Mutual fund flows (investor money moving in and out of mutual funds) are one of our core indicators. We use them to decide how much to allocate our Powerfund Portfolios into stocks, bonds, and specific sectors. 

Of course, other indications of opportunity, like valuations, are also taken into account. In fact, many of our favorite fund managers consider valuations relative to growth potential when choosing their holdings. But we still use general investor enthusiasm to select the areas in which we'll buy and sell. 

Fund managers are arguably quite skilled at determining relative valuations. For example, an emerging markets manager can tell whether a Brazilian telecom company is a better deal than a Mexican cement company, but he'll never sell his emerging markets stocks after realizing emerging markets may be in a bubble. That's our job. We pull away the punch bowl before the market gets too drunk.

Research, as well as our own experience working with funds for nearly two decades, tells us that fund investors tend to be the most optimistic when they should be wary. Which is why fund investors as a group tend to underperform the market. 

Market volatility generally highlights investment mistakes. The recent market crash and partial comeback offer plenty of examples of how buying high can destroy returns.

One way to gauge how poorly investors fared in a particular fund is to take a peek at the gains and losses on the fund's "books."

Example: If investors put $1 million into a new small fund, and it climbs 50%, the fund likely has about $500,000 in stock gains on the books. Much of that gain is probably unrealized, unless the manager sold the stocks that went up, in which case, much of the gains would have to be paid out that year as dividends.

Here's what typically happens: after a fund goes up 50%, more performance-chasing investors are bound to pile in. Let’s say $100 million goes into the fund, thanks to those nice returns. But then the fund falls 25%. The fund now has about $25 million in unrealized losses on the books (or realized losses, if investors decide to sell and force the manager to liquidate stocks to raise cash,) although the fund itself is still up in actual performance. That's how funds can show positive total returns when they've actually lost money for the typical investor.

Take the T. Rowe Price Emerging Markets Stock Fund (PRMSX). Thanks to the big recovery and strong returns before the crash, this fund is still up over the last one, three, and five years. In fact, the fund has averaged about 15% growth each year for the past five years. There were some volatile years, of course, accounting for a 60% decline in 2008, and roughly an 80% rise in 2009. 

But the fund only had about $600 million in assets five years ago. Investor money has rolled in since then, and assets under management tipped the scales at about $5 billion before the crash in emerging markets (which fell more than the crash in US stocks).

The fund's lousy 2008 meant that investors lost more money than they made in the years leading up to the drop, although if you'd invested five years ago, you would never have actually been underwater during the crash. Even today with the comeback, the fund appears to have net losses on the books.

The more targeted a fund, the greater the timing mistakes tend to be. Not to pick on T.Rowe, which is by and large a good family with quality, relatively inexpensive funds, but the T. Rowe Price Emerging Europe and Mediterranean Fund (TREMX) has been trouble for investors.

The fund, which invests heavily in Russia, enjoyed spectacular returns prior to the 75% plunge in 2008 (a good chunk of it has now come back, with a return just over 100% so far in 2009). Five years ago, the fund had a mere $115 million in investor assets, but it ran up to around $2 billion right before the plunge. That poor timing left the fund with hundreds of millions of dollars in losses on the books. (One upside: anyone getting in now probably won't see a big year-end taxable distribution for quite some time, since they're literally buying someone else's losses.) The losses were so bad that the more targeted fund lost even more investors than the broader emerging market stock fund did. And fewer bought in on the way down.

Ideally, investors should be buying stock funds after others suffer big losses. Lately, the rebound has been so strong that investors are jumping back in a bit faster than usual, given the size of the drop last year, which makes the time frame of many of these deals short-lived.

Since April 2002, we've held an emerging market stock fund in our higher-risk model portfolios. First it was Dreyfus Emerging Markets (DRFMX), which we switched to SSgA Emerging Markets (SSEMX) when Dreyfus converted to a load fund structure in February 2003. It was tough selling a no-load fund back then. Few wanted to buy emerging markets — a sign of opportunity.

We cut this stake back in May 2005 following strong returns relative to the US markets, and cut it completely at the end of February 2006. Emerging markets didn’t make a return to our model portfolios (except for a brief period of shorting in our highest-risk accounts) until February 2009, near the bottom of the emerging markets crash (we used Market Vectors Russia ETF (RSX) for four months). 

Emerging markets peaked in October 2007, but we missed it by a country mile. We'd sold out of emerging markets 20 months earlier. Yet emerging markets crashed well below the level at which we'd sold. Bottom line: we made money in emerging markets in the 2000's, while many investors lost (hopefully the recent recovery will save them). The extra returns we earned in our short-lived buy off the bottom was simply icing on the cake. In our Daredevil portfolio, two of our best three individual fund returns (206% and 80%) came from emerging markets – despite a major crash in emerging markets.

This time, we don't want to leave the party too soon, as we clearly did with our 2006 and 2009 sales in emerging markets. But the investors who pile in during the last hurrahs of strong returns often lose money. The best way to make money is to leave the party when the late crowd shows up.