As you can see, mutual funds are far from a free lunch. But you can keep more of what you earn by sticking with low-cost funds. What qualifies as low cost? That depends on how long you plan to own an investment, and what type of fund you're talking about.
When it comes to bond funds, no-load offerings with the lowest possible expense ratios are best for most investors. That's because the difference between the best- and worst-performing bond fund is pretty slim; bond-fund returns differ by just small amounts, so every dollar that goes to expenses really hurts your return. Our advice: Avoid bond funds with expense ratios above 0.60%.
On the stock side, a load fund may make a perfectly fine investment, if you're a long-term investor. But load-fund investors should look for funds with fairly low annual costs, such as those sponsored by American Funds. Their total costs (including sales fees) over a period of years are actually more moderate than those of many no-load funds.
You can find plenty of good funds investing in large-company stocks that charge less than 1% per year in expenses. As with bond funds, the range of returns doesn't vary much, so lower expenses (preferably less than 0.85%) give a fund a decided edge on the competition.
With small-company and foreign-stock funds, expect to pay closer to 1.2% annually. Fund companies contend that it takes portfolio managers and their research teams more effort—and more money—to research tiny companies and foreign firms because there isn't as much readily available information about them. Not surprisingly, these funds pass a portion of their extra costs on to you, their shareholders.
At Morningstar, we put a good deal of emphasis on mutual fund costs, not only because they're often hidden, but because we think favoring lower-cost funds is an easy way to improve your long-term results. We've found that over long time periods, lower-cost funds tend to outperform higher-cost funds. And costs are the only thing about a fund that are absolutely predictable, year in and year out.