The Measurement of Abnormal Returns - Essay Example

? Event Studies and the Measurement of Abnormal Returns Introduction Event is the occurrence of any such activity that impacts its environment and triggers many other activities and happenings. It is a continuous process of everybody’s life. Some events are very common such as birth of a child, death of a beloved one, marriage ceremonies, etc. There are events based on cultures, religions and nations. Similarly all global businesses have events that influence and affect their performance. The financial sector and capital market that deals in foreign currency and company’s stocks also have events that rapidly change their prices, costs and profits. This paper is aimed at identifying the events of stock market and making a case study of one of the events. Many studies were carried out for the events of stock market. Study was made on the influences of stock splits and stock prices by Dolley (1933). Publication of papers in the leading business journals indicate that the event studies were done by Myers (1948), by Barker (1956), (1957), (1958) and by Ashley (1962). Event studies were introduced to the financial experts and managers through two papers first by Ball Brown in 1968 and second by Fama et al in 1969. The methodology of studying events of the capital markets have developed and advanced manifold since then and yet the two papers of Brown and Fama provide the core elements of an event. MacKinlay (1997) The market model developed by Ball Brown and Fama contributed in their success. Their model was patterned after the Capital Asset Pricing Model (CAPM) developed in 1964 by Sharpe. The data from the Center for Research in Security Prices (CRSP) at University of Chicago was used by Ball Brown and Fama which also made it a standard source for research for the entire capital markets. The development of computer hardware and statistical analytical software and its increasing access and usage also played important role in the success of event studies. The key issues of capital structure market were made prominent by papers of Modigliani and Miller (1958), (1961) and (1963) which made studies of event a key empirical tool. The events that can impact capital market include declaration of dividends or earnings, splits of stock, mergers of two or more companies, listings of new companies in exchanges, initial public offerings (IPO) and changes of people at key management positions. The impact of such events can be underreaction, overreaction, abnormal returns and reversals. Corrado (2011) Literature Review There are many types of event studies in the literature such as examination of Return Variances by Beaver (1968), and Patell (1976), studies on volume of stock trading by Beaver (1968) and Campbell and Wasley (1996), analyzes of operating performance by Barber and Lyon (1996) and management of earnings through discretionary accruals by Dechow, Cloan, and Sweeney (1995) and Kothari, Leone, and Wasley (2005). However our paper is focused only the mean stock prices. Corrado (2011) The researches during past thirty years have not changed the basic statistical format and it still concentrate around the measurement of mean and cumulative mean of abnormal return before and after the event. The only major changes that took place are the periods of the data for which mean is calculated. Earlier data of returns were used on monthly basis but today data are used on daily and intraday basis. This helps in measuring the abnormal returns more accurately and determines its effects more descriptively. The second change which has come in the event studies is in the ways of estimating the abnormal returns for events that are long-horizon. The new development of French 3 factor model in pricing asset by Fama also brought some changes in event studies methodology. In spite of these changes, there are serious limitations in the methods of long-horizon and extreme caution is required while making any inferences from it. (Kothari and Warner, 1997, p.301) The model of event study constitute examination of behaviour of the stock price and return on it around occurrence of any common type of event such as ex-dividend announcement, or stock split, etc. The event can take place on any date of the calendar year. This is represented mathematically as follows:- Corrado (2011) Let t=0 as the time of event of any listed company’s stock i and the return as Rit then Where the normal return from stock trading or the expected return is represent by Kit and the abnormal return or the return which is not expected is represented by eit. By decomposition of above equation we find that the abnormal return is the difference between return on a stock less the expected return. Corrado (2011) To define an abnormal return, a model for defining expected return is required. There are many such models like market model, constant expected and returns model, and capital asset pricing model. (Brown and Warner, 1985 and Campbell, Lo, and MacKinlay, 1997) Corrado (2011) Market model establishes the relationship between the performance of an individual stock and the group of stocks that are listed in the same category or stock exchange or portfolio. A value called Beta is assigned to each stock based on this relationship. Beta is one of the main components used for calculating abnormal return. The extent of fluctuation in the price of a stock is determined by its value of Beta. Beta is a common measure of risks amongst all the stocks. It is represented by Greek Letter '?'. It is calculated by regression analysis. When returns are maximized and risks are minimized a tradeoff is achieved and this is exactly what the value of Beta represents. The value of Beta is compared with the market, if it is less than one it indicates that the stock is less volatile and if it is greater than one the stock is more volatile than the market. Investopedia (2005) Mean or Average return of a stock is another important constituent use for calculating abnormal return. The average return of the stock is found from a sample period and then compared to the expected return. The difference is determined as the abnormal return. Standard deviation a statistical measure is another important component used to asses the absolute risk. It is the value of dispersion around a mean or average. The Table 1 below shows the qualitative properties of tests conducted for event studies. It is based on three perspectives; the specification, assumptions about generating returns and the strength against alternative hypotheses. The table also differentiates the properties of event studies on short and long horizon. Corrado (2011) Events are associated with the abnormal returns that vary across the entire set of stock portfolio observed. Such association may result from chance or due to bad modelling or mispricing. This issue remains undecided until the downward adjustment is made in test statistic such that the shift in variance is reflected. The issue of variance shift can be solved by using sub samples of stocks having common characteristics in relation to abnormal performance. Corrado (2011) Table 1: Shows qualitative properties of event study tests. Corrado (2011) When the interest rates are low, the yields from dividend are also low and when the interest rates are high, then the yields from dividend are also high. Adjustment in stock market reflects general economic conditions and the impact of expected dividends are mostly included in it. Today therefore purchasing of stocks that yield highest dividend may not necessarily earn a high rate of return. Malkiel (2003) Methodology The event which is chosen for the case study of this paper is ex-dividend announcements. The data of 90 days were observed for fifty listed companies on New Zealand stock exchange. The stock price was observed for ninty days and there were fifty companies that were chosen. All of the fifty companies are registered with the New Zealand stock Exchange. The data was downloaded from the website http://banker.thomsonib.com/ta. The ninty day’s period for each of the company is different and so is the event date. For example the data of Air New Zealand (AIRN) begins from 24th October 2005 and ends on 27th February 2006. The date of event that is the (zero day) is taken as 13th March 2006. Five days before is the 6th March 2006 and five days later is 20th March 2006. Alpha and beta were obtained and then abnormal return was calculated five days before and five days after the event for each individual company. Cumulative abnormal return was then calculated for each of the fifty companies and then average cumulative of each day was found out as under:- Day -5 -4 -3 -2 -1 0 1 2 3 4 5 Average Abnormal Return -0.72 -0.85 -0.66 -0.54 -0.46 -2.17 -2.29 -2.24 -2.23 -2.31 -2.11 The graphical representation of the cumulative average abnormal return is produced below. Standard Deviation was used to calculate t-static of each day in the formula and the result is produced below: Day -5 -4 -3 -2 -1 0 1 2 3 4 5 t-Static -1.21 -1.43 -1.11 -0.90 -0.76 -3.64 -3.84 -3.76 -3.75 -3.87 -3.54 Findings The value of t-Static found 5 days prior to the event date ranges from –1.21 to -0.76 and on the day of event and five days after the t-static value ranges from -3.64 to -3.54. The threshold for critical value is 2.01. End points of the confidence interval are the critical value. The region that falls between (1-? Alpha) is known as confidence interval or acceptance region and anything outside is called the rejection or critical region. In the t-distribution, the critical value is based on the value of alpha that is the probability of rejection and degree of freedom. The critical value of 2.01 means the alpha is equal to 0.025 and confidence level of significance is 97.50% (1-0.025 = 0.975) without any limitation on degree of freedom. This is illustrated in the graphical form below:- Conclusion The hypothesis that was observed for this case study was as under:- H0: < 2.01 Event does not significantly change abnormal returns. H1: > 2.01 Event of ex-dividend announcement significantly change abnormal returns It can be observed that all the t-static values are either negative or below 2.01 and they do not fall in the rejection region. Hence we can conclude that event of ex-dividend announcement have not changed the abnormal returns significantly. The main cause of such a result is that the operating performance of each company listed in the stock market is continuously monitored by the investors and financial agents. Their reports and financial statements such as balance sheet or profit and loss account and fund flows statements are analyzed periodically and experts make their opinions about the probabilities of dividend quantum. Decisions about the declaration of dividends and their announcements are always made public as a matter of formality only. Prior to any announcements, the news by experts and analyzes made by them are spread in the market and the prices of stocks are adjusted accordingly in the daily trading. Efficiency of the market is determined by the amount of abnormal returns investors are able to make. When market is efficient it is not possible to make abnormal returns upon trading based on past trends. Market that allows investors to make abnormal returns on the basis of past trends is a weak market. Abnormal returns upon trading based on public information are also not available to investors in an efficient market. Market that allows investors to make abnormal returns based on public information is a semi strong. When the abnormal returns can not be made upon trading based on all the possible information then the market is called an efficient market. Robert and Martin (2011) It can be concluded from the results of our case study that the market from which the sample and observations were taken is an efficient market. There were no abnormal returns observed on or around the event dates of fifty companies. References Ball, R., and P. Brown, 1968, “An empirical evaluation of accounting income numbers”, Journal of Accounting Research, vol. 6, pp. 159–178. Barker, C. A., 1956, “Effective stock splits”, Harvard Business Review, vol. 34, no. 1, pp. 101–106. Barker, C. A., 1957, “Stock splits in a bull market”, Harvard Business Review, vol. 35, no. 3, pp. 72–79. Barker, C. A., 1958, “Evaluation of stock dividends”, Harvard Business Review, vol. 6, no. 4, pp. 99–114. Beaver, W. H., 1968, “The information content of annual earnings announcements, Empirical Research in Accounting: Selected Studies 1968”, Journal of Accounting Research, vol. 6, no. Suppl., pp. 67–92. Campbell, C. J., and C. E. Wasley, 1993, “Measuring security price performance using daily NASDAQ returns”, Journal of Financial Economics, vol. 33, no. 1, pp. 73–92. Corrado, J. Charles 2011, Event Studies: A methodology review, Accounting and Finance, vol. 51, pp 207-234, Deakin University, Melbourne, Australia, (viewed 26th September 2011, URL-http://www.tilburguniversity.edu/research/institutes-and-research-groups/center/staff/dejong/preprints/eventstudies.pdf) Dolley, J. C., 1933, “Characteristics and procedure of common stock split-ups”, Harvard Business Review, vol. 11, pp. 316–326. Investopedia 2005, Beta: Gauging price fluctuation, (viewed 26th September 2011, URL-http://www.investopedia.com/articles/01/102401.asp) Kothari, S. P., and J. B. Warner, 2005, “Econometrics of event studies, in: B. Eckbo Espen, ed.”, Handbook of Corporate Finance: Empirical Corporate Finance (Handbooks in Finance Series, Elsevier, North-Holland), pp. 3–36. MacKinlay, A. C., 1997, “Event studies in economics and finance”, Journal of Economic Literature, vol. 35, no. 1, pp. 13–39. Malkiel, G. Burton, 2003, The Efficient Market Hypothesis and Its Critics, Paper No. 91, Princeton University, (viewed 26th September 2011, URL-www.princeton.edu/~ceps/workingpapers/91malkiel.pdf) Modigliani, F., and M. H. Miller, 1958, “The cost of capital, corporation finance and the theory of investment”, American Economic Review, vol. 48, no. 3, pp. 261–297. Patell, J. M., 1976, “Corporate forecasts of earnings per share and stock price behavior: empirical tests”, Journal of Accounting Research, vol. 14, pp. 246–276. Robert, A. Jarrow and Martin, Larsson, 2011, The Meaning of Market Efficiency, Research Paper Series No. 07-2011, Johnson School, (viewed 26th September 2011, URL-http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1781091) Read More