Conflicting trends and lazy bureaucrats
The International Aluminium Institute has released the production figures for February, and the numbers seem to reveal a couple of trends, but beware.
Measured on a tonnes per day basis, production for the rest of the world ex-China has risen slightly over January (1%) and by 2% compared to February 2014. But it is not a uniform move, and there are no surprises in the data. Compared to 12 months ago, South America and Africa are down, GCC is up, Western Europe is up, Australia/New Zealand is down. Simply put, GCC production increases have cancelled out the cutbacks in most other parts of the world.
What this means is that the structural shortage of new metal units is yet to reveal itself fully. A 2% rise in supply is not enough to cover demand growth RoW, but combined with legacy metal exiting finance deals and semis exports from China, the new super cycle is a long time coming. But until Alba Line 6 starts up, it’s unlikely GCC can continue growing output, meaning that a structural gap in supply will be a slow reveal.
China on the other hand presents a different picture. At first glance, China seems to have grown slightly over January, achieving 89,400 tonnes per day. But that’s not a true reflection, since the IAI loads a set 300,000t per month of unreported production. Divided by 28 days in February, that number artificially inflates the daily rate for China. Without the unreported production figure, China’s daily output was unchanged. (Unreported production is a proxy for Weiqiao, which is running at 3.6 million tonnes per year, or 300,000t per month.)
But the real news is in the comparison with 2014. February shows a growth of 20% over Feb 2014. In fact, the problem seems to be in the January numbers. January’s daily rate jumped from 70,000tpd in December to 78,700tpd. At no stage in the industry’s history have we seen so much capacity added in just one month. And February’s data, leaving out unreported production, is identical to January, despite such a jump the previous month. Quite simply, the numbers appear to be made up.
Aberrations like this tend to support our case that one should treat production data in the first 3 months with a great deal of care. January’s data appears to have been loaded or guessed, and February’s data seems to be just a repeat of January. The secondary message is, beware of analysts and commentators who proclaim that China is making too much metal, based on this data. Maybe they are, but we can’t use this data to prove it, only to prove that there some lazy bureaucrats somewhere in Beijing.
0