Taxable investors should prefer capital gains over dividends. Even qualified dividends (stocks must be held for at least 61 days around dividend ex-dates) are less tax efficient than long-term capital gains. That is so even when qualified dividends and long-term capital gains are taxed at the same rate, as they are today. Since capital gains can be deferred but dividend income is taxed when received, taxable investors should favor stocks whose returns come from capital gains rather than dividend payments.
Unlike dividends, where taxes are paid on the distribution amount, if shares are sold to generate cash flow (creating a “homemade dividend”), taxes are due only on the portion of the sale representing a gain. And if there are losses on the sale (the homemade dividend), the investor gains the benefit of a tax deduction. Even in tax-advantaged accounts, investors who diversify globally (the prudent strategy) should prefer capital gains because the foreign tax credits associated with dividends have no value.
In addition, if dividends were providing more cash than needed to meet spending requirements, a total-return approach would benefit from the time value of not having to pay taxes on the “excess” dividends while avoiding the potential of the dividends pushing investors into a higher tax bracket. And finally, with the benefit of a step-up in basis upon death, capital gains taxes can be avoided. On the other hand, although dividend yield accelerates tax liabilities to the current period, it reduces future tax liabilities (as prices fall by the amount of the dividends).
The tax inefficiency of dividends versus capital gains (and a cash-flow versus a total-return approach, where homemade dividends can be used to meet cash flow needs) raises the question of whether taxable investors, instead of ignoring dividends as a factor in their investment decisions (focusing on the factors to which they want exposure, such as market beta, size, value, momentum and profitability/quality), should try to minimize or shun them altogether. Ronen Israel, Joseph Liberman, Nathan Sosner and Lixin Wang contribute to the literature with their June 2019 study, “Should Taxable Investors Shun Dividends?”
The authors examined the benefit of U.S. taxable investors shunning dividends in value (which has high exposure to dividend yield), momentum (which has low exposure to dividend yield) and quality factor strategies using approximately the Russell 1000 Index constituent universe over a 33-year period from 1985 to 2017. They rebalanced portfolios monthly. They varied factor-risk weights by 10% increments. They then separated the resulting 66 factor models into three 22-factor groups of low-, medium- and high-yield models. Using these alpha models, they constructed long-only portfolios that targeted a tracking error of 5% and a beta of 1.0 with respect to the Russell 1000 benchmark. For each strategy, they created two versions – capital-gains-agnostic and capital-gains-aware. And finally, they examined various levels of dividend aversion from 0 to 1. They assumed that the tax rate on dividends and long-term capital gains is 20%, and the tax rate on short-term capital gains is 35%. Following is a summary of their findings:
- Tax awareness results in a slight increase in realized long-term capital gains but a dramatic decrease in realized punitive short-term gains, resulting in more than a 1% per annum benefit in after-tax returns for an average study modeled.
- Dividend avoidance generally reduces implementation efficiency, thus lowering expected pretax returns by as much as 50 basis points (bps) for an average strategy.
- The reduction in implementation efficiency is particularly pronounced for strategies with naturally higher dividend yields (loss of 70 bps in pretax expected returns, which leads to a reduction in after-tax returns), such as strategies with a large exposure to the value factor. This is particularly unfortunate because this is exactly the type of strategy where one would hope to benefit from reducing punitive dividend exposure.
- Dividend avoidance detracts from the ability to manage capital gains. For example, for an average capital-gains-aware strategy, dividend aversion increases and the dividend yield decreases by 160 bps, but long-term and short-term capital gains each increase by about 50 bps. The result is that under realistic tax rate assumptions, the reduction in dividend tax costs is nearly (but not totally) offset by the increase in capital gains tax costs. In addition, dividend aversion decreases the expected pretax return by approximately 50 bps. The combination of no improvement in taxes and a reduction in pretax expected returns leads to degradation in after-tax expected returns of capital-gains-aware strategies.
Israel, Liberman, Sosner and Wang concluded: “All things considered, the tax benefit of lowering the dividend yield is not enough to compensate for the associated increase in capital gains taxes and decrease in expected pre-tax returns.”
Summary
This study provides valuable contributions to our understanding of how managing dividends can impact the performance of factor-based strategies. It shows that capital-gains-aware strategies can significantly improve after-tax returns (on the order of about 1% per annum). It then shows that shunning dividends lowers the pretax expected return of the factor strategies and has little impact on reduction of after-tax returns tax costs. Given that shunning dividends not only meaningfully reduces a strategy’s diversification implementation efficiency but also reduces the capacity of a fund to implement the strategy, shunning dividends is not a prudent approach. That said, some management of dividends on the margin should add value. For example, if you have the choice of two stocks with the same characteristics (loadings on factors), choose the one with the lower yield.
The findings also explain why fund families such as AQR, Bridgeway and Dimensional, all of which my firm recommends and uses in building portfolios, do not shun dividends in their value funds, even in their tax-aware versions.
Larry Swedroe is the director of research for The BAM Alliance, a community of more than 130 independent registered investment advisors throughout the country.
Read more articles by Larry Swedroe
