THANKS TO THE WORLD WIDE WEB, ANYBODY CAN GO WITH THE FUND FLOW: THE DATA ARE POPULAR BUT MAY NOT BE VALID

When it comes to mutual fund yardsticks, few are as closely watched — or controversial — as the daily and weekly estimates of shares bought and sold.

Hedge funds, Wall Street investment strategists and even securities salesmen seek out the daily and weekly figures in an effort to better time trades by gauging the liquidity of the market and assessing investor sentiment.

But many big fund companies have criticized the short-term estimates, noting that sophisticated traders can use the information to trade ahead of funds, compelling portfolio managers to sell stocks on the cheap in the face of shareholder redemptions, or pay more for stocks if they are under pressure to put a cascade of incoming money to work.

gone with the flow

Continuing questions over the value of estimates have prompted the Investment Company Institute, the fund industry trade group, to scrap its monthly flow estimate report. Beginning in July the ICI will publish only its final numbers, which come out about four weeks after the close of each month.

Now fund firms have more to lose sleep over: The two outfits that regularly provide short-term flow estimates — weekly reporter AMG Data Services of Arcata, Calif., and Trim Tabs Financial Services Inc. of Santa Rosa, Calif., which produces daily and weekly estimates — have begun providing fund flow data on the Internet, which means it’s available to the general public.

AMG (www.amgdata.com) updates its web page on Thursday evenings at 7 p.m. Eastern time, and customers receive their reports Friday morning. Trim Tabs, which launched its web site (www. trimtabs.com) early last month, updates its weekly reports on Sundays. Both sites provide week-old fund flow reports to surfers at no cost.

In the past, the firms distributed their short-term estimates to customers by facsimile or mail.

The firms’ figures are widely quoted in the press but do not include the same breadth of fund firms that the Investment Company Institute tracks. Consequently, the figures may differ markedly.

While most mutual fund companies regularly report their monthly fund asset totals to the ICI, and commercial firms like Boston-based Financial Research Corp. and Lipper Analytical Services and Strategic Insight in New York, few big fund companies cooperate with AMG and Trim Tabs.

“It becomes a tool of confusion, it’s counterproductive,” says fund industry consultant Avi Nachmany, of Strategic Insight.

Though the firms’ existing customers are almost exclusively institutions — from fund companies and other money managers, to brokerage firms, who use the data to track the popularity of new products and the mood of investors in general — the firms’ arrival on the Internet extends their reach to any Web surfer.

Despite the popularity of short-term flow data — Trim Tabs says its subscription revenue is up 50% over last year — there are continuing questions over the data’s relevance. AMG’s web site notes that its weekly reports include information gathered from 12,200 open-end mutual funds which represent 55% of the industry’s stock funds, 61% of taxable bond funds, and 64% of municipal bond funds. Trim Tabs says it collects daily data from 500 stock funds representing about 22% of the industry’s equity funds. The firm also samples about 300 bond funds.

But the fund companies who don’t answer these short-term polls include many of the industry’s biggest players — from No. 1- and 2-ranked Fidelity Investments and Vanguard Group, to T. Rowe Price Associates Inc., Franklin Resources and Merrill Lynch & Co. Vanguard and Fidelity alone represent about 22% of the industry’s long-term fund assets.

Says a Vanguard spokesman: “We thought there was a potential that traders could front-run our funds. With that in mind, we refrained from reporting our cash flows other than on a monthly basis.”

wide of the mark

Even when it comes to monthly fund flow estimates, results from AMG and Trim Tabs vary widely from the ICI’s Trends in Mutual Fund Activity report, which collects data from firms representing 95% of the industry’s $5.05 trillion in stock, bond and money market funds. For example, on Trim Tabs’ web site, the firm estimates that its monthly fund flow reports varied an average 21.7% against ICI’s estimates over the past 26 months.

“The times when our numbers differ significantly, in hindsight it comes out that Vanguard got big flows into their index funds,” says Charles Biderman, president of Trim Tabs. “Our clients are interested in the impact of the flows on the market, not so much that we were a few percentage points off.”

AMG President Robert Adler couldn’t be reached for comment.

The opportunity for errors is especially high in December when fund companies make their year-end capital gain distributions.

Perhaps even more important are questions over the analytical value of short-term flow estimates.

“If people are trying to use short-term mutual fund flows as a buy-and-sell signal for big moves in the market they should be looking someplace else,” says Christine Callies, chief investment strategist at Credit Suisse First Boston in New York. Though Credit Suisse subscribes to both AMG and Trim Tabs, Ms. Callies says she uses the reports as a kind of status check as opposed to a predictor of trends. “If you are a short-term trader the numbers could be manipulated to provide an edge, but I think the history of this data with respect to larger moves is questionable to say the least.”

Adds ICI Chief Economist John Rea:

“You can’t look at that short interval and determine whether some change reflects a new disposition by investors toward mutual funds, or whether it’s just the coincidence of when things were timed,” such as 401(k) contributions which tend to happen early in the month.

Indeed, the real value of short-term fund flow estimates, observes T. Neil Bathon, president of Financial Research Corp, is to assess the impact of fund marketing campaigns, as opposed to investor sentiment.