Trading Gold Volatility
Volatility, or more precisely the surreal lack thereof, is in the news more and more these days. The usual summer doldrums have been much more pronounced this year with 2005 witnessing the lowest raw volatility levels in over a decade.
Volatility, or the speed and magnitude of daily price movements, is a wonderful trading tool since it is so directly tied to prevailing sentiment. When investors are scared and selling like crazy volatility rockets higher but when they are complacent and euphoric volatility grinds lower.
Thus, volatility trends form a priceless window into the popular psyche, empirically reflecting the unseen yet immensely powerful emotions of fear and greed that drive short-term market movements. Speculators who diligently follow these volatility trends can gain a tremendous trading advantage since they often reveal when popular sentiment has swung to unsustainable extremes.
Volatility in the stock markets is well understood after being studied for centuries. While complacency today is so high that more and more investors are deluding themselves into thinking volatility is dead, contrarians aren't fooled. Abnormally low volatility periods are always followed by offsetting swings into very high volatility territory, which can only be spawned by sharp price drops.
But the greatest bull market of this new century is unfolding in commodities, not the general stock markets. Unfortunately commodities have been out of favor for so long that not many volatility studies exist on them. As a student of the markets I want to understand the volatility signatures of key commodities because they probably offer excellent trading signals to astute observers.
After establishing a stock volatility baseline with the S&P 500, last month I took a look at silver volatility. Interestingly, and surprisingly, the results were not as expected. Rather than having high volatility near major bottoms and low volatility near major tops, as the stock markets always exhibit, the silver volatility profile was curiously inverted. Silver tended to be more volatile near highs and less volatile near lows.
While no doubt unconventional and odd, at least the silver volatility signatures were consistent, and hence tradable. As an extremely small and purely speculative market, silver moves very rapidly and is easily blown about based on the capricious whims of speculators. In light of silver's inverted volatility profile I have been really curious to see how gold looked.
Gold is an enormous market compared to silver, although still far smaller than the general stock markets. And gold is a timeless investor favorite so the ratio of investors to speculators is much higher. Investors tend to hold for long time horizons, so they are a moderating influence to the endless volatility that speculators live for and love to churn up.
So, following the same methodology used with silver, this week I delved into gold volatility trends. The results of these studies were definitely illuminating, with gold being similar to silver in some ways but not quite to the same degree. If is fascinating that volatility can manifest itself so differently in different markets.
Our charts this week quantify gold interday volatility over the past decade or so. Interday volatility is the percentage price change in gold from one trading day to the next, from close to close. It doesn't matter whether gold rises or falls, we are just concerned about the absolute magnitude of these daily moves. Just as in silver, we stratified these moves into 1%+, 2%+, and 3%+ tranches.
These three levels of daily volatility are then counted over a "rolling month" and charted. Since an average month has 21 trading days, this rolling-month concept centers a 21-day window around each trading day. Then the number of 1%+, 2%+, and 3%+ days that occur within 10 days before, 10 days after, and right on a given trading day are rendered on the charts. The result is a kind of volatility frequency distribution across the seas of time.
Gold's volatility profile is unique, similar to silver's in many ways but not quite as extreme. It is certainly interesting and will help gold speculators better understand how price volatility patterns can signal superior times in probability terms to launch long or short gold trades.
For six millennia now gold has been highly prized, so it universally sought after but seldom destroyed. The total gold mined in the history of the world is believed to run about 150,000 metric tonnes. If this number is reasonable then the total global gold supply is worth $2050b or so. This utterly dwarfs the global silver market, where minimal stockpiles exist and annual demand is thought to run about 840m ounces a year, worth less than $6b at $7 an ounce.
With such massive supplies of gold floating around in the hands of countless investors, it is not surprising that gold is far less volatile than silver. The raw size of the market almost necessitates it. A $1 move in the price of gold at $425 translates into a staggering $5b change in "market capitalization" of the entire world gold supply. Thus big daily percentage moves, since they are a tail wagging such a massive capital dog, are considerably rarer in gold.
If you compare this chart to last month's silver interday volatility one, gold is far more sedate as expected. Over the last decade, through bear and bull markets, gold has seldom exceeded 10 days per rolling month of 1%+ absolute interday volatility. Silver, on the other hand, seems to spend about half its time over this benchmark.
Yellow 2%+ days are really not very common for gold and the red 3%+ days are just downright rare. This is a marked contrast to silver, where giant yellow and red interday volatility spikes are par for the course. The bigger a commodities market the more capital it takes to move it and gold just dwarfs the small and highly speculative silver market. And the legions of gold investors who don't sell very often do exert a powerful moderating influence on gold volatility.
Live silver though, the timing of the major volatility spikes in gold is not what we would expect based on conventional stock-market wisdom on volatility. In stocks volatility spikes high during the extreme fear surrounding sharp corrections, near major interim bottoms. Then it fades away as prices rise and becomes almost extinct near major interim tops when everyone is smug and complacent, like today.
If you look carefully, gold doesn't conform to this standard volatility construct. In late 1999 and early 2000 gold witnessed high volatility not on corrections or lows, but right at the tops of sharp price spikes. The large yellow and red spikes since 2001, the bull-market phase in this chart, also tend to cluster around major interim highs instead of lows.
Conversely, especially since 1998, low volatility tends to cluster near lower points in gold's price chart. Rather than gold players growing scared when prices have corrected, they seem to get complacent. This same strange phenomenon occurred in silver and I speculated on its causes in my earlier silver volatility analysis. After that essay was published, one gentleman graciously wrote in and shared an intriguing thesis on these unexpected volatility inversions that I hadn't considered.
Many contrarian investors believe gold and silver are dominated by short interests, parties that don't want to see the prices rise. Regardless if the shorts' motivations are political, like central banks trying to stave off too much scrutiny on their ruinous inflationary fiat regimes, or profit-oriented, like hedge funds, these shorts are selling gold and silver they do not have. A lot of excellent work investigating this gold-short trade has been done by contrarian analysts.
Short sellers borrow assets, sell them, and then attempt to buy them back later at a lower price to repay their loan and earn a profit. With profits earned on price drops, shorts' emotions are exactly opposed to normal longs' emotions at major interim tops and bottoms. If you are short, and a price is hitting lows, you are probably fat and happy and complacent since your profits are very high. But if a price is hitting highs, you are probably panicking and fearful of the potential unlimited losses that your shorts are exposed to in major rallies.
In gold and silver, perhaps these volatility inversions can be explained by short dominance. Fear and hence volatility runs high when shorts face rallies. And indeed, the sharp gold rallies in late 1999 and early 2000, as well as the late 2002/early 2003 spike were definitely accelerated by shorts covering. I was watching all three of these fast spikes in real time as they unfolded and I remember well the shorts scrambling. It was a beautiful thing.
And near major interim lows in gold and silver, naturally the shorts would be serenely basking in unrealized profits. As the general stock markets are so abundantly proving today, when speculators are happy and their positions are deep in the money they become lethargic and full of hubris. They are not trading too much nor are they afraid of anything. Complacency is an exceedingly dangerous thing though, long or short, because the market conditions that spawn it never last.
So I am very grateful to the gentleman who graciously shared this intriguing short thesis with me. Shorts have inverted volatility profiles, they feel greed when longs feel fear and vice versa. So perhaps this may be a major factor in the strange inverted volatility profiles of gold and silver.
I am passing this intriguing idea along merely in the interest of advancing debate, so please be aware it is not without limitations. One of the big ones is the fact that most speculators in gold or silver operate in the futures markets as opposed to physical. Even gold conspiracy theories argue that futures, or paper gold, are instrumental in any short schemes to retard its advances.
But, as all futures traders know, the total numbers of longs and shorts in any given market are always perfectly offset. Futures are the ultimate zero-sum game, for every seller there is always a buyer and any capital won by one party is directly lost by the party on the other side of that contract. I recently wrote an essay on gold futures explaining that particular market in more depth.
With short interest in gold futures always equaling long interest, the leverage of shorts does not seem as stupendous as some believe. It's not like 80% of the gold futures market is dominated by shorts, they control exactly 50% as they always have, they always will, and they do in every other futures market. So as you ponder the curious inverted volatility profiles of gold and silver, keep in mind that futures are designed to be zero-sum games by definition with perfect, perpetual parity between longs and shorts.
Back to the chart above, I also found it interesting that gold's volatility profile is abnormally low today, nearly the lowest we have seen it since 1999. Due to the surreal lack of volatility in the stock markets summer 2005 has been a slow season of lethargy for speculators. It is interesting that gold's own volatility has been trending lower for a year or so and is mirroring the general malaise.
But while abnormally low volatility is absolutely a danger sign for the general stock markets, a harbinger of a sharp fall to restore balance to sentiment, with gold's inverted volatility profile it is actually bullish. Amazingly enough, gold is much more likely to rise considerably after low volatility periods. Our next chart delves into this phenomenon, zooming into the gold interday volatility data since today's secular gold bull launched in 2001.
In order to analyze this raw gold volatility data within the context of its bull to date, we have to arbitrarily assign high volatility and low volatility benchmarks. I settled on calling times when gold had one or more 3%+ days per rolling month as high-vola episodes. Conversely, when gold had two or less 1%+ days per month I considered them low-vola periods. These benchmarks are far smaller than silver's four or more 3%+ days and seven or less 1%+ days.
These arbitrarily defined high and low volatility episodes are numbered above. By going through them systematically we can gain a better understanding of how speculators can use gold volatility as a trading tool. If a certain volatility event had a high probability of occurring before a significant move in this bull in the past, then odds are this relationship will persist into the immediate future.
Bull to date there have been seven high-volatility episodes, all marked and numbered in red above. 1 and 2, both in 2001, each occurred on sharp spikes up in the price of gold. If you weren't paying close attention back in the early days of this bull, short covering was often the reason for sharp but short-lived spikes higher. With gold languishing around $275 at the time few believed in it except belligerent contrarians. Thankfully we gold investors have come a long way from those humble beginnings.
High-vola episodes 3, 4, 5, and 7 also occurred when gold prices were relatively high compared to their surrounding technical price environment. At episode 6 the gold price was also high relative to where it had come from but after a short correction soon reversed and roared higher. And the 3, 4, 6, and 7 volatility highs are centered just to the right of their respective gold tops, so they were likely spawned by the resulting correction and not the actual initial run up to the top like 1, 2, and 5.
But regardless of whether these vola spikes occurred leading into a gold top or immediately after a top, speculators would have been well served by selling gold and waiting for a correction in six of the seven times that gold volatility has been this high bull to date.
In light of these results, speculators should really avoid throwing long when gold has been rallying and is getting fairly volatile. So far in this gold bull, even though it defies conventional wisdom, high volatility has generally been the sign of a gold market near a short-term top. When speculators see gold moving by 3%+ interday or 2%+ a day for a few days in a row, they should be aware that probabilities favor an imminent gold correction. Big gold moves, unlike the stock markets, tend to be a topping thing rather than a bottoming omen.
On the other side of the volatility pendulum there have been nine gold volatility lows in this bull to date. After each one of these gold rose, sometimes a great deal and sometimes not so much. Speculators would have been well served to buy at all of these vola lows except one, episode 6. The other eight of nine marked times when gold was priced below where it would soon be and hence a good time to throw long.
Vola low 6 occurred in late 2003 when gold was challenging $425 for the first time in this bull market. For some reason a fairly volatile and exciting gold rally just hit cruise control briefly near the end of the year. Gold traded in a tight range for a short period of time while taking a breather. During that odd period of calm volatility fell off a cliff and actually hit zero 1%+, 2%+, or 3%+ days per rolling month for two consecutive mid-December trading days. But after the holidays, during which speculators are distracted, gold soon resumed its normal rally volatility profile.
With the exception of 6, six of the seven low-vola events prior to 2005 marked excellent times for speculators to throw long to ride anything from a sharp spike to a respectable rally to a major upleg. Once again these odds are pretty good and something we should definitely pay attention to moving forward. When gold volatility is abnormally low it is not the time to sell like in stocks, but instead it is the time to throw long and ride the gold price higher.
And today, which is very obvious in this chart, gold is languishing at the lowest general volatility levels of its bull to date. It has bounced near support a few times this year and a couple low-vola signals have flashed. In light of gold's consistent bull-to-date precedent, odds are this major low-volatility episode will mark the calm before the excitement of another major upleg.
Since the markets are like a giant pendulum in sentiment and volatility terms, extreme lows are usually followed by counterbalancing extreme highs which would necessitate a major gold rally and possibly a short-covering feeding frenzy. Gold's next upleg, if it indeed rebalances gold's volatility signature, could be quite spectacular.
If you believe gold is in a secular bull market for fundamental supply and demand reasons, mirroring the general commodities bull, and you think gold volatility technicals will continue to be consistent, then today's low volatility scene demands heavy long exposure to the gold market.
At Zeal we are very bullish on gold today for a variety of technical reasons, including a topping dollar, dazzling euro-gold breakout, and now gold's abnormally low volatility. We are in the process of layering in elite unhedged gold stocks that are almost certain to thrive when gold starts moving higher in earnest. Our acclaimed monthly newsletter, Zeal Intelligence, outlines all of our favorite stocks. Please join us today and reap the rewards!
The bottom line is gold's volatility signature has been inverted like silver's in this bull to date so far. Instead of the conventional stock world where vola highs signal stock lows and vice versa, in gold vola highs often accompany major interim tops. Even if not readily explainable, gold has been profitably tradable on 6/7th of these inverted volatility buy and sell signals so far. These are good odds in the capricious markets.
Until there is clear technical evidence otherwise, I suspect speculators will do well by trading gold's volatility profile. Be wary of an interim top when gold waxes too volatile and get excited about an interim bottom when it seems too calm, like today. And trade accordingly.
Adam Hamilton, CPA
Thoughts, comments, or flames? Fire away at email@example.com. Due to my staggering and perpetually increasing e-mail load, I regret that I am not able to respond to comments personally. I will read all messages though and really appreciate your feedback!