Timeout for the Market
02/03/2010 12:00 pm EST
Jon Markman of MSN Money explains why the market may go nowhere for the next few months—and that’s a good thing.
Timeout. The pause that refreshes. Beauty sleep.
These are not terms normally associated with the stock market, but you might think differently in ten months. Because the next stretch of activity for US stocks is much more likely to resemble a restless snooze plagued by bad dreams than it is the easygoing stroll we've enjoyed since March of last year.
Part of this is fundamental:
- We're coming up on midterm Congressional elections, which are notorious for making investors queasy and putting markets on hold.
- Stubborn unemployment is confounding optimists, perplexing policy makers, and impeding consumer spending.
- The Federal Reserve [may] soon approve a small but symbolically important short-term rate hike.
- Perhaps even more importantly, rate hikes appear to be on the way from the central bank of China, which is the world's lender of last resort these days.
But there is a far more compelling statistical reason to forecast weakness, and that is mean reversion. My studies of the US market over the past 100 years show that long periods of expansion are typically followed by even longer periods of mean reversion, in which companies' valuations catch up with their share prices.
Statistically speaking, since 1950, whenever the Standard & Poor's 500 index has moved more than 10% over its ten-month average for more than seven months (as it did through mid-January), it has next slipped sideways for at least the next ten to 18 months within a range of flat to down 10%.
The most recent example of this occurred six years ago, when the 2003 bull market hit a brick wall in spring 2004 and proceeded to trade roughly sideways for the next ten months.
You don't need to be bearish to expect such a timeout now, because it just shows that you are expecting a return to normalcy.
During the 2003-07 bull market, the S&P index typically stayed 4.6% above its ten-month average, which was unusually buoyant. Until [recently], the benchmark index was ahead of its trailing-ten-month average by more than twice that amount, at 11.9%.
The only previous time in the past half-century that the index maintained a double-digit percentage lead above its ten-month average—October 1982 to June 1983—the index never crashed—the worst slip was 10% from that peak—but just slipped sideways with a downward tilt as valuations caught up with expectations.
This can happen again and probably will—not because anything is wrong with the economy but because mean reversion is just what markets do.
[Trader Tom DeMark told me recently] that his TD Absolute Retracement model suggested that a top for the 2009 rally might have been reached January 14th when the [Dow Jones Industrial Average] was at 10,722.30.
Considering that he used the model to call the September 2001 and October 2002 lows, as well as the October 2007 high, that's a call to which we must pay attention.