Anyone who believes copper’s 25% rally over the past six months is bullish has a short memory. Just this past summer copper was smarting from a 25% plunge.
If copper were a reliable leading indicator, both the U.S. economy and stock market would be much lower than they were last summer. Instead, U.S. GDP rose at an inflation-adjusted annual rate of 6% in the third and fourth quarters of 2022. The S&P 500 now is more than 7% higher than at 2022’s midyear mark.
Though these are just two data points, they should sound alarm bells for those who believe “Dr. Copper” has a Ph.D. in economics. Shooting down that myth doesn’t mean the outlook for the U.S. economy and the stock market isn’t better than it was last summer. It may be. The point is that any improvement has nothing to do with the price of copper.
To test whether Dr. Copper’s failed diagnosis last summer is the exception or the rule, I measured whether the S&P 500 tends to move up or down in synch with copper. Such a test measures copper’s record as a coincident indicator. Specifically, I measured the correlation between the two assets’ trailing one-year returns, over rolling five-year windows. What I found is plotted as the red line in the accompanying chart.
Notice that the correlation has bounced around significantly over the past four decades. Beginning in the mid-1980s and continuing through the 1990s, the coefficient was significantly negative, which meant the S&P 500 tended to do better when copper’s price was falling, and vice versa. So far this century, in contrast, the coefficient has been mostly positive, with the Global Financial Crisis being a major exception. There is no overall consistent pattern, in other words.
I next measured copper’s ability as a leading indicator: Can it predict subsequent changes in the S&P 500? To measure this, I calculated the correlation between copper’s trailing one-year return and the S&P 500’s subsequent one-year return, over rolling five-year windows. The plot of this coefficient is the dark blue line in the chart above, and it bounces around just as much as the line representing copper as a coincident indicator, though without the upward trend over time.
Needless to say, the overall correlations between copper and the S&P 500 are not significant at the 95% confidence level that statisticians often use when determining if a pattern is genuine.
The broader takeaway from this discussion is the need to subject our hunches and beliefs to historical scrutiny. In fact, to raise serious questions about Dr. Copper’s reputation, you wouldn’t have had to go to the trouble I went through for this column of calculating correlation coefficients. Simply eyeballing a price chart of the two would have revealed that copper isn’t much help for timing the stock market. Yet few bother to do even that much.
The reason, I suspect, is that we tell stories in order to make sense of a world filled with statistical noise. The stock market’s gyrations are largely random, so we comfort ourselves by saying that we just need to follow the lead of a metal that “has a Ph.D. in economics.” Given our intense need to find meaning in the randomness, we only ask that these stories be superficially plausible.
But superficial plausibility is a low bar over which to jump. If we want to have any fighting chance of beating the market, we need to set our sights much, much higher.
Mark Hulbert is a regular contributor to MarketWatch. His Hulbert Ratings tracks investment newsletters that pay a flat fee to be audited. He can be reached at email@example.com
More: Think you know where interest rates are headed? Here’s some advice