Dr. Copper moved to China

Copper, which is widely regarded as a leading indicator for economic activity and therefore dubbed as the metal with a PhD in economics, is down 12.5% since the end of 2012. Also other base metals have lost in 2013 so far. All base metals reached their high of the year in mid-February and then moved downwards. Thus, the testimony of Fed Chairman Bernanke, where he indicated the FOMC might taper the bond buying program at one of the next few FOMC meetings, could hardly explain the poor performance of base metals. Institutional investors have sold base metals already before this testimony in May.

In our fair value models, the US dollar index is also one of the explanatory variables for the price development of base metals. As many of the variables are not stationary and also show a seasonal pattern, the year-over-year percentage change had been applied in the models. The strength of the US dollar index following the change of economic policy in Japan (Abenomics) could explain the development of base metal prices to some extent.

When the fair value models were developed, the S&P 500 Index was the stock market index with the best explanatory power among the major stock indices. At this time, China had already risen to the world’s top copper consuming country. However, the S&P 500 index can not explain the decline of base metal prices because the US stock market posted strong gains while base metals moved in opposite direction.

In many reports about the base metal markets, the gradual decline of Chinese GDP growth is often cited as a reason for weaker base metal prices. Therefore, we examined various Chinese macroeconomic variables whether they explain the development of the 3mth LME copper price better than they did before. In order to have enough observation also in the second subsample, the starting date for the second sample had been set to January 2009. However, in 2009 and also for most part of 2010, the development of Chinese leading economic variables and international leading indicators was very similar.

As the interest was only in the explanatory power of some macroeconomic variables for the development of the copper price, we estimated only some bivariate linear regression models instead of multivariate models.

When we developed our fair value models for the base metal prices, the OECD leading indicator for the total OECD explained the development of copper prices better than the OECD leading indicator for China. As shown in the table below, the price changes of copper reacted stronger (higher beta-coefficient) on changes of the leading indicator for the total OECD than on the one for China. Also the R-squared is higher for the OECD region in the estimation period ranging from January 1996 until December 2008.  However, this has reversed in the period following the financial crisis. While the impact of the Chinese leading indicator is now more than 2.6 times higher, the impact of the leading indicator for the OECD region has almost halved. Also striking is the difference in the R-squared. Now, the leading indicator for China could explain more than 60% of the variation of copper prices alone.

A similar development took place for the industrial production in China. As the yoy %-change of Chinese industrial production is very erratic, a 3mth moving average had been applied to smooth the development.

One would assume that Chinese copper imports would play an important role for the copper price. Many analysts had presented charts showing the LME copper price being highly correlated with Chinese copper imports. However, those analysts manipulated the charts in their studies because they showed copper imports in value terms. And it is quite natural that the copper price is highly correlated with the product of copper prices and imported volumes. But looking at Chinese copper imports in volume terms shows a different picture. The regression coefficient for copper ore imports as well as for copper product imports (also smoothed by a 3mth moving average) is negative. A possible explanation for this negative relationship might be that the more copper China imports and then refines, the more the supply of refined copper increases and thus, weighs on the price for refined copper. However, R-squared declined considerably after the financial crisis, which implies that Chinese copper imports contribute less to the explanation of copper price movements.

Manufacturing PMIs for China are only available since mid 2004 for the HSBC index and from 2005 for the official PMI. For both PMIs, the impact on copper prices has risen after the financial crisis and is more than twice the beta coefficient of the first sample period.  But only the HSBC manufacturing PMI contributes more to the explanation of copper price movements.

Chinese macroeconomic variables have a stronger impact on copper prices after the financial crisis than they had before the collapse of Lehmann Brothers. But China was the major copper consuming country also before the financial crisis. Maybe it was the Chinese policy to stimulate the economy and to buy the metal for strategic reserves, which have strengthened the impact of Chinese macroeconomic variables on the movements of the copper price. However, the chart above indicates that the OECD leading indicator for the total OECD region moved most of the time in the same direction as the leading indicator for China. But in the second half of 2012, the leading indicator for the OECD increased, while the Chinese leading indicator headed further down. Therefore, the recovery of copper prices in the yoy-comparison was only short-lived.

Obviously, Dr. Copper moved to China. And there, the metal with a PhD in economics ignores what is going on in the rest of the world. The USA is still the second top copper consumer and the US construction sector is improving further. But this is not reflected in the copper futures the CME where large speculative investors are still holding a net short-position, albeit reduced from 28,048 to 22,380 contracts according to the latest CFTC report on the “Commitment of Traders”. As long as the focus remains only on China, base metals might have only limited recovery potential.