Getty Images/iStockphoto
 
 
It would be easy to hang our heads in shame as we approach the 10th anniversary of the 2007 bull market top.
 
Most of us failed to recognize that the top had even occurred until well after the fact, missing the early warning signs that, in retrospect, seem so obvious. From the Oct. 9, 2007, market top to the March 9, 2009, bottom, the Standard & Poor’s 500 index went on to lose 57%.
 
Consider what was happening 10 years ago this summer: On July 11, 2007, the Federal Reserve placed 612 mortgage-backed securities on its credit watch. On Aug. 6 of that year, American Home Mortgage—then the 10th-largest mortgage lender in the U.S.—filed for Chapter 11 bankruptcy protection. Small-cap stocks, as measured by the Russell 2000 index, hit their bull market high on July 13—and over the next five weeks dropped more than 12%.
 
How could we have missed these signs?
 
No doubt we won’t miss those particular ones if they were to occur again. But, needless to say, they won’t. Whenever the current bull market comes to an end, it inevitably will have been preceded by an entirely different set of warning signs.
 
And since bull market tops are characterized by widespread optimism, most of us will ignore them.
 
There is at least a ray of hope, however, if you’re willing to resist the widespread euphoria that accompanies tops. According to a number of market analysts I interviewed for this column, most major market tops do share certain telltale characteristics, even if their presence doesn’t guarantee that a bear market is imminent.
 
By focusing on them, it’s possible to avoid being taken completely by surprise by the next bear market.
 
Here are three such characteristics:
 
Overvaluation. Valuation indicators such as price/earnings, price-to-book, price-to-sales, or price-to-dividends ratios are notoriously poor at identifying market tops. Overvalued markets can remain that way for quite some time, if not become even more overvalued, before the laws of gravity set in. But we ignore valuation indicators at our peril, since it’s also true that almost all bull market tops in history have begun when they signal that the market has become overvalued.
 
In the summer of 2007, the stock market—as measured by these traditional valuation indicators—was more overvalued than it had been at almost all previous peaks since 1900. In fact, regardless of the valuation indicator chosen, there had been only one previous market top when the stock market had been more overvalued: the dot-com bubble in 2000.
 
A struggling financial sector. One of the S&P 500’s 10 sectors that typically suffers when a bull market is approaching its end is financial stocks. During the last three months of all post-1970 bull market tops prior to 2007, the sector lagged the S&P 500 two-thirds of the time, according to Ned Davis Research. And it did so again in 2007, lagging the S&P 500 by 3.1% over the three months prior to the Oct. 9 bull market top.
 
An even stronger early warning signal came from smaller regional banks, according to Hayes Martin, president of Market Extremes, an investment consulting firm that focuses on major market turning points.
 
Martin’s research has shown him that such banks are even better “canaries” than the financial sector generally. He says that regional banks peaked in December 2006, and by July 2007 they were already significantly behind the broad market—providing “early warnings of trouble ahead.”
 
These warning signs from the financial sector became progressively worse over the next few months, and by October they were screaming “sell.”
 
Wide divergences. Another indication of an unhealthy market is one in which fewer and fewer stocks are participating in the uptrend—divergences, in market-timing parlance.
 
Many market-timing systems focus on divergences in their attempts to pinpoint market tops. The Dow Theory—the oldest market-timing system that remains in widespread use today—uses divergences between the Dow industrials and Dow transports as the precondition of a sell signal. The Advance/Decline Line, a market-timing system that measures the number of rising stocks net of declining ones, points to imminent trouble if it shows the majority of stocks to be declining even as the broad averages continue rising.
 
David Aronson has been researching increasingly sensitive ways of measuring the market’s internal divergences. Aronson, an adjunct finance professor at Baruch College and the author (with Timothy Masters) of Statistically Sound Machine Learning for Algorithmic Trading of Financial Instruments, proposes a particularly systematic measure of divergence that focuses on each stock’s recent percentage change. It’s a sign of an unhealthy market, he says, when those changes fall within a wide distribution.
 
This was very much the case in July 2007, Aronson told Barron’s. Prior to that month, the stock market had exhibited more-extreme divergences less than 3% of the time. By October 2007 his measure of divergence reached even higher levels, registering readings that had been exceeded only a couple of times previously.
 
The bottom line: Though there will never be a foolproof system for pinpointing major market tops, it would be incorrect to say that those tops occur randomly and that therefore we should simply give up. The 2007 bull market top did share certain crucial characteristics with prior bull market tops.
 
So where does the current stock market stand relative to 10 years ago? Fortunately, just one of the three telltale characteristics of a top appears to be present today: overvaluation. Stocks, on average, appear to be even more overvalued today than they were at the 2007 top, according to any of the standard valuation ratios.
 
The financial sector is not struggling currently. In fact, it’s the third-strongest performer among the 10 S&P 500 sectors over the past three months, beating the S&P 500 by a margin of 4.9% to 3.8%. And though Aronson’s divergence measure did rise to alarming levels earlier this summer, he says it has backed off recently. Martin adds that the market today is “in far better shape than it was 10 years ago.”