Effects of tax reform on drop-off ratios and on the ex-dividend and ex-right prices

Abstract

The main purpose of this paper is to examine the impact of the integrated tax system introduced in Taiwan on the valuation of dividends. Based on Elton and Gruber’s (Rev Econ Stat 52:68–74, 1970) model, the ratio of ex-day price drop to cash dividend per share (i.e., the drop-off ratio) should reflect the relative taxes on dividends and capital gains. In Taiwan, the suspension of capital gains taxes, the coexistence of taxable and non-taxable stock dividends, and the change in tick sizes allow us to control for the influences of non-tax factors on drop-off ratios. In this paper, we find significant increases in drop-off ratios for both cash dividends and taxable stock dividends after Taiwan’s tax reform (in 1998), while we find no significant changes in drop-off ratios for non-taxable stock dividends. These results provide further evidence to support the argument that tax affects the valuation of firms.

Appendix

In this appendix we show that under Taiwan’s tax system, which has no capital gains taxes, even though the short-term traders proposed by Kalay (1982) maybe present, the drop-off ratio will reflect the dividend taxes. In the U.S., short-term capital gains are taxed as ordinary income. Kalay agues that if the dividend per share is smaller than the expected price drop by more than transaction costs and taxes, investors can arbitrage by selling the stock short cum-dividend and buying it back ex-dividend; otherwise they can arbitrage by behaving in the opposite direction. The condition for no arbitrage opportunities to exist is:

$$ (1 - \pi_{0} ) \cdot [D - (P_{b} - P_{a} ) - \tau \cdot p] \le 0 $$

(6)

where π₀, the marginal tax rate on ordinary income and short-term capital gains for the arbitrager; τ · p, the expected transaction costs of a “round trip” (i.e., purchase and sale).

The definitions of P _a and P _b are given in Sect. 3. Since the capital gains tax rate is the same as that of the ordinary income, the influence of the tax rate can be factored out. Thus, the drop-off ratio (P _b  − P _a )/D, derived from Eq. 6 reflects only the transaction costs. However, this is not the case in Taiwan since the capital gains taxes have been suspended. Therefore, the condition for no arbitrage opportunities to exist should be modified to be:

$$ [(1 - \pi_{0} ) \cdot D - (P_{b} - P_{a} ) - \tau \cdot p] \le 0 $$

(7)

Rearranging, we get Eq. 8, as follows:

$$ (1 - \pi_{0} ) - \frac{\tau \cdot p}{D} \le \frac{{P_{b} - P_{a} }}{D} \le (1 - \pi_{0} ) + \frac{\tau \cdot p}{D} $$

(8)

From Eq. 8, we know that even under the short-term trader hypothesis, the drop-off ratio in Taiwan will reflect the dividend tax rate π₀. If the transaction costs and tick sizes are small enough, as argued in Kalay (1982), then the drop-off ratio will be close to (1 − π₀), which is the same conclusion proposed by Elton and Gruber (1970).