Explanatory factors of bank dividend policy: revisited
John Theis and Amitabh S. Dutta
D. Abbott Turner College of Business, Columbus State University, Columbus, Georgia, USA Abstract
Purpose – The purpose of this paper is to examine the Dickens et al. model of bank holding company dividend policy. They identified five explanatory factors in a sample of bank holding companies (BHCs). Banking companies typically pay larger dividends and more often than industrial firms. Investors often look at the dividends as being important return variables. Design/methodology/approach – In this study, a sample of 99 firms with 2006 data from governmental reports and Yahoo is used in regression equations to test the relationship of the five identified variables with dividend yields. The analysis is extended to investigate non-linearities between dividend yield and insider ownership. Findings – The paper finds that the original model is robust, but not all variables keep their significance. Insider holdings have a non-linear relationship with dividend yields. Practical implications – The significant factors affecting bank dividend policy help dividend seeking investors find BHCs that return higher dividend yields. Originality/value – This paper reveals a non-linear link between insider holdings and dividend yields among BHCs. Keywords Insider holdings, Dividends, Business policy, Banks, Holding companies Paper type Research paper
Bank dividend policy
1. Introduction This study reexamines the factors explaining bank holding company (BHC) dividend policy. In the initial paper, Dickens et al. (2003) identified five factors that helped explain bank dividend policy for a sample of banks in 1998. The first model in this study replicates the Dickens et al. (2003) study to verify whether their findings hold up for a different sample of banks in another period. Other models use asset growth and other independent variables to test whether they can add more explanatory power to the model. The 99 bank holding companies (BHCs) located in Yahoo Finance’s Southeast and Mid-Atlantic regions comprise the study’s sample. Yahoo Finance not only provides daily price and dividends per share, but also the insider holdings variable. This study confirms and extends the Dickens et al. (2003) model by incorporating the impact of non-linearity in insider holdings on bank dividend policy. There are two reasons to extend the study into the effect of insider holding non-linearity. First, research suggests insider holdings affect debt and dividend policy non-linearly, both in industrial firms and banks (Morck et al., 1988; Wruck, 1989; Gorton and Rosen, 1995). The second and more important reason is Dickens et al. (2003) performed decile analysis which revealed the possibility of a non-linear relationship between the independent variables and dividend yield. This study’s conclusion offers a potential explanation for the non-linear relationships between the variables. 2. Literature review The association between insider holdings, as a measure to alleviate agency problems, and agency costs is typically researched and explained on samples of industrial firms. Rozeff (1982) and Easterbrook (1984) were among the earliest proponents of an existing Managerial Finance Vol. 35 No. 6, 2009 pp. 501-508 # Emerald Group Publishing Limited 0307-4358 DOI 10.1108/03074350910956963
agency relationship. Casey et al. (1999), in a more recent application of the Rozeff model, show dividends changed with the Tax Reform Act of 1986. Dickens et al. (2003) explicitly study bank dividend policy and a set of factors explaining it. Five of their model’s seven variables significantly impact bank dividend policy. The significant variables are investment opportunities, size, insider holdings, dividend history and risk. The relationship between insider...