Numerous literature studyingcognitive psychology argue that people tend to overestimate their abilities.This provides a basis for the assumption that managerial psychological biasesaffect corporate finance decisions. A particular and vital bias that is widelystudied is managerial overconfidence. While the existing literature focuses onthe effect of the overconfidence on investment appraisal, merger, capitalstructure and other financing decisions (Heaton, 2002; Malmendier and Tate,2005; Malmendier and Tate, 2005b; Malmendier and Tate 2008; Campbell et al.,2011), the implications for dividends are little understood (Wu and Liu, 2008;Desmukh et al.
, 2013). Wu and Liu (2008) develop atheoretical model examining the impact of managerial overconfidence oncorporate dividend policy. In their study Wu and Liu use Miller and Rock’s(1985) model assuming no agency costs and that the manager’s goal is tomaximise firm’s value. According to them, overconfident CEO is a CEO whooverestimates own ability to maintain transitory earnings. According to thisstudy, overconfident CEOs pay higher dividends than rational CEOs, due to themisleading assessment of earnings. On the other hand, other studies reportopposite results.
Cordeiro (2009) states that overconfident CEOs or managersbelieve in the assumption that their companies are undervalued. Thus, thesemanagers tend to overinvest. Cordeiro claims that overconfident CEOs are lesslikely to pay dividends but managerial overconfidence does not affect theamount of dividends. Deshmukh et al., (2013) constructed a dynamic modelexamining dividend policy and overconfident managers, and empirically tested iton the US firms using the period from 1980 to 1994. According to Deshmukh etal., overconfident managers believe that external financing is more costly thaninternal funds. Thus, overconfident CEOs overestimate the value of projectopportunities.
Deshmukh et al. reveal that overconfident managers prefer to payfewer dividends than rational managers. This assumption is based on theirdesire to invest in future projects. Deshmukh et al. managed to show thiseffect empirically and also; they revealed that the effect is weaker in highgrowth firms. The prediction of overconfident CEO’s tendency to pay fewerdividends was supported by Ben-David et al.
(2007). Ben-David et al. used asurvey-based approach to reveal that overconfident managers prefer torepurchase shares instead of paying dividends.Campbell et al. (2011) arguethat investment rate and CEO optimism are closely related.
According to theCampbell et al. (2011), investment rate might act as a measure of theoverconfidence. Thus, this paper includes the analysis of the investment rateeffect on dividend policy.
The primary objective of thispaper is to analyse the factors affecting corporate dividend policy taking intoaccount both standard and behavioural finance theories. In a standard corporatefinance framework, the paper examines the effect of age and size (life-cycletheory), ROA (signalling theory), debt and cash (agency theory) on dividendpolicy. Also, three characteristics will be taken into account to test whetherthey influence the amount of dividends. Namely, these characteristics will beboard size, investment rate and CEO overconfidence.