A few weeks ago, I

took foreign funds to task for hanging out an "Arbs Welcome" sign. These funds use stale prices to value their portfolios, thereby allowing arbitragers to make easy profits (at other shareholders' expense) by buying shares they know will rise with U.S. markets the next day.

I showed that shareholders in a dozen of these funds could have lost up to 2.5% of their assets overnight during a volatile two-day period in October 1997.

Since '97, some of these funds have removed the "Arbs Welcome" sign by improving portfolio valuation procedures, restricting frequent trading in fund shares and imposing redemption fees. But anecdotal and empirical evidence suggests that some of these efforts are half-hearted or ineffective and that the only solution to this problem may be regulatory action.

The most effective antidote to arbs is fair-value pricing. When closing market prices become stale because of events occurring after foreign exchanges have closed, portfolio managers should update the price, or

net asset value, of their funds by using their own best estimates of fair market value. But critics of fair-value pricing say its subjective nature can give rise to a new set of abuses.

This debate is not simply academic. You need look back only to April to find a prime example of how an arbitrage opportunity can hurt shareholders. On Friday, April 14, the

S&P 500

index fell 5.78% in U.S. trading. Asian markets followed suit the following Monday. Later that day, long after Asian markets closed, the S&P 500 rallied for a 3.25% gain. By 4 p.m. ET, when almost all funds price their portfolios, it was clear that Asian markets would rally on Tuesday.

How did some Asia-Pacific funds respond to these market events? They lighted up their "Arbs Welcome" signs in neon, using the lower, now-stale closing prices on Asian exchanges to value their portfolios.

Not all of the funds mentioned above are equally vulnerable to arbs, however. Class B shares for the Chase Vista Japan and Flag Investors Japanese Equity funds carry stiff 5% deferred commissions payable upon redemption -- enough to preclude arbitragers from profiting in just about any arbitrage scenario. Similarly,

(FPBSX)

Invesco Pacific Basin charges a 2% fee on shares redeemed within three months of purchase, and class C shares for the Delaware New Pacific, GAM Pacific Basin and

(MCDRX)

Merrill Lynch Dragon funds carry a 1% sales charge payable on redemption. Many funds also track frequent trading in order to identify arbitrageurs and deny them purchasing privileges.

But none of these deterrents solves the problem of the daily dilution of fund assets caused by transactions occurring at inaccurate prices. Every time a shareholder buys fund shares at a stale price, other shareholders eat the difference. As money flows into and out of funds over time, the affect on shareholders as a group is negligible. But that's little solace to individual shareholders who, unknowingly, were on the short end of the bargain.

Some Funds Fair Value

If you want to invest in Asian markets and protect yourself against arbitrager-friendly funds, there is hope. There are at least 15 Japan and Asia-Pacific funds that state in their prospecti that they may fair value their portfolio securities if market prices become stale because of intervening events.

But whether funds that reserve the right to fair value actually do it is another question. For example, the

(IVCRX)

Ivy China Region fund, which has one of the best fair-value statements, shot up 0.87% from April 17 to April 18; the

(USCOX) - Get Report

U.S. Global China Region Opportunities fund rose 1.89%. If these funds were fair valuing, the price rise should have shown up a day earlier. In fact, funds that claim to use fair-value pricing gained an average of 0.70% from April 17 to April 18, a full 18 basis points more than for the non-fair-value funds listed in the table above. Fair-value funds may not have an "Arbs Welcome" sign out, but arbs may be more than welcome once inside.

The

(GFCHX)

Guinness Flight China & Hong Kong fund has taken a different approach to the problem of arbitragers. It values its portfolio at 9:30 a.m. ET, which slices the 13-hour time lag between the close of the Hong Kong and New York exchanges to seven hours. Jim Atkinson, a director at

InvestecGuinness Flight

, believes this fresh pricing approach is superior to the admittedly subjective process of fair-value pricing.

But K. Geert Rouwenhorst, a

Yale School of Management

finance professor who co-authored a study on the topic, disagrees. "Our analysis showed that using prices that were as little as seven hours old would still allow an arbitrager to make a killing. Arbitragers could simply use information gleaned from European markets, which would already have been open for five or six hours at 9:30 am ET, to game the system," he says.

The April example provides some support for Rouwenhorst's analysis. At 9 a.m. ET April 17, electronically traded S&P 500 futures contracts were up 1.2%, thus portending a positive day for trading on the S&P 500. In theory, an arb could have invested in the Guinness fund at 9:30 and enjoyed a tidy 1.09% one-day profit.

Still, early-morning S&P 500 futures are weak indicators of the U.S. markets' performance that day. Guinness' fresh pricing approach, coupled with its 2% redemption fee, is likely to be far more effective at keeping arbs away than the efforts of funds whose fair-value policies are merely cardboard sheriffs.

Rouwenhorst argues that funds should fair value every day. He and his Yale colleagues, William Goetzmann and Zoran Ivkovic, found that one could trade foreign stock funds based solely on whether the S&P 500 had climbed or fallen during the day, and do more than twice as well, taking half the risk, as the funds' long-term shareholders.

Rouwenhorst says, "If we all agree that using NAVs based on stale prices is not a good idea, then the industry should develop fair value practices. There are approaches to fair-value pricing that would require very few computations and are based on a fairly simple model that captures most of the fair value adjustment." In fact, Rouwenhorst and his colleagues have developed such a model, which is described in his

paper.

Fair Value Faulted

Meanwhile, Atkinson argues that fair-value pricing is inferior to his approach, in part because such a subjective determination of a fund's net asset value creates the potential for abuse. For example, a fund might misprice its portfolio under cover of a fair-value policy to reduce the volatility of its share price. Or it could use fair-value pricing on Dec. 31 to get an end-of-year boost in its performance. Atkinson is correct, but there's no escaping the fact that Guinness is still using stale prices.

It may be that the only effective solution to the problem of stale prices is to prohibit them. The SEC's current position is that a fund may -- but is not required to -- fair value portfolio securities when events that materially affect the value of the securities occur after the closing of the foreign exchange on which they trade. Last December, the SEC issued its first guidance on fund pricing in 30 years, but failed to prohibit the use of stale prices. Let's hope that the SEC doesn't wait another 30 years to fix the problem.

Mercer Bullard, a former assistant chief counsel at the Securities and Exchange Commission, is the founder and CEO of Fund Democracy, a mutual fund shareholders advocacy group in Chevy Chase, Md.