HUI Bull Seasonals
As the nights lengthen, the leaves change color, and the chill winds of autumn begin to blow, the seasons are on everyone's mind this time of year. But it is not only these natural seasons driven by orbital mechanics that are changing. The most bullish seasonal time of the year for the precious metals and their miners is nearly upon us.
The mere fact that precious metals have seasonal tendencies is often surprising to traders. Everyone can understand why a soft commodity like wheat is seasonal. Due to the Earth's axial tilt and its annual revolution around the sun, there is one primary growing season in the northern hemisphere. Thus wheat supplies typically peak just after harvest before shrinking until the next harvest. Since the celestial seasons affect supply, and supply and demand drives prices, the Earth's seasons play a major role in wheat price trends.
Interestingly it is these same orbital mechanics that drive gold seasonality. The vast majority of the world's population lives in the northern hemisphere, so the importance of the autumn grain harvest is universal. In places like Asia with deep cultural affinities for gold, farmers often invest some of the profits from their annual harvests in gold. Harvest leads to big global surpluses of capital and some of these funds migrate into gold.
There are other cultural factors that also accentuate gold's seasonal strength this time of year. Indian wedding season is one of the most important. In India, the world's biggest gold consumer, little distinction is made between jewelry and investment. When brides get married, their families give them intricate 22-karat gold jewelry to help secure their financial futures. The most popular time to get married is during the autumn festival season. Thus around 40% of India's total annual gold demand tends to arise in October and November.
I realize this seems quaint to our Western minds, but we too have similar behaviors woven deeply into our own cultures. Our own holiday shopping isn't all that different. A large fraction of our total spending occurs between Thanksgiving and Christmas, a single month that often makes or breaks the entire year for retailers. Spending is high worldwide as the year wanes and people start feeling closure on their financial year. Even in the West, some of this holiday spending funnels into gold jewelry for loved ones and bullion for investment.
So world gold demand is indeed highly seasonal, for a variety of reasons. Understanding this is important as we traders can increase our odds for success if we trade with these seasonal trends. But do these gold seasonals affect the gold stocks? Each time I write about gold seasonality I receive a blizzard of e-mails asking how pure HUI seasonality looks. Does this flagship unhedged-gold-stock index follow gold's seasonal lead?
In order to address this excellent question, I applied the same methodology I have used for gold seasonals to the HUI itself. This was described in depth in my latest essay on gold seasonals, but here are the highlights. Unlike typical multi-decade futures seasonality studies that span bull and bear alike, I am only interested in how HUI seasonality has unfolded within this bull market. Past behavior within bears is probably not all that relevant to future behavior within this bull. So all of these charts only extend back to 2000 when today's bull was born.
To build these charts, each calendar year's daily closing data was individually indexed. The first trading day of each year was assigned a value of 100 and every subsequent day of that year was indexed off of it based on absolute percentage gain. All of these individual annual indexes are then date-matched and averaged together. The result, when plotted, shows the seasonal tendencies of the HUI within its bull market to date.
Since the average of annual indexing doesn't show how dispersed the underlying data actually is, standard-deviation bands are also rendered. The tighter the bands, the closer the underlying annual indexed numbers were before they were averaged. The closer the pre-average data, the higher the probability the seasonality in that region of the chart will stay relevant going forward. Closer means it isn't just a statistical anomaly created by a couple of extreme outlying years. The small inset charts show the full standard-deviation bands.
While this essay is on the HUI bull seasonals, we still need to start with gold bull seasonals since the gold price is the primary driver of the gold stocks. This chart is updated from my July essay on gold seasonality. The primary question that led to this thread of research was whether gold stocks tend to follow gold's seasonal lead. So before we get into HUI-specific seasonals, we must first have a clear picture of gold's.
Gold tends to have three big seasonal rallies every year. Provocatively for traders today, its largest by far tends to start in mid-October and then power higher into early February. This reflects the fabled autumn buying season for gold. General Asian demand, Indian wedding season, Western holiday buying, and other regional factors lead to a major surge in gold demand and often gold prices this time of year.
Gold traders know this well and usually really add to their long positions in anticipation of this tendency. While gold seasonality certainly doesn't guarantee the gold price will comply every year, it sure increases the odds. Seasonality is like a tailwind on top of the other fundamental, technical, and sentimental factors driving gold. When these other factors are already bullish, strong seasonality ramps up the probabilities of imminent gains even further.
This chart has been especially intriguing lately. Note above that gold tends to start rallying in late August, climb even higher in the first half of September, and then really shoot higher in the second half of September. Sound familiar? This was exactly what happened over the last six weeks or so. But since this chart includes data to the end of September 2007, perhaps gold's awesome performance lately unduly skewed this chart.
Thankfully this is easy to test. In my July essay on gold seasonals, the data only runs to the end of June. Check out the first chart in that summer essay. Although a bit less extreme than the chart above, gold still had the exact same tendency to do what it did from mid-August to late September. Its seasonal patterns then and now are virtually identical. So prior to September 2007, gold already had a well-established tendency to rally modestly in the first half of September and then shoot higher in the second half.
This ought to disturb you as it calls into question today's orthodox perceptions of last month's gold trading action. Almost everyone today assumes that gold rallied because the Fed cut rates while the dollar hovered on the edge of the abyss. And I agree that these factors are almost certainly the primary causes of gold's excellent month. But even last summer, the gold seasonality chart already showed a crystal-clear tendency for a very similar September price pattern to occur even without any Fed machinations.
So would gold have rallied modestly initially last month and then shot higher in the second half even without the Fed? Probably, as it sure has the tendency to do just that regardless of the Fed. Perhaps traders today should be attributing more of gold's September to usual seasonal buying and less to the Fed's throwing the dollar to the wolves. Maybe all the Fed really did for gold was modestly amplify already established seasonality.
Another interesting revelation from the July chart, which is even more pronounced here, is gold's seasonal tendency to pull back rather sharply in early October. This, of course, is exactly what we witnessed this week. The déjà vu is pretty uncanny. After this initial sharp pullback, gold gradually grinds lower and consolidates for a week or two. This early October consolidation is extremely important because it offers traders our last chance to load up on long positions ahead of the biggest seasonal rally of the year.
And gold's tendencies right now are even more relevant because the standard-deviation bands of its seasonality are fairly tight right now. It wasn't a couple of extreme annual results that distilled down to today's seasonality, but seven individually-indexed years with a rather narrow clustering of indexed levels this time of year. This renders today's narrow window of time in which to add long positions all the more important.
So back to our original question, does gold seasonality drive HUI seasonality? Absolutely! Take one more look at the gold chart above and then quickly scroll to this HUI specimen. The HUI's big seasonal rallies, as well as its seasonal weak spells, mirror gold's incredibly well. If I cut off the left axes that map the magnitude of these indexed moves, these two charts would be virtually indistinguishable to the casual eye.
When comparing these charts, realize that the HUI's goes to 135 indexed while gold's only goes to 114. So while they look very similar, with major peaks and troughs throughout a typical seasonal year matching closely, the HUI amplifies gold's volatility considerably. For example, in January and February the HUI tends to go from 100 to 108 indexed while gold only tends to move from 100 to 103.
This HUI leverage to gold is the only reason why anyone invests in gold stocks. If the far-riskier gold stocks only tended to pace gold's gains, then it would make a lot more sense to own the much-less-risky physical metal itself. But since gold stocks have amplified gold's gains on the order of 5.3 to 1 in their respective bulls to date, betting on the stocks is well worth their additional risk.
Mirroring gold, the HUI also has three big seasonal rallies each year. In indexed terms, the one ending in May tends to rise 16 points and the one ending in September 18 points. But it is the third, and largest one, that is most interesting today. It tends to start powering higher in mid-October and climb up 25 indexed points by February! It shouldn't be surprising that the biggest seasonal HUI rally of the year follows the biggest gold rally.
It is gold that drives the gold stocks. Higher gold prices mean higher profits for mining the metal. Higher profits ultimately translate into higher stock prices. Since the price of gold in a secular bull tends to rise faster than the costs of mining the metal, even during a commodities bull, it is the gold price that has the single largest impact on worldwide gold-mining profits. So if gold is going to be seasonally strong, the HUI should benefit tremendously as traders anticipate higher profits.
Now this seems simple and obvious, but an increasingly popular heresy disputes the truth in these charts. Due to a few isolated episodes in 2007 where gold stocks fell with the general stock markets, many PM traders believe that it is the general stock markets that drive the HUI, not gold. It is now fashionable to believe that no matter what gold does, if general stocks enter a bear market the HUI will be dragged down with them.
This concern is not new, and I have researched and written a great deal about it over the last seven years. While the HUI can certainly fall with general stocks for a few days during aggressive high-fear selling, overall it thrives through stock bears. From early 2000 to early 2003, the S&P 500 lost 49% of its value at worst. Meanwhile the HUI soared 322% higher at best within this bear. The HUI even rose within each of its three most vicious downlegs. PM stocks are classic alternative investments not correlated with general stocks.
These HUI seasonal charts help illuminate this general-stock-bear concern from another angle. The three bear years in the early 2000s that slaughtered general stocks are also included to arrive at this seasonal average. Thus these HUI bull seasonals transcend general-stock bulls and bears alike. While gold stocks can be temporarily distracted from time to time, they always follow gold in the end since it drives their profits.
In my gold seasonality studies, I also take a look at calendar-month seasonality. Instead of indexing calendar years, calendar months are indexed. Each month of each year is started at a level of 100 with the rest of the days indexed off of it. Then all the Januaries are averaged together, all the Februaries, etc. I was curious on how the HUI tends to perform within calendar months in this bull, so I did the same analysis here. Realize that each calendar month is a discrete individually-indexed unit, so one month does not connect to the next.
This additional perspective on seasonality is interesting. It closely follows the annual-indexed approach of course, but by distilling the data in a different way it also illuminates additional seasonal tendencies. The HUI's best calendar months of the year in its bull to date have tended to be August, November, and May. And November is rapidly approaching, a great reason to deploy more capital on this October weakness.
Now when I first saw this chart this week, I had to chuckle at August being the strongest month in this bull to date. On average, the HUI has soared almost 8% in Augusts since 2000. November and May have come close to this performance on average, but August still wins out. Obviously this past August didn't cooperate though. The index fell 5% in a month that was marked by a brutal mini-panic in the middle.
This discrepancy between what was expected and what actually happened helps illustrate some of the key limitations of seasonality for traders. Seasonality is simply a tendency, a bias for a price to move in a particular direction at a particular time of the year. But seasonality isn't always a driving factor. Technicals and sentiment, especially near extremes, can easily override seasonality and drag the HUI off of its expected seasonal vector.
August 2007 was a really unique month. Sentiment for gold stocks was horribly bad after they had consolidated for 15 months straight. In the middle of the month worries about the general stock markets temporarily spilled over into gold stocks and drove a mini-panic as many traders capitulated. For our subscribers, I explained this whole chain of events in great depth in the September issue of Zeal Intelligence.
Fear soon got out of hand as intense selling dominated gold-stock prices and they plummeted. This easily overwhelmed the usual strong August seasonals. But check out July in this chart. July tends to look like August did this year. Yet in 2007 July witnessed a 5% gain in the HUI. The usual mid-summer selling in the HUI that tends to hit in July came a month later this year in August. And after that the usual subsequent rally was compressed into September. This helps to illustrate just how elastic seasonal tendencies can be.
All traders who consider seasonality in their decisions would do well to ponder this. Seasonality doesn't offer precise timing, just general probabilities. So while gold and the HUI tend to start rallying strongly again in mid-October, if their rallies start a week or two early or late it doesn't negate the seasonal tendencies. Since seasonals aren't precise and they are so easily overridden by technicals and sentiment, I think seasonals should never be used as a primary trading tool.
But seasonals excel at increasing the odds for success of trades made for fundamental, technical, and sentimental reasons. Gold continues to be a great fundamental buy today for a wide range of reasons, including the struggling US dollar. In real inflation-adjusted terms, its price isn't even close to looking long-term overbought yet. And it hasn't yet stretched far enough above its 200-day moving average to signal the end of this upleg. Gold's sentiment is certainly not euphoric either since few have been really excited about it since May 2006. Together these factors might give us a 75% chance for success for a long trade today.
But when bullish seasonals are added on top of these primary drivers, they create an additional probability tailwind. With gold's seasonal tendency to start its strongest rally of the year in the next couple of weeks, perhaps today's probability for success rises to 85%. Although these absolute numbers are guesses, the relatively small impact of seasonality compared with that of primary drivers is not. So please realize seasonality is only relevant as a secondary indicator if primary indicators are already lining up to drive a trade.
At Zeal we are constantly studying gold and the HUI from fundamental, technical, and sentimental perspectives. Today the metal and stocks look to be in a very bullish position due to these primary drivers regardless of seasonality. But with both gold and HUI seasonality shifting massively bullish in the coming weeks as well, our odds for success on long trades climb even higher. These seasonal tailwinds are very welcome.
So in response to all the bullish fundamental, technical, sentimental, and seasonal factors that are converging today to launch gold and its miners much higher, we are aggressively adding positions in elite PM stocks. If seasonals prove true to their bull-to-date precedent, we have a narrow window here during this early-October consolidation to deploy ahead of the next upleg. If you want to join us in this probable highly-profitable ride higher, please subscribe today to our acclaimed monthly newsletter and mirror our new trades.
The bottom line is HUI bull seasonals track gold bull seasonals closely, which is no surprise for students of the markets. While not a precise primary indicator for upcoming HUI performance, when seasonals match with primary indicators the odds for success in new trades rise considerably. Today we are in such a situation, when bullish HUI seasonals match and buttress bullish HUI fundamentals, technicals, and sentiment.
Provocatively both gold and HUI seasonals expected this week's sharp pullback in early October followed by a couple of weeks of grinding consolidation. But once this short-lived buying opportunity ends, the seasonals project the start of the strongest gold and HUI rallies of the year. So if you have been waiting to add new PM-related trades, you should consider seizing this narrow window of opportunity.
Adam Hamilton, CPA
Oct 5, 2007
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!