Don’t count on a soft landing for stocks by Shawn Tully @FortuneMagazine October 8, 2014, 4:41 PM EDT E-mail Tweet Facebook Google Plus Linkedin Share icons The recent stock market selloff has investors worried that equities are overpriced, maybe extremely overpriced.If that’s the case—and all measures suggest that it is—what trajectory is the eventual selloff likely to take? Forget about a soft-landing scenario.On Wednesday, the Dow Jones Industrial Average jumped by 274 points. But bear in mind, stocks usually don’t crash in a day, or a month. Nevertheless, the damage comes fast. When stocks are this pricey, history tells us that prices swirl downwards in a continuous whirlpool that usually lasts about a year and goes to extremes, driving valuations well below rational levels. First everyone believes equities are a great deal, then no one believes. Eventually, the panic makes stocks a bargain once again.But don’t count on a benign, downward drift of 10% or 15%, just enough to bring the kind of “healthy correction” and “new buying opportunity” that equity strategists are always touting. When you start from these levels, the fall is more often brutal than soft.Since 1876, U.S. stocks have experienced eight contractions that hammered prices from between 30% to 60%, in round numbers. In the post-World War II era, where we have complete data, the crashes tended to follow similar patterns: Equities peaked with extremely high PEs, as measured by the Shiller Cyclically Adjusted Price-Earnings ratio (or CAPE), which adjusts for big swings in earnings that can greatly distort PEs based on today’s highly erratic profits. Dividend yields also tend to be far below average when selloffs commence. That’s a good description of where the market stands right now.A review of these major crashes offers a guide on what we can expect. Generally, the downdrafts last around a year, and seldom continue for more than 18 months, with the glaring exception of the Great Depression.Using data beginning in 1871, the first big bust came in the mid-Victorian period. From mid-1876 to mid-1877, stock prices tumbled 35% in a 13-month stretch. It was another three decades before investors suffered steep losses again. From January 1906 to January 1907, the S&P lost 37.7%. Another crash lasted from November 1916 to November 1917, slashing prices by 31%.The Great Depression broke the tradition of yearlong busts. From September 1929 to June 1932—a period of 33 months—the S&P contracted by 85%. It’s worth noting that at the start of that long descent, the Shiller PE stood at 33, its highest level to that date and twice its previous average. The dividend yield was also unusually low for the era, at around 3%.Stocks rebounded strongly until March 1937, when the CAPE tripled from its lows to 22. Over the next 11 months, stocks cratered by 45.3%.From the end of World War II until the late 1960s, stocks enjoyed a magical run, interspersed with only minor corrections. Then, from May 1969 through May 1970, shares fell by 27.6%. The starting Shiller PE: an elevated 21, around 30% above the then-historic average of 16.The OPEC oil crisis brought the longest market crash since the Great Depression. From January 1973 to September 1974, a span of 21 months, the S&P fell by 42.5%. This time, it was the tripling of oil prices and its damage to the economy, rather than overpriced stocks—the CAPE started out at 19—that prompted the collapse.Surprisingly, the Black Friday dive of October 19, 1987, was more severe for being sudden than big. From September 1987 to April 1988, the S&P dropped by 19.7% and then rebounded quickly. The CAPE at the start was hardly signaling rough times ahead. It stood at around 18—above average, but hardly alarming. Nor was there cause for great concern, despite the short-lived carnage of Black Friday.In fact, Black Friday is an exception to the rule that big selloffs persist for around a year. That’s hardly comforting. During those 12 to 18-month spans, stocks generally swoon straight downwards, often falling a few percentage points a month.The next crunch came in September 2000 with the bursting of the dot-com bubble. It was really two separate crashes. In September, the Shiller PE was flashing red at an incredible 42, near an all-time record. Over the next 11 months, stock prices shrank by 29%. That selloff wasn’t sufficient. The S&P snapped back until March 2002, sending the CAPE back over 30. A second selloff took shares down another 27.5% to the low point of February 2003.The pain is still fresh from the cataclysm that struck in October 2007. Stocks dropped, more or less in a straight line, until March 2009, a period of 17 months. When the din subsided, shareholders had lost 51% of their money. Once again, the Shiller PE was dangerously high at the start of the trouble, standing at 27.3. And the dividend yield was a puny 1.75%. Suddenly, investors decided they weren’t getting paid nearly enough for the danger of holding volatile shares that had just decimated many a nest egg.Right now, the CAPE of 26 tells us that stocks are pricey—unless you think equities aren’t particularly risky. It also doesn’t help that the dividend yield is far, far below its historic norm at 2%.We don’t know if a crash is imminent, or if investors are happy with low yields and the modest capital gains that, at best, they can expect at these prices. It could break either way. But if a crash is coming, it will probably mirror those of the past. Most if not all of the damage will come in around 12 months. Prices will pretty much go straight down. Don’t look for “buying opportunities” in the infrequent lulls. What’s encouraging is that the selloff will be overblown. Just look at the aftermath of 2007: By early 2009, the CAPE had dropped to 13. That was no soft landing. And that was the time to buy.