Investors avoid getting into funds likely to distribute capital gains and avoid cashing out of funds with large undistributed capital gains.
In Do After-Tax Returns Affect Mutual Fund Inflows? (NBER Working Paper No. 7595), Daniel Bergstresser and NBER Research Associate James Poterba show that at least some mutual fund investors seem to care a great deal about taxes. In a related paper -- Tax Externalities of Equity Mutual Funds (NBER Working Paper No. 7669) -- Joel Dickson, John Shoven, and Clemens Sialm explain why, demonstrating that investors who fail to pay attention to after-tax returns may well end up considerably poorer.
In 1995, just over half of all mutual fund investments were taxable, while the others were held through tax-deferred accounts, such as IRAs. Bergstresser and Poterba use asset and return data provided by Morningstar to track mutual fund inflows; their analysis includes a large sample of U.S. domestic equity funds from 1993 to 1998. The number of funds included in their sample grew from 509 in 1993 to 1607 in 1998.
They find that funds that place higher tax burdens on their investors, by earning a higher fraction of their return in the form of dividends and interest, or by distributing more realized capital gains, are likely to experience lower cash inflows than other funds with similar pretax returns but lower tax burdens. Estimates of the future taxes incurred by investors in each fund in their sample show that the individual tax burden varies by as much as 3 percentage points. Tax burdens even affect mutual fund cash inflows after adjustments for risk, pretax performance, fund age, size, investment strategy, and rating. Funds with large embedded capital gains, that is funds that are likely to distribute greater amounts of realized capital gains in the future, are likely to have both lower gross cash inflows and lower gross cash outflows. This reflects two factors: investors avoid getting into funds likely to distribute capital gains and avoid cashing out of funds with large undistributed capital gains.
For their paper, Dickson, Shoven, and Sialm calculate the aftertax returns on a variety of simulated mutual funds in order to determine the extent to which investors' aftertax returns depend on the actions of other people. Their simulated mutual funds are constructed using historical returns from 1984 to 1998 for the 50 companies with the largest market capitalization on the NYSE, AMEX, and Nasdaq in 1983.
As these three authors explain, new investors in a mutual fund dilute the overall capital gain position of the fund and make tax-sensitive accounting techniques more powerful in reducing the overall tax burden faced by investors. New cash inflows may eliminate the need for the fund to realize capital gains by selling securities to pay shareholders wishing to redeem their shares. Tax efficient accounting techniques that minimize gains by selling the highest cost share of a particular security first can improve aftertax returns. The authors look at alternative investment policies for actively managed funds and demonstrate the aftertax superiority of a policy of systematically divesting stocks with substantial losses rather than deleting securities with large capital gains.
Dickson, Shoven, and Sialm assume for this study that investors face a 39.6 percent marginal tax rate on ordinary income and a 20 percent marginal tax rate on realized long-term capital gain distributions. They conclude that a fund's aftertax return is significantly affected by its accounting method, its cash inflows, whether or not it is closed to new buyers, and whether it follows an active or a passive investment strategy. Even with very similar before-tax returns, the differences in aftertax returns for different mutual fund accounting and management policies can amount to as much as 4 percent per year.
-- Linda Gorman