| Abstract | The main aim of this doctoral thesis is to carry the dividend debate into an emerging market context and contribute more evidence to dividend literature. However, this is done differently from prior research by examining the dividend policy behaviour of an emerging market over a period of time and, therefore, attempting to uncover the behaviour of dividend policy in emerging markets. In particular, the dividend policy of the listed firms on the GCC stock market are analysed. This thesis consists of three empirical chapters that investigate the impact of dividend announcements, dividend smoothness, and the prediction of dividend changes.
The first empirical chapter examines the market response to dividend announcements in an environment where there are no taxes on capital gains and dividends. The hypotheses are tested using the event study methodology is used to estimate abnormal returns to the shares and abnormal trading volume around the announcement date. The results provide evidence for a share price reaction that partially supports the signalling hypothesis because there is no signalling effect to the public dividend announcement, but there is a signalling effect to another event (board meeting) that is reflected in the stock price. In addition, the results show that the GCC stock market is inefficient because of the leakage of information before the announcement of bad news and the delay of share price adjustment when there is good news. Further, the trading volume reacts to dividend change announcements in all three announcements clusters—where dividends increase, decrease, and are constant—thereby lending support to the hypothesis that the announcements of dividend change have an impact on trading volume response due to different investors’ preferences.
The second empirical chapter examines the dividend smoothing behaviour in GCC countries, in emerging markets where the response to news and the economic environment are different from those of developed countries. The empirical evidence shows that the dividend smoothing decision is influenced not only by public information but also by private information. In addition, for the regression analysis, the hypotheses are tested using panel regressions and GMM estimation. The empirical results can be summarised in the following manner: First, the Lintner model shows that the degree of dividend smoothing in GCC firms is approaching the degree of dividend smoothing of a developed market. Second, the results of the determinants of dividend smoothing indicate that agency-based models and information asymmetry theories affect the decisions to smooth dividends in GCC stock market. Finally, and importantly, the results reveal that the dividend smoothing in GCC firms is sensitive to private information of share prices.
The third empirical chapter attempts to investigate the factors associated with a propensity to cut or increase dividends. We use the random effect probit model estimation procedure with unbalanced panel data. We find that the longer (shorter) the time interval between dividend announcements, the larger the probability of a cut (increase) in the dividend, consistent with the view that firms delay (early) the release of bad (good) news. A further contribution to our analysis that we investigate the association between managerial proceedings (change in the capital structure) and dividend decisions. We find that, first, firms that pay back their debt using equity either pay a steady dividend or avoid cutting dividends. Second, firms that attempt to transfer wealth from debtholders to shareholders either pay a steady dividend or do not increase the dividend. The findings further indicate that the size, tangibility, leverage, free cash flow, profitability, and growth are considered the main predictors that help to understand the dividend changes that will likely occur. We further report that a few common factors influence dividend changes in both financial and non-financial firms, while certain factors affect only one group. |
|---|