TAX PLANNING AND FINANCIAL REPORTING COSTS: A STUDY OF THE CANADIAN MARKET FOR REDEEMABLE PREFERRED SHARES

This dissertation develops and tests Pareto improvement models consistent with the managerial tax planning approach to predict the effect of two episodes which involved redeemable preferred shares: the 1987 tax reform and the 1990 Abstract EIC-13. The dissertation hypothesized that the first episode would have a negative effect on the Canadian market for redeemable preferred shares by reducing the tax advantage (Pareto improvement) of preferred shares over debt. It also hypothesized that financial reporting considerations would interfere with manager's decisions to deal with the expected drop in tax advantage of preferred shares over debt. The second episode was hypothesized to have a further negative effect on the market for redeemable preferred shares, by increasing financial reporting costs. It was also hypothesized that tax considerations would interfere with the managers' decisions to deal with the expected increase in financial reporting costs.; The results were largely consistent with the hypotheses of the dissertation. It was documented that the tax reform of 1987 was associated with a decrease in the incidence of preferred-share financing. Further reductions, consistent with the expected negative effect of the 1990 EIC-13, were also documented. At the micro level, it was shown that the 1987 tax reform was associated with a reduction in the tax advantage of preferred shares over debt. The reductions in the tax advantage of preferred shares were followed by decreases in the proportions of preferred shares and increases in the proportions of debt in the capital structures of the firms affected. This implies a financing-clientele shift to debt. But the dependence between the change in the tax advantage and the change in the financing behaviour of firms was systematically lower for firms with high financial reporting costs. This finding is consistent with the theory's prediction that firms with high reporting costs sacrifice some of their tax benefits in order to look rich to investors. The micro-effect of the episode of the 1990 EIC-13 was not as significant as the micro effect of the 1987 tax reform. The results were consistent with the hypothesis that increases in financial reporting costs would be followed by redemptions of redeemable preferred shares. But the hypothesis that tax considerations interfere with contracting costs was not documented.