stocks over the next 3, 6, 9 and 12 months. This result is illustrated by the consistently negative returns to the H (high volume) - L (low volume) portfolios in their study. However, our results do not indicate any consistent negative returns for H-L. Conditional on past returns there is no evidence that low volume stocks will outperform high volume stocks over the next 12 months on any of these boards. Although we can observe that L-M generates 13

14 negative returns on the losing portfolios (R1), the values are not statistically different from zero. There are also cases where L-M generates positive returns and they usually appear in the R10. Therefore, our findings do not provide a consistent picture to support Lee and Swaminathan (2000) but is very consistent with Naughton et al. (2005) who investigate the Chinese market. In Tables 5, 6 and 7 we also report returns of R10-R1. Again, while Lee and Swaminathan (2000) and Scott, Stumpp and Xu (2003) find that price momentum is more pronounced among high volume stocks we do not find such a pattern in our sample. For example, when K=3 and J=12, the high volume zero cost portfolio R10H- R1H earns 9.32% on the main board, 11.17% on the second board and 10.23% on the combined board. On the other hand, the low volume zero cost portfolio R10L-R1L only earns 7.27% on the main board, 8.82% on the secondary board and 8.42% on the combined board. However, when K=J=6, the low volume zero cost portfolio R10L-R1L earns 3.97% on the main board, 6.05% on the secondary board and 5.53% on the combined board. Lee and Swaminathan interpret the relationship between returns and volume as an illiquidity premium. They suggest that portfolios with lower liquidity earn higher expected returns. However, they also find that momentum premium is higher in high volume portfolios which are arguably more liquid. Scott, Stumpp, and Xu (2003) argue that the relation between momentum effect and high volume stocks is a result of the underreaction of investors to earnings news and the effect is most pronounced among high growth industries. They propose that the interaction between momentum and volume will disappear once earnings related news and stocks growth rates are controlled. Our results in Tables 5, 6 and 7 show no evidence that high volume portfolios earn better momentum premium. Thus, we conclude that past trading 14

15 volume does not play an important role in predicting future returns for Singapore equities over the one-year horizon. Application of the Three-Factor Model Table 8 shows basic descriptive statistics of the proxies used in the Fama and French regressions. Our results show that the SMB portfolio generates a negative return whereas the HML portfolio generates a positive return. Some really interesting results are observed on the main board (see Table 9). First, with the exception of the one-month formation period, the intercept term (alpha) is statistically different from zero. Alpha for loser portfolios are negative while positive for both winner and momentum portfolios. It appears that there exists a pattern for these three types on portfolios on the main board for formation periods greater than one month. Loser portfolios have a tendency to produce negative risk-adjusted performance while winners and momentum portfolios follow a positive one. These patterns increase with the formation period and decrease after the six-month formation period once more reinforcing the notion of an optimal formation period of six months. The size factor consistently explains the variations in the portfolio returns of both winners and losers. In most cases the sign of the size coefficient is positive. Another important observation is that for longer formation periods, SMB loses its importance for shorter holding periods (when K is less than or equal to three). Whilst the size factor explains the variations in the portfolio returns of winners and losers, it does not explain the momentum returns. On the other hand, the book-to-market factor hardly explains any of these portfolio returns. The three-factor model implies the existence of a momentum pattern on the main board but does not explain this phenomenon. Table 10 reports the findings of the three-factor model on the secondary board. First, the intercept term (alpha) is statistically different from zero. Similar to the findings on 15

16 the main board the alpha for loser portfolios are negative. Winner and momentum portfolios exhibit negative alpha for formation periods of three months or less while becoming positive for longer formation periods. Once more we can observe a pattern for these three types on portfolios. Loser portfolios have a tendency to produce negative risk-adjusted performance while winners and momentum portfolios follow a positive or negative depending on the length of the formation period. These patterns increase with the formation period and decrease after the six-month formation period once more reinforcing the notion of an optimal formation period of six months. Furthermore, alpha is significant mostly in holding periods of 6 months or less. The size factor consistently explains the variations in the portfolio returns of both winners and losers. In most cases the sign of the size coefficient are positive. Contrary to the main board SMB better explains the returns for longer holding periods (when K is greater than or equal to six). Whilst the size explains the variations in the portfolio returns of winners and losers, it does not explain the momentum returns. Surprisingly the book-to-market factor is no longer redundant in explaining the returns for the one-month formation periods (even for the momentum portfolios). However, this factor does not work for longer formation periods. The three-factor model indicates the existence of a momentum pattern on the secondary board but fails to explain it. IV. Conclusions In this paper, we investigate various momentum trading strategies for equities listed on the Singapore Stock Exchange. We also consider the role of trading volume and use different formation and holding periods. We find evidence of substantial momentum profits during the period 1990 to A zero cost portfolio that goes long in past winners and short in past losers on average can earn up to 14% per month. We also report that returns in the second board were consistently higher than 16

17 the main board. Our results challenge the literature as we find that past trading volume does not play an explicit role in predicting future returns of stocks in shorter term horizons. However, trading volume does contain information about the extent of the continuation of stock returns in longer holding periods. In other words, trading volume can help predict the persistence and the reversal of momentum pattern in holding periods beyond one year. An attempt is also made to explain the momentum effect using the three-factor model. The evidence provided shows that the threefactor model of Fama and French cannot explain the momentum phenomenon. 17

18 References Ball, R., and Kothari, S.P., 1989, Nonstationary expected returns: implications for market efficiency and serial correlations in returns, Journal of Financial Economics 25, Ball, R., S.P. Kothari and Wasley, C.E.,1995, Can we implement research on stock trading rules: the case of short-term contrarian strategies, Journal of Portfolio Management 21, Blume, L., Easley, D., and O Hara, M., 1994, Market statistics and technical analysis: The role of volume, Journal of Finance 49, Brown G., 1999, Volatility, sentiment and noise traders. Financial Analyst Journal. 55, Campbell, J.Y., Grossman, S.J., and Wang, J., 1993, Trading volume and serial correlation in stock returns, Quarterly Journal of Economics 107, Chan, L., K.C., Hamao, Y., and J. Lakonishok, 1991, Fundamentals and stock returns in Japan, Journal of Finance 46, Chan. K., Hameed, A., and Tong, W., 2000, Profitability of momentum strategies in the international equity markets, Journal of Financial and Quantitative Analysis 35, Chordia,T., and Shivakumar, L., 2000, Momentum, business cycle and time varying expected returns, Journal of Finance 57, Cleary, S., and Inglis, M.,1998, Momentum in Canadian stock returns, Canadian Journal of Administrative Sciences 15, Connolly, R., and Stivers, C., 2003, Momentum and reversals in equity-index returns during periods of abnormal turnover and return dispersion, Journal of Finance 58, Conrad, J.S., and Kaul, G., 1998, An anatomy of trading strategies, Review of Financial Studies 11,

19 Daniel, K., Hirshleifer, D. and Subrahmanyam, A., 1998, Investor psychology and security market under and overreactions, Journal of Finance 53, Datar, V. T., Naik, N. Y. and Radcliffe, R., Liquidity and stock returns: An alternative test, Journal of Financial Markets 1, Demir I., Muthuswamy J., Walter T., 2004, Momentum returns in Australian Equities: The influences of size, risk, liquidity and return computation, Pacific-Basin Finance Journal 12, Dow J. and Gorton G. 1997, Noise trading, delegated portfolio management, and economic welfare. Journal of Political Economy 105, Drew, M.E., Naughton, T. and Veeraraghavan, M., 2003, Firm size, Book-to-market equity and security returns: Evidence from the Shanghai Stock Exchange, Australian Journal of Management 28, Drew, M., Veeraraghavan M., and Ye M., 2006, Do Momentum Strategies Work? Australian Evidence, forthcoming in Managerial Finance. Ellis M., and Dylan T. C., 2004, Momentum and the FTSE 350. Journal of Asset Management 5, Faff R. 2004, A simple test of the Fama and French model using daily data: Australian evidence. Applied Financial Economics, , Fama, E. and French, K. 1993, Common risk factors in the returns on stocks and bonds. Journal of Financial Economics 33, Fama, E. and French, K. 1996, Multifactor explanations of asset pricing anomalies. Journal of Finance 51, Fama, E.F. and French, K.R. 1998, Value versus growth: The international evidence, Journal of Finance 53, Fama, E.F., and K.R. French, 1996, Multifactor explanations of asset pricing anomalies, Journal of Finance 50,

20 Fama, E.F.,1991, Efficient Capital Markets :II, Journal of Finance 46, Griffin, J.M., Ji, X. and Martin, J.S., 2003, Momentum Investing and Business Cycle Risk: Evidence from Pole to Pole, Journal of Finance 58, Grinblatt, M., Titman, S., Wermers, R., 1995, Momentum investment strategies, portfolio performance and herding: A study of mutual fund behaviour, American Economic Review 85, Grossman, S. and Stiglitz, J. 1980, On the impossibility of informationally efficient markets. American Economic Review 70, Hameed, A., and Kusnadi, K., 2002, Momentum strategies: Evidence from the Pacific Basin Stock Markets, Journal of Financial Research 25, Heston, S., Rouwenhorst, K. and Wessels, R. 1999, The role of beta and size in the cross-section of European stock returns. European Financial Management 5, Hogan S., Jarrow R., Teo M. and Warachka M., 2004, Testing market efficiency using statistical arbitrage with applications to momentum and value strategies Journal of Financial Economics 73, Hong, H., and Stein, J.C., 1997, A unified theory of underreaction, momentum trading and overreaction in asset markets, NBER working paper. Hovey, M., Li, L. and T. Naughton 2003, The Relationship between Valuation and Ownership of Listed firms in China, Corporate Governance: An International Review 11, Jegadeesh, N., and Titman S., 1993, Returns to buying winners and selling losers: implications for stock market efficiency, Journal of Finance 48, Kaminsky G., Lyons R. K. and Schmukler S. L. 2004, Managers, investors, and crises: mutual fund strategies in emerging markets Journal of International Economics 64,

21 Kang, J., Liu M.H. and Ni, S.X., 2002, Contrarian and momentum strategies in the China stock market: , Pacific-Basin Finance Journal 10, Karpoff, J. 1986, A theory of trading volume. The Journal of Finance 41, Korajczyk, R. A. and Sadka, R., 2004, Are Momentum Profits Robust to Trading Costs? Journal of Finance 59, Kyle, A. 1985, Continuous auction and insider trading. Econometrica 53, Lakonishok, J., Shleifer, A., Vishny, R., 1994, Contrarian investment, extrapolation, and risk, Journal of Finance 49, Lakonishok, J., Shleifer, A., and Vishny, R. 1992, The structure and performance of the money management industry. Brookings Papers on Economic Activity Microeconomics Lee, C. M.C, and Swaminathan, B. 2000, Price momentum and trading volume, Journal of Finance 55, Liu, C., and Lee, Y., 2001, Does momentum strategy work universally? Evidence from the Japanese stock market, Asia-Pacific Financial Markets 8, Liu, W., Strong N., and Xu X., 1999, The Profitability of Momentum Investing, Journal of Business Finance & Accounting, 26, Lo, A., and MacKinlay, C., 1990, When are contrarian profits due to stock market overreaction?, Review of Financial Studies 3, MacNeil, L., 2002, Adaptation and convergence in corporate governance: the case of Chinese listed companies, University of Aberdeen, working paper. Maher, M., and Andersson, T., 2000, Corporate governance: Effects on firm performance and economic growth, Oxford University Press. Menkhoff l., and Schmidt. U. 2005, The use of trading strategies by fund managers: some first survey evidence. Applied Economics 37, Mok, H., and Hui, Y.V., 1998, Underpricing and aftermarket performance of IPOs in Shanghai, Pacific Basin Finance Journal 6,

22 Moskowitz, T. J., and Grinblatt, M., 1999, Do industries explain momentum? Journal of Finance 54, Naranjo, A., Nimalendran, M. and Ryngaert, M. 1998, Stock returns, dividend yields, and taxes Journal of Finance 53, Naughton A., Truong C. and Veeraraghavan, M., 2005, Yes Momentum Strategies Work? Chinese Evidence, Working Paper, School of Economics, Finance and Marketing, Royal Melbourne Institute of Technology. Naughton, T., and Hovey, M., 2002, What part do securities markets play in advancing corporate governance in Asia?, Australasian meeting of the Econometrics Society (ESAM02), Brisbane, Australia, July 7-10, O Neal, E. 2000, Industry momentum and sector mutual funds, Financial Analysts Journal 56, Odean T. 1998, Volume, volatility, price, and profit when all traders are above average. The Journal of Finance. 53, Rouwenhorst, K.G., 1998, International momentum strategies, Journal of Finance 53, Rouwenhorst, K.G., 1999, Local return factors and turnover in emerging stock markets, Journal of Finance 54, Schiereck, D., DeBondt, W. and Weber, M. 1999, Contrarian and momentum strategies in Germany, Financial Analysts Journal 55, Scott, J., Stump, M., and Xu, P., 2003, News, not trading volume, builds momentum, Financial Analysts Journal 59, Shefrin, H. and Statman, M. 1994, Behavioral Capital Asset Pricing Theory. Journal of Financial and Quantitative Analysis. 29, Su, D., and Fleisher, B.M., 1999, An empirical investigation of underpricing in Chinese IPOs, Pacific Basin Finance Journal 7,

23 Varian, H. 1985, Divergence of Opinion in Complete Markets: A note, Journal of Finance. 40, Wang, J., 1994, A model of competitive stock trading volume, Journal of Political Economy 102,

24 Table 1: Descriptive Statistics of Return, Volume, and Number of Shares (NOSH) for the three Boards from 1990 to 2004 Combined Board Main Board Secondary Board Return Volume NOSH Turnover Return Volume NOSH Turnover Return Volume NOSH Turnover (000's) (000's) Ratio (000's) (000's) Ratio (000's) (000's) Ratio Mean Standard error Median Standard Deviation Excess Kurtosis Skewness Range Minimum Maximum No. Of Stocks t-test Statistic* JB-Statistic * Testing if the Monthly Mean return is statistically different from Zero 24

25 Table 2: Price Momentum Portfolios and Stock Returns - Combined Board This table presents average monthly returns for the time period 1990 to R1 represents the loser portfolio and R10 the winner portfolio. K represents monthly holding periods where K=1,3,6,9 or 12 months. Returns are average monthly returns over the portfolio formation period. 1 to 14 represent the average monthly return for portfolios held for 1 to 14 years. The number in italics are t- values. J K=1 K=3 K=6 K=9 K= J=1 1 R1 1.94% 1.41% 1.22% 1.09% 1.20% 1.13% 1.10% 1.05% 1.02% 0.90% 0.83% T-Stats R % 0.93% 0.91% 1.05% 1.03% 0.88% 0.92% 0.89% 0.86% 0.85% 0.86% T-Stats R10-R1-1.35% -0.48% -0.31% -0.04% -0.17% -0.25% -0.18% -0.16% -0.16% -0.05% 0.02% T-Stats J=3 3 R1-1.91% 0.04% 0.69% 0.70% 1.01% 1.03% 1.04% 1.02% 1.03% 0.95% 0.80% T-Stats R % 3.39% 2.14% 1.93% 1.59% 1.19% 1.07% 0.97% 0.96% 0.96% 1.03% T-Stats R10-R1 8.52% 3.35% 1.44% 1.23% 0.58% 0.16% 0.03% -0.05% -0.07% 0.01% 0.24% T-Stats J=6 6 R1-4.55% -3.47% -1.80% -0.81% -0.22% 0.49% 0.61% 0.74% 0.83% 0.74% 0.48% T-Stats R % 7.22% 4.99% 3.68% 2.92% 1.69% 1.39% 1.15% 1.08% 0.99% 1.18% T-Stats R10-R % 10.69% 6.79% 4.49% 3.14% 1.19% 0.79% 0.41% 0.26% 0.25% 0.70% T-Stats J=9 9 R1-4.03% -3.36% -2.35% -1.33% -0.62% 0.30% 0.48% 0.65% 0.74% 0.65% 0.68% 25

26 T-Stats R % 7.07% 5.80% 4.49% 3.51% 1.98% 1.54% 1.27% 1.18% 0.98% 1.32% T-Stats R10-R % 10.43% 8.15% 5.82% 4.12% 1.67% 1.06% 0.62% 0.43% 0.33% 0.64% T-Stats J=12 12 R1-1.94% -1.99% -1.56% -1.07% -0.46% 0.34% 0.47% 0.65% 0.72% 0.71% N/A T-Stats N/A R % 5.25% 4.60% 3.94% 3.30% 1.96% 1.58% 1.31% 1.20% 0.99% N/A T-Stats N/A R10-R1 7.42% 7.23% 6.15% 5.00% 3.76% 1.61% 1.11% 0.67% 0.48% 0.28% N/A T-Stats N/A 26

27 Table 3: Price Momentum Portfolios and Stock Returns - Main Board This table presents average monthly returns for the time period 1990 to R1 represents the loser portfolio and R10 the winner portfolio. K represents monthly holding periods where K=1,3,6,9 or 12 months. Returns are average monthly returns over the portfolio formation period. 1 to 14 represent the average monthly return for portfolios held for 1 to 14 years. J K=1 K=3 K=6 K=9 K= R1 1.72% 1.33% 1.14% 1.03% 1.18% 1.08% 1.07% 1.02% 0.99% 0.85% 0.86% T-Stats R % 1.06% 1.04% 1.17% 1.13% 0.95% 0.93% 0.90% 0.87% 0.83% 0.84% T-Stats R10-R1-0.99% -0.27% -0.10% 0.14% -0.05% -0.13% -0.14% -0.12% -0.12% -0.03% -0.02% T-Stats R1-1.60% 0.06% 0.64% 0.62% 0.99% 0.99% 1.04% 1.00% 1.00% 0.89% 0.85% T-Stats R % 3.14% 2.04% 1.90% 1.62% 1.21% 1.06% 0.99% 0.96% 0.94% 0.96% T-Stats R10-R1 7.73% 3.08% 1.41% 1.28% 0.63% 0.22% 0.01% -0.01% -0.05% 0.05% 0.11% T-Stats R1-3.53% -2.74% -1.28% -0.47% 0.00% 0.51% 0.65% 0.73% 0.78% 0.72% 0.35% T-Stats R % 6.13% 4.32% 3.24% 2.65% 1.53% 1.23% 1.02% 0.96% 0.88% 0.74% T-Stats R10-R % 8.87% 5.60% 3.70% 2.65% 1.02% 0.58% 0.29% 0.18% 0.16% 0.39% T-Stats

28 9 R1-3.54% -2.93% -2.01% -1.10% -0.46% 0.31% 0.47% 0.61% 0.69% 0.53% 0.13% T-Stats R % 6.29% 5.19% 4.06% 3.24% 1.85% 1.39% 1.17% 1.05% 0.83% 0.24% T-Stats R10-R % 9.22% 7.19% 5.16% 3.70% 1.54% 0.92% 0.56% 0.36% 0.30% 0.11% T-Stats R1-3.08% -2.65% -2.01% -1.41% -0.76% 0.25% 0.45% 0.58% 0.65% 0.57% N/A T-Stats N/A R % 5.75% 5.02% 4.27% 3.44% 1.92% 1.38% 1.17% 1.01% 0.79% N/A T-Stats N/A R10-R1 9.32% 8.40% 7.03% 5.67% 4.20% 1.68% 0.94% 0.59% 0.36% 0.22% N/A T-Stats N/A 28

29 Table 4: Price Momentum Portfolios and Stock Returns - Secondary Board This table presents average monthly returns for the time period 1990 to R1 represents the loser portfolio and R10 the winner portfolio. K represents monthly holding periods where K=1,3,6,9 or 12 months. Returns are average monthly returns over the portfolio formation period. 1 to 14 represent the average monthly return for portfolios held for 1 to 14 years. J K=1 K=3 K=6 K=9 K= R1 1.47% 1.33% 1.21% -0.40% 0.00% 0.10% 0.05% 0.16% 0.15% 0.43% -0.06% T-Stats R % 0.57% 0.65% 3.81% 3.22% 2.93% 3.07% 3.01% 3.06% 2.32% 0.91% T-Stats R10-R1-2.38% -0.76% -0.55% 4.21% 3.22% 2.82% 3.02% 2.86% 2.91% 1.89% 0.97% T-Stats R1-3.59% -0.69% 0.83% -0.65% -0.17% 0.06% 0.17% 0.27% 0.30% 0.66% -0.20% T-Stats R % 3.58% 2.22% 4.28% 3.60% 3.35% 3.23% 3.10% 3.13% 2.54% 0.80% T-Stats R10-R % 4.27% 1.39% 4.93% 3.78% 3.30% 3.06% 2.83% 2.83% 1.88% 1.00% T-Stats R1-4.31% -3.34% -1.78% -1.52% -0.85% -0.07% 0.13% 0.29% 0.43% 0.51% 0.38% T-Stats R % 7.93% 5.24% 4.88% 3.96% 2.92% 2.81% 2.46% 2.48% 1.89% 0.40% T-Stats R10-R % 11.27% 7.02% 6.40% 4.81% 2.99% 2.68% 2.17% 2.06% 1.38% 0.02% T-Stats

30 9 R1-3.90% -3.38% -2.38% -2.02% -1.36% -0.17% 0.02% 0.26% 0.28% 0.39% 0.31% T-Stats R % 7.83% 6.28% 5.70% 4.63% 3.17% 2.81% 2.29% 2.21% 1.40% 0.36% T-Stats R10-R % 11.20% 8.66% 7.72% 5.99% 3.34% 2.79% 2.03% 1.94% 1.01% 0.04% T-Stats R1-3.47% -2.99% -2.33% -2.13% -1.48% -0.23% 0.07% 0.19% 0.26% 0.34% N/A T-Stats N/A R % 7.21% 6.24% 5.90% 4.81% 3.18% 2.77% 2.29% 2.14% 1.40% N/A T-Stats N/A R10-R % 10.20% 8.58% 8.03% 6.29% 3.41% 2.70% 2.09% 1.88% 1.06% N/A T-Stats N/A 30

31 Table 5: Returns for Portfolios Sorted on Past Returns and Volume - Combined Board This table presents average monthly returns for portfolios sorted on past returns and past average monthly turnover for the period 1990 to K represents monthly holding periods where K=1,3, 6, 9 or 12 months. R1 represents the loser portfolio and R10 represents the winner portfolio. L represents the lowest trading volume portfolio, M represents the medium trading volume portfolio and H the highest trading volume portfolio. K=1 K=3 K=6 K=9 K=12 J Portfolio L M H H-L L M H H-L L M H H-L L M H H-L L M H H-L 1 R1 3.33% 1.96% 0.76% -2.57% 1.89% 1.42% 0.76% -1.13% 1.54% 1.30% 0.90% -0.64% 1.45% 1.08% 0.99% -0.47% 1.47% 1.22% 1.07% -0.40% T-stats R % 1.59% 1.78% 3.57% 0.36% 1.19% 1.46% 1.10% 0.68% 1.02% 1.16% 0.48% 1.05% 1.23% 1.18% 0.13% 1.16% 1.18% 1.06% -0.11% T-stats R10-R1-5.20% -0.65% 1.50% 6.71% -1.64% -0.38% 0.98% 2.62% -0.91% -0.31% 0.39% 1.29% -0.45% 0.14% 0.34% 0.79% -0.41% -0.09% 0.09% 0.50% T-stats R1-0.74% -2.04% -2.72% -1.98% 0.30% -0.02% 0.00% -0.30% 0.00% 0.88% 0.55% 0.55% 0.79% 0.74% 0.53% -0.26% 1.01% 1.08% 0.77% -0.24% T-stats R % 5.71% 7.91% 3.87% 2.08% 2.94% 3.93% 1.84% 0.00% 0.00% 2.37% 2.37% 1.42% 2.05% 2.04% 0.62% 1.27% 1.72% 1.58% 0.31% T-stats R10-R1 4.85% 7.82% 10.74% 5.89% 1.87% 3.02% 0.00% -1.87% 0.81% 1.35% 1.83% 1.02% 0.71% 1.34% 1.53% 0.81% 0.30% 0.63% 0.84% 0.54% T-stats R1-4.31% -4.78% -4.78% -0.47% -3.36% -3.61% -3.49% -0.13% -1.84% -1.84% -1.51% 0.33% -0.89% -0.81% -0.46% 0.42% -0.30% -0.22% 0.10% 0.39% T-stats R % 8.14% 9.69% 2.65% 5.51% 6.66% 8.18% 2.67% 3.62% 4.67% 5.77% 2.15% 2.82% 3.62% 4.14% 1.31% 2.35% 2.90% 3.25% 0.89% T-stats R10-R % 12.76% 13.98% 2.49% 8.99% 10.11% 11.22% 2.23% 5.53% 6.44% 7.00% 1.47% 3.71% 4.36% 4.39% 0.68% 2.59% 3.04% 3.00% 0.41% T-stats R1-3.88% -4.19% -4.44% -0.56% -3.29% -3.45% -3.62% -0.34% -2.37% -2.31% -2.51% -0.15% -1.44% -1.22% -1.36% 0.08% -0.75% -0.52% -0.63% 0.12% T-stats R % 7.25% 8.72% 2.02% 5.76% 6.44% 7.84% 2.08% 4.46% 5.27% 6.54% 2.08% 3.39% 4.09% 5.13% 1.74% 2.77% 3.21% 4.00% 1.22% 31

32 T-stats R10-R % 11.26% 12.53% 1.95% 9.06% 9.71% 10.87% 1.81% 6.83% 7.41% 8.51% 1.68% 4.84% 5.19% 6.07% 1.23% 3.52% 3.65% 4.31% 0.79% T-stats R1-3.37% -3.79% -3.89% -0.52% -3.03% -3.28% -3.28% -0.26% -2.39% -2.52% -2.46% -0.07% -1.78% -1.75% -1.65% 0.13% -1.10% -0.96% -0.83% 0.27% T-stats R % 6.71% 7.88% 1.77% 5.43% 6.15% 7.38% 1.95% 4.46% 5.31% 6.51% 2.05% 3.66% 4.41% 5.59% 1.93% 2.87% 3.46% 4.53% 1.66% T-stats R10-R1 9.53% 10.02% 11.33% 1.80% 8.42% 8.97% 10.23% 1.81% 6.82% 7.42% 8.57% 1.75% 5.41% 5.84% 6.92% 1.51% 3.93% 4.20% 5.12% 1.19% T-stats