April 1, 2005

Goldman Sachs rocked the markets this week with its call for a pop to $105 in crude oil prices.  I don't know what time frame they have for that, but it does not appear from the evidence I see that it will be possible within the next year or two.  The following constitutes a somewhat lengthy logic chain, which I hope will not stretch too far, so please stay with me on this.

The first chart shows our leading indication for crude oil prices, which is taken from looking at gold prices set forward by 11 months.  It is not perfect, because major geopolitical events can intervene to skew prices, as noted in the chart.  But if wars or other exogenous events don't pop up, then this gives a pretty good indication of what sort of monetary influences there will be on the supply/demand balance which results in a market price for oil.

Like many leading indications, this one is only reliable for price direction, and not for determining the exact price at which a move will end.  We cannot reliably do a linear regression on this relationship to say that X price of gold will equal Y price of oil 11 months later.  We can only make vague references to price levels.  With that in mind, the fact that gold moved to a slightly higher high in late 2004 suggests that by the end of 2005 oil prices should go higher than they have already been.  It is a pretty big stretch to conclude that oil should make it all the way to $105 on that move, because the late 2004 up move in gold was just not that robust.  $65 seems reasonable, but not $105.

If we assume that this relationship between gold and oil prices is going to continue to work into the future, then to get oil up to $105 we would either need to see an exogenous geopolitical event to create a temporary spike, or else we would need to see gold prices keep on rising.  To assess the likelihood of gold rising, we turn to the second chart which presents the biggest fundamental factor I know of regarding why gold prices move up and down.  When the Fed fails to keep short term interest rates above the inflation rate, it is effectively creating too much money.  Sometimes that is what the Fed wants to do, in order to stimulate the economy out of a recession, but too much of that behavior creates big problems for the economy, as we saw from the experience of the late 1970s.  So far, the rate hikes we have seen have not yet moved short term rates ahead of the CPI growth rate, but the Fed seems to be on course to do that provided that they keep raising rates.

So to summarize the logic chain so far, a sustained oil price of $105/barrel (not a temporary war spike) would require gold to keep going higher than it has been, far higher than we have seen so far in this up move.  And to bet on gold continuing far higher, one would have to bet on the Fed failing to get short term rates ahead of the inflation rate.

There is one additional factor working against this whole gold/oil boom scenario.  That is reflected in the final chart, which shows the 8-year cycle in gold prices.  This is admittedly a goofy looking cycle pattern, and not the pretty sort of sine wave that people usually associate with cycles, but it does appear to fit the available data much better than a sine wave would.  This 8-year cycle in gold is measured from top to top, as opposed to cycles in the stock market which are measured from bottom to bottom.  The reason for this is that cycle "bottoms" are panic-driven events.  People panic out of stocks and into cash, making for stock price lows at the cycle bottoms.  People panic out of cash and INTO gold, making for gold price highs at the cycle turning points. 

Right now, this cycle shows gold as being in the early part of a 5-year downward phase.  The late 2004 extra top was a lot like the extra top seen at the 1980 cycle high, which just meant that gold had to work even hard to catch up to where it was supposed to be (i.e. lower).  This downward phase does not end until 2009, at which time a 3-year rally phase begins. 

So to review again: a bet on the price of oil going far higher requires a bet on gold prices first moving far higher.  Betting on higher gold requires that one believe the Fed will remain behind the 8-ball in terms of interest rates and inflation, and that one also believe the 8-year cycle in gold prices will stop working.  That is a lot to bet against.  Alternatively, one could believe that all prior relationships between gold, oil, interest rates, and cycles would collectively stop working, in which case oil prices can do anything they want to. 

Tom McClellan
Editor
McClellan Market Report