This paper examines whether stock prices are discounted for potential capital gains taxes. The extent to which shareholder taxes affect stock prices is central to firm valuation and tax policy and has long intrigued scholars in economics, finance, and accounting. To date, the controversy has largely centered on whether dividend taxes are capitalized in equity values. This paper extends the discussion to consider whether capital gains taxes are capitalized. The equity capitalization of capital gains taxes is assessed by evaluating stock price reactions around a likely change in expected capital gains tax rates, namely the May 1997 budget accord that led to a reduction in the long-term capital gains tax rate from 28 percent to 20 percent itat on.

Shareholders can pay capital gains taxes on trades in the secondary market, share repurchases, and liquidating distributions. Because dividends reduce current stock prices and future liquidation values, reductions in the expected capital gains tax rate should be particularly value-enhancing to firms that do not pay dividends or pay low dividends, ceteris paribus. If investors expect cross-sectional differences in the proportion of earnings ultimately to receive capital gains treatment (e.g., because of dividend policy variation), company valuation should vary cross-sectionally, based on differences between ordinary income and capital gains rates.

The purpose of this study is to determine whether such cross-sectional valuation differences can be detected when expected capital gains tax rates change.

To date, few studies have investigated the effect of capital gains taxes on stock prices and their findings have been inconclusive. Inconsistent with capital gains tax capitalization, Amoako-Adu, Rashid and Stebbins (1992) report stock prices reactions did not vary across dividend yields when the Canadian government introduced a $500,000 lifetime capital gains exemption in 1985.

Conversely, consistent with capital gains tax capitalization, they report that high-dividend yield firms experienced a significantly smaller price decrease than low dividend yield firms when the exemption was reduced to $100,000 in 1987. They do not reconcile the conflicting findings. One possible explanation is that the second test is more powerful because the second exemption amount is more binding for the marginal investor than the first.

Two concurrent papers examine capital gains tax capitalization using a residual-income valuation model (Ohlson (1995)). Harris and Kemsley (1999) infer that equity is discounted for dividend taxes because the coefficient on retained earnings (their proxy for future dividends) in their valuation model is less than the coefficient on other book value. If capital gains taxes also are capitalized, they reason that the coefficient on retained earnings should vary with the firm’s dividend yield and should vary depending on whether the firm repurchases its shares. No clear patterns of variation emerge.