When straddling the fence between load and no-load, no-load funds are often the hands down winners.

Of course, just being a no-load fund doesn't guarantee sterling performance -- just as charging loads, or commissions, doesn't engender perpetual losses.

Some load funds have terrific performance and fine management, take American Funds Washington Mutual (AWSHX: news, chart, profile), for example. And some no-load funds have dismal management and outright lousy performance.

However, objective investment advisors have to agree that when all things are equal, no-load funds are better for investors.

Still, advisors who sell load funds require a rationale to keep earning their commissions. Some say things that just don't hold up.

An advisor named Frank, responding to an earlier column, wrote: "Your advice about no-load investing, while true for the short term, one to two years, is not the best advice for long-term investors."

"I use these articles to show clients how easy it is to find bad advice about long-term investing... in the long run, the expense ratios for the upfront sales charge mutual funds will be much less than in the no-load funds."

I don't know where Frank gets his information. But in this case he's flat-out misinformed.

Frank objected to my use of a fund comparison tool at CBS.MarketWatch.com to show that in many respects the Putnam Voyager Fund (PVOYX: news, chart, profile), a load fund, has been inferior to a similar no-load fund, Jensen Fund (JENSX: news, chart, profile).

Frank accused me of deliberately skewing the results of this comparison: "Your advice sounds good when you can look back and pick the time periods you want to, so you can sound really good in your articles. How about telling the whole story next time for long-term investors."

The "whole story" on any investment can be very complex. But I'm happy to accept Frank's challenge to make an apples-to-apples comparison of load and no-load funds. The best way to do that is to eliminate all the variables except the load or absence of one. Keep the portfolio, the manager, the objective and the time periods identical.

Fortunately, there's a fund family based in our hometown of Seattle that offers several well-managed funds in load and no-load versions that are identical except for expenses and the load.

That makes these funds ideal for testing Frank's assertion that "in the long run, the expense ratios for the upfront sales charge mutual funds will be much less than in the no-load funds."

Mutual fund expense ratios aren't merely academic. Expenses come directly out of investors' pockets, and they result in lower performance. Since Safeco introduced the Class A shares in 1996, the no-load shares outperformed the Class A shares in every quarter and every year, without exception.

For the comparison, I chose the venerable Safeco Equity Fund, in operation since March 1, 1932. The fund has Class A shares (SAEAX: news, chart, profile) and no-load shares (SAFQX: news, chart, profile).

The expense ratio for Safeco Equity no-load shares is 0.89 percent, vs. 1.26 percent for the Class A shares. This directly contradicts Frank's contention.

Is this just a fluke, or is it typical?

Morningstar classifies Safeco Equity as a large-cap value fund, one of 128 it follows in that category. Among those 128 funds, the average expense ratio for no-load funds is 0.97 percent. The average expense ratio for front-end-load funds in this category is 1.28 percent.

The pattern holds for funds in other asset classes, as you can see from the following table of examples.

Annual expense ratios, load vs. no-load funds

Source: Morningstar Inc.

Frank wants us to tell "the whole story," and as we will see, it's much worse for load-fund investors than the numbers in this table suggest.

Frank wants to look at the long term. So let's go back as far as we can, to the September 30, 1996 inception of the Class A shares of Safeco Equity Fund. A $10,000 investment in the Safeco Equity no-load shares on that date would have grown to $11,440 by the end of August 2002. Total gain: $1,440, or 14.4 percent.

From looking at the fund's performance figures, which don't take the load into account, you might think that a $10,000 investment in the Class A shares on that same date would have grown to be worth $10,835 by the end of August 2002.

But because of the load, the difference is much greater than that. Frank will surely never explain this to his clients, if on September 30, 1996 one of his clients had invested $10,000 in the Class A shares, $575 of her investment would have been paid to a brokerage house for a sales commission, and only $9,425 would have been invested for her in the Safeco Equity Fund. By the end of August 2002, her account would have grown to only $10,213.

Total gain: $213, or 2.1 percent.

Same fund. Same manager. Same strategy. Same risk level. Same portfolio. Same time period. And same number of dollars out of the investor's pocket. But because of the sales commission and higher expenses, one investor got a gain of $213 while the other got $1,440.

That is the real-world comparison that Frank should be calculating for his clients. But when fund salespeople are faced with facts like these, they typically do little more than change the subject. Unless you listen (or read) carefully, you won't realize what's happening.

Frank said investors will be better off in the long run with load funds. But in fact, the longer an investor holds a fund with higher expenses, the more untrue his statement becomes.

Much of my life is devoted to informing investors of the facts, even when those facts may be inconvenient for Frank and other salespersons. If this amounts to giving "bad advice" to investors, then I happily plead guilty.