To many the recent heights achieved in gold has been astounding. Gold has soared roughly 20% since July. On a daily chart it looks like a parabolic blow off is occurring and the inevitable reaction and sell off is right around the corner. While I agree that gold will likely react from these levels, the charts suggest the reaction will be short-lived and muted.
While some may consider a three-year chart “very long-term,” we actually need to go back further than that in the case of gold. Very powerful trends can and do exist for longer than three years, and support and resistance levels can remain important for decades. If you start looking at something on a daily chart, you may think one thing or another without even realizing a very powerful contrary indicator or level on a longer-term chart. You won’t realize it because it is literally “off of the chart.” By looking at a monthly chart, we can see the historical data more clearly and over a longer period of time than what a weekly or daily chart can provide, and with less emphasis on the minor ups and downs. This is important.
The longer a trend or the longer an important support or resistance level has held, the stronger it is. This sounds obvious, but many investors and analysts fail to look back far enough into the past; they are overly concerned with the future and where price is headed in the very short-term, and yet they don’t realize that the long-term implications dictate events in the short-term. This is only a natural reaction, given the emotional state one enters into when a new or potential investment is being made, and this is why it is important to take a more objective approach. Step back, get perspective. This type of long-term historical analysis can benefit all investors, even daytraders. It is easier to trade the long side on a stock that is in a long-term uptrend, and short one that is in a long-term downtrend.
Unless you are a daytrader or someone with a very short-term time horizon, then the short-term doesn’t matter as much as you think. Do not neglect to use shorter-term charts to time entries and exits, of course, but also realize weekly and monthly charts can be used collectively for this purpose as well. The long-term, big picture, and major persisting underlying conditions are omnipotent. They provide the backdrop for the action: everything happens in context: everything is relative. When you look at the ten-year chart of Gold, you see a very strong trend. And it appears to be only getting stronger.
If you look at the chart below you will notice a persistent and relatively clean uptrend. There were three definite intermediate reactions in the last ten years that were to be expected given the multi-year history of the trend. The first was a nine-month line that formed beginning in December of 2004 and persisted into the third quarter of 2005. The price was relatively unchanged during this time-period as gold consolidated around the $430 level persistently. The second was another consolidation that began in May 2006 and lasted for fifteen months until August of 2007. The third and most severe reaction was a 30% decline from about $1,000 an ounce down to about $700. This period lasted seven months from March to October 2008.
The rest of the reactions were relatively minor given the long-term nature of the uptrend. There is a period beginning in December of 2009 and lasting to about July 2010 were price remains relatively unchanged over that period, however nearly 100% of that reaction took place in only the first two months beginning in December, and it could be construed as either a minor or intermediate reaction, depending on the technician. Whatever the case, it is important to note that most of gold’s intermediate reactions have been consolidations – with the exception of 2008 when the value of most things were plunging.
Looking from top to bottom, you will see a MACD that indicates a strong and persistent trend with Augusts’ candle showing a slightly more overbought divergence than what has been typical. However, as we are evaluating a candle that is still developing (for example Augusts’ candle will not be confirmed until 4pm Wednesday August 31st), we must be cautious about drawing any precocious conclusions. That being said I put the current move for August as in-line with the historical trend, however a lower close at the end of August, with the convergence of the MACD and signal line would be a bearish divergence.
RSI is in overbought territory, however this is indicative of a strong trend. It is also a bullish sign that RSI (at green line) is not as dangerously overbought as it was in April 2006 and February 2008 (red line). Looking below the chart you will notice these periods coincide with very steep 10-period ROC’s. Compare this to now and you have a 10-period ROC that indicates a trending environment, thus confirming a trending RSI and trending MACD. The period now is most similar to the trending environment in gold from 2002 to the end of 2004 that ended in consolidation.
The arrows mark each time price closed significantly above the upper Bollinger band on a monthly basis. A close above this band indicates either overbought or bullish conditions, depending on what other pieces of information are telling you. I have calculated a three month, six month, and one year return on each instance this has happened. The one year returns since this has happened in 2002 are 11.6%, 9.4%, 14.3%, -9.3%, and 17.1% respectively. The only move lower after a year was if you bought in late 2007 and sold at the very bottom of the market for gold in 2008. The three and six month returns are more mixed, with exactly half showing gains and half showing losses after this time. The cumulative three month gain is 14.8% and the cumulative six month gain is 18.2%, indicating that while perhaps the odds are neutral that a buy on an upper Bollinger band breakout like we’re seeing now will result in a profit three or six months later, the profits are likely to outpace any losses incurred over time, especially after a year.
Point A. marks the time when GLD broke out above the blue line (upper trend line which in this case is actually red), while point B. marks the reaction back to this line. Here the trendline resistance turns into support in a classic reversal of support of resistance roles that we would expect. It is somewhat difficult to see on the monthly chart but GLD just broke out above another upper trendline. This we will examine on the weekly chart below (the black rectangle).
The arrows on this chart point out every time in the last few years that RSI has reached overbought levels. In two of these cases (2009, 2011) the price reacted back to the 20W MA, while in the third (2010) price actually moved higher before reacting back to the same level it first showed an overbought RSI. Point A. marks the upside breakout from the resistance trendline on overbought RSI, exactly like what is happening now. Point B. is the reaction to the resistance trendline turned support line. The price breached this line briefly on a false breakdown before recovering its stride and moving higher.
Now you have a well-defined trend channel with both upper and lower bounds. The second trendline beginning at point B. is extremely bullish, because gold fell shy of reacting to the longer-term line. When this happens it is bullish because it suggests gold is even stronger than the trendline suggests, finding support at a level above the well-defined line. Now gold has gone above its upper trendline yet again, indicating an upside breakout that will likely cause a continuation of trend and if so will likely be at a higher velocity. This could in turn lead to a parabolic blow off and subsequent reaction, but everything so far has been relatively inline and until we see much higher prices this reaction does not seem to be upon us. In Technical Analysis of Stock Trends, Edwards and Magee indicate that an upper trendline breakout is a buying opportunity as long as its not a breakout from a rising wedge or an upper trendline in a downtrend. This is classic technical analysis and not to be misinterpreted.
What are the technicals saying? As of last week Gold just put out another buy signal. The trend is extremely resilient and the past ten years suggests that buying during similar periods have delivered profits after a year or less, with the exception of the 2008 crash when the baby was thrown out with the bathwater, and the value of all things dropped. While I am not saying another panic sell off would be “different this time,” but there are certain things to consider.
Since 2008 we have seen massive inflows of cash into the economy and a weaker dollar. Interest rates are at record lows on the 10-year. Its possible the trend in gold is fueling itself as a sort of proxy yield for investors with some added risk/reward. The weaker dollar trend may be helping, however gold rose even as the dollar index rose in the first half of 2010.
The 10W ROC on gold does look overbought, but keep in mind it doesn’t necessarily have to peak. It can remain elevated as trend persists over time. Point C. marks a possible short-term reaction area for gold which should move along the resistance line as time passes. Pay extra attention to the price and action of gold as it nears these levels, but also keep in mind the levels will be different depending on when price meets this line again (since it is not horizontal and could be months into the future). If the resistance line does not turn into price support, or acts only as temporary support, a decisive break below this line would negate the buy signal.
It is extremely important to go back and view 10-year charts or longer for trends that have persisted for multiple years. If you just looked at a weekly chart (black rectangle on monthly chart) you can see how small it seems in comparison. The technicals will talk to you and as a market technician it is important to remember the implications of breakouts, even if in the short-term there may appear to be a top – you just cannot ignore powerful upside breakouts in powerful trends. This is a clear indication that something major is going on and it is likely to persist for some time. Calling a top here would be very near-sighted. Or maybe I just haven’t gone far enough back in my research.