اصول سرمایه گذاری های مالیات سهامداران

We investigate how shareholder-level taxes are capitalized into stock prices using a model that incorporates the investment and payout decisions of a firm and the investment alternatives available to investors. Shareholder taxes affect stock prices both indirectly, via the effect of taxes on corporate investment decisions, and directly, by reducing both the mean and variance of after-tax returns. In our model, tax capitalization is not eliminated by the presence of tax-exempt investors, does not depend on whether equity is composed of contributed capital or retained earnings, and does not depend on the tax rate faced by a hypothetical marginal investor.

We examine a question that has long been of interest to researchers in accounting, finance, and economics: Are shareholder-level taxes capitalized into the prices of common stocks? We extend prior research by explicitly modeling both the firm's investment and dividend payout decisions, and the investment alternatives available to both tax-exempt and taxable investors. We also investigate how tax capitalization is affected by the presence of tax-exempt investors. Recent papers by Hanlon et al. (2003, hereafter HMS) and Dhaliwal et al. (2003, hereafter DEFB) argue that the existence of tax clienteles and/or the presence of tax-exempt shareholders will reduce or eliminate dividend tax capitalization. For example, HMS state (p. 123) “if taxpayers know dividends will be paid, trading will likely take place so that low tax rate or tax-exempt entities hold the securities at the time tax is assessed,” and that the conclusions of prior research “seem to be at odds with the theory of tax clienteles.”1 DEFB state (p. 180) the “conclusion that shareholder dividend taxes are fully capitalized in equity values is inconsistent with the existence of tax clienteles.” Because neither HMS nor DEFB offer a model of shareholder tax capitalization, it is difficult to evaluate their assertions about the effects of clienteles and tax-exempt shareholders. In this paper, we first develop a simple model showing the circumstances under which tax capitalization can arise, and then formally investigate how (or whether) the existence of clienteles and/or exempt shareholders affect tax capitalization. Our paper is motivated by recent suggestions that more research into tax capitalization is needed. Maydew (2001, p. 397) argues for more theoretical research, stating “without a foundation of underlying theory, tax capitalization research has been unable to address key questions. Under what conditions do we expect tax capitalization to take place? What factors determine the extent of tax capitalization?” HMS (p. 121) argue that “future research might fruitfully consider developing a valuation model that explicitly incorporates dividend-tax clienteles, alternative means of distributing corporate earnings (such as share repurchases, mergers), arbitrage activities by tax-free investors, and the dividend-tax free treatment of complete liquidations. Such a model would endogenize why firms pay dividends in the presence of costly dividend taxes.” Making a similar argument, Graham (2003, p. 1120) states “Several articles assume that companies have clienteles of investors that have similar tax characteristics, and then link these companies’ policies to the assumed investor tax rates; however, it would be helpful to make these linkages more direct.” We examine three alternative settings for an economy with only two investment assets: a taxable bond and a risky dividend-paying stock. The first setting contains only tax-exempt investors; the second setting contains only taxable investors facing identical tax rates; and the third setting contains both types of investors. Using the first two settings as benchmark cases, we investigate how investment decisions, stock prices, and tax capitalization are affected by the presence of both taxable and tax-exempt investors. In the third setting, we allow both types of investors to own stock. We address three questions in our analysis. First, how will shareholder-level taxes affect the corporation's optimal investment and dividend policies? Second, what is the tax clientele for the stock (i.e., will the stock be held by taxable investors, tax-exempt investors, or both)? Third, in the presence of tax clienteles and optimal investment and dividend policies, to what extent will shareholder taxes be capitalized into stock prices? The approach we take in this paper differs from recent papers on shareholder tax capitalization in the accounting literature.2 We jointly derive a firm's investment and dividend policies and the tax characteristics that directly and indirectly affect the price that the investors will pay for the firm's stock. Our approach is in the spirit of the public finance literature that views dividend tax capitalization as the indirect effect of dividend taxes on stock price through its effect on corporate investment decisions (Auerbach, 1984; Poterba and Summers, 1984; Sinn, 1991).3 There are three main results that come from our analysis. First, we demonstrate that shareholder-level taxes induce the firm to forgo dividends and invest in projects that would be rejected if the projects had to be financed with new equity, which illustrates the “trapped equity” problem. The optimal reinvestment policy causes Tobin's q to be less than one in our model, which shows that dividend tax capitalization is in part an internal corporate investment effect. Second, we show that, for a given corporate reinvestment policy, shareholder-level taxes can affect stock price in two ways. First, dividend taxes reduce the variance of after-tax dividends, thus reducing risk for taxable investors. Second, if the tax rate on dividends is less than the tax rate on interest, as is the case in the US since 2003, taxable investors will bid up the price of stocks because dividends are tax-favored relative to interest. These effects show that dividend tax capitalization is in part an external corporate investor effect. Third, we show that the presence of tax-exempt investors does not eliminate shareholder tax capitalization. If the tax rate on dividends and interest is the same, both taxable and tax-exempt investors buy the stock at the equilibrium market price. When the tax rate on dividends is less than the tax rate on interest, two equilibria are possible. If the riskiness of the stock is sufficiently high, risk-sharing considerations cause both taxable and tax-exempt investors to hold stock, despite its tax advantages to the taxable investor. The presence of tax-exempt investors mitigates, but does not eliminate, shareholder tax capitalization in this case. If the riskiness of the stock is sufficiently low, tax considerations for the taxable investor dominate risk-sharing considerations, yielding an equilibrium in which only taxable investors hold the stock. These results make an important contribution to tax research by first identifying the multiple ways that shareholder taxes can affect stock price, and then demonstrating how these tax-related pricing effects would be changed or mitigated by the presence of tax-exempt investors. Equally important, our results suggest which factors are not critical in understanding shareholder tax capitalization. None of the effects of shareholder taxes on stock price that we identify depend on whether equity is composed of contributed capital or retained earnings. In our model, the firm operates in perpetuity, and future dividend payments are from future earnings; current retained earnings and contributed capital are retained within the firm forever. Furthermore, there is no need to identify the tax rate faced by a hypothetical “marginal investor” to understand tax capitalization. Whenever risk-sharing considerations induce investors with different tax rates to buy stock at the equilibrium price, there is no marginal investor. Stock price reflects the tax rates of all stockholders. 4 There are several limitations to our model. First, (as in much of the tax capitalization work in economics) we do not allow stock repurchases as an alternative to dividend distributions because it is well established that corporations pay taxable dividends despite the apparent tax advantage of stock repurchases. Second, we adopt the view that any corporation will continue in perpetuity, thus avoiding the question of how the taxation of the corporation upon liquidation will affect tax capitalization. The third limitation of our model is that it assumes dividends are paid continuously, thus avoiding the tax issues surrounding ex-dividend days and how stock prices change between dividend payment dates. However, as long as any discrete dividend payment period is less than the length of time necessary for stock to achieve long-term capital gain treatment (currently more than 1 year) our results should hold for discrete dividend payment periods as well. Section 2 examines investment policy, stock price, and tax capitalization for dividend-paying corporations when all investors in the economy are tax-exempt. Section 3 examines the same issues when all investors are taxable. Section 4 considers the case in which there are both tax-exempt and taxable investors. Section 5 contains conclusions.

In this study, we investigate how shareholder-level taxes affect the prices of common stocks. We develop an analytical model that explicitly incorporates firms’ investment and dividend payout decisions, shareholder tax rates and investment alternatives, and the presence of tax-exempt investors. We show that shareholder-level taxes can affect stock prices in three distinct ways. First, the taxation of dividends induces the firm to make investments that increase stock price by less than the amount of the investment. This aspect of tax capitalization is an internal corporate investment effect. Second, for a given reinvestment policy, shareholder taxes increase stock price to the extent dividends are taxed at a lower rate than interest. Third, shareholder taxes reduce the variance of taxable investors’ after-tax returns. We emphasize that none of these effects are attributable to the way retained earnings and contributed capital would be taxed upon liquidation, since the firm in our model continues in perpetuity. We also show that tax-exempt investors can mitigate, but cannot eliminate, tax capitalization. With heterogeneous investors, the price of the risky dividend-paying stock reflects the relative risk tolerances of taxable and tax-exempt investors, as well as any tax benefits that taxable investors receive when the dividend tax rate is less than the tax rate on interest. Both taxable and tax-exempt investors are willing to buy the stock at the equilibrium price unless dividends are tax-favored relative to interest and the riskiness of the stock is sufficiently low. When dividends are tax-favored and the riskiness of the stock is sufficiently low, the tax benefit (to the taxable investors) of the lower dividend tax rate outweighs any risk sharing benefits, causing taxable investors to pay more for the stock than tax-exempt investors are willing to pay. Our model clarifies some fundamental questions surrounding tax capitalization by explicitly identifying the multiple ways that shareholder taxes can affect stock prices and by showing that tax capitalization is not eliminated by tax-exempt investors. It also shows that framing the debate in terms of the roles of retained earnings and contributed capital, or in terms of the identity of a hypothetical “marginal investor,” is unlikely to generate useful insights.