Exchange-traded funds are fashionable. Most, as measured by total expense ratios, are cheap compared with traditional open-end mutual funds. ETFs offer potential tax benefits and can be bought and sold throughout the trading day whereas mutual fund transactions occur only at daily closing quotes.

So it would seem to be a no-brainer: Forget about mutual funds and buy and sell those trendy ETFs, right?

Not necessarily!

An ETF's advantage of a lower expense ratio than a comparable no-load mutual fund can evaporate because of brokerage commissions on trades. For the same reason, small transactions for portfolio rebalancing purposes or for regular "dollar cost averaging" programs are impractical with ETFs.

Even though ETFs are great investment vehicles in many respects, investors are advised to look closely at all costs -- especially brokerage commissions -- from every angle before opting for ETFs over mutual funds.

As a simplified example, suppose you're interested in a no-load mutual fund with an annual total expense ratio of 0.20%. But an ETF with an identical portfolio exists, priced at $42 per share, with a total expense ratio of only 0.07%.

If you intend to invest slightly more than $4,200 and have an online brokerage account that charges $20 for a typical trade, it would cost $4,240 to buy 100 shares of the ETF and then sell the holding some time in the future. If the price were to remain unchanged, the annual expenses would be $2.94. But the yearly expenses for $4,240 of the no-load mutual fund would amount to $8.48.

So you can get $40 more by buying the mutual fund instead of the ETF -- equal to a $20 commission to buy the ETF and then $20 to sell it later on. But the value of the fund investment is being depleted by $5.54 more per year than with the ETF ($8.48 minus $2.94). So dividing $40 by $5.54 reveals that a rough "breakeven" on buying the ETF is 7.2 years, meaning that in this case the fund makes sense for the first 7.2 years, with the ETF having the overall cost advantage thereafter.

(The above "breakeven" is described as "rough" because those with quantitative inclinations will point out that "linear" rather than "geometric" extrapolations are used. In addition, "quants" who consider the time value of money will note that the "net present value" of the $20 brokerage commission to sell the ETF in the future is really around $16.26, assuming a normalized 3% "discount rate." Moreover, no bid/ask spreads are assumed on the purchase and sale of the ETF. Nor are they factored into the calculations, since that the sale of the ETF doesn't really "cost" $20 in commissions; rather it results in $20 being deducted from the proceeds. In most cases, all of these assumptions would net out to a value not much different than using the simple methodology above.)

In fact, a real-world example of the above "breakeven" analysis exists, as do many others. The no-load

Vanguard Large Cap Index Fund

(VLACX) - Get Report

and the

Vanguard Large Cap ETF

(VV) - Get Report

, both track the MSCI U.S. Prime Market 750 Index. If we assume 7% portfolio growth, $20 buy and sell commissions on the ETF and a 3-cent bid/ask spread on the initial purchase (note: the growth projection, commissions and bid/ask spreads are all hypothetical and used solely as examples and do not represent forecasts by Ratings. The annual expenses are based on the midpoints of the projected yearend values.), then the breakeven -- where the ETF becomes the more cost-efficient investment -- comes slightly after the end of the sixth year.

The breakeven of an ETF over an open-end mutual fund depends on a number of variables, including:

Expense ratios of the ETF and the comparable no-load fund. A wider differential between the expense ratios will result in a briefer breakeven span.

Brokerage commissions. The higher the commissions to buy and sell the ETF, the longer the breakeven. At Vanguard, commissions range from $8 per trade for to as much as $45 per trade plus 5 cents per share. The example above was for an online investor with less than $1 million in Vanguard investments.

The amount of the purchase. If a fixed-dollar commission is spread out over a larger investment, the percentage gap in cost between the ETF and the comparable no-load fund is narrowed.

The return on the investment. The higher the rate of return, the shorter the breakeven period.

The bid/ask spread of the ETF. Larger spreads translate into lengthier breakevens. In the above example, the bid/ask spread on the Vanguard ETF, over a period of time, varied between 1 cent and 4 cents, with 3 cents the most common spread.

Front-end sales charges and/or redemption fees on funds. Unless very small purchases are anticipated, the sales charges on front-end load funds and those with redemption fees render them uneconomical when compared with ETFs.

If you don't care to go through the calculations comparing the costs of no-load fund with a comparable ETF, a good rule of thumb is to go with the ETF if you anticipate holding the investment, untouched, for at least five years.

Ironically, however, ETFs can be ideal vehicles for "hedging" and active trading. Many are extremely liquid and maintain bid/ask spreads of only a penny a share. For example, $40 billion to $50 billion in shares of the


(SPY) - Get Report

consistently change hands on a typical trading day.

So, generally, ETFs might best be described as "barbell investment" -- very practical for short-term trading as well as for buy-and-hold programs lasting half a decade or more.

Of course, an ETF should be considered if it offers unique coverage of an area that an investor feels is essential. Examples might be a "triple-inverse" ETF that hedges a specific market segment of one with a narrow focus or unconventional portfolio weighting technique that an investor feels has a particularly bright future.

While commissions, total expense ratios and sizes of transactions are the primary considerations in decisions about open-end mutual funds and ETFs, a number of other factors also impact the relative appropriateness of one or the other in various situations. A nearby table summarizes the comparative attributes of each.

Richard Widows is a senior financial analyst for Ratings. Prior to joining, Widows was senior product manager for quantitative analytics at Thomson Financial. After receiving an M.B.A. from Santa Clara University in California, his career included development of investment information systems at data firms, including the Lipper division of Reuters. His international experience includes assignments in the U.K. and East Asia.