Which is better: investing in mutual funds or in exchange-traded funds?

That question is somewhat like asking a person which is his favorite organ: his heart or his lungs. There is no good answer when you put at odds two things that are not either/or.

Mutual funds are a perfectly good investment instrument, as are ETFs. If buying and selling intraday is important, then ETFs are the right choice. But, if trading during the day is not relevant, an investor is well served with either mutual funds or ETFs. ETFs tend to have lower expense ratios than mutual funds, but that is a generalization, because extremely low-cost index funds are widely available.

To measure the value of both types of diversified investment vehicles, I built two separate portfolios - one that used only ETFs and another that used only mutual funds. Each portfolio consisted of 12 asset classes: large-cap U.S. stocks, mid-cap U.S. stocks, small-cap value U.S. stocks, non-U.S. developed market stocks, non-U.S. emerging market stocks, real estate, natural resources, commodities, U.S. bonds, U.S. inflation-protected bonds, non-U.S. bonds and cash. Each asset class was weighted equally at 8.33% of the total portfolio, and each portfolio was rebalanced at the end of each year over a three-year period.



Within each of the 12 categories, the largest ETF and the largest fund in terms of total assets was used. There was no attempt to identify the best-performing ETF or mutual fund within each. The same money market mutual fund was used in both portfolios. Returns are net of expense ratios. This analysis does not account for taxes, trading costs and inflation.

The results are shown on the next page in the Titan vs. Titan chart. The ETF-based portfolio and the mutual-fund-based portfolio produced nearly identical returns in the three years from 2010 through 2012 - an 8.13% annualized return for the ETF portfolio and 8.23% for the mutual fund portfolio.

The ETF portfolio had slightly higher tax efficiency and a considerably lower aggregate expense ratio (0.22% versus 0.63%).

Some of the asset classes demonstrate a sizable performance spread between the ETF and the mutual fund. For example, in the U.S. large-cap core category, the largest ETF outperformed the largest mutual fund by 278 basis points.

Proponents of ETFs will grab this as proof that ETFs are the better choice. Yet that would not be a valid interpretation.

In the Lipper fund database, there were 115 U.S. large-cap core funds with assets of more than $500 million. Of those 115 funds, 104 are mutual funds and 11 are ETFs; the largest ETF is SPDR S&P 500 (SPY), which mimics the S&P 500.

Of the 104 U.S. large-cap core mutual funds, 23 outperformed SPY over the three years that ended on Dec. 31, 2012. On the flip side, this means that nearly 78% of those U.S. large-cap core mutual funds underperformed SPY over the same three-year period. We are left with a familiar result: Some mutual funds beat the indexes, but quite a few failed to do so.

It's more compelling to focus on what's arguably a bigger issue: building a well-diversified portfolio. If you are doing so with a large number of asset classes (in this example, 12 asset classes), there is considerably less pressure to pick the "right" fund - whether it is a mutual fund or an exchange-traded fund. The model is the real engine of performance, rather than relying on a couple of superstar funds.



If there is an advantage in favor of ETFs, it would be that the disparity of performance among ETFs tends to be smaller than the disparity of performance among mutual funds. This, of course, is attributed to the fact that nearly all ETFs are index trackers, whereas actively managed mutual funds have a long leash and can allow the manager to stray all over the playground.

This open, active mandate can produce some sparkling results for actively managed funds, but of course, it can also produce some dreadful outcomes. Conversely, the performance of ETFs may at times be more tightly clustered within each specific asset- class category.

For instance, going back to the U.S. large-cap core funds, the range of three-year annualized returns for the 104 mutual funds was from 5.63% to 25.29%, with an average three-year return of 9.74%.

Among the 11 U.S. large-cap core ETFs, the range of annualized return over the three-year period was from 9.46% to 12.43%, with an average return of 10.68%.

The difference in the range of returns between mutual funds and ETFs within the same asset-class category will not always be this pronounced. For example, in the natural resources category, there were 18 funds with assets of $500 million or more. Of those 18 funds, 10 were ETFs and eight were mutual funds. The range of three-year returns among the ETFs was from 0.66% to 16.58%, and among the mutual funds it was from 0.12% to 15.76%.

This world is big enough for both mutual funds and ETFs to productively coexist. They are both useful when used wisely. The key is to see either of them as a means to a bigger end - a well-diversified portfolio - and not as an end in and of themselves.



Read More:
"    Longer Lives, Higher Costs: Is Your Firm Ready?
"    For Advisors, a Big 401(k) Opportunity
"    How a Passive Strategy Built a $1.8 Billion Firm


Craig L. Israelsen is a Financial Planning contributing writer in Orem, Utah, and an associate professor at Utah Valley University's personal financial planning program. He is also the developer of the 7Twelve Portfolio.