Jane Bryant Quinn
Everything you thought you knew about diversification is wrong. So is everything you thought you knew about portfolio risk. It took the meltdown of September and October 2008 to flesh out the story, but the plot outlines have been in place since the 2000 tech-stock bust.
The measure of stock-market risk typically focuses on your holding period. The longer you hold a portfolio of diversified stocks, the less likely your chance of losing money over the entire term.
The classic example is the record of Standard & Poor's 500-stock index. Over rolling monthly five-year periods since 1926, it lost money 14 percent of the time, with dividends reinvested, according to Ibbotson Associates in Chicago, which publishes market data. Over 10-year periods, it lost money 4 percent of the time. Over 15-year periods, there have been no losses. (The most recent 15 years delivered a 6.5 percent annual gain.)
With data like this, stocks for the long run, plus reinvested dividends, still look like a safe bet.
But it's a mistake to focus only on your investment's theoretical end date, says money manager Mark Kritzman, chief executive of Windham Capital Management in Cambridge, Mass. You're assuming that you will be able to withstand market crashes before then, which may not be the case. On the way to its (presumed) positive long-term result, your portfolio might suffer several declines of 20 percent or more.
The longer you hold, the more likely it is that such a fall will occur, Kritzman says. Seen that way, your stock market investment grows riskier, not safer, with time. One of those retreats might drive you out of stocks entirely, or force you to lower your stock allocation to protect the money that remains.
How do you hedge against that risk? That's where diversification comes in. But the usual advice on how to diversify hasn't protected you very well.
Effective diversification depends on correlation -- a statistical measure of how two assets move in relation to each other.
Assets that produce similar returns are highly correlated. An example would be large growth stocks and an S&P 500 index fund. You get some diversification when you own them both but not a lot.
Assets with low correlations move in the same direction, up or down, but at different rates. If one asset falls a lot in price, the other might fall just a little. Assets with negative correlations move differently. When one of them produces returns that are above average, the other's returns tend to fall below average.
The best way to minimize your risk of investment loss is to own assets with low or negative correlations. The problem is that correlations change. The standard measures may be effective over full market cycles but not for the occasional, turbulent markets that come along.
As an example, take large U.S. stocks versus emerging-market stocks. The usual measures show them to have a low correlation. Diversifiers should reduce your risk and increase returns. If you hold for 15 years, they probably do.