Dividend Declaration And Abnormal Returns In The Stock Finance Essay
An event study measures the impact of a specific event on the value of a firm. The importance of the study comes from the fact that, given the rationality of the market, the effect of a particular event will be immediately on the security price. Event studies started as tests of the semi‐strong form of market efficiency, which says that all publicly‐available information gets impounded instantaneously into the stock price.
Eugene Fama first defined the term "Efficient Market" in financial literature in 1965 as one in which security prices fully reflects all available information. The market is efficient if the reaction of market prices to new information should be instantaneous and unbiased. Efficient Market Hypothesis (EMH) is the idea that information is quickly and efficiently incorporated into asset prices at any point in time, so that old information cannot be used to foretell future price movements. Consequently, three versions of EMH are being distinguished depends on the level of available information.
Degrees of Efficiency
Accepting the EMH in its purest form may be difficult; however, there are three identified classifications of the EMH, which are aimed at reflecting the degree to which it can be applied to markets.
1. Strong efficiency ‐ This is the strongest version, which states that all information in a market, whether public or private, is accounted for in a stock price. Not even insider information could give an investor an advantage.
2. Semi‐strong efficiency ‐ This form of EMH implies that all public information is calculated into a stock's current share price. Neither fundamental nor technical analysis can be used to achieve superior gains.
3. Weak efficiency ‐ This type of EMH claims that all past prices of a stock are reflected in today's stock price. Therefore, technical analysis cannot be used to predict and beat a market
The purpose of this project is to analyze how the dividend declaration of a company affects the abnormal returns in the stock.
Our objective is to analysis the Indian Stock Market (BSE SENSEX) according to the EMH (Efficient Market Hypothesis) during the period of 2001 to 2010.
We have worked on Event Based Trend Analysis in which we are taken dividend declaration as the event and we had studied the trend of such an event during the event window i.e. before and after 5 days of the event (dividend declaration).
The movement of the abnormal stock returns during this time and the possible reasons for so will also be explained. Also till how many days the effect of such an event can be seen in the
The first event study was designed and conducted by Eugene Fama, Lawrence Fisher, Michael Jensen, and Richard Roll. Their article, “The Adjustment of Stock Prices to New Information,” was published in the International Economic Review in 1969 and quickly earned itself a nickname, “the FFJR study.”
FFJR studied the stock market reaction to announcements of stock splits. Typically, stock splits are believed to be seemingly inexplicable good news for investors. One possible reason was reported by FFJR themselves: they found that 72% of firms in their sample announced above‐ average dividend increases in the year after the split. Stock splits seemed to “signal” future dividend increases. (Actually, the term “signaling” was proposed in the early 1970s by Michael Spence, who won the 2001 Nobel Prize for, among other things, his research on signaling in labour markets.)
The event study techniques were further refined by other researchers. Some of the research designs are quite clever. A bizarre example appeared in a 1985 article in the Journal of Accounting and Economics by Johnson, Magee, Nagarajan, and Newman.
The title of the article, “An Analysis of the Stock Price Reaction to Sudden Executive Deaths,” is self‐explaining. The authors found that unexpected CEO deaths are associated with stock price decreases. However, in cases when the CEO was the company founder, the stock market tends to react by a price increase, begging the inference that the ability to create a business is different from the ability to run one.
By 1975, the preponderance of evidence argued that markets were efficient. Statistical studies showed that technical analysis did not add value (consistent with the weak form of market efficiency). Event studies found that the market quickly reacts to new information (consistent with the semi‐strong form of market efficiency). And studies of professional investors’ performance made a strong case for the strong form market efficiency.
In 1976, Rozeff and Kinney published their article on stock market seasonality. They found that January stock returns were higher than in any other month. In 1981, Gibbons and Hess reported “the Monday effect” – stock prices tended to go down on Mondays. Both of these
findings were clearly inconsistent with the weak‐form market efficiency.
In the nine‐year period of 1962‐1970, the S&P 500 returned about ‐0.16% on an average Monday. In the following nine‐year period, 1970‐1978, the S&P 500 would only drop by 0.10% on average. It appears that the effect has been known to some market participants for a while, and they were taking advantage of this private information, which, in turn, caused their gains to decrease over time.
A growing body of research indicated that profitable selection rules could be based on publicly available information. In particular, stocks with low price‐earnings ratio and high dividend yield outperformed the market. And, while small capitalization stocks have a greater risk than large‐ cap stocks, the return premium seemed to be too large for the degree of additional risk taken.
The discovery of these and other “market anomalies” prompted the editorial board of the Journal of Financial Economics to publish a special issue in June 1978 on a dozen of those market anomalies.
In 1981, Henry Oppenheimer tested stock selection criteria developed by Benjamin Graham. Most of us probably know Ben Graham as the author of the classic, Security Analysis, but he also wrote another, somewhat less technical, book, called The Intelligent Investor. In each new edition of the book, Graham updated his investment advice to his readers, whom he called “defensive investors”. Oppenheimer back‐tested this advice as if he purchased every edition of The Intelligent Investor and acted on it after reading it. It turned out that Graham’s advice did have significant value. Moreover, it actually had more value than Graham himself claimed.
In 1982, Rendelman, Jones, and Latané published their article, “Empirical Anomalies Based on Unexpected Earnings and the Importance of the Risk Adjustments,” in the Journal of Financial Economics. They studied earnings surprises and their effect on the stock price. They divided their sample into ten categories (deciles in statistical parlance) according to how positive or negative the earnings surprise was. Then they calculated averaged price paths for stocks in each decile.
While the market did react to earnings surprises quickly, the prices also drifted in the direction of the earnings surprise following the announcement. In other words, the market commonly under reacts to the quarterly earnings announcements. This suggests the validity of an “earnings momentum” strategy (buying stocks that just had a positive earnings surprise). A
number of later studies produced results consistent with this thinking.
However, in a somewhat puzzling twist, there were studies which suggested that the stock market actually overreacts to certain announcements. In 1981, Robert Shiller published his article, “Do Stock Prices Move Too Much to Be Justified by Subsequent Changes in Dividends?” in the American Economic Review and concluded that they do. This phenomenon came to be known as “excess volatility”.
“Using Daily Stock Returns – the case of event study” by Stephen J Brown Yale Universiry. New Haven, CT 06520, USA Jerold B. WARNER University of Rochester, Rochester, NY 1462 7, USA of in 1984 .This paper examines properties of daily stock returns and how the particular characteristics of these data affect event study methodologies..
In 1985, Werner De Bondt and Richard Thaler published their article, “Does the Stock Market Overreact?” in the Journal of Finance. Their conclusion was that the stock market tends to overreact to long series of bad news.
So by 1985, there were enough anomalies discovered to seriously doubt the validity of the efficient market hypothesis.
Tests of Market Efficiency in the 1960s
A number of different approaches were used to test the efficient market hypothesis. One of the most obvious ones was to do more studies on serial correlation of security prices. A variation of this approach would be to test various trading strategies recommended by technical analysts to see if they have any investment value. Both have been tried, and invariably came back with mostly negative results.
An interesting area of research dealt with the nature of return distributions. There are some clearly visible asymmetries in stock returns. If we look at the ten biggest one‐day movements in S&P 500 index since 1947, nine of them would be declines. The market crash of October 1987 resulted in a negative return that was 20 standard deviations away from the mean. It turned out that stock returns are not normally distributed. They follow some sort of distribution, but, to our knowledge, no one has figured out what kind of distribution it is. On several occasions, stable Paretian distribution and Student t‐distribution were found to be better approximations
than the normal distribution.
Working paper “Econometrics of Event Study” by S.P Kothari and Jerold B Warner 2006 illustrates that the properties of event study methods can vary by calendar time period and can depend on event sample firm characteristics such as volatility; the paper also discusses various methods for event study for short as well as long time horizons.
Needless to say, this poses a huge methodological problem for researchers who, for lack of a better assumption, are still assuming normal distributions for drawing statistical inferences.
An important breakthrough in testing market efficiency came with the advent of the “Event study” methodology. In an event study, researchers take a sample of similar events that occurred in different companies at different times and determine how, on average, this event impacted the stock price.
And what would a researcher expect to see as the outcome of an event study? Assuming that we are studying favourable events, the outcome would depend on whether or not this is anticipated by the market and, of course, on whether or not the market is efficient. In all cases, we would expect the stock price to go up. Consider an unanticipated event first. If the market were efficient, the stock price would adjust upward very quickly. If not efficient, it will drift
upward for some time following the event
SINGLE INDEX MARKET MODEL
Model to be Consider for Calculating Abnormal Return during the event window:
There are various methods to calculate the expected return during the event window. The different models are Single index market model, Capital assets pricing model, Arbitrage pricing model and Market Risk Adjusted model. Each and every model has their assumptions and limitations. Since Indian stock market is developed and we also know that there is a linear relationship between market and security return. So in such a scenario, it is better to use Single index market model and regression will be used to develop a regression line, which will help us in predicting the return for the event window on the basis of the relationship between the stock return which is a dependent variable and market return which is an independent variable. The regression equation will be as follows:
Ri = α + β*Rm + ε
Ri‐ returns of the security ‘i’ Rm ‐ return of the market
ε ‐ Error equals to 0
Abnormal Return: Abnormal return is defined as the extra return which an investor makes because of the event taking place. To calculate abnormal return, we will first predict the normal return for the event window with the help of the above equation, then we will compare the actual return and the predicted return, and the difference will be called as abnormal return.
ARx = Rx – E (Rx)
ARx – Abnormal Return of Stock X Rx – Actual Return of Stock X
E (Rx) – Expected Return of Stock X
To calculate the Average Abnormal Returns for both the indices for the event dates, we average out the returns of each company on each day from ‐5 day to +5 day.
AARU,5 = (AR1,5 + AR2,5 + AR3,5 + AR4,5 + …………… +AR11,5) / 11
AARU,5 ‐‐ Average Abnormal Return for BSE Index on the 5th Day
AR1,5 – Abnormal Return on the 5th Day for Company 1 in BSE Index
Hence this result gives the average abnormal returns for Company 1 on the 5th day. Similarly we follow the same procedure for estimating the average abnormal returns for all the companies for 10 years (2001‐2010) in the indices on all the 11 days of the event window from
‐5th day to +5th day.
After the average abnormal returns, next step is to do the significance test. t‐test will be used to check if the average abnormal returns during the event window are significantly different from zero or not. The test will be done 5% & 10% level of significance to check the significance of the abnormal return during the event window.
™ The expected returns of stocks have been calculated using the regression tool in excel.
The p values have been checked for β (slope) and only those entries where the p‐value is less than 0.05 & 0.1 has been considered for our study.
The daily return of the stock and the BSE index is calculated using the below formula
ܧሺܴሻDaily = Ln( Today’s Adjusted Close/ Previous day’s adjusted close)
™ Then the abnormal return, average abnormal daily return and standard deviation is calculated for each day of the event window. Standard Error is calculated as
Std. Dev. /sqrt(n) where n is the number of different stocks.
™ Then t‐test is used as and tested at 5% & 10% level of significance. Refer Table 1 for the complete outputs of t‐statistics.
t= Mean/Std. Error
The nature of the quantitative data is secondary in nature. It has been collected from various databases. Initial data on the dividend declaration date was collected from Prowess followed by the collection of data of the stock prices and market prices (BSE SENSEX) from Capitaline.
Data is collected for a ten year period from 2001 – 2010. Only final dividends are considered for research.
Following are the companies which have been taken for analysis.
BANK OF BARODA
Based on our objectives and the previous study done related to our objective we can formulate the following hypothesis to be performed and tested with the collected data:
H0 : There is no abnormality in stock prices because of event i.e. dividend declaration
H0: μH0 = 0
H1 : There is a significant amount of abnormality in stock prices because of event i.e. dividend declaration
H1: μH0 ≠ 0 where:
H0 = the null hypothesis μH0 = the mean abnormality
1. Hypothesis: H0: μ = 0
H1: μ ≠ 0
2. T test: The values of t stat are found at 99 degrees of freedom at the levels of significance i.e. 90% and 95%
At 90% significance level: ±1.66; At 95% significance level: ±1.984
RESULT AT 90% SIGNIFICANCE LEVEL
(null hypothesis accepted or not)
RESULT AT 95% SIGNIFICANCE LEVEL
0 (EVENT DAY)
For 90% & 95%
Fig 1: At 90% significance level
Fig 2: At 95% significance level
Here the null hypothesis is not rejected. The day t+2 average abnormal stock returns are in significance level stating substantial abnormality in stock prices for rest all
days i.e. days T‐5, T‐4 ,T‐3
,T‐2 ,T‐1,T, T+1,T+3, T+4,T+5 the returns are within the confidence level so we can say that the Indian stock market ( BSE) is semi strongly efficient. That is, one cannot consistently achieve returns in excess
of average market returns on a risk‐adjusted basis, given the information publicly available at the time the investment is made.
c a r
p e cWe have proven the hypotheses for both the significan e level (90% &
RESULTS AND ANALYSIS
On the basis of the initial collection of data we observed minor abnormalities around the date of dividend declaration. However within the data huge variations were seen for instance, a 16% increase in the dividend in the year 2003 for Infosys1 showed an abnormality of ‐2.6%. In other cases dividend declaration and abnormalities moved in the same direction for instance Wipro2 in 2004 had a dividend growth rate of 2400% while the average abnormality increased by
1.42%. No general relation between the dividend declaration and abnormalities can be seen. This could be attributed to the following reasons
1. Nature of events
™ Regular Event
Dividend declaration as an event is a regular event that takes place every year. This event in comparison to other events such as mergers n acquisitions, stock splits or change in top management is more frequent in nature and hence the reaction of the stock prices on an average do not show much fluctuation.
™ Anticipated event
The event is already anticipated on the basis of previous experience as well as the announcements made by the company. And hence the information asymmetry is less
2. Better means of communication
The information flow from the company’s sites towards the investors has eased over the years. The information being mediated through various media channels, radio, newspaper, brokerage firms etc has lead to free flow of information and a greater
1 Refer Infosys calculation excel sheet 2003
2 Refer Wipro calculation excel sheet 2004
3. Increase in retail investors
The quantum of retail investors has increased over the years leading to lower fluctuations in the stock prices. Retail investors have varied opinions so the movement in prices of stocks in not abrupt as in comparisons with FII.
4. Macro economic factors
Stock prices as a whole will also depend on the current economic scenario, political developments, performance of the particular sector to which a company belongs, interest rates. Therefore the reaction of the stock prices to an event also captures the factors external to the company.
From the below graph (Fig 3) we can analysis by using a time line that is
T‐5 day to t‐2 day: if the news in the market about the dividend declaration is positive then there is an expectation of rise in stock prices in the near future so we can see positive abnormalities and vice verse.
On t‐1 day we can see a huge dip in the average abnormalities and the reason attributable to these that there are day traders/ speculators who go in for a huge selling of stock prices because they are expecting a huge dip in the prices on the event day so they sell now and buy at lower prices and book profits.
On Event day the prices of the stocks fall down because the effect of dividend from the stock prices gets reduced. A stock price has two components one is the dividend and the other is the intrinsic value of the stock.
On t+1 day to t+2 day: the prices rise because of positive results (increase in dividend declared or satisfaction of the investors with the company) on an average.
T+3 day to t + 5 day abnormalities are gradually reducing and the effect diminishes almost by the 5th
11 2 3 4 5
Fig 3: Average Abnormality Curve for the 11 day Event Window
1. As the Data is secondary in nature there is no source of reliability.
2. The impact of the various other events occurring during the event window or estimation window cannot be eliminated while calculating the abnormal return. The other various events could be
a. Merger and acquisitions
b. Stock split
c. Any local or global news like monetary or fiscal policy changes or sanctions/quotas by world bodies like OPEC etc..
3. Only ten could be taken for analysis due to time constraint.
Need help with your literature review?
Our qualified researchers are here to help. Click on the button below to find out more:
In addition to the example literature review above we also have a range of free study materials to help you with your own dissertation: