StudentShare
Contact Us
Sign In / Sign Up for FREE
Search
Go to advanced search...
Free

The Connection between Stock Market Anomalies and Microeconomic News Announcements - Literature review Example

Cite this document
Summary
There are no investment opportunities on an efficient market which can lead to abnormal returns. These abnormal returns are the actual differences…
Download full paper File format: .doc, available for editing
GRAB THE BEST PAPER98.1% of users find it useful
The Connection between Stock Market Anomalies and Microeconomic News Announcements
Read Text Preview

Extract of sample "The Connection between Stock Market Anomalies and Microeconomic News Announcements"

Introduction The Efficient Market Hypothesis (EMH) s that security process on financial markets reflects all the relevant information. There areno investment opportunities on an efficient market which can lead to abnormal returns. These abnormal returns are the actual differences between the expected returns and the actual returns of securities. One of the violations of EMH is referred to as stock market anomaly. This is a period or event that can be used to produce abnormal profits after a trade. These anomalies are classified to different categories: even anomalies’, calendar anomalies and weather anomalies. Given the fact that researchers have tried over the years to explain the anomarlies, Gerlach (288) findings happened to be very interesting. He was the first to report on the connection between stock market anomalies and microeconomic news announcements. This paper dwells more on this kind of stock market anomaly. The question at hand is whether news announcements have any effect on the stock market in regards to specific days. These are Monday, Friday and the weekdays. Day-of-the-week effect This comes to the difference in returns in the returns realized between different days of the week. The weekend effect focuses mainly on the Monday and Friday returns, stating that Monday returns are low and negative and Friday returns are high compared to the remaining trading days of the week. Extensive research has been done to find explanations for this anomaly. This section gives an overview of the most important research that was made in the last few decades regarding the day-of-the week effect and the weekend effect. Cross (69) was one of the first to report differences in returns on Fridays and Mondays compared to the rest of the week. With daily return data from 1953 – 1970 on the S&P 500, he found a statistically significant difference between Friday and Monday returns for almost every year in the sample period in both mean returns and in percentage of time that the index rose on that day. Moreover, his results showed that Monday indices following a decline on Friday rose in approximately 24% of the cases, which is significantly different from the reaction of the remaining trading days of the week following a decline of the previous trading day. Subsequent to Roll (1973), French (1980) reported that Monday returns were negative and lower than returns for other days in the week, using daily returns from the S&P 500 composite portfolio for the period 1953 – 1977. In his research, he examined if this is the consequence of the weekend prior to each Monday or the consequence of every non-trading day (holiday). His results illustrated that the return on every weekday on its own (with the exception of Tuesday) is higher when the day follows a holiday compared to when the day follows a trading day. This is different for Tuesday, because Tuesday is the first trading day after the weekend when Monday is a holiday. Therefore, French (1980) concluded that the negative Monday returns are caused by a weekend effect. Cross (70) and French (25) did not look for possible explanations for the weekend effect, although other authors did. Below a selection is given of the articles of those authors and their findings regarding the weekend effect. Settlement period By using mean returns and variances for the S&P 500 and the CRSP value- and equally-weighted portfolios from July 1962 – December 1978, Gibbons and Hess (580) also found negative Monday returns, although no Monday effect in variances. Moreover, they searched for possible explanations for the Monday effect. The settlement period explains the more negative Monday returns before 1968 compared to the returns after 1968, because the settlement period was four business days before February 1968 and five business days after February. Nonetheless, it does not explain the negative Monday returns from February 1968 – December 1978. Apart from that, the settlement period is nowadays three business days for stocks. Release of information French and Roll (58) investigated the return variances of weekdays, weekends, holidays and holiday weekends by means of daily returns on the New York and American Stock Exchanges from 1963 – 1982. They found that the returns are more volatile during exchange trading hours compared to non-trading hours. The three possible explanations that were given for this are that public information (which causes the volatility) is announced more frequently during business days (weekdays), private information probably influences prices more when the stock markets are open and the process of trading itself causes volatility. French and Roll (60) concluded that their results showed that only a small part of the difference in variances between trading hours and non-trading hours is caused by mispricing occurring during trading. The reason for this mainly lies in the difference in quantity of information announced between trading hours and non-trading hours. Internationally Jaffe and Westerfield (438) contributed to the literature by investigating the day-of-the-week effect internationally. In all countries in their research, namely the U.K., Japan, Canada and Australia, they found a day-of the-week effect with significant negative Monday returns and high Friday returns. Correlations tests suggest that there is a strong correlation between the returns of the four foreign countries and those of the U.S. Nonetheless, independent of the day-of-the-week effect in the U.S. Jaffe and Westerfield (440) found a day-of-the-week effect in each of the four countries on their own. Similar to Gibbons and Hess (54), they looked at settlement periods to explain the day-of-the-week effect. They only found little evidence for the higher Thursday and Friday returns in Australia, but for Canada, the U.K. and Japan the settlement period did not explain the day-of-the-week effect at all. In addition, Jaffe and Westerfield (442) investigate the opportunity that measurement errors cause the day-of-the-week effect. They state that if Monday returns would be influenced by mainly negative random errors and Friday returns by mainly positive random errors, the correlation between Monday and Friday returns should be low. Nonetheless, they found a higher than average correlation between Monday and Friday returns and therefore conclude that measurement errors cannot explain the day-of the- week effect. In their exploration of different anomalies on stock market indices in eighteen countries across the world, Agrawal and Tandon (88) captured the day-of-the week effect as well. They found negative returns on Mondays for thirteen countries (of which seven are statistically significant), but also negative returns on Tuesdays in twelve countries (of which eight are statistically significant). Furthermore, they found Tuesday returns to be lower compared to Monday returns in eight countries. Contrary to these negative Monday and Tuesday returns, they revealed positive Wednesday and Friday returns in the majority of the countries. After reporting the day by day returns, Agrawal and Tandon (88) discussed possible explanations for the negative Tuesday returns. They stated that the time zone hypothesis (which argues that Tuesday returns are low in some countries due to time-differences that exceed twelve hours) can explain the negative Tuesday returns in three of the five countries, but cannot explain the negative Tuesday returns in European countries. Furthermore, the difference between trading days and non-trading days is not an explanation for the negative Tuesday returns. After running day-of-the-week correlation tests and regressions, the null hypothesis that day by day variances are dependent on the US can be rejected for the majority of the countries. Moreover, Agrawal and Tandon (88) argued that the settlement procedure explains a part of the day-by-day differences (mainly the higher returns on Wednesday, Thursday and Friday) in returns, but cannot explain the negative Monday and Tuesday returns for most of the countries. Furthermore, they divided their total sample period into two sub-periods and found that in the seventies Monday returns are significantly negative in seven countries and Tuesday returns are significantly negative in nine countries, while in the eighties the Monday and Tuesday returns are not significantly negative in the majority of the countries. Finally, they found that Monday returns are negative in almost all countries if the market declined the previous week, but mainly positive when the market went up the previous week. However, this is not found for Tuesday returns. Macroeconomic News This news has effect on the stock market performance. Previous literature that examined this aspect of stock trading has a lot to say. According to Gerlach (290), there is a link between macroeconomic announcements and the returns in the stock market. He dares link the stock market calendar with the weather and macro-economic news announcements. Macroeconomic News and the Stock Market Returns The macro-economic news announcements are recurrent financial and economic indicators that give information on the state of the economy. This information can relate to: Sales: lightweight vehicle sales Prices: Producer Price Index (PPI) Trade: Trade balance Employment: Nonfarm payrolls and Employment Reports macroeconomic news announcements that are used in this research will be discussed in more detail in the next chapter. In the last few decades, there have been many studies examining the effect of acroeconomic announcements on stock market returns. Chen, Roll and Ross (1986) examine the relation between macroeconomic news and stock market returns by using returns on the equally and value weighted New York Stock Exchange for the period of January 1953 - November 1983. From the total set of macroeconomic variables that they use, the growth rate in industrial production and the changes in the risk premium and yield curve are highly significant in explaining expected stock returns, whereas the unanticipated and expected inflation (change in Consumer Price Index) is less significant. The overall conclusion of Chen, Roll and Ross (1986) is that stock market returns are influenced by macroeconomic variables and that the intensity of the influence corresponds with the total exposure to them. State of the economy Gerlach (290) investigate the impact of macroeconomic news announcements on stock market returns as well, thereby looking at the state of the economy. In doing so, they use daily stock market returns on the S&P 500 from 1977 – 1988 and data on macroeconomic announcements regarding industrial production, unemployment rate, nonfarm payroll employment, merchandise trade deficit, consumer price index, producer price index and M1. They define the economic activity as high, medium or low, which is based on the level of industrial production. Having tested the reaction of S&P 500 returns on macro-economic news announcements in the different states of the economy, Gerlach (290) conclude that the “market’s response to macroeconomic news depends on the state of the economy”. This is especially the case for higher than expected real activity, which leads to lower stock prices when there is high economic activity (a strong economy), but leads to higher stock prices when there is low economic activity (a weak economy). In their research, Boyd et al. (2005) also look at the state of the economy, although only investigating the influence of one macroeconomic announcement on the S&P 500, namely the unemployment rate. They thereby look at the reaction to the unexpected part of unemployment news and conclude that stock markets rise in response to bad employment news during expansions and drop in response to bad employment news during contractions. After presenting this conclusion, they search for an explanation by comparing the influence of two components of unemployment news on stock prices and risk-free government bonds, namely the equity risk premium and the expected future growth rate of dividends. They found that stock prices during expansions are most influenced by changes in the equity risk premium and stock prices during contractions are most influenced by changes in the expected future growth rate of dividends. Comparable research is conducted by Andersen et al. (290), although using a substantially larger set of macroeconomic announcements. They look at the returns on the S&P 500, the FTSE 100 and the DJ Euro Stoxx 50 ten minutes before every macroeconomic announcement until one and a half hours after the announcement. This is done for the whole sample period (July 1988 – December 2002), but also for two sub-periods concerning expansion (July 1988 – February 28, 2001) and contraction (February 2001 – December 2002). Gerlach (290)) found, similar to Boyd et al. (2005), that bad macroeconomic news had a negative influence on stock markets during contractions, but a positive influence during expansions. An investigation on the impact of 17 different macroeconomic announcements on daily stock returns of the value-weighted NYSE-AMEX-NASDAQ market index for the period 1980 – 1996 was done. Besides looking at the influence of these macro-economic announcements on the level of the daily stock returns, they also capture the influence on the volatility of the returns. They find that, of the 17 macro-economic announcements, CPI and PPI only affect the market returns, while Balance of Trade, Employment and Housing Starts only affect the volatility of the returns.M1 affects both returns and volatility. Furthermore, investigate if their conclusions still hold after dividing the sample period into three sub-periods based on time and three sub-periods based on economic regimes’ (by looking at the growth rate of Industrial Production, Unemployment Rate, Consumer Confidence and an index of Job Openings). Their conclusions turnout to be valid, even after the division mentioned above. 3.2 Macroeconomic news announcements and stock market anomalies As described in chapter 2, multiple explanations for calendar anomalies have been put forward in previous research. Gerlach (290) comes with another explanation for calendar and weather anomalies. He links calendar and weather anomalies to the market response to macroeconomic news announcements. For his research, he uses daily returns on the CRSP equally weighted index of the NYSE / AMEX / NASDAQ and the S&P 500 index from 1980 – 2003. Furthermore, he makes use of the announcement dates of eleven news announcements, namely the Federal Reserve’s Beige Book, Business Sales and Inventories, Consumer Price Index (CPI), Employment Report, Advance Durable Goods Shipments, Federal Open Market Committee announcements, (FOMC), Gross Domestic Product (GDP), Housing Starts, Industrial Production, Retail Sales and Lightweight Vehicle Sales. First, he presents the differences in mean daily return between the anomaly period and the remaining trading days for each of the six calendar and weather anomalies he considered. Appendix A shows table 2 of the research of Gerlach (290), where these results are illustrated. The p-values for each anomaly demonstrate that each anomaly is statistically significant at the ten percent significance level or better. Further, Gerlach (290) compares the mean daily return on days where at least one macroeconomic announcement was made (announcement day) with the mean daily return on days where no macroeconomic announcement took place (non announcement day). He finds that for both the S&P 500 and the CRSP index, the mean return on announcement days is significantly different from the mean return on non announcement days. Moreover, he demonstrates that the standard deviation is smaller on announcement days compared to non announcement days, therefore concluding that the higher return on announcement days is not caused by higher risk. After this, Gerlach (290) investigates the influence of macroeconomic announcements on six calendar and weather anomalies, namely the turn of the month effect, the January effect, the fall effect, the lunar effect, the rainfall effect and the temperature effect. For the turn-of-the-month effect, the fall effect, the lunar effect, the rain effect and the temperature effect he finds that this is entirely caused by the significantly higher returns on announcements days. The January effect is the only anomaly that is still present when only non announcement days are used in the sample, but is less strong. Appendix A presents table 4 of the research of Gerlach (290), which shows the statistical significance of the six stock market anomalies if only non announcement days are considered. So the main finding of Gerlach (290) is that five of the six anomalies do not occur when announcement days are left out. He concludes that the reaction of the market to macroeconomic announcements is the main source of calendar and weather anomalies. To corroborate this conclusion, he conducts several tests to exclude alternative explanations. A different sample of macroeconomic announcements is used, therefore testing for the possibility that the conclusion only holds for a particular set of announcements. Nevertheless, the main findings still hold for the different set of macroeconomic announcements, which contains Retail Sales, CPI, Employment Report, ISM survey and the Employment Cost Index. On top of that, the sample is divided into two sub-periods, namely 1980-1991 and 1992 – 2003. For both periods, the only significant anomaly is the January effect when controlled for macroeconomic announcements, which means that the possibility can be rejected that unusual returns in a particular period drive the conclusion. The last alternative explanation that is tested for, is the possibility that the market responds differently to macroeconomic news during periods of calendar anomalies (January, turn-of the-month and fall). The unexpected component of the market reaction to FOMC is used to determine the difference between the response of the market during calendar anomalies and the remaining of the year, but no significant differences are found. This thesis is based on the article of Gerlach (290). Comparable research will be conducted with respect to the influence of macroeconomic news announcements on stock market anomalies in the U.S. Furthermore, this will be repeated for the U.K. In the next chapter, the data and methodology used in this thesis will be described, making use of the data and methodology of Gerlach (290). Bibliography Agrawal, A., and K. Tandon, 88, Anomalies or illusions? Evidence from stock markets in eighteen countries, Journal of International Money and Finance, 13, 83-106. Cross, F., 1973, The Behavior of Stock Prices on Fridays and Mondays, Financial Analysts Journal, 67-69. French, K.R., 1980, Stock returns and the weekend effect, Journal of Financial Economics, 8, 55-69. French, K.R., and R. Roll, 1986, Stock Return Variances: The Arrival of Information and the Reaction of Traders, Journal of Financial Economics, 17, 5-26. Gerlach, J.R., 290, Macroeconomic news and stock market calendar and weather anomalies, The Journal of Financial Research, 15, 283-300. Gibbons, M.R. and P. Hess, 1981, Day of the Week Effects and Asset Returns, The Journal of Business, 54, 579-596. Jaffe, J., and R. Westerfield, 1985, The Week-End Effect in Common Stock Returns: The International Evidence, The Journal of Finance, 40, 433-454 Read More
Cite this document
  • APA
  • MLA
  • CHICAGO
(Exercise on friday effect Essay Example | Topics and Well Written Essays - 2500 words, n.d.)
Exercise on friday effect Essay Example | Topics and Well Written Essays - 2500 words. https://studentshare.org/macro-microeconomics/1815431-exercise-on-friday-effect
(Exercise on Friday Effect Essay Example | Topics and Well Written Essays - 2500 Words)
Exercise on Friday Effect Essay Example | Topics and Well Written Essays - 2500 Words. https://studentshare.org/macro-microeconomics/1815431-exercise-on-friday-effect.
“Exercise on Friday Effect Essay Example | Topics and Well Written Essays - 2500 Words”. https://studentshare.org/macro-microeconomics/1815431-exercise-on-friday-effect.
  • Cited: 0 times
sponsored ads
We use cookies to create the best experience for you. Keep on browsing if you are OK with that, or find out how to manage cookies.
Contact Us