Distribution policy and creation of shareholders' wealth : a study of firms listed on the Johannesburg Stock Exchange

Abstract

Dividend payout decisions remain one of the key functional areas in corporate finance, as it involves the means by which shareholders receive returns on their investments. For many decades, the academic debate on payout decisions has been ongoing as researchers attempted to analyse and explain how these decisions impact on the creation and maximisation of value for shareholders; the fundamental reasons why companies exist. Researchers have not found conclusive answers to put the debate to rest; rather attempts to put together pieces of the dividend dynamics have raised more questions and hence the dividend puzzle. The recognition of share repurchases as payout option (and hence distribution decisions) have made the debate quite complex. The current study, thus sought to contribute to distribution decisions’ debate in a number of ways. The study firstly reviewed the extended dividend payout models of Fama and Babiak (1968), and Andres, Betzer, Goergen and Renneboog (2009) thereby adding further explanatory variables and then tested the extended model in the South African setting. The data of 110 sample companies (Panel 1) which were also disaggregated into 85 value companies (Panel 2) and 25 growth companies (Panel 3) was used. The hypotheses were tested using the ordinary least squares (OLS), difference general method of moments (Diff GMM), system generalized method of moments (Sys GMM) and least square dummy variable correction (LSDVC) estimators. The study confirmed results of similar previous researches and also identified further trends relating to South African corporate setting. It was found that companies have target payout ratios which they adjust towards, also managers are reluctant to change (increase) dividends which may have to be cut again later and in their endeavours to create and maximise value, may have to sacrifice paying dividends. These trends are evident more with growth companies. The study secondly, tested the dividend life cycle hypothesis. A sample of 119 companies (Panel 4) were used in this regard, as well as a disaggregated sample of 86 value companies (Panel 5) and 33 growth companies (Panel 6). The hypotheses were tested using the same estimation procedures as mentioned above. The results showed that the dividend life cycle hypothesis is prevalent among South African companies. Specifically, it was observed that the considered companies pursuing growth projects paid less dividends. Furthermore, the growth companies have shown to be more aggressive in their pursuance for growth and hence are able to create more value for shareholders than value companies. Lastly, the study examined the extent to which share repurchases are used as payout option (i.e., payout flexibility), as well as factors that determine the payout flexibility. The sample number of 52 companies (Panel 7) were used in this regard and hypotheses were tested using the OLS, Diff GMM and Sys GMM. The results indicated that there is inherent flexibility of share repurchases over cash dividends; the size of company has negative and significant correlation with payout flexibility. This implies that larger companies pay out a lower fraction of payout as repurchases, and thus evidence of attitude of managers of these companies relatively different from that of smaller ones; and that share repurchases serve both substitute and complementary roles to cash dividends. This evidence collectively makes unique contribution to existing body of knowledge, particularly, for emerging economic settings, and managers will be provided with enhanced decision alternatives in their endeavours to maximise value.