Consistent with investors discounting share prices for expected capital gains taxes, we find returns are decreasing in dividend yields. During the week of the budget accord, the mean return was 12 percent for non-dividend-paying stocks and 6 percent for dividend-paying stocks. The spread exceeded the difference between the groups for any week from January 1995 to the week in question. No similar price movements were detected in the weeks preceding or following the accord. Results hold after controlling for the normal covariability between dividend-paying and non-dividend-paying firms. Among dividend-paying firms, stock returns are negatively correlated with dividend yields. Here
The inferences of the paper suggest that none of the current perspectives fully explains the role of shareholder taxes in equity valuation. The results are inconsistent with both the “traditional” view that non-tax benefits offset shareholder taxes and the tax-irrelevance perspective (Miller and Scholes (1978)) that marginal investors are tax-exempt. Instead the results suggest that anticipated shareholder taxes affect firm values. However, the findings are inconsistent with the exclusive role of dividends in the new view’s tax capitalization perspective. Additional research is needed to assess the generalizability of these results beyond the capital gains tax rate shock provided by the budget accord.
Finally, possible shareholder clientele effects are ignored throughout the paper. One explanation for the findings is that non-dividend-paying (and low-dividend) stocks are held by shareholders that are subject to greater capital gains taxation than shareholders of other companies. Suppose a firm’s dividend yield varies with its shareholders’ tax status. The marginal investors in non-dividend-paying stocks may be individuals in the maximum personal tax bracket and the marginal investors in other stocks may be pensions, tax-exempt organizations, or foreigners, which are not subject to the capital gains tax rate. If so, cuts in the capital gains tax rate should increase the value of non-dividend-paying stocks more than other stocks.
No tests of the clientele explanation are conducted in this study. Determining whether clientele effects explain the results are beyond the scope of this paper. In particular, we do not test the clientele explanation against an alternative explanation that shareholders are constant across companies and the source of cross-firm variation is the proportion of shareholder profits received in payments subject to the capital gains tax. The findings in this study are consistent with both explanations.
The remainder of the paper is organized as follows: Section 2 reviews the capital gains taxes levied on secondary trades, liquidations, and share repurchases. Section 3 provides background information concerning the budget agreement and the capital gains tax rate reduction. Section 4 presents the empirical tests and results. Concluding remarks follow.