Isaac Newton, lunar cycles, and the bond market

When I left off my last post, it appeared that ratios such as gold/platinum and gold/oil among others, such as silver/oil, had both a concurrent negative correlation and yet a positive forward correlation (by 16 months or so) with yields. If that is the case, however, then yields themselves would have, by extension, a positive correlation concurrently with themselves (obviously) and a negative correlation sixteen months out.

In other words, if everything we suggested at the end of our last post is true, then when yields move up, then sixteen months later, they should move down. Now, that is a tough proposition to consider, because bonds have been in a thirty year bull market, so that when we look at the data, it will almost inevitably be positively correlated with the sixteen months before or after a given period. Right?

We mentioned before that using the precious metal/industrial commodity ratios as forecasting tools tended to be an imperfect science. It does not always work and it generally gives us little idea of the magnitude of the coming move. Using a chart of yields inverted and forwarded by sixteen months is even more problematic, but we might as well look at the charts, or go back and revisit our old arguments.

each verticle line represents five years; the numbers refer to months

It is not especially easy to see from this perspective, but a few things stand out to me:

1. You can see something of the 5-year cycle bonds tend to go through, especially since the end of Bretton Woods, before which there seemed to be more of a 4-year cycle.

2. Yields have become more volatile.

3. There is a surprisingly not infrequent correlation between these two lines—I think we can say that much—especially in more recent decades, but also prior to that.

How seriously should we take these 16-month relationships, then?

We’ve been talking a lot about the possibility of silver/oil having a sixteen-month lead on yields in much the same way as we have been talking about gold/oil having this relationship.

Here is a chart of silver/oil forwarded sixteen months against yields from 1997-2002.

lower half courtesy of Kevin Klombies at IMRA

And, yet, concurrently, the relationship looks inverted.

courtesy Kevin Klombies at IMRA

These appear to be instances of what George Soros has called reflexivity, but even that concept, much like time cycles, provides little explanatory power. It is only a way to frame the question of causality, which I am still unable to answer.

The danger, of course, is that you are forcing patterns on to phenomena where none exist. Here, for example, is gold/oil supposedly leading the oil/gold ratio by—in this instance—40 months.

is this completely arbitrary?

Although I increasingly feel confident about the prevalence of these sixteen month cycles, one has to wonder about what would be so special about sixteen months, much like one wonders why 0.05 or 0.12 should be critical levels in the oil/gold ratio or why 80% increases in oil prices send the equity or gold markets into a tailspin while 75% can be shrugged off.

Sixteen months doesn’t seem to correspond to any economic time-frame that I know of. If it were fifteen months, at least that would provide the comfort of thinking of it as five quarters.

Unfortunately, looking for answers to that question is likely to lead one only further into the morass. Or, so it has me. For whatever reason, I decided to consider the period in terms of weeks, and the number I got was 69, which was rather meaningless, until I divided by the only number it can be divided by, three. That leaves you with three terms of 23 weeks. Bear with me.

If you try to Google “23” in reference to time periods or patterns, you get a lot of wackiness about the “23 Enigma” and its supposedly mystical or sinister qualities. There have been a couple of movies made about it, as well. The 23rd week of pregnancy may or may not be a significant milestone.

The number “69”, apart from some obvious references, usually leads one to biblical prophecy and to Isaac Newton, who kept cropping up during this portion of my investigation, although for a very simple reason. Newton had an interest in biblical prophecy, especially Daniel, and argued that Jesus and the Gospels hung on Daniel’s prophecies, including one recounting 69 weeks (or 70 weeks). I was startled for a moment, because this is also where Newton mentions the statue of four metals: gold, silver, bronze (a copper alloy), and iron. If only Daniel had known about oil—or perhaps the iron was platinum!

Irony aside….the only reference to sixteen months was the relatively new idea of  a “solar heartbeat” of sixteen month cycles which is somehow tied to 11 year solar cycles.

There’s really not a whole lot to hang your hat on. But, I did come across this reference to 23-week cycles in the market by an Eric McCurdy. I have updated the chart a little, and I may have marked the cycles differently, but you get the idea.

does gold operate on intermediate 23 week cycles?

So, naturally, it became necessary to try it on bonds.

do bonds operate on 23 week cycles?

Is this just a coincidence? Where does this kind of nonsense end?

Well, not yet.

The curious thing about these two charts is that in the case of gold, each of the alleged 23-week cycles is marked by a low or a consolidation, while in the case of the bond market—at least over the course of the last three years—yields pivot. And, this would be the mathematically required pattern, if a low were to generate a high in sixteen-months time. It is possible that the exceptionally bullish nature of the gold market in recent years actually conceals a similar high-low cycle there as well, in which case the markings on the gold chart are incorrect. But, let’s leave that for another time.

Assuming that this 23-week cycle has validity, then it would re-confirm the conclusion that we came to before on the basis of five-year cycles, the sixteen-month lead in commodities, as well as the sixteen-month bond cycle—namely that bonds are on the brink of turning down. The only problem is that none of these cyclicals are inevitable. But, there is a lot of circumstantial “evidence” to support the case for a turn down in bonds.

So, assuming that these 3 x 23-week cycles mean something, what could it possibly be? The only conclusion—which is not a conclusion but only another possible correlation—I can come to is that it is somehow tied to the lunar calendar. This is where Isaac Newton comes in again, because in his aborted attempt to reform the calendar, he attempted to include a 49-month Great Lunar Cycle (or something to that effect), with months consisting of 29.5 days, which is consistent with the synodic lunar calendar, which is slightly different from the sidereal (the things you learn when you try to make sense of the market) and is oriented around catching the more or less complete cycle of new moons. This gives us 23 weeks and 16 months (in our calendar) as subdivisions of this ancient calendrical problem.

By taking this line, however, it would seem as if we are more or less abandoning any sense of causality that is confined to market or more conventionally sociopolitical influences. That is, where we had once seen the silver/oil ratio ‘leading’, for example, this cyclical interpretation leaves us with merely a collection of ratios and assets that are somehow expressions of a lunar calendar that nobody uses, in addition to a number of other cycles, such as the five-year.

If these cycles do function, there is nothing that says that they determine the long-term direction of markets, unless of course, there are greater cycles, such as Kondratieff waves, that are deterministic. Even so, they require explanation.

Isaac Newton lost a fortune when the South Sea Bubble popped. Was his concern with 69 week lunar cycles a distraction from making sound investments or did he merely fail to apply them to his stock market speculation?

Advertisements

No comments yet

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s

%d bloggers like this: