Dividend policy, signalling and free cash flow: an integrated approach

Purpose - Scholars have examined the importance of a firm's dividend policy through two competing paradigms: the signalling hypothesis and the free cash-flow hypothesis. It has been argued that our understanding of dividend policy is hindered by the lack of a model that integrates the two hypotheses. The purpose of this paper is to address this by developing a theoretical dividend model that combines the signalling and free cash-flow motives. The objective of the analysis is to shed light on the complex relationship between dividend policy, managerial incentives and firm value. Design/methodology/approach - In order to consider the complex nature of dividend policy, a dividend signalling game is developed, in which managers possess more information than investors about the quality of the firm (asymmetric information), and may invest in value-reducing projects (moral hazard). Hence, the model combines signalling and free cash-flow motives for dividends. Furthermore, managerial communication and reputation effects are incorporated into the model. Findings - Of particular interest is the case where a firm may need to cut dividends in order to invest in a new value-creating project, but where the firm will be punished by the market, since investors are behaviourally conditioned to believe that dividend cuts are bad news. This may result in firms refusing to cut dividends, hence passing up good projects. This paper demonstrates that managerial communication to investors about the reasons for the dividend cut, supported by managerial reputation effects, may mitigate this problem. Real world examples are provided to illustrate the complexity of dividend policy. Originality/value - This work has been inspired by, and develops that of Fuller and Thakor, and Fuller and Blau, which considers the signalling and free cash-flow motives for dividends. Whereas those authors consider the case where firms only have new negative net present value (NPV) projects available (so that dividend increases provide unambiguously positive signals to the market in both the signalling and free cash-flow cases), in this paper's model, the signals may be ambiguous, since firms may need to cut dividends to take positive NPV projects. Hence, the model assists in understanding the complexity of dividend policy.