My last post on copper set me on an enquiry. Whether the enquiry turns out to be a gold mine or a rabbit hole remains to be seen. If nothing else, it has a) suggested another way to measure major market deviations, which looks promising and b) it encourages contrarian ideas about the direction of the Chinese economy.
As we know, copper market has been subject to a correction. A market full of bullish stories earlier in the year and with price rises to match, suddenly reversed course, losing ~20% of value. According to Bloomberg the pullback was the result of ‘soaring stockpiles, a relatively weak manufacturing outlook and unsatisfactory Chinese growth.’ I’m sure what Bloomberg report is true.
So far so normal. A cursory look at copper’s price history suggests this is just one of many price reversals and not even a particularly large one. Since 2010, copper prices have suffered eight 3-month price reversals larger, in percentage terms, than the current episode. They’ve suffered three larger 3-month reversals since 2016. Nothing special heh?
More detail tells a different story.
To get a true picture of copper’s behariour we need to compare its price action to the behaviour of other important markets using a common metric. As it happens, HedgeAnalytics has just the tool to help out. Our Framework takes variance and level of all major financial markets (what we call the Base Universe), extracts common drivers and reconstructs a common global financial index. The index suggests all major markets are currently closer to ‘fair value’ than copper. Not only that, it suggests copper is relatively cheaper than at any time for years. That’s quite a different picture to what the raw price behaviour tells us.
The reason for this aberration? Well, normally copper price reflects economic activity more broadly, and so tends to show co-movement with other members of the Base Universe. That co-movement is so far largely absent in copper’s current down move.
A snapshot of the underlying calculation is shown below, highlighting a selection of important markets currently shown as expensive relative to copper. Copper sits right at the top of the matrix, with all its corresponding markets marked in red across the top, signalling how cheap, relatively, copper is to each market. The larger and redder the circle, the wider the gap (see the scale on the right side to match colour to width of price gap). The list contains some of the most consequential markets for sources of global financial variance, including S&P500, 10-year TNote futures and important forex crosses. Red for ‘cheap’ and red for red metal.
As the ‘workshop of the world’, it is no surprise that China’s currency reveals a price behaviour association with copper. Odd, though, that the association has the opposite sign to that which might reasonably expected, and has for well over 12 months. Higher copper prices have been associated, until a week ago, with weaker Chinese currency (higher CNHUSD). So much so that CNHUSD is currently calculated by us to be above our estimate of fair value – i.e. the Chinese currency should be valued stronger against the dollar than currently.
Perhaps the preceding decline of the Chinese currency (higher CNHUSD) eventually triggered copper’s current price decline. Perhaps speculators acquired way too much copper in anticipation of a now-postponed Chinese economic renewal. Certainly, there are good reasons for copper to move in the opposite direction to CNHUSD. Remember Bloomberg said one reason copper prices declined was because the Chinese economy was suffering? If that were true then the Chinese renminbi (and its offshore equivalent, the CNH) has the potential to rise further above our fair value estimate. The PBoC has just cut interest rates. If lower rates mean CNH moves up to 7.5, then copper looks even cheaper against the CNH.
Yet, although there are good economic reasons why copper and CNH have an inverted association, (China weak = copper down), it is still not obvious why the relationship operated the other way round for an extended period.
One way to square the circle is to understand the recent relative price adjustment as part of a wider economic adjustment of both copper and the Chinese economy, with the PBoC rate cut as part of a countervailing adjustment. Capitulation trades are often the sign of a changing environment. Perhaps the recent fall in copper price signals the beginning of an adjustment process for the Chinese economy. That is what our metrics hint at.
How convinced am I of this? I’m pretty certain of the stationarity of our calculations, which is another way of saying that copper looks attractive against CNHUSD. But it is also attractive against a range of other important series. I confess I’m less certain that CNH will prove to be the right counterpart. It is a fascinating experiment in action. And I’m always excited by possible economic surprises from unlikely sources.