Gold/Oil Ratio and Pairs Trade Examples

As part of a further evaluation of the context of gold prices (and hence monetary inflation concerns and outlook for the US dollar) I decided to look a bit into how gold prices were relating to other inflation sensitive issues. Oil was the one that I thought would offer the most interesting comparison. The idea I wanted to explore was whether the ratio of gold prices to oil prices was informative on the future price action of either of the two.

After doing some charting I looked for some historical data on the ratio and found the useful chart below. It is not a truly long-term chart, but does go back several decades and shows probably a good sampling of the range that the ratio swings back and forth in.

Click on Chart to Enlarge

The chart above shows the gold/oil ratio in standard units oz/barrel. The chart goes back to about 1973. The basic range for the ratio is between about 8 and 30. The long term average from what I've read on it is about 15.

Click on Chart to Enlarge

In the chart above I used stock charts to create the ratio for the last 3 years of prices. Notice that at the top of the commodity bull last year, with energy related commodities leading the way, the ratio fell very low, below 7. So gold was cheap relative to oil on a historical basis. The first chart shows a point where the ratio got down to about 6.2 in 2005, but the reading last year was the second lowest. The difference was that in 2005 gold shot up about 60% or so over the next year after the ratio got very low. Last year gold dropped about 30% in the months after the low and then got back to around even by this year.

From what I've looked at as far as ratio extremes, I don't see much predictability as far as gold moving up or down on an absolute basis after extremes in the ratio to oil. However, the ratio does seem to help determine points at which gold or oil is likely to out or under perform the other. Like much of the data and analysis I post on the blog, this is best looked at as a ratio that will undergo mean reversion on a long term basis. This means that the average of the ratio is pretty stable not trending over a long period of time. So when the ratio gets far away from the average, it is best to look for it to return back toward the long term average than to get further away. This sets up the opportunity to play the expected performance or relative strength of one versus the other via a pairs trade when the ratio reaches extremes.

A pairs trade is going long one stock and short another in the same sector. So you are market neutral but will benefit if your long position goes down less or up more on a percentage basis than the short position. So you benefit by accurately determining the relative strength of two related issues. As an example, when the market has come way down and starts to turn up you may buy Apple stock and short and equal dollar amount of Dell because you believe that Apple will display superior relative strength to Dell in a rising market.

So back to gold and oil, you could create a pairs trade by buying gold futures or ETFs and shorting oil futures ETFs if you believe that gold will outperform on a relative basis. So in the example of July 2008, the gold/oil ratio reached a historically low extreme indicating that gold was cheap relative to oil. Even though gold dropped about 30% over the next 4 months, oil dropped 60%. So if you had a long gold-short oil pairs trade you would be up 30% between the two and 15% return on total dollars invested at that point.

On the chart above I made some brief notes showing a simple possible strategy of using a close above/below the 50 day moving average to initiate and close a pairs trade after the ratio reached an extreme point. So when the ratio is in a very low historical position you would look to buy gold and short oil. If the ratio gets very high, you would buy oil and short gold. The last year shows a couple nice trades, one in each direction, using that simple method.

Now in a statistical sense the long term mean of the ratio is about 15. The +1 +2 +3 standard deviations are around 20, 25 and 30 respectively. The point is that you probably should not look to pairs trade something when the ratio is right around the average. You want to wait for it to get extreme to give a higher success rate. Currently the ratio is about 15 which is right around the long term average so is probably not a time to try to create a pairs trade.

Now unfortunately from anything I've looked at so far, I have not really found anything that gives me much insight into whether gold will go up or down from here on this basis. But the more you know how markets relate, the better investment decisions you should be able to make with more confidence, and the more opportunities you may recognize and know how to take advantage of.