A number of studies have been carried to analyze the relationship of anomalies present in the stock market such as; day-of-the-week, weekend, and January effects. These anomalies have strong implications in the stock market. The most important anomaly is monthly effect on stock market. The monthly (or January) effect suggested that daily returns were higher in January than in any other month (Cads by, 1989; Lakonishok & Smidt, 1988; Cadsby & Ratner, 1992; Raj & Thurston, 1994; Mills, 2000; Floros, 2000). Although some researchers didn't find the month effect in different countries despite it was considered as most important anomaly. The January returns were significantly higher than any other months (Haugen & Jorion 2006). A similar study has been carried in low income African countries, i.e., Ghana, Nigeria, and Zimbabwe. The results revealed no January effect in Nigerian and Zimbabwean stock market whereas Ghanaian stock market showed positive January effect (Ayadi & Dufrene, 1998). Floros (2008) suggested that January effect doesn't exist in the Greek stock market.
Most of researchers found significant January effect on stock market, in which January return was higher than any other month. Most of the researchers suggested one should invest at the starting of the January to yield higher returns for following reasons:
One of the strong causes of January effect was tax-selling-hypothesis. According to tax-selling hypothesis, most of the investors sell low priced stock before the end of year to lower their tax liability. Those same investors reactivate in January to purchase shares which tends to raise stock prices.
People sell their stocks because they go on vacation.
The seasonal anomalies in the stock market have been studied in the different stock markets of the world. Rozeff and Kinney (1976) found the January effect in the USA, and then Guletkin and Guletkin (1983) reported significantly higher stock returns in January for most of 17 developed countries. Reinganum (1983) and Roll (1983) also found that January returns were higher than any other month.
However, Ramadan effect can be studied in the Muslim countries of the world. Ramadan is the ninth month of Islamic calendar and it is regarded as the holy month. Ramadan, is the month of fasting, in which eating, drinking, and smoking is prohibited from dawn to sunset. Hotels and restaurants remain closed in day times thought the holy month of Ramadan. In addition, Muslims are encouraged to perform religious rituals in the whole month, such as, fasting, salat-ul-nafil, recitation of Holy Quran, Aaitakaf, and perform social services. Apart from that, some people also visit Saudi Arabia to perform Hajj and Umrah. Muslims are refrained from sin and wrongdoings. In addition to all these rituals, timings of routine activity changed i.e. office timings reduced. In last ten days of the Ramadan people sat in mosques, which was named as Aaitakaf, in turn, it affected the activity in stock exchange. Hence, the purpose of this study was to investigate how the change in these activities affected the stock return.
1.2 Problem statement
To study the Ramadan effect on KSE-10 index.
As Ramadan starts in the Pakistan punters go to perform religious rituals such as Hajj, Umrah, and Aaitakaf, which in turns lower the activity in the Karachi Stock Exchange. Karachi stock market comprises of thousands of punters which invest on daily basis in the stock exchange. As punters refrain from investing in the stock exchange that ultimately lowers the return on the KSE-100 index.
1.3 Hypothesis
H1: Mean daily return in the month of Ramadan is lower than any other months of the year
1.4 Outline of the Study
This study was comprised of five chapters each chapter defined difference aspects of the study. Chapter one was about the introduction, which described the in detail about the topic, problem statement, hypothesis of the study and definitions. Chapter two consisted of literature review, which explained the previous studies taken in different stock exchanges of the world. In chapter three, research method was explained. This chapter comprised of method of data collection, sample size, research model developed and statistical technique. In chapter results of the research was explained, which comprised of finding and interpretation of results and hypothesis assessment summary. Chapter five comprised of conclusion, discussion, implementation and recommendations and future research of the study.
1.4 Definitions
Punters: A retailer who is investing in stock market on daily basis
CHAPTER 2:
LITERATURE REVIEW
A number of anomalies have been investigated in the different stock markets around the world like day-of-the week effect, weekend effect, holiday effect, and calendar effect, which was best known as January effect. These anomalies have strong implication in stock market. The most important anomaly was month effect. The month (or January) effect suggested than mean daily return in the month of January was higher than any other months of the year. One explanation for this was tax-loss selling hypothesis (Branch, 1977). This was defined as selling loser shares at loss in order to avoid tax levied on the gains of other investments. A study by Johnston and Cox (1996) incorporated this study through the firms which was likely claimant of tax-loss selling. Those firms which face large decline in the last half of the year witnessed the positive January returns in the successive year.
Mahendra Raj and Damini Kumari (2006) conducted the study on Day-of-the-week and other market anomalies in the Indian stock market. The purpose of this study was to find out the existence of these anomalies in the Indian stock exchange. The study was aimed at finding the performance of Indian stock market, which included number of hypotheses. Weekday effects, weekend effect, day-of-the-week, January and April effects were investigated through number of statistical techniques. The study found no January effect and Monday effect in the Indian stock market. The results were very different as compared to the results found in other stock markets of the world.
Robert A. Haugen and Philippe Jorion (1996) conducted study on The January Effect still there after all these years in the New York Stock Exchange (NYSE). The purpose of the study was to investigate the January effect presence in NYSE after the many years of emergence in the stock markets of the world. The data was collected from 1926 to 1993, which comprised of all the companies listed on the New York Stock Exchange. Regression analysis was used to find out the results of the study. The results were interesting and gave evidence that after many years January effect was present in the NYSE. The reason for this behavior was that investors sold their stock in the month of December in order to avoid capital gain tax. In January investors' reinvested money, this increased the prices of stocks. Hence, the stock return in the month of January was higher than any other months of the year.
Christos Floros (2008) analyzed the January effect in Greek Stock Exchange. The purpose of the study was to investigate the behavior of stock return January and other months of the calendar. To examine the effect the data was collected from Athens Stock Exchange (ASE), which consists of closing price of ASE General Index. This was collected from the period of 1996 to 2002. The study found that there was no January effect on ASE. Moreover, study concluded that mean daily return in the month of January was same as the mean daily return in the other months of the calendar.
O. Felix Ayadi and Uric B. Dufrene (1998) examined the turn-of-the-year or January effect in low-income stocks markets of the Africa. Ayadi and Dufrene (1998) tested whether January effect was presented only in middle-income and high-income stock exchanges of the world or it were prevailing all over the world including low-income equity markets of the Africa. The data was collected from three stock exchanges of the low-income African stock exchanges i.e. Ghanaian stock market from 1991 to 1996, Nigerian stock market from 1984-1995 and Zimbabwean stock market from 1987-1995. The Wilcoxon-Mann-Whitney pairwise test was applied to found out the January effect. The results shown that January effect was present in Ghanaian stock market, however, there was no January effect Nigerian stock exchange and Zimbabwean stock exchange.
Angel Berges, John J. McConnell and Gary G. Schlarbaum (1984) tested the presence of January effect in Canadian stock exchange. In Canada there was no capital gain tax until 1973, which was the main reason for change in the return in the month of January and other months of the year. The data was collected from the period of 1950 to 1980. The results were positive; research found that January effect was present in the Canadian stock exchange. Moreover, research also confirmed that there was relationship between high stock returns in January and tax-loss selling hypothesis.
Iryna O. Depenchuk, William S. Compto and Robert A. Kunkel (2008) investigated the presence of January effect in Ukrainian stock exchange. The purpose of the study was to found out the weekend and January effect in the Ukrainian stock exchange. In order test the January effect, regression analysis was used to found out the results. The data consisted of daily index from the period of 2003 to 2008. The daily return was calculated from the daily index of the stock exchange. The results were not supporting the previous studies carried in other stock exchanges of the world, however, results shown that there wasn't any presence of January effect in the Ukrainian stock exchange.
Gao and Kling (2005) conducted similar study in the Chinese stock market with the intend to test the calendar anomalies which were present in the other stock markets of the world. The aim of the study was to conclude whether the daily effects and monthly effects were present in the Chinese stock exchange. The data was collected from the Shanghai and Shenzhen stock exchanges, which consisted of index data from the period of 1990 to 2002. There was no January effect in Chinese stock market; however, the mean daily return in the month of February was higher than any other months of the calendar. The Chinese calendar year ends in the January, which is similar to the year end at the December according to the Georgian calendar. The explanation for this high return was that Chinese investors were armature and short term investors and investors embezzle funds to reap high returns in at the beginning of Chinese calendar, which is similar to the January.
Jaffe and Westerfield (1985) conducted study on abnormal stock return for four countries namely Australia, Canada, Japan and U.K. The data was collected from the stock exchanges of the each country. For Japan data comprised of Nikkei Dow index from the period of 1970 to 1983, for Australia data was collected from Statex Actuaries Index from the period 1973 to 1982, for Canada data was gathered from Toronto Stock Exchange from the period of 1976 to 1983, for U.K. data was collected from the Financial Times Ordinary Share Index from the period of 1950 to 1983. The main purpose of the study was to test the day-of-the-week or weekend effect in all these stock exchanges of the four countries. The results showed that there was weekend effect in stock exchanges of those countries. The study concluded that Friday returns were higher than any other days of the week and Tuesday return was lower than other days. The reason for high Friday return was that investors increase their activity in last day to rebalance portfolio.
Mark Haug and Mark Hirschey (2006) carried the study to test the January effect on U.S. stock market. Haug and Hirschey (2006) were interested to check the presence of abnormal returns in the month of January after many years of January effect emergence. The data was collected from the New York Stock Exchange. The study concluded that January effect was present in the NYSE even after the introduction of Tax Reform Act of 1986; this strongly supported the tax-loss selling hypothesis.
Paul Schultz (1985) examined the relationship between personal income taxes and January effect before War Revenue Act 1917 and for the period of 1931 to 1978. The purpose of the study was test the January effect before the 1917 and after the 1931. The data was collected on the Dow Jones Industrial Index for two periods; one was from 1900 to 1917 and other was from 1931 to 1978. The study revealed that before Revenue Act of 1917 there wasn't any presence of January effect in the NYSE. Meanwhile, study found that strong presence of January effect for the period of 1931 to 1978. The study concluded further that stock return was higher for small capitalization firms than high capitalization firms. Moreover, tax-loss selling hypothesis theory was confirmed for such high returns in the month of January.
Kaur (2004) used two Indian Stock indices, the Bombay Stock Exchange (BSE) 30 index and National Stock Exchange (NSE) S&P CNX Nifty stock index to investigate day-of-the-week effect and monthly effect. There was no January effect found on the Indian Stock Indices. However, study found positive February and December returns. Moreover, lower returns were found in the month of March and September.
Ignatius (1998) investigated the anomalies in BSE index in Standard and Poor's 500 stock indices from 1979 to 1990. Ignatius (1998) didn't find any January effect on BSE index; however, Ignatius found that mean returns in the month of December were the highest in the year. In addition, mean returns in the month of April and June were high.
Yakob, Beal and Delpachitra (2005) investigated seasonal anomalies in the ten Asian pacific stock markets, including the Indian stock market from January 2000 to March 2005. Yakob et al. (2005) found negative March and April mean returns in Indian stock market. Meanwhile, May, November, and December exhibited the positive mean returns. From these mean returns, November was found to be the most lucrative in terms of mean return while April was found to be the worst in terms of mean return. The results were astonishing because it exhibited the behavior which was different from the stock markets of the rest of the world. In addition, results were negating the tax-loss selling hypothesis.
Patel and Evans (2003) examined the seasonal behavior in the stock markets of most industrialized nations, which are also known as G7 nations. They examined the data from the period of January 1960 to December 2001. Patel and Evans (2003) found that mean stock returns for December through May were higher than mean returns for the June to November in all G7 nations.
Keppler and Xue (2003) investigated the seasonal price behavior of eighteen stock markets in developed countries for the period of 1970 to 2001. Keppler and Xue (2003) found that returns in from November through April were higher as compare to the returns from May to October. Keppler and Xue (200) termed November through April months as "Good Months" and Mat through October as "Bad Months" for investing in stock market.
Most of the studies found higher returns in the month of January and lower returns in the month of December (Keim, 1986; Chatter Jee & Maniam, 1997). The January return was also associated with the size of the firm. The fact was showing that small capitalization shares do perform better than large capitalization shares in the month of January. There were two reasons for this change in the return. One was tax-loss selling hypothesis, which entailed that investors sold their stocks in December and reinvested in the month of January that reaped high return in the month of January. Second was portfolio rebalancing, which entailed that institutions purchased the many stocks in the month of January which made possible stock returns to rose in the month of January.
In addition to the January effect, other anomalies have also enticed the attention of researchers, such as December-end holiday effect (Ariel, 1990), which states that the pre-holiday returns were higher. Another anomaly was the turn of the month effect (Ariel, 1987) where end month returns were found to be higher than the starting returns. The Friday the thirteenth effect indicates that return on these days were negative as compared to other Fridays (Kolb & Rodriguez, 1987; Dyland Maberly, 1988)
Fama (1970) introduced Efficient Market Hypothesis (EMH). The EMH states that financial markets were informational efficient. That was, the information was available to all public, thus, one cannot earn returns excessive to average market returns. The EMH states the stock returns should be same i.e. there should not be irregularity in stock prices, thus, stock returns.
Husain (1999) investigated the seasonal behavior of equity market in Pakistan. The purpose of the study was to find out the stock return behavior in the month of Ramadan. The data was collected from the Karachi Stock Exchange (KSE) from the period January 01, 1989 to December 30, 1993 on eight sector indices. GARCH model was used to test the behavior of stock returns in the month of Ramadan and other months of the calendar. The study found that there wasn't any Ramadan effect in KSE.
Shahid Ali and Muhammad Akbar (2009) conducted the study in Pakistani Stock Exchange. The aim of the study was to examine daily, weekly and calendar effects on Karachi Stock Exchange. The data was collected from 1991 to 2006 on KSE-100 index. Autoregressive Integrated Moving Averages and Ordinary Least
Squares were used to test these effects. The results revealed no month effect in Karachi Stock Exchange. The reason for such behavior was the Efficient Market Hypothesis, which entailed stock return was independent of previous days return.
Alexakis and Xanthakis (1995) examined the day-of-the-week effect in Athens Stock Exchange for the period of January 1985 to February 1994. Nikou (1997) also examined the weekend effect on Athens stock market. The results shown high returns on Monday and Fridays, negative mean returns on Tuesdays and Wednesdays.
Fazal J. Seyyed, Abraham Abraham and Mohsen Al-Hajji (2005) conducted study on the Ramadan effect in Saudi Arabian stock market. The purpose of the study was to investigate the Ramadan effect, which is similar to the January effect, in Saudi Arabian stock market which is the largest stock market among Muslim countries. The data comprised of weekly index of six sectors of the economy from the period of 1985 to 2002. Autoregressive Conditional Heteroskedastic model was used to test the effect. The study revealed that there was no Ramadan effect in Saudi Arabian stock market, the mean return in the month of Ramadan was similar to the other months of the Islamic Calendar.
CHAPTER 3:
RESEARCH METHODS
3.1 Method of Data Collection
In this study, main objective was to find out the mean daily return of stocks in the month of Ramadan and other months of the Islamic Calendar. In order to calculate the mean stock return on KSE-100 index the study required the daily closing KSE-100 index.
Secondary method of data collection has been used for this study, because the data was readily available. The data was collected from Standard Capital Securities (SCS) Pvt. Ltd. The data consists of daily Karachi Stock Exchange or KSE-100 index. There are four indices prevailing in the Karachi Stock Market i.e. KSE-All share index; KSE-100 index, which was introduced in 1991; KSE-30 index, which was introduced in 2006; and KMI-30 index, which came into existence in 2008.
The KSE-100 was launched in 1991 with a base of 10,000 points. KSE-100 index is working as a benchmark in Karachi Stock Exchange to compare the prices over the time. KSE-100 index consists of 100 companies, the criteria for selection of 100 companies was based on the sector representation and the highest market capitalization. KSE-100 index comprises 80% of total market capitalization of all the companies listed on Karachi Stock Exchange. There are 35 sectors listed on KSE, out of those 34 companies are selected from each sector on the basis of highest market capitalization. However, Open-End Mutual Fund Sector was excluded from the KSE-100 index. Remaining 66 companies are included in the KSE-100 on the basis of largest market capitalization.
3.2 Sample Size
Since secondary data was available with brokerage houses. Data from January 1995 to October 2010 has been analyzed for the research purpose. The daily closing KSE-100 index was used for the study purpose. Data was collected from the Standard Capital Securities Pvt. Ltd. and Taurus Securities Pvt. Ltd., which is subsidiary company of National Bank of Pakistan.
3.3 Research Model Developed
To calculate the return on the KSE-100 index following model was used by using the log natural.
R = ln KSE-100t - ln KSE-100t-1
Where R is the return, KSE-100t is the value on t time and KSE-100t-1 is index value on the day before t time. The closing value of the index at the end of the day was taken in order to calculate the daily return.
3.4 Statistical Technique
Test of independence was used by taking hypothesis into account; Statistical Package for Social Sciences (SPSS) is used to analyze the data.
There was two groups in the data file i.e. Ramadan (Ramadan =1) and Otherwise (Otherwise = 0), by considering these two groups test of independence was carried to find the results.
CHAPTER 4:
RESULTS
4.1 Findings and Interpretation of Results
The data consisted of total 3987 cases, out of which there was 3641 cases consisted of data on other months i.e. Muharram, Safar, and so on. Out of 3987 observations 346 observations comprises of Ramadan cases.
Table 4.1
In table 4.1 results were depicted about the means of two groups i.e. Ramadan and Otherwise. In Ramadan mean daily return was higher than other months, which was 0.30469539. While the mean daily return in other months were 0.01580514. In addition to the mean return, standard deviation of the each group was presented.
Moreover, the study revealed that mean return in the month Ramadan was higher than any other month of the Islamic Calendar. Husain (1999) tested that there was no Ramadan effect in Karachi Stock Exchange. However, this study concluded that there was Ramadan effect on KSE-100 index.
But in this study, KSE-100 index exhibited the Ramadan effect. The explanation for such a behavior was that over the recent years Karachi Stock Exchange got stabilized and it reached to the highest level of 15000 points, which enticed investors invest more money. In addition, investors used to refrain from investing in the stock market in the month of Ramadan but recent trend shown that investors put much money in order to reap high returns in the month of Ramadan.
4.2 Hypothesis Assessment Summary
The centre of every study was hypothesis and the hypothesis in this study had significant impact on the mean daily return in the different moths of the calendar.
CHAPTER 5:
DISCUSSIONS, CONCLUSION, IMPLICATIONS AND FUTURE RESEARCH
In this study, test of independence was used to examine the mean daily returns on the KSE-100 index daily data, which was from January 1995 to October 2010. Independent t test was used to arrive at the mean returns in the month of Ramadan and other months of the year.
5.1 Conclusion
There were scores of the studies, which have been carried in the stock exchanges in the different countries such as day-of-the-week effect, weekend effect, and moth effect, which is best known as January effect. In addition, some researchers have also investigated the April or November effect. Most of the studies found that January effect persists in most of the stock exchanges of the world, which means higher January returns and lower or negative returns in other months of the year. However, in Islamic countries such as Pakistan, studies can carried to find out the effects like Ramadan effect, Friday effect, and so on. This study investigated the Ramadan effect in Karachi Stock Exchange or KSE-100 index. After analyzing the more than 15 years data, the study concluded that mean return in the month of Ramadan was higher than any other month of the year.
5.2 Discussions
As for as the return on KSE-100 index in the month of Ramada that study revealed that return in the month of Ramadan was higher than any other months of the Islamic calendar. The reason for this may be the economic triggers of the country. As the economic situation is not good, investors wished to reap maximum return so they can meet their expenditures. In turn, investors had invested much in the month of Ramadan, which in turn, reaped highest return.
5.3 Implications and Recommendations
This study helped investors, brokers, fund managers and stakeholders to analyze the behavior of return over the year. Moreover, it also helped the investors to decide in which month they should invest in order to reap the highest return on their investments.
However, the major issue for this study was that it gives results on the average basis, which may undermine the returns. Moreover, it also includes the period of 1998 when Pakistan tested nuclear weapons, in turn, world imposed restriction on trade. In addition, it also includes the recession period of 2008.
5.4 Future Research
In future so many anomalies can be investigated of these types. Researchers can check the effect on Islamic indices i.e. Karachi Meezan Index or KMI-30 index, Dow Jones Islamic index, and so on. Moreover, researcher can investigate the Moharram effect, Rabi-ul-awal effect.
