Note: In this series, I am attempting to describe the fundamental relationship between yields and inflation.
Commodity prices have been highly correlated with equity yields for the last 140 years (inversely correlated with P/E ratios), and there is every indication that they were correlated with the rate of profit during the 140 years prior to that.
Under the dollar standard, all goods prices, including commodities, have been relatively suppressed. That is, when you take into account the decay of the value of the dollar, you will find that goods prices of nearly all varieties have been in a state of decline since Bretton Woods. Even taking into account the incredible commodity bull market of the last decade, this is still true.
This is counterintuitive for a couple of reasons which I will presently recount, but it is important for investors and the general public to disabuse themselves of this anachronistic and archaic notion.
The long collapse in commodity prices
The first reason people believe commodities go hand-in-hand with inflation is most likely because it used to be true. But, that was under the old gold standard.
Up until the early 20th Century, consumer prices moved up and down with commodities, although there appears to have been an ever so slight inflationary bias in consumer goods. Under the dollar standard, however, almost all systemic, chronic inflation originates in labor-intensive goods and services. Especially services, e.g. medical care, education, finance, law, etc.
The prices for goods (except technological goods) have risen in nominal terms over the decades, of course. But not nearly as much as would expect by the decline of the dollar.
That assumes, of course, that there is some reliable way of measuring the decline in the dollar. The only way that I know of is to estimate it by means of a broad basket of goods and services, such as CPI. This may be wrong, but I simply don't know of any alternative. Even dogmatic gold bugs out there likely would not argue that using the price of gold is an adequate measure. Otherwise, they would have to argue that during the 1980s and 1990s, the dollar strengthened and there was a general state of deflation, although clearly that was not true for services and a number of goods.
If one uses a broadly representative basket of goods and services, therefore, one will be forced to agree with me, I think, that the price of goods, including commodities, has generally fallen for the last sixty years. This is even more true if you subscribe to the numbers put out by ShadowStats. That site claims that consumer inflation has been far above the rates reported by the BLS for a good quarter of a century. But, if that is the case, then commodities have been made even cheaper than the BLS CPI data suggests.
Let me reiterate, for the benefit of the ShadowStats crowd, that the more attracted you are by the belief that inflation and the weakness of the dollar is underreported, the more you must accept my claim. My purpose is not to trick anyone into believing that inflation does not exist, which would be absurd, but to replace the widespread misapprehensions about the way inflation manifests itself in an economy that runs on a fiat currency and heavy credit expansion with a more precise rendering.
Gold and equity yields, reconsidered
This downward trajectory in real commodity prices coincides with the downward trajectory of equity yields (e.g., the higher highs and lows of P/E ratios). This is especially evident in dividend yields, which have been flattened since World War II.
Even the spread between equity and bond yields was often negative as well. This has not been caused by falling earnings, but rather by a robust stock market, another post-Fed development that has accelerated since World War II. But, stocks are not my primary concern in this article.
Gold is, and I'm afraid I have to make something of a confession.
There are some exceptions to the broad claims I made about the behavior of commodity prices and their relationship with equity yields. Let me enumerate them.
First, there are a handful of primary commodities that have not suffered under the reign of the dollar: precious metals and energy.
Below is a chart of the best-performing commodities and the Grilli-Yang Commodity Price Index (GYCPI). The GYCPI is not perfect for our purposes. It lacks energy commodities and includes a number of non-primary commodities (e.g., beef, lamb), and its weights are skewed to the 1970s, which seems to over-emphasize the price of certain metals in the early 20th century. But, it gives us much the same picture as producer prices and individual commodity prices, so it's good enough.
In the chart below, you can see the relative decline of the broad GYCPI index compared to the either horizontal or upward-sloping energy and metals commodities.
Second, because gold was money up until the 20th Century, in nominal terms, it was never correlated with equity yields. If one deflates it by CPI or PPI, it would have been inversely correlated with equity yields up until the mid-century revolution in prices.
So, the positive correlation between gold and equity yields, which I explained in a previous article, is a relatively new phenomenon.
This is where things get a bit weird.
The Quandary of the Gold Standard
I have been at pains to show that the correlation between commodities and equity yields is not a function of the gold standard but rather a function of the operations of the real economy and that it is a "real price" phenomenon.
If that is the case, this results in something of a quandary. It means that the conventional view of gold standard prices is false. Prices did not rise and fall because of the rising and falling value of gold, but due to the relatively autonomous operations of the economy. It also suggests that gold did not cause fluctuations in the value of the dollar. As a commodity, gold should rise and fall with equity yields, but it did not. Again, the nominal price remained flat; the price deflated by CPI was inversely correlated. Gold only began to correlate with equity yields once the gold standard was abandoned.
So, what was the function of gold?
Knowing that the dollar standard has primarily impacted consumer goods-raising services prices and equities, lowering goods prices-it almost seems as if gold's role was to keep consumer prices roughly in line with commodity prices and yields in line with each other in that happy equilibrium termed "Gibson's Paradox."
"Gibson's Paradox" was the observation made famous by Keynes that price levels and bond yields were correlated with one another during the gold standard, not the rate of inflation and bond yields. I have shown elsewhere that this was especially true for commodities and that this was probably more true for equity yields.
To put it in the words of intermarket analysis, during the gold standard, commodity/gold ratios were highly correlated with equity yields. This relationship then shifted to commodity/CPI ratios with the displacement of the gold standard by the dollar standard.
Gold as a price deflator
If I am right in all of this, it puts us in the awkward situation of having to speak of "real prices" as nominal prices under the gold standard and then "deflated prices" (i.e. prices deflated by CPI) afterwards. But, where does one end and the other begin? The dollar did not displace gold all at once. The process began with the establishment of the Fed, accelerated with Bretton Woods, and generally concluded with the end of Bretton Woods. Almost all of the essential pieces were in place by 1960, but no single date seems perfectly adequate to mark that transition.
It would be very nice to have some consistent way of representing prices that spans the gold standard, the dollar standard, and the muddle in the middle. Deflating commodity prices by CPI works well enough over the last century and a half, but it is theoretically and practically problematic, as I have discussed here and in previous articles. So, what about gold? What if we used gold to deflate commodity prices? In other words, gold ratios, just as in Keynes's formulation and, although in a somewhat convoluted fashion, the solution arrived at by Barsky and Summers in their standard account of Gibson's Paradox.
Unfortunately, as I have pointed out again and again, we already know that gold now conforms to the behavior of the broader commodity complex. For the last half-century, and especially in the last forty years, gold correlates with equity yields. So, it seems rather doubtful that gold ratios would still be meaningful.
But, what I found with respect to these commodity/gold ratios was rather surprising and contrary to my expectation.
Let me demonstrate it this way:
First, here is a snapshot of WPI/gold ratios (or, PPI/gold), commodity/gold ratios, and the dividend yield since 1900.
Again, the conditions that applied in the early parts of the 20th Century were in operation as far back as 1730: these gold ratios were correlated with yields. Things generally hold together until the Depression, but from then until 1950, the relationship between yields and gold ratios is hard to trace, and after 1950, it has completely broken down.
The New Gold Standard
But, actually, that's not what has happened. From 1950, this relationship simply reversed.
It was not just the relationship between gold/CPI and equity yields that reversed when the gold standard died off mid-century. With very few exceptions, from about 1950, all gold ratios reversed their correlations with equity yields. The exceptions are extremely important, but let's set those aside for the moment.
The other thing one has to note is that the level of the gold ratios has shifted significantly off of its historic axis. Roy Jastram's classic work on gold showed that for as far back as we have price data, producer prices (read: commodity prices) and gold have swung around a single axis.
In the current instance, even though gold fell from $800 in the early 1980s to somewhere around $250 at the end of the century, this was not enough to erase gold's outperformance, since gold was officially demonetized in the "Nixon Shock."
Moreover, considering the durability of the connection between commodities and equity yields, it is interesting that the relative strength of gold has coincided with the relative weakness of equity yields, especially the dividend yield.
I suspect that we will find that that is not coincidence, but here seems as good a place as any to pause in our treatment of the relationship between prices and yields.
In my next installment, assuming all goes to plan, I would like to show how those other commodities that I mentioned at the beginning of this essay, particularly oil, are the primary exceptions to this reversal of gold ratio correlations and how their behavior relates to equity yields and, by extension, stocks.