Is That Fund's Performance for Real?

Plain-Vanilla Tuesday's column touched on an unusual topic: whether a fund's reported results match what its shareholders received.

(This issue should not be confused with Morningstar's statistic of "investor return." That measure attempts to gauge whether investors, in aggregate, used a mutual fund effectively. In contrast, today's subject has nothing to do with investor decisions. Instead, it addresses quirks in performance calculations that can create differences between what is published and what a fund's shareholders realize.)

Discrepancies rarely occur with the plainest of vanilla investments, U.S. mutual funds, save for the effects of one-time sales charges. By convention, mutual fund performance is computed assuming that investors held the fund entering and exiting the time period. Official returns account for all ongoing fund expenses, meaning that for shareholders who did not pay loads during that time period, those figures will be fully representative.

The exceptions occur from misreporting. Sometimes, the entities that send mutual fund data to outside parties will err on a detail, such as the reinvestment date for a fund's distributions. (All fund performance calculations assume that dividend, capital gains, and return-of-capital distributions are reinvested back into a fund, as soon as possible.) The databases usually identify and correct such mistakes, but on occasion the mistakes persist.

Slightly Flavored Exchange-traded funds and closed-end funds operate similarly, with the additional wrinkle that such funds possess two prices, 1) market value and 2) net asset value. The first price is actual and the second theoretical. In other words, market value tracks what investors pay to buy the fund (or receive when selling it), while the NAV shows the underlying worth of those assets. For ETFs, those two figures usually are almost identical. Not so for CEFs, which may sell at large premiums or (more commonly) discounts to their NAVs.

Morningstar publishes performance numbers based on both prices, but its primary figure--the green number at the top left of this page--is derived from market value. That calculation should therefore match investor outcomes, aside from the negligible effects of bid-ask spreads and (even smaller these days) brokerage commissions. In short, the market-value performance figures for ETFs and CEFs line up very closely with what investors actually achieve.

So, too, is the case for most--but not all--collective investment trusts. CITs are unregistered versions of mutual funds. They are often found in large 401(k) plans, carrying lower expense ratios than the originals. CITs usually report prices and distributions in the same fashion as do mutual funds. Sometimes, though, CITs supply only their internally calculated monthly returns, which means that outsiders cannot calculate the performance figures independently, but instead must assume that investment organization got the details right. That situation is less desirable.

Becoming Complicated More faith yet is required to trust the returns cited for separate-account composites. Investment managers frequently create individual portfolios for those investors who wish to hold securities on their own, rather than as part of a pool. Such portfolios might be institutional, as with a $75 million commitment from an endowment fund; or they might come from retail buyers who seek "tailored solutions." But how to discuss the performance of such portfolios? After all, the buyer will hold something new, not something that already exists.

The answer lies with "composite" returns. The investment organization determines which of its accounts follow a substantially similar strategy, calculates the monthly results for each of those accounts, and then sums them into a single composite "strategy" result, which it makes available to outside parties. This figure does not include expenses, because separate-account management fees are negotiated.

Such a structure poses several challenges. One is that regardless of how the composite is created, it is merely an average. Mutual funds, ETFs, CEFs, and CITs are single entities. There may be some difficulties in calculating their performances, but there at least we can attach the stated returns to one particular fund. Not so with separate-account composites. They are merely general guidelines.

What's more, the investment organization decides what goes into the composite, rather than outsiders. Happily, the practice of fattening up the composite by selecting only the healthiest separate accounts has sharply declined, thanks to the 1999 launch of Global Investment Performance Standards. But the compiler must nonetheless exercise judgment, which means that even if all intentions are pure, one company's decisions will vary from another's. Comparability is impaired.

The third concern is that the composite calculation excludes costs. This feature is convenient for consultants, who wish first to evaluate which managers have been successful, and then to consider how much they will pay them. It is less helpful to prospective buyers, though. While it may seem simple to adjust a composite's totals by applying an expense ratio, in practice, investors often oversee that step, thereby comparing the apple of a composite to the orange of a registered fund.

A Hot Mess Finally come those funds that are only available for accredited investors: hedge funds, venture capital funds, buyout funds, private equity funds. As discussed on Tuesday, the difficulties associated with their reporting range from large to massive. Hedge funds are the simplest of the four groups, yet, states HedgeCo.net:

"Most funds report their returns from previous years 'net of all fees.' This means net of management fees and net of incentive/performance fees. However ... some funds report gross returns or returns net of management fees but gross of incentive/performance fees. Still others will report audited net-of-all-fees returns with estimated/pre-audited net-of-all-fees performance for the current year's performance."

And those are the relatively straightforward funds. Things get a lot trickier from there. In one veteran but particularly notable case, an investor in a private equity fund that quoted a 27.1% annualized rate of return discovered that the return on his personal account, over that very same time period, was ... 7.3%. Now that's grim.

Whether the unregistered funds that serve the elite are better than the registered funds that are created for the masses is open to question; each side of the argument has its proponents. From the perspective of trusting the performance figures, however, the contest is a runaway. By and large, everyday shareholders need ask few questions about their funds' reported performance figures, unless they buy separate accounts. Those who buy funds for accredited investors, need ask many.

John Rekenthaler has been researching the fund industry since 1988. He is now a columnist for Morningstar.com and a member of Morningstar's investment research department. John is quick to point out that while Morningstar typically agrees with the views of the Rekenthaler Report, his views are his own.