Hedge-fund net equity exposure as a leading indicator of stock market moves. Gordon Fowler Jr. is CIO of Glenmede Trust Company, an independent wealth-management firm based in Philadelphia.

Summary

There hasn't been much in the way of economic data released in May that would change our generally constructive view on the market.
The market has taken a more optimistic view of the future -- that alone is news.
Based on one indicator -- relatively high levels of net equity exposure amongst hedge funds -- it wouldn't be surprising to see a flat equity market over the next six months.

Review and outlook

The stock market vacillates between two types of fear. There is the fear that the world is about to fall apart, and the fear that a good thing is getting away. At the beginning of the year, joy was in the air and investors felt the need to pile in as equity prices rose rapidly. At the end of February, the market deflated, giving up nearly 5%. Since then, the market's confidencehas marched back up; so have price levels. The S&P 500 is within striking distance of an all-time closing high.

A healthy correction, however, may be in the near future. Why do we say this? Because there seems to be a sense among the "smart money" crowd that prices are rising and a good thing is in danger of getting away.

As measured by the S&P 500, the market rose 1.2% last week. Energy, up 4.0%, and telecoms, up 2.7%, led the way. Small caps had a rough week, however, with the Russell 2000 falling 0.7%. International equity results were mixed with the EAFE Index rising by 0.3% despite a stronger dollar. Year to date, U.S. large-cap equity is up 8.1%, small-cap stocks are up 5.0% and the international-equity index is up 9.5%.

Can the market continue to rise? We're inclined to think that the rally can, at the very least, hold its gains. But we would add that a key measure of sentiment may be pointing towards a bit of a pullback.

More important

Given their share of average daily trading volume, the most important marginal players in the stock market are hedge-fund managers. They have mandates to produce absolute rates of return. To ensure capital protection they frequently add equity-market exposure when the stock market rises and shrink that exposure when the market begins to fall. In practice, this means changing the ratio of long (buy) to short (sell) positions.

The research firm ISI surveys hedge-fund managers every week to determine how much exposure they have to the equity markets. The chart below shows the results. According to the survey, they are right now at a very high level of equity exposure.

Once upon a time, when hedge funds were a small part of the market, this wouldn't have been terribly important. Since the collapse of the Internet bubble, however, hedge funds have grown tremendously as investors -- in effect fighting the last war -- have striven to protect themselves from another precipitous decline. The in process, hedge funds may have gone from being smart money worth following to -- perhaps -- the kind of big dumb money that's worth betting against.

|image1|

The anecdotal evidence is that hedge-fund movements have a large impact on day-to-day price moves. Much of the market rally late last year was accompanied by hedge-fund buying. When the Chinese stock market cracked late in February 2007, hedge-fund moves to bring equity exposures back to neutral set off a wave of selling. Given their impact on day-to-day prices, excessive pessimism or optimism by hedge-fund investors would seem to signal that the market is near a relative high or low. We can illustrate the relationship between the net equity exposure of hedge funds and the subsequent six-month return to the S&P 500 by plotting the relationship between the two numbers on a graph.

In the graph below, we mark a point for each combination of the weekly measure from ISI's net equity hedge-fund survey and the return to the S&P 500 over the subsequent six months using data from 2003 to 2007.

|image2|

These points create a "data cloud." You can see that there is a relationship between the two variables. When the net equity exposure for hedge funds is relatively low, the return to the S&P 500 is relatively high over the next six months. When hedge funds reach a relatively high net equity exposure, the return tends to be relatively low over the next six months.

Party stopper

The line that runs through the center of data cloud is a statistical measure of the relationship between these two numbers -- a "regression line." Based on the current net equity exposure of hedge funds (60.4 in the latest ISI survey), the regression line would tell us that the market rally is going to go on hold, and six months from now, the S&P 500 will be at current levels.

But this isn't a precise relationship. You can see that the data points don't line up in a perfectly straight line. But it's nothing to sneeze at either. This equation has an R2 of 0.44, and the coefficient on the "x" variable has a very significant T-stat of greater than 12. Now if the world was fair and data analysis was a required course for all undergraduates, you would say, with a bit of awe in your voice, "Wow, that equation sure has some very impressive statistics." (Mind you, when I drop lines like that at cocktail parties, and conversation comes to an awkward standstill, my wife is apt to smooth things over by saying, "Yes, he's a nerd; but he's my nerd." We celebrate our twentieth anniversary in June.)

But life isn't fair, so you'll probably roll your eyes and hand this analysis to your snarky brother-in-law; a guy who took a statistics course a long time ago and specializes in one-upsmanship. And he'll point out that the standard error of the predicted six-month return is probably close to 3.821%. "In other words," he'll say, "if the expected return from your equation is zero, there is a 16% chance the return to the market over the next six months will be 3.821% or more." Your brother-in-law is right, but since you find your brother-in-law annoying and his children just spilled fruit juice on your new couch, you may be prepared to accept my argument.

I realize this isn't the mother of all forecasts. Constructing regression equations with similar statistical significance is about as easy as developing a well-reasoned, verbal argument. With a little work, it usually can be done. However, it's worth noting that when markets get excessively optimistic -- think the summer of 1987 or the late 1990s -- they tend to take a breather. Given how much the market has gone up over the 12 months or so, this would not be a terrible outcome, and could even be healthy for the market over the longer term. -FWR

.