The death of asset allocation, and its replacement with another potentially bad idea

Writes Tom Lauricella in a fascinating article in today's WSJ:

Asset allocation, a bedrock of investing for decades, appeared to fail miserably in 2008. The conviction shared by most investors -- that they should spread their money across myriad asset classes to minimize losses -- was shaken as nearly all markets tumbled in unison.

The main reason asset allocation failed is that it had become so successful. That is, everybody was spreading their money across the same myriad asset classes. So when the market panic came, it came everywhere.

"When people start buying an asset, the act of them diversifying ultimately makes the asset less of a diversifier," says Pimco's Mr. Bhansali.

Asset allocation was a product of the great quantification of finance that came (mostly) out of the University of Chicago in the 1950s and 1960s (I'm told there's a book available on the subject). It was often promoted hand-in-hand with the efficient market hypothesis, which held that everything you needed to know about an asset's value was already reflected in its price. A key (and common-sense) plank of the efficient market platform is that any easy-to-spot pattern or imperfection in pricing would be quickly corrected by the mere fact that lots of investors/speculators would rush to take advantage of it. But for some reason the asset allocators never seemed to think this truth would apply to them. They found a free lunch of sorts in the strategy of spreading money among uncorrelated asset classes, and seem surprised that following their advice eventually made the free lunch go away.

The most intriguing (and disturbing) element of Lauricella's story is the lesson some investors and consultants are drawing from it. He cites the chief economist at Ibbotson Associates saying that the asset classes that have done best when stock markets fall are intermediate-term government bonds, gold and—most of all—the inflation-linked government bonds known as TIPs. So Ibbotson is advising its clients to put more money into U.S. government bonds. (As is TIPophile Zvi Bodie, as I wrote a few weeks ago in TIME.)

Well what could go wrong with that? Something has to eventually, right? The simple, Peter Schiff explanation is that the U.S. is headed for a fiscal collapse in which the country will default on its debt or at the very least pseudodefault by allowing inflation to run rampant. Of course, if demand keeps growing for TIPs and they come to make up a larger share of the country's debt, then we won't be able to inflate our way out of trouble—because the debt will keep rising with inflation.

Less apocalyptically, high demand for Treasuries could simply lead to decades of returns so low as to outweigh any advantage gained by not being correlated with the stock market in a crisis. Which is sort of how things have worked with gold over the last half century.