Gibson’s Paradox + Greenspan’s Conundrum = Overstreet’s Enigma

After all of my fooling around with variations of the gold/copper ratio in my most recent posts, who would have guessed I could have missed such an important variation as this:  gold/dollar=copper/10-year yields?

Over the course of the last decade, it is not hard to imagine that with commodities breaking records to the upside and the dollar and yields hitting record lows that the gold/dollar ratio will look a lot like the copper/yield ratio or an industrial metals/yield ratio.

industrial metals index/10-year yield ratio

dollar index

copper/10-year yield ratio

For whatever reason, the ratio of industrial metals/10-year yields is, if any conclusions can be drawn from these charts, a better ‘predictor’ of the gold/dollar ratio than it is of either gold or the dollar individually. I think that this is another confirmation of the gold/copper=dollar/yield ratio that we have been arguing for.

But, there are two questions that this raises, I think.

First, this seems to be a variation of Gibson’s Paradox and the observations of Barsky’s and Summer’s paper on that topic years ago (as far as I could understand it). They noted that gold and industrial metals had an inverse relationship with real interest rates. But, gold and industrial metals don’t appear to be interchangeable. If you run gold/yield vs copper/dollar, for example, you will not get the same results.

Insofar as rates fail to keep up with inflation (represented by copper, as well as other industrial metals in this instance), either gold will benefit and/or the dollar will be punished. In Summer’s paper, although he noted that industrial metals appeared to be more ‘sensitive’ to real interest rates than gold was, it would actually appear as if gold is just as sensitive as other metals are, provided you put it in the right context. (Silver appears to stand somewhere between gold and industrial metals on this score, unsurprisingly).

It seems rather curious, though, that neither gold nor the dollar (i.e., inversely) should manage to keep up with the copper/yield ratio individually. Why would gold fail to keep up with the copper/yield ratio in the first place, and why would a countermove in the dollar make up for gold’s ‘failure’? Or, is the failure the dollar’s? Or fiat currencies in general? After all, what we are calling the “dollar” here is just an index of the dollar against other currencies.

Frankly, I don’t know.

But, I would like to get to my second problem with some of this. That is where these formulae break down. If you look at the gold/dollar ratio above during the 1999-2001 period, which unusually for a P&F chart is quite distinct this time, it is not the action ‘predicted’ by the copper/yield ratio, which you can see below.

copper/yield ratio 1999-2002

It is 2000 which is the most curious year and the most relevant (for me) for trying to make sense of the oil/gold ratio. In December of 1999, the price of year-on-year oil hit the 80% mark—if you don’t know about my “Supersystem” or about Leeb’s Oil Indicator, this won’t make much sense perhaps—after which, unsurprisingly, the dollar rose and stocks buckled. What is unusual about this period is that industrial commodities continued to rise, even though interest rates were rising and the yield curve inverting. Copper, as you can see, outpaced the ten-year yield during the latter halves of 1999 and 2000, but gold fell and the dollar rose.

Oil, which is so often lumped together with gold, after already having jumped up against metals in 1999, continued to outperform them during 2000. In some sense, oil seems to have somehow made up for the failure of gold and the dollar to react ‘properly’ to the conditions represented by the copper/yield ratio.

oil/gold ratio 1999-2001

The only other thing that catches my eye as being perhaps somewhat unusual is the behavior of the yield curve. I have mentioned before that the oil/copper ratio seems to have a positive correlation with the yield curve, although I am not entirely sure which one leads the other. Oil/copper led the yield curve by a quarter or so during the 2003 peak and and lagged the 2007 bottom, but the bottom in oil/copper in early 1999 (or before, since I can’t see before 1999), came almost two years before the yield curve inverted.

Something of these weird relationships during the 1997-2003 period can be seen on the following chart.

the numbers along the x-axis refer to months; each vertical line marks a year

The problem for central bankers at this time, I would imagine, is that the markets were caught between commodities hitting unbelievable lows (remember $10 oil?) in the wake of the Asian currency crisis and the LTCM collapse and equities being surprisingly resilient relative to interest rates being raised, both in 1994 and in 1999-2000.

I would have to think that this is part of the conditions caused by an unprecedented high in the Dow/gold and P/E ratios, but I have no idea why. The nature of modern monetary policy creates these huge waves in Dow/gold which central bankers are ultimately unable to control when the extremes are reached. One month, it’s preventing catastrophic worldwide deflation, the next month is pricking an equity market that has gone parabolic. I can only guess that when the peak of a Dow/gold ratio arrives, that it sends a combination of signals that baffles the central banker (and everybody else).

Then there is Greenspan’s “conundrum”, the flattening of the yield curve that began in 2004 and culminated in the inversion from the end of 2005-2007. Apart from the fact that the oil/copper ratio had peaked the year before, it is difficult to draw any direct parallels between oil/gold and the yield curve, but I have a hunch that there is some sort of relationship that feeds into market events and conditions from —to return to our main example—1999-2000, which, to repeat, saw a shift from economic collapse to NASDAQ bubble, a virtual doubling in oil prices, alongside a rise in industrial commodities and the dollar and yields, along with a fall in gold (and silver, I believe). And, a violation of the 0.12 oil/gold ratio for the first time in over two decades and the action in the Dow/gold ratio which coincidentally (ha!) had also topped.

oil/copper ratio 1999-Oct 2008

10 year/2 year yield ratio 2000-2008

Of course, each market is driven by its own fundamentals (ha! ha!), so I shouldn’t be tearing my hair out over this or torturing people with P&F charts to try to drive home a point that I cannot prove or even articulate. I wish I could stay and work on this question more or clean up this post, but I have already spent all my energy and time for the day on this problem. I suspect that the solution lies in finding some way to combine the gold/copper=dollar/yield formula with the oil/copper:yield curve relationship in order to uncover the cause behind oil/gold, but I have to say oil/gold feels as far away as it ever has right now.


No comments yet

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: