E-Wave Counts Revisited

In late August, we looked at a few short, medium and long term wave count alternatives for the HUI (see “Gold and Gold Stocks, What Comes Next” for details). We want to revisit the topic to see where things now stand. At the time, our main assumption was (and remains) that the entire long term structure since the March 2008 high is a large degree corrective wave. Since the low in late June 2013, it became clear that some sort of bottoming formation was likely underway, not least as the extent of the bear market from the 2011 high was roughly the same as that of the most severe bear markets in the sector since the 1940s. However, it was of course not certain how things would play out in detail. With hindsight, it turned out that the near term bearish alternate count reproduced below was the correct idea (we also considered a more bullish alternative, as well as a 'short term bullish, medium term bearish alternative' shown further below).

This was an idea from our late August update that turned out to be correct – click to enlarge.

We were wondering at the time whether the long term wave 'C' was or wasn't finished yet. As you will see further below, that question can still not be answered with absolute certainty, but we have a few positive indications from ancillary technical indicators that suggest wave C may have ended. First a look at the above mentioned 'short term bullish, medium term bearish alternative' we also pondered in August, in the context of the formation of the fourth and fifth waves of C. We now know that this view was too optimistic in the short term, but otherwise, namely in terms of the general shape and timing of the formation subsequently traced out by the market, it was fairly close to the mark. Had we used the point marked 'c' in the chart above as the high point for wave four, we may well have gotten it perfectly right:

The idea proposed in August for waves 4 and 5 of C – too optimistic regarding the short term, but otherwise it seems the idea was sound – click to enlarge.

If we look at what has since then happened, the market did indeed make a low in late December, only the formation started from a lower peak and was more downwardly slanted, in the process making a new low. However, in spite of following the above template pretty closely, this has created a new problem. The problem is that the putative fourth wave looks a bit small relative to wave 2, both in terms of size and duration. It can therefore not be ruled out that wave 4 is  not finished yet (as per the 'alt' count in red on the next chart).

On the other hand, if one looks at the daily chart, the decline from the late August high looks actually very much like an ending diagonal. Furthermore, numerous divergences have developed at the lows between prices, RSI, MACD, trading volume, the bullish percent index and sentiment indicators on both daily and weekly charts in the shares and the metal. Typically such divergences are associated with major trend changes, or let us better say: the longer prices manage to hold above the levels they inhabited at the time the divergences were created, the more likely it is that a major trend change has occurred.

So there is still a possibility that wave C needs at least one more downturn  (namely a 5^th wave), but there is a lot of evidence mitigating against that outcome. On the other hand, even if it were to happen, it would still be part of a bottoming process, so from a longer term perspective it is nothing to be feared. In the short term, the recent bullish trend must be regarded as intact, as long as pullbacks hold above recently broken resistance levels.

Here is the current weekly chart of the HUI:

HUI weekly – our preferred wave count in blue, the alternate in red. If the alternate turns out to be correct, than a final fifth wave down will still be required, but the many price/momentum divergences on the weekly and daily charts argue in favor of the preferred view – click to enlarge.

The HUI daily, with lateral resistance levels, divergences and the possible 'ending diagonal' indicated – click to enlarge.

Wedges, Relative Performance, Divergences

Note that major lows in the HUI preceded by some form of diagonal or wedge have occurred on several occasions. One prominent example is the low of late 2000. What is noteworthy about this particular low is that the HUI actually bottomed a full six months before the gold price did and subsequently outperformed the gold price sharply (after underperforming prior to establishing the low). Currently the two are far more closely correlated, even though the HUI exhibits a notable degree of outperformance as well.

The HUI and gold (green line) in 2000-2001. Note the wedge at the end of the decline and how strongly the HUI subsequently outperformed gold – click to enlarge.

At the moment the HUI outperforms the gold price as well, as the HUI-gold ratio shows. However, the degree of outperformance is not (yet) similarly pronounced:

Whether it makes sense to assume that lateral support and resistance levels exist on a ratio chart may be debatable; essentially the levels are a derivative of the HUI's own support and resistance levels. What is important though is the trend of the ratio. In bull markets it tends to trend up. Whether the most recent uptrend can be maintained remains to be seen of course – click to enlarge.

Finally, here is an updated chart of the GDM bullish percent index. This shows the percentage of index components currently sporting a point & figure buy signal. It is useful as an overbought/oversold indicator that gives buy and sell signals, as well as a measure of divergences. The latter function is especially interesting. E.g. when the index makes a new low and the bullish percent index doesn't, it indicates that there is a certain degree of internal strength developing in spite of the weakness of the index. This is in fact what occurred on occasion of the lows made in December:

GDM bullish percent index vs. the HUI (green line above). The red vertical lines align lows in the HUI with the BP measure. The blue dotted lines indicate the divergences recorded at previous short term lows relative to the December low – click to enlarge.

Gold and GLD

Gold itself indicates very similar possibilities with regard to its 'wave position'. It is too early to tell for sure whether wave C is finished (always assuming that the 'A-B-C' correction idea is actually correct of course, but this remains our operating assumption for now) or whether it requires one more wave down to find its final low.

In this case the idea that a durable low has in fact been put in is also supported by numerous price/momentum and price/sentiment divergences (we have discussed the divergences with sentiment and positioning data previously, seethis update. Here are links to the charts: price vs. CoTs, and price vs. 'public opinion').

Nevertheless, we cannot yet rule the 'alternate count' out. What creates the remaining uncertainty is that there is nothing that initially differentiates a new bull market from a bear market rebound, at least nothing that is obvious. If one looks for instance at the July-August 2013 rebound on a line chart, there was nothing in the price chart itself that indicated that the rebound would end after  just three counter-trend waves had been put in. The same is true of the most recent advance – we simply don't know yet whether it is merely another corrective rebound or something more. What we do however know is that the retest of the late June low occurred on lower volume and was marked by very strong price/momentum divergences on the weekly chart. This is definitely a positive indication. Upcoming pullbacks should ideally be contained in the $1,280-1,290 region, a previous area of resistance.

Gold weekly, with divergences, preferred and alternate (in red) wave count. Here too the alternate would imply one more down wave, but the many divergences argue in favor of this not happening – click to enlarge.

The support area mentioned above is more easily visible on the daily chart (if one is generous, one could widen it a bit, to $1,270 to $1,290). We would argue that in order for the bullish alternative to remain intact, this support zone should ideally not be violated on a closing basis on pullbacks. If it is broken, then the alternate count would likely be the correct one.

Gold, April contract daily. A small triangle has recently been built, which is normally a short term continuation formation. The red lines indicate major lateral resistance, the blue lines the support area that needs to hold on pullbacks – click to enlarge.

A brief comment on GLD: we have recently come across an opinion piece in which it was argued that the recent lack of additions to the trust's gold holdings proves that gold demand in the West is still too weak and that the recent price gains therefore won't hold.

However, we would point out that after gold topped in 2011, GLD's holdings continued to rise even after gold's correction/bear market had begun. In fact, there was a bearish divergence at the secondary high (a lower gold price concurrently with GLD holdings reaching a new high). Changes in GLD's holdings are largely a side effect of the short term demand for GLD shares. If the shares trade at a large enough premium to the trust's NAV, authorized participants will sell GLD shares and concurrently buy gold to exchange it for new share baskets. They can make a risk-free arbitrage profit that way, while at the same time ensuring that the trust's price won't stray very far from its NAV. Conversely, they will buy shares and sell gold if  GLD shares trade at a discount to NAV. The legs of these transactions that remain 'open' at the moment the arbitrage is executed are closed out either by a delivery of gold to the trust in exchange for new shares, or by dissolving a share basket and taking delivery of gold from the trust.

In other words, all that can really be said about the fact that the trust's holdings haven't grown much is that there hasn't been sufficient enthusiasm among market participants to push GLD's shares to large premiums to NAV in intraday trading. It is not obvious to us that this is necessarily a negative sign at this stage of the rebound. It mainly seems to indicate that there is still a lot of doubt about the rally, so one could well interpret it as a contrarian bullish sign. That the amounts bought and sold by GLD as such don't matter much to the gold price should be obvious: the total supply of gold as well as the amount of gold traded daily on a global basis both vastly exceed the movements of gold in and out of GLD's vaults. The size of the trust's holdings is largely an effect, not a cause of gold's price movements. If traders become more enthusiastic about the rally and regularly bid up GLD shares above their NAV in the short term, the fund's gold holdings will tend to rise.

Conclusion:

For the moment, the bulls have the upper hand in the gold market. As long as the newly established support levels survive on pullbacks that will remain to be the case. Nevertheless, one should not lose sight of the fact that rebounds in bear markets cannot be objectively differentiated from initial rallies in new bull markets. One must therefore keep an open mind regarding the short term possibilities. We believe that nothing would materially change for long term oriented investors, regardless of whether or not one more wave down occurs. The region between current levels and the late December lows strikes us as an area in which it definitely makes sense to invest in the sector if one has a long term view. Short term traders however no doubt need to keep a close eye on the support levels discussed above.

Addendum:

Readers may recall that we have occasionally mentioned the fact that early in gold bull markets, there tends to be a 'sweet spot' for gold mining margins, as very often gold rallies coincide in their early stages with falling input costs. In this context, note this recent Bloomberg report on the bear market in rubber and the falling input costs enjoyed by tire makers. To this we would point out that bear markets in rubber don't just fall from the sky – usually they are a sign that demand for tires is falling or is expected to fall. Large industrial tires are quite an important input cost item for large open pit mines.

“Rubber’s bear market is poised to deepen as the global supply glut stretches into a fourth year and stockpiles in China swell, cutting costs for tire makers.

[...]

The world’s largest tire maker projects a 5 billion yen benefit this year from lower prices that will also help Michelin (ML) & Cie. and Goodyear (GT) Tire and Rubber Co.

“You can’t be bullish on the outlook for this market now with huge volumes of inventory,” said Takaki Shigemoto, an analyst at Tokyo-based researcher JSC Corp., who was among a majority observers forecasting price gains in October. “Rubber reserves in China almost doubled in the past six months and rising stockpiles signal ample supply from producer nations.”

Rubber dropped into a bear market on Jan. 28 amid concern that global production was outpacing demand and as inventories in China climbed to a nine-year high. Vehicle demand in China, the world’s largest car market and the biggest user of rubber, is slowing as anti-pollution and austerity campaigns spread.

(emphasis added)

We conclude from the above that falling costs for tire makers will eventually translate into markedly lower costs for major industrial users of tires as well.