Citation
Abdorreza, Asadi
(2011)
Signaling effects and determinants of corporate dividend policy in Iran and Saudi Arabia.
Doctoral thesis, Universiti Putra Malaysia.
Abstract
Corporate dividend policy is considered as one of the most important financial decisions that managers encounter. Researchers have listed this financial decision as one of the top ten important unresolved issues in the field of advance corporate finance (Brealey & Myers, 2005). Its importance is due to not only the amount of money involved but also the repeated nature of the decision, and interacts with other financing and investment decisions. The previous researches of dividend assert that the dividend decisions tend to differ across countries, especially between developed and emerging capital markets. The researchers have found that dividend policies in emerging markets differed from those in developed markets, reporting that dividend payout ratios in developing markets appeared to be lower than that in developed markets (Glen, Karmokoiias, Miller, & Shah, 1995). In recent decade, Middle Eastern financial markets have experienced a substantial development in size, activity and trading value to be considered as attractive emerging markets for foreign investors. It is notable that the stock exchanges in these financial markets are going to play more important role as investable markets in the international financial system. In this regard, firms’ financial policies would be able to influence this development process by affecting on investment returns and risks involved. Therefore, from this prospective, firms’ dividend decisions are taken into account as one of the most important financial policies by investors and policy makers. Although so far, limited empirical research have been made on the topic in this region and further research need to be done to clarify the various aspects of the decision. The present study investigates the determinants of dividend policy as well as the directional effect of dividend announcements on the stock price in two large Middle Eastern countries Iran and Saudi Arabia. The study employs panel data of 193 listed companies comprising 133 firms in sample of Iran together with 60 firms in sample of Saudi Arabia over the period of 1997 to 2008. The previous pattern of dividend payments documents that the majority of listed firms in both markets are willing to pay out high dividend, indicating the important role of dividend policy in the markets. The first objective of the study is to investigate the directional effects of dividend announcements on the stock returns. Using Risk Adjusted Market model the parameters of a and B are estimated to calculate the abnormal returns around the announcement date. The t-statistic is used to test the significance of the abnormal returns and cumulative abnormal returns for short-term and long-term window around the event date. The results confirm the revaluation effect of dividend change announcements in both markets. In sample of Iran, the daily abnormal return on the announcement day appears to be around 0.2 percent, with significant t-statistics for dividend increase category. Positive abnormal returns are created by dividend increase announcements while the announcements of dividend decrease are followed by negative abnormal returns. Results of no-change in dividend are similar to the case of dividend increase by positive abnormal return around the event day. The cumulative abnormal returns show significant effect of dividend increase announcements especially over the pre-event periods. In Saudi Arabia, the daily abnormal returns vary from -0.2 percent for dividend decreases to 0.3 percent for dividend increases. The effect of dividend decrease announcements is stronger than that for dividend increases on stock price. The results of no-change in dividend, similar to the case of Iran, create positive abnormal return on the event day but not significant. The cumulative abnormal returns show a strong negative pattern for dividend decrease over both pre-event and post-event periods. While for dividend increase category, the cumulative abnormal returns are positive only over pre-event periods. The results confirms that dividend change announcements have informational content in both markets, but the effects of dividend changes announcements in Iran are stronger than that in Saudi Arabia especially for dividend increase category. Therefore, the signaling hypothesis of dividend appears to be supported in both markets by positive reactions to the announcement of dividend increases and negative reactions to dividend decreases in both countries. For the second objective, some static and dynamic regression models are employed respectively to investigate the determinants of firms’ dividend policy in both markets. At the first stage, the study applies static regression models such as pooled OLS regression and GLS model to estimate the parameters, and then several diagnostic tests were used in the study. The results of the specification tests show that for both samples Fixed Effect model is the best static model. Secondly, the dynamic analysis is conducted using a combination of GMM approach and instrumental variable. Under the GMM model, the Arellano and Bond’s (1991) estimator is used to estimate the dynamic determinants of dividend policy. The results suggest that firms’ dividend policy can be highly influenced by several determinants. The findings indicate that for sample of Iran, market to book value, firm size, profitability, lagged payout ratio, market share, and majority ownership positively affect dividend payout ratio, whereas beta coefficient is associated with dividend payout ratio with largest negative effect. But for sample of Saudi Arabia the variables of GDP growth rate, lagged payout ratio, free cash flow, retained earnings, majority ownership, market share, and oil price positively determine dividend payout ratio, while beta coefficient, market to book value, and leverage are negatively associated with payout ratio. This study also established a stylized fact that besides firm specific variables, some macroeconomic variables such as GDP growth rate, interest rate, and oil price influence firms’ dividend policy in Saudi Arabia. In summary, dividend decision is one of the most important financial decisions for listed companies in Iran and Saudi Arabia. The findings of this study show that firms’ dividend policy in selected markets, is affected by various firm characteristics as well as economic factors, and the two most important determinants of dividend decision are market risk (Beta) and market to book value (Tobin’s Q) in both markets. This study reveals the signaling effects of dividend change announcements, in which the announcements of dividend increase create positive abnormal returns while the announcements of dividend decrease result in negative abnormal returns. The findings support the signaling hypothesis of dividend for both Middle Eastern countries.
Download File
Additional Metadata
Item Type: |
Thesis
(Doctoral)
|
Subject: |
Dividends - Government policy - Iran |
Subject: |
Dividends - Government policy - Saudi Arabia |
Subject: |
Economic policy - Iran |
Call Number: |
GSM 2011 3 |
Chairman Supervisor: |
Professor Dr. Shamsher Mohamad |
Divisions: |
Graduate School of Management |
Depositing User: |
Haridan Mohd Jais
|
Date Deposited: |
26 Aug 2014 04:12 |
Last Modified: |
19 Aug 2024 07:23 |
URI: |
http://psasir.upm.edu.my/id/eprint/26998 |
Statistic Details: |
View Download Statistic |
Actions (login required)
|
View Item |