Empirical Relationship between Accounting Disclosure and Market Returns in the GCC Countries - Dissertation Example

?Exploring the empirical relationship between accounting disclosure and market returns in the GCC countries Introduction In today’s dynamic and global economy, several businesses, especially those that trade on stock markets, operate under the ownership of numerous investors around the world. Besides the opportunity for portfolio diversification, investors prefer stocks in international markets due to a number of other factors like high returns, favorable laws etc. Irrespective of their geographical location, investors remain vigilant over the prospects and performances of the companies they have invested in. The rapid advances in technology and communication have provided instant access of news on companies and capital markets to investors through mediums such as the Internet and television (Mayer, 2003). Thus, information is now analyzed by investors in a matter of minutes or seconds and decisions on stock investments are taken much more frequently. Companies disclose information on their performance or other aspects through various mediums such as earnings reports, corporate communications and management interviews (Weston, 2009). The investor base is constantly on the lookout for any new information from such events that may help them increase their returns or reduce their risk exposure (Schillhofer, 2008). Based on these factors, it is believed that an empirical relationship exists between accounting disclosures of firms and the performance of related stocks in the market. The paper analyzes the potential existence of such an empirical relationship among companies within the Gulf Cooperation Council (GCC) region by using a numerical analysis on historical information on related parameters. Schillhofer (2008) has shown that earnings disclosures have a direct influence on investor preferences that determine the prices and returns from traded securities. Since then, numerous studies have been carried out to understand the relationship between corporate disclosures and the performances of securities related to the target firms. While this strong relationship is demonstrated among firms based in the developed and western economies, behavior demonstrating the relationship between accounting disclosures and stock returns is rather sparse and relatively unexplored among firms based in Oman and other countries belongings to the Gulf Cooperation Council (GCC). This paper is an attempt to provide some further investigation into this phenomenon with a specific focus on the firms based out of the GCC member countries. McCahery (2007) believes that research exploring the value of information disclosed through annual reports and other corporate disclosures for investment analysis and valuation purposed has been rather limited in nature. This however leads one to question the importance of accounting disclosures and why they are so avidly analyzed by investors. Hirschey (2009) have conducted elaborate surveys of investors and investment analysts and have arrived at a broad consensus over the relevance and importance of the target company’s financial statements, footnotes, Management discussions & analysis (MD & A) as well as the various accounting assumptions and estimates of the company. For example, the release of financial statements and accounts is followed by the annual general meeting (AGM), which in the opinion of Ang (2008) perhaps offers the only opportunity for investors to understand the company’s management, their behavior and their practices to improve the sustainability and prospects of the business. However, Heinrich (2006) conducted some analytical studies in this regard only to conclude that such meetings did not generally evoke a collective and unilateral reaction from the market. According to Weston (2009), the extent of information available in the market about a company is directly related to the range of disclosures made public by the organization as well as the extent to which the company’s stock is followed by intermediaries such as financial analysts and brokers. Thus, most researchers including Baker (2010) and Choi (2008) agree that accounting disclosures have an inverse relationship with the firm size and stock market returns. The current study has analyzed the role of three major events related to accounting disclosures – prior disclosures (announcements or PA), release of financial statements (FSR) and annual general meetings (AGM) – in influencing stock market returns. Specific attention has been given to the relative influence of financial reports and accounting disclosures in aggregate on investor decisions. For this purpose, daily share prices have been used as data for the analysis and relevant statistical procedures have been applied. The next sections provide a description of the research aims and methodology employed for this study. The subsequent chapter details the results of the analysis which are then used to arrive at suitable conclusions in relation to the research question. This is followed by a suitable conclusion that summarizes the findings with some plausible recommendations for further study. Research Aim and Motivation Aim The primary aim of the study is to examine and establish whether any empirical relationship exists between accounting disclosures and stock market returns among companies operating in the Middle East and GCC regions. Motivations Studies by various researchers including Chan (1996) and Morse (2001) have identified a strong influence of several corporate disclosure instruments on the subsequent trading in the stock markets. Mayer (2003) has shown the influence of earnings disclosure on the performance of stock prices and share turnover. In fact, he estimates that the impact of such disclosures has been more pronounced since the 1960s. Several other studies like Eaton (2007) have also applied similar methodologies to large sample sizes and utilized complex empirical techniques to understand the role of disclosures in the performance of capital markets. Most of these studies establish a positive influence of accounting disclosure on the returns available to investors in the period following such events. However, such studies have been limited in the case of stock markets in the Middle East. Companies trading on stock markets within the GCC region are supported by a thriving economic environment and corporate activity. The role of factors like globalization and foreign investment are becoming more prominent in the dynamics of the local economy. As such, an analysis of the relationship between disclosures and stock returns from a Middle Eastern perspective seemed appropriate and necessary to reflect current trends in these domains. Research Methodology Data The sample used for this study consists of information pertaining to stock price returns of 18 prominent companies that trade on the various stock exchanges in the GCC region. A set of 3 companies have been selected from every stock exchange in the GCC markets except for Qatar as the Qatar Exchange was formed in a relatively recent period (in 1997) and consisted mostly of securities with infrequent trading and lower market capitalization in comparison to other markets. The data on stock and market returns was sourced from the popular financial information provider – Bloomberg – which also facilitated the analysis carried out as part of this study through inbuilt functions and provisions. The price information for all sample companies is between Jan 2008 and Dec 2008 (total duration of 205 days – 100 days prior to a disclosure event and 100 days after a disclosure event. The remaining 5 days in-between these two time-based categories are denoted collectively as an event period when the market takes time to react to new information). Keeping in mind the issue related to the use of daily returns data (Heller, 2008), the selected companies possessed a rating between 1 and 2 as per the figures published by the Global Risk Management Inc. (for the year 2006). The ratings are based on the market liquidity and the trade frequency. For example, a rating of 1 denotes that a trade occurred within 0.1th of a day while a rating of 2 indicates that the previous trade occurred within a day. All the selected companies have fulfilled both these criterion thereby indicating that none were thinly traded stocks. Methodology As mentioned earlier, this study analyzes 4 major events related to financial disclosures based on the information and impact associated with each of them. The notion here is that if any of these 3 events contains any valuable information, the stock price should move accordingly to result in an abnormal yield or residual which is higher than the average estimate during other normal periods. These abnormal returns are therefore a direct indicator of specific information related to the firm. In simpler words, if the PA, FSR and AGM events provide any new and useful information to investors regarding the firm, there should be a resulting impact on the movement of the share price leading to a significant abnormal return. Thus, the abnormal return, which is the difference between the actual and expected returns on any given day for the stock, serves as one of the metrics employed for this study. Nicolo (2006) has performed a simulation of various events using daily stock price movements. The results of this study were consistent with a previous exercise involving monthly return data based on the OLS approach (Ordinary Least Squares) in identifying abnormal returns in response to such events. To test this perceived relationship, two individual models were applied – the market model and one based on the Capital Asset Pricing Model (CAPM). The market model associated the performance of the target stock with the market return. Based on the widely used market model (Fama, 1969), the basic approach assumes that there exists a constant term for the stock return, denoted by ai in the equation below. Another advantage of this assumption is that possible biased values of betas, determined through stock returns of thinly traded shares are not included. This model can thus be expressed mathematically as: ERi = ai + bi (Rmkt), Where Rmkt = market return, ai = the intercept or constant part of the return model and bi = an estimated parameter denoting the slope of the market return with the expected return of the stock. The CAPM approach determines the returns for a stock based on the market return and the stock’s level of systematic or non-diversifiable risk. The CAPM equation can be expressed as: ER = RFR + ? (Rmkt – RFR), Where ER = expected return of the stock RFR = the risk-free rate ? = the systematic risk of the stick with respect to the market. The analysis for this study is based on the null hypothesis that event disclosures such as PA, FSR and AGM do not have any influence on stock returns. Thus, H0 (null Hypothesis): Event disclosures do not have any effect on stock returns. Ha (alternate Hypothesis): Event disclosures have an effect on stock returns. Potential Limitations The study utilizes stock performance data pertaining to a set of 18 companies operating within the GCC region. While all the selected companies are characterized by the largest market capitalization with respect to their individual stock markets, there is a possibility that similar performance may not be replicated in other stocks of smaller companies that were not included in this study. Thus, a wider study could encompass many more companies for obtaining more precise results. Secondly, the study uses data pertaining to a single year. While the number of individual input values (205 sets of returns data for each company) is large, the potential vulnerability to macroeconomic trends does exist. For example, the sample data belongs to the year 2008 when the global economy was experiencing a financial crisis. As such, trends discovered during this period may not be indicative of overall performance. Thus, a future elaborate study can include a much larger sample, perhaps for the last 10 years, to take all phases of business cycles into account. Variable Estimation For each stock, the daily returns were calculated using the following equation: ARi = ln(Pi + Di) – ln(Pi -1), Where, ’i’ refers to the ith stock in the sample selection and ARi is the continuous compounded return of the stock i. Pi is the closing price for the security in the given month while Di is the declared dividend for the stock during the same period. The market return for the stock exchange to which a particular stock belongs (and trades) has been estimated using the following equation: Rmkt = ln(Ii * (1 + Yi / 260)) – ln(Ii – 1) Where, Ii = the share price index of the relevant stock market, Yi = Dividend yield on a daily basis for stock i. Estimation of the Parameters The parameters ai and bi for the sample firms were determined by applying an OLS (Ordinary Least Squares) regression of the daily returns on the equation. The values of these coefficients were thus determined using a data size of 205 daily returns. For each day within the sample period, the input values were segregated based on the PA event during the year into the pre-disclosure and post-disclosure periods. The results of this regression are summarized in the table below and are based on the market model mentioned previously. Period a b R2 Pre -0.000312 0.752 0.241 Post -0.000375 0.824 0.156 Baker (2010) has noted the influence of the size effect apart from ‘?’ when estimating the abnormal returns. To account for this, a subsequent OLS regression was performed upon including an additional log(size) variable into the equation. The new equation and corresponding results of this regression are presented in the table below: ERi = ai + bi (Rmkt) + ci (log(size)) Period a b c Mean -0.000257 0.812 -0.000008 Standard Deviation 0.0032 0.34 0.0007 The above results clearly demonstrate that the firm size does not have any notable influence on the stock return. Moreover, the constant term ‘a’ as well as the coefficient for the firm size, ‘c’, is not statistically significant. Test for the influence of event disclosures A test to analyze the effects of information content based on the PA, FSR and AGM events was performed using the average cross-sectional abnormal returns data within the sample period. These returns were initially ranked by size. The basis for this test is that release of information on any given day may have an influence on the prices implying that any reaction from the market would be noticeable in the closing prices of the respective stocks during the subsequent period. Thus, the prices during the following day are indicated by the events themselves using the notation ‘PA + 1’, ‘FSR + 1’ and ‘AGM _ 1’. Thus, each member of the 18-company sample generated an abnormal return, R for each of the 200 normal days (non-event) and 5 event days. As mentioned previously, the PA event day was preceded by 100 non-event days. The results for the rankings of the average daily abnormal returns for the two models are listed in the table below: Rank Model Market AR CAPM AR 1 PA 0.043 PA 0.045 2 FSR 0.0402 FSR 0.04 3 FSR + 1 0.0246 FSR + 1 0.0244 4 PA + 1 0.0244 PA + 1 0.0245 5 AGM + 1 0.0187 AGM + 1 0.0186 6 AGM 0.0174 AGM 0.0173 From the above results, it is evident that the results for both the models are identical in nature. The reason behind this phenomenon can be attributed to the use of daily returns data whose ‘?’ values are generally irrelevant in relation to small movements within the market. In such cases, Baker (2010) notes that such similar results may be unlikely when using weekly or monthly information on stock returns. Amongst the event variables sampled, PA was found to possess the highest average abnormal returns across all sample firms at 4.3%, followed by FSR at 4.02%. Likewise, the FSR + 1 and PA + 1 have returns of 2.46% and 2.44% respectively. On a comparative basis, the reactions to AGM disclosures were largely smaller in nature at a mere 1.8%. The analysis of returns for the day after an event suggests the existence of certain spillover effects into the day following a corporate disclosure. In other words, these results also suggest that the market takes beyond a day to react to new incoming information on corporate trends signifying the interest among investors beyond the close of trading on an event day. Conclusion The study has applied an OLS based regression study to identify possible empirical relationships between events pertaining to accounting disclosures and the resulting performance of stocks in the capital markets within the Gulf Cooperation Council (GCC). The study considers three main disclosure events typical of most companies: public announcements (PA), release of financial statements (FSR) and annual general meetings (AGM). The analysis has also ranked all three events based on their relative influence on stock performance. The findings indicate that PA and FSR had the most profound influence on stock returns while AGM had the least effect. These results are consistent with the nature of these disclosure events. Announcements and earnings disclosures occur more frequently (usually every quarter) while AGMs are annual events. As such, the market seeks more information from the first two events while AGMs are usually analyzed later to develop a broader consensus (Mayer, 2003). The analysis also leads to the rejection of the null hypothesis (H0) and acceptance of the alternate hypothesis (Ha) that disclosures have a notable influence on stock market returns, especially in the case of PA and FSR. The results also indicate that any news related to corporate performance are not absorbed immediately by the markets. Instead, trading continues beyond the first day since a disclosure into further trading sessions until the effects of the disclosure reflect completely in the stock price. Further research should expand the methods used in this study to include more companies and a wider sample size to develop more comprehensive, unbiased and qualitative results. References 1. Ang (2008), Corporate governance and corporate finance: a European perspective. New York: Taylor & Francis. 2. Baker (2010), Corporate Governance: A Synthesis of Theory, Research, and Practice. New York: John Wiley. 3. Chan (1996), An Empirical Examination of Information, Differences of Opinion, and Trading Activity. Journal of Financial Economics, 50 (1): 102–128. 4. Choi (2008), Institutional approach to global corporate governance: business systems and beyond. Emerald Group. 5. Eaton (2007), Disclosure and the Cost of Equity in International Cross-Listing. Review of Quantitative Finance and Accounting, 32 (1): 2–21. 6. Heinrich (2006), Complementarities in corporate governance. Berlin: Springer. 7. Heller (2008), Corporate governance lessons from transition economy reforms. Princeton University Press. 8. Hirschey (2009), Corporate governance and finance. Emerald Group. 9. Mayer (2003), Finance, Investment, and Growth. Journal of Financial Economics, 70 (1): 201–215. 10. McCahery (2007), Corporate governance regimes: convergence and diversity. Oxford University Press. 11. Morse (2001), Price and Trading Volume Reaction Surrounding Earnings Announcements: A Closer Examination. Journal of Accounting Research, 20 (1): 328-347. 12. Nicolo (2006), Corporate governance quality: trends and real effects, Issues 2006-2293. International Monetary Fund. 13. Schillhofer (2008), Corporate governance and expected stock returns: empirical evidence from Germany. DUV. 14. Weston (2009), Takeovers, Restructuring, and Corporate Governance. Chicago: Pearson. Read More