Should I Use Index Exchange Traded or Mutual Funds?

While not quite in the same league as "how many angels can dance on the head of a pin?", the debate over whether to use index exchange traded funds (ETFs) or mutual funds to implement your portfolio's asset allocations comes close. Let's look at the arguments on both sides of this issue:

Exchange Traded Funds' supporters point to a number of this product's advantages:

(1) Unlike mutual funds, ETFs are continuously priced throughout the trading day, whereas mutual fund sales take place at the end of the day price.

(2) In theory, ETFs should be able to more closely track an index than a mutual fund. Both index ETFs and index mutual funds face the need to reinvest dividends and interest they receive on the securities they own. They both also need to adjust their holdings in response to changes in the companies included in the underlying index they track. However, because they are "open ended" mutual funds, index fund managers are also confronted with the need to provide liquidity to buyers and sellers of their fund's shares, which requires them to hold a percentage of their assets in cash. ETF managers don't have to do this, because purchases and sales of their funds' shares only take place in the secondary market (ETFs are closed end funds). Because they don't have to hold cash to provide liquidity, they should be able to track an index more closely than an index mutual fund.

(3) Because ETFs trade like a stock, an investor can employ a wider range of trading techniques with them, such as stop loss and limit orders, and short sales. An increasing variety of futures and options products have become available on the more liquid ETFs, which creates more potential trading strategies.

(4) The operating expenses on many ETFs have historically tended to be lower than on mutual funds which track the same index, because ETFs don't provide the same level of service to their owners that mutual fund owners receive (e.g., telephone service centers, free fund transfers, check writing privileges, etc.).

(5) Finally, supporters claim ETFs are more tax efficient. A mutual fund is an open ended investment company. When you sell shares in an index mutual fund, you sell them back to the mutual fund company. If offsetting buyers aren't available, the number of outstanding mutual fund shares is reduced, and some underlying shares in the companies that make up the index being tracked will have to be sold by the fund to finance the outflow of cash caused by the net redemptions. This sale of the underlying company shares triggers capital gains distributions for all the owners of the index mutual fund.

In comparison, an ETF trades on a stock exchange; when you dispose of your ETF shares, you are selling them to other buyers, not back to a mutual fund company. As a result, a sale of the ETF shares does not have the potential to trigger a sale of the underlying shares in the companies that make up the index being tracked. In this manner, the potential for unwanted capital gains arriving on your tax return is minimized. This is not to say that ETFs never make distributions. When the composition of the index they track changes (as recently happened with the S&P 500), the ETF trust has to sell and buy shares, and this can trigger capital gains distributions. ETFs also receive dividends on some of the shares they own, which they may also distribute.

Supporters of index mutual funds often respond to these arguments with ones of their own:

(A) Operating expenses are only part of the story. When you buy an ETF, you also pay a brokerage commission, which you usually avoid when you buy an index mutual fund (which rarely carry front end sales loads).

(B) For people who dollar cost average -- investing an amount of money each month into the index fund or funds they own, the ability to avoid trading commissions makes mutual funds a much better deal over time (that is, the avoided sales commissions on ETF purchases more than offset the slightly higher operating expenses charged by the index mutual fund).

(C) If you are a long term, buy and hold investor (as many index fund investors tend to be), the ability to trade ETFs throughout the day, and to employ a wide range of trading strategies really isn't very useful.

(D) Mutual fund companies provide a range of services (e.g., a knowledgeable person on the other end of an 800 number to answer your questions) that many discount brokerages do not (this assumes that, in order to minimize sales commissions, people buy ETFs through discount rather than full service stockbrokers).

(E) In practice, many ETFs have had larger tracking errors versus the index than comparable mutual funds. And in the event of a sharp drop in market prices that causes many investors to sell an ETF, these tracking errors may significantly widen.

As you can see, there are good arguments on both sides of this issue. On balance we are agnostic -- we have concluded that the right index vehicle to use really depends on an investor's individual circumstances, and not on one or two arguments that apply equally to everyone.

Consider the case of an investor who is considering a single investment in either an ETF or a mutual fund that tracks the same index. Which one makes the most economic sense?

As you would expect, there are a number of variables at work. First, the size of the investment matters. Second, the difference in expenses between the index mutual fund and ETF matters. Third, the expected holding period matters. Any difference in tracking errors (assuming both funds track the same index) also matters. Finally, differences in taxes can be important if the investment is held in a taxable account. As we said, theoretically, tax treatment should favor the ETF. And with lower turnover in its holdings, the ETF should generate less investor-level taxes.

Because of the profusion of deep discount commission structures, we've approached this issue as a breakeven problem. Assuming no difference in tracking errors (which, as noted above, may not be a good assumption to make), and leaving the tax treatment aside (because in theory it favors the ETF), we calculate the present value of the expense savings, based on different holding periods. Since these savings are quite certain (i.e., the initial investment is known, as is the difference in expenses), we discount them at 3%, our estimate of the long term real risk free rate of return (we use real rates because we aren't including any inflation in our estimate of future expenses). If the present value of the future expense savings is less than the commission you would have to pay to purchase the ETF, you would be better off buying a no-load mutual fund that tracks the same index.

Here are three examples. First, let's assume a $1,000 initial investment, a 5 basis point (0.05%) difference in the expense ratios (in favor of the ETF), and a flat $10.99 commission at a discount broker. Over all holding periods up to 20 years, you would be better off with the mutual fund (even at 20 years, the present value of the expense savings is only about $7). However, if the expense savings rises to 25 basis points, then you'd be better off with the ETF if you expected to hold it for six years or more. If the expense savings rise to 50 basis points per year, the ETF is a better deal if you expect to hold it for 3 years or more.

Next, let's increase the investment to $10,000. At 5 basis points expense ratio difference, you would prefer the ETF if you expected to hold it for three years or more. When the expense ratio rises to 15bp/year or more, you would prefer the ETF under all circumstances.

For a $100,000 investment, you always prefer the ETF. However, there are two other important points to keep in mind. First, we emphasize that this analysis is for a one-time investment. If you are making regular investments over time (and incurring commissions on each purchase), you may well be better off with the index mutual fund.

Second, if the tracking errors were greater for the ETF than the mutual fund, this would reduce the relative attractiveness of the ETF, as it would effectively reduce, or even reverse, the difference in expense ratios.

All of the topics we've reviewed in these brief introductory notes are covered in more depth in back issues of The Index Investor.