Gold and Dollar Decoupling
Occasionally in the financial markets an event happens that generates enough interest to transcend the usual sector boundaries. It captures the attention of contrarians and mainstreamers alike, and leads to universal discussions on its implications. The recent sharp slide of the US dollar is looking like one of these events.
While the dollar has hit record lows against some currencies, the metric of choice for following the dollar is generally still the US Dollar Index. Several decades old, it is a geometrically-weighted average of the dollar's exchange rate with six major world currencies. It is dominated by the euro, which accounts for about 58% of its weight. Next comes the yen near 14%, the British pound around 12%, and the Canadian dollar near 9%.
Since mid-October, the USDX has plunged by 5.4%. Fully two-thirds of this sharp decline occurred between mid-November and this past Monday. Now for stock traders who can sometimes see 5% swings in the opening minutes of a trading day this doesn't sound particularly impressive. But for currencies that typically move with glacial slowness, the dollar has seemingly just plummeted over a steep cliff.
This sharp decline is all the more noteworthy since the US Dollar Index is geometrically averaged. As traders of the pre-July-2005 CRB Commodities Index remember, the math under geometric averaging aggressively smoothes out underlying component volatility. So the dollar really has to get hit hard in most of the six currencies for its index to slide as sharply as it has in recent weeks.
The implications of this dollar slide are legion and many essays could be written barely scratching the surface of discussing them all. As a gold investor and speculator though there is one in particular that I find exceptionally provocative today. All over the contrarian world in recent weeks, people are ascribing gold's recent strength to dollar weakness. While no doubt a material factor, we are no longer in the purely mechanical dollar-weak-equals-gold-strong world of a few years ago.
Interestingly, gold's potential is "sold short" in a proverbial sense by relapsing into the old gold and dollar paradigm that defined the initial years of gold's current secular bull. Back then, gold was only strong when the dollar was weak. Then the dollar was indeed the primary driver of gold. But over the last 20 months or so, gold has increasingly been decoupling from the dollar. Now pure investment demand is gold's primary driver.
This may seem like a trivial distinction at first glance, but it is not. If gold is still dominated by the dollar, then the only way that this gold bull can continue is if the dollar bear keeps spiraling lower. Of course like all fiat-currency experiments in world history the US dollar is ultimately trending towards its true value of zero, but this process will probably take many decades. After all, it took many centuries in ancient Rome for its own empire-ending currency debasement to fully run its course.
Believing that the dollar is still the key to gold is no longer technically correct as I will illustrate below. Understanding this could have major psychological implications for gold investors and speculators. If they still believe the dollar is the key, they could risk getting discouraged and selling out far too early because the dollar happens to bounce in a bear-market rally. But if they realize it is investment demand, not the dollar, driving gold, then they are freed from the tyranny of cowering each time the dollar flexes its muscles.
Despite the recent strong negative correlation between gold and the dollar that conjures up memories of years past, gold and the dollar are either in the process of decoupling or essentially decoupled. Investors are now buying gold around the world for its own merit. Gold's own independent supply and demand is driving it today and the relentlessly inflating US dollar has been relegated to mere peripheral status.
This thesis is considered controversial, even heretical, among certain sects of gold investors. I would have a hard time believing it myself if I hadn't personally done the underlying research over the last couple years. But when you actually look at gold's price behavior compared to the dollar's over gold's bull to date, it is crystal clear that the dollar is fading in importance. Like it or not, gold and the dollar are really decoupling.
Before we get into the charts, it is essential to understand the concept of a decoupling. Several years ago gold and the dollar had a strong negative correlation. If one was up the other was down or vice versa. A decoupling doesn't mean the opposite, a prevailing positive correlation, but actually no correlation. In a decoupled no-correlation environment there will be times when both gold and the dollar are up, both are down, or they are moving in opposite directions.
The key to a no-correlation environment is none of these tactical correlation conditions will last for long. Since any price can only move up or down, there are only four combinations of how two prices can interact. They are up up, down down, up down, and down up. So even assuming these are randomly distributed over time, in a no-correlation environment the dollar and gold could still move in traditional mirrored opposition 50% of the time (up down and down up, two of the four possibilities).
As this first chart of the US Dollar Index and gold shows, this decoupling probably started in Q2 2005. I was studying it back then when it happened and have written quite a bit on it since. While the decoupling wasn't as clear at first, the deeper we march into this gold bull the more readily apparent it is becoming. The decoupling is rendered on this chart as the dotted-yellow line at the beginning of Q2 2005.
While the gold bull was stealthily born in April 2001, the beginning of the parallel dollar bear is not as well-defined. The once mighty dollar initially topped in the summer of 2001 but recovered to carve a double-top in early 2002. As such, most technicians including me tend to see the dollar bear as officially starting on January 31st, 2002, the last time it closed above 120. Since then it has been a long grind lower.
With the world's reserve currency having a vastly larger global market than gold, it is useful to view the early years of our gold bull through the lens of the dollar bear. From 2002 to Q1 2005, seemingly the only time when gold could catch a bid was when the dollar was weak. This strong negative correlation led to the widespread belief today that dollar weakness is still essential in order to see gold strength.
During those initial several years, gold carved five major interim highs as its bull market gradually clawed higher. They are all numbered above. The interesting thing to note is that every one of these major interim gold highs, without exception, occurred right near the end of a long slide in the US Dollar Index. Thus gold uplegs only happened during dollar downlegs. And during dollar bear-market rallies gold subserviently corrected.
This powerful negative correlation is very evident visually as well as mathematically. On the visual side, check out the precisely mirrored price patterns of gold and the dollar until Q2 2005. This tendency was so strong and so ironclad in these years that successful gold trading systems often watched the dollar to know when to trade gold. Not only were price patterns mirrored, but the opposing moves were proportional and the interim extremes were roughly synchronized. Gold made highs near dollar lows and vice versa.
Mathematically this correlation was every bit as strong as it is visually. The daily closes of the US Dollar Index and gold had an utterly massive negative correlation of -0.956 up until Q2 2005. Squaring a correlation gives an r-square value, which statisticians use to explore potential causation. Until the decoupling, gold and the dollar ran a stellar r-square of 91.4%! Thus over 91% of the daily price action in gold could be statistically explained by and/or attributed to the dollar! The dollar truly was the primary driver of the gold price.
But in Q2 2005 a peculiar thing happened. The dollar surged but gold held its own. It kind of reminded me of a boxing match where a beaten-up underdog cowers through several rounds of brutal punches from the champion but then in round four the underdog suddenly stands up and blasts the champion in the jaw. In Q2 2005, for the first time in this bull, gold was holding its own. Gold stood up to its dollar dominator and scoffed.
Now this event was anticipated, we were looking for it in advance at Zeal and were ecstatic to finally see it arrive. Great gold bulls have three stages. The first third or so is currency-devaluation driven, and indeed this happened in our bull when gold only rose when the dollar fell. But this currency devaluation merely primes the pump for the much larger second stage, when gold rises on its own intrinsic merits on ever-increasing global investment demand. In Stage Two gold decouples from the dominant currency and its bull really starts thriving.
Before this decoupling I was trying to figure out how to know when it arrived, and after much research I decided on watching gold denominated in euros for the most likely sign. Euro gold had challenged €350 and failed to break above it for several years in a row. So I figured when euro gold finally broke €350 it would unleash a surge of gold investment demand from old pro-gold European money and would ignite Stage Two where gold decouples from the dollar. This awesome €350 breakout happened in Q2 2005.
The €350 breakout was so crucial because until that point non-American investors largely believed the so-called gold bull was really just a dollar bear. Gold was only moving locally in dollar terms as it responded to a dollar devaluation. But when euro gold broke out and started to carve new highs, they were forced to acknowledge this bull as the real deal. European (and global) capital starting bidding on gold and this marginal demand caused it to rise despite the dollar. This created a virtuous circle where more gold demand created a stronger gold-dollar decoupling enticing in new investors to buy ever-more gold.
So the Stage Two transition started in mid-2005 and has only gathered steam since. There is all kinds of evidence that the gold bull since Q2 2005 is radically different than the one before the decoupling. Visually it is readily apparent that gold's latest massive upleg that ended in May was an entirely new beast compared to its earlier comparatively anemic bull-to-date uplegs. In six months gold soared 54% while the dollar merely fell 8%. The previous years' proportional opposing moves had totally vaporized.
In addition, note that the biggest upleg of the bull to that time, the one that led to interim high 6 earlier this year, happened during the biggest dollar bear rally of its entire bear. By the time gold surged through $550 the dollar was actually in a minor pullback in a major bear rally, not at the end of a long downleg as it had been near all previous major interim gold highs. The dollar was thankfully losing its influence over the gold bull.
But the most telling evidence for gold's decoupling from the dollar is not visual but mathematical. From April 1st, 2005 until this week, the daily correlation between the dollar and gold plummeted to a mere -0.400. This is a breathtaking decrease. The r-square of this is a trivial 16%, not at all correlated. Prior to Q2 2005, 91% of the daily moves in gold could be explained by opposing dollar moves, but since then only 16% of gold's moves are explainable by or statistically attributable to the dollar.
Now as a life-long speculator I effectively gamble for a living, I love risk more than most folks love oxygen. While I would not hesitate to bet when my odds for success are 91% in my favor, I wouldn't bet in a million years if I only had a 16% chance of winning. The old dollar-weak-gold-strong thesis of past years was very true a few years ago, but this thesis is no longer valid. Hence it isn't wise to trade on it today. We are now in a brave new Stage Two world where gold's supply and demand is independent of the dollar's machinations. Gold and the dollar have decoupled!
The moral of this story is don't get too eager to ascribe all gold's strength of recent months to dollar weakness. While a falling dollar does get more investors interested in gold and hence probably drives indirect gold demand, gold is trading independently of its old nemesis these days in Stage Two.
I am pretty convinced right now that the dollar could bounce and surge yet gold's new upleg would continue higher on balance with nary a worry. Gold just finished a necessary consolidation in early October so it is technically ready to rise regardless of the fortunes of the dollar. So there is really no reason for gold investors to get particularly excited about a dollar downleg in Stage Two nor get worried about the consequences when the dollar inevitably bounces in its next major bear-market rally.
Our most successful trading tool for gaming gold back in Stage One that netted us enormous realized profits in gold stocks in the early years was a comparison of gold and the dollar relative to their respective 200-day moving averages. Per my Relativity trading theory, dividing each by its own 200dma creates horizontal trading bands. In effect the 200dmas of gold and the dollar are flattened to horizontal and each price is charted over time as a continuously comparable multiple of its 200dma. You can see this in the next chart.
The decoupling seen above in the raw dollar and gold price data is even more striking in Relative Dollar and Relative Gold terms. For whatever reason in Q2 2005 the character and nature of our gold bull radically changed and it hasn't looked back since. The once-king dollar-dominated-gold paradigm is no longer valid. Gold is marching to the beat of its own supply/demand drummer now.
The blue line is gold expressed as a constant multiple of its 200dma, and as you can see it largely traded in a horizontal band until the decoupling in early 2005. For trading gold in Stage One, we were using an rGold range of 0.99 to 1.14, also rendered above. When gold was low in this range we expected a major upleg and when it was high we prepared for a major correction. Bull markets surge above their 200dmas in uplegs and then retreat back to them in corrections.
The red line is the rDollar, the US Dollar Index divided by its own 200dma. Since the dollar was in a bear market its range is below the 1.00 line that indicates its 200dma. In bears prices plunge below their 200dmas in downlegs before surging back up to them in bear-market rallies. We were watching an rDollar range of 0.92 to 1.00 during Stage One. When the dollar was low in this range a major bear-market rally was due and when it was high in this range another downleg was likely approaching.
Now considering rGold and the rDollar together, during Stage One they had mirrored price patterns, proportional opposing moves, and largely synchronized extremes. Both diverged away from their 200dmas (uplegs in gold, downlegs in the dollar) at the same time and then converged back to them (corrections in gold, bear rallies in the dollar) at the same time as well. Gold was indeed slaved to the dollar then as Relativity so vividly illustrates.
But in Q2 2005 a strange thing happened. The dollar didn't only surge up in a normal bear rally to top near its 200dma, but it blasted well above its 200dma as if it was entering a new bull. It was the highest the dollar had been relative to its 200dma in its entire bear. If there was anything that should have scared Stage One gold investors, the possibility of the end of the dollar bear was it. Yet rather than plunge in fear in Q2 2005, gold held its own despite unprecedented-within-this-bear dollar strength. It decoupled.
In late Q3 2005 gold surged right when the dollar was rallying higher itself, an event that never would have happened back in Stage One. And while rGold and the rDollar have still had roughly opposing patterns since, it is very clear that the old lockstep inverted relationship simply no longer exists. Yes you can find episodes where the dollar is down and gold is up, but you can also see up-up times, down-down times, and opposing up-down times. This is no-correlation behavior, a stark contrast to Stage One's persistent negative-correlation realm.
I believe that understanding this new gold-dollar paradigm is very important for a couple reasons. First, the better we understand the markets the higher our odds are of successfully buying low and selling high. People who remain stuck in the obsolete dollar-dominated Stage One paradigm today risk making poor decisions if they are still operating under now-faulty Stage One logic. The dollar is no longer gold's primary driver.
Second, if gold investors continue to give the dollar more reverence than it is due they risk getting psychologically whipsawed as soon as the dollar inevitably bounces. Sure, it is fun to watch gold rise as the dollar falls. But when the dollar bounces is it a good decision to immediately sell all gold-related speculations? Almost certainly not! Gold is rising today because investors are buying it after a necessary consolidation. While the dollar may be a factor in some of these decisions, it no longer steers gold.
Since gold just apparently finished a necessary and expected correction in early October, and since gold finally retreated down to its 200dma from whence all major uplegs launch, I am very bullish on gold today. As the first Stage Two upleg earlier this year vividly illustrated, the gains to be won are getting much larger as more investors drive up prices. As such, we have been aggressively buying gold stocks in our newsletters in recent months.
As of this week some of our earlier trades, just a couple months old, were already up as high as 60%. If this is indeed the beginning of a major new upleg as I increasingly suspect, the gains so far are only the tip of the iceberg. If you want to leverage this probable gold move higher in elite gold stocks, please subscribe to our acclaimed monthly newsletter today. We are continuing to layer into high-potential gold stocks but this rare opportunity won't last for long.
The bottom line is gold has decoupled from the US dollar. While the dollar will ultimately migrate down to its true value of nothingness since it is backed by nothing but faith in Washington, gold is no longer dependent on that long slow slide. Investors around the world are increasingly discovering gold again and bidding it up for its own merits. The dollar really doesn't matter all that much anymore in a Stage Two gold bull.
Thus, it is probably prudent not to get too caught up in the latest dollar slide. While it certainly doesn't hurt gold, the last couple years have proven that gold doesn't need a weak dollar anymore to rocket higher on its own accord. Today gold investment demand is gold's primary driver, not dollar weakness.
Adam Hamilton, CPA
December 8, 2006
Thoughts, comments, or flames? Fire away at firstname.lastname@example.org. 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!