By Mark Hulbert, MarketWatch
Over the past 45 years, the stock market has lost more than 20% each time three warning signs flashed simultaneously.
After a selloff this past week dragged the Dow Jones Industrial Average into negative territory for the year, it’s worth noting that all three are flashing today.
The signals are excessive levels of bullish enthusiasm; significant overvaluation, based on measures like price/earnings ratios; and extreme divergences in the performances of different market sectors.
They have gone off in unison six times since 1970, according to Hayes Martin, president of Market Extremes, an investment consulting firm in New York whose research focus is major market turning points.
Bear in the air
The S&P 500’s average subsequent decline on those earlier occasions was 38%, with the smallest drop at 22%. A bear market is considered a selloff of at least 20%, with bull markets defined as rallies of at least 20%.
In fact, no bear market has occurred without these three signs flashing at the same time. Once they do, the average length of time to the beginning of a decline is about one month, according to Martin.
The first two of these three market indicators — an overabundance of bulls and overvaluation of stocks — have been present for several months. Back in December, for example, the percentage of advisers who described themselves as bullish rose above 60%, a level Investors Intelligence, an investment service, considers “danger territory.” Its latest reading, as of Wednesday, was 56%.
Also beginning late last year, the price/earnings ratio for the Russell 2000 index of smaller-cap stocks, after excluding negative earnings, rose to its highest level since the benchmark was created in 1984 — higher even than at the October 2007 bull-market high or the March 2000 top of the Internet bubble.
Three strikes and you’re out
The third of Martin’s trio of bearish omens emerged just recently, which is why in late July he advised clients to sell stocks and hold cash. That’s when the fraction of stocks participating in the bull market, which already had been slipping, declined markedly.
One measure of this waning participation is the percentage of stocks trading above an average of their prices over the previous four weeks. Among stocks listed on the New York Stock Exchange, this proportion fell from 82% at the beginning of July to just 50% on the day the S&P 500 hit its all-time high.
It was one of “the sharpest breakdowns in market breadth that I’ve ever seen in so short a period of time,” Martin says.
Another sign of diverging market sectors: When the S&P 500 hit its closing high on July 24, it was ahead 1.4% for the month, in contrast to a 3.1% decline for the Russell 2000.
Expect up to a 20% S&P 500 decline
How big of a decline is likely? Martin’s best guess is a loss of between 13% and 20% for the S&P 500, less than the 38% average decline following past occasions when his triad of unfavorable indicators was present. The reason? He expects the Federal Reserve to quickly “step in to provide extreme liquidity to blunt the decline.”
To be sure, Martin focuses on a small sample, which makes it difficult to draw robust statistical conclusions. But David Aronson, a former finance professor at Baruch College in New York who now runs a websitethat makes complex statistical tests available to investors, says that this limitation is unavoidable when focusing on past market tops, since “by definition it will involve a small sample.”
He says that he has closely analyzed Martin’s research and takes his forecast of a market drop “very seriously.”
Martin says that expanding his sample isn’t possible because most of his current indicators didn’t exist before the 1970s and “the comparative math gets very unreliable.” But he says he does use several statistical techniques for dealing with small samples that increase his confidence in the conclusions that his research draws.
Russell 2000 could take 30% hit
He says stocks with smaller market capitalizations will be the hardest hit in the decline he is anticipating, in part because they currently are so overvalued. He forecasts that the Russell 2000 will fall by as much as 30%.
Also among the hardest-hit stocks during a decline will be those with the highest “betas” — that is, those with the most pronounced historical tendencies to rise or fall by more than the overall market. Martin singles out semiconductors in particular — and technology stocks generally — as high-beta sectors.
He predicts that blue-chip stocks, particularly those that pay a large dividend, will lose the least in any decline. One exchange-traded fund that invests in such stocks is iShares Select Dividend, which charges annual expenses of 0.40%, or $40 per $10,000 invested.
The average dividend yield of the stocks the fund owns is 3%; that yield is calculated by dividing a company’s annual dividend by its stock price. Though the fund’s yield is higher than the S&P 500’s 2% yield, it nevertheless pursues a defensive strategy. It invests in the highest-dividend-paying blue-chip stocks only after excluding firms whose five-year dividend growth rate is negative, those whose dividends as a percentage of earnings per share exceed 60% and those whose average daily trading volume is less than 200,000 shares.
The consumer-staples sector has also held up relatively well during past declines. The Consumer Staples Select Sector SPDR ETF currently has a dividend yield of 2.5% and an annual expense ratio of 0.16%.
If the broad market’s loss is in the 13%-to-20% range that Martin anticipates, and you have a large amount of unrealized capital gains in your taxable portfolio, you could lose in taxes what you gain by selling to sidestep the decline. But the larger losses he anticipates for smaller-cap stocks could be big enough to justify selling and paying the taxes on your gains.