Modelling the Dividend Determinants: An Insight into Panel Data Analysis of Companies Indexed in Bse Sensex in India

Abstract

This study investigates the determinants of dividend policy of companies listed in the BSE Sensex in India. The analysis examines how firm-specific financial factors influence three key dividend measures: Dividend Payout (DP), Dividend Yield (DY), and Dividend Per Share (DPS). The study utilizes panel data of Sensex companies covering the period from 2010–11 to 2024–25. Preliminary econometric diagnostics were conducted to ensure reliability of the model. The normality test confirmed that the residuals follow an approximately normal distribution. The correlation matrix revealed no serious multicollinearity among the explanatory variables. The heteroskedasticity test supported the assumption of homoskedastic errors, while the LM test indicated that pooled OLS is inappropriate, thereby necessitating a panel regression approach. The panel unit root test confirmed that all variables are integrated at first difference, and the Kao residual cointegration test revealed the presence of a long-run equilibrium relationship among the variables. The Granger causality analysis identified both unidirectional and bidirectional causal relationships between certain determinants and dividend per share. Based on the Hausman specification test, the Fixed Effect Panel Regression Model was found to be the most appropriate for estimating the determinants of dividend policy. The regression results indicate that firm size (SIZE) and price-to-book value (PB) have a positive and statistically significant influence on all three dividend measures. Return on assets (ROA) positively affects dividend yield and dividend per share, reflecting the role of profitability in dividend decisions. Conversely, debt–equity ratio (DE) and price–earnings ratio (PE) exhibit significant negative relationships with dividends, suggesting that firms with higher leverage or growth expectations tend to retain earnings rather than distribute them. Free cash flow variables (FCF and FCE) show negative but statistically insignificant effects on dividend measures. Market capitalization (MC) demonstrates mixed effects across the models, while earnings per share (EPS) exhibits limited statistical significance. The findings emphasize that firm size, profitability, market valuation indicators, and capital structure play significant roles in shaping dividend policy among major Indian firms. This study contributes to the empirical literature on dividend determinants by providing evidence from leading Indian companies and offers insights useful for investors, financial analysts, and corporate managers in understanding dividend behavior in emerging markets.