Gold & Business Cycle Relationships
Here is a simplified graphic representation of The Business Cycle:
We can see how bonds do well during deflation, making their highs in the business cycle low. Stocks do well after the recession low passes, and commodities do best as inflation is peaking.
There is an overlap within each asset class of course. The stocks phase begins with bond-substitute defensive type stocks, which pay a dividend, highly valued during recessions. But the best performance soon rotates over to cyclical stocks, and then to growth stocks, and finally into high flyers, which tends to include some fresh paper, a surge of IPOs and then into junk as the market takes it’s percentage of loose money before it goes down.
Likewise, bonds are more or less risky depending upon the type of bond, and corporate bonds do best at a different phase of the bond bull market from eg sovereign bonds which are reputedly safer, but pay a lower rate of return.
So where is gold in all this?
I prefer to regard gold as occupying the place between “late” stocks and “early” commodities. In other words, a leading commodity. That reflects gold’s perceived function as an inflation hedge.
If only it was all so simple! But it isn’t!
For example, gold can act as “a fear indicator” rising to highs during times of conflict. At those times gold becomes regarded as the safest kind of money, and it takes over from bonds as the receptacle of risk avoiding capital flows.
So gold can move opposite to bonds and the same direction as late stocks ... a good amount of the time. But the rest of the time it can move in a different way, to reflect the cycles of war and disruption which sweep mankind. Oil tends to rise in those times too, so at those phases gold would correlate with the price of oil more than usual. Therefore we can expect a correlation to hold at times, and break at other times, returning to "normal mode" again sometime later.
Let’s look at the same asset classes, but in real life situations.
First the rotation from bonds to stocks :
Notice the times, marked when a top of bonds preceded a following top of stocks several months later. In fact, at these times when the stock market finally made it’s high, bonds had fallen so much that their first bear market low was already visible. This seems to confirm a rotation of capital from bonds into stocks, while also admitting that it is a minority of all the highs shown which demonstrate what is happening. So bonds have about 3 highs, or which only one of the three is an interesting high for the purposes of this discussion.
Case “E” in the chart above remains to be seen how it will work out.
How about rotation from stocks into gold?
During the 1970s the S&P (red) bottomed before gold. And at the 1987 crash it topped before gold. The 1982 low in both SPX and gold was at the same time. Other turns do not reveal anything.
How about later?
From the 90s to the 00s we see a different picture. In 1995 stocks (SPX in red) began to rise fast. This upturn decoupled from gold, which after the expected delay turned down, instead of up. In 2001 stocks topped and subsequently broke down, and a period later gold made it 2nd low, continuing the inverse correlation, but retaining the lead-lag relationship between the two.
In 2002 they recoupled, and both rise. In 2007 the SPX topped and turned down. Several months after the S&P gold made it’s high, and broke down into a decline.
And more recently – with a change from monthly to weekly timeframe in the timescale this is what we see:
The S&P made high in March 2011, continuing it’s role as “leader”, and gold went on to make following highs in August and a September 2011 high before breaking down. The gold lows in mid 2012, and the high in late 2012 were all preceded by the SPX.
But at end of 2012 the correlation reversed again, and the SPX moved up, with gold sinking. This has happened before.
The question arises as to causation and origin of these reversals in correlation between US stocks and gold measured in US dollars.
How about comparing gold to bonds?
Before getting tied up in that, it might be a good idea to first compare gold to bonds and see what is to be seen. I’ll skip to the last one and get to the point:
And it really appears that gold (green) is, in recent years, leading the TBond (blue) by about a year. This is interesting.
To sum up: a significant portion of the time, bonds lead stocks, stocks lead gold, gold leads bonds.
Occasionally the correlation between eg stocks and gold reverses direction, but the lead-lag relationship to stocks still seems to persist, though inverted. Not every turn is a significant turn for these purposes.
Right now, gold is leading the TBond downwards, and running about a year ahead of it. The TBond which leads stocks is going down (monthly basis) but stocks have gone in the opposite direction.
The implication might be that doing what it takes to support stocks is at the same time hurting the TBond.
That’s an interesting thought and it might be giving headaches to the-people-who-like-to-consider-themselves-in-charge. How far things can be stretched is always a tough question. But I think at CB level there might be a wakening acceptance, that they already know the answer, and every time they have new bonds to sell a reminder is served by the bond market. Playtime is coming to an end could be the message.
A last observation: while the anti-correlation is running, gold and stocks will probably retain their lead-lag relationship, inversely. We see to have gotten caught in a time warp operated by Central Bankers and it looks like we could be in the year 2000.
If the correlation was to go normal-wise for the coming year or two we could get something like this:
I have not gone into the geopolitical situations which pertained during the reverse correlation times, and that would be worth following up at a later stage. The conditions for reversal and return to normal of the correlations are also not investigated here, but worthy of looking into.
Still the clues dropped by the market are few, and always difficult to decipher, and this relationship appears to be one of them.