|
Page 1 of 3
 Since peaking in early, July commodity prices have plunged by more than 25% as measured by the Dow Jones-AIG Commodity Index. The main catalyst for the decline was a sharp fall in the oil price following weaker-than-expected U.S. and European economic data as well as continued indications of weakness in global auto sales and gasoline consumption. The price corrections have been indiscriminate, with prices of nearly all commodities ranging from natural gas to cocoa falling sharply regardless of underlying fundamentals. The main question on investors' minds now is whether the fall in commodities prices represents the end of the bull market that began in 2002, or is just a temporary correction in a bull market that still has many years to go.
Anyone who has observed financial markets for more than one business cycle recognizes that markets rarely maintain a straight-line trajectory for long (up or down). Often, the more sustained and substantial the move, the more violent the reversal. Two clear hints that perhaps markets were moving too far too fast on the upside earlier this year (and likely are now doing the same on the downside) were the sharp rise in correlations within the commodity sector and the sharp rise in the (negative) correlation of commodity prices with the U.S. dollar. When assets that previously had low correlations see those correlations rise sharply in a short period of time, it is often a sign that there is some speculative froth in the market, and short-term overshooting may be occurring.
Rising Correlations Highlight Herd Behavior
As shown in Figure 2, the correlations of most commodities to energy prices have soared since the correction began in July. For example, agricultural prices—which are normally relatively immune to business cycle fluctuations—have seen their correlation to energy prices double compared with the first half of 2008 and more than triple since the first half of 2007. This is one quantitative indication that investor activity was playing a greater role than before in price formation in the first half of this year and that a very large part of the recent sell-off has been a result of herd behavior by financial players.
Separately, it is interesting to note that the correlation between oil and precious metals rose sharply in the first half of 2008 compared with 2007, likely reflecting increasing flows into commodities as an asset class, as well as also attempts by investors to find natural U.S. dollar and inflation hedges. Since July, however, the correlation has fallen quite abruptly as the oil price has declined and the gold price has outperformed, reflecting concern that economically sensitive commodities such as oil and industrial metals will be in less demand; gold has been relatively resilient as investors have held onto safe-haven assets. Concerns about demand growth would also explain why the correlation between energy and industrial metals has risen so sharply since the correction began.
Another indication that more than purely fundamental factors have been affecting commodity prices in recent months has been the sharp rise in the negative correlation between energy prices—and therefore the rest of the commodity complex—and the U.S. dollar. There are some fundamental reasons for a negative correlation between the two during some periods (a weaker dollar makes oil cheaper in other currencies; a higher oil price can hurt U.S. growth and rate expectations, and therefore the dollar and vice versa, etc). However, as Figure 3 illustrates, since the middle of this year, the negative correlation has risen to much higher than normal levels.
|
