History suggests that primary commodity prices have always approximated equity yields. Unfortunately, it does not tell us why, nor if commodity prices should continue to behave in this manner in the future, and theory is equally dumb. This has also been true for gold since it has been freely traded, and in my last article, I demonstrated that this relationship between commodities and yields does more to account for the fluctuations in the price of gold than do more popular explanations such as inflation, real interest rates, the value of the dollar, fear and manipulation. In this article I will show how the history of silver backs up this hypothesis.
Gold has only been priced by the market for four decades. In that time, it has demonstrated a few peculiar characteristics that set it somewhat apart from many other commodities. Having covered these in greater detail elsewhere, I will be brief.
1. Over the last forty years, gold has fared well against consumer inflation and better than most other commodities.
2. Gold has behaved like a "super-commodity," rising faster than other commodities (with a handful of exceptions) when equity yields rise and falling faster when equity yields fall.
3. Gold has a tendency to explode in spectacular fashion when equity yields top out. This was the case most notably in 1980 and again in 2011, as well as in 1974 and 2008.
These tendencies are even more pronounced in silver. In both 1980 and 2011, silver exploded and then collapsed amid finger pointing and accusations of manipulation. And silver, whether priced in terms of other commodities or in terms of CPI, looks much like the earnings yield.
In the chart below is silver priced in terms of inflation-adjusted dollars and the World Bank's ex-fuel index of commodities (also ex-precious metals) set beside the current and cyclically adjusted earnings yields. You can see points (2) and (3) demonstrated fairly clearly, as well as point (1) to a lesser degree.
In the chart below are correlations between the ratios of silver to World Bank commodity indexes and the earnings yields marked from both 1960 and shortly after the Nixon Shock in 1971. It appears that the silver/agriculture ratios were the weakest correlations, while industrial and precious metals correlations were the strongest. In the case of energy, I think the numbers are slightly misleading for the 1960-1972 period because oil had been pegged to the dollar during that time.
Although somewhat more difficult to see, here are correlations of the silver ratios for all commodities listed by the World Bank.
Blank spaces indicate that there was not enough data to return a correlation coefficient. The weakest correlations were for the silver/coffee and silver/tin ratios. Although the silver/coffee ratio is rather interesting, especially since the silver/tea ratio is rather strongly correlated with equity yields, the silver/tin one is more interesting due to the behavior of the tin market, about which I hope to write something soon. (Incidentally, silver/platinum and silver/potash possess the two strongest correlations with earnings yields.)
The real point of this demonstration is to show that both silver and gold not only behave like commodities but like super commodities. All of this is surprising for a few reasons.
1. As relatively useless (in the narrow sense) commodities, especially gold, it seems odd that they should be more susceptible to the laws governing commodities than are most other commodities.
2. This is precisely the opposite relationship that gold ratios had with equity yields prior to Bretton Woods. That is, under the gold standard and its dying remains up until World War II, commodity/gold ratios were positively correlated with equity yields, not inversely correlated.
If I am not mistaken, this would tend to suggest that prices under the gold standard were not, in fact, a function of the vicissitudes of gold but were a function of the real economy. Without going on too much of a tangent, in the wake of the uncoupling of commodity and consumer prices since the establishment of the Fed and especially after World War II, it would also seem that the gold standard's effect was to keep consumer prices (including services) roughly in sync with commodities.
But this is really only forty years of evidence, a time-span which only includes one clear top (early 1980s) in equity yields and one clear bottom (late 1990s). Even if I could squeeze in the low marked in the 1960s and the high of 2011, it's not overwhelming evidence on its own. It does coincide with a transformation in the behavior of the internal structure of consumer inflation and how consumer inflation interacts with bond and equity yields, and it is clearly a reversal of the gold standard correlations, so I think it is suggestive but hardly overwhelming.
This is where the history of silver prices gives us a little perspective and support.
Gold has only been demonetized and freely traded for forty years. Silver, however, was demonetized after the Civil War, over 140 years ago, and what we find is that, a) like the broader commodity sector, silver has been correlated with equity yields over that entire time, but b) its status or role as a super-commodity has only been evident after World War II. This suggests to me that there is something about the unbridled dollar that elevates these two former monetary metals from commodity to super-commodity.
But first, here is a chart of the nominal price of silver and the earnings yield since 1871. As you can see, after World War II, consumer inflation became a routine part of economic life, but the nominal price of silver continued to react to the earnings yield.
Below is the relative price of silver (i.e., silver/CPI) with the earnings yield. From the establishment of the Fed and especially from 1960, the relative price of silver more closely approximates the earnings yield.
(Sources: Shiller, GPIH, World Bank)
The transition from the nominal to the relative price is, as I have argued elsewhere, simply because that, while the relative price is the "real price" under the inflationary dollar standard, the nominal price was the "real price" under the gold standard. Again, this assumes that price fluctuations under the gold standard were a function of the real economy and not the shifting fortunes of gold.
To reiterate, the doubly interesting thing about this transition from nominal to relative prices was that gold and silver acquired the special super-commodity characteristics described above simultaneously. For most other commodities, one can gloss over this transition from nominal to relative prices because relative prices reflect the correlation throughout the last 140 years, but gold and silver have reacted in somewhat unique fashion to the shift from what appears to have been a non-inflationary monetary standard to an inflationary one.
As interesting as this is (at least to me), most readers will be curious about what this suggests about the future prospects for silver.
And, with respect to that, I think we can argue this much:
1. Apart from a metal being remonetized in the near future, it is likely that it will continue to correlate with equity yields.
2. Precious metals will likely continue to exhibit greater volatility than other commodities because of this super-commodity tendency.
3. Because the size of the collapse in precious metals prices confirms what the rest of the commodity sector and equities have been increasingly signaling since 2008-2009, it would appear that most of the commodity (GSC) action will be to the downside until equity yields find a new bottom.
That is the relatively short-term (probably a few years at most) outlook. The other thing to keep in mind is that the earnings yield is already rather low. One assumes that in a equity bull market (DIA), it could revisit its old lows from the 1990s or go even lower over the next few years, but there doesn't seem to be the kind of room that equities had to run from 1980 to 2000. Therefore, this commodity bear market will probably not be as long, but it could be quite painful in the interim. Needless to say, there are other possible scenarios, but at the moment, there would appear to be continued downward pressure on commodity prices and especially precious metals in our immediate future.
As a final note, I know that precious metals bugs take a keen interest in the behavior of the silver/gold ratio, and I thought they might be interested in seeing its performance relative to the earnings yield. The monthly chart from 1972 suggests that the relationship between silver and gold is at least partly a function of their relative dispositions towards equity yields.
An annual chart is less conclusive, although I think a case could be made that the transition from the gold standard to the dollar standard might have played some role in the breakdown in this relationship from World War II up until the Nixon Shock.
(Sources: GPIH, Shiller)
Additional disclosure: I am long June Dow futures and short gold and Brent futures.