Thursday, March 12, 1998

The Daily Workshop Report
by Robert Sheard (TMF Sheard)

LEXINGTON, KY. (Mar. 12, 1998) -- I will undoubtedly regret bringing the topic up again, but many readers have e-mailed me with their own thoughts, so I'd like to revisit the idea of measuring total portfolio returns when one has non-periodic cash flows (either in or out) of the portfolio.

From what I can gather in my responses from several mutual fund companies, the returns they report officially are nothing more than the NAV growth from January 1 to December 31. And that's accurate so far as an investor who’s in the fund for a whole year is concerned.

But I maintain that such a calculation is useless as a total return measure for the entire fund. Let's look at an example. On January 1, the Small Fry Fund opens its doors with $1 million under management and a Net Asset Value of $20 per share.

No one knows about the fund as it's brand new so new money doesn't come in. But from January through June, the Small Fry Fund has a great run, doubling its NAV from $20 to $40 per share. The original million is now worth $2 million.

This fantastic 6-month return attracts the notice of a pension plan, though, which subsequently plows $15 million into the Small Fry Fund. New money, of course, doesn't change the Net Asset Value since funds issue an unlimited number of shares. Even though the fund went from $2 million under management to $17 million under management in the blink of an eye, its NAV is still $40 a share.

Zoom ahead to New Year's Eve. After the $15 million infusion at the end of the first half, the Small Fry Fund didn't take in any new money, but neither did it make a single penny. The total value of the fund is still $17 million on 12/31 and the NAV is still $40 a share.

So what was the fund's total return for the year? Using the unit value approach, it would be a 100% return, but logically, that's absurd. Since $16 million of the fund's $17 million in assets didn't come from growth, how can it claim a 100% return? Yet that's the method apparently accepted as convention in reporting mutual fund returns. Neither the size of the cash flows nor their duration in the portfolio is accounted for in such a calculation, so in fact, it's not a measure of the fund's overall return at all, but merely a measure of how a single share fared from January through December -- not at all the same thing.

The XIRR function in Excel, however, does take those factors into account (size and dates of the cash flows), and gives us a more representative figure for the entire portfolio's return over the whole year. Using the XIRR function for this scenario, the annual return was really only 12.04%, which I think you'll agree is a more accurate figure for the performance of the whole portfolio, not just the shares invested on January 1.

Some will argue that my example is an extreme one, and I agree. But even for extreme cases, the procedures we use to calculate portfolio returns should be accurate or the procedure itself is flawed. So if you're trying to measure a return for a portfolio with irregular cash flows, I wouldn't rely on the unit value method. It only measures how well your units performed that were in the portfolio for the whole year and disregards the effects of the cash flows themselves.

If you don't have Excel, you might try the newest version of Quicken, which several readers have reported calculates the same answers as XIRR on random tests we performed. The more I learn about the fund industry, the more I realize how misleading the marketing can be. Fool on!

[Want to be the first Fool on your block to get a copy of Robert Sheard's forthcoming book? Click here to pre-order your copy of The Unemotional Investor.]