Wednesday, April 10, 2013

Say it With Me, There is Little Relationship Between Gold and GDP

Before I dive in here, let me say, yes, I am biased in this argument. However, I embarked on the following analysis with an open mind, testing the thesis of a correlative causation between GDP and gold. Additionally, I think you should know that I could buy into the thesis being thrown around by various the research groups and by the financial media that the weakness (and future expected weakness) in gold is due to a strengthening economy. For many, gold has been a safe haven investment, turned to when the markets were failing. A strengthening economy could cause some investors to sell out of the precious metal space in a move into other asset classes. Other contradictory anecdotes aside and considering that gold prices have increased since the beginning of the last decade, well before the 2008 market implosion, the historical data shows little to no relationship between gold price and GDP.

The following charts show the year-over-year and quarter-over-quarter change in the price of gold and real GDP.

Year-over-year


Quarter-over-quarter


See the relationship? Neither do I. Aside from from the period in the 1970's (and the great recession) where gold and GDP exhibited somewhat of a POSITIVE relationship, there appears to be little to no relationship in the data. Not one to rely on my eyes however, I ran a correlative study using the above data on a coincident basis and by lagging GDP by up to six months. The coincident correlation statistics showed their was little to no relationship between gold prices and GDP over the entire time period since the late 1960's Additionally and as I increased the lag time in GDP, the correlative statistics increased, topping out at about 27% at the six month mark. This does not bode well for the thesis that an increase in GDP will lead to lower gold prices.

Now, You may be saying, but the 1970's is skewing the data, as the highly inflationary environment had an out-sized affect on the results. Well, I also looked at the period since 1980, excluding the 1970's from the data set. The results show here show a more negative relationship between gold price and GDP, supporting the thesis. However, the relationship appears to be weak, at best, and is likely not economically significant. In fact, the greatest absolute negative correlation tops out at a -35% using a three month lag in GDP. Hardly a statistic to build a thesis around. More so, the relationship between gold and GDP since 2000 can be more described as one showing little to no correlation, positively or negatively. The relationship between changes in the price of gold and GDP seems flimsy, leading me to conclude that the thesis built around it is also weak.


No comments:

Post a Comment