Dividend clienteles, financial leverage clienteles, and the determinants of dividend policy under the New Zealand full imputation tax regime : this dissertation is submitted in partial fulfilment for the award of the degree of Doctor of Philosophy in Accounting, Massey University, Auckland, New Zealand

Abstract

Under the New Zealand imputation tax regime, and in contrast to the classical tax regime, investors are to large extent indifferent between receiving their income in the form of cash dividends or capital gains. This study capitalizes on the unique features of the imputation tax regime with regard to tax neutrality, the minimum effect of discreteness, the homogeneity of investors relative to a classical tax regime, and the introduction of FITC, to examine the tax-induced dividend, financial leverage, and the joint clienteles. Furthermore, the study examines institutional and foreign investor's clienteles as an alternative explanation to the ex-dividend day share price behaviour. It also develops a cross-sectional model to explain the determinants of dividend policy. The results indicate that share prices drop by less than the amount of dividend on the ex-dividend day. The implication of this finding is that the behaviour of share prices on the ex-dividend day is not due to tax differential between dividend and capital gains. The distribution of dividend yields and leverage ratios is not found to exhibit a bimodal behaviour. Their correlation with the implied marginal tax rate is negative and is not significant, negating the dividend and financial leverage clientele hypotheses. In addition, companies with high dividends-low leverage, and vice versa, have approximately the same implied marginal tax rate as those firms with a high dividend yield-high leverage and firms with low dividend yield-low leverage. This suggests that there is nothing unique about the clienteles of firms with either high dividends-high leverage or low dividends- low leverage, and does not support the joint clientele hypothesis. The relationship between the percentage of financial institutional holdings and the degree of foreign ownership and dividend yield is not significant and does not exhibit any bimodal behaviour. This finding does not support the hypothesis that firms with high dividend yield attract more institutions (relative to retail), and more foreign (relative to domestic) investors. To the extent that institutional (relative to retail), and foreign (relative to domestic) investors represents heterogeneous groups, the results suggest that investors heterogeneity does not affect the drop off ratio or influence the results of this study. The results are consistent with the argument that marginal investors in aggregate determine market prices by their collective activity and thus the implied marginal tax rate. The results of the cross-sectional analysis indicate that dividend policy is determined by a firm's growth opportunities, profitability, firm's size, riskness of operating income, and firm's growth rate. These are the same fundamental factors that determine dividend policy under the classical tax regime. This suggests that regardless whether the market is small, such as the New Zealand capital market, or large, such as the USA capital market, corporate dividend policies are determined by basic fundamental factors and tax has a minor role to play. Give that the percentage of financial institution and firm size is negatively related with dividend yield, the positive and significant correlation between firm's size and the percentage of institutional holdings suggest that institutions tend to invest in large companies seeking for liquidity and willing to accept lower dividend yields in exchange for this liquidity premium.