ANOMALIES IN MALAYSIA'S EQUITY MARKET: AN INVESTIGATION OF THE PRE-FESTIVAL EFFECT Khong Wye Leong Roy Hong Kheng Ngee Seng Mei Chen Lim Kwee Pheng ABSTRACT Previous researches in finance have mainly concentrated on the study of pre-holiday effects but not many on festival effects on the stock price movement. Malaysia is a multi racial country and its people are made up of mainly Malay, Chinese, Indians and other indigenous people. Because of their differences in cultures and faiths, these people celebrated different religious and cultural festivals. Therefore, the objective of this study is to find out the effects of the 4 main festivals on the stock price movement and in an emerging market like Malaysia. The results based on dummy regressions suggested that positive abnormal returns existed one day prior to Chinese New Year, while only two sectors namely the Finance and Properties sectors exhibited superior returns one day before the Hari Raya Puasa. Keywords: Abnormal Returns, Efficient Market Hy pothesis, Festival Effect INTRODUCTION In an efficient financial market, the price of financial assets reflects all the relevant information. Under this circumstance, investors and speculators are unable to earn any abnormal returns. Despite this, much of the research in finance has focused on ways to 'beat the market'. Since then, various stock market anomalies were documented across the globe especially the seasonal/calendar anomalies such as January effect, weekend effect, day of the week effect, turn of the month effect and preholiday effect. Of particular interest is the pre-holiday effect whereby the stock returns one day prior to holidays is significantly different from other trading days. In recent years, numerous researches were conducted on pre-holiday effect which, mainly focused on developed markets (see for example Lakonishok and Smidt, 1988; Pettengill, 1989; Ariel, 1990; Ziemba, 1991; Cadsby and Ratner, 1992; Kim and Park, 1994; Mills and Coutts, 1995; Arsad and Coutts, 1997; Brockman and Michayluk, 1998; Vergin and McGinnis, 1999) with notable exception of Tan and Wong (1998) and Coutts et al. (2000). In addition, with a background of different cultures and religions, Asians would definitely celebrate different holidays/festivals compare with their western counterparts. Moreover, previous studies on this region have focused only on one festival namely the Chinese New Year. In particular, Davidson and Peker (1996) have found no abnormal returns due to Chinese New Year effect, while Wong et al. (1990) and Yen et al. (2001) have recorded higher returns prior to Chinese New Year. However, as a multi racial country, Malaysians celebrated various festivals. Thus, this study examines whether stock market anomalies exists one day prior to four main festivals celebrated by Malaysians namely the Chinese New Year, Hari Raya Puasa, Deepavali and Christmas. The rest of the paper proceeds as follows. Section 2 describes the methods used. Section 3 is devoted to the empirical results and Section 4 concludes. 281
METHODOLOGY In this study, the data set consists of daily closing values of nine stock indices covering the period from January 1990 to July 2002. These indices comprised of Kuala Lumpur Composite Index and eight others sectoral indices listed on the main board of the exchange. All the data were collected from Daily Dairy of Kuala Lumpur Stock Exchange. Daily returns of each index are calculated as: Pt R t = 1 (1) P t 1 where R t is the daily return in day t and P t is the closing value of day t stock index. In this study, a pre-festival is defined as one trading day prior to the four festivals. In order to test for any existence of pre-festival effect, a dummy regression was employed. Formally, the regression equation is stated as: R t = α 0 + α 1 DCNY t + α 2 DHRP t + α 3 DD t + α 4 DC t + ε t (2) where DCNY t = 1 if day t is one day before Chinese New Year and 0 otherwise; DHRP t = 1 if day t is one day before Hari Raya Puasa and 0 otherwise; DD t = 1 if day t is one day before Deepavali and 0 otherwise; DC t = 1 if day t is one day before Christmas and 0 otherwise and ε t = i.d.d. error term. In order to obtain heteroskedasticity and autocorrelation consistent standard errors, Newey and West (1987) procedures was used. The estimated coefficients (excluding the constant term) will be significant different from zero if the stock returns exhibit a particular pre-festival effect. RESULTS AND DISCUSSIONS The descriptive summary statistics of all the nine sectoral returns are presented in Table 1. In general, the market yields an average return of 2.2% throughout the period examined. The highest return obtained is from the mining sector with an average return of 5.0% while the industrial product and trading and services yielded negative returns. Interestingly, the consumer product recorded the lowest risk. On the other hand, the mining sector documented the highest risk, which is consistent with the high-risk high- return relationship. The results presented in Table 2 and Table 3 served as basis to examine the festival effect. Basically, the findings in Table 2 tests whether a particular festival effect exists using either the 1% or 5% level of significance while Table 3 shows the returns 1 day prior to the festivals and the average returns for all the 9 indices over the 1990-2002 period. Based on Table 2, it is found that for the 9 indices examined, the Chinese New Year (CNY) effect is consistent over the period of 1990-2002. This suggests that abnormal returns occurred in the market before the festival was celebrated. As seen in Table 3, the Composite, Finance, Properties, Consumer Product and Construction indexes had continuously produced positive returns one day prior to the festival while others had a few but minimal negative returns for th e 13-year period. However, all pre- CNY average returns were found to be positive for each of the indexes. Thus, the CNY effect yields positive average returns consistently. For the Hari Raya Puasa effect, it should be noted that in year 2000, the festival was celebrated twice, one in January and the other in December. For the purpose of analysis, the returns one day before Hari Raya Puasa for the year 2000 presented in Table 3 is calculated as an average of the two returns. Based on the findings, only the Finance and Properties indices indicated the Hari Raya Puasa effect, as shown in Table 2. This implies that only these two sectors gained abnormal returns prior to the festival. From Table 3, the abnormal returns can be ascertained as positive values. The average returns for the 13-year period were 0.94% and 1.40% for Finance and Properties sectors respectively. 282
The results in Table 2 show that the Deepavali and Christmas effect are not significant in the 1990-2002 period for all the 9 indices and indicate no superior returns can be earned one day prior to both festivals. CONCLUSIONS In this paper, the existence of the festival effect in the Malaysian stock market was examined. The daily stock returns of the nine indices examined for the period from January 1990 to July 2002 were used to calculate the average returns and the dummy regressions was employed to test for the existence of pre-festival effect. From the empirical results, the evidences have suggested the existence of the Chinese New Year effect and Hari Raya Puasa effect. It is found that the Chinese New Year effect consistently exists throughout the 1990-2002 period in all the sectors, while the Hari Raya effect was detected only in two sectors, namely Finance and Properties sectors. The existence of these anomalies imply that investors can systematically beat the market, obtaining abnormal returns one day prior to Chinese New Year in the market as a whole. In addition, the Finance and Properties sectors can also generate abnormal returns one day prior to Hari Raya Puasa. Thus, the results are partially supported by Wong et al. (1990) and Yen et al. (2001) but contradicted with the Efficient Market Hypothesis (EMH). On the basis of these findings, investors and speculators of the Malaysian stock market can take advantage of this information to obtain superior returns. With the knowledge that the stock prices increase one day before the festival, investors should purchase stocks 2 days before Chinese New Year and to certain extent the Hari Raya Puasa and sell them the next day to yield abnormal returns. REFERENCES Ariel, R. A. (1990). "High Stock Returns before Holidays: Existence and Evidence on Possible Causes". Journal of Finance, 45(6), 1611-1626. Arsad, Z. and Coutts, J. A. (1997). " Security Price Anomalies in the London International Stock Exchange: A 60 Year Perspective". Applied Financial Economics, 7(5), 455-464. Brockman, P. and Michayluk, D. (1998). "The Persistent Holiday Effect: Additional Evidence". Applied Economics Letters, 5(4), 205-209. Cadsby, C. B. and Ratner, M. (1992). "Turn-of-month and Pre-holiday Effects on Stock Returns: Some International Evidence". Journal of Banking and Finance, 16(3), 497-509. Coutts, J. A., Kaplanidis, C. and Roberts, J. (2000). "Security Price Anomalies in an Emerging Market: The Case of the Athens Stock Exchange". Applied Financial Economics, 10(5), 561-571. Davidson, S. and Peker, A. (1996). "Is there a Chinese New Year Effect in Malaysia?". Capital Markets Review, 4(1), 49-61. Kim, C. W. and Park, J. (1994). "Holiday Effects and Stock Returns: Further Evidence". Journal of Financial and Quantitative Analysis, 29(1), 145-157. Lakonishok, J. and Smidt, S. (1988). "Are Seasonal Anomalies Real? A Ninety-Year Perspective". Review of Financial Studies, 1(4), 403-425. Mills, T. C. and Coutts, J. A. (1995). "Calendar Effects in the London Stock Exchange FT-SE Indices". European Journal of Finance, 1(1), 79-93. Newey, W. K. and West, K. D. (1987). "A Simple, Positive Semi-Definite, Heteroskedasticity and Autocorrelation Consistent Covariance Matrix". Econometrica, 55(3), 703-708. Pettengill, G. N. (1989). "Holiday Closing and Security Returns". Journal of Financial Research, 12(1), 57-67. Tan, R. S. K. and Wong, N. T. (1998). "The Diminishing Calendar Anomalies in the Stock Exchange of Singapore". Applied Financial Economics, 8(2), 119-125. 283
Vergin, R. C. and McGinnis, J. (1999). "Revisiting the Holiday Effect: Is it on Holiday?". Applied Financial Economics, 9(5), 477-482. Wong, P. L., Neoh, S. K., Lee, K. H. and Thong, T. S. (1990). "Seasonality in the Malaysian Stock Market". Asia Pacific Journal of Management, 7(Special Issue), 43-62. Yen, G., Lee, C. F., Chen, C. L. and Lin, W. C. (2001). "On the Chinese Lunar New Year Effect in Six Asian Stock Markets: An Empirical Analysis (1991-2000)". Review of Pacific Basin Financial Markets and Policies, 4(4), 463-478. Ziemba, W. T. (1991). "Japanese Security Market Regularities". Japan and the World Economy, 3(2), 119-146. Mean Composite Index 0.02241 1.73579-0.21458 Finance 0.04843 2.00308-0.18588 TABLE 1 Summary of Descriptive Statistics Standard Min Max Number of Deviation Observations Period 0.23143 3098 2 Jan 1990 0.25392 3098 2 Jan 1990 Properties 0.00339 2.52350-0.38099 Mining 0.05085 3.23422-0.34320 Plantation 0.03253 1.79893-0.15345 Consumer Products 0.01870 1.55604-0.15191 Industrial Products - 2.05064-0.01257 0.21955 Construction 0.01827 2.55531-0.20374 Trading and - 2.11513 - Services 0.01568 0.19021 0.44915 3098 2 Jan 1990 0.37330 3098 2 Jan 1990 0.18455 3098 2 Jan 1990 0.17502 2195 1 Sept 1993 0.27704 2195 1 Sept 1993 0.27023 2195 1 Sept 1993 0.25070 2107 3 Jan 1994 284
TABLE 2 Relationships among Pre -Festival Anomalies Coefficients Constant DCNY DHRP DD DC Composite 0.00016 0.00912 ** 0.00199 0.00218 0.00266 (0.47395) (4.61040) (0.62681) (0.58976) (0.97040) Finance 0.00037 0.01273 ** 0.00613 * -0.00049 0.00858 (0.87554) (5.82560) (2.17009) (-0.11342) (1.33106) Properties -0.00007 0.01513 ** 0.01063 ** -0.00275 0.00214 (-0.15792) (4.71202) (3.00314) (-0.58071) (0.58721) Mining 0.00044 0.01415 ** 0.00527 0.00301-0.00604 (0.78551) (4.01322) (1.09172) (0.29373) (-1.22076) Plantation 0.00024 0.01077 * 0.00525 0.00172 0.00346 (0.69919) (2.55125) (1.40558) (0.56113) (0.86008) Consumer Products 0.00013 0.00974 ** 0.00300 0.00133-0.00091 (0.36371) (3.60447) (0.80343) (0.46902) (-0.29853) Industrial Products -0.00022 0.01709 ** -0.00078 0.00371 0.00165 (-0.48960) (3.78321) (-0.12647) (1.22016) (0.42171) Construction 0.00006 0.01293* 0.00650-0.00207 0.01053 (0.10937) (2.55635) (1.22863) (-0.63608) (1.95302) Trading/Services -0.00023 0.01171 ** 0.00088 0.00660-0.00240 (-0.47470) (3.05056) (0.17789) (1.44941) (-0.75232) Notes: The figures in the parentheses represent the t-statistics. ** and * denote significance at 1% and 5% respectively. 285
TABLE 3: Stocks Returns One -Day Prior to Festival Mean Number of Mean Number of Mean Number of (%) +ve -ve (%) +ve -ve (%) +ve -ve Returns Returns Returns Returns Returns Returns 1. Composite Index (1990-2002) 4. Mining (1990-2002) 7. Industrial Products (1993-2002) CNY 0.93486 13 0 CNY 1.58010 11 2 CNY 1.66153 8 1 HRP 0.38810 9 3 HRP 0.89767 9 3 HRP 0.46999 6 2 DE 0.39934 8 4 DE 0.34470 4 8 DE 0.35019 4 5 CH 0.17378 8 4 CH -0.56045 5 7 CH 0.24281 5 4 Others 0.01429 7 6 Others 0.04687 7 6 Others 0.00849 6 4 Total 0.01972 8 5 Total 0.05597 7 6 Total 0.02047 6 4 2. Finance (1990-2002) 5. Plantation (1990-2002) 8. Construction (1993-2002) CNY 1.45199 13 0 CNY 1.22149 9 4 CNY 1.51622 9 0 HRP 0.94425 11 1 HRP 0.79727 11 1 HRP 1.08789 7 1 DE -0.09500 7 5 DE 0.19550 7 5 DE -0.20033 4 5 CH 0.89576 8 4 CH 0.37008 9 3 CH 1.05982 5 4 Others 0.03739 8 5 Others 0.02451 5 8 Others 0.04004 7 3 Total 0.04990 9 4 Total 0.03437 7 6 Total 0.05372 7 3 3. Properties (1990-2002) 6. Consumer Products (1993-2002) 9. Trading and Services (1994-2002) CNY 1.75153 13 0 CNY 1.08752 9 0 CNY 1.17746 8 1 HRP 1.40504 11 1 HRP 0.63817 7 1 HRP 0.45510 6 2 DE -0.36547 6 6 DE 0.25761 5 4 DE 0.63741 6 2 CH 0.20666 7 5 CH -0.13527 3 6 CH -0.26306 3 5 Others -0.00813 6 7 Others 0.04321 6 4 Others -0.02303 4 5 Total 0.00494 6 7 Total 0.08382 7 3 Total -0.01410 6 3 Note: CNY, HRP, DE, CH denote as Chinese New Year, Hari Raya Puasa, Deepavali and Christmas respectively. Others refer to non pre-festival trading days. 287