In my last main blog post a few weeks before Christmas I drew attention to a very strong MIDAS warning on the US dollar index based on two very powerful fractal Bottomfinder signals. Within a week we saw the turn predicted by the Bottomfinders in the dollar index and a precipitate collapse in gold futures. This was an excellent signal.

In this blog post (a fairly long one with at times a fair amount of technical detail) I want to draw attention to a similar Topfinder warning in the copper market. Having presented the evidence, I could leave things there, but I also want to take this opportunity to discuss China’s stock market, because I believe it has extremely important implications for my Elliott wave count on the European stock indexes. Although I have yet to extend these implications to the US markets (the subject is a large one and deserves a detailed post of its own), I believe the implications for the US markets will be the same. First, I’ll discuss copper and then turn to the Shanghai SE Composite.

Comex copper futures

On New Year’s Eve 2009/2010 copper hit a 15 month high at $7,420 per tonne, taking copper up a massive 140.2 per cent since its bottom roughly coinciding with the bottoming out of the stock markets. Some commentators have blamed the recent parabolic rise in copper to concerns about supply disruptions as strikes threatened Chilean production.

Figure 1 below shows a daily chart of copper futures bottoming in December 2008. This was approximately a month after China’s stockmarket bottomed in late October 2008 and a good three months before Germany’s DAX and the S&P 500. The relationship here between China’s growth and its inexorable demand for industrial metals such as copper is clear.

A Topfinder was launched from the very start of this turnabout in copper futures and fitted to two main junctures of this trend at a cumulative volume (D) of 5,000,000. At the time of this blog entry the Topfinder is now 99.2% done, with remaining volume of a mere 39,159. If this Topfinder reading proves to be anything like as good as the recent Bottomfinder readings on the US dollar index, it’s time again to take note.

Figure 1. Copper with Topfinder at 99.2%

The temptation to see China as the main cause of the parabolic rise in copper futures is obvious, given the lag between copper bottoming and the slower reactivation in European and US stocks. There’ll be more on this below. In the meantime, the Topfinder implications in Figure 1 are supported by the data in Figure 2. Figure 2 is a weekly chart of copper futures with total open interest in the middle pane derived from the CFTC’s weekly Commitments of Traders report.

Figure 2: copper futures with total open interest and normalized stochastic

Aficiandos of the COT report such as Larry Williams will often normalize data such as total open interest by creating a stochastic-like indicator (sometimes called the COT Index among other names) with a lookback period varying between 1 and 3 years (here it is 1 year). The calculation for the indicator is this:

((current week value – Lowest value of lookback period) / (Highest high of lookback period – Lowest low of lookback period)) * 100

This indicator can be programmed into Metsastock, as I have done here, with 50 as the average normalization and the 80 and 20 levels representing significant overextended areas. As the total open interest data in the middle pane show, total open interest in copper is even higher now than it was at the 2006 and 2008 tops. Note too that the current reading of the indicator is also as high (normalized) as it was at the 2008 top.

Experts who prefer working with net positioning of the Commercials will similarly identify an extreme low offsetting the extreme long position of the NonCommercials (funds) in Figure 3 below. Experts in the net positioning of the Commericals also often separate the data into the producers and the consumers. Producers should always be net short whereas consumers should always be net long. In Figure 3, then, it is the commerical producers of copper (the exporting countries like Chile) who are net short; producers go net short because this reduces their exposure to the risk of falling commodity prices. We’ll note this for now and return to it a little later when discussing China. Here we’ll merely note that while total open interest is at an extreme not seen for a decade, commercial producer hedgers are as net short as they can be. The Topfinder is 99.2 per cent done. I’d call this a very strong signal.

Figure 3: net positioning of the commercials in blue in middle pane, www.timingcharts.com

******

Intermarket reflections

In the first edition of his book on intermarket analysis John Murphy argued that copper is so highly correlated with the stockmarkets because it’s a key industrial commodity, being used in the automotives, housing, and electronics industries. Thus, demand for copper during increased industrial production explains the high correlation.

In the last couple of years, Murphy has been forced to write a new edition of his original book in light of a number of changes in the intermarket alignments he identified in his first study. I suspect that with the emergence of the BRIC economies as well as with lesser countries such as Chile in direct service to these economies we’ll require a third rewriting of many of these intermarket relationships. This is well illustrated by the Chilean peso having risen 25.6 per cent against the US dollar as a direct result of copper exports to China. Moreover, recent rises in copper prices were attributed directly not merely to an accelerating Chinese market but to supply disruptions in Chile as a result of industrial strikes.

Implications for China

So far, we have volume data in the Topfinder and extreme readings in the weekly COT report total open interest and net Commercial positioning data all warning on copper. Why? The weekend financial papers were full of interest in China, particularly in the idea that China is a high-risk play right now. Here are a few facts:

1. China’s foreign exchange reserves rose again in the fourth quarter to hit $2,399bn, in part due to a flow of foreign speculative capital into its asset markets.

2. Due in large part to a collapse in China’s exports in 2008, China’s state-sponsored banks made double the amount of bank loans in 2008 as a result of government orders, flooding the economy with cheap credit. In January alone of this year the banks lent $50bn. This lending helped China beat the 8 per cent annual GDP target set by the government at the start of the year, but it is also highly inflationary and has already started creating rapid asset bubbles, particularly in the property market (copper?).

3. In a sobering article in the weekend FT, Merryn Somerset Webb made several additional points:

• China has been growing at a rate of 10 per cent a year for 30 years — Pivot Asset Management notes that China’s capital spending boom has now outstripped all other post-war country transformations, including Japan’s and Germany’s. China’s growth cycle is mature not emerging.
• This is reflected in a variety of facts and statistics, including: a newly built city called Ordos, purportedly home to 1m people, yet lying empty, with only property speculation driving up prices at 50 per cent a year. The huge number of ports, offices, airports and steel plants in China, suggesting a surplus to requirements, plus rental yields in Shanghai as high as 50 per cent — yet China is still trying to press ahead with its building work.
• A recent National Bureau of Economic Research report that cited China as a potential beneficiary of an indicator common to every financial crisis going back 140 years: rapid credit growth leading to disastrous deflationary implications.
• A Graham and Dodd p/e ratio (using a 10 year average) of around 50 times earnings for the Shanghai SE Composite — not a good value investment in anyone’s book.

The People’s Bank of China has responded by tightening liquidity recently, especially by increasing the bank reserve requirement ratio. It has also raised the yield on 1 year central bank bills and did the same with 3 month bills last week. An additional measure would be to allow the renminbi to appreciate within looser settings. As some analysts have pointed out, the latter is looking likely in view of a rapid recovery in exports and increasing CPI-related inflation data. The COT net positioning data reflects the view strongly that the PBoC moves could stifle demand for copper, as commercial producers are hedging by shorting futures heavily against this eventuality at near record levels of total open interest.

Longer-term implications

My own longer-term Elliott wave count on China is provided below in Figure 4. The most salient part of this count is that the second deflation wave (wave C) is expected to begin in 2011, with a fairly quiet 2010 as wave B of this massive ABC corrective cycle completes on its two subwaves, b and c. The bearish implications of China’s credit overextension are consistent with this count. So far, since rallying from its late October 2008 low, The Shanghai SE Composite terminated on a putative wave A of a larger wave B at a Fibonacci level of 0.382. According to this count, subwave b of B is now in progress before 2011. The middle pane contains an indicator known as the Elliott Wave Oscillator, which was developed by the engineer and trader Tom Joseph, who developed Advanced GET. The EW Oscillator is a version of the MACD with specialized settings. Here it’s programmed in Metastock. Notice that wave 3 of the large wave A decline produced the deepest low (highlighted by the vertical blue line, which is precisely what the indicator is supposed to do on wave 3s). Thereafter it positively diverged by several months, forewarning of the upside. Now it is starting to diverge heavily again (negatively). This is consistent with the 2011 wavecount for a large credit deflation in wave C.

Figure 4: wavecount on Shanghai SE Composite

Implications for European markets

I’ve included this putative wavecount on the Shanghai SE Composite here because it’s identical to one of the two Elliott wave patterns I’m seeing in Europe at the present time.

The same one of these patterns is identical to the pattern exhibited by the S&P 500. Having recently viewed 14 European stock indices, two patterns stand out — one I’m calling the Austrian/Danish pattern and the other the Finnish/German pattern. The latter differs in certain respects from the former and is deserving of a separate blog post, because it’s far too much to include in a single post. However, the Austrian/Danish pattern (which is also on the UK’s FTSE 100) is exemplified in Figure 5. Figure 5 is the Danish index and we can see straightaway the similarities with the Chinese stock market. As such, the wavecount is the same and the implications are also the same as regards price and time. The timing is set for early 2011 and the price target follows the Fibonacci proportions of a typical ABC zigzag pattern, which is that wave C will be proportionate to wave A which terminated in 2008.

Figure 5: the Austrian/Danish pattern also in the FTSE 100, the S&P 500, and the Shanghai SE Composite

In Figure 5 notice as the market reaches the Fibonacci 50% level that we have the same negative divergence brewing in the EW Oscillator. Also the weekly RSI is now approaching overbought for the first time since the 2007 top (note too the oversold reading in late 2008). Finally, Figure 6 has a Topfinder launched since the bottom in early 2009 and it too has just expired on what is the final putative 5th subwave before we see the expected market turn in this ongoing waveount.

Figure 6: expired Topfinder in the final 5th subwave

Final note on the Elliott wave count

Readers who follow Elliott wave counts closely on the world indices will know that this wavecount and its projection for the start of wave C in 2011 is in direct opposition to the current wavecount on Robert Prechter’s site, www.elliottwave.com. According to the Prechter count, wave C is already in progress, with the current upside being a wave 2 correction of this massive deflation C wave. This count depends on reading the decline between 2000 and 2003 as wave A, whereas the present count offered here depends on taking the collapse from the 2008 high as wave A. It must be acknowledged that this is a massive topic which requires looking at wave patterns in other markets such as the Nasdaq as well as the long-term historical wavecount. Such a topic is the subject of an entirely separate blog. For now I merely post this putative count.

Whichever wavecount proves to be correct, I close this blog post with the following data, which offer a sober indication of how close we are to a market correction, irrespective of whether the market will turn down in a continuing wave C immanently or whether 2010 will prove to be a quiet year before the next deflation wave (wave C) in 2011.

• The risk appetite for high yield (junk) bonds is now at pre-financial crisis levels, as high yield bond issuance is surging and credit spreads are narrowing alarmingly.
• As noted by the FT over the weekend, the cost to insure junk bonds against default — a good gauge of market sentiment — has fallen to its lowest level since the market highs of 2008. The iTraxx Crossover index is trading at 407 basis points and has fallen over 100 basis points since peaking at 1,142bp at the market low.
• An alarming interest too in emerging market bond issuance, where yields have also narrowed dramatically with US Treasuries.
• The VIX (the so-called fear index) has now fallen to levels last seen in the 2008 top (now at 18.47), which means that traders think there is less chance of future volatility and less risk of losing money in stockmarkets.
• Sovereign credit default swaps have rapidly risen in recent weeks as part of a trend that has been in place since early autumn. These swaps measure the risk of government default and their cost has continued to rise, especially with regard to Greece, Spain, and the UK.

A. Coles 18th Jan ‘10

Tweet This Post