Contributed Commentaries
Will silver rise again?
Kitco Commentaries | Opinions, Ideas and Markets Talk
Featuring views and opinions written by market professionals, not staff journalists.
Monitoring gold-to-silver ratio stability provides an important clue
Sept 9, 2020
My July 27 Kitco commentary “Is Silver Near a Top?” posited an affirmative answer and the white metal made a new closing high on August 10. The column also presented a technique and criterion for confirming that the top would hold, at least for the near term. That condition was also satisfied and there have been no new highs for nearly a month. It is now reasonable to ask whether there will be a new high (or low) for silver for the remainder of the year. As explained below, the same technique applied to recent price action suggests there may indeed be more drama ahead for silver.
Gold and silver have already enjoyed an extraordinary year. On an intraday basis, Comex gold advanced 43% from its mid-March bottom to score an all-time high above the mercurial $2,000 per troy ounce in August. Silver rallied a stunning 154% to make its August high. Greater than three-times performance of the white metal compared to the yellow is reminiscent of 2011 when silver made a mad dash at $50 per troy-ounce in April of that year. Silver stands on the shoulders of gold when they both reach for the sky but can easily tumble down a mineshaft when it loses its balance.
The 3-month beta of silver relative to gold rose from less than unity in mid-May to over 3.0. This implies that silver gains are significantly amplified with respect to its precious metal cousin as evidenced by the performance from the March lows. The converse is also true: a downturn in gold price could be accompanied by a serious correction for silver. Do recent moves in beta foreshadow silver’s fortunes in the coming months?
Gold-to-silver ratio stability
As explained in the prior commentary, monitoring the stability of the gold-to-silver ratio (GSR) is a powerful technique for confirming ratio and price local extrema. A measure of stability is the volatility of the ratio over two different time periods. Volatility for this analysis is taken to be the standard deviation of the ratio divided by its mean over a given time expressed as a percent. For gold ratios, I have found 3- and 1-month periods to be useful and consider volatilities exceeding 4% to represent "instability."
As shown in Figure 1, plotting the 3- and 1-month volatilities (Y-axis and X-axis respectively) results in a stability lambda-Map©. Connecting the market day-to-day points produces a "stability trajectory." By this method, the lower-left quadrant is the domain of "very stable" ratios; and the upper-right, "unstable" ratios. The other two quadrants are considered transitional.
Figure 1 – Gold-to-Silver Ratio (GSR) Stability lambda-Map©
KEY INDICATION: Major trajectory turns in the unstable quadrant typically occur when ratio and price are near their respective extremum. These highs (lows) generally hold through a period of ratio stabilization (typically weeks or months). Major turns are a confirming indicator - anticipatory, coincident or lagging but usually within several market days of the high (low).
Figure 1 illustrates three examples that have occurred this year. Each extremum (E) and the associated GSR and silver price are given in the table in the bottom-right corner. The trajectory starts February 14 during a quiescent time in the very stable quadrant. On March 11, The World Health Organization declares covid-19 a pandemic as the trajectory rapidly enters the unstable quadrant (bottom red arrow, volatilities>4%). Silver price makes a low and the GSR scores a high March 18 (E1) followed by a major turn three (+3) market days later March 23.
The ratio begins to stabilize as it enters the upper-left transitional quadrant (green arrow, second from the top) but eventually returns to the unstable region. A second major turn occurs June 1 coincident with a silver high and GSR low (E2). Re-stabilization follows (note July 7) and then it is off to the races again (top red arrow).
By the July Kitco commentary, the trajectory was still in the unstable quadrant and appeared to be decelerating to initiate a major turn. That did not happen. The failed turn is indicated by the red dotted ellipse. On August 10, a new GSR low and silver price high of $29.26 per ounce (E3) occurred one (+1) market day before the third major turn.
It is noteworthy that the GSR low at the third extremum is remarkably close to its 10-year average (70.2 versus 69.5 ounce per ounce). During the March market panic the ratio hit a peak of 124. Moving down so swiftly means silver rapidly gained relative value compared to gold. This breaks a long-term trend of gold gaining value on silver established from the April 2011 low (GSR=31) – a very bullish indication for the white metal.
By Friday’s close the trajectory finds itself in the stable left-half plane and decelerating again (blue arrow). What happens next? There are three main cases to consider:
Case #1 - The short-term stability of the GSR persists for some time. Presently the 1-month volatility is almost currency like at 2%. This will eventually erode the high 3-month volatility (13%) and the direction of the trajectory will slowly direct toward the very stable quadrant. In this scenario, it is reasonable to expect no new highs or lows for silver in 2020.
Case # 2 - Volatility could return to the precious metals as the trajectory re-enters the unstable quadrant for a fourth try at a major turn this year. Silver might plunge to a new low during a market liquidation like March or just be caught up in a significant market rotation away from safe havens.
Case #3 - The last case exhibits a similar trajectory and volatility as case #2 except gold and silver post new highs.
We can find clues about which of the three cases is more likely by examining the silver-to-gold beta.
Silver-to-gold beta
The silver-to-gold beta tells you how gains (losses) in silver are related to gains (losses) in gold. You prefer a high beta (i.e. greater than 1.0) when gold rallies. For example, a beta of 2.0 implies a 1% bump in gold price should yield a 2% jump for silver.
Beta is the product of correlation and volatility. The silver-to-gold beta is the correlation of silver and gold times their relative volatility over a given time period. Typically silver and gold move together in price so their correlation is positive and near 1.0. For this precious metal pair, beta is therefore mostly a measure of comparative volatility.
For example, on August 10 when silver made its high the correlation of silver and gold was +0.96 on a 1-month basis. The silver volatility over the same time period was a high 13%; and gold, a lower 5%. Silver was therefore 2.6 times more volatile than gold. Multiplying that number by the correlation results in a beta of +2.5. In other words, a 1% move in gold yields a 2.5% increase in silver based on a 1-month daily price record. The 3-month beta on that date was also a high at +2.8.
Since GSR stability is also related to volatility over similar time periods, it is reasonable to expect the notions of beta and ratio stability are closely related. Figure 2 illustrates that relation by plotting the percentage change in the GSR (Y-axis) for a percentage change in gold price (X-axis) given a range of betas (0.0 to 8.0). The Figure 2 inset summarizes the relevant equation set.
Figure 2 – Gold-to-Silver Ratio (GSR) sensitivity & potential for instability
If beta=1.0 (blue trace), there is no change in the GSR - the percentage price changes in gold and silver are equal. This is the most stable condition for the ratio. For betas less than one, the GSR increases with a positive change in gold but by a smaller absolute amount (i.e. silver loses value to gold). For example, a beta of 0.5 (yellow trace) shows that a +1% change in gold results in a +0.5% change in ratio.
The GSR ratio is increasingly more sensitive to changes in gold price for betas greater than 1.0. and thereby has an increasing potential for instability. For this condition, the GSR decreases with a positive change in gold (i.e. silver gains value on gold) and the percentage decrease grows with increasing beta. For example, a +1% change in gold produces a 3% decline in ratio for beta = 4.0 (gray trace); 6.5% for beta = 8.0 (green trace). A sudden rise in gold price volatility can cause exceptionally large swings in silver price resulting in GSR instability. The August 10 example is one such case where the GSR stability trajectory is far into the unstable quadrant (Figure 1).
Silver-to-gold beta-Map©
Since beta is a good indication of ratio stability, a silver-to-gold beta-Map© is helpful in assessing stability direction (i.e. improving or decaying stability). Figure 3 is such a map, plotting the 3- versus the 1-month beta (Y-axis and X-axis respectively) from July 7, 2020 to last Friday’s close.
.jpg)
Figure 3 – Silver-to-Gold beta-Map©
After July 7, the trajectory of Figure 1 quickly moved from the stable left-half plane (LHP) to the unstable quadrant. In Figure 3, we note that change is coincident with rapidly increasing beta to a peak in both time bases July 22. Beta declines some as silver makes its high August 10, followed by the third major turn August 11. As previously noted, the stability trajectory returns to the stable LHP as the short-term beta briefly falls below 1.0.
Importantly, the beta trajectory is now pointed towards increasing short- and longer-term betas greater than one.
Will silver rise again?
We can now assess the three cases given the above analysis. If short-term GSR stability persists (case #1), Comex gold will remain in a trading range as headlines ebb and flow into the coming Presidential election. The limits of that range are the highs and lows since mid-August: $2,025 (8/18) and $1,874 per ounce (8/12). For this case, the 1-month silver-to-gold beta remains near 1.0. The GSR is accordingly stable and holds close to the present 1-month average or about 72 ounce per ounce. Applying that number to the gold range gives a high/low range for silver of $28.1 to $26.0 per ounce.
Presently, this is the least likely case as suggested by Figure 3 which shows a migration away from beta=1.0 as noted. The stability trajectory of Figure 1 is also showing early signs of deceleration and turn back to the unstable quadrant. This suggests volatility will return to precious metals and an increasing potential for ratio instability – the necessary environment for new lows or highs.
Whether this manifests in case #2 (new lows) or case #3 (new highs) depends on the outlook for the remainder of the year. For instance, implementing a safe and efficacious vaccine before the end of the year could spark a significant rotation away from safe havens pressuring gold and silver to break below the ranges given for case #1. A 10% correction from gold’s all-time closing high puts prices in mid-$1,800 territory, there is July support at $1,820. Since this would likely occur in high-beta territory, silver could fare far worse. A 30% downdraft from the August closing high is still above $20; greater than that brings prices to June support around $17. Revisiting the March low would probably require liquidation associated with a financial crisis, a quite unlikely scenario for 2020.
I believe case #3 to be the most likely scenario. Gold and silver prices will rise again before the new year. The current environment favors this choice: 1) there remains plenty of domestic and geo-political uncertainty to support safe havens: U.S. elections, a stalled recovery package in Congress, a stubbornly pervasive covid-19 virus, civil unrest and escalating U.S./China tensions, 2) the fortunes of the U.S. dollar grow dimmer with more expected Federal Reserve and U.S. Treasury largesse, and 3) negative U.S. benchmark real rates continue - bullish for noninterest earning assets like gold and silver.
For gold, the 2011 record brought forward in time by inflation is a reasonable target or about $2,200 per ounce. The GSR will likely fall back to the 10-year average of 69.5 or lower. Applying the average to the gold target results in a silver price of $32 per ounce. The inflation-adjusted silver high of 2011 is $57 which I believe is a bridge too far for 2020.
Please follow me on twitter @Eurekaminer for updates of the above graphs. When the fourth major turn occurs, that is probably it for silver – at least for a while.
Cheers!
 |
