March 22, 2006
WILL THE COMMODITIES' DIVERSIFICATION BENEFIT SURVIVE?
Commodities have been discovered by the masses in recent years. Rediscovered is probably the more accurate description, since raw materials were formerly all the rage in investing circles in the 1970s and through the first half of the 1980s. The fact that Wall Street and Main Street are finding reason anew to embrace a broad mix of commodities is arguably a more-strategic move this time around, courtesy of the widely publicized data that show the diversification power of adding commodities to a portfolio of stocks and bonds.
But some worry that the diversification value of commodities may fade over time as more and more investors climb on the bandwagon. One way to monitor that diversification value is by watching the correlations of returns between commodities and other asset classes, primarily stocks and bonds. On that front, there's reason to wonder if the price of popularity is starting to take a toll.
Consider that the rolling 36-month correlations between stocks and commodities have been rising sharply for more than a year, as the chart below illustrates. The correlations between commodities and bonds have been rising recently too, albeit with less drama.
Granted, even after the jump in correlations between stocks and commodities there's still a hefty diversification advantage left. Indeed, the trailing 36-month correlation of total returns as of last month between the Russell 3000 and Dow Jones-AIG Commodity Index was a mere 0.20. (1.0 is perfect correlation, 0.0 is no correlation, and -1.0 is perfect negative correlation.)
By comparison, there's a lot correlation these days between U.S. stocks and foreign stocks, as measured by the Russell 3000 vs. MSCI EAFE. The two equity indices post a 0.82 correlation for the past 36 months as of February 2006. The two indices, in other words, are generally moving in line with one another.
Simply put, commodities still offer a strong diversification counterweight to stocks and bonds. But if the stocks/commodities and bonds/commodities combos still yield diversification benefits, one might reasonable ask: how long will the strategic value last? Not much longer if the rise in correlations between the two asset classes maintains its current pace.
Then again, correlations aren't set in stone. They evolve, ebbing and flowing over time. Bull markets in one, a bear market in another, keep the relationship dynamic and fluid. It may well be that the recent rise in correlations between stocks and commodities is just part of the natural process of ups and downs in correlations.
We'd be inclined to leave it at that, except for one thing: money's pouring into commodities at a robust clip. What's more, a lot of that new money is chasing commodities specifically to tap into the long history of low correlations.
The oldest mutual fund targeting commodities, Oppenheimer Real Asset, has an asset base of $1.9 billion since launching nine years ago. Several other commodities mutual funds have come on the scene since then. Notably, the mutual fund Pimco Real Return, has accumulated $15 billion--up by $3 billion last year alone, according to Morningstar. And the choices keep expanding: as of February, the first commodities ETF in the U.S. debuted, the DB Commodity Index Tracking Fund. Meanwhile, there are additional commodities ETFs are in registration with the Securities and Exchange Commission. In addition, there are many more billions rolling into private commodities funds, courtesy of institutional investors.
The recent inflow of money is no doubt driven, at least in part, by the strong returns in commodities. Hot asset classes always attract money, and commodities have clearly been hot. Consider the annualized five-year total returns through February 28, 2006 for commodities, stocks and bonds:
Commodities (Dow Jones-AIG Commodity Index) 10.31%
Bonds (Lehman Bros. Aggregate Bond) 5.16%
Stocks (Russell 3000) 3.28%
If commodities stumble, perhaps the correlations will fall. Or, maybe a selloff in stocks or bonds will bring correlations lower.
It doesn't help to consider that foreign stocks once posted much lower correlations with domestic stocks. Indeed, in the 1980s and early 1990s, there was a much greater bang for allocating money into foreign equities of developed markets. Today, after so many investors went overseas, the diversification value of foreign stocks has faded. Not completely, but it's faded relative to what it once was. Is the same process now attacking commodities?
The future, as always, is unclear. The past, by contrast, yields no secrets. Bridging the gap between the two is always the great chasm in investing. Correlations threaten no less of a challenge.
Posted by jp at March 22, 2006 9:40 AM
I agree. It is getting harder to find any investments that are negatively correlated to the markets. Smart manager want to diversify away from what everyone else is doing, but then everyone else wants to join in.
Whatever the masses invest in, that is the investment that will lose its luster. With modern communication, the masses find out faster the latest thing, and thus it is harder and harder to stay ahead of the pack.
Posted by: Luke at September 22, 2007 11:25 AM
I used to do lots of these types of analyses. They have their uses, but I am leery: the correlations most of us use refer only to LINEAR relationships. So things can correlate/not correlate for a variety of reasons. Perhaps because of the reasons that the able author cites. Or perhaps because the linearity of one or the other series has changed... There are no rights or wrongs in the thinking game, just a series of views - correlated or not!
Posted by: Enzio von Pfeil at March 23, 2006 4:09 AM