Morningstar will publish dollar-weighted fund
returns so investors can better assess what they get.

    Morningstar is well known for its mutual fund star ratings, but later this year the research company will introduce another standard for assessing funds-by measuring how much of their gains end up in investors' pockets.
    The Chicago-based financial research company has announced that before the year is out it will start publishing the dollar-weighted returns of the funds on its popular Web site, and a "success ratio" that will represent the percentage of total returns captured by dollar-weighted returns.
    Dollar-weighted returns, as opposed to the time-weighted returns normally used to measure fund performance, are calculated based on cash flows into and out of a fund. Hence, returns achieved during a time of high assets levels are weighted more than returns at a time when assets levels are low. Losses, likewise, count more heavily when more money is invested in a fund, as opposed to when outflows are high.
    Cash flows, in essence, are counted as part of the return-and-loss equation when measuring funds to gain a measure of how much investors actually benefited or were hurt by fund performance. Time-weighted returns, by comparison, measure fund performance with the underlying assumption that investor funds stay in the fund from beginning to end.
    The reality, according to proponents of dollar-weighting, is that investor funds often do not sit in a fund for the long term, meaning an investor's actual gains often trail a fund's official returns. So, while dollar-weighted returns are not an accurate indicator of an active manager's performance, they may be a way of assessing the kind of job fund families do in encouraging investors to use their funds wisely, says Morningstar Managing Director Don Phillips.   
    "What is different about dollar-weighted returns is we weight each month's performance based on the assets in the fund at the time," Phillips says. "The dollar-weighted returns will come closer to capturing the cumulative investor experience."
    Of course, one of the reasons for the new slant on fund ratings-as well as other attempts to refine the traditional investment measuring sticks-is because the cumulative investor experience hasn't been too great the past several years. In addition to the attempt by Morningstar to refine its fund ratings, market analysts have of late been hammering away at new ways for investors to dig deeper into investments-hopefully gaining a few hundred basis points in the process.
    Cap-weighted indexing-a market measure that has been in use for decades by funds emulating the S&P 500 and other indices-has come under increased scrutiny because the measure gives added weight to overvalued companies, as was illustrated by the technology bust of 2000. The new wave of thinking has led to products such as equal-weighted S&P 500 funds, as well as indexes that scrap cap-weighting altogether and focus on "true" company value, to get a more accurate reading of the equity market.
    In a similar vein, the idea of using dollar-weighted returns as a measure has bounced around for several years, as a way to get a truer read on the amount of published market returns that is actually being captured by investors. In December 2004, a study by Ilia Dichev, associate professor of accounting at the University of Michigan's Stephen M. Ross School of Business, concluded that dollar-weighted returns were consistently lower than time-weighted returns through the entire history of stock returns.
    The study found that the return differential is 1.3% for the NYSE/AMEX markets between 1926 and 2002, 5.3% for the Nasdaq between 1973 and 2002, and an average of 1.5% for 19 major stock markets around the world between 1973 and 2004.
    Noting that most investors are actively trading and varying their stock exposure, Dichev said at the time, "While buy-and-hold returns provide a good benchmark to track the investment performance of stocks, they can be a poor measure of the actual return experience of investors, if capital-flow timing affects stock returns."
    What dollar-weighted returns do not measure is the performance of fund managers, says Paul Kaplan, Morningstar's vice president of quantitative research. "A dollar-weighted return is not a reflection of how the manager has done in generating the performance," Kaplan says. "The idea is, that while a fund may have provided good returns on a time-weighted basis, the investor experience could be quite different if investors went in and out of funds at the wrong time."
    What can advisors and their clients get out of dollar-weighted returns? Analysts say at the very least, if a fund's dollar-weighted returns significantly lag its time-weighted returns, it sends a signal that investors are not using it wisely. Moreover, it may suggest that the fund complex is aggressively selling something just because it's hot.
    "If you know that a fund has a big gap in dollar-weighted versus actual and that historically people have done a pretty poor job of investing in the fund, it should tell you this is probably an erratic fund or a misunderstood fund that you should get a handle on before you buy," says Russel Kinnel, Morningstar's director of mutual fund research.
    Phillips feels it can give advisors a better sense of the mistakes investors make, and use it to counsel clients on how to correctly use funds. As an example, he cites a study in which Morningstar looked at ten years of data on high- and low-risk funds for the period ending December 31. Total returns showed little difference, with lowest quartile of low standard deviation funds having an annualized return of 8.70% and the highest quartile of high standard deviation funds with 8.25%.
    The difference was stark, however, on a dollar-weighted basis. Low-risk funds held steady with an annualized 8.53% return, and a success ratio of 98%, while high-risk funds dropped to 5.11%, and a success ratio of 62%.
    The figures indicate that investors lost out on a big chunk of the high-risk funds' returns because they weren't pouring in money until after they had big run-ups, Phillips says. For investors, he says, this indicates that "if you're out buying an energy ETF or a gold fund, you're in more dangerous territory. ... It doesn't mean don't do it, but that other investors haven't done as well" in these types of funds.
The discrepancy is more glaring when dollar-weighted returns are calculated for mutual funds that employ hedge-like strategies-presumably to achieve absolute-return, uncorrelated results. Morningstar looked at 80 hedge-like funds for the ten-year period ending December 31 and found that they had a total annualized return of 3.41% but a dollar-weighted return of 0.48%.
    The results were worse in recent years, with a five-year analysis showing a total return of 1.62% and a dollar-weighted annualized loss of 1.42%. "Investor's timing on these funds has been very weak," he says.
    Phillips believes dollar-weighted returns have more value as a measure of a fund family's performance, as opposed to a measure of individual funds-although he says Morningstar still hasn't decided if it will use dollar-weighting as part of its fund stewardship grades.
    "A fund company doesn't have complete control over how someone uses funds, but that doesn't mean they can't influence control," he says.   
    How funds are sold and marketed are one source of control, as is whether or not they close or raise minimums when returns are overheated and expectations high, he says. Does a fund family run ads touting three-digit returns, or does it take a more long-term approach to performance?
    "There are steps a fund company can take that would improve the investor experience and increase the success ratio," he says.
    He cited DFA Funds as an example of a fund family that has successfully managed investor expectations as well as its investments. For the ten-year period ending December 31, the annualized dollar-weighted return of DFA funds average out to 10.81%, which is actually higher than its time-weighted return of 9.90% for the same period, for a success ratio of 109%.
    Phillips notes that among DFA's policies are mandatory education for advisors who sell their funds, nonparticipation in NTF platforms and a fund lineup that encourages portfolio construction.
    Other large fund families that score high on a dollar-weighted basis include Dodge & Cox, with a success ratio of 98%; American Funds, 95%; Franklin Templeton, 94%; Fidelity, 91%, and Vanguard, 86%. Among the poorer dollar-weighted performers among large fund families were Putnam, 67%, and Janus, 25%, according to Morningstar. In Putnam's case, their score was significantly hurt by the results of one fund, OTC Emerging Growth, in the late 1990s.
    The success of DFA, however, isn't simply a result of their passive investment style, Phillips notes. Again looking at ten-year average returns, active fund investors have had more success in capturing gains. For the universe of active funds, Morningstar's study showed a ten-year total annualized return of 9.18% versus a dollar-weighted return of 7.53%. The gap for passive funds was 9.11% and 7.09%, respectively. The publishing of fund dollar-weighted averages, and their success ratios in capturing total returns, raises another question: How will investors use the information?
    While Phillips says he wouldn't use a success ratio to pick an individual fund, he acknowledges that investors could do so if not properly educated on what the number means. "I think that's a fair point. I think it's going to be importantt how we display this," he says. "We're certainly not going to just throw it out with just the term 'success ratio.'"
    He does believe the numbers will encourage better behavior by fund companies and put the focus back on what's best for investors. "At the end of the day, the way you judge a fund a success or failure is by asking, does it make money for people, does it meet their goals?" he says.