CEO termination risk and the likelihood of paying dividends

Abstract

Purpose The purpose of this paper is to empirically examine the relation between the ex-ante termination risk and the likelihood of paying and increasing dividends. Design/methodology/approach Logit regressions are used to estimate the relation between the ex-ante termination risk and the likelihood of paying and increasing dividends. Furthermore, instrumental variable (IV-Probit) and propensity score matching (PSM) techniques are used to check the robustness of empirical results. Findings Empirical results show that termination risk is negatively associated with the likelihood of paying and increasing dividends. However, this relationship is sensitive to the degree of termination risk. When termination risk is high (above the sample median), termination risk is negatively related to the likelihood of paying and increasing dividends, which conforms to the predictions of prospect theory. At low levels of risk (below sample median), termination risk has a positive effect and supports the predictions of agency theory. Research limitations/implications The results of this study are based on data from large US corporations in non-financial, non-regulated industries. Future research should investigate if these results hold for firms in other countries and in financial and regulated industries to better our understanding of the relation between the ex-ante termination risk and the likelihood of paying and increasing dividends. Practical implications Termination risk influences chief executive officer (CEO) discretion over corporate payout policies and CEOs change dividend policy to manage this risk. Originality/value This study contributes to the literature on career concerns and corporate payout policies and documents that high (low) termination risk has negative (positive) effect on the likelihood of paying and increasing dividends. To my knowledge, this is the first study that documents the direct link between termination risk and payout policy.