Why Gas Prices and Gold Prices Are Not Always Correlated

Recently, an interesting article in the Washington Times discussed a direct relationship between gold and oil prices. Although historically there have been periods when this was true, there are many periods since 1986 when the movement of these prices diverged. Here is the statistical relationship found by the author of the article: 

"Downloading daily gold and West Texas oil price data from readily available sources, one can perform a calculation to estimate the price of oil based on the price of gold. For example, if the price of gold for the given day is $326.30, multiply that number by .0602 ounces of gold to get a value of $19.643 per barrel...On a sample of daily interval data from January 2, 1986 to March 3, 2012 (6,457 matching intervals), using basic statistics, a 2-sample t test was performed and the results are eye-opening. 

We can say that .0602 ounces of gold times the price of gold produces the price of a barrel of oil, with an standard error of just .421%, or an average difference of just $6.10 dollars between the calculated price and the actual price since 1986. For the last 26 years, the price of gold dictated the price of oil with 99.579 percent accuracy."

However, Figure 1 illustrates the natural log of both prices for an ounce of gold and a barrel of oil since 1986. 

Figure 1: Gold and Oil Price Relationship

Source: Fed FRED

Since these are in logs, the change between two periods is the percentage change. Therefore, one data point below the period before will indicate a negative percentage change and vice-versa. If no other market fundamentals affected these prices except for the value of the dollar, considering that both commodities are priced in dollars, then these two should increase and decrease together. As you can see, there are periods when these two commodities move closely together and other times when they do not.

The data indicate there is a relationship between the two, but there are many times when this link is not as strong. Since the prices of gold and oil are measured in dollars and in the long run, however long that may be, they should revert to an average relationship. The article notes this relationship to be ".0602 ounces of gold times the price of gold produces the price of a barrel of oil". 

Using this simple calculation with the price of gold closing at $1,607 per ounce this past Friday, then multiply this price by .0602 provides an oil price of $96.74 per barrel. However, oil closed at $91.39 per barrel on Friday. This $5.35 difference between the projected and actual oil price indicates that there are other market fundamentals at work. 

Economic research indicates that the fundamentals of supply and demand in the oil market are important to understand for short run and long run fluctuations. In particular, these fundamental changes allow more pricing information for which endeavors are profitable and which ones are not. Moreover, the primary source of oil price fluctuations from oil supply, oil demand, or precautionary oil demand can lead to significantly different economic effects. 

In my research, I find that a global demand for oil shock, represented by a real global economic activity measure that reflects global demand for commodities that is dominated by crude oil, has significantly greater economic effects than disturbances specific to the oil price. A global demand for oil shock is the major contributor to oil price fluctuations since the 1970s and has a larger impact on the unemployment rate, real GDP growth, and the inflation rate in the U.S. compared with a shock to the price of oil. 

What I hope to research further is what impact the Fed has on global demand for oil from monetary policy because other central banks tend to follow the Fed's actions. In particular, this synchronization of monetary policies would affect the value of the dollar, reflected by the price of gold, and change relative prices; thereby, changing global demand and the impacts on the U.S. economy that are not reflected from an oil price disturbance. 

Simply, the price of oil does not matter much. The sources of oil price movements are what matters. In addition, the Fed's contribution to these underlying causes from its impact on gold, the U.S. dollar, and global demand for oil also change the macroeconomic effects of an oil price shock. 

As noted in the article above, the author's simple calculation was wildly off from time to time from his conversion of the price of gold to forecast the price of oil. Therefore, the price of gold gives a valuation of the value of the dollar and provides an indication of the price of oil, all else constant, but it is not perfect and this is when the oil market fundamentals are helpful.