Wall Street Self-defense

When They Say “Buy,’ Sell

The least-bad market prediction tool.

Read Henry Blodget’s detailed disclosure statement here; find the Complete Guide to Wall Street Self-Defense here.

So if assessing whether stocks are “cheap” or “expensive” isn’t a big help in determining what the market will do over the next year, what is? As it turns out, almost nothing. But this week, let’s look at what might be described as the “least-bad” market prediction tool.

Richard Bernstein, the chief U.S. strategist at Merrill Lynch, has tested the effectiveness of several metrics commonly used to predict the performance of the market over time horizons of up to one year, including price-earnings ratios, the slope of the yield curve, changes in interest rates, and several so-called “sentiment indicators,” which measure the relative optimism or pessimism of investors and analysts. Of these, it turns out that the most effective is not P/E ratios or any other commonly used valuation or sentiment tool, but a sentiment indicator of Bernstein’s own invention—the “sell-side indicator.” (Bernstein updates the sell-side indicator every month. The reports are available to Merrill Lynch clients on the firm’s Web site.)

The sell-side indicator measures the consensus outlook of Wall Street strategists, as determined by their recommended asset-allocation models (suggestions as to what percentage of portfolios the average investor should place in stocks, bonds, and cash). Bernstein assumes (probably logically) that, when the strategists suggest that investors place a high percentage of their portfolios in stocks, they are most optimistic about future stock performance, and when they suggest a low percentage, most pessimistic. The sell-side indicator averages the strategists’ recommendations and then plots the result on a chart tracking their relative optimism or pessimism over time.

Wall Street strategists are highly trained professionals with decades of experience, so when they think the market is going to go up, it’s probably going to go up—right? Well, no. The sell-side indicator, like some other sentiment indicators, is a “contrarian” indicator: It holds that when Wall Street strategists are most confident the market is going to rise, it is most likely to fall, and vice versa. The idea behind contrarian sentiment indicators is that most analysts and investors put their money where their mouths are: When they think the market is going up, they own stocks, and when they think it is going down, they don’t. What causes changes in stock prices is not absolute demand but changes in demand, so when everyone’s already bullish, there are few people left to get bullish, etc. Extreme bullishness among strategists, in other words, triggers a “sell” signal on the sell-side indicator, and extreme bearishness, a “buy.”

According to Bernstein, over the last 20 years, the sell-side indicator has predicted about 34 percent of the variability of future S&P 500 returns, a level significantly higher than P/E ratios (20 percent), interest rates (9 percent), and other popular metrics. In recent history, for example, Wall Street strategists were most bearish about stocks in 1989 (S&P 500 up 27 percent the following year); 1995 (S&P 500 up 34 percent, 20 percent, and 31 percent the next three years); and 1997 (S&P 500 up 27 percent and 20 percent the next two years). Similarly, strategists were most bullish about stocks in 2000 and 2001, at the beginning of the worst bear market since the early 1970s. Now, with the S&P 500 down about 25 percent from its 2000 peak, Wall Street strategists have gotten more cautious—which, ironically, is good news. Several months ago, the strategists’ optimism waned enough that the sell-side indicator was finally predicting a positive performance for the following year.

Commentators love to use measures like the sell-side indicator to suggest that Wall Street professionals are a bunch of idiots. The real takeaways, however, are far more profound. First, Wall Street strategists are the furthest thing from idiots; on the contrary, they are smart, well-educated, well-informed, and well-trained, and they still often get it wrong (so you can imagine the result for most people, who, in comparison, are inexperienced, untrained, and/or uninformed). Second, as described in this previous piece, almost all of us, professionals and amateurs alike, form expectations about the future by assessing what has happened in the recent past, by looking in the “rearview mirror.” Consequently, we are most bullish when we should be most bearish (at the end of long bull markets) and most bearish when we should be most bullish (at the end of long bear markets). Right now, we’re still close to the end of a long bull market.

The third profound point about the sell-side indicator is that it is Richard Bernstein’s best predictor of one-year market performance—and it is far from perfect. The indicator was wrong last year, for example, and wrong in the last years of the bull market. The lesson of this, unfortunately, is that predicting the market’s movements over one- to three-year time frames is difficult—so difficult that, no matter what tools they use, only a small percentage of investors can do it with an accuracy that exceeds that of luck. And the lesson of this, unfortunately, is that the vast majority of the effort devoted to trying to predict what the market is going to do is a waste of time.

There is a silver lining, however. Unlike most professional investors and strategists, whose careers depend on short-term performance—weeks, months, quarters, and years—you have the luxury of being able to invest for the true long term (decades). And over the true long term, the market is more predictable.

Next week: The (Vast and Underappreciated) Role of Luck in Investing.