The Current Situation
We have last discussed the gold sector in a series of posts between August 11 and September 1, arguing that an interesting risk-reward proposition could be discerned, both from a longer term investment perspective and a shorter term trading perspective. In particular, with a major support level nearby, and a great many similarities in the technical set-up to previous significant lows (plus a fundamental backdrop with growing potential to shift to a more bullish configuration), an opportunity combining potentially high return with minimal risk had emerged again (meaning that risk could be minimized by using the nearby support level as a stop).
Photo via stockmarketgps.com
These posts can be reviewed here in chronological order: “Gold Stocks at an Interesting Juncture”, “A Playable Rally May be Beginning” and “Update on a Tricky Situation”. Considering that in the brief rally between late December 2014 to late January 2015, numerous individual gold stocks rose between 100-200%, opportunities of this sort are nothing to be sneezed at, even if the bear market resumes again later. This view was a fairly lonely one at the time, which is no surprise given the awful performance prior to the low being put in. In the meantime a worthwhile advance has indeed gotten underway, even if it has taken a few more retests of the low before things really got going:
The action in the HUI index over the past year. During the recent period within the blue rectangle on the right hand side we discussed the emerging new opportunity (we incidentally did the same after the capitulation in late October 2014).
One thing that should be immediately clear is that there is nothing yet that can either confirm or disprove that anything more significant than another short term advance is underway. However, there are a few differences to the last rally, which we will discuss further below. However, the main point is this: it is entirely possible that the current rally will once again fail in the vicinity of the 200-day ma – but even if it does that, it will have been well worth playing it, as even now, with the index up a little less than 30% from its low, numerous individual issues have already posted quite impressive gains and even bigger gains seem likely over coming weeks.
There still remains some distance to be covered before serious chart resistance is encountered, resp. the 200 dma is reached. Naturally, this is not going to happen all at once – in fact, it appears as though a short term pullback may have begun, resp. is fairly imminent. Such a pullback would likely represent a buying opportunity. It should ideally last a few days at most and be characterized by positive internal divergences (with some individual stocks performing noticeably better than the index).
There are two reasons why the advance may encounter resistance in the short term: Gold is close to its own 200 dma, and in silver, the previous positive divergence between its market price and its fundamental price suggested by the cobasis is no longer in evidence as Keith Weiner reported on Monday (incidentally this happened just as silver reached its 200 dma).
Gold and the HUI-gold ratio. Although resistance is close, the general backdrop remains positive for now.
In summary, we can state that it is highly likely that there is more room to the upside following a short term pullback; and secondly, we cannot be certain yet if the bear market will resume thereafter, or if a more significant rally is in store. Initially, a major trend change is in almost all respects indistinguishable from a bear market bounce.
A Comparative Analysis
Our ability to furnish you with a prediction is therefore limited, although we do of course have a personal opinion. What we can however do is try to answer the question “if the current advance turns out to be more than just another bounce, what is likely to happen?” In order to do so, we will simply take a close look at the past.
What happened after significant (medium to long term) lows were put in previously? How could one recognize that a major trend change had taken place? Naturally, there are always some differences, but it turns out that certain patterns tend to repeat over and over again. To keep it simple, we will focus on moving averages and what prices have done relative to them (primarily the already mentioned 200 dma, which seems to be playing an important role every time).
First we will take a look at three quite significant lows, which were followed by multi-month or even multi-year rallies: 1986, 1992/3 and 2000. We use the XAU for 1986 and 1992, and the HUI for later charts. The 1986 low and subsequent rally is in fact a textbook example of how big advances in the sector tend to begin:
1986: initially, the 200 dma provides resistance, but is breached decisively a short time later. Once this has happened, a multi-month consolidation begins, with the 200 dma turning into support. Then an accelerated rally gets underway.
Next comes the 1992/3 chart – as we have already pointed out in a previous post, this was quite a tricky situation, as the low was made after several failed rally attempts (we won’t repeat here what the index did relative to gold, as this was already discussed on that occasion – here is the chart).
XAU – the 1992-1993 low and subsequent rally. Here the consolidation period after breaching the 200 dma was much shorter, and we would argue this was because the moving average had already flattened prior to being overcome, and the distance between the low and the 200 dma was much smaller. Also, the 50 dma played a somewhat more significant role during the initial rise, as it served as strong resistance for a while. The current situation has more similarities with 1986 and 2000 in this respect.
And here is the pattern that could be observed in 2000-2001:
The major low in the HUI in late 2000 and the subsequent rally. One again, we see a similar pattern, with both the 50 and the 200 dma playing an important role, first as resistance and and then as support after they have been breached. The successful retest of the 200 dma from above is usually the time when the accelerated rally phase begins.
So you can see that the sector follows certain patterns with respect to its 200-day moving average (with the 50 day playing a subordinated role) whenever a more significant rally is in the works. The most important characteristics to keep in mind are: at first, the 200 dma will provide resistance. Then it tends to be breached after only a very brief and shallow pullback. This is essentially the first sign that indicates that something different to a bear market bounce is happening. Final confirmation occurs once the 200 dma is successfully retested from above. After that, an accelerated and usually quite sizable rally can be expected to begin.
Now we want to briefly look at two different examples. One is the 1998 low, which was followed by a failing rally (even though it was definitely a rally worth playing). Although a breach of the 200 dma occurred at the time as well (in fact, without the preceding pullback), the retest failed immediately, and the 200 dma right away turned into resistance again:
The failed rally off the 1998 low – it was worth playing it, but after the failure to overcome the 200 dma decisively, the bear market resumed (and would only end in November 2000, in spite of one more spirited short term rally in 1999 after the Washington agreement was announced).
As you can see, the failed rally in 1998 was very similar to the failed rally of early 2015 – in both cases, the index was immediately rejected again after the initial breach of the 200-dma.
The 2008 post crash low was in many ways exceptional due to the circumstances attending it. It also produced a pattern that is slightly different from the 1986, 1992 and 2000 examples. We include it for the sake of completeness though, not least because the 200 dma once again appeared to be an important pivot.
The 2008 low – once again, the 200 dma seemed to be an important threshold, with prices struggling for a while in its vicinity.
Additional Technical Evidence and Ratio Charts
Looking at a weekly chart of the gold price in dollar terms, we can see that it remains slightly below a lateral resistance level (1180-1200). If this resistance level can be overcome, it would obviously greatly improve the chances that the recent low in gold stocks was a significant one.
Gold, weekly – whether the bear market is over remains an open question. The resistance level indicated by the blue lines strikes us as quite important in the short to medium term in this context – click to enlarge.
There are a few positive indications for the gold sector that could make a difference compared to the previous rally attempts that have failed in recent years. For one thing, gold has entered a new medium term uptrend relative to commodities since mid 2014. This tends to be good for gold mining margins, as it indicates that the real price of gold (specifically its purchasing power relative to mining input costs) is rising.
Gold vs. the CRB Index – gold’s real price has been rising a lot more than its nominal price. This should be increasingly reflected in cash flows and earnings reported by gold mining firms. Note that gold has also strengthened in non-dollar currencies – click to enlarge.
The bullish percent index of the GDM (broad gold mining index) is rising after having revisited the zero line repeatedly, but remains far from overbought. In a new bull market, it should eventually become overbought (this means conversely that a failure to become overbought should actually be seen as a negative).
The GDM bullish percent index, which shows the percentage of stocks in the index that are currently on a point & figure buy signal – click to enlarge.
Gold sentiment as reflected by sentimentrader’s optimism index or Optix – an average of the most important surveys and positioning data – shows that bullish sentiment remains at a historically subdued level. This means there is room for a bigger rally from a sentiment perspective, but once again, it should eventually rise to overbought status in order to confirm that a bull market has begun (with sentiment following a rising gold price).
Gold Optix – the current level of 35 remains historically quite low.
Lastly, there is one correlation that continues to give us pause and is actually a bit worrisome. This correlation is a relatively new phenomenon, which we have briefly discussed on one previous occasion. As the next chart shows, the HUI-gold ratio follows the action in the broader metals ETF XME quite closely. XME primarily reflects the action in base metal stocks, which by rights should exhibit a cyclical behavior that is different from that of gold stocks over certain time periods – especially during times of clear economic weakness, resp. declining economic confidence and/or economic strength/ rising confidence.
Why has this correlation become so pronounced? The only explanation we have is the proliferation of ETFs in recent years, as well as of resource-focused funds and trackers. It appears that when base metals stocks (many of which have far larger market caps than the gold producers) are sold, gold stocks are sold by these ETFs/funds in sympathy, essentially on autopilot. Hopefully this correlation will at some point disappear again, but so far it hasn’t – although at the recent lows, there was actually a small divergence in evidence, with XME making a new low that was not “confirmed” by the HUI-gold ratio.
Mining and metals ETF XME and the HUI-gold ratio – a close correlation that is actually illogical. When the gold-CRB ratio rises as it has done since mid 2014, HUI-gold should normally be expected to decouple from XME. In other words, this correlation is likely a purely technical phenomenon. At the recent low, the HUI-gold ratio was slightly stronger than XME, so perhaps the two will indeed decouple at some point.
As we have said at the outset already, we are not predicting a specific outcome, although the current technical and fundamental evidence leads us personally to believe that the recent low in gold stocks is likely to turn out to be of the medium to long term variety, i.e., we believe a significant low has finally been put in.
However, our personal beliefs, resp. interpretation of the data may turn out to be wrong – we cannot know with certainty what the future will bring. If e.g. the global and specifically the US economy were to unexpectedly strengthen, the fundamental backdrop for gold would worsen again. We think this is unlikely, but it is not something we can categorically rule out over the short to medium term.
We can however state with confidence that the bubble will eventually burst and that the greatest monetary policy experiment of the post WW2 era will fail – in all likelihood quite spectacularly. So we have every reason to remain long term bullish on gold and gold-related investments.
Moreover, by looking closely at past lows of significance we have hopefully been able to provide a bit of a road map in case the recent low does indeed represent a major pivot point. Although the sample size we presented is small, we have no reason to expect that things will be much different from how they played out at previous lows. We haven’t shown the lows of the 1930s – 1970s period in this post, but can actually tell you that the patterns were very similar as well.