Is gold stretched?

 Posted by at 5:00 pm  Investing, Ratios
Jan 132016

Last week I wrote about the gold silver ratio as a way of determining which represents better value. Since then the ratio has moved higher, with gold outperforming silver on its move above $1,100. This has brought with it a number of bullish articles and while the move is encouraging and supports the idea that gold may have bottomed, in relative terms gold looks stretched to me at this time if we take a step back and look at the bigger picture.

First, consider gold relative to the other precious metals and oil. In the charts below a higher number indicates relative overvaluation. Only for palladium could you argue that gold is not stretched. By the way, look at the volatility in the oil ratio. You often see people talking about the stable gold:oil ratio but movements from 10 to 30 is an increase of 200% and a fall from 30 to 10 is a 66% reduction – hardly stable.


Next, if we look at agricultural commodities (click here for a bigger version). The left hand side has the breakfast club commodities – OJ, coffee, sugar, corn and bacon – and the right hand side has cotton, coca, soybeans, lumber and cattle. Only two of these ratios are within long term ranges with the rest at elevated levels.


Finally, the charts below compare gold to other metals (in these charts the ratio is inverted, so a low figure indicates gold is relatively more expensive – click here for a bigger version). Again, most of the ratios are at the extreme end of long term high/low trading ranges.


Of course the above charts may be reflecting the relative weakness of commodities but on that logic when commodities recover that will just put the ratios back into their normal ranges, meaning gold will not get a boost from that general money flow into commodities when (if) it happens. No doubt gold is attracting fear money at this time and moving up along with the US dollar – another example of its non-correlated behaviour with asset classes – but it is still too early to call as the chart below puts the move above $1,100 into context.


Gold has been above the 200 day moving average before and failed and it doesn’t take a technical analysis genius to look at the chart above and see that gold has to move to at least $1,150 and hold before shorts will lose their confidence (although recent equity jitters and questions about the strength of economic recovery would be giving them pause for thought).

In terms of Asian demand, Perth Mint is seeing good premiums on kilobars in the fact of this price rise. Normally demand dries up on price surges so this is positive but we wouldn’t classify demand as crazy either, which considering Yuan devaluation and the fact that early February is Chinese New Year could also be read negatively.

All up, I feel that gold is a bit stretched and thus likely to grind sideways in the medium term as it builds a base and other metals and commodities catch up.

Read this article in German on GoldSeiten.

Jan 042016

In August I did an analysis of the ideal percentage allocation between gold & silver. This assumed one picks a percentage allocation and sticks with it. Another investing approach is to switch between gold and silver based on one’s view of which metal will outperform the other in the future. One way to determine the point at which to switch is to use the gold/silver ratio.

Below is a chart of the gold/silver ratio, calculated by dividing the gold price by the silver price. Another way of looking at this chart is that is shows the price of gold in ounces of silver.


On the chart I have marked some reasonable high and low trading bands based on historical movements in the ratio. When the ratio is high, it is saying that gold’s price in ounces of silver is high, so sell gold and buy silver (light grey line). When it is low, gold is relatively cheap, so sell your silver and buy gold (gold line).

It is important when dealing with ratios to keep in mind that they are just a relative performance measure. Switching between metals at a high or low ratio does not guarantee dollar profits, as the ratio can change when both gold and silver are falling in price (in which case it is just telling you which metal is losing you less money).

You can see from the chart above that in February 2011 the ratio broke through 45, which based on history looked like a reasonable ratio low and thus a signal to sell silver and buy gold. The ratio then had a relatively steady climb to 75 in November 2014, a historically high point and with some commentators suggesting silver will be the outperforming metal going forward based on this high ratio.

However, the chart below shows the dollar performance of gold and silver since 2011.


First, note that the ratio had you out of silver and missing its 50% increase. Gold still went up initially but you can see from the chart that while the ratio moved from 45 to 75 over nearly four years, both gold and silver fell over that time period. The ratio signal did “work” in the sense that gold did outperform, only falling 15% compared to silver which fell 49% between Feb 2011 and Nov 2014.

The other caveat is that the ratio trading bands (or any trading bands) are not guaranteed to work in the future. You will notice on the ratio chart that prior to the mid 1980s a reasonable gold:silver ratio trading band appeared to be 30 for a switch to gold and 40 to switch to silver. However in 1983 you would have got into silver at 40 waiting for the ratio to get back to 30 as it had done many times in the past 12 years but instead it only got into the low 30s before climbing to nearly 100 in 1992.

It should be clear from the chart that something structural or fundamental changed in the gold or silver market during the mid 80s to early 90s with gold and silver moving into a new “relative” relationship. Possibly that structural change was the decision by the United States to liquidate its Strategic Stockpile of silver, as detailed here, see the chart below.


Note that the rate of stockpile reduction increases from 1985 onwards, around the time the gold:silver ratio begins its dramatic run to 100, and that when the stockpile drops to its first low point in 1993 the ratio breaks below 90 (which it has never seen again).

The takeaway is don’t just trade a ratio purely based on historical numbers without some theory as to how the two assets are related and whether the environment in which that relationship exists is likely to continue in the future.

May 112015

Each month The Perth Mint releases its minted coin/bar sales for gold and silver (see here for the latest). Data hoarder Nick at Sharelynx takes those figures and produces the chart below, which converts the ounces into dollar amounts (using a month end rate as an approximate).

Perth Mint Au:Ag Ratio

On average, our minted coin/bar customers spend between 20%-25% on silver as they do on gold . Contrast that with the US Mint, which as you can see from the chart below, often spend dollar for dollar on gold and silver.

US Mint Au:Ag Ratio

The average ratio for the US Mint over the same time period as The Perth Mint chart is 74%. The Royal Canadian Mint’s customers spent 62% on silver compared to gold in 2014. The difference between the mints is probably due to a different mix of customer preferences – for gold, silver, or for both.

Trying to establish the ratio for other precious metal businesses is difficult, because few report sales volumes in such detail. However we can work it out backwards for those who do report their total holdings of gold and silver.

To get a reliable figure we need a business which offers both gold and silver to the same customer base over the same time period. To eliminate the effect of the different price peaks in gold and silver and any reductions in holdings thereafter, I also think we need to compare the period March 2008 to March 2011, which is a period where inflows into gold and silver were strong and consistent. The ETFs are a potential source of inflow information, but the data I think is unreliable as the ETFs have different mixes of customers (individuals vs institutional) so we aren’t comparing like customer with like customer.

From Nick’s data there are three businesses that fit these criteria: Gold Money, Bullion Management Group and Central Fund of Canada. The respective ratios of dollar amount of silver bought to gold bought over those three years are: 103%, 99.4% and 99.8% – surprisingly consistent.

I also found a note from Eric Sprott in late 2012 that observed that the “last Gold Trust issue in September 2012 raised US$393 million and the last Silver Trust issue raised US$310 million”, which is a ratio of 78.8%, close to the US Mint ratio.

This prevalence of equal dollar investment in gold and silver is something that we see in the Perth Mint’s Depository business. When we segment our Depository customers by the metal they buy, around 35% buy equal dollar amounts of gold and silver. However, there are also clear groups of customers who either only buy gold, or only buy silver. There aren’t many who have mixes outside of those three groups.

So it seems that precious metal investors are either goldbugs, silverbugs, or balancedbugs.