***Stock Forecasts? Election Cycles Deserve a Look***
BY JAMES K. GLASSMAN
Washington - What will 1999 bring in the stock market? A new Reuters survey finds that analysts believe the Dow Jones industrial average will rise by an average of about 500 points, or 6 percent.
Sounds reasonable but, of course, it's utter nonsense. No one has the slightest idea what stocks will do next year, and smart investors who take the long view shouldn't care.
But we do care. It's only human. We turn to experts to tell us the future, even though they have not the foggiest idea. It's a practice that goes back thousands of years.
Daniel Boorstin writes in "The Seekers," his new book about philosophers, "The saga of the prophets records efforts to cease being the victim of the gods' whims by deciphering divine intentions in advance. . . . Diviners watched smoke curling up from burning incense, they . . . answered questions about the future by pouring oil into a bowl of water held on their lap and noting its movement on the surface or toward the rim."
Oil in a bowl of water, huh? Sounds strange, but it probably works as well as the techniques of financial analysts, whose record seems worse than the laws of chance would allow.
Just one example: In his book, "The New Contrarian Investment Strategy," David Dreman examined 52 surveys over 50 years in which experts in publications such as the Financial Analysts Journal picked their top stocks or industries for the coming 12 months. In 77 percent of the cases, Dreman found, the experts' selections actually underperformed the market as a whole.
Still, there is one predictor of stock market performance that, while crude, has been uncannily accurate since at least 1832. It's the election-year cycle, and it indicates that the market is almost certain to rise, and rise significantly, in 1999 - no matter what happens to Bill Clinton, Saddam Hussein or the Japanese banking system.
Why? Well, 1999 is the year before a presidential election, and data since the administration of Andrew Jackson show that the market loves pre-election years. For three decades, Yale Hirsch, who publishes the "Stock Trader's Almanac," has been following the long- -term effects of what economists call "exogenous," or outside, factors on the stock market, and he has discovered that the election cycle is particularly powerful.
Since 1832, in pre-election years the Dow (and its 19th century predecessors) gained an average of 10 percent annually, not including dividends; in election years, the Dow gained an average of 7 percent, also a good advance, especially when you add dividend yields averaging almost 5 percent.
But in the first year after a presidential election - that is, a post-election year (such as 2001 will be) - the Dow gained an average of only 2 percent; in mid-term election years (like 1998), the gain has averaged just 4 percent.
Adding up the net advances in each of the cycle years, Hirsch reports in the 1999 edition of the Almanac that "the last two years of the 42 administrations since 1832 produced a total net market gain of 703.2 percent, dwarfing the 235.7 percent gain of the first two years of these administrations."
In all, the Dow (again, not counting dividends) has gained ground in three- -quarters of pre-election years but has lost points in more than half of post-election years. Since the election of Lyndon Johnson in 1964, the Standard & Poor's 500-Stock index, a broader measure than the Dow, has produced 17 percent returns in pre-election years, 14 percent in election years, 5 percent in post-election years, and barely 1 percent in mid-term years.
Since 1940, Hirsch's research shows, the Dow has never fallen in a pre-election year like the one coming up. It has scored double-digit gains in 10 of the 14 years. By contrast, the Dow has fallen in seven post-election years since 1940, notching double-digit losses four times.
What if Bill Clinton is removed as president? Not a problem, says Hirsch, who told me simply that "the biggest bull market of all time would continue." After all, 1999 would still be a pre-election year; there would simply be a different incumbent.
Also, look at the last time such an event occurred. In 1974, a mid-term year in which the market was down 32 percent (following 1973, a post-election year, in which it fell 14 percent), Gerald Ford became president with the resignation of Richard Nixon. But in 1975, a pre-election year, the S&P shot up 32 percent, its best performance until 1995, which was also (you guessed it) a pre-election year.
Or consider 1963, the year Johnson acceded to the White House after the assassination of John F. Kennedy. In that pre-election year, the S&P rose 19 percent - following a decline of 12 percent in 1962, a dreaded mid-term year.
Why does the presidential cycle work so well as a predictor of the market? Hirsch believes that, facing presidential elections, politicians do the best they can to boost the economy. Farther from elections, without voters to discipline them, they often take risks and get the nation into trouble.
"Presidents want to be re-elected," says Bill Staton, a Charlotte, N.C., financial analyst who is another longtime follower of the cycle. "They tend to do negative things to the economy in the first two years of a term, then turn up the juice for the following two. It's the equivalent of Popeye eating a can of spinach."
Hirsch writes, "Wars, recessions and bear markets tend to start or occur in the first half of the (presidential) term; prosperous times and bull markets, in the latter half."
Before you dismiss these explanations, think about all the terrible things that have happened in post-election years: The start of the Civil War, 1861; U.S. entry into World War I, 1917; Wall Street Crash, 1929; U.S. entry into World War II, 1941; Bay of Pigs, 1961; start of the Vietnam War, 1965.
Of course, the pattern is not perfect. In recent years, the stock market has been soaring, with not much regard to presidential cycles.
Yes, the only two losing years for the S&P in the 1990s were mid-term years, and the best year, 1995, was a pre-election year. But in 1997, a post-election year, the S&P registered a gain of 31 percent, and in 1998, a mid-term year, it is up about 20 percent so far.
Another strong pattern - this one with no discernible explanation - is the decennial cycle, originally identified by Edgar Lawrence Smith in his book, "Common Stocks and Business Cycles."
Since the 1880s, years ending in "5" have a perfect record - never a single decline in the S&P. Years ending in "2," "8," and "9" are second-best with only two declines each. The worst years by far are those ending in "0," during which the S&P has dropped eight out of 11 times.
My guess is decennial cycles are merely random events - and, truth to tell, presidential cycles probably are, too, but they can serve an important purpose.
The quadrennial pattern absolutely should not affect your investment strategy, which should simply be to find good businesses at decent prices and hold them for a long time. But as we approach a scary year, the good news that history supplies about pre-election markets may give you the courage to stay the course.
It is discipline, more than good stock-picking, that separates successful investors from mediocre ones. |
