Barrons

Heeding History

How to spot a market slide before it occurs

By LAUREN R. RUBLIN

For most of us, crummy weather can kill a summer weekend.

But Richard Arms, the well-known market technician, had a recent weekend ruined by his comparative study of the three charts displayed below. They show how the famous bull markets of the 1920s and 1980s both peaked in the summer, in 1929 and 1987, to be exact, and how those bull runs ended shortly thereafter in infamous crashes. And that's why the bottom chart, depicting the Dow's majestic rise since 1994, gives Arms the chills.

The charts bear shocking similarities, "not just in the major moves but in the minor blips," Arms told us last week from his perch in New Mexico. At the least, they suggest, he says, that Mark Twain was onto something when he observed that while history may not repeat itself, it does rhyme.

"Essentially the market doubled in each time period, going from about 150 to 350 in the 1920s, 1200 to 2800 in the 1980s, and 4000 to 8000 in the past two years," Arms points out.

Besides tracing the Dow's movements, Arms charted weekly volume for each of the three bull markets, observing in all cases that light volume in the early stages grew progressively heavier as the market climbed. "If history repeats, we are within weeks of a major market top," he concludes. "The charts are implying that the party is very nearly over."

Ever the technician, Arms will not venture a guess as to why, or how, today's bull will meet his sad demise. But Arms is quick to point out that the aging beast himself will offer numerous clues that the end is nigh. First and foremost, look for day upon day of "extremely heavy" volume, accompanied by "lack of progress" to the upside. "The market just runs into a wall," he says.

Volume on the New York Stock Exchange indeed has been rising in the past few years, as it did in the 1920s and in the years preceding the "87 top. "A few years ago we were trading 300 million shares a day," Arms observes. "In June, average daily volume was 520 million shares, compared with 482 million shares in May, and 476 million shares in April. Before it's over, I would expect to see unprecedented volume, with perhaps 700 million-800 million share days." Last week, Big Board volume exceeded 600 million shares in several sessions.

Another telltale sign of a market top is a series of "preliminary" sharp drops and rebounds, Arms says. Wrenching moves haven't become commonplace-at least, not yet-although the Dow's near 200-point decline June 23, and its 120-point drop in July's first full week, conceivably could be early warnings. By the same token, the NASDAQ's explosive rally in recent days also hints of potential trouble. In Arms's view, it suggests a pickup in speculative fever: "Investors are moving away from the tried-and-true Dow stocks, and into things from which they might get better performance. But they are taking more risks to do so."

To be sure, Arms is not the first to notice the eerie parallels between the 1920s, 1980s and 1990s. Folks who brush off the implications can take comfort from the fact that the economic backdrop today is markedly different, particularly in comparison with the 1985-87 period. A decade ago, the economy was in overdrive; inflation and interest rates were rising, and the dollar was falling. Today's economic expansion is a much more balanced affair, with inflation nearly mute. "I get the fundamentals quoted to me all the time, but I really don't look at them," says Arms. "Everything that is known about the market is in the price of the stocks. When the fundamentals change, the market already will be reacting to them."

Arms, who has been cautious about stocks for a while, only recently turned more bearish. Of his little chart experiment, he says, "Human emotions remain the same, and the market gets pushed up in the same rhythmic manner."

If Richard Arms, and perhaps Mark Twain, are right, this market could topple in the same way, too.


Questions:

  1. This paper, like the earlier one by Sobel, looks at historical parallels between the 1920s and the 1990s. It bases its analysis on technical indicators instead of the qualitative variables suggested by Sobel. What are the technical indicators used to make this analysis?
  2. One of the indicators used by Arms is trading volume. Why might volume convey information about future market directions? What are some of the concerns you would have about comparing trading volume in earlier markets with this one?