The Myth of Stock-Market Tops – Wall Street Journal (subscription)

This post was originally published on this site



Illustration:

Andrew Bannecker for The Wall Street Journal

Many investors are trying to pinpoint the exact day on which the bull market eventually reaches its top. My advice to them: Stop.

I say that not because it is difficult to predict when the stock market changes direction—though that is true, too. But there’s an additional reason: Stock-market tops typically are a gradual rolling-over by the market rather than a sharp trend reversal. Some market averages will hit their peak months before or after others, and individual sectors can behave differently than the market as a whole.

In other words, the top doesn’t occur on just one single day, or even in one week or one month. Even if you successfully predict the precise day, or week or month, on which a particular market average hits its top, you can still lose money—or leave a lot of it on the table—because other market averages and sectors likely aren’t at their peaks.

A gradual process

This is an important insight not just for market historians. With the stock market up more than 300% from its March 2009 lows, and with the economy in its ninth year of recovery from the latest recession, many investors worry that a bear market is overdue. Rather than anxiously trying to determine the exact day when to shift out of stocks, they should instead view market tops as a gradual process in which equity exposure is slowly and deliberately reduced over time.

Take the 2007 bull-market top. The most widely followed market benchmarks, such as the Dow Jones Industrial Average and the S&P 500 index, hit their highs on Oct. 9 of that year, and that is the date that most market timers record as the “top” of the 2002-07 bull market. Yet the Russell 2000 index, a widely used proxy for the small-cap sector, registered its top on July 13, three months prior. The Dow Jones Utility Average topped out even earlier, on May 21. And some sectors began their bear markets even longer before; the SPDR S&P Regional Banking ETF, for example, hit its bull-market high in December 2006.

What this means: Pinpointing that Oct. 9 top wouldn’t have been very helpful unless you were investing in the S&P 500. In the event you were investing in the average small-cap, utility or regional bank, among others, you would have lost money if you had waited until Oct. 9 to sell.

Perhaps the most spectacular example of a divergence at a market top, however, came when the internet bubble burst. Though the S&P 500 topped out on March 24, 2000, and didn’t hit bottom until October 2002, the average small-cap value stock actually rose during that bear market, according to data from University of Chicago professor

Eugene Fama

and Dartmouth professor Ken French. So small-cap value investors left a lot of money on the table by going to cash during the 2000-2002 bear market.

Divergences this stark and spread out aren’t unusual at market tops, according to

David Aronson,

a former finance professor at Baruch College and now president of Hood River Associates, a research firm that uses machine learning to enhance stock-market trading systems. “The process of topping out can take a really long period of time, evolving over a year or more,” he told me in an interview.


Take the end of the bull market in the early 1970s, for example. Though the S&P 500 didn’t hit its high until January 1973, major divergences “started to manifest as early as the spring of 1971,” nearly two years prior, he says.

In contrast, Mr. Aronson adds, “market bottoms tend to be sharper.” The bottom at the end of the 2007-09 bear market is a good case in point: Virtually all major averages hit their bear-market lows on March 9, 2009. Most individual sectors did so also, and those that didn’t mostly hit their bottoms only a few days before or after. As Mr. Aronson argues in a just-published book, “bottoms are easier to identify, in real time, than tops.”

The narrow focus

Instead of trying to pinpoint the day of a particular market’s average top, therefore, you might want to focus on the prospects for the individual stocks or mutual funds that you own. Bragging rights for calling the top of this or that market benchmark mean little if your stocks or funds start declining months before that top.

One helpful focus would be on internal market divergences between different sectors of the market. A healthy market is one in which most stocks are participating. As divergences emerge and become more pronounced, odds increase that the stocks you own may suffer even if the major market averages such as the Dow and S&P continue rising.

One of the flashiest market ‘tops’ was Nasdaq’s record close above 5000 on March 9, 2000.

One of the flashiest market ‘tops’ was Nasdaq’s record close above 5000 on March 9, 2000.


Photo:

Andrew Serban. Bloomberg News

By the same token, you need to be alert to the possibility that your individual stocks will continue to rise even if the major averages begin a major decline.

It’s worth noting in this regard that some potentially worrisome divergences have materialized in recent months, according to

Hayes Martin,

president of Market Extremes, an investment consulting firm that focuses on market turning points. In an email to me, he pointed out that there have been some significant divergences among groups that historically have been reliable “coal-mine canaries,” such as small banks, which peaked on March 1, and smaller-cap stocks, which have been “battered” since late July. This suggests to him that we are probably in a long-term topping process.

Mr. Aronson agrees, though he says it’s unclear from his work whether the coming top will precede a major bear market or something of more intermediate-term significance.

Regardless, these divergences suggest that risk is elevated. We may want to focus more on which stocks or funds we would want to sell to reduce our equity exposure rather than what we’d buy to increase it.

Mr. Hulbert is the founder of the Hulbert Financial Digest and a senior columnist for MarketWatch. He can be reached at reports@wsj.com.