Can Your Fund Protect You from Market-Timers?

It's time to deal with the market-timers--those rapid-fire traders who dart into a fund one day and out the next for a quick profit. Problem is, foolproof solutions are tough to come by.

First, the fund companies have to want to foil them. If the allegations made last week by New York Attorney General Eliot Spitzer are correct, one reason some mutual funds haven't succeeded in keeping out market-timers is because fund-company executives were actually working with the timers. Or, at least they cooperated with arbitragers who invested enough money to make it worthwhile for the funds to ignore the detrimental effects such quick round-trips can have on their long-term shareholders.

Those fund companies that sincerely do want to stymie the timers have several options. We wrote about one of them--fair-value pricing--two years ago, and under the circumstances, the issue is worth revisiting. Since we last addressed it, several academic papers have delved into the subject, including an insightful one written by Eric Zitzewitz of the Stanford Graduate School of Business in October 2002.

Market-Timing and Its Effects
Why do timers do what they do? Briefly, because of time-zone differences or other quirks, market-timers know they are practically guaranteed to profit by buying a fund one day and selling it the next.

For example, if the U.S. market rises sharply on a Monday, with technology firms particularly strong, the Japanese market, which has its eye on the U.S. and which features many prominent tech stocks, will almost certainly rise overnight. If you buy Jumpin' Jack's Japan Fund on Monday afternoon at 3 p.m. Eastern time, you'll get the fund at its NAV as of market close at 4 p.m., which would not yet have priced in the expected Japan gains. If Japanese prices do rise sharply, the timers can sell the fund the next afternoon and make a tasty profit in 24 hours. Not bad.

Why would funds want to discourage this practice--which, it should be noted, is legal? For one thing, it wreaks havoc on the portfolio manager's strategy. It's tough for a manager to concentrate on which stocks to buy, or how much, when the fund's asset base is swinging wildly from one day to the next. And in order to make sure the fund has money available to pay when the timers redeem shares, it must hold more cash than the manager might deem appropriate. That holds down returns when the market rises. Or if it doesn't have enough cash, the manager might have to sell shares against his or her better instincts. And the one-day inflows themselves can boost the fund's cash level, diluting the gains long-term shareholders otherwise would've made that day. Finally, all the administrative costs raise expenses.

Redemption Fees and Other Sanctions
To thwart them, some funds have imposed redemption fees, typically 1% or 2% of the shareholder's assets, on those who withdraw money within 60 or 90 days after investing it. That has proved only partly effective in regulating asset flows. We found that to be the case in a study a few years ago that looked at a slightly different issue, and anecdotal information we've received in the past couple of years supports that conclusion. Zitzewitz's research also confirms it. One reason is that the timers can still make money even if they have to pay the fee. And trying to apply the fee to shareholders in different types of accounts can get complicated.

Fund officials tell us they also use a second method: tracking down the culprits and barring them from investing in the funds. But this technique also meets with only partial success. In the past, some fund managers and representatives have told me timers are causing them problems, and they're going after them very aggressively. Six months later, they tell me the same thing, and concede it's still a problem.

Though funds should continue to use this tactic, one reason it is not 100% effective may be that some market-timing stockbrokers were purposely trying to evade the fund-company rules by falsifying their identification numbers. TheWall Street Journal reported last week that Massachusetts authorities are investigating that possibility.

Pros and Cons of Fair-Value Pricing
That leaves fair-value pricing. With this method, funds can foil the timers by adjusting their NAVs ahead of time. Instead of relying on the actual closing price of the stocks in the portfolio, they use estimates to account for up-to-the-minute information. In the example given above, Jumpin' Jack's Japan Fund, in calculating its 4 p.m. Monday NAV, would assume the Japanese stocks it owns are going to rise overnight, and would set their prices on Monday afternoon accordingly. The timers thus wouldn't get an artificially cheap price by buying on Monday. The easy profit opportunity is gone.

Some funds do use this technique, but it's not as widespread as it could be. Further, those that do use it, often apply it only in extreme cases. 

Zitzewitz was quoted in The Boston Globe last week saying he's surprised the practice hasn't become more widespread. Maybe it should be. But it's no panacea. For one thing, NAV accuracy is critically important to the industry's credibility. Once a fund starts estimating NAVs, instead of relying on the actual closing prices, the door is opened to vagueness and ambiguity in an area that, ideally, should be based on hard fact. That holds even if funds do their fair-value pricing with the utmost thoroughness and integrity. A guesstimate done with the best intentions is still a guesstimate.

Second, while fair-value pricing an index-fund portfolio is easy and straightforward--you can just check how the futures prices for that index are trading, for example--figuring the estimate for an entire portfolio that does not resemble an index is complex and time-consuming. Although they're not particularly talkative about this issue, fund representatives have told us they use a variety of inputs, such as prices on other exchanges that happen to be open, futures contracts, or quotes from dealers they call. As we learned when such an exercise was necessary for Japanese funds in early 2000, coming to a conclusion about the fair-value price of just one stock takes effort.

Not only would it thus take a long time each day to fair-value every stock in a portfolio, but different funds holding the same securities could come out with different numbers. If that happens often enough, investors could start to wonder what the heck is going on.

Zitzewitz argues that using index futures or third-party evaluators can help discourage timers even for actively managed portfolios, and maybe it would. Nevertheless, relying on that method for non-index funds would not only be complicated, it would raise questions about the accuracy of their NAVs.

And don't forget the worst-case scenario--a fund knowingly assigning inflated values to its stocks in order to pump up the fund's returns for a certain period. No doubt the overwhelming majority of funds would not even consider that. But as we've seen in the corporate world in the past few years (if you didn't already know this), not all people are scrupulously honest. Allowing funds to continually estimate their own NAVs rather than use the hard numbers provided by closing prices raises a temptation to which someone, sometime, could succumb.

No Silver Bullet
Those points are not reasons to shun fair-value pricing. It does seem to be the most effective counterweight to market-timing, and funds should use it, along with the other methods, to combat this nasty and growing problem. But don't mistake it for a silver bullet.

In the end, what's most troubling is that some fund companies--their public pronouncements aside--may not even want to foil the timers. Spitzer's charges make that case--and present some potentially damning evidence. In such instances, the effectiveness of all of the above methods would be moot. The approaches themselves are not ideal, but one thing's certain in the quest to thwart market-timers: Without the will, there certainly won't be a way.