Are ETFs Really More Tax-Efficient Than Mutual Funds?

Purveyors of exchange-traded funds hawk their wares as the cure for the common capital gains distribution. For a while, it was hard to assess the claim that ETFs are more tax-efficient than conventional mutual funds because few ETFs had significant track records. Now that more than a third of ETFs have been around for five years or more, we can assess if ETFs have delivered the tax efficiency they promised. Overall, the answer is yes. ETFs have been much more tax-efficient, as measured by Morningstar's tax-cost ratio, than the typical conventional mutual fund. A few exceptions, however, show that ETFs' tax advantage, while large, is not unassailable.

As tax-friendly as ETFs are, it also pays to compare their records with similar conventional mutual fund competitors. For instance,  Vanguard 500 Index's (VFINX) tax-cost ratios for the trailing one-, three-, and five-year periods ended Jan. 31, 2006, are lower than those of both of its ETF rivals:  SPDR (SPY) and  iShares S&P 500 Index (IVV). That means Vanguard shareholders lost less of their returns to taxes than investors in the ETFs. Vanguard 500's actual aftertax returns (assuming an investor doesn't sell the fund at the end of the time period) also are better than its exchange-traded rivals'. Similarly, at the end of January, the five-year tax-cost ratios and tax-adjusted returns of traditional index funds such as the  Schwab 1000 Index  (SNXFX) and the  TIAA-CREF Equity Index  were better than those of their ETF counterparts-- iShares Russell 1000 Index  (IWB) and  iShares Russell 3000 Index  (IWV), respectively. This shows traditional fund managers who pay attention to taxes--for example, by assiduously harvesting losses to offset gains--can be more than competitive with ETFs.

Tax-Advantaged
Before we examine the particulars, let's review why ETFs should be more tax-efficient than traditional mutual funds. ETFs are not immune from capital gains distributions; they may make them, for example, if the indexes they track change or if one company in a benchmark acquires another and pays a premium to do so.

A variety of factors should make distributions rare, though: ETFs are currently all index funds, which--for the most part--have lower turnover than actively managed funds. This helps limit their realization of capital gains. ETF investors also trade shares among themselves, not with the fund, so ETF managers don't have to sell securities to pay off redeeming shareholders. Only large investors, known as authorized participants, deal directly with the funds, and ETFs can satisfy those redemptions by giving those large investors baskets of their underlying portfolios' stocks instead of cash. Finally, ETF managers also can use that in-kind redemption process to get rid of the stock shares with the biggest unrealized gains, thereby limiting the ETF's potential for distributing gains.

The system seems to work. Capital gains distributions have been rare in recent years at most ETF shops. Indeed, we looked at the tax-cost ratios (which measure how much a fund’s annualized return is reduced by the taxes that an investor in the highest tax bracket would pay on distributions) of ETFs that have been around for at least five years. We found that in six of the nine diversified domestic-stock fund categories the average ETF had a lower five-year tax-cost ratio through the end of January 2006 than the typical conventional open-end mutual fund. ETFs showed the biggest advantages in the mid- and small-cap blend and growth squares of the style box, where high-turnover strategies that can generate a lot of capital gains are common among traditional funds. The exceptions were the large-cap categories, where the average traditional fund's tax-cost ratios (0.39% for large blend, 0.13% for large growth, and 0.69% for large value) were lower than those of the typical ETFs (respectively, 0.48%, 0.26%, and 0.73%). Lower expense ratios may be working against ETFs here, because, like regular funds, ETFs tap their income to pay expenses. Because large-cap ETF expense ratios are a fraction of those charged by the average conventional offering in the category, the ETFs had more income to distribute.

Those are the broad category averages. When you compare individual ETFs head-to-head with traditional counterparts, they look pretty good, too. The tax-cost ratios of the vast majority of domestic and international ETFs with five-year records ranked in the lowest half of their respective categories and broad asset classes.

Exceptions to the Rule
There were just a few funds, such as  StreetTracks Dow Jones Wilshire Small Cap Value (DSV),  iShares Cohen & Steers Realty Majors (ICF), and  iShares Dow Jones US Real Estate Index (IYR), with higher than average tax costs.

StreetTracks Dow Jones Wilshire Small Cap Value has issued capital gains distributions in every year of its existence, including one that amounted to 4% of its net asset value in 2004. This is due, in part, to the fact that the index this ETF tracks has an unusually high turnover rate: It reached 54% in 2004. The offering's turnover could be lower and tax efficiency better in the future, though. In 2005 it adopted a new benchmark that has rules designed to keep the fund from automatically kicking out stocks on the benchmark's size and style borders.

IShares C&S Realty Majors and iShares Dow Jones US Real Estate Index haven't made many capital gains distributions, but they have distributed a lot of income. The Dow Jones ETF also has made payments consisting of return of capital, essentially giving back all or part of an original investment. Neither REIT income nor return of capital qualifies for the lower 15% tax on dividends enacted in 2003. In general, the government treats REIT distributions as regular income for tax purposes.

Tax Takeaways
These examples, however, may be exceptions that prove the rule: In most cases ETFs are more tax efficient than conventional mutual funds in the same asset classes or categories. Nevertheless, ETFs can surprise you. We'd be wary of those tracking benchmarks that require a lot of turnover, such as  Rydex S&P Equal Weight (RSP), which has a turnover rate of 55%, or  PowerShares Dynamic OTC , which has a turnover rate of 112%. These ETFs have been successful at avoiding capital gains thus far, but it's not certain they can keep that up indefinitely.

Disclosure: Barclays Global Investors (BGI), which is owned by Barclays, currently licenses Morningstar's 16 style-based indexes for use in BGI's iShares exchange-traded funds. iShares are not sponsored, issued, or sold by Morningstar. Morningstar does not make any representation regarding the advisability of investing in iShares that are based on Morningstar indexes.